file0_1490

THIS IS YOUR BRAIN ON~ MUSIC~

DUTTONPublished by Penguin Group (USA) Inc.375 Hudson Street, New York, New York 10014, U.S.A.Penguin Group (Canada), 90 Eglinton Avenue East, Suite 700, Toronto, Ontario M4P 2Y3,Canada (a division of Pearson Penguin Canada Inc.); Penguin Books Ltd, 80 Strand, LondonWC2R 0RL, England; Penguin Ireland, 25 St Stephen’s Green, Dublin 2, Ireland (a divisionof Penguin Books Ltd); Penguin Group (Australia), 250 Camberwell Road, Camberwell,Victoria 3124, Australia (a division of Pearson Australia Group Pty Ltd); Penguin BooksIndia Pvt Ltd, 11 Community Centre, Panchsheel Park, New Delhi – 110 017, India; PenguinGroup (NZ), cnr Airborne and Rosedale Roads, Albany, Auckland 1310, New Zealand(a division of Pearson New Zealand Ltd); Penguin Books (South Africa) (Pty) Ltd,24 Sturdee Avenue, Rosebank, Johannesburg 2196, South AfricaPenguin Books Ltd, Registered Offices: 80 Strand, London WC2R 0RL, EnglandPublished by Dutton, a member of Penguin Group (USA) Inc.First electronic edition, August 200610 9 8 7 6 5 4 3 2 1Copyright © 2006 by Daniel J. LevitinAll rights reservedREGISTERED TRADEMARK—MARCA REGISTRADALIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATALevitin, Daniel J.This is your brain on music : the science of a human obsession / Daniel J. Levitin.p. cm.Includes bibliographical references and index.MSR ISBN 0-7865-8404-1AEB ISBN 0-7865-8405-XSet in ITC Century BookWithout limiting the rights under copyright reserved above, no part of this publication maybe reproduced, stored in or introduced into a retrieval system, or transmitted, in any form,or by any means (electronic, mechanical, photocopying, recording, or otherwise), withoutthe prior written permission of both the copyright owner and the above publisher of thisbook.The scanning, uploading, and distribution of this book via the Internet or via any othermeans without the permission of the publisher is illegal and punishable by law. Please purchaseonly authorized electronic editions, and do not participate in or encourage electronicpiracy of copyrighted materials. Your support of the author’s rights is appreciated.While the author has made every effort to provide accurate telephone numbers and Internetaddresses at the time of publication, neither the publisher nor the author assumes any responsibilityfor errors, or for changes that occur after publication. Further, the publisherdoes not have any control over and does not assume any responsibility for author or thirdpartyWeb sites or their content.Making or distributing electronic copies of this book constitutes copyright infringement and could subjectthe infringer to criminal and civil liability.www.us.penguingroup.com

CONTENTSIntroductionI Love Music and I Love Science—Why Would I Wantto Mix the Two? 11. What Is Music?From Pitch to Timbre 132. Foot TappingDiscerning Rhythm, Loudness, and Harmony 553. Behind the CurtainMusic and the Mind Machine 814. AnticipationWhat We Expect From Liszt (and Ludacris) 1095. You Know My Name, Look Up the NumberHow We Categorize Music 1296. After Dessert, Crick Was Still Four Seats Away from MeMusic, Emotion, and the Reptilian Brain 165

viContents7. What Makes a Musician?Expertise Dissected 1898. My Favorite ThingsWhy Do We Like the Music We Like? 2179. The Music InstinctEvolution’s #1 Hit 241Appendices 263Bibliographic Notes 271Acknowledgments 301Index 303

IntroductionI Love Music and I Love Science—Why Would I Want to Mix the Two?I love science, and it pains me to think that so many are terrified of thesubject or feel that choosing science means you cannot also choosecompassion, or the arts, or be awed by nature. Science is not meant tocure us of mystery, but to reinvent and reinvigorate it.—Robert Sapolsky, Why Zebras Don’t Get Ulcers, p. xiiIn the summer of 1969, when I was eleven, I bought a stereo system atthe local hi-fi shop. It cost all of the hundred dollars I had earnedweeding neighbors’ gardens that spring at seventy-five cents an hour.I spent long afternoons in my room, listening to records: Cream, theRolling Stones, Chicago, Simon and Garfunkel, Bizet, Tchaikovsky,George Shearing, and the saxophonist Boots Randolph. I didn’t listenparticularly loud, at least not compared to my college days when I actuallyset my loudspeakers on fire by cranking up the volume too high, butthe noise was evidently too much for my parents. My mother is a novelist;she wrote every day in the den just down the hall and played the pianofor an hour every night before dinner. My father was a businessman;he worked eighty-hour weeks, forty of those hours in his office at homeon evenings and weekends. Being the businessman that he was, my fathermade me a proposition: He would buy me a pair of headphones if Iwould promise to use them when he was home. Those headphones foreverchanged the way I listened to music.The new artists that I was listening to were all exploring stereo mixingfor the first time. Because the speakers that came with my hundreddollarall-in-one stereo system weren’t very good, I had never beforeheard the depth that I could hear in the headphones—the placement of

2 Introductioninstruments both in the left-right field and in the front-back (reverberant)space. To me, records were no longer just about the songs anymore,but about the sound. Headphones opened up a world of sonic colors, apalette of nuances and details that went far beyond the chords andmelody, the lyrics, or a particular singer’s voice. The swampy Deep Southambience of “Green River” by Creedence, or the pastoral, open-spacebeauty of the Beatles’ “Mother Nature’s Son”; the oboes in Beethoven’sSixth (conducted by Karajan), faint and drenched in the atmosphere of alarge wood-and-stone church; the sound was an enveloping experience.Headphones also made the music more personal for me; it was suddenlycoming from inside my head, not out there in the world. This personalconnection is ultimately what drove me to become a recording engineerand producer.Many years later, Paul Simon told me that the sound is always whathe was after too. “The way that I listen to my own records is for thesound of them; not the chords or the lyrics—my first impression is of theoverall sound.”I dropped out of college after the incident with the speakers in mydorm room, and I joined a rock band. We got good enough to record at atwenty-four-track studio in California with a talented engineer, MarkNeedham, who went on to record hit records by Chris Isaak, Cake, andFleetwood Mac. Mark took a liking to me, probably because I was theonly one interested in going into the control room to hear back what wesounded like, while the others were more interested in getting high in betweentakes. Mark treated me like a producer, although I didn’t knowwhat one was at the time, asking me what the band wanted to sound like.He taught me how much of a difference to the sound a microphone couldmake, or even the influence of how a microphone was placed. At first, Ididn’t hear some of the differences he pointed out, but he taught mewhat to listen for. “Notice that when I put this microphone closer to theguitar amp, the sound becomes fuller, rounder, and more even; but whenI put it farther back, it picks up some of the sound of the room, giving ita more spacious sound, although you lose some of the midrange if I dothat.”

Introduction 3Our band became moderately well known in San Francisco, and ourtapes played on local rock radio stations. When the band broke up—dueto the guitarist’s frequent suicide attempts and the vocalist’s nasty habitof taking nitrous oxide and cutting himself with razor blades—I foundwork as a producer of other bands. I learned to hear things I had neverheard before: the difference between one microphone and another, evenbetween one brand of recording tape and another (Ampex 456 tape hada characteristic “bump” in the low-frequency range, Scotch 250 had acharacteristic crispness in the high frequencies, and Agfa 467 a luster inthe midrange). Once I knew what to listen for, I could tell Ampex fromScotch or Agfa tape as easily as I could tell an apple from a pear or an orange.I progressed to work with other great engineers, like Leslie AnnJones (who had worked with Frank Sinatra and Bobby McFerrin), FredCatero (Chicago, Janis Joplin), and Jeffrey Norman (John Fogerty, theGrateful Dead). Even though I was the producer—the person in chargeof the sessions—I was intimidated by them all. Some of the engineers letme sit in on their sessions with other artists, such as Heart, Journey, Santana,Whitney Houston, and Aretha Franklin. I got a lifetime of educationwatching them interact with the artists, talking about subtle nuances inhow a guitar part was articulated or how a vocal performance had beendelivered. They would talk about syllables in a lyric, and choose amongten different performances. They could hear so well; how did they traintheir ears to hear things that mere mortals couldn’t?While working with small, unknown bands, I got to know the studiomanagers and engineers, and they steered me toward better and betterwork. One day an engineer didn’t show up and I spliced some tape editsfor Carlos Santana. Another time, the great producer Sandy Pearlmanwent out for lunch during a Blue Öyster Cult session and left me incharge to finish the vocals. One thing led to another, and I spent over adecade producing records in California; I was eventually lucky enoughto be able to work with many well-known musicians. But I also workedwith dozens of musical no-names, people who are extremely talentedbut never made it. I began to wonder why some musicians becomehousehold names while others languish in obscurity. I also wondered

4 Introductionwhy music seemed to come so easily to some and not others. Wheredoes creativity come from? Why do some songs move us so and othersleave us cold? And what about the role of perception in all of this, the uncannyability of great musicians and engineers to hear nuances that mostof us don’t?These questions led me back to school for some answers. While stillworking as a record producer, I drove down to Stanford University twicea week with Sandy Pearlman to sit in on neuropsychology lectures byKarl Pribram. I found that psychology was the field that held the answersto some of my questions—questions about memory, perception, creativity,and the common instrument underlying all of these: the human brain.But instead of finding answers, I came away with more questions—as isoften the case in science. Each new question opened my mind to an appreciationfor the complexity of music, of the world, and of the humanexperience. As the philosopher Paul Churchland notes, humans havebeen trying to understand the world throughout most of recorded history;in just the past two hundred years, our curiosity has revealed muchof what Nature had kept hidden from us: the fabric of space-time, theconstitution of matter, the many forms of energy, the origins of the universe,the nature of life itself with the discovery of DNA, and the completionof the mapping of the human genome just five years ago. But onemystery has not been solved: the mystery of the human brain and how itgives rise to thoughts and feelings, hopes and desires, love, and the experienceof beauty, not to mention dance, visual art, literature, and music.What is music? Where does it come from? Why do some sequences ofsounds move us so, while others—such as dogs barking or cars screeching—makemany people uncomfortable? For some of us, these questionsoccupy a large part of our life’s work. For others, the idea of picking musicapart in this way seems tantamount to studying the chemical structurein a Goya canvas, at the expense of seeing the art that the painterwas trying to produce. The Oxford historian Martin Kemp points out asimilarity between artists and scientists. Most artists describe their workas experiments—part of a series of efforts designed to explore a com-

Introduction 5mon concern or to establish a viewpoint. My good friend and colleagueWilliam Forde Thompson (a music cognition scientist and composer atthe University of Toronto) adds that the work of both scientists andartists involves similar stages of development: a creative and exploratory“brainstorming” stage, followed by testing and refining stages that typicallyinvolve the application of set procedures, but are often informed byadditional creative problem-solving. Artists’ studios and scientists’ laboratoriesshare similarities as well, with a large number of projects goingat once, in various stages of incompletion. Both require specializedtools, and the results are—unlike the final plans for a suspension bridge,or the tallying of money in a bank account at the end of the businessday—open to interpretation. What artists and scientists have in commonis the ability to live in an open-ended state of interpretation and reinterpretationof the products of our work. The work of artists and scientistsis ultimately the pursuit of truth, but members of both camps understandthat truth in its very nature is contextual and changeable, dependent onpoint of view, and that today’s truths become tomorrow’s disproven hypothesesor forgotten objets d’art. One need look no further than Piaget,Freud, and Skinner to find theories that once held widespread currencyand were later overturned (or at least dramatically reevaluated). In music,a number of groups were prematurely held up as of lasting importance:Cheap Trick were hailed as the new Beatles, and at one time theRolling Stone Encyclopedia of Rock devoted as much space to Adamand the Ants as they did to U2. There were times when people couldn’timagine a day when most of the world would not know the names PaulStookey, Christopher Cross, or Mary Ford. For the artist, the goal of thepainting or musical composition is not to convey literal truth, but an aspectof a universal truth that if successful, will continue to move and totouch people even as contexts, societies, and cultures change. For thescientist, the goal of a theory is to convey “truth for now”—to replace anold truth, while accepting that someday this theory, too, will be replacedby a new “truth,” because that is the way science advances.Music is unusual among all human activities for both its ubiquity andits antiquity. No known human culture now or anytime in the recorded

6 Introductionpast lacked music. Some of the oldest physical artifacts found in humanand protohuman excavation sites are musical instruments: bone flutesand animal skins stretched over tree stumps to make drums. Wheneverhumans come together for any reason, music is there: weddings, funerals,graduation from college, men marching off to war, stadium sportingevents, a night on the town, prayer, a romantic dinner, mothers rockingtheir infants to sleep, and college students studying with music as abackground. Even more so in nonindustrialized cultures than in modernWestern societies, music is and was part of the fabric of everyday life.Only relatively recently in our own culture, five hundred years or so ago,did a distinction arise that cut society in two, forming separate classes ofmusic performers and music listeners. Throughout most of the worldand for most of human history, music making was as natural an activityas breathing and walking, and everyone participated. Concert halls, dedicatedto the performance of music, arose only in the last several centuries.Jim Ferguson, whom I have known since high school, is now a professorof anthropology. Jim is one of the funniest and most fiercely intelligentpeople I know, but he is shy—I don’t know how he manages toteach his lecture courses. For his doctoral degree at Harvard, he performedfieldwork in Lesotho, a small nation completely surrounded bySouth Africa. There, studying and interacting with local villagers, Jim patientlyearned their trust until one day he was asked to join in one oftheir songs. So, typically, when asked to sing with these Sotho villagers,Jim said in a soft voice, “I don’t sing,” and it was true: We had been inhigh school band together and although he was an excellent oboe player,he couldn’t carry a tune in a bucket. The villagers found his objectionpuzzling and inexplicable. The Sotho consider singing an ordinary,everyday activity performed by everyone, young and old, men andwomen, not an activity reserved for a special few.Our culture, and indeed our very language, makes a distinction betweena class of expert performers—the Arthur Rubinsteins, EllaFitzgeralds, Paul McCartneys—and the rest of us. The rest of us paymoney to hear the experts entertain us. Jim knew that he wasn’t much of

Introduction 7a singer or dancer, and to him, a public display of singing and dancing impliedhe thought himself an expert. The villagers just stared at Jim andsaid, “What do you mean you don’t sing?! You talk!” Jim told me later, “Itwas as odd to them as if I told them that I couldn’t walk or dance, eventhough I have both my legs.” Singing and dancing were a natural activityin everybody’s lives, seamlessly integrated and involving everyone. TheSesotho verb for singing (ho bina), as in many of the world’s languages,also means to dance; there is no distinction, since it is assumed thatsinging involves bodily movement.A couple of generations ago, before television, many families wouldsit around and play music together for entertainment. Nowadays there isa great emphasis on technique and skill, and whether a musician is “goodenough” to play for others. Music making has become a somewhat reservedactivity in our culture, and the rest of us listen. The music industryis one of the largest in the United States, employing hundreds ofthousands of people. Album sales alone bring in $30 billion a year, andthis figure doesn’t even account for concert ticket sales, the thousands ofbands playing Friday nights at saloons all over North America, or thethirty billion songs that were downloaded free through peer-to-peer filesharing in 2005. Americans spend more money on music than on sex orprescription drugs. Given this voracious consumption, I would say thatmost Americans qualify as expert music listeners. We have the cognitivecapacity to detect wrong notes, to find music we enjoy, to rememberhundreds of melodies, and to tap our feet in time with the music—an activitythat involves a process of meter extraction so complicated thatmost computers cannot do it. Why do we listen to music, and why are wewilling to spend so much money on music listening? Two concert ticketscan easily cost as much as a week’s food allowance for a family of four,and one CD costs about the same as a work shirt, eight loaves of bread,or basic phone service for a month. Understanding why we like musicand what draws us to it is a window on the essence of human nature.To ask questions about a basic, and omnipresent human ability is to implicitlyask questions about evolution. Animals evolved certain physical

8 Introductionforms as a response to their environment, and the characteristics thatconferred an advantage for mating were passed down to the next generationthrough the genes.A subtle point in Darwinian theory is that living organisms—whetherplants, viruses, insects, or animals—coevolved with the physical world.In other words, while all living things are changing in response to theworld, the world is also changing in response to them. If one species developsa mechanism to keep away a particular predator, that predator’sspecies is then under evolutionary pressure either to develop a means toovercome that defense or to find another food source. Natural selectionis an arms race of physical morphologies changing to catch up with oneanother.A relatively new scientific field, evolutionary psychology, extends thenotion of evolution from the physical to the realm of the mental. Mymentor when I was a student at Stanford University, the cognitive psychologistRoger Shepard, notes that not just our bodies but our mindsare the product of millions of years of evolution. Our thought patterns,our predispositions to solve problems in certain ways, our sensory systems—suchas the ability to see color (and the particular colors wesee)—are all products of evolution. Shepard pushes the point still further:Our minds coevolved with the physical world, changing in responseto ever-changing conditions. Three of Shepard’s students, Leda Cosmidesand John Tooby of the University of California at Santa Barbara,and Geoffrey Miller of the University of New Mexico, are among those atthe forefront of this new field. Researchers in this field believe that theycan learn a lot about human behavior by considering the evolution of themind. What function did music serve humankind as we were evolvingand developing? Certainly the music of fifty thousand and one hundredthousand years ago is very different from Beethoven, Van Halen, or Eminem.As our brains have evolved, so has the music we make with them,and the music that we want to hear. Did particular regions and pathwaysevolve in our brains specifically for making and listening to music?Contrary to the old, simplistic notion that art and music are processedin the right hemisphere of our brains, with language and mathe-

Introduction 9matics in the left, recent findings from my laboratory and those of mycolleagues are showing us that music is distributed throughout the brain.Through studies of people with brain damage, we’ve seen patients whohave lost the ability to read a newspaper but can still read music, or individualswho can play the piano but lack the motor coordination tobutton their own sweater. Music listening, performance, and compositionengage nearly every area of the brain that we have so far identified,and involve nearly every neural subsystem. Could this fact account forclaims that music listening exercises other parts of our minds; that listeningto Mozart twenty minutes a day will make us smarter?The power of music to evoke emotions is harnessed by advertisingexecutives, filmmakers, military commanders, and mothers. Advertisersuse music to make a soft drink, beer, running shoe, or car seem more hipthan their competitors’. Film directors use music to tell us how to feelabout scenes that otherwise might be ambiguous, or to augment our feelingsat particularly dramatic moments. Think of a typical chase scene inan action film, or the music that might accompany a lone woman climbinga staircase in a dark old mansion: Music is being used to manipulateour emotions, and we tend to accept, if not outright enjoy, the power ofmusic to make us experience these different feelings. Mothers throughoutthe world, and as far back in time as we can imagine, have used softsinging to soothe their babies to sleep, or to distract them from somethingthat has made them cry.Many people who love music profess to know nothing about it. I’vefound that many of my colleagues who study difficult, intricate topicssuch as neurochemistry or psychopharmacology feel unprepared to dealwith research in the neuroscience of music. And who can blame them?Music theorists have an arcane, rarified set of terms and rules that are asobscure as some of the most esoteric domains of mathematics. To thenonmusician, the blobs of ink on a page that we call music notationmight just as well be the notations of mathematical set theory. Talk ofkeys, cadences, modulation, and transposition can be baffling.Yet every one of my colleagues who feels intimidated by such jargon

10 Introductioncan tell me the music that he or she likes. My friend Norman White is aworld authority on the hippocampus in rats, and how they remember differentplaces they’ve visited. He is a huge jazz fan, and can talk expertlyabout his favorite artists. He can instantly tell the difference betweenDuke Ellington and Count Basie by the sound of the music, and can eventell early Louis Armstrong from late. Norm doesn’t have any knowledgeabout music in the technical sense—he can tell me that he likes a certainsong, but he can’t tell me what the names of the chords are. He is, however,an expert in knowing what he likes. This is not at all unusual, ofcourse. Many of us have a practical knowledge of things we like, and cancommunicate our preferences without possessing the technical knowledgeof the true expert. I know that I prefer the chocolate cake at onerestaurant I often go to, over the chocolate cake at my neighborhoodcoffee shop. But only a chef would be able to analyze the cake—to decomposethe taste experience into its elements—by describing thedifferences in the kind of flour, or the shortening, or the type of chocolateused.It’s a shame that many people are intimidated by the jargon musicians,music theorists, and cognitive scientists throw around. There isspecialized vocabulary in every field of inquiry (try to make sense of afull blood-analysis report from your doctor). But in the case of music,music experts and scientists could do a better job of making their workaccessible. That is something I tried to accomplish in this book. The unnaturalgap that has grown between musical performance and music listeninghas been paralleled by a gap between those who love music (andlove to talk about it) and those who are discovering new things abouthow it works.A feeling my students often confide to me is that they love life and itsmysteries, and they’re afraid that too much education will steal awaymany of life’s simple pleasures. Robert Sapolsky’s students have probablyconfided much the same to him, and I myself felt the same anxiety in1979, when I moved to Boston to attend the Berklee College of Music.What if I took a scholarly approach to studying music and, in analyzing

Introduction 11it, stripped it of its mysteries? What if I became so knowledgeable aboutmusic that I no longer took pleasure from it?I still take as much pleasure from music as I did from that cheap hi-fithrough those headphones. The more I learned about music and aboutscience the more fascinating they became, and the more I was able to appreciatepeople who are really good at them. Like science, music overthe years has proved to be an adventure, never experienced exactly thesame way twice. It has been a source of continual surprise and satisfactionfor me. It turns out science and music aren’t such a bad mix.This book is about the science of music, from the perspective of cognitiveneuroscience—the field that is at the intersection of psychologyand neurology. I’ll discuss some of the latest studies I and other researchersin our field have conducted on music, musical meaning, andmusical pleasure. They offer new insights into profound questions. Ifall of us hear music differently, how can we account for pieces thatseem to move so many people—Handel’s Messiah or Don McLean’s“Vincent (Starry Starry Night)” for example? On the other hand, if we allhear music in the same way, how can we account for wide differences inmusical preference—why is it that one man’s Mozart is another man’sMadonna?The mind has been opened up in the last few years by the explodingfield of neuroscience and the new approaches in psychology due to newbrain-imaging technologies, drugs able to manipulate neurotransmitterssuch as dopamine and serotonin, and plain old scientific pursuit. Lesswell known are the extraordinary advances we have been able to makein modeling how our neurons network, thanks to the continuing revolutionin computer technology. We are coming to understand computationalsystems in our head like never before. Language now seems to besubstantially hardwired into our brains. Even consciousness itself is nolonger hopelessly shrouded in a mystical fog, but is rather somethingthat emerges from observable physical systems. But no one until nowhas taken all this new work together and used it to elucidate what is forme the most beautiful human obsession. Your brain on music is a way to

12 Introductionunderstand the deepest mysteries of human nature. That is why I wrotethis book.By better understanding what music is and where it comes from, wemay be able to better understand our motives, fears, desires, memories,and even communication in the broadest sense. Is music listening morealong the lines of eating when you’re hungry, and thus satisfying an urge?Or is it more like seeing a beautiful sunset or getting a backrub, whichtriggers sensory pleasure systems in the brain? Why do people seemto get stuck in their musical tastes as they grow older and cease experimentingwith new music? This is the story of how brains and music coevolved—whatmusic can teach us about the brain, what the braincan teach us about music, and what both can teach us about ourselves.

1. What Is Music?From Pitch to TimbreWhat is music? To many, “music” can only mean the great masters—Beethoven, Debussy, and Mozart. To others, “music” is BustaRhymes, Dr. Dre, and Moby. To one of my saxophone teachers at BerkleeCollege of Music—and to legions of “traditional jazz” aficionados—anything made before 1940 or after 1960 isn’t really music at all. I hadfriends when I was a kid in the sixties who used to come over to myhouse to listen to the Monkees because their parents forbade them to listento anything but classical music, and others whose parents wouldonly let them listen to and sing religious hymns. When Bob Dylan daredto play an electric guitar at the Newport Folk Festival in 1965, peoplewalked out and many of those who stayed, booed. The Catholic Churchbanned music that contained polyphony (more than one musical partplaying at a time), fearing that it would cause people to doubt the unityof God. The church also banned the musical interval of an augmentedfourth, the distance between C and F-sharp and also known as a tritone(the interval in Leonard Bernstein’s West Side Story when Tony sings thename “Maria”). This interval was considered so dissonant that it musthave been the work of Lucifer, and so the church named it Diabolus inmusica. It was pitch that had the medieval church in an uproar. And itwas timbre that got Dylan booed.

14 This Is Your Brain on MusicThe music of avant-garde composers such as Francis Dhomont,Robert Normandeau, or Pierre Schaeffer stretches the bounds of whatmost of us think music is. Going beyond the use of melody and harmony,and even beyond the use of instruments, these composers use recordingsof found objects in the world such as jackhammers, trains, and waterfalls.They edit the recordings, play with their pitch, and ultimatelycombine them into an organized collage of sound with the same type ofemotional trajectory—the same tension and release—as traditional music.Composers in this tradition are like the painters who stepped outsideof the boundaries of representational and realistic art—the cubists,the Dadaists, many of the modern painters from Picasso to Kandinsky toMondrian.What do the music of Bach, Depeche Mode, and John Cage fundamentallyhave in common? On the most basic level, what distinguishesBusta Rhymes’s “What’s It Gonna Be?!” or Beethoven’s “Pathétique”Sonata from, say, the collection of sounds you’d hear standing in themiddle of Times Square, or those you’d hear deep in a rainforest? Asthe composer Edgard Varèse famously defined it, “Music is organizedsound.”This book drives at a neuropsychological perspective on how musicaffects our brains, our minds, our thoughts, and our spirit. But first, it ishelpful to examine what music is made of. What are the fundamentalbuilding blocks of music? And how, when organized, do they give rise tomusic? The basic elements of any sound are loudness, pitch, contour, duration(or rhythm), tempo, timbre, spatial location, and reverberation.Our brains organize these fundamental perceptual attributes into higherlevelconcepts—just as a painter arranges lines into forms—and theseinclude meter, harmony, and melody. When we listen to music, we are actuallyperceiving multiple attributes or “dimensions.” Here is a brief summaryof them.~ A discrete musical sound is usually called a tone. The word note isalso used, but scientists reserve that word to refer to somethingthat is notated on a page or score of music. The two terms, tone

What Is Music? 15and note, refer to the same entity in the abstract, where the wordtone refers to what you hear, and the word note refers to what yousee written on a musical score.~ Pitch is a purely psychological construct, related both to the actualfrequency of a particular tone and to its relative position in the musicalscale. It provides the answer to the question “What note is that?”(“It’s a C-sharp.”) I’ll define frequency and musical scale below.~ Rhythm refers to the durations of a series of notes, and to the waythat they group together into units. For example, in the “AlphabetSong” (the same as “Twinkle, Twinkle Little Star”) the notes of thesong are all equal in duration for the letters A B C D E F G H I J K(with an equal duration pause, or rest, between G and H), and thenthe following four letters are sung with half the duration, or twiceas fast per letter: L M N O (leading generations of schoolchildrento spend several early months believing that there was a letter inthe English alphabet called ellemmenno).~ Tempo refers to the overall speed or pace of the piece.~ Contour describes the overall shape of a melody, taking into accountonly the pattern of “up” and “down” (whether a note goes upor down, not the amount by which it goes up or down).~ Timbre is that which distinguishes one instrument from another—say, trumpet from piano—when both are playing the same writtennote. It is a kind of tonal color that is produced in part by overtonesfrom the instrument’s vibrations.~ Loudness is a purely psychological construct that relates (nonlinearlyand in poorly understood ways) to the physical amplitude ofa tone.~ Spatial location is where the sound is coming from.~ Reverberation refers to the perception of how distant the source isfrom us in combination with how large a room or hall the music is

16 This Is Your Brain on Musicin; often referred to as “echo” by laypeople, it is the quality thatdistinguishes the spaciousness of singing in a large concert hallfrom the sound of singing in your shower. It has an underappreciatedrole in communicating emotion and creating an overall pleasingsound.These attributes are separable. Each can be varied without alteringthe others, allowing the scientific study of one at a time, which is why wecan think of them as dimensions. The difference between music and arandom or disordered set of sounds has to do with the way these fundamentalattributes combine, and the relations that form between them.When these basic elements combine and form relationships with one anotherin a meaningful way, they give rise to higher-order concepts suchas meter, key, melody, and harmony.~ Meter is created by our brains by extracting information fromrhythm and loudness cues, and refers to the way in which tonesare grouped with one another across time. A waltz meter organizestones into groups of three, a march into groups of two or four.~ Key has to do with a hierarchy of importance that exists betweentones in a musical piece; this hierarchy does not exist in-the-world,it exists only in our minds, as a function of our experiences with amusical style and musical idioms, and mental schemas that all ofus develop for understanding music.~ Melody is the main theme of a musical piece, the part you singalong with, the succession of tones that are most salient in yourmind. The notion of melody is different across genres. In rock music,there is typically a melody for the verses and a melody for thechorus, and verses are distinguished by a change in lyrics andsometimes by a change in instrumentation. In classical music, themelody is a starting point for the composer to create variations onthat theme, which may be used throughout the entire piece in differentforms.

What Is Music? 17~ Harmony has to do with relationships between the pitches of differenttones, and with tonal contexts that these pitches set up thatultimately lead to expectations for what will come next in a musicalpiece—expectations that a skillful composer can either meetor violate for artistic and expressive purposes. Harmony can meansimply a parallel melody to the primary one (as when two singersharmonize) or it can refer to a chord progression—the clusters ofnotes that form a context and background on which the melodyrests.The idea of primitive elements combining to create art, and of the importanceof relationships between elements, also exists in visual art anddance. The fundamental elements of visual perception include color(which can be decomposed into the three dimensions of hue, saturation,and lightness), brightness, location, texture, and shape. But a painting ismore than these—it is not just a line here and another there, or a spot ofred in one part of the picture and a patch of blue in another. What makesa set of lines and colors into art is the relationship between this line andthat one; the way one color or form echoes another in a different part ofthe canvas. Those dabs of paint and lines become art when form andflow (the way in which your eye is drawn across the canvas) are createdout of lower-level perceptual elements. When they combine harmoniouslythey ultimately give rise to perspective, foreground and background,emotion, and other aesthetic attributes. Similarly, dance is notjust a raging sea of unrelated bodily movements; the relationship ofthose movements to one another is what creates integrity and integrality,a coherence and cohesion that the higher levels of our brain process.And as in visual art, music plays on not just what notes are sounded, butwhich ones are not. Miles Davis famously described his improvisationaltechnique as parallel to the way that Picasso described his use of a canvas:The most critical aspect of the work, both artists said, was not theobjects themselves, but the space between objects. In Miles’s case, hedescribed the most important part of his solos as the empty space be-

18 This Is Your Brain on Musictween notes, the “air” that he placed between one note and the next.Knowing precisely when to hit the next note, and allowing the listenertime to anticipate it, is a hallmark of Davis’s genius. This is particularlyapparent in his album Kind of Blue.To nonmusicians, terms such as diatonic, cadence, or even key andpitch can throw up an unnecessary barrier. Musicians and critics sometimesappear to live behind a veil of technical terms that can sound pretentious.How many times have you read a concert review in thenewspaper and found you have no idea what the reviewer is saying? “Hersustained appoggiatura was flawed by an inability to complete theroulade.” Or, “I can’t believe they modulated to C-sharp minor! Howridiculous!” What we really want to know is whether the music was performedin a way that moved the audience. Whether the singer seemed toinhabit the character she was singing about. You might want the reviewerto compare tonight’s performance to that of a previous night or adifferent ensemble. We’re usually interested in the music, not the technicaldevices that were used. We wouldn’t stand for it if a restaurant reviewerstarted to speculate about the precise temperature at which thechef introduced the lemon juice in a hollandaise sauce, or if a film critictalked about the aperture of the lens that the cinematographer used; weshouldn’t stand for it in music either.Moreover, many of those who study music—even musicologists andscientists—disagree about what is meant by some of these terms. Weemploy the term timbre, for example, to refer to the overall sound ortonal color of an instrument—that indescribable character that distinguishesa trumpet from a clarinet when they’re playing the same writtennote, or what distinguishes your voice from Brad Pitt’s if you’re sayingthe same words. But an inability to agree on a definition has caused thescientific community to take the unusual step of throwing up its handsand defining timbre by what it is not. (The official definition of theAcoustical Society of America is that timbre is everything about a soundthat is not loudness or pitch. So much for scientific precision!)What is pitch? This simple question has generated hundreds of scien-

What Is Music? 19tific articles and thousands of experiments. Pitch is related to the frequencyor rate of vibration of a string, column of air, or other physicalsource. If a string is vibrating so that it moves back and forth sixty timesin one second, we say that it has a frequency of sixty cycles per second.The unit of measurement, cycles per second, is often called Hertz (abbreviatedHz) after Heinrich Hertz, the German theoretical physicistwho was the first to transmit radio waves (a dyed-in-the-wool theoretician,when asked what practical use radio waves might have, he reportedlyshrugged, “None”). If you were to try to mimic the sound of a fireengine siren, your voice would sweep through different pitches, or frequencies(as the tension in your vocal folds changes), some “low” andsome “high.”Keys on the left of the piano keyboard strike longer, thicker stringsthat vibrate at a relatively slow rate. Keys to the right strike shorter, thinnerstrings that vibrate at a higher rate. The vibration of these strings displacesair molecules, and causes them to vibrate at the same rate—withthe same frequency as the string. These vibrating air molecules are whatreach our eardrum, and they cause our eardrum to wiggle in and out atthe same frequency. The only information that our brains get about thepitch of sound comes from that wiggling in and out of our eardrum; ourinner ear and our brain have to analyze the motion of the eardrum in orderto figure out what vibrations out-there-in-the-world caused theeardrum to move that way.By convention, when we press keys nearer to the left of the keyboard,we say that they are “low” pitch sounds, and ones near the right side ofthe keyboard are “high” pitch. That is, what we call “low” are thosesounds that vibrate slowly, and are closer (in vibration frequency) to thesound of a large dog barking. What we call “high” are those sounds thatvibrate rapidly, and are closer to what a small yip-yip dog might make.But even these terms high and low are culturally relative—the Greekstalked about sounds in the opposite way because the stringed instrumentsthey built tended to be oriented vertically. Shorter strings or pipeorgan tubes had their tops closer to the ground, so these were calledthe “low” notes (as in “low to the ground,”) and the longer strings and

20 This Is Your Brain on Musictubes—reaching up toward Zeus and Apollo—were called the “high”notes. Low and high—just like left and right—are effectively arbitraryterms that ultimately have to be memorized. Some writers have arguedthat “high” and “low” are intuitive labels, noting that what we call highpitchedsounds come from birds (who are high up in trees or in the sky)and what we call low-pitched sounds often come from large, close-tothe-groundmammals such as bears or the low sounds of an earthquake.But this is not convincing, since low sounds also come from up high(think of thunder) and high sounds can come from down low (cricketsand squirrels, leaves being crushed underfoot).As a first definition of pitch, let’s say it is that quality that primarilydistinguishes the sound that is associated with pressing one piano keyversus another.Pressing a piano key causes a hammer to strike one or more stringsinside the piano. Striking a string displaces it, stretching it a bit, and itsinherent resiliency causes it to return toward its original position. But itovershoots that original position, going too far in the opposite direction,and then attempts to return to its original position again, overshooting itagain, and in this way it oscillates back and forth. Each oscillation coversless distance, and, in time, the string stops moving altogether. This iswhy the sound you hear when you press a piano key gets softer until ittrails off into nothing. The distance that the string covers with each oscillationback and forth is translated by our brains into loudness; the rateat which it oscillates is translated into pitch. The farther the string travels,the louder the sound seems to us; when it is barely traveling at all,the sound seems soft. Although it might seem counterintuitive, the distancetraveled and the rate of oscillation are independent. A string canvibrate very quickly and traverse either a great distance or a small one.The distance it traverses is related to how hard we hit it—this correspondsto our intuition that hitting something harder makes a loudersound. The rate at which the string vibrates is principally affected by itssize and how tightly strung it is, not by how hard it was struck.It might seem as though we should simply say that pitch is the sameas frequency; that is, the frequency of vibration of air molecules. This is

What Is Music? 21almost true. Mapping the physical world onto the mental world is seldomso straightforward. However, for most musical sounds, pitch andfrequency are closely related.The word pitch refers to the mental representation an organism hasof the fundamental frequency of a sound. That is, pitch is a purely psychologicalphenomenon related to the frequency of vibrating air molecules.By “psychological,” I mean that it is entirely in our heads, not inthe world-out-there; it is the end product of a chain of mental events thatgives rise to an entirely subjective, internal mental representation orquality. Sound waves—molecules of air vibrating at various frequencies—donot themselves have pitch. Their motion and oscillations canbe measured, but it takes a human (or animal) brain to map them to thatinternal quality we call pitch.We perceive color in a similar way, and it was Isaac Newton who firstrealized this. (Newton, of course, is known as the discoverer of the theoryof gravity, and the inventor, along with Leibniz, of calculus. LikeEinstein, Newton was a very poor student, and his teachers often complainedof his inattentiveness. Ultimately, Newton was kicked out ofschool.)Newton was the first to point out that light is colorless, and that consequentlycolor has to occur inside our brains. He wrote, “The wavesthemselves are not colored.” Since his time, we have learned that lightwaves are characterized by different frequencies of oscillation, andwhen they impinge on the retina of an observer, they set off a chain ofneurochemical events, the end product of which is an internal mentalimage that we call color. The essential point here is: What we perceive ascolor is not made up of color. Although an apple may appear red, itsatoms are not themselves red. And similarly, as the philosopher DanielDennett points out, heat is not made up of tiny hot things.A bowl of pudding only has taste when I put it in my mouth—when itis in contact with my tongue. It doesn’t have taste or flavor sitting in myfridge, only the potential. Similarly, the walls in my kitchen are not“white” when I leave the room. They still have paint on them, of course,but color only occurs when they interact with my eyes.

22 This Is Your Brain on MusicSound waves impinge on the eardrums and pinnae (the fleshy parts ofyour ear), setting off a chain of mechanical and neurochemical events,the end product of which is an internal mental image we call pitch. If atree falls in a forest and no one is there to hear it, does it make a sound?(The question was first posed by the Irish philosopher George Berkeley.)Simply, no—sound is a mental image created by the brain in response tovibrating molecules. Similarly, there can be no pitch without a human oranimal present. A suitable measuring device can register the frequencymade by the tree falling, but truly it is not pitch unless and until it isheard.No animal can hear a pitch for every frequency that exists, just as thecolors that we actually see are a small portion of the entire electromagneticspectrum. Sound can theoretically be heard for vibrations from justover 0 cycles per second up to 100,000 cycles per second or more, buteach animal hears only a subset of the possible sounds. Humans who arenot suffering from any kind of hearing loss can usually hear sounds from20 Hz to 20,000 Hz. The pitches at the low end sound like an indistinctrumble or shaking—this is the sound we hear when a truck goes by outsidethe window (its engine is creating sound around 20 Hz) or when atricked-out car with a fancy sound system has the subwoofers crankedup really loud. Some frequencies—those below 20 Hz—are inaudible tohumans because the physiological properties of our ears aren’t sensitiveto them.The range of human hearing is generally 20 Hz to 20,000 Hz, but thisdoesn’t mean that the range of human pitch perception is the same; althoughwe can hear sounds in this entire range, they don’t all sound musical;that is, we can’t unambiguously assign a pitch to the entire range.By analogy, colors at the infrared and ultraviolet ends of the spectrumlack definition compared to the colors closer to the middle. The figure onpage 23 shows the range of musical instruments, and the frequency associatedwith them. The sound of the average male speaking voice isaround 110 Hz, and the average female speaking voice is around 220 Hz.The hum of fluorescent lights or from faulty wiring is 60 Hz (in NorthAmerica; in Europe and countries with a different voltage/current stan-

What Is Music? 23PiccoloViolinWomanʼs voiceTrumpetManʼs voiceTubaA-440Middle C3729.33322.42960.02489.02217.51864.71661.21480.01244.51108.7932.33830.61739.99622.25554.37466.16415.30369.99311.13277.18233.08207.65185.00155.56138.59116.54103.8392.49977.78269.26958.27051.91346.24938.89134.64829.1354186.03951.13520.03136.02793.02637.02349.32093.01975.51760.01568.01396.91318.51174.71046.5987.77880.00783.99698.46659.26587.33523.55493.88440.00392.00349.23329.63293.66261.63246.94220.00196.00174.61164.81146.83130.81123.47110.0097.99987.30782.40773.41665.40661.73555.00048.99943.65441.20336.70832.70330.86327.500B C D E F G A B C D E F G A B C D E F G A B C D E F G A B C D E F G A B C D E F G A B C D E F G A B CA

24 This Is Your Brain on Musicdard, it can be 50 Hz). The sound that a singer hits when she causes aglass to break might be 1000 Hz. The glass breaks because it, like allphysical objects, has a natural and inherent vibration frequency. You canhear this by flicking your finger against its sides or, if it’s crystal, by runningyour wet finger around the rim of the glass in a circular motion.When the singer hits just the right frequency—the resonant frequency ofthe glass—it causes the molecules of the glass to vibrate at their naturalrate, and they vibrate themselves apart.A standard piano has eighty-eight keys. Very rarely, pianos can have afew extra ones at the bottom and electronic pianos, organs, and synthesizerscan have as few as twelve or twenty-four keys, but these are specialcases. The lowest note on a standard piano vibrates with a frequencyof 27.5 Hz. Interestingly, this is about the same rate of motion that constitutesan important threshold in visual perception. A sequence of stillphotographs—slides—displayed at or about this rate of presentationwill give the illusion of motion. “Motion pictures” are a sequence of stillimages alternating with pieces of black film presented at a rate (oneforty-eighth of a second) that exceeds the temporal resolving propertiesof the human visual system. We perceive smooth, continuous motionwhen in fact there is no such thing actually being shown to us. Whenmolecules vibrate at around this speed we hear something that soundslike a continuous tone. If you put playing cards in the spokes of your bicyclewheel when you were a kid, you demonstrated to yourself a relatedprinciple: At slow speeds, you simply hear the click-click-click of thecard hitting the spokes. But above a certain speed, the clicks run togetherand create a buzz, a tone you can actually hum along with; a pitch.When this lowest note on the piano plays, and vibrates at 27.5 Hz, tomost people it lacks the distinct pitch of sounds toward the middle of thekeyboard. At the lowest and the highest ends of the piano keyboard, thenotes sound fuzzy to many people with respect to their pitch. Composersknow this, and they either use these notes or avoid them depending onwhat they are trying to accomplish compositionally and emotionally.Sounds with frequencies above the highest note on the piano keyboard,around 6000 Hz and more, sound like a high-pitched whistling to most

What Is Music? 25people. Above 20,000 Hz most humans don’t hear a thing, and by the ageof sixty, most adults can’t hear much above 15,000 Hz or so due to a stiffeningof the hair cells in the inner ear. So when we talk about the rangeof musical notes, or that restricted part of the piano keyboard that conveysthe strongest sense of pitch, we are talking about roughly threequarters of the notes on the piano keyboard, between about 55 Hz and2000 Hz.Pitch is one of the primary means by which musical emotion is conveyed.Mood, excitement, calm, romance, and danger are signaled by anumber of factors, but pitch is among the most decisive. A single highnote can convey excitement, a single low note sadness. When notes arestrung together, we get more powerful and more nuanced musical statements.Melodies are defined by the pattern or relation of successivepitches across time; most people have no trouble recognizing a melodythat is played in a higher or lower key than they’ve heard it in before. Infact, many melodies do not have a “correct” starting pitch, they just floatfreely in space, starting anywhere. “Happy Birthday” is an example ofthis. One way to think about a melody, then, is as an abstract prototypethat is derived from specific combinations of key, tempo, instrumentation,and so on. A melody is an auditory object that maintains its identityin spite of transformations, just as a chair maintains its identity whenyou move it to the other side of the room, turn it upside down, or paint itred. So, for example, if you hear a song played louder than you are accustomedto, you still identify it as the same song. The same holds forchanges in the absolute pitch values of the song, which can be changedso long as the relative distances between them remain the same.The notion of relative pitch values is seen readily in the way that wespeak. When you ask someone a question, your voice naturally rises inintonation at the end of the sentence, signaling that you are asking. Butyou don’t try to make the rise in your voice match a specific pitch. It isenough that you end the sentence somewhat higher in pitch than you beganit. This is a convention in English (though not in all languages—wehave to learn it), and is known in linguistics as a prosodic cue. There aresimilar conventions for music written in the Western tradition. Certain

26 This Is Your Brain on Musicsequences of pitches evoke calm, others, excitement. The brain basis forthis is primarily based on learning, just as we learn that a rising intonationindicates a question. All of us have the innate capacity to learn thelinguistic and musical distinctions of whatever culture we are born into,and experience with the music of that culture shapes our neural pathwaysso that we ultimately internalize a set of rules common to that musicaltradition.Different instruments use different parts of the range of availablepitches. The piano has the largest range of any instrument, as you cansee from the previous illustration. The other instruments each use a subsetof the available pitches, and this influences the ways that instrumentsare used to communicate emotion. The piccolo, with its high-pitched,shrill, and birdlike sound, tends to evoke flighty, happy moods regardlessof the notes it’s playing. Because of this, composers tend to use the piccolofor happy music, or rousing music, as in a Sousa march. Similarly, inPeter and the Wolf, Prokofiev uses the flute to represent the bird, andthe French horn to indicate the wolf. The characters’ individuality inPeter and the Wolf is expressed in the timbres of different instrumentsand each has a leitmotiv—an associated melodic phrase or figure thataccompanies the reappearance of an idea, person, or situation. (This isespecially true of Wagnerian music drama.) A composer who picks socalledsad pitch sequences would only give these to the piccolo if hewere trying to be ironic. The lumbering, deep sounds of the tuba ordouble bass are often used to evoke solemnity, gravity, or weight.How many unique pitches are there? Because pitch comes from acontinuum—the vibration frequencies of molecules—there are technicallyan infinite number of pitches: For every pair of frequencies youmention, I could always come up with one between them, and a theoreticallydifferent pitch would exist. But not every change in frequencygives rise to a noticeable difference in pitch, just as adding a grain ofsand to your backpack will not change the weight perceptibly. So not allfrequency changes are musically useful. People differ in their ability todetect small changes in frequency; training can help, but generallyspeaking, most cultures don’t use distances much smaller than a semi-

What Is Music? 27tone as the basis for their music, and most people can’t reliably detectchanges smaller than about one tenth of a semitone.The ability to detect differences in pitch is based on physiology, andvaries from one animal to another. The basilar membrane of the humaninner ear contains hair cells that are frequency selective, firing only inresponse to a certain band of frequencies. These are stretched outacross the membrane from low frequencies to high; low-frequencysounds excite hair cells on one end of the basilar membrane, mediumfrequency sounds excite the hair cells in the middle, and high-frequencysounds excite them at the other end. We can think of the membrane ascontaining a map of different pitches very much like a piano keyboardsuperimposed on it. Because the different tones are spread out acrossthe surface topography of the membrane, this is called a tonotopic map.After sounds enter the ear, they pass by the basilar membrane, wherecertain hair cells fire, depending on the frequency of the sounds. Themembrane acts like a motion-detector lamp you might have in your garden;activity in a certain part of the membrane causes it to send an electricalsignal on up to the auditory cortex. The auditory cortex also has atonotopic map, with low to high tones stretched out across the corticalsurface. In this sense, the brain contains a “map” of different pitches, anddifferent areas of the brain respond to different pitches. Pitch is so importantthat the brain represents it directly; unlike almost any other musicalattribute, we could place electrodes in the brain and be able todetermine what pitches were being played to a person just by looking atthe brain activity. And although music is based on pitch relations ratherthan absolute pitch values, it is, paradoxically, these absolute pitch valuesthat the brain is paying attention to throughout its different stages ofprocessing.A scale is just a subset of the theoretically infinite number of pitches, andevery culture selects these based on historical tradition or somewhat arbitrarily.The specific pitches chosen are then anointed as being part ofthat musical system. These are the letters that you see in the figureabove. The names “A,” “B,” “C,” and so on are arbitrary labels that we as-

28 This Is Your Brain on Musicsociate with particular frequencies. In Western music—music of the Europeantradition—these pitches are the only “legal” pitches; most instrumentsare designed to play these pitches and not others. (Instrumentslike the trombone and cello are an exception, because they can slide betweennotes; trombonists, cellists, violinists, etc., spend a lot of timelearning how to hear and produce the precise frequencies required toplay each of the legal notes.) Sounds in between are considered mistakes(“out of tune”) unless they’re used for expressive intonation (intentionallyplaying something out of tune, briefly, to add emotionaltension) or in passing from one legal tone to another.Tuning refers to the precise relationship between the frequency of atone being played and a standard, or between two or more tones beingplayed together. Orchestral musicians “tuning up” before a performanceare synchronizing their instruments (which naturally drift in their tuningas the wood, metal, strings, and other materials expand and contractwith changes in temperature and humidity) to a standard frequency, oroccasionally not to a standard but to each other. Expert musicians oftenalter the frequency of tones while they’re playing for expressive purposes(except, of course, on fixed-pitch instruments such as keyboardsand xylophones); sounding a note slightly lower or higher than its nominalvalue can impart emotion when done skillfully. Expert musiciansplaying together in ensembles will also alter the pitch of tones they playto bring them more in tune with the tones being played by the other musicians,should one or more musicians drift away from standard tuningduring the performance.The note names in Western music run from A to G, or, in an alternativesystem, as Do - re - mi - fa - sol - la - ti - do (the alternate system isused as lyrics to the Rodgers and Hammerstein song “Do-Re-Mi” fromThe Sound of Music: “Do, a deer, a female deer, Re, a drop of goldensun . . .”). As frequencies get higher, so do the letter names; B has ahigher frequency than A (and hence a higher pitch) and C has a higherfrequency than either A or B. After G, the note names start all over againat A. Notes with the same name have frequencies that are multiples ofeach other. One of the several notes we call A has a frequency of 55 Hz

What Is Music? 29and all other notes called A have frequencies that are two, three, four,five (or a half) times this frequency.Here is a fundamental quality of music. Note names repeat because ofa perceptual phenomenon that corresponds to the doubling and halvingof frequencies. When we double or halve a frequency, we end up with anote that sounds remarkably similar to the one we started out with. Thisrelationship, a frequency ratio of 2:1 or 1:2, is called the octave. It is soimportant that, in spite of the large differences that exist between musicalcultures—between Indian, Balinese, European, Middle Eastern, Chinese,and so on—every culture we know of has the octave as the basisfor its music, even if it has little else in common with other musical traditions.This phenomenon leads to the notion of circularity in pitch perception,and is similar to circularity in colors. Although red and violetfall at opposite ends of the continuum of visible frequencies of electromagneticenergy, we see them as perceptually similar. The same is truein music, and music is often described as having two dimensions, onethat accounts for tones going up in frequency (and sounding higher andhigher) and another that accounts for the perceptual sense that we’vecome back home again each time we double a tone’s frequency.When men and women speak in unison, their voices are normally anoctave apart, even if they try to speak the exact same pitches. Childrengenerally speak an octave or two higher than adults. The first two notesof the Harold Arlen melody “Somewhere Over the Rainbow” (from themovie The Wizard of Oz) make an octave. In “Hot Fun in the Summertime”by Sly and the Family Stone, Sly and his backup singers are singingin octaves during the first line of the verse “End of the spring and hereshe comes back.” As we increase frequencies by playing the successivenotes on an instrument, there is a very strong perceptual sense thatwhen we reach a doubling of frequency, we have come “home” again. Theoctave is so basic that even some animal species—monkeys and cats, forexample—show octave equivalence, the ability to treat as similar, theway that humans do, tones separated by this amount.An interval is the distance between two tones. The octave in Westernmusic is subdivided into twelve (logarithmically) equally spaced tones.

30 This Is Your Brain on MusicThe intervallic distance between A and B (or between “do” and “re”) iscalled a whole step or a tone. (This latter term is confusing, since we callany musical sound a tone; I’ll use the term whole step to avoid ambiguity).The smallest division in our Western scale system cuts a whole step perceptuallyin half: This is the semitone, which is one twelfth of an octave.Intervals are the basis of melody, much more so than the actualpitches of notes; melody processing is relational, not absolute, meaningthat we define a melody by its intervals, not the actual notes used to createthem. Four semitones always create the interval known as a majorthird regardless of whether the first note is an A or a G# or any othernote. Here is a table of the intervals as they’re known in our (Western)musical system:The table could continue on: Thirteen semitones is a minor ninth,Distance in semitonesInterval name0 unison1 minor second2 major second3 minor third4 major third5 perfect fourth6 augmented fourth, diminished fifth, or tritone7 perfect fifth8 minor sixth9 major sixth10 minor seventh11 major seventh12 octave

What Is Music? 31fourteen semitones is a major ninth, etc., but these names are typicallyused only in more advanced discussions. The intervals of the perfectfourth and perfect fifth are so called because they sound particularlypleasing to many people, and since the ancient Greeks, this particularfeature of the scale is at the heart of all music. (There is no “imperfectfifth,” this is just the name we give the interval.) Ignore the perfect fourthand fifth or use them in every phrase, they have been the backbone ofmusic for at least five thousand years.Although the areas of the brain that respond to individual pitcheshave been mapped, we have not yet been able to find the neurological basisfor the encoding of pitch relations; we know which part of the cortexis involved in listening to the notes C and E, for example, and for F andA, but we do not know how or why both intervals are perceived as a majorthird, or the neural circuits that create this perceptual equivalency.These relations must be extracted by computational processes in thebrain that remain poorly understood.If there are twelve named notes within an octave, why are there onlyseven letters (or do-re-mi syllables)? After centuries of being forced toeat in the servants’ quarters and to use the back entrance of the castle,this may just be an invention by musicians to make nonmusicians feel inadequate.The additional five notes have compound names, such as E♭pronounced “E-flat”) and F# (pronounced “F-sharp”). There is no reasonfor the system to be so complicated, but it is what we’re stuck with.The system is a bit clearer looking at the piano keyboard. A piano haswhite keys and black keys spaced out in an uneven arrangement—sometimestwo white keys are adjacent, sometimes they have a black keybetween them. Whether the keys are white or black, the perceptual distancefrom one adjacent key to the next always makes a semitone, and adistance of two keys is always a whole step. This applies to many Westerninstruments; the distance between one fret on a guitar and the nextis also a semitone, and pressing or lifting adjacent keys on woodwind instruments(such as the clarinet or oboe) typically changes the pitch by asemitone.

32 This Is Your Brain on MusicThe white keys are named A, B, C, D, E, F, and G. The notes between—theblack keys—are the ones with compound names. The notebetween A and B is called either A-sharp or B-flat, and in all but formalmusic theoretic discussions, the two terms are interchangeable. (In fact,this note could also be referred to as C double-flat, and similarly, A couldbe called G double-sharp, but this is an even more theoretical usage.)Sharp means high, and flat means low. B-flat is the note one semitonelower than B; A-sharp is the note one semitone higher than A. In the paralleldo-re-mi system, unique syllables mark these other tones: di and raindicate the tone between do and re, for example.The notes with compound names are not in any way second-class musicalcitizens. They are just as important, and in some songs and somescales they are used exclusively. For example, the main accompanimentto “Superstition” by Stevie Wonder is played on only the black keys ofthe keyboard. The twelve tones taken together, plus their repeatingcousins one or more octaves apart, are the basic building blocks formelody, for all the songs in our culture. Every song you know, from“Deck the Halls” to “Hotel California,” from “Ba Ba Black Sheep” to thetheme from Sex and the City, is made up from a combination of thesetwelve tones and their octaves.To add to the confusion, musicians also use the terms sharp and flatto indicate if someone is playing out of tune; if the musician plays thetone a bit too high (but not so high as to make the next note in the scale)we say that the tone being played is sharp, and if the musician plays thetone too low we say that the tone is flat. Of course, a musician can beonly slightly off and nobody would notice. But when the musician is offby a relatively large amount—say one quarter to one half the distance betweenthe note she was trying to play and the next one—most of us canusually detect this and it sounds off. This is especially apparent whenthere is more than one instrument playing, and the out-of-tune tone weare hearing clashes with in-tune tones being played simultaneously byother musicians.The names of pitches are associated with particular frequency values.Our current system is called A440 because the note we call A that is in

What Is Music? 33the middle of the piano keyboard has been fixed to have a frequency of440 Hz. This is entirely arbitrary. We could fix A at any frequency, such as439, 444, 424, or 314.159; different standards were used in the time ofMozart than today. Some people claim that the precise frequencies affectthe overall sound of a musical piece and the sound of instruments. LedZeppelin often tuned their instruments away from the modern A440 standardto give their music an uncommon sound, and perhaps to link it withthe European children’s folk songs that inspired many of their compositions.Many purists insist on hearing baroque music on period instruments,both because the instruments have a different sound and becausethey are designed to play the music in its original tuning standard, somethingthat purists deem important.We can fix pitches anywhere we want because what defines music isa set of pitch relations. The specific frequencies for notes may be arbitrary,but the distance from one frequency to the next—and hence fromone note to the next in our musical system—isn’t at all arbitrary. Eachnote in our musical system is equally spaced to our ears (but not necessarilyto the ears of other species). Although there is not an equal changein cycles per second (Hz) as we climb from one note to the next, the distancebetween each note and the next sounds equal. How can this be?The frequency of each note in our system is approximately 6 percentmore than the one before it. Our auditory system is sensitive both to relativechanges and to proportional changes in sound. Thus, each increasein frequency of 6 percent gives us the impression that we have increasedpitch by the same amount as we did last time.The idea of proportional change is intuitive if you think aboutweights. If you’re at a gym and you want to increase your weight liftingof the barbells from 5 pounds to 50 pounds, adding 5 pounds each weekis not going to change the amount of weight you’re lifting in an equalway. After a week of lifting 5 pounds, when you move to 10 you are doublingthe weight; the next week when you move to 15 you are adding 1.5times as much weight as you had before. An equal spacing—to give yourmuscles a similar increase of weight each week—would be to add a constantpercentage of the previous week’s weight each time you increase.

34 This Is Your Brain on MusicFor example, you might decide to add 50 percent each week, and so youwould then go from 5 pounds to 7.5, then to 11.25, then to 16.83, and soon. The auditory system works the same way, and that is why our scaleis based on a proportion: Every tone is 6 percent higher than the previousone, and when we increase each step by 6 percent twelve times, weend up having doubled our original frequency (the actual proportion isthe twelfth root of two = 1.059463 ...).The twelve notes in our musical system are called the chromaticscale. Any scale is simply a set of musical pitches that have been chosento be distinguishable from each other and to be used as the basis for constructingmelodies.In Western music we rarely use all the notes of chromatic scale incomposition; instead, we use a subset of seven (or less often, five) ofthose twelve tones. Each of these subsets is itself a scale, and the type ofscale we use has a large impact on the overall sound of a melody, and itsemotional qualities. The most common subset of seven tones used inWestern music is called the major scale, or Ionian mode (reflecting itsancient Greek origins). Like all scales, it can start on any of the twelvenotes, and what defines the major scale is the specific pattern or distancerelationship between each note and its successive note. In any majorscale, the pattern of intervals—pitch distances between successive keys—is: whole step, whole step, half step, whole step, whole step, whole step,half step.Starting on C, the major scale notes are C - D - E - F - G - A - B - C, allwhite notes on the piano keyboard. All other major scales require one ormore black notes to maintain the required whole step/half step pattern.The starting pitch is also called the root of the scale.The particular placement of the two half steps in the sequence of themajor is crucial; it is not only what defines the major scale and distinguishesit from other scales, but it is an important ingredient in musicalexpectations. Experiments have shown that young children, as well asadults, are better able to learn and memorize melodies that are drawnfrom scales that contain unequal distances such as this. The presence ofthe two half steps, and their particular positions, orient the experienced,

What Is Music? 35acculturated listener to where we are in the scale. We are all experts inknowing, when we hear a B in the key of C—that is, when the tones arebeing drawn primary from the C major scale—that it is the seventh note(or “degree”) of that scale, and that it is only a half step below the root,even though most of us can’t name the notes, and may not even knowwhat a root or a scale degree is. We have assimilated the structure of thisand other scales through a lifetime of listening and passive (rather thantheoretically driven) exposure to the music. This knowledge is not innate,but is gained through experience. By a similar token, we don’t needto know anything about cosmology to have learned that the sun comesup every morning and goes down at night—we have learned this sequenceof events through largely passive exposure.Different patterns of whole steps and half steps give rise to alternativescales, the most common of which (in our culture) is the minorscale. There is one minor scale that, like the C major scale, uses only thewhite notes of the piano keyboard: the A minor scale. The pitches forthat scale are A - B - C - D - E - F - G - A. (Because it uses the same set ofpitches, but in a different order, A minor is said to be the “relative minorof the C major scale.”) The pattern of whole steps and half steps isdifferent from that of the major scale: whole–half–whole–whole–half–whole–whole. Notice that the placement of the half steps is very differentthan in the major scale; in the major scale, there is a half step justbefore the root that “leads” to the root, and another half step just beforethe fourth scale degree. In the minor scale, the half steps are before thethird scale degree and before the sixth. There is still a momentum whenwe’re in this scale to return to the root, but the chords that create thismomentum have a clearly different sound and emotional trajectory.Now you might well ask: If these two scales use exactly the same setof pitches, how do I know which one I’m in? If a musician is playing thewhite keys, how do I know if he is playing the A minor scale or the C majorscale? The answer is that—entirely without our conscious awareness—ourbrains are keeping track of how many times particular notesare sounded, where they appear in terms of strong versus weak beats,and how long they last. A computational process in the brain makes an

36 This Is Your Brain on Musicinference about the key we’re in based on these properties. This is anotherexample of something that most of us can do even without musicaltraining, and without what psychologists call declarative knowledge—the ability to talk about it; but in spite of our lack of formal musical education,we know what the composer intended to establish as the tonalcenter, or key, of the piece, and we recognize when he brings us backhome to the tonic, or when he fails to do so. The simplest way to establisha key, then, is to play the root of the key many times, play it loud, andplay it long. And even if a composer thinks he is writing in C major, if hehas the musicians play the note A over and over again, play it loud andplay it long; if the composer starts the piece on an A and ends the pieceon an A, and moreover, if he avoids the use of C, the audience, musicians,and music theorists are most probably going to decide that thepiece is in A minor, even if this was not his intent. In musical keys as inspeeding tickets, it is the observed action, not the intention, that counts.For reasons that are largely cultural, we tend to associate majorscales with happy or triumphant emotions, and minor scales with sad ordefeated emotions. Some studies have suggested that the associationsmight be innate, but the fact that these are not culturally universal indicatesthat, at the very least, any innate tendency can be overcome byexposure to specific cultural associations. Western music theory recognizesthree minor scales and each has a slightly different flavor. Bluesmusic generally uses a five note (pentatonic) scale that is a subset of theminor scale, and Chinese music uses a different pentatonic scale. WhenTchaikovsky wants us to think of Arab or Chinese culture in the Nutcrackerballet, he chooses scales that are typical to their music, andwithin just a few notes we are transported to the Orient. When Billie Holidaywants to make a standard tune bluesy, she invokes the blues scaleand sings notes from a scale that we are not accustomed to hearing instandard classical music.Composers know these associations and use them intentionally. Ourbrains know them, too, through a lifetime of exposure to musical idioms,patterns, scales, lyrics, and the associations between them. Each timewe hear a musical pattern that is new to our ears, our brains try to make

What Is Music? 37an association through whatever visual, auditory and other sensory cuesaccompany it; we try to contextualize the new sounds, and eventually,we create these memory links between a particular set of notes and aparticular place, time, or set of events. No one who has seen Hitchcock’sPsycho can hear Bernard Hermann’s screeching violins without thinkingof the shower scene; anyone who has ever seen a Warner Bros. “MerrieMelody” cartoon will think of a character sneakily climbing stairs wheneverthey hear plucked violins playing an ascending major scale. Theassociations are so powerful—and the scales distinguishable enough—that only a few notes are needed: The first three notes of David Bowie’s“China Girl” or Mussorgsky’s “Great Gate of Kiev” (from Pictures at anExhibition) instantly convey a rich and foreign (to us) musical context.Nearly all this variation in context and sound comes from differentways of dividing up the octave and, in virtually every case we know of,dividing it up into no more than twelve tones. Although it has beenclaimed that Indian and Arab-Persian music use “microtuning”—scaleswith intervals much smaller than a semitone—close analysis reveals thattheir scales also rely on twelve or fewer tones and the others are simplyexpressive variations, glissandos (continuous glides from one tone toanother), and momentary passing tones, similar to the American bluestradition of sliding into a note for emotional purposes.In any scale, a hierarchy of importance exists among scale tones;some are more stable, structurally significant, or final sounding than others,causing us to feel varying amounts of tension and resolution. In themajor scale, the most stable tone is the first degree, also called the tonic.In other words, all other tones in the scale seem to point toward thetonic, but they point with varying momentum. The tone that points moststrongly to the tonic is the seventh scale degree, B in a C major scale.The tone that points least strongly to the tonic is the fifth scale degree, Gin the C major scale, and it points least strongly because it is perceivedas relatively stable; this is just another way of saying that we don’t feeluneasy—unresolved—if a song ends on the fifth scale degree. Music theoryspecifies this tonal hierarchy. Carol Krumhansl and her colleaguesperformed a series of studies establishing that ordinary listeners have

38 This Is Your Brain on Musicincorporated the principles of this hierarchy in their brains, through passiveexposure to music and cultural norms. By asking people to rate howwell different tones seemed to fit with a scale she would play them, sherecovered from their subjective judgments the theoretical hierarchy.A chord is simply a group of three or more notes played at the sametime. They are generally drawn from one of the commonly used scales,and the three notes are chosen so that they convey information aboutthe scale they were taken from. A typical chord is built by playing thefirst, third, and fifth notes of a scale together. Because the sequence ofwhole steps and half steps is different for minor and major scales, the intervalsizes are different for chords taken in this way from the two differentscales. If we build a chord starting on C and use the tones from theC major scale, we use C, E, and G. If instead we use the C minor scale,the first, third, and fifth notes are C, E-flat, and G. This difference in thethird degree, between E and E-flat, turns the chord itself from a majorchord into a minor chord. All of us, even without musical training, cantell the difference between these two even if we don’t have the terminologyto name them; we hear the major chord as sounding happy and theminor chord as sounding sad, or reflective, or even exotic. The most basicrock and country music songs use only major chords: “Johnny B.Goode,” “Blowin’ in the Wind,” “Honky Tonk Women,” and “MammasDon’t Let Your Babies Grow Up to Be Cowboys,” for example.Minor chords add complexity; in “Light My Fire” by the Doors, theverses are played in minor chords (“You know that it would be untrue. . .”) and then the chorus is played in major chords (“Come on baby,light my fire”). In “Jolene,” Dolly Parton mixes minor and major chordsto give a melancholy sound. Pink Floyd’s “Sheep” (from the album Animals)uses only minor chords.Like single notes in the scale, chords also fall along a hierarchy of stability,depending on context. Certain chord progressions are part ofevery musical tradition, and even by the age of five, most children haveinternalized rules about what chord progressions are legal, or typical oftheir culture’s music; they can readily detect deviations from the standardsequences just as easily as we can detect when an English sentence

What Is Music? 39is malformed, such as this one: “The pizza was too hot to sleep.” Forbrains to accomplish this, networks of neurons must form abstract representationsof musical structure, and musical rules, something that theydo automatically and without our conscious awareness. Our brains aremaximally receptive—almost spongelike—when we’re young, hungrilysoaking up any and all sounds they can and incorporating them into thevery structure of our neural wiring. As we age, these neural circuits aresomewhat less pliable, and so it becomes more difficult to incorporate,at a deep neural level, new musical systems, or even new linguisticsystems.Now the story about pitch becomes a bit more complicated, and it’s allthe fault of physics. But this complication gives rise to the rich spectrumof sounds we hear in different instruments. All natural objects in theworld have several modes of vibration. A piano string actually vibrates atseveral different rates at once. The same thing is true of bells that we hitwith a hammer, drums that we hit with our hands, or flutes that we blowair into: The air molecules vibrate at several rates simultaneously, notjust a single rate.An analogy is the several types of motion of the earth that are simultaneouslyoccurring. We know that the earth spins on its axis once everytwenty-four hours, that it travels around the sun once every 365.25 days,and that the entire solar system is spinning along with the Milky Waygalaxy. Several types of motion, all occurring at once. Another analogy isthe many kinds of vibration that we often feel when riding a train. Imaginethat you’re sitting on a train in an outdoor station, with the engine off.It’s windy, and you feel the car rock back and forth just a little bit. It doesso with a regularity that you can time with your handy stopwatch, andyou feel the train moving back and forth about twice a second. Next, theengineer starts the engine, and you feel a different kind of vibrationthrough your seat (due to the oscillations of the motor—pistons andcrankshafts turning around at a certain speed). When the train startsmoving, you experience a third sensation, the bump the wheels makeevery time they go over a track joint. Altogether, you will feel several dif-

40 This Is Your Brain on Musicferent kinds of vibrations, all of them likely to be at different rates, or frequencies.When the train is moving, you are no doubt aware that there isvibration. But it is very difficult, if not impossible, for you to determinehow many vibrations there are and what their rates are. Using specializedmeasuring instruments, however, one might be able to figure this out.When a sound is generated on a piano, flute, or any other instrument—includingpercussion instruments like drums and cowbells—itproduces many modes of vibration occurring simultaneously. When youlisten to a single note played on an instrument, you’re actually hearingmany, many pitches at once, not a single pitch. Most of us are not awareof this consciously, although some people can train themselves to hearthis. The one with the slowest vibration rate—the one lowest in pitch—is referred to as the fundamental frequency, and the others are collectivelycalled overtones.To recap, it is a property of objects in the world that they generally vibrateat several different frequencies at once. Surprisingly, these otherfrequencies are often mathematically related to each other in a very simpleway: as integer multiples of one another. So if you pluck a string andits slowest vibration frequency is one hundred times per second, theother vibration frequencies will be 2 x 100 (200 Hz), 3 x 100 Hz (300 Hz),etc. If you blow into a flute or recorder and cause vibrations at 310 Hz,additional vibrations will be occurring at twice, three times, four times,etc., this rate: 620 Hz, 930 Hz, 1240 Hz, etc. When an instrument createsenergy at frequencies that are integer multiples such as this, we say thatthe sound is harmonic, and we refer to the pattern of energy at differentfrequencies as the overtone series. There is evidence that the brain respondsto such harmonic sounds with synchronous neural firings—theneurons in auditory cortex responding to each of the components of thesound synchronize their firing rates with one another, creating a neuralbasis for the coherence of these sounds.The brain is so attuned to the overtone series that if we encounter asound that has all of the components except the fundamental, the brainfills it in for us in a phenomenon called restoration of the missing fundamental.A sound composed of energy at 100 Hz, 200 Hz, 300 Hz, 400

What Is Music? 41Hz, and 500 Hz is perceived as having a pitch of 100 Hz, its fundamentalfrequency. But if we artificially create a sound with energy at 200 Hz, 300Hz, 400 Hz, and 500 Hz (leaving off the fundamental), we still perceive itas having a pitch of 100 Hz. We don’t perceive it as having a pitch of 200Hz, because our brain “knows” that a normal, harmonic sound with apitch of 200 Hz would have an overtone series of 200 Hz, 400 Hz, 600 Hz,800 Hz, etc. We can also fool the brain by playing sequences that deviatefrom the overtone series such as this: 100 Hz, 210 Hz, 302 Hz, 405 Hz, etc.In cases like these, the perceived pitch shifts away from 100 Hz in a compromisebetween what is presented and what a normal harmonic serieswould imply.When I was in graduate school, my advisor, Mike Posner, told meabout the work of a graduate student in biology, Petr Janata. Although hehadn’t been raised in San Francisco like me, Petr had long bushy hairthat he wore in a ponytail, played jazz and rock piano, and dressed in tiedye:a true kindred spirit. Peter placed electrodes in the inferior colliculusof the barn owl, part of its auditory system. Then, he played the owlsa version of Strauss’s “The Blue Danube Waltz” made up of tones fromwhich the fundamental frequency had been removed. Petr hypothesizedthat if the missing fundamental is restored at early levels of auditory processing,neurons in the owl’s inferior colliculus should fire at the rate ofthe missing fundamental. This was exactly what he found. And becausethe electrodes put out a small electrical signal with each firing—and becausethe firing rate is the same as a frequency of firing—Petr sent theoutput of these electrodes to a small amplifier, and played back thesound of the owl’s neurons through a loudspeaker. What he heard wasastonishing; the melody of “The Blue Danube Waltz” sang clearly fromthe loudspeakers: ba da da da da, deet deet, deet deet. We were hearingthe firing rates of the neurons and they were identical to the frequencyof the missing fundamental. The overtone series had an instantiation notjust in the early levels of auditory processing, but in a completely differentspecies.One could imagine an alien species that does not have ears, or thatdoesn’t have the same internal experience of hearing that we do. But it

42 This Is Your Brain on Musicwould be difficult to imagine an advanced species that had no abilitywhatsoever to sense vibrating objects. Where there is atmosphere thereare molecules that vibrate in response to movement. And knowingwhether something is generating noise or moving toward us or awayfrom us, even when we can’t see it (because it is dark, our eyes aren’t attendingto it, or we’re asleep) has a great survival value.Because most physical objects cause molecules to vibrate in severalmodes at once, and because for many, many objects the modes bear simpleinteger relations to one another, the overtone series is a fact-of-theworldthat we expect to find everywhere we look: in North America, inFiji, on Mars, and on the planets orbiting Antares. Any organism thatevolved in a world with vibrating objects is likely—given enough evolutionarytime—to have evolved a processing unit in the brain that incorporatedthese regularities of its world. Because pitch is a fundamentalcue to an object’s identity, we would expect to find tonotopic mappingsas we do in human auditory cortex, and synchronous neural firings fortones that bear octave and other harmonic relations to one another; thiswould help the brain (alien or terrestrial) to figure out that all thesetones probably originated from the same object.The overtones are often referred to by numbers: The first overtone isthe first vibration frequency above the fundamental, the second overtoneis the second vibration frequency above the fundamental, etc. Becausephysicists like to make the world confusing for the rest of us, thereis a parallel system of terminology called harmonics, and I think it wasdesigned to make undergraduates go crazy. In the lingo of harmonics,the first harmonic is the fundamental frequency, the second harmonic isequal to the first overtone, and so on. Not all instruments vibrate inmodes that are so neatly defined. Sometimes, as with the piano (becauseit is a percussive instrument), the overtones can be close, but not exact,multiples of the fundamental frequency, and this contributes to theircharacteristic sound. Percussion instruments, chimes, and other objects—depending on composition and shape—often have overtones that areclearly not integer multiples of the fundamental, and these are calledpartials or inharmonic overtones. Generally, instruments with inhar-

What Is Music? 43monic overtones lack the clear sense of pitch that we associate with harmonicinstruments, and the cortical basis for this may relate to a lack ofsynchronous neural firing. But they still do have a sense of pitch, and wehear this most clearly when we can play inharmonic notes in succession.Although you may not be able to hum along with the sound of a singlenote played on a woodblock or a chime, we can play a recognizablemelody on a set of woodblocks or chimes because our brain focuses onthe changes in the overtones from one to another. This is essentiallywhat is happening when we hear people playing a song on their cheeks.A flute, a violin, a trumpet, and a piano can all play the same tone—that is, you can write a note on a musical score and each instrument willplay a tone with an identical fundamental frequency, and we will (tendto) hear an identical pitch. But these instruments all sound very differentfrom one another.This difference is timbre (pronounced TAM-ber), and it is the mostimportant and ecologically relevant feature of auditory events. The timbreof a sound is the principal feature that distinguishes the growl of alion from the purr of a cat, the crack of thunder from the crash of oceanwaves, the voice of a friend from that of a bill collector one is trying tododge. Timbral discrimination is so acute in humans that most of us canrecognize hundreds of different voices. We can even tell whether someoneclose to us—our mother, our spouse—is happy or sad, healthy orcoming down with a cold, based on the timbre of that voice.Timbre is a consequence of the overtones. Different materials havedifferent densities. A piece of metal will tend to sink to the bottom of apond; an identically sized and shaped piece of wood will float. Partly dueto density, and partly due to size and shape, different objects also makedifferent noises when you strike them with your hand, or gently tap themwith a hammer. Imagine the sound that you’d hear if you tap a hammer(gently, please!) against a guitar—a hollow, wooden plunk sound. Or ifyou tap a piece of metal, like a saxophone—a tinny plink. When you tapthese objects, the energy from the hammer causes the molecules withinthem to vibrate, to dance at several different frequencies, frequenciesdetermined by the material the object is made out of, its size, and its

44 This Is Your Brain on Musicshape. If the object is vibrating at, say, 100 Hz, 200 Hz, 300 Hz, 400 Hz,etc., the intensity of vibration doesn’t have to be the same for each ofthese harmonics, and in fact, typically, it is not.When you hear a saxophone playing a tone with a fundamental frequencyof 220 Hz, you are actually hearing many tones, not just one. Theother tones you hear are integer multiples of the fundamental: 440, 660,880, 1200, 1420, 1640, etc. These different tones—the overtones—havedifferent intensities, and so we hear them as having different loudnesses.The particular pattern of loudnesses for these tones is distinctive of thesaxophone, and they are what give rise to its unique tonal color, itsunique sound—its timbre. A violin playing the same written note (220Hz) will have overtones at the same frequencies, but the pattern of howloud each one is with respect to the others will be different. Indeed, foreach instrument, there exists a unique pattern of overtones. For one instrument,the second overtone might be louder than in another, while thefifth overtone might be softer. Virtually all of the tonal variation wehear—the quality that gives a trumpet its trumpetiness and that gives apiano its pianoness—comes from the unique way in which the loudnessesof the overtones are distributed.Each instrument has its own overtone profile, which is like a fingerprint.It is a complicated pattern that we can use to identify the instrument.Clarinets, for example, are characterized by having relatively highamounts of energy in the odd harmonics—three times, five times, andseven times the multiples of the fundamental frequency, etc. (This is aconsequence of their being a tube that is closed at one end and open atthe other.) Trumpets are characterized by having relatively evenamounts of energy in both the odd and the even harmonics (like the clarinet,the trumpet is also closed at one end and open at the other, but themouthpiece and bell are designed to smooth out the harmonic series). Aviolin that is bowed in the center will yield mostly odd harmonics and accordinglycan sound similar to a clarinet. But bowing one third of theway down the instrument emphasizes the third harmonic and its multiples:the sixth, the ninth, the twelfth, etc.All trumpets have a timbral fingerprint, and it is readily distinguish-

What Is Music? 45able from the timbral fingerprint for a violin, piano, or even the humanvoice. To the trained ear, and to most musicians, there even exist differencesamong trumpets—all trumpets don’t sound alike, nor do all pianosor all accordions. (Well, to me all accordions sound alike, and the sweetest,most enjoyable sound I can imagine is the sound they would makeburning in a giant bonfire.) What distinguishes one particular piano fromanother is that their overtone profiles will differ slightly from each other,but not, of course, as much as they will differ from the profile for a harpsichord,organ, or tuba. Master musicians can hear the difference betweena Stradivarius violin and a Guarneri within one or two notes. I canhear the difference between my 1956 Martin 000-18 acoustic guitar, my1973 Martin D-18, and my 1996 Collings D2H very clearly; they sound likedifferent instruments, even though they are all acoustic guitars; I wouldnever confuse one with another. That is timbre.Natural instruments—that is, acoustic instruments made out of realworldmaterials such as metal and wood—tend to produce energy at severalfrequencies at once because of the way the internal structure oftheir molecules vibrates. Suppose that I invent an instrument that, unlikeany natural instruments we know of, produces energy at one, and onlyone, frequency. Let’s call this hypothetical instrument a generator (becauseit can generate tones of specific frequencies). If I line up a bunchof generators, I could set each one of them to play a specific frequencycorresponding to the overtone series for a particular instrument playinga particular tone. I could have a bank of these generators making soundsat 110, 220, 330, 440, 550, and 660 Hz, which would give the listener theimpression of a 110 Hz tone played by a musical instrument. Furthermore,I could control the amplitude of each of my generators and makeeach of the tones play at a particular loudness, corresponding to theovertone profile of a natural musical instrument. If I did that, the resultingbank of generators would approximate the sound of a clarinet, orflute, or any other instrument I was trying to emulate.Additive synthesis such as the above approach achieves a syntheticversion of a musical-instrument timbre by adding together elementalsonic components of the sound. Many pipe organs, such as those found

46 This Is Your Brain on Musicin churches, have a feature that will let you play around with this. Onmost pipe organs you press a key (or a pedal), which sends a blast of airthrough a metal pipe. The organ is constructed of hundreds of pipes ofdifferent sizes, and each one produces a different pitch, correspondingto its size, when air is shot through it; you can think of them as mechanicalflutes, in which the air is supplied by an electric motor rather than bya person blowing. The sound that we associate with a church organ—itsparticular timbre—is a function of there being energy at several differentfrequencies at once, just as with other instruments. Each pipe of the organproduces an overtone series, and when you press a key on the organkeyboard, a column of air is blasted through more than one pipe at atime, giving a very rich spectrum of sounds. These supplementary pipes,in addition to the one that vibrates at the fundamental frequency of thetone you’re trying to play, either produce tones that are integer multiplesof the fundamental frequency, or are closely related to it mathematicallyand harmonically.The organ player typically has control over which of these supplementarypipes he wants to blow air through by pulling and pushinglevers, or drawbars, that direct the flow of air. Knowing that clarinetshave a lot of energy in the odd harmonics of the overtone series, a cleverorgan player could simulate the sound of a clarinet by manipulatingdrawbars in such a way as to re-create the overtone series of that instrument.A little bit of 220 Hz here, a dash of 330 Hz, a dollop of 440 Hz, aheaping helping of 550 Hz, and voilà!—you’ve cooked yourself up a reasonablefacsimile of an instrument.Starting in the late 1950s, scientists began experimenting with buildingsuch synthesis capabilities into smaller, more compact electronic devices,creating a family of new musical instruments known collectivelyas synthesizers. By the 1960s, synthesizers could be heard on records bythe Beatles (on “Here Comes the Sun” and “Maxwell’s Silver Hammer”)and Walter/Wendy Carlos (Switched-On Bach), followed by groups whosculpted their sound around the synthesizer, such as Pink Floyd andEmerson, Lake and Palmer.

What Is Music? 47Many of these synthesizers used additive synthesis as I’ve describedit here, and later ones used more complex algorithms such as wave guidesynthesis (invented by Julius Smith at Stanford) and FM synthesis (inventedby John Chowning at Stanford). But merely copying the overtoneprofile, while it can create a sound reminiscent of the actual instrument,yields a rather pale copy. There is more to timbre than just the overtoneseries. Researchers still argue about what this “more” is, but it is generallyaccepted that, in addition to the overtone profile, timbre is definedby two other attributes that give rise to a perceptual difference from oneinstrument to another: attack and flux.Stanford University sits on a bucolic stretch of land just south of SanFrancisco and east of the Pacific Ocean. Rolling hills covered with pasturelandlie to the west, and the fertile Central Valley of California is justan hour or so to the east, home of a large proportion of the world’sraisins, cotton, oranges, and almonds. To the south, near the town ofGilroy, are vast fields of garlic. Also to the south is Castroville, known asthe “artichoke capitol of the world.” (I once suggested to the CastrovilleChamber of Commerce that they change capitol to heart. The responsewas not enthusiastic.)Stanford has become something of a second home for computer scientistsand engineers who love music. John Chowning, who was wellknown as an avant-garde composer, has had a professorship in the musicdepartment there since the 1970s, and was among a group of pioneeringcomposers at the time who were using the computer to create,store, and reproduce sounds in their compositions. Chowning later becamethe founding director of the Center for Computer Research in Musicand Acoustics at Stanford, known as CCRMA (pronounced CAR-ma;insiders joke that the first c is silent). Chowning is warm and friendly.When I was an undergraduate at Stanford, he would put his hand on myshoulder and ask what I was working on. You got the feeling talking to astudent was for him an opportunity to learn something. In the early1970s, while fiddling with the computer and with sine waves—the sortsof artificial sounds that are made by computers and used as the building

48 This Is Your Brain on Musicblocks of additive synthesis—Chowning noticed that changing the frequencyof these waves as they were playing created sounds that weremusical. By controlling these parameters just so, he was able to simulatethe sounds of a number of musical instruments. This new technique becameknown as frequency modulation synthesis, or FM synthesis, andbecame embedded first in the Yamaha DX9 and DX7 line of synthesizers,which revolutionized the music industry from the moment of their introductionin 1983. FM synthesis democratized music synthesis. Before FM,synthesizers were expensive, clunky, and hard to control. Creating newsounds took a great deal of time, experimentation, and know-how. Butwith FM, any musician could obtain a convincing instrumental soundat the touch of a button. Songwriters and composers who could not affordto hire a horn section or an orchestra could now play around withthese textures and sounds. Composers and orchestrators could test outarrangements before taking the time of an entire orchestra to see whatworked and what didn’t. New Wave bands like the Cars and the Pretenders,as well as mainstream artists like Stevie Wonder, Hall andOates, and Phil Collins, started to use FM synthesis widely in theirrecordings. A lot of what we think of as “the eighties sound” in popularmusic owes its distinctiveness to the particular sound of FM synthesis.With the popularization of FM came a steady stream of royalty incomethat allowed Chowning to build up CCRMA, attracting graduate studentsand top-flight faculty members. Among the first of many famous electronicmusic/music-psychology celebrities to come to CCRMA wereJohn R. Pierce and Max Mathews. Pierce had been the vice president ofresearch at the Bell Telephone Laboratories in New Jersey, and supervisedthe team of engineers who built and patented the transistor—and itwas Pierce who named the new device (TRANSfer resISTOR). In his distinguishedcareer, he also is credited with inventing the traveling wavevacuum tube, and launching the first telecommunications satellite, Telstar.He was also a respected science fiction writer under the pseudonymJ. J. Coupling. Pierce created a rare environment in any industry or researchlab, one in which the scientists felt empowered to do their best

What Is Music? 49and in which creativity was highly valued. At the time, the Bell TelephoneCompany/AT&T had a complete monopoly on telephone service in theU.S. and a large cash reserve. Their laboratory was something of a playgroundfor the very best and brightest inventors, engineers, and scientistsin America. In the Bell Labs “sandbox,” Pierce allowed his people to becreative without worrying about the bottom line or the applicability oftheir ideas to commerce. Pierce understood that the only way true innovationcan occur is when people don’t have to censor themselves and canlet their ideas run free. Although only a small proportion of those ideasmay be practical, and a smaller proportion still would become products,those that did would be innovative, unique, and potentially very profitable.Out of this environment came a number of innovations includinglasers, digital computers, and the Unix operating system.I first met Pierce in 1990 when he was already eighty and was givinglectures on psychoacoustics at CCRMA. Several years later, after I hadearned my Ph.D. and moved back to Stanford, we became friends andwould go out to dinner every Wednesday night and discuss research. Heonce asked me to explain rock and roll music to him, something he hadnever paid any attention to and didn’t understand. He knew about myprevious career in the music business, and he asked if I could come overfor dinner one night and play six songs that captured all that was importantto know about rock and roll. Six songs to capture all of rock androll? I wasn’t sure I could come up with six songs to capture the Beatles,let alone all of rock and roll. The night before he called to tell me that hehad heard Elvis Presley, so I didn’t need to cover that.Here’s what I brought to dinner:1) “Long Tall Sally,” Little Richard2) “Roll Over Beethoven,” the Beatles3) “All Along the Watchtower,” Jimi Hendrix4) “Wonderful Tonight,” Eric Clapton5) “Little Red Corvette,” Prince6) “Anarchy in the U.K.,” the Sex Pistols

50 This Is Your Brain on MusicA couple of the choices combined great songwriters with differentperformers. All are great songs, but even now I’d like to make some adjustments.Pierce listened and kept asking who these people were, whatinstruments he was hearing, and how they came to sound the way theydid. Mostly, he said that he liked the timbres of the music. The songsthemselves and the rhythms didn’t interest him that much, but he foundthe timbres to be remarkable—new, unfamiliar, and exciting. The fluidromanticism of Clapton’s guitar solo in “Wonderful Tonight,” combinedwith the soft, pillowy drums. The sheer power and density of the Sex Pistols’brick-wall-of-guitars-and-bass-and-drums. The sound of a distortedelectric guitar wasn’t all that was new to Pierce. The ways in which instrumentswere combined to create a unified whole—bass, drums, electricand acoustic guitars, and voice—that was something he had neverheard before. Timbre was what defined rock for Pierce. And it was a revelationto both of us.The pitches that we use in music—the scales—have remained essentiallyunchanged since the time of the Greeks, with the exception of thedevelopment—really a refinement—of the equal tempered scale duringthe time of Bach. Rock and roll may be the final step in a millennium-longmusical revolution that gave perfect fourths and fifths a prominence inmusic that had historically been been given only to the octave. Duringthis time, Western music was largely dominated by pitch. For the pasttwo hundred years or so, timbre has become increasingly important. Astandard component of music across all genres is to restate a melody usingdifferent instruments—from Beethoven’s Fifth and Ravel’s “Bolero”to the Beatles’ “Michelle” and George Strait’s “All My Ex’s Live in Texas.”New musical instruments have been invented so that composers mighthave a larger palette of timbral colors from which to draw. When a countryor popular singer stops singing and another instrument takes up themelody—even without changing it in any way—we find pleasurable therepetition of the same melody with a different timbre.The avant-garde composer Pierre Schaeffer (pronounced Sheh-FEHR,using your best imitation of a French accent) performed some crucial

What Is Music? 51experiments in the 1950s that demonstrated an important attribute oftimbre in his famous “cut bell” experiments. Schaeffer recorded a numberof orchestral instruments on tape. Then, using a razor blade, he cutthe beginnings off of these sounds. This very first part of a musical instrumentsound is called the attack; this is the sound of the initial hit,strum, bowing, or blowing that causes the instrument to make sound.The gesture our body makes in order to create sound from an instrumenthas an important influence on the sound the instrument makes. Butmost of that dies away after the first few seconds. Nearly all of the gestureswe make to produce a sound are impulsive—they involve short,punctuated bursts of activity. In percussion instruments, the musiciantypically does not remain in contact with the instrument after this initialburst. In wind instruments and bowed instruments, on the other hand,the musician continues to be in contact with the instrument after the initialimpulsive contact—the moment when the air burst first leaves hermouth or the bow first contacts the string; the continued blowing andbowing has a smooth, continuous, and less impulsive quality.The introduction of energy to an instrument—the attack phase—usually creates energy at many different frequencies that are not relatedto one another by simple integer multiples. In other words, for the briefperiod after we strike, blow into, pluck, or otherwise cause an instrumentto start making sound, the impact itself has a rather noisy qualitythat is not especially musical—more like the sound of a hammer hittinga piece of wood, say, than like a hammer hitting a bell or a piano string,or like the sound of wind rushing through a tube. Following the attack isa more stable phase in which the musical tone takes on the orderly patternof overtone frequencies as the metal or wood (or other material)that the instrument is made out of starts to resonate. This middle part ofa musical tone is referred to as the steady state—in most instances theovertone profile is relatively stable while the sound emanates from theinstrument during this time.After Schaeffer edited out the attack of orchestral instrument recordings,he played back the tape and found that it was nearly impossible formost people to identify the instrument that was playing. Without the at-

52 This Is Your Brain on Musictack, pianos and bells sounded remarkably unlike pianos and bells, andremarkably similar to one another. If you splice the attack of one instrumentonto the steady state, or body, from another, you get varied results:In some cases, you hear an ambiguous hybrid instrument that soundsmore like the instrument that the attack came from than the one thesteady state came from. Michelle Castellengo and others have discoveredthat you can create entirely new instruments this way; for example, splicinga violin bow sound onto a flute tone creates a sound that strongly resemblesa hurdy-gurdy street organ. These experiments showed theimportance of the attack.The third dimension of timbre—flux—refers to how the soundchanges after it has started playing. A cymbal or gong has a lot of flux—its sound changes dramatically over the time course of its sound—whilea trumpet has less flux—its tone is more stable as it evolves. Also, instrumentsdon’t sound the same across their range. That is, the timbre ofan instrument sounds different when playing high and low notes. WhenSting reaches up toward the top of his vocal range in “Roxanne” (by ThePolice), his straining, reedy voice conveys a type of emotion that he can’tachieve in the lower parts of his register, such as we hear on the openingverse of “Every Breath You Take,” a more deliberate, longing sound. Thehigh part of Sting’s register pleads with us urgently as his vocal cordsstrain, the low part suggests a dull aching that we feel has been going onfor a long time, but has not yet reached the breaking point.Timbre is more than the different sounds that instruments make.Composers use timbre as a compositional tool; they choose musical instruments—andcombinations of musical instruments—to express particularemotions, and to convey a sense of atmosphere or mood. There isthe almost comical timbre of the bassoon in Tchaikovsky’s NutcrackerSuite as it opens the “Chinese Dance,” and the sensuousness of StanGetz’s saxophone on “Here’s That Rainy Day.” Substitute a piano for theelectric guitars in the Rolling Stones’ “Satisfaction” and you’d have anentirely different animal. Ravel used timbre as a compositional device inBolero, repeating the main theme over and over again with different timbres;he did this after he suffered brain damage that impaired his ability

What Is Music? 53to hear pitch. When we think of Jimi Hendrix, it is the timbre of his electricguitars and his voice that we are likely to recall the most vividly.Composers such as Scriabin and Ravel talk about their works assound paintings, in which the notes and melodies are the equivalent ofshape and form, and the timbre is equivalent to the use of color andshading. Several popular songwriters—Stevie Wonder, Paul Simon, andLindsey Buckingham—have described their compositions as soundpaintings, with timbre playing a role equivalent to the one that color doesin visual art, separating melodic shapes from one another. But one of thethings that makes music different from painting is that it is dynamic,changing across time, and what moves the music forward are rhythmand meter. Rhythm and meter are the engine driving virtually all music,and it is likely that they were the very first elements used by our ancestorsto make protomusics, a tradition we still hear today in tribal drumming,and in the rituals of various preindustrial cultures. While I believetimbre is now at the center of our appreciation of music, rhythm has heldsupreme power over listeners for much longer.

2. Foot TappingDiscerning Rhythm, Loudness, and HarmonyIsaw Sonny Rollins perform in Berkeley in 1977; he is one of the mostmelodic saxophone players of our time. Yet nearly thirty years later,while I can’t remember any of the pitches that he played, I clearly remembersome of the rhythms. At one point, Rollins improvised for threeand a half minutes by playing the same one note over and over again withdifferent rhythms and subtle changes in timing. All that power in onenote! It wasn’t his melodic innovation that got the crowd to their feet—it was rhythm. Virtually every culture and civilization considers movementto be an integral part of music making and listening. Rhythm iswhat we dance to, sway our bodies to, and tap our feet to. In so manyjazz performances, the part that excites the audience most is the drumsolo. It is no coincidence that making music requires the coordinated,rhythmic use of our bodies, and that energy be transmitted from bodymovements to a musical instrument. At a neural level, playing an instrumentrequires the orchestration of regions in our primitive, reptilianbrain—the cerebellum and the brain stem—as well as higher cognitivesystems such as the motor cortex (in the parietal lobe) and the planningregions of our frontal lobes, the most advanced region of the brain.Rhythm, meter, and tempo are related concepts that are often confusedwith one another. Briefly, rhythm refers to the lengths of notes,

56 This Is Your Brain on Musictempo refers to the pace of a piece of music (the rate at which you wouldtap your foot to it), and meter refers to when you tap your foot hard versuslight, and how these hard and light taps group together to form largerunits.One of the things we usually want to know when performing music ishow long a note is to be played. The relationship between the length of onenote and another is what we call rhythm, and it is a crucial part of whatturns sounds into music. Among the most famous rhythms in our cultureis the rhythm often called “shave-and-a-haircut, two bits,” sometimes usedas the “secret” knock on a door. An 1899 recording by Charles Hale, “At aDarktown Cakewalk,” is the first documented use of this rhythm. Lyricswere later attached to the rhythm in a song by Jimmie Monaco and JoeMcCarthy called “Bum-Diddle-De-Um-Bum, That’s It!” in 1914. In 1939,the same musical phrase was used in the song “Shave and a Haircut—Shampoo” by Dan Shapiro, Lester Lee, and Milton Berle. How the wordshampoo became two-bits is a mystery. Even Leonard Bernstein got intothe act by scoring this rhythm in the song “Gee, Officer Krupke” from themusical West Side Story. In “shave-and-a-haircut” we hear a series of notesof two different lengths, long and short; the long notes are twice as long asthe short ones: long-short-short-long-long (rest) long-long.In the William Tell overture by Rossini (what many of us know as thetheme from The Lone Ranger) we also hear a series of notes of two differentlengths, long and short; again, the long notes are twice as long asthe short ones: da-da-bump da-da-bump da-da-bump bump bump (hereI’ve used the “da” syllable for short, and the “bump” syllable for long).“Mary Had a Little Lamb” uses short and long syllables, too, in this casesix equal duration notes (Ma-ry had a lit-tle) followed by a long one(lamb) roughly twice as long as the short ones. The rhythmic ratio of 2:1,like the octave in pitch ratios, appears to be a musical universal. We seeit in the theme from The Mickey Mouse Club (bump-ba bump-ba bumpbabump-ba bump-ba bump-ba baaaaah) in which we have three levels ofduration, each one twice as long as the other. We see it in The Police’s“Every Breath You Take” (da-da-bump da-da baaaaah), in which thereare again three levels:

Foot Tapping 57Ev-ry breath you-oo taaake1 1 2 2 4(The 1 represents one unit of some arbitrary time just to illustrate thatthe words breath and you are twice as long as the syllables Ev and ry,and that the word take is four times as long as Ev or ry and twice as longas breath or you.)Rhythms in most of the music we listen to are seldom so simple. Inthe same way that a particular arrangement of pitches—the scale—canevoke music of a different culture, style, or idiom, so can a particulararrangement of rhythms. Although most of us couldn’t reproduce a complexLatin rhythm, we recognize as soon as we hear it that it is Latin,as opposed to Chinese, Arabic, Indian, or Russian. When we organizerhythms into strings of notes, of varying lengths and emphases, we developmeter and establish tempo.Tempo refers to the pace of a musical piece—how quickly or slowly itgoes by. If you tap your foot or snap your fingers in time to a piece of music,the tempo of the piece will be directly related to how fast or slow youare tapping. If a song is a living, breathing entity, you might think of thetempo as its gait—the rate at which it walks by—or its pulse—the rate atwhich the heart of the song is beating. The word beat indicates the basicunit of measurement in a musical piece; this is also called the tactus.Most often, this is the natural point at which you would tap your foot orclap your hands or snap your fingers. Sometimes, people tap at half ortwice the beat, due to different neural processing mechanisms from oneperson to another as well as differences in musical background, experience,and interpretation of a piece. Even trained musicians can disagreeon what the tapping rate should be. But they always agree on the underlyingspeed at which the piece is unfolding, also called tempo; the disagreementsare simply about subdivisions or superdivisions of thatunderlying pace.Paula Abdul’s “Straight Up” and AC/DC’s “Back in Black” have atempo of 96, meaning that there are 96 beats per minute. If you dance to“Straight Up” or “Back in Black,” it is likely that you will be putting a foot

58 This Is Your Brain on Musicdown 96 times per minute or perhaps 48, but not 58 or 69. In “Back inBlack” you can hear the drummer playing a beat on his high-hat cymbalat the very beginning, steadily, deliberately, at precisely 96 beats perminute. Aerosmith’s “Walk This Way” has a tempo of 112, Michael Jackson’s“Billie Jean” has a tempo of 116, and the Eagles’ “Hotel California”has a tempo of 75.Two songs can have the same tempo but feel very different. In “Backin Black,” the drummer plays his cymbal twice for every beat (eighthnotes) and the bass player plays a simple, syncopated rhythm perfectlyin time with the guitar. On “Straight Up” there is so much going on, it isdifficult to describe it in words. The drums play a complex, irregular patternwith beats as fast as sixteenth notes, but not continuously—the“air” between drum hits imparts a sound typical of funk and hip-hop music.The bass plays a similarly complex and syncopated melodic line thatsometimes coincides with and sometimes fills in the holes of the drumpart. In the right speaker (or the right ear of headphones) we hear theonly instrument that actually plays on the beat every beat—a Latin instrumentcalled an afuche or cabasa that sounds like sandpaper or beansshaking inside a gourd. Putting the most important rhythm on a light,high-pitched instrument is an innovative rhythmic technique that turnsupside down the normal rhythmic conventions. While all this is goingon, synthesizers, guitar, and special percussion effects fly in and out ofthe song dramatically, emphasizing certain beats now and again toadd excitement. Because it is hard to predict or memorize where manyof these are, the song holds a certain appeal over many, many listenings.Tempo is a major factor in conveying emotion. Songs with fast tempostend to be regarded as happy, and songs with slow tempos as sad. Althoughthis is an oversimplification, it holds in a remarkable range ofcircumstances, across many cultures, and across the lifespan of an individual.The average person seems to have a remarkable memory fortempo. In an experiment that Perry Cook and I published in 1996, weasked people to simply sing their favorite rock and popular songs frommemory and we were interested to know how close they came to the actualtempo of the recorded versions of those songs. As a baseline, we

Foot Tapping 59considered how much variation in tempo the average person can detect;that turns out to be 4 percent. In other words, for a song with a tempo of100 bpm, if the tempo varies between 96–100, most people, even someprofessional musicians, won’t detect this small change (although mostdrummers would—their job requires that they be more sensitive to tempothan other musicians, because they are responsible for maintainingtempo when there is no conductor to do it for them). A majority ofpeople in our study—nonmusicians—were able to sing songs within 4percent of their nominal tempo.The neural basis for this striking accuracy is probably in the cerebellum,which is believed to contain a system of timekeepers for our dailylives and to synchronize to the music we are hearing. This means thatsomehow, the cerebellum is able to remember the “settings” it uses forsynchronizing to music as we hear it, and it can recall those settingswhen we want to sing a song from memory. It allows us to synchronizeour singing with a memory of the last time we sang. The basal ganglia—what Gerald Edelman has called “the organs of succession”—are almostcertainly involved, as well, in generating and shaping rhythm, tempo,and meter.Meter refers to the way in which the pulses or beats are grouped together.Generally when we’re tapping or clapping along with music,there are some beats that we feel more strongly than others. It feels as ifthe musicians play this beat louder and more heavily than the others.This louder, heavier beat is perceptually dominant, and other beats thatfollow it are perceptually weaker until another strong one comes in.Every musical system that we know of has patterns of strong and weakbeats. The most common pattern in Western music is for the strong beatsto occur once every 4 beats: STRONG-weak-weak-weak STRONG-weakweak-weak.Usually the third beat in a four-beat pattern is somewhatstronger than the second and fourth: There is a hierarchy of beatstrengths, with the first being the strongest, the third being next, followedby the second and fourth. Somewhat less often the strong beatoccurs once in every three in what we call the “waltz” beat: STRONGweak-weakSTRONG-weak-weak. We usually count to these beats as

60 This Is Your Brain on Musicwell, in a way that emphasizes which one is the strong beat: ONE-twothree-four,ONE-two-three-four, or ONE-two-three, ONE-two-three.Of course music would be boring if we only had these straight beats.We might leave one out to add tension. Think of “Twinkle, Twinkle LittleStar,” written by Mozart when he was six years old. The notes don’t occuron every beat:ONE-two-three-fourONE-two-three-(rest)ONE-two-three-fourONE-two-three-(rest):TWIN-kle twin-kleLIT-tle star (rest)HOW-I won-derWHAT you are (rest).A nursery rhyme written to this same tune, “Ba Ba Black Sheep” subdividesthe beat. A simple ONE-two-three-four can be divided intosmaller, more interesting parts:BA ba black sheepHAVE-you-any-wool?Notice that each syllable in “have-you-any” goes by twice as fast asthe syllables “ba ba black.” The quarter notes have been divided in half,and we can count this asONE-two-three-fourONE-and-two-and-three-(rest).In “Jailhouse Rock,” performed by Elvis Presley and written by twooutstanding songwriters of the rock era, Jerry Leiber and Mike Stoller,

Foot Tapping 61the strong beat occurs on the first note Presley sings, and then everyfourth note after that:[Line 1:] WAR-den threw a party at the[Line 2:] COUN-ty jail (rest) the[Line 3:] PRIS-on band was there and they be-[Line 4:] GAN to wailIn music with lyrics, the words don’t always line up perfectly withthe downbeats; in “Jailhouse Rock” part of the word began starts beforea strong beat and finishes on that strong beat. Most nursery rhymesand simple folk songs, such as “Ba Ba Black Sheep” or “Frère Jacques,”don’t do this. This lyrical technique works especially well on “JailhouseRock” because in speech the accent is on the second syllable of began;spreading the word across lines like this gives the song additional momentum.By convention in Western music, we have names for the note durationssimilar to the way we name musical intervals. A musical interval of a“perfect fifth” is a relative concept—it can start on any note, and then bydefinition, notes that are either seven semitones higher or seven semitoneslower in pitch are considered a perfect fifth away from the startingnote. The standard duration is called a whole note and it lasts four beats,regardless of how slow or how fast the music is moving—that is, irrespectiveof tempo. (At a tempo of sixty beats per minute—as in the FuneralMarch—each beat lasts one second, so a whole note would lastfour seconds.) A note with half the duration of a whole note is called,logically enough, a half note, and a note half as long as that is called aquarter note. For most music in the popular and folk tradition, the quarternote is the basic pulse—the four beats that I was referring to earlierare beats of a quarter note. We talk about such songs as being in 4/4 time:The numerator tells us that the song is organized into groups of fournotes, and the denominator tells us that the basic note length is a quarter

62 This Is Your Brain on Musicnote. In notation and conversation, we refer to each of these groups offour notes as a measure or a bar. One measure of music in 4/4 time hasfour beats, where each beat is a quarter note. This does not imply thatthe only note duration in the measure is the quarter note. We can havenotes of any duration, or rests—that is to say, no notes at all; the 4/4 indicationis only meant to describe how we count the beats.“Ba Ba Black Sheep” has four quarter notes in its first measure, andthen eighth notes (half the duration of a quarter note) and a quarter noterest in the second measure. I’ve used the symbol ⎢ to indicate a quarternote, and ⎣ to indicate an eighth note, and I’ve kept the spacing betweensyllables proportional to how much time is spent on them:[measure 1:] ba ba black sheep⎢ ⎢ ⎢ ⎢[measure 2:] have you an- y wool (rest)⎣ ⎣ ⎣ ⎣ ⎢ ⎢You can see in the diagram that the eighth notes go by twice as fast asthe quarter notes.In “That’ll Be the Day” by Buddy Holly, the song begins with a pickupnote; the strong beat occurs on the next note and then every fourth noteafter that, just as in “Jailhouse Rock”:THAT’ll be the day (rest) whenYOU say good-bye-yes;THAT’ll be the day (rest) whenYOU make me cry-hi; youSAY you gonna leave (rest) youKNOW it’s a lie ’causeTHAT’ll be the day-ay-AY when I die.Notice how, like Elvis, Holly cuts a word in two across lines (day inthe last two lines). To most people, the tactus is four beats between

Foot Tapping 63downbeats of this song, and they would tap their feet four timesfrom one downbeat to the next. Here, all caps indicate the downbeat asbefore, and bold indicates when you would tap your foot against thefloor:WellTHAT’ll be the day (rest) whenYOU say good-bye-yes;THAT’ll be the day (rest) whenYOU make me cry-hi; youSAY you gonna leave (rest) youKNOW it’s a lie ’causeTHAT’ll be the day-ay-AY when I die.If you pay close attention to the song’s lyrics and their relationship tothe beat, you’ll notice that a foot tap occurs in the middle of some of thebeats. The first say on the second line actually begins before you putyour foot down—your foot is probably up in the air when the word saystarts, and you put your foot down in the middle of the word. The samething happens with the word yes later in that line. Whenever a note anticipatesa beat—that is, when a musician plays a note a bit earlier thanthe strict beat would call for—this is called syncopation. This is a veryimportant concept that relates to expectation, and ultimately to the emotionalimpact of a song. The syncopation catches us by surprise, andadds excitement.As with many songs, some people feel “That’ll Be the Day” in halftime; there’s nothing wrong with this—it is another interpretation and avalid one—and they tap their feet twice in the same amount of time otherpeople tap four times: once on the downbeat, and again two beats later.The song actually begins with the word Well that occurs before astrong beat—this is called a pickup note. Holly uses two words, Well,you, as pickup notes to the verse, also, and then right after them we’re insync again with the downbeats:

64 This Is Your Brain on Music[pick up][line 1][line 2][line 3][line 4]Well, youGAVE me all your lovin’ and your(REST) tur-tle dovin’ (rest)ALL your hugs and kisses and your(REST) money too.What Holly does here that is so clever is that he violates our expectationsnot just with anticipations, but by delaying words. Normally, therewould be a word on every downbeat, as in children’s nursery rhymes. Butin lines two and four of the song, the downbeat comes and he’s silent!This is another way that composers build excitement, by not giving uswhat we would normally expect.When people clap their hands or snap their fingers with music, theysometimes quite naturally, and without training, keep time differentlythan they would do with their feet: They clap or snap not on the downbeat,but on the second beat and the fourth beat. This is the so-calledbackbeat that Chuck Berry sings about in his song “Rock and Roll Music.”John Lennon said that the essence of rock and roll songwriting forhim was to “Just say what it is, simple English, make it rhyme, and put abackbeat on it.” In “Rock and Roll Music” (which John sang with theBeatles), as on most rock songs, the backbeat is what the snare drum isplaying: The snare drum plays only on the second and fourth beat ofeach measure, in opposition to the strong beat which is on one, and asecondary strong beat, on three. This backbeat is the typical rhythmic elementof rock music, and Lennon used it a lot as in “Instant Karma”(*whack* below indicates where the snare drum is played in the song, onthe backbeat):Instant karma’s gonna get you(rest) *whack* (rest) *whack*“Gonna knock you right on the head”(rest) *whack* (rest) *whack*...

Foot Tapping 65But we all *whack* shine *whack*on *whack* (rest) *whack*Like the moon *whack* and the stars *whack*and the sun *whack* (rest) *whack*In “We Will Rock You” by Queen, we hear what sounds like feet stampingon stadium bleachers twice in a row (boom-boom) and then hand-clapping(CLAP) in a repeating rhythm: boom-boom-CLAP, boom-boom-CLAP;the CLAP is the backbeat.Imagine now the John Philip Sousa march, “The Stars and StripesForever.” If you can hear it in your mind, you can tap your foot alongwith the mental rhythm. While the music goes “DAH-dah-ta DUM-dumdah DUM-dum dum-dum DUM,” your foot will be tapping DOWN-upDOWN-up DOWN-up DOWN-up. In this song, it is natural to tap your footfor every two quarter notes. We say that this song is “in two,” meaningthat the natural grouping of rhythms is two quarter notes per beat.Now imagine “My Favorite Things” (words and music by RichardRodgers and Oscar Hammerstein). This song is in waltz time, or what iscalled 3/4 time. The beats seem to arrange themselves in groups of three,with a strong beat followed by two weak ones. “RAIN-drops-on ROSE-esand WHISK-ers-on KIT-tens (rest).” ONE-two-three ONE-two-three ONEtwo-threeONE-two-three.As with pitch, small-integer ratios of durations are the most common,and there is accumulating evidence that they are easier to process neurally.But, as Eric Clarke notes, small-integer ratios are almost neverfound in samples of real music. This indicates that there is a quantizationprocess—equalizing durations—occurring during our neural processingof musical time. Our brains treat durations that are similar as beingequal, rounding some up and some down in order to treat them as simpleinteger ratios such as 2:1, 3:1 and 4:1. Some musics use more complex ratiosthan these; Chopin and Beethoven use nominal ratios of 7:4 and and5:4 in some of their piano works, in which seven or five notes are playedwith one hand while the other hand plays four. As with pitch, any ratio is

66 This Is Your Brain on Musictheoretically possible, but there are limitations to what we can perceiveand remember, and there are limitations based on style and convention.The three most common meters in Western music are: 4/4, 2/4, and 3/4.Other rhythmic groupings exist, such as 5/4, 7/4, and 9/4. A somewhatcommon meter is 6/8, in which we count six beats to a measure, andeach eighth note gets one beat. This is similar to 3/4 waltz time, the differencebeing that the composer intends for the musicians to “feel” themusic in groups of six rather than groups of three, and for the underlyingpulse to be the shorter-duration eighth note rather than a quarter note.This points to the hierarchy that exists in musical groupings. It is possibleto count 6/8 as two groups of 3/8 (ONE-two-three ONE-two-three) oras one group of six (ONE-two-three-FOUR-five-six) with a secondary accenton the fourth beat, and to most listeners these are uninterestingsubtleties that only concern a performer. But there may be brain differences.We know that there are neural circuits specifically related to detectingand tracking musical meter, and we know that the cerebellum isinvolved in setting an internal clock or timer that can synchronize withevents that are out-there-in-the-world. No one has yet done the experimentto see if 6/8 and 3/4 have different neural representations, but becausemusicians truly treat them as different, there is a high probabilitythat the brain does also. A fundamental principle of cognitive neuroscienceis that the brain provides the biological basis for any behaviorsor thoughts that we experience, and so at some level there must be neuraldifferentiation wherever there is behavioral differentiation.Of course, 4/4 and 2/4 time are easy to walk to, dance to, or march tobecause (since they are even numbers) you always end up with thesame foot hitting the floor on a strong beat. Three-quarter is less naturalto walk to; you’ll never see a military outfit or infantry division marchingto 3/4. Five-quarter time is used once in a while, the most famous examplesbeing Lalo Shiffrin’s theme from Mission: Impossible, and the DaveBrubeck song “Take Five.” As you count the pulse and tap your foot tothese songs, you’ll see that the basic rhythms group into fives: ONE-twothree-four-five,ONE-two-three-four-five. There is a secondary strongbeat in Brubeck’s composition on the four: ONE-two-three-FOUR-five. In

Foot Tapping 67this case, many musicians think of 5/4 beats as consisting of alternating3/4 and 2/4 beats. In “Mission: Impossible,” there is no clear subdivisionof the five. Tchaikovsky uses 5/4 time for the second movement of hisSixth Symphony. Pink Floyd used 7/4 for their song “Money,” as did PeterGabriel for “Salisbury Hill”; if you try to tap your foot or count along,you’ll need to count seven between each strong beat.I left discussion of loudness for almost-last, because there really isn’tmuch to say about loudness in terms of definition that most people don’talready know. One counterintuitive point is that loudness is, like pitch,an entirely psychological phenomenon, that is, loudness doesn’t exist inthe world, it only exists in the mind. And this is true for the same reasonthat pitch only exists in the mind. When you’re adjusting the output ofyour stereo system, you’re technically increasing the amplitude of thevibration of molecules, which in turn is interpreted as loudness by ourbrains. The point here is that it takes a brain to experience what we call“loudness.” This may seem largely like a semantic distinction, but it isimportant to keep our terms straight. Several odd anomalies exist in themental representation of amplitude, such as loudnesses not being additivethe way that amplitudes are (loudness, like pitch, is logarithmic), orthe phenomenon that the pitch of a sinusoidal tone varies as a functionof its amplitude, or the finding that sounds can appear to be louder thanthey are when they have been electronically processed in certain ways—such as dynamic range compression—that are often done in heavy metalmusic.Loudness is measured in decibels (named after Alexander GrahamBell and abbreviated dB) and it is a dimensionless unit like percent; itrefers to a ratio of two sound levels. In this sense, it is similar to talkingabout musical intervals, but not to talking about note names. The scaleis logarithmic, and doubling the intensity of a sound source results in a3 dB increase in sound. The logarithmic scale is useful for discussingsound because of the ear’s extraordinary sensitivity: The ratio betweenthe loudest sound we can hear without causing permanent damage andthe softest sound we can detect is a million to one, when measured as

68 This Is Your Brain on Musicsound-pressure levels in the air; on the dB scale this is 120 dB. The rangeof loudnesses we can perceive is called the dynamic range. Sometimescritics talk about the dynamic range that is achieved on a high-qualitymusic recording; if a record has a dynamic range of 90 dB, it means thatthe difference between the softest parts on the record and the loudestparts is 90 dB—considered high fidelity by most experts, and beyond thecapability of most home audio systems.Our ears compress sounds that are very loud in order to protect thedelicate components of the middle and inner ear. Normally, as soundsget louder in the world, our perception of the loudness increases proportionatelyto them. But when sounds are really loud, a proportional increasein the signal transmitted by the eardrum would cause irreversibledamage. The compression of the sound levels—of the dynamic range—means that large increases in sound level in the world create muchsmaller changes of level in our ears. The inner hair cells have a dynamicrange of 50 decibels (dB) and yet we can hear over a 120 dB dynamicrange. For every 4 dB increase in sound level, a 1 dB increase is transmittedto the inner hair cells. Most of us can detect when this compressionis taking place; compressed sounds have a different quality.Acousticians have developed a way to make it easy to talk aboutsound levels in the environment—because dBs express a ratio betweentwo values, they chose a standard reference level (20 micropascals ofsound pressure) which is approximately equal to the threshold of humanhearing for most healthy people—the sound of a mosquito flying ten feetaway. To avoid confusion, when decibels are being used to reflect thisreference point of sound pressure level, we refer to them as dB (SPL).Here are some landmarks for sound levels, expressed in dB (SPL):0 dB Mosquito flying in a quiet room, ten feet away from yourears20 dB A recording studio or a very quiet executive office35 dB A typical quiet office with the door closed and computersoff50 dB Typical conversation in a room

Foot Tapping 6975 dB Typical, comfortable music listening level in headphones100–105 dB Classical music or opera concert during loud passages;some portable music players go to 105 dB110 dB A jackhammer three feet away120 dB A jet engine heard on the runway from three hundred feetaway; a typical rock concert126–130 dB Threshold of pain and damage; a rock concert by the Who(note that 126 dB is four times as loud as 120 dB)180 dB Space shuttle launch250–275 dB Center of a tornado; volcanic eruptionConventional foam insert earplugs can block about 25 dB of sound,although they do not do so across the entire frequency range. Earplugsat a Who concert can minimize the risk of permanent damage by bringingdown the levels that reach the ear close to 100–110 dB (SPL). Theover-the-ear type of ear protector worn at rifle firing ranges and by airportlanding personnel is often supplemented by in-the-ear plugs to affordmaximum protection.A lot of people like really loud music. Concertgoers talk about a specialstate of consciousness, a sense of thrills and excitement, when themusic is really loud—over 115 dB. We don’t yet know why this is so. Partof the reason may be related to the fact that loud music saturates the auditorysystem, causing neurons to fire at their maximum rate. Whenmany, many neurons are maximally firing, this could cause an emergentproperty, a brain state qualitatively different from when they are firing atnormal rates. Still, some people like loud music, and some people don’t.Loudness is one of the seven major elements of music along withpitch, rhythm, melody, harmony, tempo, and meter. Very tiny changes inloudness have a profound effect on the emotional communication of music.A pianist may play five notes at once and make one note only slightlylouder than the others, causing it to take on an entirely different role inour overall perception of the musical passage. Loudness is also an importantcue to rhythms, as we saw above, and to meter, because it is theloudness of notes that determines how they group rhythmically.

70 This Is Your Brain on Music* * *Now we have come full circle and return to the broad subject of pitch.Rhythm is a game of expectation. When we tap our feet we are predictingwhat is going to happen in the music next. We also play a game of expectationsin music with pitch. Its rules are key and harmony. A musicalkey is the tonal context for a piece of music. Not all musics have a key.African drumming, for instance, doesn’t, nor does the twelve-tone musicof contemporary composers such as Schönberg. But virtually all of themusic we listen to in Western culture—from commercial jingles on theradio to the most serious symphony by Bruckner, from the gospel musicof Mahalia Jackson to the punk of the Sex Pistols—has a central set ofpitches that it comes back to, a tonal center, the key. The key can changeduring the course of the song (called modulation), but by definition, thekey is generally something that holds for a relatively long period of timeduring the course of the song, typically on the order of minutes.If a melody is based on the C major scale, for example, we generallysay that the melody is “in the key of C.” This means that the melody hasa momentum to return to the note C, and that even if it doesn’t end on aC, the note C is what listeners are keeping in their minds as the dominantand focal note of the entire piece. The composer may temporarily usenotes from outside the C major scale, but we recognize those as departures—somethinglike a quick edit in a movie to a parallel scene or aflashback, in which we know that a return to the main plotline is imminentand inevitable. (For a more detailed look at music theory see Appendix2.)The attribute of pitch in music functions within a scale or a tonal/harmoniccontext. A note doesn’t always sound the same to us every timewe hear it: We hear it within the context of a melody and what has comebefore, and we hear it within the context of the harmony and chords thatare accompanying it. We can think of it like flavor: Oregano tastes goodwith eggplant or tomato sauce, maybe less good with banana pudding.Cream takes on a different gustatory meaning when it is on top of strawberriesfrom when it is in coffee or part of a creamy garlic salad dressing.In “For No One” by the Beatles, the melody is sung on one note for

Foot Tapping 71two measures, but the chords accompanying that note change, giving ita different mood and a different sound. The song “One Note Samba” byAntonio Carlos Jobim actually contains many notes, but one note is featuredthroughout the song with changing chords accompanying it, andwe hear a variety of different shades of musical meaning as this unfolds.In some chordal contexts, the note sounds bright and happy, in others,pensive. Another thing that most of us are expert in, even if we are nonmusicians,is recognizing familiar chord progressions, even in the absenceof the well-known melody. Whenever the Eagles play this chordsequence in concertB minor / F-sharp major / A major / E major / G major / D major /E minor / F-sharp majorthey don’t have to play more than three chords before thousands of nonmusicianfans in the audience know that they are going to play “HotelCalifornia.” And even as they have changed the instrumentation over theyears, from electric to acoustic guitars, from twelve-string to six-stringguitars, people recognize those chords; we even recognize them whenthey’re played by an orchestra coming out of cheap speakers in a Muzakversion in the dentist’s office.Related to the topic of scales and major and minor is the topic oftonal consonance and dissonance. Some sounds strike us as unpleasant,although we don’t always know why. Fingernails screeching on a chalkboardare a classic example, but this seems to be true only for humans;monkeys don’t seem to mind (or at least in the one experiment that wasdone, they like that sound as much as they like rock music). In music,some people can’t stand the sound of distorted electric guitars; otherswon’t listen to anything else. At the harmonic level—that is, the level ofthe notes, rather than the timbres involved—some people find particularintervals or chords particularly unpleasant. Musicians refer to thepleasing-sounding chords and intervals as consonant and the unpleasingones as dissonant. A great deal of research has focused on the problemof why we find consonant some intervals and not others, and there is

72 This Is Your Brain on Musiccurrently no agreement about this. So far, we’ve been able to figure outthat the brain stem and the dorsal cochlear nucleus—structures that areso primitive that all vertebrates have them—can distinguish betweenconsonance and dissonance; this distinction happens before the higherlevel, human brain region—the cortex—gets involved.Although the neural mechanisms underlying consonance and dissonanceare debated, there is widespread agreement about some of the intervalsthat are deemed consonant. A unison interval—the same noteplayed with itself—is deemed consonant, as is an octave. These createsimple integer frequencies ratios of 1:1 and 2:1 respectively. (From anacoustics standpoint, half of the peaks in the waveform for octaves lineup with each other perfectly, the other half fall exactly in between twopeaks.) Interestingly, if we divide the octave precisely in half, the intervalwe end up with is called a tritone and most people find it the most disagreeableinterval possible. Part of the reason for this may be related tothe fact that the tritone does not come from a simple integer ratio, its ratiobeing 43:32. We can look at consonance from an integer ratio perspective.A ratio of 3:1 is a simple integer ratio, and that defines twooctaves. A ratio of 3:2 is also a simple integer ratio, and that defines theinterval of a perfect fifth. This is the distance between, for example, Cand the G above it. The distance from that G to the C above it forms aninterval of a perfect fourth, and its frequency ratio is 4:3.The particular notes found in our major scale trace their roots back tothe ancient Greeks and their notions of consonance. If we start with anote C and simply add the interval of a perfect fifth to it iteratively, weend up generating a set of frequencies that are very close to the currentmajor scale: C - G - D - A - E - B - F-sharp - C-sharp - G-sharp - D-sharp -A-sharp - E-sharp (or F ), and then back to C. This is known as the circleof fifths because after going through the cycle, we end up back at thenote we started on. Interestingly, if we follow the overtone series, we cangenerate frequencies that are somewhat close to the major scale as well.A single note cannot, by itself, be dissonant, but it can sound dissonantagainst the backdrop of certain chords, particularly when the chordimplies a key that the single note is not part of. Two notes can sound dis-

Foot Tapping 73sonant together, both when played simultaneously or in sequence, if thesequence does not conform to the customs we have learned that go withour musical idioms. Chords can also sound dissonant, especially whenthey are drawn from outside the key that has been established. Bringingall these factors together is the task of the composer. Most of us are verydiscriminating listeners, and when the composer gets the balance justslightly wrong, our expectations have been betrayed more than we canstand, and we switch radio stations, pull off the earphones, or just walkout of the room.I’ve reviewed the major elements that go into music: pitch, timbre, key,harmony, loudness, rhythm, meter, and tempo. Neuroscientists deconstructsound into its components to study selectively which brain regionsare involved in processing each of them, and musicologists discuss theirindividual contributions to the overall aesthetic experience of listening.But music—real music—succeeds or fails because of the relationshipamong these elements. Composers and musicians rarely treat these in totalisolation; they know that changing a rhythm may also require changingpitch or loudness, or the chords that accompany that rhythm. Oneapproach to studying the relationship between these elements traces itsorigins back to the late 1800s and the Gestalt psychologists.In 1890, Christian von Ehrenfels was puzzled by something all of ustake for granted and know how to do: melodic transposition. Transpositionis simply singing or playing a song in a different key or with differentpitches. When we sing “Happy Birthday” we just follow along withthe first person who started singing, and in most cases, this person juststarts on any note that she feels like. She might even have started on apitch that is not a recognized note of the musical scale, falling between,say, C and C-sharp, and almost no one would notice or care. Sing “HappyBirthday” three times in a week and you might be singing three completelydifferent sets of pitches. Each version of the song is called atransposition of the others.The Gestalt psychologists—von Ehrenfels, Max Wertheimer, WolfgangKöhler, Kurt Koffka, and others—were interested in the problem of

74 This Is Your Brain on Musicconfigurations, that is, how it is that elements come together to formwholes, objects that are qualitatively different from the sum of theirparts, and cannot be understood in terms of their parts. The wordGestalt has entered the English language to mean a unified whole form,applicable to both artistic and nonartistic objects. One can think of a suspensionbridge as a Gestalt. The functions and utility of the bridge arenot easily understood by looking at pieces of cable, girders, bolts, andsteel beams; it is only when they come together in the form of a bridgethat we can apprehend how a bridge is different from, say, a constructioncrane that might be made out of the same parts. Similarly, in painting,the relationship between elements is a critical aspect of the finalartistic product. The classic example is a face—the Mona Lisa wouldnot be what it is if the eyes, nose, and mouth were painted entirely asthey are but were scattered across the canvas in a different arrangement.The Gestaltists wondered how it is that a melody—composed of a setof specific pitches—could retain its identity, its recognizability, evenwhen all of its pitches are changed. Here was a case for which they couldnot generate a satisfying theoretical explanation, the ultimate triumph ofform over detail, of the whole over the parts. Play a melody using any setof pitches, and so long as the relation between those pitches is held constant,it is the same melody. Play it on different instruments and peoplestill recognize it. Play it at half speed or double speed, or impose all ofthese transformations at the same time, and people still have no troublerecognizing it as the original song. The influential Gestalt school wasformed to address this particular question. Although they never answeredit, they did go on to contribute enormously to our understandingof how objects in the visual world are organized, through a set of rulesthat are taught in every introductory psychology class, the “Gestalt Principlesof Grouping.”Albert Bregman, a cognitive psychologist at McGill University, hasperformed a number of experiments over the last thirty years to developa similar understanding of grouping principles for sound. The music theoristFred Lerdahl from Columbia University and the linguist Ray Jack-

Foot Tapping 75endoff from Brandeis University (now at Tufts University) tackled theproblem of describing a set of rules, similar to the rules of grammar inspoken language, that govern musical composition, and these includegrouping principles for music. The neural basis for these principles hasnot been competely worked out, but through a series of clever behavioralexperiments we have learned a great deal about the phenomenologyof the principles.In vision, grouping refers to the way in which elements in the visualworld combine or stay separate from one another in our mental image ofthe world. Grouping is partly an automatic process, which means thatmuch of it happens rapidly in our brains and without our consciousawareness. It has been described simply as the problem of “what goeswith what” in our visual field. Hermann von Helmholtz, the nineteenthcenturyscientist who taught us much of what we now accept as thefoundations of auditory science, described it as an unconscious processthat involved inferencing, or logical deductions about what objects inthe world are likely to go together based on a number of features or attributesof the objects.If you’re standing on a mountaintop overlooking a varied landscape,you might describe seeing two or three other mountains, a lake, a valley,a fertile plain, and a forest. Although the forest is composed of hundredsor thousands of trees, the trees form a perceptual group, distinct fromother things we see, not necessarily because of our knowledge of forests,but because the trees share similar properties of shape, size, andcolor—at least when they stand in opposition to fertile plains, lakes, andmountains. But if you’re in the center of a forest with a mixture of aldertrees and pines, the smooth white bark of the alders will cause them to“pop out” as a separate group from the craggy dark-barked pines. If I putyou in front of one tree and ask you what you see, you might start to focuson details of that tree: bark, branches, leaves (or needles), insects,and moss. When looking at a lawn, most of us don’t typically see individualblades of grass, although we can if we focus our attention on them.Grouping is a hierarchical process and the way in which our brains form

76 This Is Your Brain on Musicperceptual groups is a function of a great many factors. Some groupingfactors are intrinsic to the objects themselves—shape, color, symmetry,contrast, and principles that address the continuity of lines and edges ofthe object. Other grouping factors are psychological, that is, mind based,such as what we’re consciously trying to pay attention to, what memorieswe have of this or similar objects, and what our expectations areabout how objects should go together.Sounds group too. This is to say that while some group with one another,others segregate from each other. Most people can’t isolate thesound of one of the violins in an orchestra from the others, or one of thetrumpets from the others—they form a group. In fact, the entire orchestracan form a single perceptual group—called a stream in Bregman’sterminology—depending on the context. If you’re at an outdoor concertwith several ensembles playing at once, the sounds of the orchestra infront of you will cohere into a single auditory entity, separate from theother orchestras behind you and off to the side. Through an act of volition(attention) you can then focus on just the violins of the orchestra infront of you, just as you can follow a conversation with the person nextto you in a crowded room full of conversations.One case of auditory grouping is the way that the many differentsounds emanating from a single musical instrument cohere into a perceptof a single instrument. We don’t hear the individual harmonics of anoboe or of a trumpet, we hear an oboe or we hear a trumpet. This is allthe more remarkable if you imagine an oboe and a trumpet playing at thesame time. Our brains are capable of analyzing the dozens of differentfrequencies reaching our ears, and putting them together in just the rightway. We don’t have the impression of dozens of disembodied harmonics,nor do we hear just a single hybrid instrument. Rather, our brains constructfor us separate mental images of an oboe and of a trumpet, andalso of the sound of the two of them playing together—the basis for ourappreciation of timbral combinations in music. This is what Pierce wastalking about when he marveled at the timbres of rock music—thesounds that an electric bass and an electric guitar made when they wereplaying together—two instruments, perfectly distinguishable from one

Foot Tapping 77another, and yet simultaneously creating a new sonic combination thatcan be heard, discussed, and remembered.Our auditory system exploits the harmonic series in grouping soundstogether. Our brains coevolved in a world in which many of the soundsthat our species encountered—over the tens of thousands of years ofevolutionary history—shared certain acoustical properties with one another,including the harmonic series as we now understand it. Throughthis process of “unconscious inference” (as von Helmholtz called it), ourbrains assume that it is highly unlikely that several different soundsources are present, each producing a single component of the harmonicseries. Rather, our brains use the “likelihood principle” that it must be asingle object producing these harmonic components. All of us can makethese inferences, even those of us who can’t identify or name the instrument“oboe” as distinct, from, say, a clarinet or bassoon, or even a violin.But just as people who don’t know the names of the notes in the scalecan still tell when two different notes are being played as opposed to thesame notes, nearly all of us—even lacking a knowledge of the names ofmusical instruments—can tell when there are two different instrumentsplaying. The way in which we use the harmonic series to group soundsgoes a long way toward explaining why we hear a trumpet rather the individualovertones that impinge on our ears—they group together likeblades of grass that give us the impression of “lawn.” It also explainshow we can distinguish a trumpet from an oboe when they’re each playingdifferent notes—different fundamental frequencies give rise to a differentset of overtones, and our brains are able to effortlessly figure outwhat goes with what, in a computational process that resembles what acomputer might do. But it doesn’t explain how we might be able to distinguisha trumpet from an oboe when they’re playing the same note,because then the overtones are very nearly the same in frequency (althoughwith different amplitudes characteristic of the instrument). Forthat, the auditory system relies on a principle of simultaneous onsets.Sounds that begin together—at the same instant in time—are perceivedas going together, in the grouping sense. And it has been known since thetime Wilhelm Wundt set up the first psychological laboratory in the 1870s

78 This Is Your Brain on Musicthat our auditory system is exquisitely sensitive to what constitutes simultaneousin this sense, being able to detect differences in onset timesas short as a few milliseconds.So when a trumpet and an oboe are playing the same note at the sametime, our auditory system is able to figure out that two different instrumentsare playing because the full sound spectrum—the overtoneseries—for one instrument begins perhaps a few thousandths of a secondbefore the sound spectrum for the other. This is what is meant by agrouping process that not only integrates sounds into a single object, butsegregates them into different objects.This principle of simultaneous onsets can be thought of more generallyas a principle of temporal positioning. We group all the sounds thatthe orchestra is making now as opposed to those it will make tomorrownight. Time is a factor in auditory grouping. Timbre is another, and this iswhat makes it so difficult to distinguish one violin from several that areall playing at once, although expert musicians and conductors can trainthemselves to do this. Spatial location is a grouping principle, as our earstend to group together sounds that come from the same relative positionin space. We are not very sensitive to location in the up-down plane, butwe are very sensitive to position in the left-right plane and somewhatsensitive to distance in the forward-back plane. Our auditory system assumesthat sounds coming from a distinct location in space are probablypart of the same object-in-the-world. This is one of the explanations forwhy we can follow a conversation in a crowded room relatively easily—our brains are using the cues of spatial location of the person we’re conversingwith to filter out other conversations. It also helps that theperson we’re speaking to has a unique timbre—the sound of his voice—that works as an additional grouping cue.Amplitude also affects grouping. Sounds of a similar loudness grouptogether, which is how we are able to follow the different melodies inMozart’s divertimenti for woodwinds. The timbres are all very similar,but some instruments are playing louder than others, creating differentstreams in our brains. It is as though a filter or sieve takes the sound of

Foot Tapping 79the woodwind ensemble and separates it out into different parts dependingon what part of the loudness scale they are playing in.Frequency, or pitch, is a strong and fundamental consideration ingrouping. If you’ve ever heard a Bach flute partita, there are typically momentswhen some flute notes seem to “pop out” and separate themselvesfrom one another, particularly when the flautist is playing a rapid passage—theauditory equivalent of a “Where’s Waldo?” picture. Bach knewabout the ability of large frequency differences to segregate sounds fromone another—to block or inhibit grouping—and he wrote parts that includedlarge leaps in pitch of a perfect fifth or more. The high notes, alternatingwith a succession of lower-pitched notes, create a separatestream and give the listener the illusion of two flutes playing when thereis only one. We hear the same thing in many of the violin sonatas by Locatelli.Yodelers can accomplish the same effect with their voices, bycombining pitch and timbral cues; when a male yodeler jumps into hisfalsetto register, he is creating both a distinct timbre and, typically, alarge jump in pitch, causing the higher notes to again separate out into adistinct, perceptual stream, giving the illusion of two people singing interleavedparts.We now know that the neurobiological subsystems for the differentattributes of sound that I’ve described separate early on, at low levels ofthe brain. This suggests that grouping is carried out by general mechanismsworking somewhat independently of one another. But it is alsoclear that the attributes work with or against each other when they combinein particular ways, and we also know that experience and attentioncan have an influence on grouping, suggesting that portions of the groupingprocess are under conscious, cognitive control. The ways in whichconscious and unconscious processes work together—and the brainmechanisms that underlie them—are still being debated, but we’ve comea long way toward understanding them in the last ten years. We’ve finallygotten to the point where we can pinpoint specific areas of the brain thatare involved in particular aspects of music processing. We even think weknow which part of the brain causes you to pay attention to things.

80 This Is Your Brain on MusicHow are thoughts formed? Are memories “stored” in a particular partof the brain? Why do songs sometimes get stuck in your head and youcan’t get them out? Does your brain take some sick pleasure in slowlydriving you crazy with inane commercial jingles? I take up these andother ideas in the coming chapters.

3. Behind the CurtainMusic and the Mind MachineFor cognitive scientists, the word mind refers to that part of each ofus that embodies our thoughts, hopes, desires, memories, beliefs,and experiences. The brain, on the other hand, is an organ of the body,a collection of cells and water, chemicals and blood vessels, that residesin the skull. Activity in the brain gives rise to the contents of the mind.Cognitive scientists sometimes make the analogy that the brain is like acomputer’s CPU, or hardware, while the mind is like the programs orsoftware running on the CPU. (If only that were literally true and wecould just run out to buy a memory upgrade.) Different programs canrun on what is essentially the same hardware—different minds can arisefrom very similar brains.Western culture has inherited a tradition of dualism from RenéDescartes, who wrote that the mind and the brain are two entirely separatethings. Dualists assert that the mind preexisted, before you wereborn, and that the brain is not the seat of thought—rather, it is merely aninstrument of the mind, helping to implement the mind’s will, move muscles,and maintain homeostasis in the body. To most of us, it certainlyfeels as though our minds are something unique and distinctive, separatefrom just a bunch of neurochemical processes. We have a feeling of whatit is like to be me, what it is like to be me reading a book, and what it is

82 This Is Your Brain on Musiclike to think about what it is like to be me. How can me be reduced so unceremoniouslyto axons, dendrites, and ion channels? It feels like we aresomething more.But this feeling could be an illusion, just as it certainly feels as thoughthe earth is standing still, not spinning around on its axis at a thousandmiles per hour. Most scientists and contemporary philosophers believethat the brain and mind are two parts of the same thing, and some believethat the distinction itself is flawed. The dominant view today is thatthat the sum total of your thoughts, beliefs, and experiences is representedin patterns of firings—electrochemical activity—in the brain. Ifthe brain ceases to function, the mind is gone, but the brain can still exist,thoughtless, in a jar in someone’s laboratory.Evidence for this comes from neuropsychological findings of regionalspecificity of function. Sometimes, as a result of stroke (a blockage ofblood vessels in the brain that leads to cell death), tumors, head injury,or other trauma, an area of the brain becomes damaged. In many of thesecases, damage to a specific brain region leads to a loss of a particularmental or bodily function. When dozens or hundreds of cases show lossof a specific function associated with a particular brain region, we inferthat this brain region is somehow involved in, or perhaps responsible for,that function.More than a century of such neuropsychological investigation hasallowed us to make maps of the brain’s areas of function, and to localizeparticular cognitive operations. The prevailing view of the brain isthat it is a computational system, and we think of the brain as a type ofcomputer. Networks of interconnected neurons perform computationson information and combine their computations in ways that lead tothoughts, decisions, perceptions, and ultimately consciousness. Differentsubsystems are responsible for different aspects of cognition. Damageto an area of the brain just above and behind the left ear—Wernicke’sarea—causes difficulty in understanding spoken language; damage to aregion at the very top of the head—the motor cortex—causes difficultymoving your fingers; damage to an area in the center of the brain—thehippocampal complex—can block the ability to form new memories,

Behind the Curtain 83while leaving old memories intact. Damage to an area just behind yourforehead can cause dramatic changes in personality—it can rob aspectsof you from you. Such localization of mental function is a strong scientificargument for the involvement of the brain in thought, and the thesisthat thoughts come from the brain.We have known since 1848 (and the medical case of Phineas Gage)that the frontal lobes are intimately related to aspects of self and personality.Yet even one hundred and fifty years later, most of what we cansay about personality and neural structures is vague and quite general.We have not located the “patience” region of the brain, nor the “jealousy”or “generous” regions, and it seems unlikely that we ever will. The brainhas regional differentiation of structure and function, but complex personalityattributes are no doubt distributed widely throughout the brain.The human brain is divided up into four lobes—the frontal, temporal,parietal, and occipital—plus the cerebellum. We can make some grossgeneralizations about function, but in fact behavior is complex and notreadily reducible to simple mappings. The frontal lobe is associated withplanning, and with self-control, and with making sense out of the denseand jumbled signals that our senses receive—the so-called “perceptualorganization” that the Gestalt psychologists studied. The temporal lobeis associated with hearing and memory. The parietal lobe is associatedwith motor movements and spatial skill, and the occipital lobe with vision.The cerebellum is involved in emotions and the planning of movements,and is the evolutionarily oldest part of our brain; even manyanimals, such as reptiles, that lack the “higher” brain region of the cortexstill have a cerebellum. The surgical separation of a portion of the frontallobe, the prefrontal cortex, from the thalamus is called a lobotomy. Sowhen the Ramones sang “Now I guess I’ll have to tell ’em/That I gotno cerebellum” in their song “Teenage Lobotomy” (words and music byDouglas Colvin, John Cummings, Thomas Erdely, and Jeffrey Hyman)they were not being anatomically accurate, but for the sake of artistic license,and for creating one of the great rhymes in rock music, it is hardto begrudge them that.Musical activity involves nearly every region of the brain that we

84 This Is Your Brain on Musicknow about, and nearly every neural subsystem. Different aspects of themusic are handled by different neural regions—the brain uses functionalsegregation for music processing, and employs a system of feature detectorswhose job it is to analyze specific aspects of the musical signal,such as pitch, tempo, timbre, and so on. Some of the music processinghas points in common with the operations required to analyze othersounds; understanding speech, for example, requires that we segmenta flurry of sounds into words, sentences, and phrases, and that we beable to understand aspects beyond the words, such as sarcasm (isn’tthat interesting). Several different dimensions of a musical sound needto be analyzed—usually involving several quasi-independent neuralprocesses—and they then need to be brought together to form a coherentrepresentation of what we’re listening to.Listening to music starts with subcortical (below-the-cortex) structures—thecochlear nuclei, the brain stem, the cerebellum—and thenmoves up to auditory cortices on both sides of the brain. Trying to followalong with music that you know—or at least music in a style you’re familiarwith, such as baroque or blues—recruits additional regions of thebrain, including the hippocampus—our memory center—and subsectionsof the frontal lobe, particularly a region called inferior frontal cortex,which is in the lowest parts of the frontal lobe, i.e., closer to yourchin than to the top of your head. Tapping along with music, either actuallyor just in your mind, involves the cerebellum’s timing circuits. Performingmusic—regardless of what instrument you play, or whether yousing, or conduct—involves the frontal lobes again for the planning ofyour behavior, as well as the motor cortex in the parietal lobe just underneaththe top of your head, and the sensory cortex, which providesthe tactile feedback that you have pressed the right key on your instrument,or moved the baton where you thought you did. Reading music involvesthe visual cortex, in the back of your head in the occipetal lobe.Listening to or recalling lyrics invokes language centers, including Broca’sand Wernicke’s area, as well as other language centers in the temporaland frontal lobes.

Behind the Curtain 85At a deeper level, the emotions we experience in response to musicinvolve structures deep in the primitive, reptilian regions of the cerebellarvermis, and the amygdala—the heart of emotional processing in thecortex. The idea of regional specificity is evident in this summary but acomplementary principle applies as well, that of distribution of function.The brain is a massively parallel device, with operations distributedwidely throughout. There is no single language center, nor is there a singlemusic center. Rather, there are regions that peform component operations,and other regions that coordinate the bringing together of thisinformation. Finally, we have discovered only recently that the brain hasa capacity for reorganization that vastly exceeds what we thought before.This ability is called neuroplasticity, and in some cases, it suggeststhat regional specificity may be temporary, as the processing centers forimportant mental functions actually move to other regions after traumaor brain damage.It is difficult to appreciate the complexity of the brain because the numbersare so huge they go well beyond our everyday experience (unlessyou are a cosmologist). The average brain consists of one hundred billion(100,000,000,000) neurons. Suppose each neuron was one dollar,and you stood on a street corner trying to give dollars away to people asthey passed by, as fast as you could hand them out—let’s say one dollarper second. If you did this twenty-four hours a day, 365 days a year, withoutstopping, and if you had started on the day that Jesus was born, youwould by the present day only have gone through about two thirds ofyour money. Even if you gave away hundred-dollar bills once a second,it would take you thirty-two years to pass them all out. This is a lot ofneurons, but the real power and complexity of the brain (and of thought)come through their connections.Each neuron is connected to other neurons—usually one thousand toten thousand others. Just four neurons can be connected in sixty-threeways, or not at all, for a total of sixty-four possibilities. As the number ofneurons increases, the number of possible connections grows exponen-

86 This Is Your Brain on Musictially (the formula for the way that n neurons can be connected to eachother is 2 (n*(n-1)/2) ):For 2 neurons there are 2 possibilities for how they can be connectedFor 3 neurons there are 8 possibilitiesFor 4 neurons there are 64 possibilitiesFor 5 neurons there are 1,024 possibilitiesFor 6 neurons there are 32,768 possibilitiesThe number of combinations becomes so large that it is unlikely thatwe will ever understand all the possible connections in the brain, orwhat they mean. The number of combinations possible—and hence thenumber of possible different thoughts or brain states each of us canhave—exceeds the number of known particles in the entire known universe.Similarly, you can see how it is that all the songs we have everheard—and all those that will ever be created—could be made up of justtwelve musical notes (ignoring octaves). Each note can go to anothernote, or to itself, or to a rest, and this yields twelve possibilities. But eachof those possibilities yields twelve more. When you factor in rhythm—each note can take on one of many different note lengths—the numberof possibilities grows very, very rapidly.Much of the brain’s computational power comes from this enormouspossibility for interconnection, and much of it comes from the fact thatbrains are parallel processing machines, rather than serial processors. Aserial processor is like an assembly line, handling each piece of informationas it comes down the mental conveyor belt, performing some operationon that piece of information, and then sending it down the line forthe next operation. Computers work like this. Ask a computer to downloada song from a Web site, tell you the weather in Boise, and save a fileyou’ve been working on, and it will do them one at a time; it does thingsso fast that it can seem as though it is doing them at the same time—inparallel—but it isn’t. Brains, on the other hand, can work on many things

Behind the Curtain 87at once, overlapping and in parallel. Our auditory system processessound in this way—it doesn’t have to wait to find out what the pitch of asound is to know where it is coming from; the neural circuits devoted tothese two operations are trying to come up with answers at the sametime. If one neural circuit finishes its work before another, it just sendsits information to other connected brain regions and they can begin usingit. If late-arriving information that affects an interpretation of whatwe’re hearing comes in from a separate processing circuit, the brain can“change its mind” and update what it thinks is out there. Our brains areupdating their opinions all the time—particularly when it comes to perceivingvisual and auditory stimuli—hundreds of times per second, andwe don’t even know it.Here’s an analogy to convey how neurons connect to each other.Imagine that you’re sitting home alone one Sunday morning. You don’tfeel much of one way or another—you’re not particularly happy, not particularlysad, neither angry, excited, jealous, nor tense. You feel more orless neutral. You have a bunch of friends, a network of them, and you cancall any of them on the phone. Let’s say that each of your friends is ratherone dimensional and that they can exert a great influence on your mood.You know, for example, that if you telephone your friend Hannah she’llput you in a happy mood. Whenever you talk to Sam it makes you sad,because the two of you had a third friend who died and Sam reminds youof that. Talking to Carla makes you calm and serene, because she has asoothing voice and you’re reminded of the times you sat in a beautifulforest clearing with her, soaking up the sun and meditating. Talking toEdward makes you feel energized; talking to Tammy makes you feeltense. You can pick up your telephone and connect to any of thesefriends and induce a certain emotion.You might have hundreds or thousands of these one-dimensionalfriends, each capable of evoking a particular memory, experience, ormood state. These are your connections. Accessing them causes you tochange your mood, or state. If you were to talk to Hannah and Sam at thesame time, or one right after the other, Hannah would make you feel

88 This Is Your Brain on Musichappy, Sam would make you feel sad, and in the end you’d be back towhere you were—neutral. But we can add an additional nuance, whichis the weight or force-of-influence of these connections—how close youfeel to an individual at a particular point in time. That weight determinesthe amount of influence the person will have on you. If you feel twice asclose to Hannah as you do to Sam, talking to Hannah and Sam for anequal amount of time would still leave you feeling happy, although not ashappy as if you had talked to Hannah alone—Sam’s sadness brings youdown, but only halfway from the happiness you gained from talking toHannah.Let’s say that all of these people can talk to one another, and in so doing,their states can be modified to some extent. Although your friendHannah is dispositionally cheery, her cheerfulness can be attenuated bya conversation she has with Sad Sam. If you phone Edward the energizerafter he’s just spoken with Tense Tammy (who has just gotten off thephone with Jealous Justine), Edward may make you feel a new mix ofemotions you’ve never experienced before, a kind of tense jealousy thatyou have a lot of energy to go out and do something about. And any ofthese friends might telephone you at any time, evoking these states inyou as a complex chain of feelings or experiences that has gone around,each one influencing the other, and you, in turn, will leave your emotionalmark on them. With thousands of friends interconnected like this,and a bunch of telephones in your living room ringing off the hook allday long, the number of emotional states you might experience would indeedbe quite varied.It is generally accepted that our thoughts and memories arise fromthe myriad connections of this sort that our neurons make. Not all neuronsare equally active at one time, however—this would cause a cacophonyof images and sensations in our heads (in fact, this is whathappens in epilepsy). Certain groups of neurons—we can call them networks—becomeactive during certain cognitive activities, and they inturn can activate other neurons. When I stub my toe, the sensory receptorsin my toe send signals up to the sensory cortex in my brain. This setsoff a chain of neural activations that causes me to experience pain, with-

Behind the Curtain 89draw my foot from the object I stubbed it against, and that might causemy mouth to open involuntarily and shout “& % @ !”When I hear a car horn, air molecules impinging on my eardrum causeelectrical signals to be sent to my auditory cortex. This causes a cascadeof events that recruits a very different group of neurons than toe stubbing.First, neurons in the auditory cortex process the pitch of the soundso that I can distinguish the car horn from something with a differentpitch like a truck’s air horn, or the air-horn-in-a-can at a football game. Adifferent group of neurons is activated to determine the location fromwhich the sound came. These and other processes invoke a visual orientingresponse—I turn toward the sound to see what made it, and instantaneously,if necessary, I jump back (the result of activity from theneurons in my motor cortex, orchestrated with neurons in my emotionalcenter, the amygdala, telling me that danger is imminent).When I hear Rachmaninoff’s Piano Concerto no. 3, the hair cells inmy cochlea parse the incoming sound into different frequency bands,sending electrical signals to my primary auditory cortex—area A1—telling it what frequencies are present in the signal. Additional regions inthe temporal lobe, including the superior temporal sulcus and the superiortemporal gyrus on both sides of the brain, help to distinguish thedifferent timbres I’m hearing. If I want to label those timbres, the hippocampushelps to retrieve the memory of similar sounds I’ve heardbefore, and then I’ll need to access my mental dictionary—which willrequire using structures found at the junction between the temporal,occipetal, and parietal lobes. So far, these regions are the same ones,although activated in different ways and with different populationsof neurons, that I would use to process the car horn. Whole new populationsof neurons will become active, however, as I attend to pitchsequences (dorsalateral prefrontal cortex, and Brodmann areas 44and 47), rhythms (the lateral cerebellum and the cerebellar vermis), andemotion (frontal lobes, cerebellum, the amygdala, and the nucleusaccumbens—part of a network of structures involved in feelings of pleasureand reward, whether it is through eating, having sex, or listening topleasurable music).

90 This Is Your Brain on MusicTo some extent, if the room is vibrating with the deep sounds of thedouble bass, some of those same neurons that fired when I stubbed mytoe may fire now—neurons sensitive to tactile input. If the car horn hasa pitch of A440, neurons that are set to fire when that frequency is encounteredwill most probably fire, and they’ll fire again when an A440 occursin Rachmaninoff. But my inner mental experience is likely to bedifferent because of the different contexts involved and the differentneural networks that are recruited in the two cases.My experience with oboes and violins is different, and the particularway that Rachmaninoff uses them may cause me to have the oppositereaction to his concerto than I have to the car horn; rather than feelingstartled, I feel relaxed. The same neurons that fire when I feel calm andsafe in my environment may be triggered by the calm parts of the concerto.Through experience, I’ve learned to associate car horns with danger,or at least with someone trying to get my attention. How did this happen?Some sounds are intrinsically soothing while others are frightening.Although there is a great deal of interpersonal variation, we are bornwith a predisposition toward interpreting sounds in particular ways.Abrupt, short, loud sounds tend to be interpreted by many animals as analert sound; we see this when comparing the alert calls of birds, rodents,and apes. Slow onset, long, and quieter sounds tend to be interpreted ascalming, or at least neutral. Think of the sharp sound of a dog’s bark, versusthe soft purring of a cat who sits peacefully on your lap. Composersknow this, of course, and use hundreds of subtle shadings of timbre andnote length to convey the many different emotional shadings of humanexperience.In the “Surprise Symphony” by Haydn (Symphony no. 94 in G Major,second movement, andante), the composer builds suspense by usingsoft violins in the main theme. The softness of the sound is soothing, butthe shortness of the pizzicato accompaniment sends a gentle, contradictorymessage of danger, and together they give a soft sense of suspense.The main melodic idea spans barely more than half an octave, a perfect

Behind the Curtain 91fifth. The melodic contour further suggests complacency—the melodyfirst goes up, then down, then repeats the “up” motif. The parallelism impliedby the melody, the up/down/up, gets the listener ready for another“down” part. Continuing with the soft, gentle violin notes, the maestrochanges the melody by going up—just a little—but holds the rhythmsconstant. He rests on the fifth, a relatively stable tone harmonically. Becausethe fifth is the highest note we’ve encountered so far, we expectthat when the next note comes in, it will be lower—that it will begin thereturn home toward the root (or tonic), and “close the gap” created bythe distance between the tonic and the current note—the fifth. Then,from out of nowhere, Haydn sends us a loud note an octave higher, withthe brash horns and timpani carrying the sound. He has violated our expectationsfor melodic direction, for contour, for timbre, and for loudnessall at once. This is the “Surprise” in the “Surprise Symphony.”This Haydn symphony violates our expectations of how the worldworks. Even someone with no musical knowledge or musical expectationswhatsoever finds the symphony surprising because of this timbraleffect, switching from the soft purring of the violins to the alert call ofhorns and drums. For someone with a musical background, the symphonyviolates expectations that have been formed based on musicalconvention and style. Where do surprises, expectations, and analyses ofthis sort occur in the brain? Just how these operations are carried out inneurons is still something of a mystery, but we do have some clues.Before going any farther, I have to admit a bias in the way I approach thescientific study of minds and brains: I have a definite preference forstudying the mind rather than the brain. Part of my preference is personalrather than professional. As a child I wouldn’t collect butterflieswith the rest of my science class because life—all life—seems sacred tome. And the stark fact about brain research over the course of the lastcentury is that it generally involves poking around in the brains of liveanimals, often our close genetic cousins, the monkeys and apes, andthen killing (they call it “sacrificing”) the animal. I worked for one mis-

92 This Is Your Brain on Musicerable semester in a monkey lab, dissecting the brains of dead monkeysto prepare them for microscopic examination. Every day I had to walkby cages of the ones that were still alive. I had nightmares.At a different level, I’ve always been more fascinated by the thoughtsthemselves, not the neurons that give rise to them. A theory in cognitivescience named functionalism—which many prominent researcherssubscribe to—asserts that similar minds can arise from quite differentbrains, that brains are just the collection of wires and processing modulesthat instantiate thought. Regardless of whether the functionalistdoctrine is true, it does suggest that there are limits to how much we canknow about thought from just studying brains. A neurosurgeon once toldDaniel Dennett (a prominent and persuasive spokesperson for functionalism)that he had operated on hundreds of people and seen hundreds oflive, thinking brains, but he had never seen a thought.When I was trying to decide where to attend graduate school, andwho I wanted to have as a mentor, I was infatuated with the work of ProfessorMichael Posner. He had pioneered a number of ways of lookingat thought processes, among them mental chronometry (the idea thatmuch can be learned about the organization of the mind by measuringhow long it takes to think certain thoughts), ways to investigate thestructure of categories, and the famous Posner Cueing Paradigm, a novelmethod for studying attention. But rumor had it that Posner was abandoningthe mind and had started studying the brain, something I was certainI did not want to do.Although still an undergraduate (albeit a somewhat older one thanusual), I attended the annual meeting of the American Psychological Association,which was held in San Francisco that year, just forty miles upthe road from Stanford, where I was finishing up my B.A. I saw Posner’sname on the program and attended his talk, which was full of slides containingpictures of people’s brains while they were doing one thing or another.After his talk was over he took some questions, then disappearedout a back door. I ran around to the back and saw him way ahead, rushingacross the conference center to get to another talk. I ran to catch upto him. I must have been quite a sight to him! I was out of breath from

Behind the Curtain 93running. Even without the panting, I was nervous meeting one of thegreat legends of cognitive psychology. I had read his textbook in my firstpsychology class at MIT (where I began my undergraduate training beforetransferring to Stanford); my first psychology professor, SusanCarey, spoke of him with what could only be described as reverence inher voice. I can still remember the echoes of her words, reverberatingthrough the lecture hall at MIT: “Michael Posner, one of the smartest andmost creative people I’ve ever met.”I started to sweat, I opened my mouth, and ...nothing. I started“Mmm . . .” All this time we were walking rapidly side by side—he’s a fastwalker—and every two or three steps I’d fall behind again. I stammeredan introduction and said that I had applied to the University of Oregon towork with him. I’d never stuttered before, but I had never been this nervousbefore. “P-p-p-professor P-p-posner, I hear that you’ve shifted yourresearch focus entirely to the b-b-brain—is that true? Because I reallywant to study cognitive psychology with you,” I finally told him.“Well, I am a little interested in the brain these days,” he said. “But Isee cognitive neuroscience as a way to provide constraints for our theoriesin cognitive psychology. It helps us to distinguish whether a modelhas a plausible basis in the underlying anatomy.”Many people enter neuroscience from a background in biology orchemistry and their principal focus is on the mechanisms by which cellscommunicate with each other. To the cognitive neuroscientist, understandingthe anatomy or physiology of the brain may be a challengingintellectual exercise (the brain scientists’ equivalent of a really complicatedcrossword puzzle), but it is not the ultimate goal of the work. Ourgoal is to understand thought processes, memories, emotions, and experiences,and the brain just happens to be the box that all this happens in.To return to the telephone analogy and conversations you might havewith different friends who influence your emotions: If I want to predicthow you’re going to feel tomorrow, it will be of only limited value for meto map the layout of the telephone lines connecting all the differentpeople you know. More important is to understand their individual proclivities:Who is likely to call you tomorrow and what are they likely to

94 This Is Your Brain on Musicsay? How are they apt to make you feel? Of course, to entirely ignore theconnectivity question would be a mistake too. If a line is broken, or ifthere is no evidence of a connection between person A and person B, orif person C can never call you directly but can only influence youthrough person A who can call you directly—all this information providesimportant constraints to a prediction.This perspective influences the way I study the cognitive neuroscienceof music. I am not interested in going on a fishing expedition totry every possible musical stimulus and find out where it occurs in thebrain; Posner and I have talked many times about the current mad rushto map the brain as just so much atheoretical cartography. The point forme isn’t to develop a map of the brain, but to understand how it works,how the different regions coordinate their activity together, how the simplefiring of neurons and shuttling around of neurotransmitters leads tothoughts, laughter, feelings of profound joy and sadness, and how allthese, in turn, can lead us to create lasting, meaningful works of art.These are the functions of the mind, and knowing where they occurdoesn’t interest me unless the where can tell us something about howand why. An assumption of cognitive neuroscience is that it can.My perspective is that, of the infinite number of experiments that arepossible to do, the ones worth doing are those that can lead us to a betterunderstanding of how and why. A good experiment is theoreticallymotivated, and makes clear predictions as to which one of two or morecompeting hypotheses will be supported. An experiment that is likely toprovide support for both sides of a contentious issue is not one worthdoing; science can only move forward by the elimination of false or untenablehypotheses.Another quality of a good experiment is that it is generalizable toother conditions—to people not studied, to types of music not studied,and to a variety of situations. A great deal of behavioral research is conductedon only a small number of people (“subjects” in the experiment),and with very artificial stimuli. In my laboratory we use both musiciansand nonmusicians whenever possible, in order to learn about the broadestcross section of people. And we almost always use real-world music,

Behind the Curtain 95actual recordings of real musicians playing real songs, so that we canbetter understand the brain’s responses to the kind of music that mostpeople listen to, rather than the kind of music that is found only in theneuroscientific laboratory. So far this approach has panned out. It ismore difficult to provide rigorous experimental controls with this approach,but it is not impossible; it takes a bit more planning and carefulpreparation, but in the long run, the results are worth it. In using this naturalisticapproach, I can state with reasonable scientific certainty thatwe’re studying the brain doing what it normally does, rather than what itdoes when assaulted by rhythms without any pitch, or melodies withoutany rhythms. In an attempt to separate music into its components, werun the risk—if the experiments are not done properly—of creatingsound sequences that are very unmusical.When I say that I am less interested in the brain than in the mind, thisdoes not mean that I have no interest in the brain. I believe that we allhave brains, and I believe brains are important! But I also believe similarthoughts can arise from different brain architectures. By analogy, I canwatch the same television program on an RCA, a Zenith, a Mitsubishi,even on my computer screen with the right hardware and software. Thearchitectures of all these are sufficiently distinct from one another thatthe patent office—an organization charged with the responsibility of decidingwhen something is sufficiently different from something else thatit constitutes an invention—has issued different patents to these variouscompanies, establishing that the underlying architectures are significantlydifferent. My dog Shadow has a very different brain organization,anatomy, and neurochemistry from mine. When he is hungry or hurts hispaw, it is unlikely that the pattern of nerve firings in his brain bears muchresemblance to the pattern of firings in my brain when I’m hungry or stubmy toe. But I do believe that he is experiencing substantially similarmind states.Some common illusions and misconceptions need to be set aside.Many people, even trained scientists in other disciplines, have the strongintuition that inside the brain there is a strictly isomorphic representationof the world around us. (Isomorphic comes from the Greek word iso,

96 This Is Your Brain on Musicmeaning “same,” and morphus, meaning “form.”) The Gestalt psychologists,who were right about a great many things, were among the first toarticulate this idea. If you look at a square, they argued, a square-shapedpattern of neurons is activated in your brain. Many of us have the intuitionthat if we’re looking at a tree, the image of the tree is somewhererepresented in the brain as a tree, and that perhaps seeing the tree activatesa set of neurons in the shape of a tree, with roots at one end andleaves at the other. When we listen to or imagine a favorite song, it feelslike the song is playing in our head, over a set of neural loudspeakers.Daniel Dennett and V. S. Ramachandran have eloquently argued thatthere is a problem with this intuition. If a mental picture of something(either as we see it right now or imagine it in memory) is itself a picture,there has to be some part of our mind/brain that is seeing that picture.Dennett talks about the intuition that visual scenes are presented onsome sort of a screen or theater in our minds. For this to be true, therewould have to be someone in the audience of that theater watching thescreen, and holding a mental image inside his head. And who would thatbe? What would that mental image look like? This quickly leads to aninfinite regress. The same argument applies to auditory events. No oneargues that it doesn’t feel like we have an audio system in our minds.Because we can manipulate mental images—we can zoom in on them,rotate them, in the case of music we can speed up or slow down the songin our heads—we’re compelled to think there is a home theater in themind. But logically this cannot be true because of the infinite regressproblem.We are also under the illusion that we simply open our eyes and—wesee. A bird chirps outside the window and we instantly hear. Sensoryperception creates mental images in our minds—representations of theworld outside our heads—so quickly and seamlessly that it seems thereis nothing to it. This is an illusion. Our perceptions are the end productof a long chain of neural events that give us the illusion of an instantaneousimage. There are many domains in which our strongest intuitionsmislead us. The flat earth is one example. The intuition that our sensesgive us an undistorted view of the world is another.

Behind the Curtain 97It has been known at least since the time of Aristotle that our sensescan distort the way we perceive the world. My teacher Roger Shepard, aperception psychologist at Stanford University, used to say that whenfunctioning properly, our perceptual system is supposed to distort theworld we see and hear. We interact with the world around us through oursenses. As John Locke noted, everything we know about the world isthrough what we see, hear, smell, touch, or taste. We naturally assumethat the world is just as we perceive it to be. But experiments haveforced us to confront the reality that this is not the case. Visual illusionsare perhaps the most compelling proof of sensory distortion. Many of ushave seen these sorts of illusions as children, such as when two lines ofthe same length appear to be different lengths (the Ponzo illusion).Roger Shepard drew an illusion he calls “Turning the Tables” that isrelated to the Ponzo. It’s hard to believe, but these tabletops are identicalin size and shape (you can check by cutting out a piece of paper orcellophane the exact shape of one and then placing it over the other).This illusion exploits a principle of our visual system’s depth perceptionmechanisms. Even knowing that it is an illusion does not allow us to turnoff the mechanism. No matter how many times we view this figure, it

98 This Is Your Brain on Musiccontinues to surprise us because our brains are actually giving us misinformationabout the objects.In the Kaniza illusion there appears to be a white triangle lying on topof a black-outlined one. But if you look closely, you’ll see that there areno triangles in the figure. Our perceptual system completes or “fills in”information that isn’t there.●▼▼●●

Behind the Curtain 99Why does it do this? Our best guess is that it was evolutionarily adaptiveto do so. Much of what we see and hear contains missing information.Our hunter-gatherer ancestors might have seen a tiger partiallyhidden by trees, or heard a lion’s roar partly obscured by the sound ofleaves rustling much closer to us. Sounds and sights often come to us aspartial information that has been obscured by other things in the environment.A perceptual system that can restore missing informationwould help us make quick decisions in threatening situations. Better torun now than sit and try to figure out if those two separate, brokenpieces of sound were part of a single lion roar.The auditory system has its own version of perceptual completion.The cognitive psychologist Richard Warren demonstrated this particularlywell. He recorded a sentence, “The bill was passed by both housesof the legislature,” and cut out a piece of the sentence from the recordingtape. He replaced the missing piece with a burst of white noise(static) of the same duration. Nearly everyone who heard the alteredrecording could report that they heard both a sentence and static. But alarge proportion of people couldn’t tell where the static was! The auditorysystem had filled in the missing speech information, so that the sentenceseemed to be uninterrupted. Most people reported that there wasstatic and that it existed apart from the spoken sentence. The static andthe sentence formed separate perceptual streams due to differences intimbre that caused them to group separately; Bregman calls this streamingby timbre. Clearly this is a sensory distortion; our perceptual systemis telling us something about the world that isn’t true. But just as clearly,this has an evolutionary/adaptive value if it can help us make sense ofthe world during a life-or-death situation.According to the great perception psychologists Hermann vonHelmholtz, Richard Gregory, Irvin Rock, and Roger Shepard, perceptionis a process of inference, and involves an analysis of probabilities. Thebrain’s task is to determine what the most likely arrangement of objectsin the physical world is, given the particular pattern of information thatreaches the sensory receptors—the retina for vision, the eardrum for

100 This Is Your Brain on Musichearing. Most of the time the information we receive at our sensory receptorsis incomplete or ambiguous. Voices are mixed in with othervoices, the sounds of machines, wind, footsteps. Wherever you are rightnow—whether you’re in an airplane, a coffee shop, a library, at home, ina park, or anywhere else—stop and listen to the sounds around you. Unlessyou’re in a sensory isolation tank, you can probably identify at leasta half-dozen different sounds. Your brain’s ability to make these identificationsis nothing short of remarkable when you consider what it startsout with—that is, what the sensory receptors pass up to it. Groupingprinciples—by timbre, spatial location, loudness, and so on—help tosegregate them, but there is still a lot we don’t know about this process;no one has yet designed a computer that can perform this task of soundsource separation.The eardrum is simply a membrane that is stretched across tissue andbone. It is the gateway to hearing. Virtually all of your impressions of theauditory world come from the way in which it wiggles back and forth inresponse to air molecules hitting it. (To a degree, the pinnae—the fleshyparts of your ear—are also involved in auditory perception, as are thebones in your skull, but for the most part, the eardrum is the primarysource of what we know about what is out there in the auditory world.)Let’s consider a typical auditory scene, a person sitting in her living roomreading a book. In this environment, let’s suppose that there are sixsources of sound that she can readily identify: the whooshing noise ofthe central heating (the fan or blower that moves air through the ductwork),the hum of a refrigerator in the kitchen, traffic outside on thestreet (which itself could be several or dozens of distinct sounds comprisingdifferent engines, brakes squeaking, horns, etc.), leaves rustlingin the wind outside, a cat purring on the chair next to her, and a recordingof Debussy preludes. Each of these can be considered an auditoryobject or a sound source, and we are able to identify them because eachhas its own distinctive sound.Sound is transmitted through the air by molecules vibrating at certainfrequencies. These molecules bombard the eardrum, causing it to wigglein and out depending on how hard they hit it (related to the volume or

Behind the Curtain 101amplitude of the sound) and on how fast they’re vibrating (related towhat we call pitch). But there is nothing in the molecules that tells theeardrum where they came from, or which ones are associated withwhich object. The molecules that were set in motion by the cat purringdon’t carry an identifying tag that says cat, and they may arrive on theeardrum at the same time and in the same region of the eardrum as thesounds from the refrigerator, the heater, Debussy, and everything else.Imagine that you stretch a pillowcase tightly across the opening of abucket, and different people throw Ping-Pong balls at it from differentdistances. Each person can throw as many Ping-Pong balls as he likes,and as often as he likes. Your job is to figure out, just by looking at howthe pillowcase moves up and down, how many people there are, whothey are, and whether they are walking toward you, away from you, orare standing still. This is analogous to what the auditory system has tocontend with in making identifications of auditory objects in the world,using only the movement of the eardrum as a guide. How does the brainfigure out, from this disorganized mixture of molecules beating against amembrane, what is out there in the world? In particular, how does it dothis with music?It does this through a process of feature extraction, followed by anotherprocess of feature integration. The brain extracts basic, low-levelfeatures from the music, using specialized neural networks that decomposethe signal into information about pitch, timbre, spatial location,loudness, reverberant environment, tone durations, and the onset timesfor different notes (and for different components of tones). These operationsare carried out in parallel by neural circuits that compute thesevalues and that can operate somewhat independently of one another—that is, the pitch circuit doesn’t need to wait for the duration circuit to bedone in order to perform its calculations. This sort of processing—where only the information contained in the stimulus is considered bythe neural circuits—is called bottom-up processing. In the world and inthe brain, these attributes of the music are separable. We can change onewithout changing the other, just as we can change shape in visual objectswithout changing their color.

102 This Is Your Brain on MusicLow-level, bottom-up processing of basic elements occurs in the peripheraland phylogenetically older parts of our brains; the term lowlevelrefers to the perception of elemental or building-block attributes ofa sensory stimulus. High-level processing occurs in more sophisticatedparts of our brains that take neural projections from the sensory receptorsand from a number of low-level processing units; this refers to thecombining of low-level elements into an integrated representation. Highlevelprocessing is where it all comes together, where our minds come toan understanding of form and content. Low-level processing in yourbrain sees blobs of ink on this page, and perhaps even allows you to putthose blobs together and recognize a basic form in your visual vocabulary,such as the letter A. But it is high-level processing that puts togetherthree letters to let you read the word ART and to generate a mental imageof what the word means.At the same time as feature extraction is taking place in the cochlea,auditory cortex, brain stem, and cerebellum, the higher-level centers ofour brain are receiving a constant flow of information about what hasbeen extracted so far; this information is continually updated, and typicallyrewrites the older information. As our centers for higher thought—mostly in the frontal cortex—receive these updates, they are working hardto predict what will come next in the music, based on several factors:~ what has already come before in the piece of music we’re hearing;~ what we remember will come next if the music is familiar;~ what we expect will come next if the genre or style is familiar,based on previous exposure to this style of music;~ any additional information we’ve been given, such as a summaryof the music that we’ve read, a sudden movement by a performer,or a nudge by the person sitting next to us.These frontal-lobe calculations are called top-down processing andthey can exert influence on the lower-level modules while they are per-

Behind the Curtain 103forming their bottom-up computations. The top-down expectations cancause us to misperceive things by resetting some of the circuitry in thebottom-up processors. This is partly the neural basis for perceptual completionand other illusions.The top-down and bottom-up processes inform each other in an ongoingfashion. At the same time as features are being analyzed individually,parts of the brain that are higher up—that is, that are morephylogenetically advanced, and that receive connections from lowerbrain regions—are working to integrate these features into a perceptualwhole. The brain constructs a representation of reality, based on thesecomponent features, much as a child constructs a fort out of Legoblocks. In the process, the brain makes a number of inferences, dueto incomplete or ambiguous information; sometimes these inferencesturn out to be wrong, and that is what visual and auditory illusions are:demonstrations that our perceptual system has guessed incorrectlyabout what is out-there-in-the-world.The brain faces three difficulties in trying to identify the auditory objectswe hear. First, the information arriving at the sensory receptors isundifferentiated. Second, the information is ambiguous—different objectscan give rise to similar or identical patterns of activation on theeardrum. Third, the information is seldom complete. Parts of the soundmay be covered up by other sounds, or lost. The brain has to make a calculatedguess about what is really out there. It does so very quickly andgenerally subconsciously. The illusions we saw previously, along withthese perceptual operations, are not subject to our awareness. I can tellyou, for example, that the reason you see triangles where there are nonein the Kaniza figure is due to perceptual completion. But even after youknow the principles that are involved, it is impossible to turn them off.Your brain keeps on processing the information in the same way, andyou continue to be surprised by the outcome.Helmholtz called this process “unconscious inference.” Rock called it“the logic of perception.” George Miller, Ulrich Neisser, Herbert Simon,and Roger Shepard have described perception as a “constructive process.”These are all ways of saying that what we see and hear is the end

104 This Is Your Brain on Musicof a long chain of mental events that give rise to an impression, a mentalimage, of the physical world. Many of the ways in which our brains function—includingour senses of color, taste, smell, and hearing—arose dueto evolutionary pressures, some of which no longer exist. The cognitivepsychologist Steven Pinker and others have suggested that our musicperceptionsystem was essentially an evolutionary accident, and thatsurvival and sexual-selection pressures created a language and communicationsystem that we learned to exploit for musical purposes. This isa contentious point in the cognitive-psychology community. The archaeologicalrecord has left us some clues, but it rarely leaves us a “smokinggun” that can settle such issues definitively. The filling-in phenomenonI’ve described is not just a laboratory curiosity; composers exploit thisprinciple as well, knowing that our perception of a melodic line will continue,even if part of it is obscured by other instruments. Whenever wehear the lowest notes on the piano or double bass, we are not actuallyhearing 27.5 or 35 Hz, because those instruments are typically incapableof producing much energy at these ultralow frequencies: Our ears are fillingin the information and giving us the illusion that the tone is that low.We experience illusions in other ways in music. In piano works suchas Sindig’s “The Rustle of Spring” or Chopin’s Fantasy-Impromptu inC-sharp Minor, op. 66, the notes go by so quickly that an illusory melodyemerges. Play the tune slowly and it disappears. Due to stream segregation,the melody “pops out” when the notes are close enough together intime—the perceptual system holds the notes together—but the melody islost when its notes are too far apart in time. As studied by Bernard Lortat-Jacob at the Musée de l’Homme in Paris, the Quintina (literally “fifthone”) in Sardinian a capella vocal music also conveys an illusion: A fifthfemale voice emerges from the four male voices when the harmony andtimbres are performed just right. (They believe the voice is that of the VirginMary coming to reward them if they are pious enough to sing it right.)In the Eagles’ “One of These Nights” (the title song from the album ofthe same name) the song opens with a pattern played by bass and guitarthat sounds like one instrument—the bass plays a single note, and the

Behind the Curtain 105guitar adds a glissando, but the perceptual effect is of the bass sliding,due to the Gestalt principle of good continuation. George Shearing createda new timbral effect by having guitar (or in some cases, vibrophone)double what he was playing on the piano so precisely that listeners comeaway wondering, “What is that new instrument?” when in reality it is twoseparate instruments whose sounds have perceptually fused. In “LadyMadonna,” the four Beatles sing into their cupped hands during an instrumentalbreak and we swear that there are saxophones playing, basedon the unusual timbre they achieve coupled with our (top-down) expectationthat saxophones should be playing in a song of this genre.Most contemporary recordings are filled with another type of auditoryillusion. Artificial reverberation makes vocalists and lead guitarssound like they’re coming from the back of a concert hall, even whenwe’re listening in headphones and the sound is coming from an inchaway from our ears. Microphone techniques can make a guitar soundlike it is ten feet wide and your ears are right where the soundhole is—an impossibility in the real world (because the strings have to go acrossthe soundhole—and if your ears were really there, the guitarist would bestrumming your nose). Our brains use cues about the spectrum of thesound and the type of echoes to tell us about the auditory world aroundus, much as a mouse uses his whiskers to know about the physical worldaround him. Recording engineers have learned to mimic those cues toimbue recordings with a real-world, lifelike quality even when they’remade in sterile recording studios.There is a related reason why so many of us are attracted to recordedmusic these days—and especially now that personal music players arecommon and people are listening in headphones a lot. Recording engineersand musicians have learned to create special effects that tickle ourbrains by exploiting neural circuits that evolved to discern importantfeatures of our auditory environment. These special effects are similar inprinciple to 3-D art, motion pictures, or visual illusions, none of whichhave been around long enough for our brains to have evolved specialmechanisms to perceive them; rather, they leverage perceptual systems

106 This Is Your Brain on Musicthat are in place to accomplish other things. Because they use these neuralcircuits in novel ways, we find them especially interesting. The sameis true of the way that modern recordings are made.Our brains can estimate the size of an enclosed space on the basis ofthe reverberation and echo present in the signal that hits our ears. Eventhough few of us understand the equations necessary to describe howone room differs from another, all of us can tell whether we’re standingin a small, tiled bathroom, a medium-sized concert hall, or a large churchwith high ceilings. And we can tell when we hear recordings of voiceswhat size room the singer or speaker is in. Recording engineers createwhat I call “hyperrealities,” the recorded equivalent of the cinematographer’strick of mounting a camera on the bumper of a speeding car. Weexperience sensory impressions that we never actually have in the realworld.Our brains are exquisitely sensitive to timing information. We are ableto localize objects in the world based on differences of only a few millisecondsbetween the time of arrival of a sound at one of our ears versusthe other. Many of the special effects we love to hear in recordedmusic are based on this sensitivity. The guitar sound of Pat Metheny orDavid Gilmour of Pink Floyd use multiple delays of the signal to give anotherwordly, haunting effect that triggers parts of our brains in ways thathumans had never experienced before, by simulating the sound of an enclosedcave with multiple echoes such as would never actually occur inthe real world—an auditory equivalent of the barbershop mirrors that repeatedinfinitely.Perhaps the ultimate illusion in music is the illusion of structure andform. There is nothing in a sequence of notes themselves that creates therich emotional associations we have with music, nothing about a scale, achord, or a chord sequence that intrinsically causes us to expect a resolution.Our ability to make sense of music depends on experience, andon neural structures that can learn and modify themselves with eachnew song we hear, and with each new listening to an old song. Our brainslearn a kind of musical grammar that is specific to the music of our culture,just as we learn to speak the language of our culture.

Behind the Curtain 107Noam Chomsky’s contribution to modern linguistics and psychologywas proposing that we are all born with an innate capacity to understandany of the world’s languages, and that experience with a particular languageshapes, builds, and then ultimately prunes a complicated and interconnectednetwork of neural circuits. Our brain doesn’t know beforewe’re born which language we’ll be exposed to, but our brains and naturallanguages coevolved so that all of the world’s languages share certainfundamental principles, and our brains have the capacity to incorporateany of them, almost effortlessly, through mere exposure during a criticalstage of neural development.Similarly, it seems that we all have an innate capacity to learn any ofthe world’s musics, although they, too, differ in substantive ways fromone another. The brain undergoes a period of rapid neural developmentafter birth, continuing for the first years of life. During this time, newneural connections are forming more rapidly than at any other time inour lives, and during our midchildhood years, the brain starts to prunethese connections, retaining only the most important and most oftenused ones. This becomes the basis for our understanding of music, andultimately the basis for what we like in music, what music moves us, andhow it moves us. This is not to say that we can’t learn to appreciate newmusic as adults, but basic structural elements are incorporated into thevery wiring of our brains when we listen to music early in our lives.Music, then, can be thought of as a type of perceptual illusion inwhich our brain imposes structure and order on a sequence of sounds.Just how this structure leads us to experience emotional reactions ispart of the mystery of music. After all, we don’t get all weepy eyed whenwe experience other kinds of structure in our lives, such as a balancedcheckbook or the orderly arrangement of first-aid products in a drugstore(well, at least most of us don’t). What is it about the particular kindof order we find in music that moves us so? The structure of scales andchords has something to do with it, as does the structure of our brains.Feature detectors in our brains work to extract information from thestream of sounds that hits our ears. The brain’s computational systemcombines these into a coherent whole, based in part on what it thinks it

108 This Is Your Brain on Musicought to be hearing, and in part based on expectations. Just where thoseexpectations come from is one of the keys to understanding how musicmoves, when it moves us, and why some music only makes us want toreach for the off button on our radios or CD players. The topic of musicalexpectations is perhaps the area in the cognitive neuroscience ofmusic that most harmoniously unites music theory and neural theory,musicians and scientists, and to understand it completely, we have tostudy how particular patterns of music give rise to particular patterns ofneural activations in the brain.

4. AnticipationWhat We Expect from Liszt (and Ludacris)When I’m at a wedding, it is not the sight of the hope and love of thebride and groom standing in front of their friends and family, theirwhole life before them, that makes my eyes tear up. It is when the musicbegins that I start to cry. In a movie, when two people are at long last reunitedafter some great ordeal, the music again pushes me and my emotionsover the sentimental edge.I said earlier that music is organized sound, but the organization hasto involve some element of the unexpected or it is emotionally flat androbotic. The appreciation we have for music is intimately related to ourability to learn the underlying structure of the music we like—the equivalentto grammar in spoken or signed languages—and to be able to makepredictions about what will come next. Composers imbue music withemotion by knowing what our expectations are and then very deliberatelycontrolling when those expectations will be met, and when theywon’t. The thrills, chills, and tears we experience from music are the resultof having our expectations artfully manipulated by a skilled composerand the musicians who interpret that music.Perhaps the most documented illusion—or parlor trick—in Westernclassical music is the deceptive cadence. A cadence is a chord sequencethat sets up a clear expectation and then closes, typically with a satisfy-

110 This Is Your Brain on Musicing resolution. In the deceptive cadence, the composer repeats the chordsequence again and again until he has finally convinced the listeners thatwe’re going to get what we expect, but then at the last minute, he givesus an unexpected chord—not outside the key, but a chord that tells usthat it’s not over, a chord that doesn’t completely resolve. Haydn’s use ofthe deceptive cadence is so frequent, it borders on an obsession. PerryCook has likened this to a magic trick: Magicians set up expectationsand then defy them, all without you knowing exactly how or whenthey’re going to do it. Composers do the same thing. The Beatles’ “ForNo One” ends on the V chord (the fifth degree of the scale we’re in) andwe wait for a resolution that never comes—at least not in that song. Butthe very next song on the album Revolver starts with the very chord wewere waiting to hear.The setting up and then manipulating of expectations is the heart ofmusic, and it is accomplished in countless ways. Steely Dan do it by playingsongs that are essentially the blues (with blues structure and chordprogressions) but by adding unusual harmonies to the chords that makethem sound very unblues—for example on their song “Chain Lightning.”Miles Davis and John Coltrane made careers out of reharmonizing bluesprogressions to give them new sounds that were anchored partly inthe familiar and partly in the exotic. On his solo album Kamakiriad,Donald Fagen (of Steely Dan) has songs with blues/funk rhythms thatlead us to expect the standard blues chord progression, but the entiresong is played on only one chord, never moving from that harmonic position.In “Yesterday,” the main melodic phrase is seven measures long; theBeatles surprise us by violating one of the most basic assumptions ofpopular music, the four- or eight-measure phrase unit (nearly all rock/pop songs have musical ideas that are organized into phrases of thoselengths). In “I Want You (She’s So Heavy),” the Beatles violate expectationsby first setting up a hypnotic, repetitive ending that sounds like itwill go on forever; based on our experience with rock music and rockmusic endings, we expect that the song will slowly die down in volume,

Anticipation 111the classic fade-out. Instead, they end the song abruptly, and not even atthe end of a phrase—they end right in the middle of a note!The Carpenters use timbre to violate genre expectations; they wereprobably the last group people expected to use a distorted electric guitar,but they did on “Please Mr. Postman” and some other songs. TheRolling Stones—one of the hardest rock bands in the world at the time—had done the opposite of this just a few years before by using violins (asfor example, on “Lady Jane”). When Van Halen were the newest, hippestgroup around they surprised fans by launching into a heavy metal versionof an old not-quite-hip song by the Kinks, “You Really Got Me.”Rhythm expectations are violated often as well. A standard trick inelectric blues is for the band to build up momentum and then stop playingaltogether while the singer or lead guitarist continues on, as in StevieRay Vaughan’s “Pride and Joy,” Elvis Presley’s “Hound Dog,” or the AllmanBrothers’ “One Way Out.” The classic ending to an electric bluessong is another example. The song charges along with a steady beat fortwo or three minutes and—wham! Just as the chords suggest an endingis imminent, rather than charging through at full speed, the band suddenlystarts playing at half the tempo they were before.In a double whammy, Creedence Clearwater Revival pulls out thisslowed-down ending in “Lookin’ Out My Back Door”—by then such anending was already a well-known cliché—and they violate the expectationsof that by coming in again for the real ending of the song at fulltempo.The Police made a career out of violating rhythmic expectations. Thestandard rhythmic convention in rock is to have a strong backbeat onbeats two and four. Reggae music turns this around putting the snaredrum on beats one and two, and (typically) a guitar on two and four. ThePolice combined reggae with rock to create a new sound that fulfilledsome and violated other rhythmic expectations simultaneously. Sting oftenplayed bass guitar parts that were entirely novel, avoiding the rockclichés of playing on the downbeat or of playing synchronously with thebass drum. As Randy Jackson of American Idol fame, and one of the top

112 This Is Your Brain on Musicsession bass players, told me (back when we shared an office in a recordingstudio in the 1980s), Sting’s basslines are unlike anyone else’s,and they wouldn’t even fit in anyone else’s songs. “Spirits in the MaterialWorld” from their album Ghost in the Machine takes this rhythmic playto such an extreme it can be hard to tell where the downbeat even is.Modern composers such as Schönberg threw out the whole idea ofexpectation. The scales they used deprive us of the notion of a resolution,a root to the scale, or a musical “home,” thus creating the illusion ofno home, a music adrift, perhaps as a metaphor for a twentieth-centuryexistentialist existence (or just because they were trying to be contrary).We still hear these scales used in movies to accompany dream sequencesto convey a lack of grounding, or in underwater or outer space scenes toconvey weightlessness.These aspects of music are not represented directly in the brain, atleast not during initial stages of processing. The brain constructs its ownversion of reality, based only in part on what is there, and in part on howit interprets the tones we hear as a function of the role they play ina learned musical system. We interpret spoken language analogously.There is nothing intrinsically catlike about the word cat or even any of itssyllables. We have learned that this collection of sounds represents thefeline house pet. Similarly, we have learned that certain sequences oftones go together, and we expect them to continue to do so. We expectcertain pitches, rhythms, timbres, and so on to co-occur based on a statisticalanalysis our brain has performed of how often they have gone togetherin the past. We have to reject the intuitively appealing idea thatthe brain is storing an accurate and strictly isomorphic representation ofthe world. To some degree, it is storing perceptual distortions, illusions,and extracting relationships among elements. It is computing a realityfor us, one that is rich in complexity and beauty. A basic piece of evidencefor such a view is the simple fact that light waves in the worldvary along one dimension—wavelength—and yet our perceptual systemtreats color as two dimensional (the color circle described on page 29).Similarly with pitch: From a one-dimensional continuum of moleculesvibrating at different speeds, our brains construct a rich, multidimen-

Anticipation 113sional pitch space with three, four, or even five dimensions (according tosome models). If our brain is adding this many dimensions to what is outthere in the world, this can help explain the deep reactions we have tosounds that are properly constructed and skillfully combined.When cognitive scientists talk about expectations and violating them,we mean an event whose occurrence is at odds with what might havebeen reasonably predicted. It is clear that we know a great deal about anumber of different standard situations. Life presents us with similar situationsthat differ only in details, and often those details are insignificant.Learning to read is an example. The feature extractors in our brainhave learned to detect the essential and unvarying aspect of letters of thealphabet, and unless we explicitly pay attention, we don’t notice detailssuch as the font that a word is typed in. Even though surface details are different,all these words are equally recognizable, as are their individualletters. (It may be jarring to read sentences in which every word is in adifferent font, and of course such rapid shifting causes us to notice, butthe point remains that our feature detectors are busy extracting thingslike “the letter a” rather than processing the font it is typed in.)An important way that our brain deals with standard situations is thatit extracts those elements that are common to multiple situations andcreates a framework within which to place them; this framework iscalled a schema. The schema for the letter a would be a description ofits shape, and perhaps a set of memory traces that includes all the a’swe’ve ever seen, showing the variability that accompanies the schema.Schemas inform a host of day-to-day interactions we have with theworld. For example, we’ve been to birthday parties and we have a generalnotion—a schema—of what is common to birthday parties. Thebirthday party schema will be different for different cultures (as is music),and for people of different ages. The schema leads to clear expectations,as well as a sense of which of those expectations are flexible andwhich are not. We can make a list of things we would expect to find ata typical birthday party. We wouldn’t be surprised if these weren’t allpresent, but the more of them that are absent, the less typical the partywould be:

114 This Is Your Brain on Music~ A person who is celebrating the anniversary of their birth~ Other people helping that person to celebrate~ A cake with candles~ Presents~ Festive food~ Party hats, noisemakers, and other decorationsIf the party was for an eight-year-old we might have the additional expectationthat there would be a rousing game of pin-the-tail-on-thedonkey,but not single-malt scotch. This more or less constitutes ourbirthday party schema.We have musical schemas, too, and these begin forming in the womband are elaborated, amended, and otherwise informed every time we listento music. Our musical schema for Western music includes implicitknowledge of the scales that are normally used. This is why Indian orPakistani music, for example, sounds “strange” to us the first time wehear it. It doesn’t sound strange to Indians and Pakistanis, and it doesn’tsound strange to infants (or at least not any stranger than any other music).This may be an obvious point, but it sounds strange by virtue of itsbeing inconsistent with what we have learned to call music. By the ageof five, infants have learned to recognize chord progressions in the musicof their culture—they are forming schemas.We develop schemas for particular musical genres and styles; style isjust another word for “repetition.” Our schema for a Lawrence Welk concertincludes accordions, but not distorted electric guitars, and ourschema for a Metallica concert is the opposite. A schema for Dixielandincludes foot-tapping, up-tempo music, and unless the band was tryingto be ironic, we would not expect there to be overlap between theirrepertoire and that of a funeral procession. Schemas are an extension ofmemory. As listeners, we recognize when we are hearing somethingwe’ve heard before, and we can distinguish whether we heard it earlier

Anticipation 115in the same piece, or in a different piece. Music listening requires, accordingto the theorist Eugene Narmour, that we be able to hold in memorya knowledge of those notes that have just gone by, alongside aknowledge of all other musics we are familiar with that approximate thestyle of what we’re listening to now. This latter memory may not havethe same level of resolution or the same amount of vividness as noteswe’ve just heard, but it is necessary in order to establish a context for thenotes we’re hearing.The principal schemas we develop include a vocabulary of genresand styles, as well as of eras (1970s music sounds different from 1930smusic), rhythms, chord progressions, phrase structure (how many measuresto a phrase), how long a song is, and what notes typically followwhat. When I said earlier that the standard popular song has phrases thatare four or eight measures long, this is a part of the schema we’ve developedfor late twentieth-century popular songs. We’ve heard thousands ofsongs thousands of times and even without being able to explicitly describeit, we have incorporated this phrase tendency as a “rule” aboutmusic we know. When “Yesterday” plays with its seven-measure phrase,it is a surprise. Even though we’ve heard “Yesterday” a thousand or eventen thousand times, it still interests us because it violates schematic expectationsthat are even more firmly entrenched than our memory forthis particular song. Songs that we keep coming back to for years playaround with expectations just enough that they are always at least a littlebit surprising. Steely Dan, the Beatles, Rachmaninoff, and Miles Davisare just a few of the artists that some people say they never tire of, andthis is a big part of the reason.Melody is one of the primary ways that our expectations are controlledby composers. Music theorists have identified a principle calledgap fill; in a sequence of tones, if a melody makes a large leap, either upor down, the next note should change direction. A typical melody includesa lot of stepwise motion, that is, adjacent tones in the scale. If themelody makes a big leap, theorists describe a tendency for the melody to“want” to return to the jumping-off point; this is another way to say thatour brains expect that the leap was only temporary, and tones that fol-

116 This Is Your Brain on Musiclow need to bring us closer and closer to our starting point, or harmonic“home.”In “Somewhere Over the Rainbow,” the melody begins with one of thelargest leaps we’ve ever experienced in a lifetime of music listening: anoctave. This is a strong schematic violation, and so the composer rewardsand soothes us by bringing the melody back toward home again, but notby too much—he does come down, but only by one scale degree—because he wants to continue to build tension. The third note of thismelody fills the gap. Sting does the same thing in “Roxanne”: He leaps upan interval of roughly a half octave (a perfect fourth) to hit the first syllableof the word Roxanne, and then comes down again to fill the gap.We also hear gap fill in the andante cantabile from Beethoven’s“Pathétique” Sonata. As the main theme climbs upward, it moves from aC (in the key of A-flat, this is the third degree of the scale) to the A-flatthat is an octave above what we consider the “home” note, and it keepson climbing to a B-flat. Now that we’re an octave and a whole step higherthan home, there is only one way to go, back toward home. Beethovenactually jumps toward home, down an interval of a fifth, landing on thenote (E-flat) that is a fifth above the tonic. To delay the resolution—Beethoven was a master of suspense—instead of continuing the descentdown to the tonic, Beethoven moves away from it. In writing the jumpdown from the high B-flat to the E-flat, Beethoven was pitting twoschemas against each other: the schema for resolving to the tonic, andthe schema for gap fill. By moving away from the tonic at this point, he isalso filling the gap he made by jumping so far down to get to this midpoint.When Beethoven finally brings us home two measures later, it is assweet a resolution as we’ve ever heard.Consider now what Beethoven does to expectations with the melodyto the main theme from the last movement of his Ninth Symphony (“Odeto Joy”). These are the notes of the melody, as solfège, the do-re-misystem:mi - mi - fa - sol - sol - fa - me - re - do - do -re - mi - mi - re - re

Anticipation 117(If you’re having trouble following along, it will help if you sing inyour mind the English words to this part of the song: “Come and sing asong of joy for peace a glory gloria ...”)The main melodic theme is simply the notes of the scale! The bestknown,overheard, and overused sequence of notes we have in Westernmusic. But Beethoven makes it interesting by violating our expectations.He starts on a strange note and ends on a strange note. He starts on thethird degree of the scale (as he did on the “Pathétique” Sonata), ratherthan the root, and then goes up in stepwise fashion, then turns aroundand comes down again. When he gets to the root—the most stabletone—rather than staying there he comes up again, up to the note westarted on, then back down so that we think and we expect he will hit theroot again, but he doesn’t; he stays right there on re, the second scale degree.The piece needs to resolve to the root, but Beethoven keeps ushanging there, where we least expect to be. He then runs the entire motifagain, and only on the second time through does he meet our expectations.But now, that expectation is even more interesting because ofthe ambiguity: We wonder if, like Lucy waiting for Charlie Brown, he willpull the football of resolution away from us at the last minute.What do we know about the neural basis for musical expectations andmusical emotion? If we acknowledge that the brain is constructing a versionof reality, we must reject that the brain has an accurate and strictlyisomorphic representation of the world. So what is the brain holding inits neurons that represents the world around us? The brain represents allmusic and all other aspects of the world in terms of mental or neuralcodes. Neuroscientists try to decipher this code and understand itsstructure, and how it translates into experience. Cognitive psychologiststry to understand these codes at a somewhat higher level—not at thelevel of neural firings, but at the level of general principles.The way in which a picture is stored on your computer is similar, inprinciple, to how the neural code works. When you store a picture onyour computer, the picture is not stored on your hard drive the way that

118 This Is Your Brain on Musica photograph is stored in your grandmother’s photo album. When youopen your grandmother’s album, you can pick up a photo, turn it upsidedown, give it to a friend; it is a physical object. It is the photograph, nota representation of a photograph. On the other hand, a photo in yourcomputer is stored in a file made up of 0s and 1s—the binary code thatcomputers use to represent everything.If you’ve ever opened a corrupt file, or if your e-mail program didn’tproperly download an attachment, you’ve probably seen a bunch of gibberishin place of what you thought was a computer file: a string of funnysymbols, squiggles, and alphanumeric characters that looks like theequivalent of a comic-strip swear word. (These represent a sort of intermediatehexadecimal code that itself is resolved into 0s and 1s, but thisintermediate stage is not crucial for understanding the analogy.) In thesimplest case of a black-and-white photograph, a 1 might represent thatthere is a black dot at a particular place in the picture, and a 0 might indicatethe absence of a black dot, or a white dot. You can imagine thatone could easily represent a simple geometric shape using these 0s and1s, but the 0s and 1s would not themselves be in the shape of a triangle,they would simply be part of a long line of 0s and 1s, and the computerwould have a set of instructions telling it how to interpret them (and towhat spatial location each number refers). If you got really good at readingsuch a file, you might be able to decode it, and guess what sort of imageit represents. The situation is vastly more complicated with a colorimage, but the principle is the same. People who work with image filesall the time are able to look at the stream of 0s and 1s and tell somethingabout the nature of the photograph—not at the level of whether it is ahuman or a horse, perhaps, but things like how much red or gray is in thepicture, how sharp the edges are, and so forth. They have learned to readthe code that represents the picture.Similarly, audio files are stored in binary format, as sequences of 0sand 1s. The 0s and 1s represent whether or not there is any sound at particularparts of the frequency spectrum. Depending on its position in thefile, a certain sequence of 0s and 1s will indicate if a bass drum or a piccolois playing.

Anticipation 119In the cases I’ve just described, the computer is using a code to representcommon visual and auditory objects. The objects themselves aredecomposed into small components—pixels in the case of a picture, sinewaves of a particular frequency and amplitude in the case of sound—andthese components are translated into the code. Of course, the computer(brain) is running a lot of fancy software (mind) that translates the codeeffortlessly. Most of us don’t have to concern ourselves with the code itselfat all. We scan a photo or rip a song to our hard drive, and when wewant to see it or hear it, we double-click on it and there it appears, in allits original glory. This is an illusion made possible by the many layers oftranslation and amalgamation going on, all of it invisible to us. This iswhat the neural code is like. Millions of nerves firing at different ratesand different intensities, all of it invisible to us. We can’t feel our nervesfiring; we don’t know how to speed them up, slow them down, turn themon when we’re having trouble getting started on a bleary-eyed morning,or shut them off so we can sleep at night.Years ago, my friend Perry Cook and I were astonished when we readan article about a man who could look at phonograph records and identifythe piece of music that was on them, by looking at the grooves, withthe label obscured. Did he memorize the patterns of thousands of recordalbums? Perry and I took out some old record albums and we noticedsome regularities. The grooves of a vinyl record contain a code that is“read” by the needle. Low notes create wide grooves, high notes createnarrow grooves, and a needle dropped inside the grooves is movingthousands of times per second to capture the landscape of the innerwall. If a person knew many pieces of music well, it would be possible tocharacterize them in terms of how many low notes there were (rap musichas a lot, baroque concertos don’t), how steady versus percussive thelow notes are (think of a jazz-swing tune with walking bass as opposedto a funk tune with slapping bass), and to learn how these shapes are encodedin vinyl. This fellow’s skills are extraordinary, but they’re not inexplicable.We encounter gifted auditory-code readers every day: the mechanicwho can listen to the sound of your engine and determine whether your

120 This Is Your Brain on Musicproblems are due to clogged fuel injectors or a slipped timing chain; thedoctor who can tell by listening to your heart whether you have an arrhythmia;the police detective who can tell when a suspect is lying by thestress in his voice; the musician who can tell a viola from a violin or aB-flat clarinet from an E-flat clarinet just by the sound. In all these cases,timbre is playing an important role in helping us to unlock the code.How can we study neural codes and learn to interpret them? Someneuroscientists start by studying neurons and their characteristics—what causes them to fire, how rapidly they fire, what their refractory periodis (how long they need to recover between firings); we study howneurons communicate with each other and the role of neurotransmittersin conveying information in the brain. Much of the work at this level ofanalysis concerns general principles; we don’t yet know much about theneurochemistry of music, for example, although I’ll reveal some excitingnew results along this line from my laboratory in Chapter 5.But I’ll back up for a minute. Neurons are the primary cells of thebrain; they are also found in the spinal cord and the peripheral nervoussystem. Activity from outside the brain can cause a neuron to fire—suchas when a tone of a particular frequency excites the basilar membrane,and it in turn passes a signal up to a frequency-selective neurons in theauditory cortex. Contrary to what we thought a hundred years ago, theneurons in the brain aren’t actually touching; there’s a space betweenthem called the synapse. When we say a neuron is firing, it is sending anelectrical signal that causes the release of a neurotransmitter. Neurotransmittersare chemicals that travel throughout the brain and bind toreceptors attached to other neurons. Receptors and neurotransmitterscan be thought of as locks and keys respectively. After a neuron fires, aneurotransmitter swims across that synapse to a nearby neuron, andwhen it finds the lock and binds with it, that new neuron starts to fire.Not all keys fit all locks; there are certain locks (receptors) that are designedto accept only certain neurotransmitters.Generally, neurotransmitters cause the receiving neuron to fire orprevent it from firing. The neurotransmitters are then absorbed through

Anticipation 121a process called reuptake; without reuptake, the neurotransmitters wouldcontinue to stimulate or inhibit the firing of a neuron.Some neurotransmitters are used throughout the nervous system,and some only in certain brain regions and by certain kinds of neurons.Serotonin is produced in the brain stem and is associated with the regulationof mood and sleep. The new class of antidepressants, includingProzac and Zoloft, are known as selective serotonin reuptake inhibitors(SSRIs) because they inhibit the reuptake of serotonin in the brain, allowingwhatever serotonin is already there to act for a longer periodof time. The precise mechanism by which this alleviates depression,obsessive-compulsive disorder, and mood and sleep disorders is notknown. Dopamine is released by the nucleus accumbens and is involvedin mood regulation and the coordination of movement. It is most famousfor being part of the brain’s pleasure and reward system. When drug addictsget their drug of choice, or when compulsive gamblers win a bet—even when chocoholics get cocoa—this is the neurotransmitter thatis released. Its role—and the important role played by the nucleusaccumbens—in music was unknown until 2005.Cognitive neuroscience has been making great leaps in understandingover the last decade. We now know so much more about how neuronswork, how they communicate, how they form networks, and howneurons develop from their genetic recipes. One finding at the macrolevel about the function of the brain is the popular notion about hemisphericspecialization—the idea that the left half of the brain and theright half of the brain perform different cognitive functions. This is certainlytrue, but as with much of the science that has permeated popularculture, that real story is somewhat more nuanced.To begin with, the research on which this is based was performed onright-handed people. For reasons that aren’t entirely clear, people whoare left-handed (approximately 5 to 10 percent of the population) orambidextrous sometimes have the same brain organization as righthanders,but more often have a different brain organization. When thebrain organization is different, it can take the form of a simple mirror

122 This Is Your Brain on Musicimage, such that functions are simply flipped to the opposite side. Inmany cases, however, left-handers have a neural organization that is differentin ways that are not yet well documented. Thus, any generalizationswe make about hemispheric asymmetries are applicable only tothe right-handed majority of the population.Writers, businessmen, and engineers refer to themselves as left-braindominant, and artists, dancers, and musicians as right-brain dominant.The popular conception that the left brain is analytical and the rightbrain is artistic has some merit, but is overly simplistic. Both sides of thebrain engage in analysis and both sides in abstract thinking. All of theseactivities require coordination of the two hemispheres, although some ofthe particular functions involved are clearly lateralized.Speech processing is primarily left-hemisphere localized, although certainglobal aspects of spoken language, such as intonation, emphasis, andthe pitch pattern, are more often disrupted following right-hemispheredamage. The ability to distinguish a question from a statement, or sarcasmfrom sincerity, often rests on these right-hemisphere lateralized, nonlinguisticcues, known collectively as prosody. It is natural to wonder whethermusic shows the opposite asymmetry, with processing located primarilyon the right. There are many cases of individuals with brain damage tothe left hemisphere who lost the power of speech, but retained their musicalfunction, and vice versa. Cases like these suggest that music andspeech, although they may share some neural circuits, cannot use completelyoverlapping neural structures.Local features of spoken language, such as distinguishing one speechsound from another, appear to be left-hemisphere lateralized. We’vefound lateralization in the brain basis of music as well. The overall contourof a melody—simply its melodic shape, while ignoring intervals—isprocessed in the right hemisphere, as is making fine discriminations oftones that are close together in pitch. Consistent with its language functions,the left hemisphere is involved in the naming aspects of music—such as naming a song, a performer, an instrument, or a musical interval.Musicians using their right hands or reading music from their right eyealso use the left brain because the left half of the brain controls the right

Anticipation 123half of the body. There is also new evidence that tracking the ongoing developmentof a musical theme—thinking about key and scales andwhether a piece of music makes sense or not—is lateralized to the leftfrontal lobes.Musical training appears to have the effect of shifting some musicprocessing from the right (imagistic) hemisphere to the left (logical)hemisphere, as musicians learn to talk about—and perhaps thinkabout—music using linguistic terms. And the normal course of developmentseems to cause greater hemispheric specialization: Children showless lateralization of musical operations than do adults, regardless ofwhether they are musicians or not.The best place to begin to look at expectation in the musical brain is inhow we track chord sequences in music over time. The most importantway that music differs from visual art is that it is manifested over time.As tones unfold sequentially, they lead us—our brains and our minds—to make predictions about what will come next. These predictions arethe essential part of musical expectations. But how to study the brain basisof these?Neural firings produce a small electric current, and consequentlythe current can be measured with suitable equipment that allows us toknow when and how often neurons are firing; this is called the electroencephalogram,or EEG. Electrodes are placed (painlessly) on the surfaceof the scalp, much as a heart monitor might be taped to your finger,wrist, or chest. The EEG is exquisitely sensitive to the timing of neuralfirings, and can detect activity with a resolution of one thousandth of asecond (one millisecond). But it has some limitations. EEG is not able todistinguish whether the neural activity is releasing excitatory, inhibitory,or modulatory neurotransmitters, the chemicals such as serotonin anddopamine that influence the behavior of other neurons. Because theelectrical signature generated by a single neuron firing is relatively weak,the EEG only picks up the synchronous firing of large groups of neurons,rather than individual neurons.EEG also has limited spatial resolution—that is, a limited ability to

124 This Is Your Brain on Musictell us the location of the neural firings, due to what is called the inversePoisson problem. Imagine that you’re standing inside a football stadiumthat has a large semitransparent dome covering it. You have a flashlight,and you point it up to the inside surface of the dome. Meanwhile, I’mstanding on the outside, looking down at the dome from high above, andI have to predict where you’re standing. You could be standing anywhereon the entire football field and shining your light at the same particularspot in the center of the dome, and from where I’m standing, it will alllook the same to me. There might be slight differences in the angle or thebrightness of the light, but any prediction I make about where you’restanding is going to be a guess. And if you were to bounce your flashlightbeam off of mirrors and other reflective surfaces before it reached thedome, I’d be even more lost. This is the case with electrical signals inthe brain that can be generated from multiple sources in the brain, fromthe surface of the brain or deep down inside the grooves (sulci), and thatcan bounce off of the sulci before reaching the electrode on the outerscalp surface. Still, EEG has been helpful in understanding musical behaviorbecause music is time based, and EEG has the best temporal resolutionof the tools we commonly employ for studying the human brain.Several experiments conducted by Stefan Koelsch, Angela Friederici,and their colleagues have taught us about the neural circuits involved inmusical structure. The experimenters play chord sequences that eitherresolve in the standard, schematic way, or that end on unexpectedchords. After the onset of the chord, electrical activity in the brain associatedwith musical structure is observed within 150–400 milliseconds(ms), and activity associated with musical meaning about 100–150 mslater. The structural processing—musical syntax—has been localized tothe frontal lobes of both hemispheres in areas adjacent to and overlappingwith those regions that process speech syntax, such as Broca’s area,and shows up regardless of whether listeners have musical training. Theregions involved in musical semantics—associating a tonal sequencewith meaning—appear to be in the back portions of the temporal lobe onboth sides, near Wernicke’s area.The brain’s music system appears to operate with functional inde-

Anticipation 125pendence from the language system—the evidence comes from manycase studies of patients who, postinjury, lose one or the other faculty butnot both. The most famous case is perhaps that of Clive Wearing, a musicianand conductor, whose brain was damaged as a result of herpes encephalitis.As reported by Oliver Sacks, Clive lost all memory except formusical memories, and the memory of his wife. Other cases have beenreported for which the patient lost music but retained language andother memories. When portions of his left cortex deteriorated, the composerRavel selectively lost his sense of pitch while retaining his sense oftimbre, a deficit that inspired his writing of Bolero, a piece that emphasizesvariations in timbre. The most parsimonious explanation is thatmusic and language do, in fact, share some common neural resources,and yet have independent pathways as well. The close proximity of musicand speech processing in the frontal and temporal lobes, and theirpartial overlap, suggests that those neural circuits that become recruitedfor music and language may start out life undifferentiated. Experienceand normal development then differentiate the functions of what beganas very similar neuronal populations. Consider that at a very early age,babies are thought to be synesthetic, to be unable to differentiate the inputfrom the different senses, and to experience life and the world as asort of psychedelic union of everything sensory. Babies may see thenumber five as red, taste cheddar cheeses in D-flat, and smell roses in triangles.The process of maturation creates distinctions in the neural pathwaysas connections are cut or pruned. What may have started out as aneuron cluster that responded equally to sights, sound, taste, touch, andsmell becomes a specialized network. So, too, may music and speechhave started in us all with the same neurobiological origins, in the sameregions, and using the same specific neural networks. With increasingexperience and exposure, the developing infant eventually creates dedicatedmusic pathways and dedicated language pathways. The pathwaysmay share some common resources, as has been proposed most prominentlyby Ani Patel in his SSIRH—shared syntactic integration resourcehypothesis.

126 This Is Your Brain on MusicMy collaborator and friend Vinod Menon, a systems neuroscientist atStanford Medical School, shared with me an interest in being able to pindown the findings from the Koelsch and Friederici labs, and in being ableto provide solid evidence for Patel’s SSIRH. For that, we had to use a differentmethod of studying the brain, since the spatial resolution of EEGwasn’t fine enough to really pinpoint the neural locus of musical syntax.Because the hemoglobin of the blood is slightly magnetic, changes inthe flow of blood can be traced with a machine that can track changesin magnetic properties. This is what a magnetic resonance imaging machine(MRI) is, a giant electromagnet that produces a report showing differencesin magnetic properties, which in turn can tell us where, at anygiven point in time, the blood is flowing in the body. (The research on thedevelopment of the first MRI scanners was performed by the Britishcompany EMI, financed in large part from their profits on Beatlesrecords. “I Want to Hold Your Hand” might well have been titled “I Wantto Scan Your Brain.”) Because neurons need oxygen to survive, and theblood carries oxygenated hemoglobin, we can trace the flow of blood inthe brain too. We make the assumption that neurons that are actively firingwill need more oxygen than neurons that are at rest, and so those regionsof the brain that are involved in a particular cognitive task will bejust those regions with the most blood flow at a given point in time.When we use the MRI machine to study the function of brain regions inthis way, the technology is called functional MRI, or fMRI.fMRI images let us see a living, functioning human brain while it isthinking. If you mentally practice your tennis serve, we can see the flowof blood move up to your motor cortex, and the spatial resolution offMRI is good enough that we can see that it is the part of your motor cortexthat controls your arm that is active. If you then start to solve a mathproblem, the blood moves forward, to your frontal lobes, and in particularto regions that have been identified as being associated with arithmeticproblem solving, and we see this movement and ultimately thecollection of blood in the frontal lobes on the fMRI scan.Will this Frankenstein science I’ve just described, the science of brainimaging, ever allow us to read people’s minds? I’m happy to report that

Anticipation 127the answer is probably not, and absolutely not for the foreseeable future.The reason is that thoughts are simply too complicated and involve toomany different regions. With fMRI I can tell that you are listening to musicas opposed to watching a silent film, but we can’t yet tell if you’re listeningto hip-hop versus Gregorian chants, let alone what specific songyou’re listening to or thought you’re thinking.With the high spatial resolution of fMRI, one can tell within just acouple of millimeters where something is occurring in the brain. Theproblem, however, is that the temporal resolution of fMRI isn’t particularlygood because of the amount of time it takes for blood to becomeredistributed in the brain—known as hemodynamic lag. But othershad already studied the when of musical syntax/musical structure processing;we wanted to know the where and in particular if the where involvedareas already known to be dedicated to speech. We found exactlywhat we predicted. Listening to music and attending to its syntacticfeatures—its structure—activated a particular region of the frontal cortexon the left side called pars orbitalis—a subsection of the region knownas Brodmann Area 47. The region we found in our study had some overlapwith previous studies of structure in language, but it also had someunique activations. In addition to this left hemisphere activation, we alsofound activation in an analogous area of the right hemisphere. This toldus that attending to structure in music requires both halves of the brain,while attending to structure in language only requires the left half.Most astonishing was that the left-hemisphere regions that we foundwere active in tracking musical structure were the very same ones thatare active when deaf people are communicating by sign language. Thissuggested that what we had identified in the brain wasn’t a region thatsimply processed whether a chord sequence was sensible, or whether aspoken sentence was sensible. We were now looking at a region that respondedto sight—to the visual organization of words conveyed throughAmerican Sign Language. We found evidence for the existence of a brainregion that processes structure in general, when that structure is conveyedover time. Although the inputs to this region must have come fromdifferent neural populations, and the outputs of it had to go through dis-

128 This Is Your Brain on Musictinctive networks, there it was—a region that kept popping up in anytask that involved organizing information over time.The picture about neural organization for music was becomingclearer. All sound begins at the eardrum. Right away, sounds get segregatedby pitch. Not much later, speech and music probably diverge intoseparate processing circuits. The speech circuits decompose the signalin order to identify individual phonemes—the consonants and vowelsthat make up our alphabet and our phonetic system. The music circuitsstart to decompose the signal and separately analyze pitch, timbre, contour,and rhythm. The output of the neurons performing these tasks connectsto regions in the frontal lobe that put all of it together and try tofigure out if there is any structure or order to the temporal patterning ofit all. The frontal lobes access our hippocampus and regions in the interiorof the temporal lobe and ask if there is anything in our memorybanks that can help to understand this signal. Have I heard this particularpattern before? If so, when? What does it mean? Is it part of a largersequence whose meaning is unfolding right now in front of me?Having nailed down some of the neurobiology of musical structureand expectation, we were now ready to ask about the brain mechanismsunderlying emotion and memory.

5. You Know My Name,Look Up the NumberHow We Categorize MusicOne of my earliest memories of music is as a three-year-old, lying onthe floor underneath the family’s grand piano as my mother played.Lying on our shaggy green wool carpet, with the piano above me, all Icould see were my mother’s legs moving the pedals up and down, but thesound—it engulfed me! It was all around, vibrating through the floor andthrough my body, the low notes to the right of me, the high notes to theleft. The loud, dense chords of Beethoven; the flurry of dancing, acrobaticnotes of Chopin; the strict, almost militaristic rhythms of Schumann,a German like my mother. In these—among my first memories ofmusic—the sound held me in a trance, it transported me to sensoryplaces I had never been. Time seemed to stand still while the music wasplaying.How are memories of music different from other memories? Why canmusic trigger memories in us that otherwise seemed buried or lost? Andhow does expectation lead to the experience of emotion in music? Howdo we recognize songs we have heard before?Tune recognition involves a number of complex neural computationsinteracting with memory. It requires that our brains ignore certain featureswhile we focus only on features that are invariant from one listeningto the next—and in this way, extract invariant properties of a song.

130 This Is Your Brain on MusicThat is, the brain’s computational system must be able to separate the aspectsof a song that remain the same each time we hear it from those thatare one-time-only variations, or from those that are peculiar to a particularpresentation. If the brain didn’t do this, each time we heard a songat a different volume, we’d experience it as an entirely different song!And volume isn’t the only parameter that potentially changes without affectingthe underlying identity of the song. Instrumentation, tempo, andpitch can be considered irrelevant from a tune-recognition standpoint.In the process of abstracting out the features that are essential to a song’sidentity, changes to these features must be set aside.Tune recognition dramatically increases the complexity of the neuralsystem necessary for processing music. Separating the invariant propertiesfrom the momentary ones is a huge computational problem. I workedfor an Internet company in the late 1990s that developed software toidentify MP3 files. Lots of people have soundfiles on their computers, butmany of the files are either misnamed or not named at all. No one wantsto go through file by file and correct bad spellings, like “Etlon John,” orrename songs like “My Aim Is True” to “Alison” by Elvis Costello (thewords my aim is true are the refrain in the chorus, but not the name ofthe song).Solving this automatic naming problem was relatively easy; each songhas a digital “fingerprint,” and all we needed to do was to learn how to efficientlysearch a database of a half-million songs in order to correctlyidentify the song. This is called a “lookup table” by computer scientists.It is equivalent to looking up your Social Security number in a databasegiven your name and date of birth: Only one Social Security number ispresumably associated with a given name and DOB. Similarly, only onesong is associated with a specific sequence of digital values that representthe overall sound of a particular performance of that song. The programworks fabulously well at looking up. What it cannot do is find otherversions of the same song in the database. I might have eight versions of“Mr. Sandman” on my hard drive, but if I submit the version by ChetAtkins to a program and ask it to find other versions (such as the ones byJim Campilongo or the Chordettes), it can’t. This is because the digital

You Know My Name, Look Up the Number 131stream of numbers that starts the MP3 file doesn’t give us anything thatis readily translated to melody, rhythm, or loudness, and we don’t yetknow how to make this translation. Our program would have to be ableto identify relative constancies in melodic and rhythmic intervals, whileignoring performance-specific details. The brain does this with ease, butno one has invented a computer that can even begin to do this.This different ability of computers and humans is related to a debateabout the nature and function of memory in humans. Recent experimentsof musical memory have provided decisive clues in sorting out thetrue story. The big debate among memory theorists over the last hundredyears has been about whether human and animal memory is relational orabsolute. The relational school argues that our memory system storesinformation about the relations between objects and ideas, but not necessarilydetails about the objects themselves. This is also called the constructivistview, because it implies that, lacking sensory specifics, weconstruct a memory representation of reality out of these relations (withmany details filled in or reconstructed on the spot). The constructivistsbelieve that the function of memory is to ignore irrelevant details, whilepreserving the gist. The competing theory is called the record-keepingtheory. Supporters of this view argue that memory is like a tape recorderor digital video camera, preserving all or most of our experiences accurately,and with near perfect fidelity.Music plays a role in this debate because—as the Gestalt psychologistsnoted over one hundred years ago—melodies are defined by pitchrelations (a constructivist view) and yet, they are composed of precisepitches (a record-keeping view, but only if those pitches are encoded inmemory).A great deal of evidence has accumulated in support of both viewpoints.The evidence for the constructivists comes from studies in whichpeople listen to speech (auditory memory) or are asked to read text (visualmemory) and then report what they’ve heard or read. In study afterstudy, people are not very good at re-creating a word-for-word account.They remember general content, but not specific wording.Several studies also point to the malleability of memory. Seemingly

132 This Is Your Brain on Musicminor interventions can powerfully affect the accuracy of memory retrieval.An important series of studies was carried out by Elizabeth Loftusof the University of Washington, who was interested in the accuracyof witnesses’ courtroom testimonies. Subjects were shown videotapesand asked leading questions about the content. If shown two cars thatbarely scraped each other, one group of subjects might be asked, “Howfast were the cars going when they scraped each other?” and anothergroup would be asked, “How fast were the cars going when they smashedeach other?” Such one-word substitutions caused dramatic differencesin the eyewitnesses’ estimates of the speeds of the two vehicles. ThenLoftus brought the subjects back, sometimes up to a week later, andasked, “How much broken glass did you see?” (There really was no brokenglass.) The subjects who were asked the question with the wordsmashed in it were more likely to report “remembering” broken glass inthe video. Their memory of what they actually saw had been reconstructedon the basis of a simple question the experimenter had asked aweek earlier.Findings like these have led researchers to conclude that memory isnot particularly accurate, and that it is constructed out of disparatepieces that may themselves not be accurate. Memory retrieval (and perhapsstorage) undergoes a process similar to perceptual completion orfilling in. Have you ever tried to tell someone about a dream you had overbreakfast the next morning? Typically our memory of the dream appearsto us in imagistic fragments, and the transitions between elements arenot always clear. As we tell the dream, we notice gaps, and we almostcan’t help but fill them in as we unfold the narrative. “I was standing ontop of a ladder outside listening to a Sibelius concert, and the sky wasraining Pez candy . . .” you might begin. But the next image is of yourselfhalfway down the ladder. We naturally and automatically fill in this missinginformation when retelling the dream. “And I decided to protect myselffrom this Pez pelting, so I started climbing down the ladder where Iknew there was shelter. ...”This is the left brain talking (and probably the region called orbitofrontalcortex, just behind your left temple). When we fabricate a story,

You Know My Name, Look Up the Number 133it is almost always the left brain doing the fabricating. The left brainmakes up stories based on the limited information it gets. Usually it getsthe story right, but it will go to great lengths to sound coherent. MichaelGazzaniga discovered this in his work with commissurotomized patients—patientswho had the two hemispheres of the brain surgicallyseparated for the relief of intractable epilepsy. Much of the inputs andoutputs of the brain are contralateral—the left brain controls movementin the right half of the body, and the left brain processes information thatyour right eye sees. A picture of a chicken’s talon was shown to a patient’sleft brain, and a snow-covered house to his right brain (throughhis right and left eyes respectively). A barrier limited the sight of eacheye to only one picture. The patient was then asked to select from an arrayof pictures the one that was most closely associated with each of thetwo items. The patient pointed to a chicken with his left brain (that is, hisright hand) and he pointed to a shovel with his right brain. So far, sogood; chicken goes with talon, and shovel with a snow-covered house.But when Gazzaniga removed the barrier and asked the patient why hehad chosen the shovel, his left hemisphere saw both the chicken and theshovel and generated a story that was consistent with both images. “Youneed a shovel to clean out the chicken shed,” the patient answered, withno awareness that he had seen a snowbound house (with his nonverbalright brain), or that he was inventing an explanation on the spot. Scoreanother piece of evidence for the constructivists.At MIT in the early 1960s, Benjamin White took up the mantle of theGestalt psychologists, who wondered how it is that a song is able to retainits identity in spite of transposition in pitch and time. White systematicallyaltered well-known songs like “Deck the Halls” and “Michael,Row Your Boat Ashore.” In some cases, he would transpose all thepitches, in others he would alter the pitch distances so that contour waspreserved, but the interval sizes were shrunk or stretched. He playedtunes backward and forward, and changed their rhythms. In almostevery case, the deformed tune was recognized more often than chancecould account for.White demonstrated that most listeners can recognize a song in trans-

134 This Is Your Brain on Musicposition almost immediately and without error. And they could recognizeall kinds of deformations of the original tune as well. The constructivistinterpretation of this is that the memory system must be extractingsome generalized, invariant information about songs and storing that. Ifthe record-keeping account were true, they say, it would require new calculationseach time we hear a song in transposition as our brains workto compare the new version to the single, stored representation we haveof the actual performance. But here, it seems that memory extracts anabstract generalization for later use.The record-keeping account follows an old idea of my favorite researchers,the Gestalt psychologists, who said that every experienceleaves a trace or residue in the brain. Experiences are stored as traces,they said, that are reactivated when we retrieve the episodes from memory.A great deal of experimental evidence supports this theory. RogerShepard showed people hundreds of photographs for a few secondseach. A week later, he brought the subjects back into the laboratory andshowed them pairs of photographs that they had seen before, along withsome new ones that they hadn’t. In many cases, the “new” photos hadonly subtle differences from the old, such as the angle of the sail ona sailboat, or the size of a tree in the background. Subjects were ableto remember which ones they had seen a week earlier with astonishingaccuracy.Douglas Hintzman performed a study in which people were shownletters that differed in font and capitalization. For example:F l u t eContrary to studies of gist memory, subjects were able to remember thespecific font.We also know anecdotally that people can recognize hundreds, if notthousands, of voices. You can probably recognize the sound of yourmother’s voice within one word, even if she doesn’t identify herself. Youcan tell your spouse’s voice right away, and whether he or she has a cold

You Know My Name, Look Up the Number 135or is angry with you, all from the timbre of the voice. Then there are wellknownvoices—dozens, if not hundreds, that most people can readilyidentify: Woody Allen, Richard Nixon, Drew Barrymore, W. C. Fields,Groucho Marx, Katharine Hepburn, Clint Eastwood, Steve Martin. Wecan hold in memory the sound of these voices, often as they’re utteringspecific content or catchphrases: “I’m not a crook,” “Say the magic woidand win a hundred dollars,” “Go ahead—make my day,” “Well, excuuuuuseme!” We remember the specific words and specific voices, not just thegist. This supports the record-keeping theory.On the other hand, we enjoy listening to impressionists who do comedyroutines by mimicking the voices of celebrities, and often the funniestroutines involve phrases that the real celebrity never said. In orderfor this to work, we have to have some sort of stored memory trace forthe timbre of the person’s voice, independent of the actual words. Thiscould contradict the record-keeping theory by showing that it is only theabstract properties of the voice that are encoded in memory, rather thanthe specific details. But, we might argue that timbre is a property ofsounds that is separable from other attributes; we can hold on to our“record-keeping” theory of memory by saying that we are encoding specifictimbre values in memory and still explain why we can recognize thesound of a clarinet, even if it is playing a song we’ve never heard before.One of the most famous cases in the neuropsychological literature isthat of a Russian patient known only by his initial S, who saw the physicianA. R. Luria. S. had hypermnesia, the opposite of amnesia—insteadof forgetting everything, he remembered everything. S. was unable torecognize that different views of the same person were related to a singleindividual. If he saw a person smiling, that was one face; if the personlater was frowning, that was another face. S. found it difficult to integratethe many different expressions and viewing angles of a person intoa single, coherent representation of that person. He complained to Dr.Luria, “Everyone has so many faces!” S. was unable to form abstractgeneralizations, only his record-keeping system was intact. In order forus to understand spoken language, we need to set aside variations in

136 This Is Your Brain on Musichow different people pronounce words, or how the same person pronouncesa given phoneme as it appears in different contexts. How canthe record-keeping account be consistent with this?Scientists like having their world organized. Allowing two theories tostand that make different predictions is scientifically unappealing. We’dlike to tidy up our logical world and choose one theory over the other, orgenerate a third, unifying theory that accounts for everything. So whichaccount is right? Record-keeping or constructivist? In short, neither.The research I’ve just described occurred contemporaneously withground-breaking new work on categories and concepts. Categorizationis a basic function of living creatures. Every object is unique, but we oftenact toward different objects as members of classes or categories.Aristotle laid the methods by which modern philosophers and scientiststhink about how concepts form in humans. He argued that categories resultfrom lists of defining features. For example, we have in our minds aninternal representation for the category “triangle.” It contains a mentalimage or picture of every triangle we’ve ever seen, and we can imaginenew triangles as well. At its heart, what constitutes this category and determinesthe boundaries of category membership (what goes in andwhat stays out) is a definition that might be something like this: “A triangleis a three-sided figure.” If you have mathematical training, your definitionmight be more elaborate: “A triangle is a three-sided, closed figure,the sum of whose interior angles is 180 degrees.” Subcategories of trianglesmight be attached to this definition, such as “an iscosceles trianglehas two sides of equal length; an equilateral triangle has three sides ofequal length; in a right triangle, the sum of the squares of the sides equalsthe square of the hypotenuse.”We have categories for all kinds of things, living and inanimate. Whenwe’re shown a new item—a new triangle, a dog we’ve never seenbefore—we assign the item to a category based on an analysis of itsproperties and a comparison with the category definition, according toAristotle. From Aristotle, through to Locke and the present day, cate-

You Know My Name, Look Up the Number 137gories were assumed to be a matter of logic, and objects were either insideor outside of a category.After 2,300 years of no substantial work on the topic, Ludwig Wittgensteinasked a simple question: What is a game? This launched a renaissanceof empirical work on category formation. Enter Eleanor Rosch,who did her undergraduate philosophy thesis at Reed College in Portland,Oregon, on Wittgenstein. Rosch had planned for years to go tograduate school in philosophy, but a year with Wittgenstein, she says,“cured her” of philosophy completely. Feeling that contemporary philosophyhad hit a dead end, Rosch wondered how she could study philosophicalideas empirically, how she could discover new philosophicalfacts. When I was teaching at UC Berkeley, where she is a professor, shetold me that she thought that philosophy had done all it could do with respectto problems of the brain and the mind, and that experimentationwas necessary to move forward. Today, following Rosch, many cognitivepsychologists consider an apt description of our field to be “empiricalphilosophy”; that is, experimental approaches to questions and problemsthat have been traditionally in the domain of philosophers: What isthe nature of mind? Where do thoughts come from? Rosch ended up atHarvard, and took her Ph.D. there in cognitive psychology. Her doctoralthesis changed the way we think about categories.Wittgenstein dealt the first blow to Aristotle by pulling the rug outfrom strict definitions of what a category is. Using the category “games”as an example, Wittgenstein argued that there is no definition or set ofdefinitions that can encompass all games. For example, we might saythat a game (a) is done for fun or recreation, (b) is a leisure activity, (c)is an activity most often found among children, (d) has certain rules, (e)is in some way competitive, (f) involves two or more people. Yet, we cangenerate counterexamples for each of these elements, showing that thedefinitions break down: (a) In the Olympic Games, are the athletes havingfun? (b) Is pro football a leisure activity? (c) Poker is a game, as is jaialai, but not most often found among children. (d) A child throwing aball against a wall is having fun, but what are the rules? (e) Ring-around-

138 This Is Your Brain on Musicthe-rosy isn’t competitive. (f) Solitaire doesn’t involve two or morepeople. How do we get out of this reliance on definitions? Is there an alternative?Wittgenstein proposed that category membership is determined notby a definition, but by family resemblance. We call something a game if itresembles other things we have previously called games. If we go to theWittgenstein family reunion, we might find that certain features areshared by members of the family, but that there is no single physical featurethat one absolutely, positively must have to be a family member. Acousin might have Aunt Tessie’s eyes; another might have the Wittgensteinchin. Some family members will have Grandpa’s forehead, otherswill have Grandma’s red hair. Rather than using a static list of definitions,family resemblance relies on a list of features that may or may not bepresent. The list may also be dynamic; at some point red hair may die outof the family line (if not dye out), and so we simply remove it from ourlist of features. If it pops up again several generations later, we can reintroduceit to our conceptual system. This prescient idea forms the basisfor the most compelling theory in contemporary memory research,the multiple-trace memory models that Douglas Hintzman worked on,and which have been recently taken up by a brilliant cognitive scientistnamed Stephen Goldinger from Arizona.Can we define music by definitions? What about types of music, suchas heavy metal, classical, or country? Such attempts would certainly failas they did for “games.” We could, for example, say that heavy metal is amusical genre that has (a) distorted electric guitars; (b) heavy, louddrums; (c) three chords, or power chords; (d) sexy lead singers, usuallyshirtless, dripping sweat and swinging the microphone stand around thestage like it was a piece of rope; (e) ümlauts in the gröup names. But thisstrict list of definitions is easy to refute. Although most heavy metalsongs have distorted electric guitars, so does “Beat It” by Michael Jackson—infact, Eddie Van Halen (the heavy metal god) plays the guitar soloin that song. Even the Carpenters have a song with a distorted guitar, andno one would call them “heavy metal.” Led Zeppelin—the quintessentialheavy metal band and arguably the band that spawned the genre—has

You Know My Name, Look Up the Number 139several songs with no distorted guitars at all (“Bron-y-aur,” “Down bythe Seaside,” “Goin’ to California,” “The Battle of Nevermore”). “Stairwayto Heaven” by Led Zeppelin is a heavy metal anthem, and there are noheavy, loud drums (or distorted guitars for that matter) in 90 percent ofthat song. Nor does “Stairway to Heaven” have only three chords. Andlots of songs have three chords and power chords that are not heavymetal, including most songs by Raffi. Metallica is a heavy metal band forsure, but I’ve never heard anyone call their lead singer sexy, and althoughMötley Crüe, Blue Öyster Cult, Motörhead, Spin¨ al Tap, andQueensrÿche have gratuitous umlauts, many heavy metal bands do not:Led Zeppelin, Metallica, Black Sabbath, Def Leppard, Ozzie Osbourne,Triumph, etc. Definitions of musical genres aren’t very useful; we saythat something is heavy metal if it resembles heavy metal—a family resemblance.Armed with her knowledge of Wittgenstein, Rosch decided thatsomething can be more or less a category member; rather than being allor none as Aristotle had believed, there are shades of membership, degreesof fit to a category, and subtle shadings. Is a robin a bird? Mostpeople would answer yes. Is a chicken a bird? Is a penguin? Most peoplewould say yes after a slight pause, but then would add that chickens andpenguins are not very good examples of birds, nor typical of the category.This is reflected in everyday speech when we use linguistic hedgessuch as “A chicken is technically a bird,” or “Yes, a penguin is a bird, butit doesn’t fly like most other birds.” Rosch, following Wittgenstein,showed that categories do not always have clear boundaries—they havefuzzy boundaries. Questions of membership are a matter of debate andthere can be differences of opinion: Is white a color? Is hip-hop reallymusic? If the surviving members of Queen perform without FreddieMercury, am I still seeing Queen (and is it worth $150 a ticket)? Roschshowed that people can disagree about categorizations (is a cucumber afruit or a vegetable?), and that the same person can even disagree withhimself at different times about a category (is so-and-so my friend?).Rosch’s second insight was that all of the experiments on categoriesthat had been done before her used artificial concepts and sets of artifi-

140 This Is Your Brain on Musiccial stimuli that had little to do with the real world. And these controlledlaboratory experiments were inadvertently constructed in ways thatended up with a bias toward the experimenters’ theories! This underscoresan ongoing problem that plagues all of empirical science: the tensionbetween rigorous experimental control and real-world situations.The trade-off is that in achieving one, there is often a compromise of theother. The scientific method requires that we control all possible variablesin order to be able to draw firm conclusions about the phenomenonunder study. Yet such control often creates stimuli or conditions thatwould never be encountered in the real world, situations that are so farremoved from the real world as not even to be valid. The British philosopherAlan Watts, author of The Wisdom of Insecurity, put it this way: Ifyou want to study a river, you don’t take out a bucketful of water andstare at it on the shore. A river is not its water, and by taking the waterout of the river, you lose the essential quality of river, which is its motion,its activity, its flow. Rosch felt that scientists had disrupted the flow ofcategories by studying them in such artificial ways. This, incidentally, isthe same problem with a lot of the research that has been done in theneuroscience of music for the past decade: Too many scientists study artificialmelodies using artificial sounds—things that are so removed frommusic, it’s not clear what we’re learning.Rosch’s third insight was that certain stimuli hold a privileged positionin our perceptual system or our conceptual system, and that thesebecome prototypes for a category: Categories are formed around theseprototypes. In the case of our perceptual system, categories like “red”and “blue” are a consequence of our retinal physiology; certain shades ofred are universally going to be regarded as more vivid, more central,than others because a specific wavelength of visible light will cause the“red” receptors in our retina to fire maximally. We form categoriesaround these central, or focal, colors. Rosch tested this idea on a tribe ofNew Guinea people, the Dani, who have only two words in their languagefor colors, mili and mola, which essentially correspond to lightand dark.

You Know My Name, Look Up the Number 141Rosch wanted to show that what we call red, and what we would pickout as an example of the best red, is not culturally determined orlearned. When shown a bunch of different shades of red, we don’t pick aparticular one because we’ve been taught that it is the best red, we pickit out because our physiology bestows a privileged perceptual positionon it. The Dani have no word for red in their language, and therefore notraining in what constitutes a good red versus a bad red. Rosch showedher Dani subjects chips colored with dozens of different shades of redand asked them to pick out the best example of this color. They overwhelminglyselected the same “red” that Americans do, and they werebetter at remembering it. And they did this for other colors that theycouldn’t name, like greens and blues. This led Rosch to conclude that (a)categories are formed around prototypes; (b) these prototypes can havea biological or physiological foundation; (c) category membership canbe thought of as a question of degree, with some tokens being “better”exemplars than others; (d) new items are judged in relation to the prototypes,forming gradients of category membership; and the final blow forAristotelian theory, (e) there don’t need to be any attributes which allcategory members have in common, and boundaries don’t have to bedefinite.We’ve done some informal experiments in my laboratory with musicalgenres and have found similar results. People appear to agree as towhat are prototypical songs for musical categories, such as “country music,”“skate punk,” and “baroque music.” They are also inclined to considercertain songs or groups as less good examples than the prototype:the Carpenters aren’t really rock music; Frank Sinatra is not really jazz,or at least not as much as John Coltrane is. Even within the category ofa single artist, people make graded distinctions that imply a prototypestructure. If you ask me to pick out a Beatles song, and I select “Revolution9” (an experimental tape piece written by John Lennon and Paul Mc-Cartney, with no original music, no melody or rhythm, which begins withan announcer repeating, “Number 9, Number 9,” over and over again)you might complain that I was being difficult. “Well, technically that’s a

142 This Is Your Brain on MusicBeatles piece—but that’s not what I meant!” Similarly, Neil Young’s onealbum of fifties doo-wop (Everybody’s Rockin’) is not representative (ortypical) Neil Young; Joni Mitchell’s jazz foray with Charles Mingus is notwhat we usually think of when we think of Joni Mitchell. (In fact, NeilYoung and Joni Mitchell were each threatened with contract cancellationsby their record labels for making music that was not deemed NeilYoung–like and Joni Mitchell–like, respectively.)Our comprehension of the world around us begins with specific andindividual cases—a person, a tree, a song—and through experience withthe world, these particular objects are almost always dealt with in ourbrains as members of a category. Roger Shepard has described the generalissue in all of this discussion in terms of evolution. There are threebasic appearance-reality problems that need to be solved by all higheranimals, he says. In order to survive, to find edible food, water, shelter,to escape predators, and to mate, the organism must deal with three scenarios.First, objects, though in presentation they may be similar, are inherentlydifferent. Objects that may create identical, or nearly identical,patterns of stimulation on our eardrums, retinas, taste buds, or touchsensors may actually be different entities. The apple I saw on the tree isdifferent from the one I am holding in my hand. The different violinsounds I hear coming from the symphony, even when they’re all playingthe same note, represent several different instruments.Second, objects, though in presentation they may be different, are inherentlyidentical. When we look at an apple from above, or from theside, it appears to be an entirely different object. Successful cognition requiresa computational system that can integrate these separate viewsinto a coherent representation of a single object. Even when our sensoryreceptors receive distinct and nonoverlapping patterns of activation, weneed to abstract out information that is critical to creating a unified representationof the object. Although I may be used to hearing your voicein person, through both ears, when I hear you over the phone, in one ear,I need to recognize that you’re the same person.

You Know My Name, Look Up the Number 143The third appearance-reality problem invokes higher-order cognitiveprocesses. The first two are perceptual processes: understanding that asingle object may manifest itself in multiple viewpoints, or that severalobjects may have (nearly) identical viewpoints. The third problem statesthat objects, although different in presentation, are of the same naturalkind. This is an issue in categorization, and it is the most powerful andadvanced principle of all. All higher mammals, many lower mammalsand birds, and even fish, can categorize. Categorization entails treatingobjects that appear different as of the same kind. A red apple may lookdifferent from a green apple, but they are both still apples. My motherand father may look very different, but they are both caregivers, to betrusted in an emergency.Adaptive behavior, then, depends on a computational system that cananalyze the information available at the sensory surfaces into (1) the invariantproperties of the external object or scene, and (2) the momentarycircumstances of the manifestation of that object or scene. Leonard Meyernotes that classification is essential to enable composers, performers, andlisteners to internalize the norms governing musical relationships, andconsequently, to comprehend the implications of patterns, and experiencedeviations from stylistic norms. Our need to classify, as Shakespeare saysin A Midsummer Night’s Dream, is to give “to airy nothing/A local habitationand a name.”Shepard’s characterization recast the categorization problem as an evolutionary/adaptiveone. In the meantime, Rosch’s work was beginning toshake up the research community, and dozens of leading cognitive psychologistsbegan to study to challenge her theory. Posner and Keele hadshown that people store prototypes in memory. In a clever experiment,they created tokens that contained patterns of dots placed in a square—something like the face of dice, but with the dots more or less randomlyplaced on each face. They called these the prototypes. Then they shiftedsome of the dots a millimeter or so in one random direction or another.This created a set of distortions from the prototype—that is, variations—

144 This Is Your Brain on Musicthat differed in their relationship to the prototype. Due to random variation,some of the tokens could not be easily identified with one prototypeor another, the distortions were just too great.This is like what a jazz artist does with a well-known song, or standard.When we compare Frank Sinatra’s version of “A Foggy Day” withthe version by Ella Fitzgerald and Louis Armstrong, we hear that some ofthe pitches and rhythms are the same and some are different; we expecta good vocalist to interpret the melody, even if that means changingit from the way the composer originally wrote it. In the courts of Europeduring the baroque and enlightenment eras, musicians like Bach andHaydn would regularly perform variations of themes. Aretha Franklin’sversion of “Respect” differs from that written and performed by OtisRedding in interesting ways—but we still consider it the same song.What does this say about prototypes and the nature of categories? Canwe say that the musical variations share a family resemblance? Are eachof these versions of a song variations on an ideal prototype?Posner and Keele addressed the general question of categories andprototypes using their dot stimuli. Subjects were shown pieces of paperwith version after version of these squares with dots in them, each ofthem different, but they were never shown the prototypes from whichthe variations were made. The subjects weren’t told how these dots patternshad been constructed, or that prototypes for these various formsexisted. A week later, they asked the subjects to look at more pieces ofpaper, some old and some new, and to indicate which ones they had seenbefore. The subjects were good at identifying which ones they had seenbefore and which ones they hadn’t. Now, unbeknownst to the subjects,Posner and Keele had slipped in the prototypes from which all the figureshad been derived. Astonishingly, the subjects often identified thetwo previously unseen prototypes as figures they had seen before. Thisprovided the foundation for an argument that prototypes are stored inmemory—how else could the subjects have misidentified the unseen tokens?In order to store in memory something that wasn’t seen, the memorysystem must be performing some operations on the stimuli; theremust be a form of processing going on at some stage that goes beyond

You Know My Name, Look Up the Number 145merely preserving the information that was presented. This seemed likethe death of any record-keeping theory; if prototypes are stored in memory,memory must be constructive.What we learned from Ben White, and subsequent work by Jay Dowlingof the University of Texas and others, is that music is quite robust inthe face of transformations and distortions of its basic features. We canchange all of the pitches used in the song (transposition), the tempo, andthe instrumentation, and the song is still recognized as the same song.We can change the intervals, the scales, even the tonality from major tominor or vice versa. We can change the arrangement—say from bluegrassto rock, or heavy metal to classical—and, as the Led Zepplin lyricgoes, the song remains the same. I have a recording of a bluegrass group,the Austin Lounge Lizards, playing “Dark Side of the Moon” by theprogressive rock group Pink Floyd, using banjos and mandolins. I haverecordings of the London Symphony Orchestra playing the songs of theRolling Stones and Yes. With such dramatic changes, the song is still recognizableas the song. It seems, then, that our memory system extractsout some formula or computational description that allows us to recognizesongs in spite of these transformations. It seems that the constructivistaccount most closely fits the music data, and from Posner and Keele,it fits visual cognition as well.In 1990, I took a course at Stanford called “Psychoacoustics and CognitivePsychology for Musicians,” jointly offered by the departments ofmusic and psychology. The course was team-taught by an all-star cast:John Chowning, Max Mathews, John Pierce, Roger Shepard, and PerryCook. Each student had to complete a research project, and Perry suggestedthat I look at how well people can remember pitches, and specificallywhether they can attach arbitrary labels to those pitches. Thisexperiment would unite memory and categorization. The prevailing theoriespredicted that there was no reason for people to retain absolutepitch information—the fact that people can so easily recognize tunes intransposition argues for that. And most people cannot name the notes,except for the one in ten thousand who have absolute pitch.Why is absolute pitch (AP) is so rare? People with AP can name notes

146 This Is Your Brain on Musicas effortlessly as most of us name colors. If you play someone with AP aC-sharp on the piano, he or she can tell you it was a C-sharp. Most peoplecan’t do that, of course—even most musicians can’t unless they’re lookingat your fingers. Most AP possessors can name the pitch of other sounds,too, like car horns, the hum of fluorescent lights, and knives clinkingagainst dinner plates. As we saw earlier, color is a psychophysical fiction—itdoesn’t exist in the world, but our brains impose a categoricalstructure, such as broad swatches of red or blue, on the unidimensionalcontinuum of frequency of light waves. Pitch is also a psychophysical fiction,the consequence of our brains’ imposing a structure on the unidimensionalcontinuum of frequency of the sound waves. We can instantlyname a color just by looking at it. Why can’t we name sounds just by listeningto them?Well, most of us can identify sounds as effortlessly as we identify colors;it’s simply not the pitch we identify, but rather, the timbre. We can instantlysay of a sound, “That’s a car horn,” or “That’s my grandmotherSadie with a cold,” or “That’s a trumpet.” We can identify tonal color, justnot pitch. Still, it remains an unsolved problem why some people haveAP and others don’t. The late Dixon Ward from the University of Minnesotanoted wryly that the real question isn’t “Why do only a few peoplehave AP?” but “Why don’t we all?”I read everything I could about AP. In the 130 years from 1860 to 1990,roughly a hundred research articles were published on the subject. In thefifteen years since 1990 there has been an equal number! I noticed that allthe AP tests required the subjects to use a specialized vocabulary—thenote names—that only musicians would know. There seemed to be noway to test for absolute pitch among nonmusicians. Or was there?Perry suggested that we find out how easily the proverbial man in thestreet could learn to name pitches by associating particular pitches witharbitrary names, like Fred or Ethel. We thought about using piano notes,pitch pipes, and all kinds of things (except for kazoos, for obvious reasons),and decided that we’d get a bunch of tuning forks and hand themout to nonmusicians. Subjects were instructed to bang the tuning forksagainst their knees several times a day for a week, hold it up to their ears,

You Know My Name, Look Up the Number 147and try to memorize the sound. We told half the people that their soundwas called Fred and we told the other half it was called Ethel (after theneighbors of Lucy and Ricky on I Love Lucy; their last name was Mertz,which rhymes with Hertz, a pleasing coincidence that we didn’t realizeuntil years later).Half of each group had forks tuned to middle C, the other half hadforks tuned to G. We turned them loose, then took the forks away fromthem for a week, and then had them come back into the laboratory. Halfof the subjects were asked to sing back “their pitch” and half were askedto pick it out from three notes that I played on a keyboard. The subjectswere overwhelmingly able to reproduce or recognize “their” note. Thissuggested to us that ordinary people could remember notes with arbitrarynames.This got us thinking about the role that names play in memory. Althoughthe course was over and I had handed in my term paper, we werestill curious about this phenomenon. Roger Shepard asked if nonmusiciansmight be able to remember the pitches of songs even though theydon’t have names for them. I told him about a study by Andrea Halpern.Halpern had asked nonmusicians to sing well-known songs such as“Happy Birthday” or “Frère Jacques” from memory on two different occasions.She found that although people tended not to sing in the samekeys as one another, they did tend to sing a song consistently, in thesame key from one occasion to the other. This suggested that they hadencoded the pitches of the songs in long-term memory.Naysayers suggested that these results could be accounted for withoutmemory for pitch if the subjects had simply relied on muscle memoryfor the position of their vocal chords from one time to another. (Tome, muscle memory is still a form of memory—labeling the phenomenondoes nothing to change it.) But an earlier study by Ward and his colleagueEd Burns from the University of Washington had shown thatmuscle memory isn’t actually all that good. They asked trained singerswith absolute pitch to “sight-read” from a musical score; that is, thesingers had to look at music they had never seen before and sing it usingtheir knowledge of absolute pitch and their ability to read music. This is

148 This Is Your Brain on Musicsomething they’re usually very good at. Professional singers can sightsingif you give them a starting pitch. Only professional singers with AP,however, can sing in the right key just by looking at the score; this is becausethey have some internal template, or memory, for how note namesand sounds match up with each other—that’s what AP is. Now, Ward andBurns had their AP singers wear headphones, and they blasted thesingers with noise so that they couldn’t hear what they were singing—they had to rely on muscle memory alone. The surprising finding wasthat their muscle memory didn’t do very well. On average, it only gotthem to within a third of an octave of the correct tone.We knew that nonmusicians tended to sing consistently. But wewanted to push the notion further—how accurate is the average person’smemory for music? Halpern had chosen well-known songs that don’thave a “correct” key—each time we sing “Happy Birthday,” we’re likelyto sing it in a different key; someone begins on whatever pitch firstcomes to mind and we follow. Folk and holiday songs are sung so oftenand by so many people that they don’t have an objectively correct key.This is reflected in the fact that there is no standard recording that couldbe thought of as a reference for these songs. In the jargon of my field, wewould say that a single canonical version does not exist.The opposite is true with rock/pop songs. Songs by the Rolling Stones,the Police, the Eagles, and Billy Joel do exist in a single canonical version.There is one standard recording (in most cases) and that is the only versionanyone has ever heard (with the exception of the occasional barband playing the song, or if we go see the group live). We’ve probablyheard these songs as many times as we’ve heard “Deck the Halls.” Butevery time we’ve heard, say, M. C. Hammer’s “U Can’t Touch This” or U2’s“New Year’s Day,” they’ve been in the same key. It is difficult to recall aversion other than the canonical one. After hearing a song thousands oftimes, might the actual pitches become encoded in memory?To study this, I used Halpern’s method of simply asking people to singtheir favorite songs. I knew from Ward and Burns that their muscle memorywouldn’t be good enough to get them there. In order to reproduce the

You Know My Name, Look Up the Number 149correct key, they’d have to be keeping stable, accurate memory traces ofpitches in their heads. I recruited forty nonmusicians from around campusand asked them to come into the laboratory and sing their favoritesong from memory. I excluded songs that existed in multiple versions andsongs that had been recorded more than once, which would exist outthere-in-the-worldin more than one key. I was left with songs for whichthere is a single well-known recording that is the standard, or reference—songs such as “Time and Tide” by Basia or “Opposites Attract” by PaulaAbdul (this was 1990, after all), as well as songs such as “Like a Virgin” byMadonna and “New York State of Mind” by Billy Joel.I recruited subjects with a vague announcement for a “memory experiment.”Subjects would receive five dollars for ten minutes. (This isusually how cognitive psychologists get subjects, by putting up noticesaround campus. We pay more for brain imaging studies, usually aroundfifty dollars, just because it is somewhat unpleasant to be in a confined,noisy scanner.) A lot of subjects complained vociferously upon discoveringthe details of the experiment. They weren’t singers, they couldn’tcarry a tune in a bucket, they were afraid they’d ruin my experiment. Ipersuaded them to try anyway. The results were surprising. The subjectstended to sing at, or very near, the absolute pitches of their chosensongs. I asked them to sing a second song and they did it again.This was convincing evidence that people were storing absolute pitchinformation in memory; that their memory representation did not justcontain an abstract generalization of the song, but details of a particularperformance. In addition to singing with the correct pitches, other performancenuances crept in; subjects’ reproductions were rich with thevocal affectations of the original singers. For example, they would reproducethe high-pitched “ee-ee” of Michael Jackson in “Billie Jean,” orthe enthusiastic “Hey!” of Madonna in “Like a Virgin”; the syncopation ofKaren Carpenter in “Top of the World” as well as the raspy voice of BruceSpringsteen on the first word of “Born in the U.S.A.” I created a tape thathad the subjects’ productions on one channel of a stereo signal and theoriginal recording on the other; it sounded as though the subjects were

150 This Is Your Brain on Musicsinging along with the record—but we hadn’t played the record to them,they were singing along with the memory representation in their head,and that memory representation was astonishingly accurate.Perry and I also found that the majority of subjects sang at the correcttempo. We checked to see if all the songs were merely sung at the sametempo to begin with, which would mean that people had simply encodedin memory a single, popular tempo. But this wasn’t the case, there was alarge range of tempos. In addition, in their own subjective accounts ofthe experiment, the subjects told us that they were “singing along withan image” or “recording” inside their heads. How does this mesh with aneural account of the findings?By now I was in graduate school with Mike Posner and Doug Hintzman.Posner, always on the watch for neural plausibility, told me aboutthe newest work of Petr Janata. Petr had just completed a study in whichhe kept track of people’s brain waves while they listened to music andwhile they imagined music. He used EEG, placing sensors that measureelectrical activity emanating from the brain across the surface of thescalp. Both Petr and I were surprised to see that it was nearly impossibleto tell from the data whether people were listening to or imagining music.The pattern of brain activity was virtually indistinguishable. Thissuggested that people use the same brain regions for remembering asthey do for perceiving.What does this mean exactly? When we perceive something, a particularpattern of neurons fire in a particular way for a particular stimulus.Although smelling a rose and smelling rotten eggs both invoke the olfactorysystem, they use different neural circuits. Remember, neurons canconnect to one another in millions of different ways. One configurationof a group of olfactory neurons may signal “rose” and another may signal“rotten eggs.” To add to the complexity of the system, even the same neuronsmay have different settings associated with a different event-in-theworld.The act of perceiving then entails that an interconnected set ofneurons becomes activated in a particular way, giving rise to our mentalrepresentation of the object that is out-there-in-the-world. Rememberingmay simply be the process of recruiting that same group of neurons we

You Know My Name, Look Up the Number 151used during perception to help us form a mental image during recollection.We re-member the neurons, pulling them together again from theirdisparate locations to become members of the original club of neuronsthat were active during perception.The common neural mechanisms that underlie perception of musicand memory for music help to explain how it is that songs get stuck inour heads. Scientists call these ear worms, from the German Ohrwurm,or simply the stuck song syndrome. There has been relatively little scientificwork done on the topic. We know that musicians are more likelyto have ear worm attacks than nonmusicians, and that people withobsessive-compulsive disorder (OCD) are more likely to report beingtroubled by ear worms—in some cases medications for OCD can minimizethe effects. Our best explanation is that the neural circuits representinga song get stuck in “playback mode,” and the song—or worse, alittle fragment of it—plays back over and over again. Surveys have revealedthat it is rarely an entire song that gets stuck, but rather a piece ofthe song that is typically less than or equal in duration to the capacity ofauditory short-term (“echoic”) memory: about 15 to 30 seconds. Simplesongs and commercial jingles seem to get stuck more often than complexpieces of music. This predilection for simplicity has a counterpartin our formation of musical preference, which I’ll discuss in Chapter 8.The findings from my study of people singing their favorite songs withaccurate pitch and tempo have been replicated by other laboratories, sowe know now that they’re not just the result of chance. Glenn Schellenbergat the University of Toronto—incidentally, an original member ofthe New Wave group Martha and the Muffins—performed an extensionof my study in which he played people snippets of Top 40 songs thatlasted a tenth of a second or so, about the same duration as a finger snap.People were given a list of song names and had to match them up withthe snippet they heard. With such a short excerpt, they could not rely onthe melody or rhythm to identify the songs—in every case, the excerptwas less than one or two notes. The subjects could only rely on timbre,the overall sound of the song. In the introduction, I mentioned the importancethat timbre holds for composers, songwriters, and producers.

152 This Is Your Brain on MusicPaul Simon thinks in terms of timbre; it is the first thing he listens for inhis music and the music of others. Timbre also appears to hold this privilegedposition for the rest of us; the nonmusicians in Schellenberg’sstudy were able to identify songs using only timbral cues a significantpercentage of the time. Even when the excerpts were presented backward,so that anything overtly familiar was disrupted, they still recognizedthe songs.If you think about the songs that you know and love, this should holdsome intuitive sense. Quite apart from the melody, the specific pitchesand rhythms, some songs simply have an overall sound, a sonic color. Itis similar to that quality that makes the plains of Kansas and Nebraskalook one way, the coastal forests of northern California, Oregon, andWashington another, the mountains of Colorado and Utah yet another.Before recognizing any details in a picture of these places, you apprehendthe overall scene, the landscape, the way that things look together.The auditory landscape, the soundscape, also has a presentation that isunique in much of the music we hear. Sometimes it is not song specific.This is what allows us to identify musical groups even when we cannotrecognize a specific song. Early Beatles albums have a particular timbralquality such that many people can identify a recording as the Beatles ifthey don’t immediately recognize the song—even if it is a song theynever heard before. This same quality allows us to identify imitations ofthe Beatles, when Eric Idle and his colleagues from Monty Python formedthe fictitious group the Rutles as a Beatles satire band, for example. By incorporatingmany of the distinctive timbral elements of the Beatlessoundscape, they were able to create a realistic satire that sounds like theBeatles.Overall timbral presentations, soundscapes, can also apply to wholeeras of music. Classical records from the 1930s and early 1940s havea particular sound to them due to the recording technology of the day.Nineteen eighties rock, heavy metal, 1940s dance hall music, and late1950s rock and roll are fairly homogeneous eras or genres. Record producerscan re-create these sounds in the studio by paying close attentionto details of the soundscape: the microphones they use, the way they

You Know My Name, Look Up the Number 153mix instruments, and so on. And many of us can hear a song and accuratelyguess what era it belongs to. One clue is often the echo, or reverberation,used on the voice. Elvis Presley and Gene Vincent had a verydistinctive “slap-back” echo, in which you hear a sort of instant repeat ofthe syllable the vocalist just sang. You hear it on “Be-Bop-A-Lula” byGene Vincent and by Ricky Nelson, on “Heartbreak Hotel” by Elvis, andon “Instant Karma” by John Lennon. Then there is the rich, warm echomade by a large tiled room on recordings by the Everly Brothers, such as“Cathy’s Clown” and “Wake Up Little Susie.” There are many distinctiveelements in the overall timbre of these records that we identify with theera in which they were made.Taken together, the findings from memory for popular songs providestrong evidence that absolute features of music are encoded in memory.And there is no reason to think that musical memory functions differentlyfrom, say, visual, olfactory, tactile, or gustatory memory. It wouldseem, then, that the record-keeping hypothesis has enough support forus to adopt it as a model for how memory works. But before we do, whatdo we do with the evidence supporting the constructivist theory? Sincepeople can so readily recognize songs in transposition, we need to accountfor how this information is stored and abstracted. And there is yetanother feature of music that is familiar to all of us, which an adequatetheory of memory needs to account for: We can scan songs in our mind’sear and we can imagine transformations of them.Here’s a demonstration, based on an experiment that Andrea Halpernconducted: Does the word at appear in the American national anthem(“The Star-Spangled Banner”)? Think about it before you read on.If you’re like most people, you “scanned” through the song in yourhead, singing it to yourself at a rapid rate, until you got to the phrase“What so proudly we hailed, at the twilight’s last gleaming.” Now, a numberof interesting things happened here. First, you probably sang thesong to yourself faster than you’ve ever heard it. If you were only able toplay back a particular version you had stored in memory, you wouldn’tbe able to do this. Second, your memory is not like a tape recorder; if youwant to speed up a tape recorder or video or film to make the song go

154 This Is Your Brain on Musicfaster, you have to also raise the pitch. But in our minds, we can varypitch and tempo independently. Third, when you did finally reach theword at in your mind—your “target” in answering the question I posed—you probably couldn’t help yourself from continuing, pulling up the restof the phrase, “the twilight’s last gleaming.” This suggests that our memoryfor music involves hierarchical encoding—not all words are equallysalient, and not all parts of a musical phrase hold equal status. We havecertain entry points and exit points that correspond to specific phrasesin the music—again, unlike a tape recorder.Experiments with musicians have confirmed this notion of hierarchicalencoding in other ways. Most musicians cannot start playing a pieceof music they know at any arbitrary location; musicians learn music accordingto a hierarchical phrase structure. Groups of notes form unitsof practice, these smaller units are combined into larger units, and ultimatelyinto phrases; phrases are combined into structures such asverses and choruses or movements, and ultimately everything is strungtogether as a musical piece. Ask a performer to begin playing from a fewnotes before or after a natural phrase boundary, and she usually cannotdo it, even when reading from a score. Other experiments have shownthat musicians are faster and more accurate at recalling whether a certainnote appears in a musical piece if that note is at the beginning of aphrase or is on a downbeat, rather than being in the middle of a phraseor on a weak beat. Even musical notes appear to fall into categories, asto whether they are the “important” notes of a piece or not. Many amateursingers don’t store in memory every note of a musical piece. Rather,we store the “important” tones—even without any musical training, weall have an accurate and intuitive sense of which those are—and westore musical contour. Then, when it comes time to sing, the amateurknows that she needs to go from this tone to that tone, and she fills in themissing tones on the spot, without having explicitly memorized each ofthem. This reduces memory load substantially, and makes for greater efficiency.From all of these phenomena, we can see that a principal developmentin memory theory over the last hundred years was its convergence

You Know My Name, Look Up the Number 155with the research on concepts and categories. One thing is for sure now:Our decision about which memory theory is right—the constructivist orthe record-keeping/tape-recorder theory—will have implications for theoriesof categorization. When we hear a new version of our favorite song,we recognize that it is fundamentally the same song, albeit in a differentpresentation; our brains place the new version in a category whose membersinclude all the versions of that song we’ve heard.If we’re real music fans, we might even displace a prototype in favorof another based on knowledge that we gain. Take, for example, the song“Twist and Shout.” You might have heard it countless times by live bandsin various bars and Holiday Inns, and you might also have heard therecordings by the Beatles and the Mamas and the Papas. One of these lattertwo versions may even be your prototype for the song. But if I tell youthat the Isley Brothers had a hit with the song two years before the Beatlesrecorded it, you might reorganize your category to accommodate thisnew information. That you can accomplish such reorganization basedon a top-down process suggests that there is more to categories thanRosch’s prototype theory states. Prototype theory has a close connectionto the constructivist theory of memory, in that details of individualcases are discarded, and the gist or abstract generalization is stored—both in the sense of what is being stored as a memory trace, and what isbeing stored as the central memory of the category.The record-keeping memory account has a correlate in categorizationtheory, too, and it is called exemplar theory. As important as prototypetheory was, and as well as it accounted for both our intuitions and experimentaldata on category formation, scientists started to find problemswith it in the 1980s. Led by Edward Smith, Douglas Medin, and BrianRoss, researchers identified some weaknesses in prototype theory. First,when the category is broad and category members differ widely, how canthere be a prototype? Think, for example, of the category “tool.” What isthe prototype for it? Or for the category “furniture”? What is the prototypicalsong by a female pop artist?Smith, Medin, Ross, and their colleagues also noticed that withinthese kinds of heterogeneous categories, context can have a strong im-

156 This Is Your Brain on Musicpact on what we consider to be the prototype. The prototypical tool at anautomobile repair garage is more likely to be a wrench than a hammer,but at a home construction site the opposite would be true. What is theprototypical instrument in a symphony orchestra? I’m willing to bet thatyou didn’t say “guitar” or “harmonica,” but asked the same question for acampfire I doubt you would say “French horn” or “violin.”Contextual information is part of our knowledge about categoriesand category members, and prototype theory doesn’t account for this.We know, for example, that within the category “birds” the ones that singtend to be small. Within the category “my friends,” there are some that Iwould let drive my car and some I wouldn’t (based on their accident historyand whether or not they have a license). Within the category “FleetwoodMac songs,” some are sung by Christine McVie, some by LindseyBuckingham, and some by Stevie Nicks. Then there is knowledge aboutthe three distinct eras of Fleetwood Mac: the blues years with Peter Greenon guitar, the middle pop years with Danny Kirwan, Christine McVie, andBob Welch as songwriters, and the later years after Buckingham-Nicksjoined. If I ask you for the prototypical Fleetwood Mac song, context isimportant. If I ask you for the prototypical Fleetwood Mac member,you’ll throw up your hands and tell me there is something wrong withthe question! Although Mick Fleetwood and John McVie, the drummerand bassist, are the only two members who have been with the groupfrom its beginning, it doesn’t seem quite right to say that the prototypicalmember of Fleetwood Mac is the drummer or the bassist, neither ofwhom sings or wrote the major songs. Contrast this with the Police, forwhom we might say that Sting was the prototypical member, as songwriter,singer, and bassist. But if someone said that, you could just asforcefully argue that she’s wrong, Sting is not the prototypical member,he is merely the best known and the most crucial member, not the samething. The trio we know as the Police is a small but heterogeneous category,and to talk about a prototypical member doesn’t seem to be inkeeping with the spirit of what a prototype is—the central tendency, theaverage, the seen or unseen object that is most typical of the category.Sting is not typical of the Police in the sense of being any kind of average;

You Know My Name, Look Up the Number 157he is rather atypical in that he is so much better known than the othertwo, Andy Summers and Stewart Copeland, and his history since the Policehas followed such a different course.Another problem is that, although Rosch doesn’t explicitly state this,her categories seem to take some time to form. Although she explicitly allowsfor fuzzy boundaries, and the possibility that a given object couldoccupy more than one category (“chicken” could occupy the categories“bird,” “poultry,” “barnyard animals,” and “things to eat”), there isn’t aclear provision for our being able to make up new categories on the spot.And we do this all the time. The most obvious example is when we makeplaylists for our MP3 players, or load up our car with CDs to listen to on along drive. The category “music I feel like listening to now” is certainly anew and dynamic one. Or consider this: What do the following items havein common: children, wallet, my dog, family photographs, and car keys?To many people, these are things to take with me in the event of a fire.Such collections of things form ad hoc categories, and we are adept atmaking these. We form them not from perceptual experience with thingsin-the-world,but from conceptual exercises such as the ones above.I could form another ad hoc category with the following story: “Carolwas in trouble. She had spent all her money and she wouldn’t be getting apaycheck for another three days. There was no food in the house.” Thisleads to the ad hoc functional category “ways to get food for the nextthree days,” which might include “go to a friend’s house,” “write a badcheck,” “borrow money from someone,” or “sell my copy of This Is YourBrain on Music.” Thus, categories are formed not just by matching properties,but by theories about how things are related. We need a theory ofcategory formation that will account for (a) categories that have no clearprototype, (b) contextual information, and (c) the fact that we form newcategories all the time, on the spot. To accomplish this, it seems that wemust have retained some of the original information from the items, becauseyou never know when you’re going to need it. If (according to theconstructivists) I’m only storing abstract, generalized gist information,how could I construct a category like “songs that have the word love inthem without having the word love in the title”? For example, “Here,

158 This Is Your Brain on MusicThere and Everywhere” (the Beatles), “Don’t Fear the Reaper” (Blue ÖysterCult), “Something Stupid” (Frank and Nancy Sinatra), “Cheek toCheek” (Ella Fitzgerald and Louis Armstrong), “Hello Trouble (Come OnIn)” (Buck Owens), “Can’t You Hear Me Callin’” (Ricky Skaggs).Prototype theory suggests the constructivist view, that an abstractgeneralization of the stimuli we encounter becomes stored. Smith andMedin proposed exemplar theory as an alternative. The distinguishingfeature of exemplar theory is that every experience, every word heard,every kiss shared, every object seen, every song you’ve ever listened to,is encoded as a trace in memory. This is the intellectual descendant ofthe so-called residue theory of memory proposed by the Gestalt psychologists.Exemplar theory accounts for how we are able to retain so many detailswith such accuracy. Under it, details and context are retained in theconceptual memory system. Something is judged a member of a categoryif it resembles other members of that category more than it resemblesmembers of an alternative, competing category. Indirectly, exemplar theorycan also account for the experiments that suggested that prototypesare stored in memory. We decide whether a token is a member of a categoryby comparing it with all the other category members—memories ofeverything we encountered that is a category member and every time weencountered it. If we are presented with a previously unseen prototype—as in the Posner and Keele experiment—we categorize it correctly andswiftly because it bears a maximum resemblance to all the other storedexamples. The prototype will be similar to examples from its own categoryand not similar to examples from alternative categories, so it remindsyou of examples from the correct category. It makes more matchesthan any previously seen example because, by definition, the prototype isthe central tendency, the average category member. This has powerfulimplications for how we come to enjoy new music we’ve never heard before,and how we can like a new song instantly—the topic of Chapter 6.The convergence of exemplar theory and memory theory comes in theform of a relatively new group of theories, collectively called “multipletracememory models.” In this class of models, each experience we have

You Know My Name, Look Up the Number 159is preserved with high fidelity in our long-term memory system. Memorydistortions and confabulations occur when, in the process of retrieving amemory, we either run into interference from other traces that are competingfor our attention—traces with slightly different details—or someof the details of the original memory trace have degraded due to normallyoccurring neurobiological processes.The true test of such models is whether they can account for and predictthe data on prototypes, constructive memory, and the formation andretention of abstract information—such as when we recognize a song intransposition. We can test the neural plausibility of these models throughneuroimaging studies. The director of the U.S. NIH (National Institutesof Health) brain laboratories, Leslie Ungerleider, and her colleagues performedfMRI studies showing that representations of categories are locatedin specific parts of the brain. Faces, animals, vehicles, foods, andso on have been shown to occupy specific regions of the cortex. Andbased on lesion studies, we’ve found patients who have lost the ability toname members of some categories, while other categories remain intact.These data speak to the reality of conceptual structure and conceptualmemory in the brain; but what about the ability to store detailed informationand still end up with a neural system that acts like it has storedabstractions?In cognitive science, when neurophysiological data is lacking, neuralnetwork models are often used to test theories. These are essentiallybrain simulations that run on computers, with models of neurons, neuronalconnections, and neuronal firings. The models replicate the parallelnature of the brain, and so are often referred to as parallel distributedprocessing or PDP models. David Rumelhardt from Stanford and Jay Mc-Clelland from Carnegie Mellon University were at the forefront of thistype of research. These aren’t ordinary computer programs. PDP modelsoperate in parallel (like real brains), they have several layers of processingunits (as do the layers of the cortex), the simulated neurons can beconnected in myriad different ways (like real neurons), and simulatedneurons can be pruned out of the network or added into the network asnecessary (just as the brain reconfigures neural networks as incoming

160 This Is Your Brain on Musicinformation arrives). By giving PDP models problems to solve—such ascategorization or memory storage and retrieval problems—we can learnwhether the theory in question is plausible; if the PDP model acts theway humans do, we take that as evidence that things may work in humansthat way as well.Douglas Hintzman built the most influential PDP model demonstratingthe neural plausibility of multiple-trace memory models. His model,named MINERVA after the Roman goddess of knowledge, was introducedin 1986. She stored individual examples of the stimuli she encountered,and still managed to produce the kind of behavior we would expectto see from a system that stored only prototypes and abstract generalizations.She did this in much the way that Smith and Medin describe, bycomparing new instances to stored instances. Stephen Goldinger foundfurther evidence that multiple-trace models can produce abstractionswith auditory stimuli, specifically with words spoken in specific voices.There is now an emerging consensus among memory researchers thatneither the record-keeping nor the constructivist view is correct, but thata third view, a hybrid of sorts, is the correct theory: the multiple-tracememory model. The experiments on the accuracy of memory for musicalattributes are consistent with the Hintzman/Goldinger multiple-tracemodels. This is the model that most closely resembles the exemplarmodel of categorization, for which there is also an emerging consensus.How does a multiple-trace memory model account for the fact thatwe extract invariant properties of melodies as we are listening to them?As we attend to a melody, we must be performing calculations on it; in additionto registering the absolute values, the details of its presentation—details such as pitch, rhythms, tempo, and timbre—we must also becalculating melodic intervals and tempo-free rhythmic information. Neuroimagingstudies from Robert Zatorre and his colleagues at McGill havesuggested this is the case. Melodic “calculation centers” in the dorsal(upper) temporal lobes—just above your ears—appear to be payingattention to interval size and distances between pitches as we listen tomusic, creating a pitch-free template of the very melodic values we willneed in order to recognize songs in transposition. My own neuroimaging

You Know My Name, Look Up the Number 161studies have shown that familiar music activates both these regions andthe hippocampus, a structure deep in the center of the brain that isknown to be crucial to memory encoding and retrieval. Together, thesefindings suggest that we are storing both the abstract and the specific informationcontained in melodies. This may be the case for all kinds ofsensory stimuli.Because they preserve context, multiple-trace memory models canalso explain how we sometimes retrieve old and nearly forgotten memories.Have you ever been walking down the street and suddenly smelledan odor that you hadn’t smelled in a long time, and that triggered a memoryof some long-ago event? Or heard an old song come on the radio thatinstantly retrieved deeply buried memories associated with when thatsong was first popular? These phenomena get to the heart of what itmeans to have memories. Most of us have a set of memories that we treatsomething like a photo album or scrapbook. Certain stories we are accustomedto telling to our friends and families, certain past experienceswe recall for ourselves during times of struggle, sadness, joy, or stress, toremind us of who we are and where we’ve been. We can think of this asthe repertoire of our memories, those memories that we are used to playingback, something like the repertoire of a musician and the pieces heknows how to play.According to the multiple-trace memory models, every experience ispotentially encoded in memory. Not in a particular place in the brain, becausethe brain is not like a warehouse; rather, memories are encoded ingroups of neurons that, when set to proper values and configured in aparticular way, will cause a memory to be retrieved and replayed in thetheater of our minds. The barrier to being able to recall everything wemight want to is not that it wasn’t “stored” in memory, then; rather, theproblem is finding the right cue to access the memory and properly configureour neural circuits. The more we access a memory, the more activebecome the retrieval and recollection circuits, and the more facilewe are with the cues necessary to get at the memory. In theory, if we onlyhad the right cues, we could access any past experience.Think for a moment of your third-grade teacher—this is probably

162 This Is Your Brain on Musicsomething you haven’t thought about in a long time, but there it is—aninstant memory. If you continue to think about your teacher, your classroom,you might be able to recall some other things about third grade suchas the desks in the classroom, the hallways of your school, your playmates.These cues are rather generic and not very vivid. However, if Icould show you your third-grade class photo, you might suddenly beginto recall all kinds of things you had forgotten—the names of your classmates,the subjects you learned in class, the games you played atlunchtime. A song playing comprises a very specific and vivid set of memorycues. Because the multiple-trace memory models assume that contextis encoded along with memory traces, the music that you havelistened to at various times in your life is cross-coded with the events ofthose times. That is, the music is linked to events of the time, and thoseevents are linked to the music.A maxim of memory theory is that unique cues are the most effectiveat bringing up memories; the more items or contexts a particular cue isassociated with, the less effective it will be at bringing up a particularmemory. This is why, although certain songs may be associated withcertain times of your life, they are not very effective cues for retrievingmemories from those times if the songs have continued to play all alongand you’re accustomed to hearing them—as often happens with classicrock stations or the classical radio stations that rely on a somewhat limitedrepertoire of “popular” classical pieces. But as soon as we hear asong that we haven’t heard since a particular time in our lives, the floodgatesof memory open and we’re immersed in memories. The song hasacted as a unique cue, a key unlocking all the experiences associatedwith the memory for the song, its time and place. And because memoryand categorization are linked, a song can access not just specific memories,but more general, categorical memories. That’s why if you hear one1970s disco song—“YMCA” by the Village People, for example—youmight find other songs from that genre playing in your head, such as “ILove the Nightlife” by Alicia Bridges and “The Hustle” by Van McCoy.Memory affects the music-listening experience so profoundly that itwould be not be hyperbole to say that without memory there would be

You Know My Name, Look Up the Number 163no music. As scores of theorists and philosophers have noted, as well asthe songwriter John Hartford in his song “Tryin’ to Do Something to GetYour Attention,” music is based on repetition. Music works because weremember the tones we have just heard and are relating them to the onesthat are just now being played. Those groups of tones—phrases—mightcome up later in the piece in a variation or transposition that tickles ourmemory system at the same time as it activates our emotional centers. Inthe past ten years, neuroscientists have shown just how intimately relatedour memory system is with our emotional system. The amygdala,long considered the seat of emotions in mammals, sits adjacent to thehippocampus, long considered the crucial structure for memory storage,if not memory retrieval. Now we know that the amygdala is involved inmemory; in particular, it is highly activated by any experience or memorythat has a strong emotional component. Every neuroimaging studythat my laboratory has done has shown amygdala activation to music,but not to random collections of sounds or musical tones. Repetition,when done skillfully by a master composer, is emotionally satisfying toour brains, and makes the listening experience as pleasurable as it is.

6. After Dessert, Crick Was StillFour Seats Away from MeMusic, Emotion, and the Reptilian BrainAs I’ve discussed, most music is foot-tapping music. We listen to musicthat has a pulse, something you can tap your foot to, or at leasttap the foot in your mind to. This pulse, with few exceptions, is regularand evenly spaced in time. This regular pulse causes us to expect eventsto occur at certain points in time. Like the clickety-clack of a railroadtrack, it lets us know that we’re continuing to move forward, that we’rein motion, that everything is all right.Composers sometimes suspend the sense of pulse, such as in the firstfew measures of Beethoven’s Fifth Symphony. We hear “bump-bumpbump-baaaah”and the music stops. We’re not sure when we’re going tohear a sound again. The composer repeats the phrase—using differentpitches—but after that second rest, we’re off and running, with a regularfoot-tappable meter. Other times, composers give us the pulse explicitly,but then intentionally soften its presentation before coming in with aheavy articulation of it for dramatic effect. “Honky Tonk Women” by theRolling Stones begins with cowbell, followed by drums, followed byelectric guitar; the meter stays the same and our sense of the beat does,too, but the intensity of the strong beats unfolds. (And when we listen onheadphones the cowbell comes out of only one ear for more dramatic effect.)This is typical of heavy metal and rock anthems. “Back in Black” by

166 This Is Your Brain on MusicAC/DC begins with the high-hat cymbal and muted guitar chords thatsound almost like a small snare drum for eight beats until the onslaughtof electric guitar comes in. Jimi Hendrix does the same thing in opening“Purple Haze”—eight quarter notes on the guitar and bass, single notesthat explicitly set up the meter for us before Mitch Mitchell’s thunderousdrums are ushered in. Sometimes composers tease us, setting up expectationsfor the meter and then taking them away before settling onsomething strong—a sort of musical joke that they let us in on. StevieWonder’s “Golden Lady” and Fleetwood Mac’s “Hypnotized” establish ameter that is changed when the rest of the instruments come in. FrankZappa was a master at this.Some types of music seem more rhythmically driven than others, ofcourse. Although “Eine Kleine Nachtmusik” and “Stayin’ Alive” both havea definable meter, the second one is more likely to make most people getup and dance (at least that’s the way we felt in the 1970s). In order to bemoved by music (physically and emotionally) it helps a great deal tohave a readily predictable beat. Composers accomplish this by subdividingthe beat in different ways, and accenting some notes differently thanothers; a lot of this has to do with performance as well. When we talkabout a great groove in music, we’re not talking in some jive sixtiesAustin Powers fab lingo, baby; we’re talking about the way in whichthese beat divisions create a strong momentum. Groove is that qualitythat moves the song forward, the musical equivalent to a book that youcan’t put down. When a song has a good groove, it invites us into a sonicworld that we don’t want to leave. Although we are aware of the pulse ofthe song, external time seems to stand still, and we don’t want the songto ever end.Groove has to do with a particular performer or particular performance,not with what is written on paper. Groove can be a subtle aspectof performance that comes and goes from one day to another, even withthe same group of musicians. And, of course, listeners disagree aboutwhether something has a good groove or not, but to establish some commonground for the topic here, most people feel that “Shout” by the IsleyBrothers and “Super Freak” by Rick James have a great groove, as does

After Dessert, Crick Was Still Four Seats Away from Me 167“Sledgehammer” by Peter Gabriel. “I’m On Fire” by Bruce Springsteen,“Superstition” by Stevie Wonder, and “Ohio” by the Pretenders all havegreat grooves, and are very different from one another. But there is noformula for how to create a great one, as every R & B musician who hastried to copy the groove of classic tunes like those by the Temptationsand Ray Charles will tell you. The fact that we can point to relatively fewsongs that have it is evidence that copying it is not so easy.One element that gives “Superstition” its great groove is Stevie Wonder’sdrumming. In the opening few seconds of “Superstition,” whenStevie’s high-hat cymbal is playing alone, you can hear part of the secretto the song’s groove. Drummers consider the high-hat to be their timekeeper.Even if listeners can’t hear it in a loud passage, the drummeruses it as a point of reference for himself. The beat Stevie plays on thehigh-hat is never exactly the same way twice; he throws in little extrataps, hits, and rests. Moreover, every note that he plays on the cymbalhas a slightly different volume—nuances in his performance that add tothe sense of tension. The snare drum starts with bum-(rest)-bum-bum-paand we’re into the high-hat pattern:DOOT-doot-doot-dootah DOOtah-doot-doot-dootahDOOT-daat-doot-dootah DOOT-dootah-dootah-dootThe genius of his playing is that he keeps us on our mental toes bychanging aspects of the pattern every time he plays it, holding justenough of it the same to keep us grounded and oriented. Here, he playsthe same rhythm at the beginning of each line, but changes the rhythm inthe second part of the line, in a “call-and-response” pattern. He also useshis skill as a drummer to alter the timbre of his high-hat in one key place:for the second note of the second line, in which he has kept the rhythmthe same, he hits the cymbal differently to make it “speak” in a separatevoice; if his cymbal were a voice, it’s as if he changed the vowel soundthat was speaking.Musicians generally agree that groove works best when it is notstrictly metronomic—that is, when it is not perfectly machinelike. Al-

168 This Is Your Brain on Musicthough some danceable songs have been made with drum machines(Michael Jackson’s “Billie Jean” and Paula Abdul’s “Straight Up,” for example),the gold standard of groove is usually a drummer who changesthe tempo slightly according to aesthetic and emotional nuances of themusic; we say then that the rhythm track, that the drums, “breathe.”Steely Dan spent months trying to edit, reedit, shift, push, and pull thedrum-machine parts on their album Two Against Nature in order to getthem to sound as if a human had played them, to balance groove withbreathing. But changing local, as opposed to global, tempos like thisdoesn’t change meter, the basic structure of the pulse; it only changesthe precise moment that beats will occur, not whether they group intwos, threes, or fours, and not the global pace of the song.We don’t usually talk about groove in the context of classical music,but most operas, symphonies, sonatas, concertos, and string quartetshave a definable meter and pulse, which generally corresponds to theconductor’s movements; the conductor is showing the musicians wherethe beats are, sometimes stretching them out or compressing them foremotional communication. Real conversations between people, realpleas of forgiveness, expressions of anger, courtship, storytelling, planning,and parenting don’t occur at the precise clips of a machine. To theextent that music is reflecting the dynamics of our emotional lives, andour interpersonal interactions, it needs to swell and contract, to speedup and slow down, to pause and reflect. The only way that we can feel orknow these timing variations is if a computational system in the brainhas extracted information about when the beats are supposed to occur.The brain needs to create a model of a constant pulse—a schema—sothat we know when the musicians are deviating from it. This is similar tovariations of a melody: We need to have a mental representation of whatthe melody is in order to know—and appreciate—when the musician istaking liberties with it.Metrical extraction, knowing what the pulse is and when we expect itto occur, is a crucial part of musical emotion. Music communicates to usemotionally through systematic violations of expectations. These violationscan occur in any domain—the domain of pitch, timbre, contour,

After Dessert, Crick Was Still Four Seats Away from Me 169rhythm, tempo, and so on—but occur they must. Music is organized sound,but the organization has to involve some element of the unexpected or itis emotionally flat and robotic. Too much organization may technicallystill be music, but it would be music that no one wants to listen to.Scales, for example, are organized, but most parents get sick of hearingtheir children play them after five minutes.What of the neural basis for this metrical extraction? From lesionstudies we know that rhythm and metrical extraction aren’t neurally relatedto each other. Patients with damage to the left hemisphere can losethe ability to perceive and produce rhythm, but they can still extract meter,and patients with damage to the right hemisphere have shown theopposite pattern. Both of these are neurally separate from melody processing:Robert Zatorre found that lesions to the right temporal lobe affectthe perception of melodies more than lesions to the left; IsabellePeretz discovered that the right hemisphere of the brain contains a contourprocessor that in effect draws an outline of a melody and analyzesit for later recognition, and this is dissociable from rhythm and meter circuitsin the brain.As we saw with memory, computer models can help us grasp the innerworkings of the brain. Peter Desain and Henkjan Honing of the Netherlandsdeveloped a computer model that could extract the beat from apiece of music. It relied mainly on amplitude, the fact that meter is definedby loud versus soft beats occurring at regular intervals of alternation.To demonstrate the effectiveness of their system—and becausethey recognize the value of showmanship, even in science—they hookedup the output of their system to a small electric motor mounted inside ashoe. Their beat-extraction demonstration actually tapped its foot (or atleast a shoe on a metal rod) to real pieces of music. I saw this demonstratedat CCRMA in the mid nineties. It was quite impressive. Spectators(I’m calling us that because the sight of men’s size-nine blackwingtip shoe hanging from a metal rod and connected via a snake ofwires to the computer was quite a spectacle) could give a CD to Desainand Honing, and their shoe would, after a few seconds of “listening,”start to tap against a piece of plywood. (When the demonstration was

170 This Is Your Brain on Musicover, Perry Cook went up to them and said, “Very nice work ...but doesit come in brown?”)Interestingly, the Desain and Honing system had some of the sameweaknesses that real, live humans do: It would sometimes tap its foot inhalf time or double time, compared to where professional musicians feltthat the beat was. Amateurs do this all the time. When a computerizedmodel makes similar mistakes to a human, it is even better evidence thatour program is replicating human thought, or at least the types of computationalprocesses underlying thought.The cerebellum is the part of the brain that is involved closely withtiming and with coordinating movements of the body. The word cerebellumderives from the Latin for “little brain,” and in fact, it looks like asmall brain hanging down underneath your cerebrum (the larger, mainpart of the brain), right at the back of your neck. The cerebellum has twosides, like the cerebrum, and each is divided into subregions. From phylogeneticstudies—studies of the brains of different animals up anddown the genetic ladder—we’ve learned that the cerebellum is one ofthe oldest parts of the brain, evolutionarily speaking. In popular language,it has sometimes been referred to as the reptilian brain. Althoughit weighs only 10 percent as much as the rest of the brain, it contains 50to 80 percent of the total number of neurons. The function of this oldestpart of the brain is something that is crucial to music: timing.The cerebellum has traditionally been thought of as that part of thebrain that guides movement. Most movements made by most animalshave a repetitive, oscillatory quality. When we walk or run, we tend to doso at a more or less constant pace; our body settles into a gait and wemaintain it. When fish swim or birds fly, they tend to flip their fins or flaptheir wings at a more or less constant rate. The cerebellum is involved inmaintaining this rate, or gait. One of the hallmarks of Parkinson’s diseaseis difficulty walking, and we now know that cerebellar degeneration accompaniesthis disease.But what about music and the cerebellum? In my laboratory we foundstrong activations in the cerebellum when we asked people to listen tomusic, but not when we asked them to listen to noise. The cerebellum

After Dessert, Crick Was Still Four Seats Away from Me 171appears to be involved in tracking the beat. And the cerebellum hasshown up in our studies in another context: when we ask people to listento music they like versus music they don’t like, or familiar music versusunfamiliar music.Many people, including ourselves, wondered if these cerebellar activationsto liking and familiarity were in error. Then, in the summer of2003, Vinod Menon told me about the work of Harvard professor JeremySchmahmann. Schmahmann has been swimming upstream against thetide of traditionalists who said that the cerebellum is for timing andmovement and nothing else. But through autopsies, neuroimaging, casestudies, and studies of other species, Schmahmann and his followershave amassed persuasive evidence that the cerebellum is also involvedin emotion. This would account for why it becomes activated whenpeople listen to music they like. He notes that the cerebellum containsmassive connections to emotional centers of the brain—the amygdala,which is involved in remembering emotional events, and the frontallobe, the part of the brain involved in planning and impulse control.What is the connection between emotion and movement, and why wouldthey both be served by the same brain region, a region found even insnakes and lizards? We don’t know for sure, but some informed speculationcomes through the very best of sources: the codiscoverers of DNA’sstructure, James Watson and Francis Crick.Cold Spring Harbor Laboratory is an advanced, high-tech institution onLong Island, specializing in research on neuroscience, neurobiology,cancer, and—as befits an institution whose director is the Nobel laureateJames Watson—genetics. Through SUNY Stony Brook, CSHL offers degreesand advanced training in these fields. A colleague of mine, AmandinePenel, was a postdoctoral fellow there for a couple of years. Shehad taken her Ph.D. in music cognition in Paris while I was earning mineat the University of Oregon; we knew each other from the annual musiccognition conferences. Every so often, CSHL sponsors a workshop, anintensive gathering of scientists who are specialists on a particular topic.These workshops span several days; everyone eats and sleeps at the lab-

172 This Is Your Brain on Musicoratory, and spends all day together hashing out the chosen scientificproblem. The idea behind such a gathering is that if the people whoare world experts on the topic—often contentiously holding oppositeviews—can come to some sort of an agreement about certain aspects ofthe problem, science can move forward more quickly. The CSHL workshopsare famous in genomics, plant genetics, and neurobiology.I was taken by surprise one day when, buried in between rather mundanee-mails about the undergraduate curriculum committee and finalexamination schedules at McGill, I saw one inviting me to participate ina four-day workshop at Cold Spring Harbor. Here is what I found in myin-box:Neural Representation and Processing of Temporal PatternsHow is time represented in the brain? How are complex temporalpatterns perceived or produced? Processing of temporal patternsis a fundamental component of sensory and motor function. Giventhe inherent temporal nature of our interaction with the environment,understanding how the brain processes time is a necessarystep towards understanding the brain. We aim to bring togetherthe top psychologists, neuroscientists, and theorists in the worldworking on these problems. Our goals are twofold: First, we wishto bring together researchers from different fields that share acommon focus on timing and would benefit greatly from crossfertilizationof ideas. Second, much significant work to date hasbeen carried out on single-temporal-interval processing. Looking tothe future, we wish to learn from these studies while extending thediscussion to the processing of temporal patterns that are composedof multiple intervals. Temporal pattern perception is growingas a multi-disciplinary field; we anticipate that this meetingmay help to discuss and set a cross-disciplinary research agenda.At first, I thought that the organizers had made a mistake by includingmy name on the list. I knew all the names of the invited participants that

After Dessert, Crick Was Still Four Seats Away from Me 173came with the e-mail. They were the giants in my field—the GeorgeMartins and Paul McCartneys, the Seiji Ozawas and Yo-Yo Mas of timingresearch. Paula Tallal had discovered, with her collaborator MikeMerzenich of UC San Francisco, that dyslexia was related to a timingdeficit in children’s auditory systems. She had also published some of themost influential fMRI studies of speech and the brain, showing where inthe brain phonetic processing occurs. Rich Ivry was my intellectualcousin, one of the brightest cognitive neuroscientists of my generation,who had received his Ph.D. from Steve Keele at the University of Oregonand had done ground-breaking work on the cerebellum and on thecognitive aspects of motor control. Rich has a very low-key, down-toearthmanner, and he can cut to the heart of a scientific issue with razorprecision.Randy Gallistel was a top mathematical psychologist who modeledmemory and learning processes in humans and mice; I had read his papersforward and backward. Bruno Repp had been Amandine Penel’sfirst postdoctoral advisor, and had been a reviewer on the first two papersI ever published (the experiments of people singing pop songs verynear the correct pitch and tempo). The other world expert on musicaltiming, Mari Reiss Jones, was also invited. She had done the most importantwork on the role of attention in music cognition, and had an influentialmodel of how musical accents, meter, rhythm, and expectationsconverge to create our knowledge of musical structure. And John Hopfield,the inventor of Hopfield nets, one of the most important classes ofPDP neural-network models, was going to be there! When I arrived atCold Spring Harbor, I felt like a girl backstage at a 1957 Elvis concert.The conference was intense. Researchers there couldn’t agree onbasic issues, such as how to distinguish an oscillator from a timekeeper,or whether different neural processes were involved in estimating thelength of a silent interval, versus the length of a time span that was filledwith regular pulses.As a group, we realized—just as the organizers had hoped—thatmuch of what impeded true progress in the field was that we were usingdifferent terminology to mean the same things, and in many cases, we

174 This Is Your Brain on Musicwere using a single word (such as timing) to mean very different things,and following very different elementary assumptions.When you hear someone use a word like planum temporale (a neuralstructure), you think he’s using it the same way you are. But in science,as in music, assumptions can be the death of you. One person consideredthat the planum temporale had to be defined anatomically, anotherthat it had to be defined functionally. We argued about the importance ofgray matter versus white matter, about what it means to have two eventsbe synchronous—do they actually have to happen at exactly the sametime, or just at what appears perceptually to be the same time?At night, we had catered dinners and lots of beer and red wine, andwe continued discussions as we ate and drank. My doctoral studentBradley Vines came down as an observer, and he played saxophone foreveryone. I played guitar with a few of the group who were musicians,and Amandine sang.Because the meeting was about timing, most of the people therehadn’t paid much attention to Schmahmann’s work or to the possibleconnection between emotion and the cerebellum. But Ivry had; he knewSchmahmann’s work and was intrigued by it. In our discussions, he casta light on similarities between music perception and motor action planning,which I hadn’t been able to see in my own experiment. He agreedthat the heart of the mystery of music must involve the cerebellum.When I met Watson, he told me he also felt there to be a plausible connectionamong the cerebellum, timing, music, and emotion. But whatcould that connection be? What was its evolutionary basis?A few months later, I visited my close collaborator Ursula Bellugi atthe Salk Institute, in La Jolla, California. The Salk Institute sits on a pristinepiece of land overlooking the Pacific Ocean. Bellugi, a student of thegreat Roger Brown at Harvard in the 1960s, runs the Cognitive NeuroscienceLaboratory there. Among many, many “firsts” and landmark findingsin her career, she was the first to show that sign language is truly alanguage (with syntactic structure, it is not just an ad hoc or disorganizedbunch of gestures), and she thus showed that Chomsky’s linguisticmodule is not for spoken language only. She also has done ground-

After Dessert, Crick Was Still Four Seats Away from Me 175breaking work on spatial cognition, gesture, neurodevelopmental disorders,and the ability of neurons to change function—neuroplasticity.Ursula and I have been working together for ten years to uncover thegenetic basis of musicality. What better place was there for the researchto be based than an institute headed by Francis Crick, the man who, withWatson, discovered the structure of DNA? I had gone there, as I do everyyear, so that we could look at our data together, and work on preparingarticles for publication. Ursula and I like sitting in the same room together,looking at the same computer screen, where we can point to thechromosome diagrams, look at brain activations, and talk over whatthey mean for our hypotheses.Once a week, the Salk Institute had a “professors’ lunch” at whichvenerable scientists sat around a large square table with Francis Crick,the Institute’s director. Visitors were seldom allowed; this was a privateforum at which scientists felt free to speculate. I’d heard of this hallowedground and dreamed of visiting it.In Crick’s book The Astonishing Hypothesis, he argued that consciousnessarises from the brain, that the sum total of our thoughts, beliefs,desires, and feelings comes from the activities of neurons, glialcells, and the molecules and atoms that make them up. This was interesting,but as I’ve said, I am somewhat biased against mapping the mindfor mapping’s own sake, and biased toward understanding how the machinerygives rise to human experience.What really made Crick interesting to me was not his brilliant workon DNA or his stewardship of the Salk Institute, or even The AstonishingHypothesis. It was his book What Mad Pursuit, about his earlyyears in science. In fact, it was precisely this passage, because I, too, hadbegun my scientific career somewhat late in life.When the war finally came to an end, I was at a loss as to what todo....I took stock of my qualifications. A not-very-good degree,redeemed somewhat by my achievements at the Admiralty. Aknowledge of certain restricted parts of magnetism and hydrodynamics,neither of them subjects for which I felt the least bit of

176 This Is Your Brain on Musicenthusiasm. No published papers at all. ...Only gradually did I realizethat this lack of qualification could be an advantage. By thetime most scientists have reached age thirty they are trapped bytheir own expertise. They have invested so much effort in one particularfield that it is often extremely difficult, at that time in theircareers, to make a radical change. I, on the other hand, knew nothing,except for a basic training in somewhat old-fashioned physicsand mathematics and an ability to turn my hand to new things. . . .Since I essentially knew nothing, I had an almost completely freechoice....Crick’s own search had encouraged me to take my lack of experience asa license to think about cognitive neuroscience differently than otherpeople, and it inspired me to reach beyond what seemed to be the shallowlimits of my own grasp.I drove to Ursula’s lab from my hotel one morning to get an early start.“Early” for me was seven A.M., but Ursula had been in the lab since six.While we worked together in her office, typing on our computer keyboards,Ursula put down her coffee and looked at me with a pixieliketwinkle in her eye. “Would you like to meet Francis today?” The coincidenceof my having met Watson, Crick’s Nobel laureate twin, only a fewmonths before was striking.I felt a rush of panic as an old memory assaulted me. When I was justgetting started as a record producer, Michelle Zarin, the manager of thetop recording studio in San Francisco, the Automatt, would have Fridayafternoon wine-and-cheese get-togethers in her office to which only theinner circle were invited. For months as I worked with unknown bandslike the Afflicted and the Dimes, I saw rock’s royalty file into her officeon Friday afternoons: Carlos Santana, Huey Lewis, the producers JimGaines and Bob Johnston. One Friday she told me that Ron Nevison wasgoing to be in town—he had engineered my favorite Led Zeppelinrecords, and had worked with the Who. Michelle led me into her officeand showed me where to stand in the semicircle that began to form.People drank and chatted, and I listened respectfully. But Ron Nevison

After Dessert, Crick Was Still Four Seats Away from Me 177seemed oblivious to me, and he was the one I really wanted to meet. Ilooked at my watch—fifteen minutes went by. Boz Scaggs (anotherclient) was on the stereo in the corner. “Lowdown.” “Lido.” Twenty minuteshad gone by. Was I ever going to meet Nevison? “We’re All Alone”came on, and—as music can sometimes do—the lyrics got under my skin.I had to take matters into my own hands. I walked over to Nevison and introducedmyself. He shook my hand and returned to the conversation hewas having. That was it. Michelle scolded me later—this sort of thing issimply not done. If I had waited until she introduced me, she would havereminded him that I was the young producer she had spoken to himabout, the potential apprentice, the respectful and thoughtful young manthat she wanted him to meet. I never saw Nevison again.At lunchtime, Ursula and I walked out into the warm spring San Diegoair. I could hear seagulls calling overhead. We walked to the corner ofthe Salk campus with the best view of the Pacific, and walked up threeflights of stairs to the professors’ lunchroom. I immediately recognizedCrick, although he looked quite frail—he was in his late eighties, knockingtentatively on ninety’s door. Ursula showed me to a seat about fourpeople away from him to his right.The lunch conversation was a cacophony. I heard snippets of conversationsabout a cancer gene that one of the professors had just identified,and about decoding the genetics of the visual system in the squid. Someoneelse was speculating on a pharmaceutical intervention to slow thememory loss associated with Alzheimer’s. Crick mostly listened, but heoccasionally spoke, in a voice so soft I couldn’t hear a word. The lunchroomthinned out as the professors finished eating.After dessert, Crick was still four seats away from me, animatedlytalking to someone on his left, facing away from us. I wanted to meethim, to talk about The Astonishing Hypothesis, to find out what hethought about the relationship among cognition, emotion, and motorcontrol. And what did the codiscoverer of DNA’s structure have to sayabout a possible genetic basis for music?Ursula, sensing my impatience, said that she’d introduce me to Francison our way out. I was disappointed, anticipating a “hello-goodbye.”

178 This Is Your Brain on MusicUrsula took me by the elbow; she is only four foot ten and has to reachup to get to my elbow. She brought me over to Crick, who was talkingabout leptons and muons with a colleague. She interrupted him. “Francis,”she said, “I just wanted to introduce you to my colleague Dan Levitin,from McGill, who works on Williams and music with me.” Before Crickcould say a word, Ursula pulled me by the elbow toward the door. Crick’seyes lit up. He sat up straight in his chair. “Music,” he said. He brushedaway his lepton colleague. “I’d like to talk to you about that sometime,”he said. “Well,” Ursula said slyly, “we have some time right now.”Crick wanted to know if we had done any neuroimaging studies ofmusic; I told him about our studies on music and the cerebellum. He wasintrigued by our results, and at the possibility that the cerebellum mightbe involved in musical emotion. The cerebellum’s role in helping performersand conductors keep track of musical time and to maintain aconstant tempo was well known. Many also assumed it was involved inkeeping track of musical time in listeners. But where did emotion fit in?What might have been the evolutionary connection between emotion,timing, and movement?To begin with, what might be the evolutionary basis for emotions?Scientists can’t even agree about what emotions are. We distinguishbetween emotions (temporary states that are usually the result of someexternal event, either present, remembered, or anticipated), moods (notso-temporary,longer-lasting states that may or may not have an externalcause), and traits (a proclivity or tendency to display certain states, suchas “She is generally a happy person,” or “He never seems satisfied”).Some scientists use the word affect to refer to the valence (positive ornegative) of our internal states, and reserve the word emotion to refer toparticular states. Affect can thus take on only two values (or a thirdvalue if you count “no affective state”) and within each we have a rangeof emotions: Positive emotions would include happiness and satiety,negative would include fear and anger.Crick and I talked about how in evolutionary history, emotions wereclosely associated with motivation. Crick reminded me that emotionsfor our ancient hominid ancestors were a neurochemical state that

After Dessert, Crick Was Still Four Seats Away from Me 179served to motivate us to act, generally for survival purposes. We see alion and that instantly generates fear, an internal state—an emotion—that results when a particular cocktail of neurotransmitters and firingrates is achieved. This state that we call “fear” motivates us to stop whatwe’re doing and—without thinking about it—run. We eat a piece of badfood and we feel the emotion of disgust; immediately certain physiologicalreflexes kick in, such as a scrunching up of the nose (to avoid lettingin a possible toxic odor) and a sticking out of the tongue (to eject the offendingfood); we also constrict our throat to limit the amount of foodthat gets into our stomach. We see a body of water after we’ve been wanderingfor hours, and we’re elated—we drink and the satiety fills us witha sense of well-being and contentment, emotions that cause us to rememberwhere that watering hole is for next time.Not all emotional activities lead to motor movements, but many ofthe important ones do, and running is prime among them. We can runfaster and far more efficiently if we do so with a regular gait—we’re lesslikely to stumble or lose our balance. The role of the cerebellum is clearhere. And the idea that emotions might be bound up with cerebellar neuronsmake sense too. The most crucial survival activities often involverunning—away from a predator or toward escaping prey—and our ancestorsneeded to react quickly, instantly, without analyzing the situationand studying the best course of action. In short, those of our ancestorswho were endowed with an emotional system that was directly connectedto their motor system could react more quickly, and thus live toreproduce and pass on those genes to another generation.What really interested Crick wasn’t evolutionary origins of behaviorso much as the data. Crick had read the work of Schmahmann, who wasattempting to resurrect many old ideas that had fallen into disfavor orhad simply been forgotten, such as a 1934 paper suggesting that the cerebellumwas involved in the modulation of arousal, attention, and sleep.During the 1970s, we learned that lesions to particular regions of thecerebellum could cause dramatic changes in arousal. Monkeys with a lesionto one portion of their cerebellum would experience rage—calledsham rage by scientists because there was nothing in the environment to

180 This Is Your Brain on Musiccause this reaction. (Of course, the monkeys had every reason to be enragedbecause some surgeon had just lesioned parts of their brains, butthe experiments show that they only exhibit rage after these cerebellar—but not other—lesions.) Lesions to other parts of the cerebellum causecalm and have been used clinically to soothe schizophrenics. Electricalstimulation of a thin strip of tissue at the center of the cerebellum, calledthe vermis, can lead to aggression in humans, and in a different region toa reduction in anxiety and depression.Crick’s dessert plate was still in front of him, and he pushed it away.He clutched a glass of ice water in his hands. I could see the veins of hishands through his skin. For a moment I thought I could actually see hispulse. He became quiet for a moment, staring, thinking. The room wascompletely still now, but through an open window we could hear thecrashing of the waves below.We discussed the work of neurobiologists who had shown in the1970s that the inner ear doesn’t send all of its connections to the auditorycortex, as was previously believed. In cats and rats, animals whose auditorysystems are well known and bear a marked resemblance to our own,there are projections directly from the inner ear to the cerebellum—connections that come into the cerebellum from the ear—that coordinatethe movements involved in orienting the animal to an auditorystimulus in space. There are even location-sensitive neurons in the cerebellum,an efficient way of rapidly orienting the head or body to a source.These areas in turn send projections out to the areas in the frontal lobethat my studies with Vinod Menon and Ursula found to be active in processingboth language and music—regions in the inferior frontal andorbitofrontal cortex. What was going on here? Why would the connectionsfrom the ear bypass the auditory cortex, the central receiving areafor hearing, and send masses of fibers to the cerebellum, a center of motorcontrol (and perhaps, we were learning, of emotion)?Redundancy and distribution of function are crucial principles ofneuroanatomy. The name of the game is that an organism has to live longenough to pass on its genes through reproduction. Life is dangerous;there are a lot of opportunities to get whacked in the head and poten-

After Dessert, Crick Was Still Four Seats Away from Me 181tially lose some brain function. To continue to function after a brain injuryrequires that a blow to a single part of the brain doesn’t shut downthe whole system. Important brain systems evolved additional, supplementarypathways.Our perceptual system is exquisitely tuned to detect changes in theenvironment, because change can be a signal that danger is imminent.We see this in each of the five senses. Our visual system, while endowedwith a capacity to see millions of colors and to see in the dark when illuminationis as dim as one photon in a million, is most sensitive to suddenchange. An entire region of the visual cortex, area MT, specializes in detectingmotion; neurons there fire when an object in our visual fieldmoves. We’ve all had the experience of an insect landing on our neck andwe instinctively slap it—our touch system noticed an extremely subtlechange in pressure on our skin. And although it is now a staple of children’scartoons, the power of a change in smell—the odor wafting through theair from an apple pie cooling on a neighbor’s windowsill—can cause analerting and orienting reaction in us. But sounds typically trigger thegreatest startle reactions. A sudden noise causes us to jump out of ourseats, to turn out heads, to duck, or to cover our ears.The auditory startle is the fastest and arguably the most importantof our startle responses. This makes sense: In the world we live in,surrounded by a blanket of atmosphere, the sudden movement of anobject—particularly a large one—causes an air disturbance. This movementof air molecules is perceived by us as sound. The principle of redundancydictates that our nervous system needs to be able to react tosound input even if it becomes partially damaged. The deeper we look insidethe brain, the more we find redundant pathways, latent circuits, andconnections among systems that we weren’t aware of before. These secondarysystems serve an important survival function. The scientific literaturehas recently featured articles on people whose visual pathwayswere cut, but who can still “see.” Although they aren’t consciously awareof seeing anything—in fact they claim to be blind—they can still orienttoward objects, and in some cases identify them.A vestigial or supplementary auditory system also appears to be in

182 This Is Your Brain on Musicplace involving the cerebellum. This preserves our ability to reactquickly—emotionally and with movement—to potentially dangeroussounds.Related to the startle reflex, and to the auditory system’s exquisitesensitivity to change, is the habituation circuit. If your refrigerator has ahum, you get so used to it that you no longer notice it—that is habituation.A rat sleeping in his hole in the ground hears a loud noise above.This could be the footstep of a predator, and he should rightly startle.But it could also be the sound of a branch blowing in the wind, hitting theground above him more or less rhythmically. If, after one or two dozentaps of the branch against the roof of his house, he finds he is in no danger,he should ignore these sounds, realizing that they are no threat. If theintensity or frequency should change, this indicates that environmentalconditions have changed and that he should start to notice. Maybe thewind has picked up and its added velocity will cause the branch to pokethrough his rodentine residence. Maybe the wind has died down, and itis safe for him to go out and seek food and mates without fear of beingblown away by torrential winds. Habituation is an important and necessaryprocess to separate the threatening from the nonthreatening. Thecerebellum acts as something of a timekeeper, so when it is damaged, itsability to track the regularity of sensory stimulation is compromised, andhabituation goes out the window.Ursula told Crick of Albert Galaburda’s discovery, at Harvard, that individualswith Williams syndrome (WS) have defects in the way theircerebellums form. Williams occurs when about twenty genes turn upmissing on one chromosome (chromosome 7). This happens in one outof twenty thousand births, and so it is about one fourth as common asthe better-known developmental disorder Down syndrome. Like Downsyndrome, Williams results from a mistake in genetic transcription thatoccurs early in the stages of fetal development. Out of the twenty-fivethousand or so genes that we have, the loss of these twenty is devastating.People with Williams can end up with profound intellectual impairment.Few of them learn to count, tell time, or read. Yet, they have moreor less intact language skills, they are very musical, and they are unusu-

After Dessert, Crick Was Still Four Seats Away from Me 183ally outgoing and pleasant; if anything, they are more emotional than therest of us, and they are certainly more friendly and gregarious than theaverage person. Making music and meeting new people tend to be two oftheir favorite things to do. Schmahmann had found that lesions to thecerebellum can create Williams-like symptoms, with people suddenly becomingtoo outgoing, and acting overly familiar with strangers.A couple of years ago I was asked to visit a teenage boy with WS.Kenny was outgoing, cheerful, and loved music, but he had an IQ of lessthan fifty, meaning that at the age of fourteen he had the mental capacityof a seven-year-old. In addition, as with most people struck with Williamssyndrome, he had very poor eye-hand coordination, and had difficultybuttoning up his sweater (his mother had to help him), tying his ownshoes (he had Velcro straps instead of laces), and he even had difficultyclimbing stairs or getting food from his plate to his mouth. But he playedthe clarinet. There were a few pieces that he had learned, and he wasable to execute the numerous and complicated finger movements to playthem. He could not name the notes, and couldn’t tell me what he was doingat any one point of the piece—it was as though his fingers had a mindof their own. Suddenly the eye-hand coordination problems were gone!But then as soon as he stopped playing, he needed help opening the caseto put the clarinet back.Allan Reiss at Stanford University Medical School has shown that theneocerebellum, the newest part of the cerebellum, is larger than normalin those with WS. Something about movement when it could be entrainedto music was different in people with WS than other kinds of movement.Knowing that their cerebellar morphometry was different from others’suggested that the cerebellum might be the part of them that had a “mindof its own,” and that could tell us something about how the cerebellumnormally influences music processing in people without WS. The cerebellumis central to something about emotion—startle, fear, rage, calm,gregariousness. It was now implicated in auditory processing.Still sitting with me, long after the lunch plates were cleared, Crickmentioned “the binding problem,” one of the most difficult problems incognitive neuroscience. Most objects have a number of different fea-

184 This Is Your Brain on Musictures that are processed by separate neural subsystems—in the case ofvisual objects, these might be color, shape, motion, contrast, size, and soon. Somehow the brain has to “bind together” these different, distinctcomponents of perception into a coherent whole. I have described howcognitive scientists believe that perception is a constructive process, butwhat are the neurons actually doing to bring it all together? We knowthis is a problem from the study of patients with lesions or particularneuropathic diseases such as Balint’s syndrome, in which people canrecognize only one or two features of an object but cannot hold them together.Some patients can tell you where an object is in their visual fieldbut not its color, or vice versa. Other patients can hear timbre andrhythm but not melody or vice versa. Isabelle Peretz discovered a patientwho has absolute pitch but is tone deaf ! He can name notes perfectly,but he cannot sing to save his life.One solution to the binding problem, Crick proposed, was the synchronousfiring of neurons throughout the cortex. Part of the “astonishinghypothesis” of Crick’s book was that consciousness emerges fromthe synchronous firing, at 40 Hz, of neurons in the brain. Neuroscientistshad generally considered that the operations of the cerebellum occurredat a “preconscious” level because it coordinates things like running,walking, grasping, and reaching that are normally not under consciouscontrol. There’s no reason that the cerebellar neurons can’t fire at 40 Hzto contribute to consciousness, he said, although we don’t normally attributehumanlike consciousness to those organisms that have only acerebellum, such as the reptiles. “Look at the connections,” Crick said.Crick had taught himself neuroanatomy during his time at Salk, and hehad noticed that many researchers in cognitive neuroscience were notadhering to their own founding principles, to use the brain as a constraintfor hypotheses; Crick had little patience for such people, and believedthat true progress would only be made by people rigorouslystudying details about brain structure and function.The lepton colleague was now back, reminding Crick of an impendingappointment. We all stood up to leave, and Crick turned to me one last

After Dessert, Crick Was Still Four Seats Away from Me 185time and repeated, “Look at the connections. . . .” I never saw him again.He died a few months later.The connection between the cerebellum and music wasn’t that hardto see. The Cold Spring Harbor participants were talking about how thefrontal lobe—the center of the most advanced cognitions in humans—isconnected directly to the cerebellum, the most primitive part of the humanbrain. The connections run in both directions, with each structureinfluencing the other. Regions in the frontal cortex that Paula Tallal wasstudying—those that help us to distinguish precise differences in speechsounds—were also connected to the cerebellum. Ivry’s work on motorcontrol showed connections between the frontal lobes, occipital cortex(and the motor strip), and the cerebellum. But there was another playerin this neural symphony, a structure deep inside the cortex.In a landmark study in 1999, Anne Blood, a postdoctoral fellow workingwith Robert Zatorre at the Montreal Neurological Institute, hadshown that intense musical emotion—what her subjects described as“thrills and chills”—was associated with brain regions thought to be involvedin reward, motivation, and arousal: the ventral striatum, theamygdala, the midbrain, and regions of the frontal cortex. I was particularlyinterested in the ventral striatum—a structure that includes the nucleusaccumbens—because the nucleus accumbens (NAc) is the centerof the brain’s reward system, playing an important role in pleasure andaddiction. The NAc is active when gamblers win a bet, or drug users taketheir favorite drug. It is also closely involved with the transmission ofopioids in the brain, through its ability to release the neurotransmitterdopamine. Avram Goldstein had shown in 1980 that the pleasure of musiclistening could be blocked by administering the drug nalaxone, believedto interfere with dopamine in the nucleus accumbens. But theparticular type of brain scan that Blood and Zatorre had used, positronemission tomography, doesn’t have a high enough spatial resolution todetect whether the small nucleus accumbens was involved. VinodMenon and I had lots of data collected from the higher-resolution fMRI,and we had the resolving power to pinpoint the nucleus accumbens if it

186 This Is Your Brain on Musicwas involved in music listening. But to really nail down the story abouthow pleasure in the brain occurs in response to music, we’d have toshow that the nucleus accumbens was involved at just the right time in asequence of neural structures that are recruited during music listening.The nucleus accumbens would have to be involved following activationof structures in the frontal lobe that process musical structure andmeaning. And in order to know that it was the nucleus accumbens’s roleas a modulator of dopamine, we would have to figure out a way to showthat its activation occurred at the same time as activation of other brainstructures that were involved in the production and transmission ofdopamine—otherwise, we couldn’t argue that the nucleus accumbens involvementwas anything more than coincidence. Finally, because so muchevidence seemed to point to the cerebellum, which we know to also havedopamine receptors, it would have to show up in this analysis as well.Menon had just read some papers by Karl Friston and his colleaguesabout a new mathematical technique, called functional and effectiveconnectivity analysis, that would allow us to address these questions, byrevealing the way that different brain regions interact during cognitiveoperations. These new connectivity analyses would allow us to detectassociations between neural regions in music processing that conventionaltechniques cannot address. By measuring the interaction of onebrain region with another—constrained by our knowledge of the anatomicalconnections between them—the technique would permit us tomake a moment-by-moment examination of the neural networks inducedby music. This is surely what Crick would have wanted to see. Thetask was not easy; brain scan experiments produce millions and millionsof data points; a single session can take up the entire hard drive on anordinary computer. Analyzing the data in the standard way—just tosee which areas are activated, not the new type of analyses we wereproposing—can take months. And there was no “off the shelf” statisticalprogram that would do these new analyses for us. Menon spent twomonths working through the equations necessary to do these analyses,and when he was done, we reanalyzed the data of people listening toclassical music we had collected.

After Dessert, Crick Was Still Four Seats Away from Me 187We found exactly what we had hoped. Listening to music caused acascade of brain regions to become activated in a particular order: first,auditory cortex for initial processing of the components of the sound.Then the frontal regions, such as BA44 and BA47, that we had previouslyidentified as being involved in processing musical structure and expectations.Finally, a network of regions—the mesolimbic system—involvedin arousal, pleasure, and the transmission of opioids and the productionof dopamine, culminating in activation in the nucleus accumbens. Andthe cerebellum and basal ganglia were active throughout, presumablysupporting the processing of rhythm and meter. The rewarding andreinforcing aspects of listening to music seem, then, to be mediated byincreasing dopamine levels in the nucleus accumbens, and by the cerebellum’scontribution to regulating emotion through its connections tothe frontal lobe and the limbic system. Current neuropsychological theoriesassociate positive mood and affect with increased dopamine levels,one of the reasons that many of the newer antidepressants act on thedopaminergic system. Music is clearly a means for improving people’smoods. Now we think we know why.Music appears to mimic some of the features of language and to conveysome of the same emotions that vocal communication does, but in anonreferential, and nonspecific way. It also invokes some of the sameneural regions that language does, but far more than language, musictaps into primitive brain structures involved with motivation, reward,and emotion. Whether it is the first few hits of the cowbell on “HonkyTonk Women,” or the first few notes of “Sheherazade,” computationalsystems in the brain synchronize neural oscillators with the pulse of themusic, and begin to predict when the next strong beat will occur. As themusic unfolds, the brain constantly updates its estimates of when newbeats will occur, and takes satisfaction in matching a mental beat with areal-in-the-world one, and takes delight when a skillful musician violatesthat expectation in an interesting way—a sort of musical joke that we’reall in on. Music breathes, speeds up, and slows down just as the realworld does, and our cerebellum finds pleasure in adjusting itself to staysynchronized.

188 This Is Your Brain on MusicEffective music—groove—involves subtle violations of timing. Justas the rat has an emotional response to a violation of the rhythm of thebranch hitting his house, we have an emotional response to the violationof timing in music that is groove. The rat, with no context for the timingviolation, experiences it as fear. We know through culture and experiencethat music is not threatening, and our cognitive system interprets these violationsas a source of pleasure and amusement. This emotional responseto groove occurs via the ear–cerebellum–nucleus accumbens–limbic circuitrather than via the ear–auditory cortex circuit. Our response togroove is largely pre- or unconscious because it goes through the cerebellumrather than the frontal lobes. What is remarkable is that all these differentpathways integrate into our experience of a single song.The story of your brain on music is the story of an exquisite orchestrationof brain regions, involving both the oldest and newest parts of thehuman brain, and regions as far apart as the cerebellum in the back ofthe head and the frontal lobes just behind your eyes. It involves a precisionchoreography of neurochemical release and uptake between logicalprediction systems and emotional reward systems. When we love a pieceof music, it reminds us of other music we have heard, and it activatesmemory traces of emotional times in our lives. Your brain on music is allabout, as Francis Crick repeated as we left the lunchroom, connections.

7. What Makes a Musician?Expertise DissectedOn his album Songs for Swinging Lovers, Frank Sinatra is awesomelyin control of his emotional expression, rhythm, and pitch.Now, I am not a Sinatra fanatic. I only have a half dozen or so of the morethan two hundred albums he’s released, and I don’t like his movies.Frankly, I find most of his repertoire to be just plain sappy; in everythingpost-1980, he sounds too cocky. Years ago Billboard hired me to reviewthe last album he made, duets with popular singers such as Bono andGloria Estefan. I panned it, writing that Frank “sings with all the satisfactionof a man who just had somebody killed.”But on Swinging Lovers, every note he sings is perfectly placed intime and pitch. I don’t mean “perfectly” in the strict, as-notated sense; hisrhythms and timing are completely wrong in terms of how the music iswritten on paper, but they are perfect for expressing emotions that gobeyond description. His phrasing contains impossibly detailed and subtlenuances—to be able to pay attention to that much detail, to be able tocontrol it, is something I can’t imagine. Try to sing along with any songon Swinging Lovers. I’ve never found anyone who could match hisphrasing precisely—it is too nuanced, too quirky, too idiosyncratic.How do people become expert musicians? And why is that of the millionsof people who take music lessons as children, relatively few con-

190 This Is Your Brain on Musictinue to play music as adults? When they find out what I do for a living,many people tell me that they love music listening, but their music lessons“didn’t take.” I think they’re being too hard on themselves. Thechasm between musical experts and everyday musicians that has grownso wide in our culture makes people feel discouraged, and for some reasonthis is uniquely so with music. Even though most of us can’t playbasketball like Shaquille O’Neal, or cook like Julia Child, we can still enjoyplaying a friendly backyard game of hoops, or cooking a holiday mealfor our friends and family. This performance chasm does seem to be cultural,specific to contemporary Western society. And although manypeople say that music lessons didn’t take, cognitive neuroscientists havefound otherwise in their laboratories. Even just a small exposure to musiclessons as a child creates neural circuits for music processing thatare enhanced and more efficient than for those who lack training. Musiclessons teach us to listen better, and they accelerate our ability to discernstructure and form in music, making it easier for us to tell what musicwe like and what we don’t like.But what about that class of people that we all acknowledge are truemusical experts—the Alfred Brendels, Sarah Changs, Wynton Marsalises,and Tori Amoses? How did they get what most of us don’t have, anextraordinary facility to play and perform? Do they have a set of abilities—orneural structures—that are of a totally different sort than therest of us have (a difference of kind) or do they just have more of thesame basic stuff all of us are endowed with (a difference of degree)? Anddo composers and songwriters have a fundamentally different set ofskills than players?The scientific study of expertise has been a major topic within cognitivescience for the past thirty years, and musical expertise has tended tobe studied within the context of general expertise. In almost all cases,musical expertise has been defined as technical achievement—masteryof an instrument or of compositional skills. The late Michael Howe, andhis collaborators Jane Davidson and John Sloboda, launched an internationaldebate when they asked whether the lay notion of “talent” is scientificallydefensible. They assumed the following dichotomy: Either

What Makes a Musician? 191high levels of musical achievement are based on innate brain structures(what we refer to as talent) or they are simply the result of training andpractice. They define talent as something (1) that originates in geneticstructures; (2) that is identifiable at an early stage by trained people whocan recognize it even before exceptional levels of performance havebeen acquired; (3) that can be used to predict who is likely to excel; and(4) that only a minority can be identified as having because if everyonewere “talented,” the concept would lose meaning. The emphasis on earlyidentification entails that we study the development of skills in children.They add that in a domain such as music, “talent” might be manifesteddifferently in different children.It is evident that some children acquire skills more rapidly than others:The age of onset for walking, talking, and toilet training vary widelyfrom one child to another, even within the same household. There maybe genetic factors at work, but it is difficult to separate out ancillaryfactors—with a presumably environmental component—such as motivation,personality, and family dynamics. Similar factors can influencemusical development and can mask the contributions of genetics to musicalability. Brain studies, so far, haven’t been of much use in sorting outthe issues because it has been difficult to separate cause from effect.Gottfried Schlaug at Harvard collected brain scans of individuals withabsolute pitch (AP) and showed that a region in the auditory cortex—theplanum temporale—is larger in the AP people than the non-AP people.This suggests that the planum is involved in AP, but it’s not clear if itstarts out larger in people who eventually acquire AP, or rather, if the acquisitionof AP causes the planum to increase in size. The story is clearerin the areas of the brain that are involved in skilled motor movements.Studies of violin players by Thomas Elbert have shown that the region ofthe brain responsible for moving the left hand—the hand that requiresthe most precision in violin playing—increases in size as a result of practice.We do not know yet if the propensity for increase preexists in somepeople and not others.The strongest evidence for the talent position is that some peoplesimply acquire musical skills more rapidly than others. The evidence

192 This Is Your Brain on Musicagainst the talent account—or rather, in favor of the view that practicemakes perfect—comes from research on how much training the expertsor high achievement people actually do. Like experts in mathematics,chess, or sports, experts in music require lengthy periods of instructionand practice in order to acquire the skills necessary to truly excel. In severalstudies, the very best conservatory students were found to havepracticed the most, sometimes twice as much as those who weren’tjudged as good.In another study, students were secretly divided into two groups (notrevealed to the students so as not to bias them) based on teachers’ evaluationsof their ability, or the perception of talent. Several years later, thestudents who achieved the highest performance ratings were those whohad practiced the most, irrespective of which “talent” group they hadbeen assigned to previously. This suggests that practice is the cause ofachievement, not merely something correlated with it. It further suggeststhat talent is a label that we’re using in a circular fashion: When wesay that someone is talented, we think we mean that they have some innatepredisposition to excel, but in the end, we only apply the term retrospectively,after they have made significant achievements.Anders Ericsson, at Florida State University, and his colleagues approachthe topic of musical expertise as a general problem in cognitivepsychology involving how humans become experts in general. In otherwords, he takes as a starting assumption that there are certain issuesinvolved in becoming an expert at anything; that we can learn aboutmusical expertise by studying expert writers, chess players, athletes,artists, mathematicians, in addition to musicians.First, what do we mean by “expert”? Generally we mean that it issomeone who has reached a high degree of accomplishment relative toother people. As such, expertise is a social judgment; we are making astatement about a few members of a society relative to a larger population.Also, the accomplishment is normally considered to be in a fieldthat we care about. As Sloboda points out, I may become an expert atfolding my arms or pronouncing my own name, but this isn’t generallyconsidered the same as becoming, say, an expert at chess, at repairing

What Makes a Musician? 193Porsches, or being able to steal the British crown jewels without beingcaught.The emerging picture from such studies is that ten thousand hours ofpractice is required to achieve the level of mastery associated with beinga world-class expert—in anything. In study after study, of composers,basketball players, fiction writers, ice skaters, concert pianists, chessplayers, master criminals, and what have you, this number comes upagain and again. Ten thousand hours is equivalent to roughly three hoursa day, or twenty hours a week, of practice over ten years. Of course, thisdoesn’t address why some people don’t seem to get anywhere when theypractice, and why some people get more out of their practice sessionsthan others. But no one has yet found a case in which true world-classexpertise was accomplished in less time. It seems that it takes the brainthis long to assimilate all that it needs to know to achieve true mastery.The ten-thousand-hours theory is consistent with what we knowabout how the brain learns. Learning requires the assimilation and consolidationof information in neural tissue. The more experiences wehave with something, the stronger the memory/learning trace for that experiencebecomes. Although people differ in how long it takes them toconsolidate information neurally, it remains true that increased practiceleads to a greater number of neural traces, which can combine to createa stronger memory representation. This is true whether you subscribe tomultiple-trace theory or any number of variants of theories in the neuroanatomyof memory: The strength of a memory is related to how manytimes the original stimulus has been experienced.Memory strength is also a function of how much we care about theexperience. Neurochemical tags associated with memories mark themfor importance, and we tend to code as important things that carry withthem a lot of emotion, either positive or negative. I tell my students ifthey want to do well on a test, they have to really care about the materialas they study it. Caring may, in part, account for some of the early differenceswe see in how quickly people acquire new skills. If I really like aparticular piece of music, I’m going to want to practice it more, and becauseI care about it, I’m going to attach neurochemical tags to each as-

194 This Is Your Brain on Musicpect of the memory that label it as important: The sounds of the piece,the way I move my fingers, if I’m playing a wind instrument the way thatI breathe—all these become part of a memory trace that I’ve encoded asimportant.Similarly, if I’m playing an instrument I like, and whose sound pleasesme in and of itself, I’m more likely to pay attention to subtle differencesin tone, and the ways in which I can moderate and affect the tonal outputof my instrument. It is impossible to overestimate the importance ofthese factors; caring leads to attention, and together they lead to measurableneurochemical changes. Dopamine, the neurotransmitter associatedwith emotional regulation, alertness, and mood, is released, and thedopaminergic system aids in the encoding of the memory trace.Owing to various factors, some people who take music lessons areless motivated to practice; their practice is less effective because of motivationaland attentional factors. The ten-thousand-hours argument isconvincing because it shows up in study after study across many domains.Scientists like order and simplicity, so if we see a number or aformula that pops up in different contexts, we tend to favor it as an explanation.But like many scientific theories, the ten-thousand-hours theoryhas holes in it, and it needs to account for counterarguments andrebuttals.The classic rebuttal to the ten-thousand-hours argument goes somethinglike this: “Well, what about Mozart? I hear that he was composingsymphonies at the age of four! And even if he was practicing forty hoursa week since the day he was born, that doesn’t make ten thousandhours.” First, there are factual errors in this account: Mozart didn’t begincomposing until he was six, and he didn’t write his first symphony untilhe was eight. Still, writing a symphony at age eight is unusual, to say theleast. Mozart demonstrated precociousness early in his life. But that isnot the same as being an expert. Many children write music, and someeven write large-scale works when they’re as young as eight. And Mozarthad extensive training from his father, who was widely considered to bethe greatest living music teacher in all of Europe at the time. We don’tknow how much Mozart practiced, but if he started at age two and

What Makes a Musician? 195worked thirty-two hours a week at it (quite possible, given his father’sreputation as a stern taskmaster) he would have made his ten thousandhours by the age of eight. Even if Mozart hadn’t practiced that much, theten-thousand-hours argument doesn’t say that it takes ten thousandhours to write a symphony. Clearly Mozart became an expert eventually,but did the writing of that first symphony qualify him as an expert, or didhe attain his level of musical expertise sometime later?John Hayes of Carnegie Mellon asked just this question. DoesMozart’s Symphony no. 1 qualify as the work of a musical expert? Putanother way, if Mozart hadn’t written anything else, would this symphonystrike us as the work of a musical genius? Maybe it really isn’tvery good, and the only reason we know about it is because the childwho wrote it grew up to become Mozart—we have a historical interest init, but not an aesthetic one. Hayes studied the performance programs ofthe leading orchestras and the catalog of commercial recordings, assumingthat better musical works are more likely to be performed andrecorded than lesser works. He found that the early works of Mozartwere not performed or recorded very often. Musicologists largely regardthem as curiosities, compositions that by no means predicted the expertworks that were to follow. Those of Mozart’s compositions that are consideredtruly great are those that he wrote well after he had been at it forten thousand hours.As we have seen in the debates about memory and categorization, thetruth lies somewhere between the two extremes, a composite of the twohypotheses confronting each other in the nature/nurture debate. To understandhow this particular synthesis occurs, and what predictions itmakes, we need to look more closely at what the geneticists have to say.Geneticists seek to find a cluster of genes that are associated withparticular observable traits. They assume that if there is a genetic contributionto music, it will show up in families, since brothers and sistersshare 50 percent of their genes with one another. But it can be difficult toseparate out the influence of genes from the influence of the environmentin this approach. The environment includes the environment of thewomb: the food that the mother eats, whether she smokes or drinks, and

196 This Is Your Brain on Musicother factors that influence the amount of nutrients and oxygen the fetusreceives. Even identical twins can experience very different environmentsfrom one another within the womb, based on the amount of spacethey have, their room for movement, and their position.Distinguishing genetic from environmental influences on a skill thathas a learned component, such as music, is difficult. Music tends to runin families. But a child with parents who are musicians is more likely toreceive encouragement for her early musical leanings than a child in anonmusical household, and siblings of that musically raised child arelikely to receive similar levels of support. By analogy, parents who speakFrench are likely to raise children who speak French, and parents whodo not are unlikely to do so. We can say that speaking French “runs infamilies,” but I don’t know anyone who would claim that speakingFrench is genetic.One way that scientists determine the genetic basis of traits or skillsis by studying identical twins, especially those who have been rearedapart. The Minnesota twins registry, a database kept by the psychologistsDavid Lykken, Thomas Bouchard, and their colleagues, has followedidentical and fraternal twins reared apart and reared together.Because fraternal twins share 50 percent of their genetic material, andidentical twins share 100 percent, this allows scientists to tease apart therelative influences of nature versus nurture. If something has a geneticcomponent, we would expect it to show up more often in each individualwho is an identical twin than in each who is a fraternal twin. Moreover,we would expect it to show up even when the identical twins havebeen raised in completely separate environments. Behavioral geneticistslook for such patterns and form theories about the heritability ofcertain traits.The newest approach looks at gene linkages. If a trait appears to beheritable, we can try to isolate the genes that are linked to that trait. (Idon’t say “responsible for that trait,” because interactions among genesare very complicated, and we cannot say with certainty that a singlegene “causes” a trait.) This is complicated by the fact that we can have agene for something without its being active. Not all of the genes that we

What Makes a Musician? 197have are “turned on,” or expressed, at all times. Using gene chip expressionprofiling, we can determine which genes are and which genes aren’texpressed at a given time. What does this mean? Our roughly twenty-fivethousand genes control the synthesis of proteins that our bodies andbrains use to perform all of our biological functions. They control hairgrowth, hair color, the creation of digestive fluids and saliva, whether weend up being six feet tall or five feet tall. During our growth spurt aroundthe time of puberty, something needs to tell our body to start growing,and a half dozen years later, something has to tell it to stop. These are thegenes, carrying instructions about what to do and how to do it.Using gene chip expression profiling, I can analyze a sample of yourRNA and—if I know what I’m looking for—I can tell whether yourgrowth gene is active—that is, expressed—right now. At this point, theanalysis of gene expression in the brain isn’t practical because current(and foreseeable) techniques require that we analyze a piece of brain tissue.Most people find that unpleasant.Scientists studying identical twins who’ve been reared apart havefound remarkable similarities. In some cases, the twins were separatedat birth, and not even told of each other’s existence. They might have beenraised in environments that differed a great deal in geography (Maineversus Texas, Nebraska versus New York), in financial means, and in religiousor other cultural values. When tracked down twenty or moreyears later, a number of astonishing similarities emerged. One womanliked to go to the beach and when she did, she would back into the water;her twin (whom she had never met) did exactly the same thing. Oneman sold life insurance for a living, sang in his church choir, and woreLone Star beer belt buckles; so did his completely-separated-from-birthidentical twin. Studies like these suggested that musicality, religiosity,and criminality had a strong genetic component. How else could you explainsuch coincidences?One alternative explanation is statistical, and can be stated like this:“If you look hard enough, and make enough comparisons, you’re goingto find some really weird coincidences that don’t really mean anything.”Take any two random people off the street who have no relationship to

198 This Is Your Brain on Musicone another, except perhaps through their common ancestors Adam andEve. If you look at enough traits, you’re bound to find some in commonthat aren’t obvious. I’m not talking about things like “Oh, my gosh! Youbreathe the atmosphere too!!” but things like “I wash my hair on Tuesdaysand Fridays, and I use an herbal shampoo on Tuesdays—scrubbingwith only my left hand, and I don’t use a conditioner. On Fridays I use anAustralian shampoo that has a conditioner built in. Afterward, I read TheNew Yorker while listening to Puccini.” Stories like these suggest thatthere is an underlying connection between these people, in spite of thescientists’ assurances that their genes and environment are maximallydissimilar. But all of us differ from one another in thousands upon thousandsof different ways, and we all have our quirks. Once in a while wefind co-occurrences, and we’re surprised. But from a statistical standpoint,it isn’t any more surprising than if I think of a number between oneand one hundred and you guess it. You may not guess it the first time, butif we play the game long enough, you’re going to guess it once in a while(1 percent of the time, to be exact).A second alternative explanation is social psychological—the waysomeone looks influences the way that others treat him (with “looks” assumedto be genetic); in general, an organism is acted on by the world inparticular ways as a function of its appearance. This intuitive notion hasa rich tradition in literature, from Cyrano de Bergerac to Shrek: Shunnedby people who were repulsed by their outward appearance, they rarelyhad the opportunity to show their inner selves and true nature. As a culturewe romanticize stories like these, and feel a sense of tragedy abouta good person suffering for something he had nothing to do with: hislooks. It works in the opposite way as well: good-looking people tend tomake more money, get better jobs, and report that they are happier. Evenapart from whether someone is considered attractive or not, his appearanceaffects how we relate to him. Someone who was born with facialfeatures that we associate with trustworthiness—large eyes, for example,with raised eyebrows—is someone people will tend to trust. Someonetall may be given more respect than someone short. The series of

What Makes a Musician? 199encounters we have over our lifetimes are shaped to some extent by theway others see us.It is no wonder, then, that identical twins may end up developing similarpersonalities, traits, habits, or quirks. Someone with downturned eyebrowsmight always look angry, and the world will treat them that way.Someone who looks defenseless will be taken advantage of; someonewho looks like a bully may spend a lifetime being asked to fight, and eventuallywill develop an aggressive personality. We see this principle atwork in certain actors. Hugh Grant, Judge Reinhold, Tom Hanks, andAdrien Brody have innocent-looking faces; without doing anything, Granthas an “awww, shucks” look, a face that suggests he has no guile or deceit.This line of reasoning says that some people are born with particularfeatures, and their personalities develop in large part as a reflection ofhow they look. Genes here are influencing personality, but only in an indirect,secondary way.It is not difficult to imagine a similar argument applying to musicians,and in particular to vocalists. Doc Watson’s voice sounds completely sincereand innocent; I don’t know if he is that way in person, and at onelevel it doesn’t matter. It’s possible that he became the successful artisthe is because of how people react to the voice that he was born with. I’mnot talking about being born with (or acquiring) a “great” voice, like EllaFitzgerald’s or Placido Domingo’s, I’m talking about expressiveness apartfrom whether the voice itself is a great instrument. Sometimes as AimeeMann sings, I hear the traces of a little girl’s voice, a vulnerable innocencethat moves me because I feel that she is reaching down deep insideand confessing feelings that normally are expressed only to a closefriend. Whether she intends to convey this, or really feels this, I don’tknow—she may have been born with a vocal quality that makes listenersinvest her with those feelings, whether she is experiencing them or not.In the end, the essence of music performance is being able to conveyemotion. Whether the artist is feeling it or was born with an ability tosound as if she’s feeling it may not be important.I don’t mean to imply that the actors and musicians I’ve mentioned

200 This Is Your Brain on Musicdon’t have to work at what they do. I don’t know any successful musicianswho haven’t worked hard to get where they are; I don’t know anywho had success fall into their laps. I’ve known a lot of artists whom thepress has called “overnight sensations,” but who spent five or ten yearsbecoming that! Genetics are a starting point that may influence personalityor career, or the specific choices one makes in a career. Tom Hanksis a great actor, but he’s not likely to get the same kinds of roles asArnold Schwarzenegger, largely owing the differences in their geneticendowments. Schwarzenegger wasn’t born with a body-builder’s body;he worked very hard at it, but he had a genetic predisposition toward it.Similarly, being six ten creates a predisposition toward becoming abasketball player rather than a jockey. But it is not enough for someonewho is six ten to simply stand on the court—he needs to learn the gameand practice for years to become an expert. Body type, which is largely(though not exclusively) genetic, creates predispositions for basketballas it does for acting, dancing, and music.Musicians, like athletes, actors, dancers, sculptors and painters, usetheir bodies as well as their minds. The role of the body in the playing ofa musical instrument or in singing (less so, of course, in composing andarranging) means that genetic predispositions can contribute strongly tothe choice of instruments a musician can play well—and to whether aperson chooses to become a musician.When I was six years old, I saw the Beatles on The Ed Sullivan Show,and in what has become a cliché for people of my generation, I decidedthen that I wanted to play the guitar. My parents, who were of the oldschool, did not view the guitar as a “serious instrument” and told me toplay the family piano instead. But I wanted desperately to play. I wouldcut out pictures of classical guitarists like Andrés Segovia from magazinesand casually leave them around the house. At six, I was still speakingwith a prominent lisp that I had had all my life; I didn’t get rid of ituntil age ten when I was embarrassingly plucked out of my fourth-gradeclass by the public-school speech therapist who spent a grueling twoyears (at three hours a week) teaching me to change the way that I saidthe letter s. I pointed out that the Beatles must be therious to share the

What Makes a Musician? 201stage of The Ed Sullivan Show with such therious artithts as BeverlyThills, Rodgers and Hammerthtein, and John Gielgud. I was relentless.By 1965, when I was eight, the guitar was everywhere. With San Franciscojust fifteen miles away, I could feel a cultural and musical revolutiongoing on, and the guitar was at the center of it all. My parents werestill not enthusiastic about me studying the guitar, perhaps because of itsassociation with hippies and drugs, or perhaps as a result of my failurethe previous year to practice the piano diligently. I pointed out that bynow, the Beatles had been on The Ed Sullivan Show four times and myparents finally quasi-relented, agreeing to ask a friend of theirs for advice.“Jack King plays the guitar,” my mother said at dinner one night tomy father. “We could ask him if he thinks Danny is old enough to beginguitar lessons.” Jack, an old college friend of my parents, dropped by thehouse one day on his way home from work. His guitar sounded differentfrom the ones that had mesmerized me on television and radio; it was aclassical guitar, not made for the dark chords of rock and roll. Jack was abig man with large hands, and a short black crew cut. He held the guitarin his arms as one might cradle a baby. I could see the intricate patternsof wood grain bending around the curves of the instrument. He playedsomething for us. He didn’t let me touch the guitar, instead he asked meto hold my hand out, and he pressed his palm against mine. He didn’t talkto me or look at me, but what he said to my mother I can still hearclearly: “His hands are too small for the guitar.”I now know about three-quarter size and half-size guitars (I even ownone), and about Django Reinhardt, one of the greatest guitarists of alltime, who had only two fingers on his left hand. But to an eight-year-old,the words of adults can seem unbreachable. By 1966, when I had grownsome, and the Beatles were egging me on with electric guitar strains of“Help,” I was playing the clarinet and happy to at least be making music.I finally bought my first guitar when I was sixteen and with practice, Ilearned to play reasonably well; the rock and jazz that I play don’t requirethe long reach that classical guitar does. The very first song Ilearned to play—in what has become another cliché for my generation—was Led Zeppelin’s “Stairway to Heaven” (hey, it was the seventies).

202 This Is Your Brain on MusicSome musical parts that guitarists with different hands can play will alwaysbe difficult for me, but that is always the case with every instrument.On Hollywood Boulevard in Hollywood, California, some of thegreat rock musicians have placed their handprints in the cement. I wassurprised last summer when I put my hands in the imprint left by JimmyPage (of Led Zeppelin), one of my favorite guitarists, that his hands wereno bigger than mine.Some years ago I shook hands with Oscar Peterson, the great jazz pianist.His hands were very large; the largest hands I have ever shaken, atleast twice the size of my own. He began his career playing stride piano,a style dating back to the 1920s in which the pianist plays an octave basswith his left hand and the melody with his right. To be a good strideplayer, you need to be able to be able to reach keys that are far apart witha minimum of hand movements, and Oscar can stretch a whopping octaveand a half with one hand! Oscar’s style is related to the kinds of chordshe is able to play, chords that someone with smaller hands could not. IfOscar Peterson had been forced to play violin as a child it would havebeen impossible with those large hands; his wide fingers would make itdifficult to play a semitone on the relatively small neck of the violin.Some people have a biological predisposition toward particular instruments,or toward singing. There may also be a cluster of genes thatwork together to create the component skills that one must have tobecome a successful musician: good eye-hand coordination, muscle control,motor control, tenacity, patience, memory for certain kinds of structuresand patterns, a sense of rhythm and timing. To be a good musician,one must have these things. Some of these skills are involved in becominga great anything, especially determination, self-confidence, and patience.We also know that, on average, successful people have had manymore failures than unsuccessful people. This seems counterintuitive.How could successful people have failed more often than everyone else?Failure is unavoidable and sometimes happens randomly. It’s what youdo after the failure that is important. Successful people have a stick-toit-iveness.They don’t quit. From the president of FedEx to the novelist

What Makes a Musician? 203Jerzy Kosinsky, from van Gogh to Bill Clinton to Fleetwood Mac, successfulpeople have had many, many failures, but they learn from themand keep going. This quality might be partly innate, but environmentalfactors must also play a role.The best guess that scientists currently have about the role of genesand the environment in complex cognitive behaviors is that each is responsiblefor about 50 percent of the story. Genes may transmit a propensityto be patient, to have good eye-hand coordination, or to bepassionate, but certain life events—life events in the broadest sense,meaning not just your conscious experiences and memories, but thefood you ate and the food your mother ate while you were in herwomb—can influence whether a genetic propensity will be realized ornot. Early life traumas, such as the loss of a parent, or physical or emotionalabuse, are only the obvious examples of environmental influencescausing a genetic predisposition to become either heightened or suppressed.Because of this interaction, we can only make predictions abouthuman behavior at the level of a population, not an individual. In otherwords, if you know that someone has a genetic predisposition towardcriminal behavior, you can’t make any predictions about whether he willend up in jail in the next five years. On the other hand, knowing that ahundred people have this predisposition, we can predict that some percentageof them will probably wind up in jail; we simply don’t knowwhich ones. And some will never get into any trouble at all.The same applies to musical genes we may find someday. All we cansay is that a group of people with those genes is more likely to produceexpert musicians, but we cannot know which individuals will becomethe experts. This, however, assumes that we’ll be able to identify the geneticcorrelates of musical expertise, and that we can agree on what constitutesmusical expertise. Musical expertise has to be about more thanstrict technique. Music listening and enjoyment, musical memory, andhow engaged with music a person is are also aspects of a musical mindand a musical personality. We should take as inclusive an approach aspossible in identifying musicality, so as not to exclude those who, whilemusical in the broad sense, are perhaps not so in a narrow, technical

204 This Is Your Brain on Musicsense. Many of our greatest musical minds weren’t considered experts ina technical sense. Irving Berlin, one of the most successful composers ofthe twentieth century, was a lousy instrumentalist and could barely playthe piano.Even among the elite, top-tier classical musicians, there is more to beinga musician than having excellent technique. Both Arthur Rubinsteinand Vladimir Horowitz are widely regarded as two of the greatest pianistsof the twentieth century but they made mistakes—little technicalmistakes—surprisingly often. A wrong note, a rushed note, a note thatisn’t fingered properly. But as one critic wrote, “Rubinstein makes mistakeson some of his records, but I’ll take those interpretations that arefilled with passion over the twenty-two-year-old technical wizard whocan play the notes but can’t convey the meaning.”What most of us turn to music for is an emotional experience. Wearen’t studying the performance for wrong notes, and so long as theydon’t jar us out of our reverie, most of us don’t notice them. So much ofthe research on musical expertise has looked for accomplishment in thewrong place, in the facility of fingers rather than the expressiveness ofemotion. I recently asked the dean of one of the top music schools inNorth America about this paradox: At what point in the curriculum isemotion and expressivity taught? Her answer was that they aren’t taught.“There is so much to cover in the approved curriculum,” she explained,“repertoire, ensemble, and solo training, sight singing, sight reading, musictheory—that there simply isn’t time to teach expressivity.” So how dowe get expressive musicians? “Some of them come in already knowinghow to move a listener. Usually they’ve figured it out themselves somewherealong the line.” The surprise and disappointment in my face musthave been obvious. “Occasionally,” she added, almost in a whisper, “ifthere’s an exceptional student, there’s time during the last part of theirlast semester here to coach them on emotion. ...Usually this is forpeople who are already performing as soloists in our orchestra, and wehelp them to coax out more expressivity from their performance.” So, atone of the best music schools we have, the raison d’être for music is

What Makes a Musician? 205taught to a select few, and then, only in the last few weeks of a four- orfive-year curriculum.Even the most uptight and analytic among us expect to be moved byShakespeare and Bach. We can marvel at the craft these geniuses havemastered, a facility with language or with notes, but ultimately that facilitymust be brought into service for a different type of communication.Jazz fans, for example, are especially demanding of their post-big-banderaheroes, starting with the Miles Davis/John Coltrane/Bill Evans era.We say of lesser jazz musicians who appear detached from their trueselves and from emotion that their playing is nothing more than “shuckingand jiving,” attempts to please the audience through musical obsequiesrather than through soul.So—in a scientific sense—why are some musicians superior to otherswhen it comes to the emotional (versus the technical) dimension of music?This is the great mystery, and no one knows for sure. Musicianshaven’t yet performed with feeling inside brain scanners, due to technicaldifficulties. (The scanners we currently use require the subject tostay perfectly still, so as not to blur the brain image; this may change inthe coming five years.) Interviews with, and diary entries of, musiciansranging from Beethoven and Tchaikovsky to Rubinstein and Bernstein,B. B. King, and Stevie Wonder suggest that part of communicating emotioninvolves technical, mechanical factors, and part of it involves somethingthat remains mysterious.The pianist Alfred Brendel says he doesn’t think about notes when he’sonstage; he thinks about creating an experience. Stevie Wonder told mein 1996 that when he’s performing, he tries to get himself into the sameframe of mind and “frame of heart” that he was in when he wrote thesong; he tries to capture the same feelings and sentiment, and that helpshim to deliver the performance. What this means in terms of how he singsor plays differently is something no one knows. From a neuroscientificperspective, though, this makes perfect sense. As we’ve seen, rememberingmusic involves setting the neurons that were originally active in theperception of a piece of music back to their original state—reactivating

206 This Is Your Brain on Musictheir particular pattern of connectivity, and getting the firing rates asclose as possible to their original levels. This means recruiting neurons inthe hippocampus, amygdala, and temporal lobes in a neural symphonyorchestrated by attention and planning centers in the front lobe.The neuroanatomist Andrew Arthur Abbie speculated in 1934 a linkagebetween movement, the brain, and music that is only now becomingproven. He wrote that pathways from the brain stem and cerebellum tothe frontal lobes are capable of weaving all sensory experience and accuratelycoordinated muscular movements into a “homogeneous fabric”and that when this occurs, the result is “man’s highest powers as expressed...inart.” His idea of this neural pathway was that it is dedicatedto motor movements that incorporate or reflect a creative purpose. Newstudies by Marcelo Wanderley of McGill, and by my former doctoral studentBradley Vines (now at Harvard) have shown that nonmusician listenersare exquisitely sensitive to the physical gestures that musiciansmake. By watching a musical performance with the sound turned off, andattending to things like the musician’s arm, shoulder, and torso movements,ordinary listeners can detect a great deal of the expressive intentionsof the musician. Add in the sound, and an emergent qualityappears—an understanding of the musician’s expressive intentions thatgoes beyond what was available in the sound or the visual image alone.If music serves to convey feelings through the interaction of physicalgestures and sound, the musician needs his brain state to match the emotionalstate he is trying to express. Although the studies haven’t beenperformed yet, I’m willing to bet that when B.B. is playing the blues andwhen he is feeling the blues, the neural signatures are very similar. (Ofcourse there will be differences, too, and part of the scientific hurdle willbe subtracting out the processes involved in issuing motor commandsand listening to music, versus just sitting on a chair, head in hands, andfeeling down.) And as listeners, there is every reason to believe thatsome of our brain states will match those of the musicians we are listeningto. In what is a recurring theme of your brain on music, even those ofus who lack explicit training in music theory and performance have musicalbrains, and are expert listeners.

What Makes a Musician? 207In understanding the neurobehavioral basis of musical expertise andwhy some people become better performers than others, we need toconsider that musical expertise takes many forms, sometimes technical(involving dexterity) and sometimes emotional. The ability to draw usinto a performance so that we forget about everything else is also a specialkind of ability. Many performers have a personal magnetism, orcharisma, that is independent of any other abilities they may or may nothave. When Sting is singing, we can’t take our ears off of him. WhenMiles Davis is playing the trumpet, or Eric Clapton the guitar, an invisibleforce seems to draw us toward him. This doesn’t have to do so muchwith the actual notes they’re singing or playing—any number of goodmusicians can play or sing those notes, perhaps even with better technicalfacility. Rather, it is what record company executives call “star quality.”When we say of a model that she is photogenic, we’re talking abouthow this star quality manifests itself in photographs. The same thing istrue for musicians, and how their quality comes across on records—I call this phonogenic.It is also important to distinguish celebrity from expertise. The factorsthat contribute to celebrity could be different from, maybe whollyunrelated to, those that contribute to expertise. Neil Young told me thathe did not consider himself to be especially talented as a musician,rather, he was one of the lucky ones who managed to become commerciallysuccessful. Few people get to pass through the turnstiles of a dealwith a major record label, and fewer still maintain careers for decades asNeil has done. But Neil, along with Stevie Wonder and Eric Clapton, attributesa lot of his success not to musical ability but to a good break.Paul Simon agrees. “I’ve been lucky to have been able to work with someof the most amazing musicians in the world,” he said, “and most of themare people no one’s ever heard of.”Francis Crick turned his lack of training into a positive aspect of his life’swork. Unbound by scientific dogma, he was free—completely free, hewrote—to open his mind and discover science. When an artist bringsthis freedom, this tabula rasa, to music, the results can be astounding.

208 This Is Your Brain on MusicMany of the greatest musicians of our era lacked formal training, includingSinatra, Louis Armstrong, John Coltrane, Eric Clapton, Eddie VanHalen, Stevie Wonder, and Joni Mitchell. And in classical music, GeorgeGershwin, Mussorgsky, and David Helfgott are among those who lackedformal training, and Beethoven considered his own training to have beenpoor according to his diaries.Joni Mitchell had sung in choirs in public school, but had never takenguitar lessons or any other kind of music lessons. Her music has a uniquequality that has been variously described as avant-garde, ethereal, and asbridging classical, folk, jazz, and rock. Joni uses a lot of alternate tunings;that is, instead of tuning the guitar in the customary way, she tunesthe strings to pitches of her own choosing. This doesn’t mean that sheplays notes that other people don’t—there are still only twelve notes in achromatic scale—but it does mean that she can easily reach with her fingerscombinations of notes that other guitarists can’t reach (regardlessof the size of their hands).An even more important difference involves the way the guitar makessound. Each of the six strings of the guitar is tuned to a particular pitch.When a guitarist wants a different one, of course, she presses one ormore strings down against the neck; this makes the string shorter, whichcauses it to vibrate more rapidly, making a tone with a higher pitch. Astring that is pressed on (“fretted”) has a different sound from one thatisn’t, due to a slight deadening of the string caused by the finger; the unfrettedor “open” strings have a clearer, more ringing quality, and theywill keep on sounding for a longer time than the ones that are fretted.When two or more of these open strings are allowed to ring together, aunique timbre emerges. By retuning, Joni changed the configuration ofwhich notes are played when a string is open, so that we hear notes ringingthat don’t usually ring on the guitar, and in combinations we don’tusually hear. You can hear it on her songs “Chelsea Morning” and “Refugeof the Roads” for example.But there is something more to it than that—lots of guitarists usetheir own tunings, such as David Crosby, Ry Cooder, Leo Kottke, and

What Makes a Musician? 209Jimmy Page. One night, when I was having dinner with Joni in Los Angeles,she started talking about bass players that she had worked with. Shehas worked with some of the very best of our generation: Jaco Pastorius,Max Bennett, Larry Klein, and she wrote an entire album with CharlesMingus. Joni will talk compellingly and passionately about alternate tuningsfor hours, comparing them to the different colors that van Goghused in his paintings.While we were waiting for the main course, she went off on a storyabout how Jaco Pastorius was always arguing with her, challenging her,and generally creating mayhem backstage before they would go on. Forexample when the first Roland Jazz Chorus amplifier was hand-deliveredby the Roland Company to Joni to use at a performance, Jaco picked itup, and moved it over to his corner of the stage. “It’s mine,” he growled.When Joni approached him, he gave her a fierce look. And that was that.We were well into twenty minutes of bass-player stories. Because Iwas a huge fan of Jaco when he played with Weather Report, I interruptedand asked what it was like musically to play with him. She saidthat he was different from any other bass player she had every playedwith; that he was the only bass player up to that time that she felt reallyunderstood what she was trying to do. That’s why she put up with his aggressivebehaviors.“When I first started out,” she said, “the record company wanted togive me a producer, someone who had experience churning out hitrecords. But [David] Crosby said, ‘Don’t let them—a producer will ruinyou. Let’s tell them that I’ll produce it for you; they’ll trust me.’ So basically,Crosby put his name as producer to keep the record company outof my way so that I could make the music the way that I wanted to.“But then the musicians came in and they all had ideas about howthey wanted to play. On my record! The worst were the bass players becausethey always wanted to know what the root of the chord was.” The“root” of a chord, in music theory, is the note for which the chord isnamed and around which it is based. A “C major” chord has the note C asits root, for example, and an “E-flat minor” chord has the note E-flat as its

210 This Is Your Brain on Musicroot. It is that simple. But the chords Joni plays, as a consequence of herunique composition and guitar-playing styles, aren’t typical chords: Jonithrows notes together in such a way that the chords can’t be easily labeled.“The bass players wanted to know the root because that’s whatthey’ve been taught to play. But I said, ‘Just play something that soundsgood, don’t worry about what the root is.’ And they said, ‘We can’t dothat—we have to play the root or it won’t sound right.’”Because Joni hadn’t had music theory and didn’t know how to readmusic, she couldn’t tell them the root. She had to tell them what notesshe was playing on the guitar, one by one, and they had to figure it out forthemselves, painstakingly, one chord at a time. But here is where psychoacousticsand music theory collide in an explosive conflagration: Thestandard chords that most composers use—C major, E-flat minor, D7,and so on—are unambiguous. No competent musician would need toask what the root of a chord like those is; it is obvious, and there is onlyone possibility. Joni’s genius is that she creates chords that are ambiguous,chords that could have two or more different roots. When there isno bass playing along with her guitar (as in “Chelsea Morning” or “SweetBird”), the listener is left in a state of expansive aesthetic possibilities.Because each chord could be interpreted in two or more different ways,any prediction or expectation that a listener has about what comes nextis less grounded in certainty than with traditional chords. And when Jonistrings together several of these ambiguous chords, the harmonic complexitygreatly increases; each chord sequence can be interpreted indozens of different ways, depending on how each of its constituents isheard. Since we hold in immediate memory what we’ve just heard and integrateit with the stream of new music arriving at our ears and brains,attentive listeners to Joni’s music—even nonmusicians—can write andrewrite in their minds a multitude of musical interpretations as the pieceunfolds; and each new listening brings a new set of contexts, expectations,and interpretations. In this sense, Joni’s music is as close to impressionistvisual art as anything I’ve heard.As soon as a bass player plays a note, he fixes one particular musicalinterpretation, thus ruining the delicate ambiguity the composer has so

What Makes a Musician? 211artfully constructed. All of the bass players Joni worked with beforeJaco insisted on playing roots, or what they perceived to be roots. Thebrilliance of Jaco, Joni said, is that he instinctively knew to wanderaround the possibility space, reinforcing the different chord interpretationswith equal emphasis, sublimely holding the ambiguity in a delicate,suspended balance. Jaco allowed Joni to have bass guitar on her songswithout destroying one of their most expansive qualities. This, then, wefigured out at dinner that night, was one of the secrets of why Joni’s musicsounds unlike anyone else’s—its harmonic complexity born out ofher strict insistence that the music not be anchored to a single harmonicinterpretation. Add in her compelling, phonogenic voice, and we becomeimmersed in an auditory world, a soundscape unlike any other.Musical memory is another aspect of musical expertise. Many of usknow someone who remembers all kinds of details that the rest of uscan’t. This could be a friend who remembers every joke he’s ever heardin his life, while some of us can’t even retell one we’ve heard that sameday. My colleague Richard Parncutt, a well-known musicologist and musiccognition professor at the University of Graz in Vienna, used to playpiano in a tavern to earn money for graduate school. Whenever he comesto Montreal to visit me he sits down at the piano in my living room andaccompanies me while I sing. We can play together for a long time: Anysong I name, he can play from memory. He also knows the different versionsof songs: If I ask him to play “Anything Goes,” he’ll ask if I want theversion by Sinatra, Ella Fitzgerald, or Count Basie! Now, I can probablyplay or sing a hundred songs from memory. That is typical for someonewho has played in bands or orchestras, and who has performed. ButRichard seems to know thousands and thousands of songs, both thechords and lyrics. How does he do it? Is it possible for mere memorymortals like me to learn to do this too?When I was in music school, at the Berklee College of Music inBoston, I ran into someone with an equally remarkable form of musicalmemory, but different from Richard’s. Carla could recognize a piece ofmusic within just three or four seconds and name it. I don’t actually

212 This Is Your Brain on Musicknow how good she was at singing songs from memory, because wewere always busy trying to come up with a melody to stump her, and thiswas hard to do. Carla eventually took a job at the American Society ofComposers and Publishers (ASCAP), a composers’ rights organizationthat monitors radio station playlists in order to collect royalties forASCAP members. ASCAP workers sit in a room in Manhattan all day, listeningto excerpts from radio programs all over the country. To be efficientat their job, and indeed to be hired in the first place, they have to beable to name a song and the performer within just three to five secondsbefore writing it down in the log and moving on to the next one.Earlier, I mentioned Kenny, the boy with Williams syndrome whoplays the clarinet. Once when Kenny was playing “The Entertainer” (thetheme song from The Sting), by Scott Joplin, he had difficulty with a certainpassage. “Can I try that again?” he asked me, with an eagerness toplease that is typical of Williams syndrome. “Of course,” I said. Instead ofbacking up just a few notes or a few seconds in the piece, however, hewent all the way back to the beginning! I had seen this before, in recordingstudios, with master musicians from Carlos Santana to the Clash—atendency to go back, if not to the beginning of the entire piece, to the beginningof a phrase. It is as though the musician is executing a memorizedsequence of muscle movements, and the sequence has to beginfrom the beginning.What do these three demonstrations of memory for music have incommon? What is going on in the brains of someone with a fantastic musicalmemory like Richard and Carla, or the “finger memory” that Kennyhas? How might those operations be different from—or similar to—thenormal neural processes in someone with a merely ordinary musicalmemory? Expertise in any domain is characterized by a superior memory,but only for things within the domain of expertise. My friend Richarddoesn’t have a superior memory for everything in life—he still loses hiskeys just like anyone else. Grandmaster chess players have memorizedthousands of board and game configurations. However, their exceptionalmemory for chess extends only to legal positions of the chesspieces. Asked to memorize random arrangements of pieces on a board,

What Makes a Musician? 213they do no better than novices; in other words, their knowledge of chesspiecepositions is schematized, and relies on knowledge of the legalmoves and positions that pieces can take. Likewise, experts in musicrely on their knowledge of musical structure. Expert musicians excel atremembering chord sequences that are “legal” or make sense within theharmonic systems that they have experience with, but they do no betterthan anyone else at learning sequences of random chords.When musicians memorize songs, then, they are relying on a structurefor their memory, and the details fit into that structure. This is an efficientand parsimonious way for the brain to function. Rather than memorizingevery chord or every note, we build up a framework within which manydifferent songs can fit, a mental template that can accommodate a largenumber of musical pieces. When learning to play Beethoven’s “Pathétique”Sonata, the pianist can learn the first eight measures and then, forthe next eight, simply needs to know that the same theme is repeated butan octave higher. Any rock musician can play “One After 909” by the Beatleseven if he’s never played it before, if he is simply told that it is a“standard sixteen-bar blues progression.” That phrase is a frameworkwithin which thousands of songs fit. “One After 909” has certain nuancesthat constitute variations of the framework. The point is that musiciansdon’t typically learn new pieces one note at a time once they havereached a certain level of experience, knowledge, and proficiency. Theycan scaffold on the previous pieces they know, and just note any variationsfrom the standard schema.Memory for playing a musical piece therefore involves a process verymuch like that for music listening as we saw in Chapter 4, through establishingstandard schemas and expectation. In addition, musicians usechunking, a way of organizing information similar to the way chess players,athletes, and other experts organize information. Chunking refersto the process of tying together units of information into groups, and rememberingthe group as a whole rather than the individual pieces. We dothis all the time without much conscious awareness when we have to remembersomeone’s long-distance phone number. If you’re trying to rememberthe phone number of someone in New York City—and if you

214 This Is Your Brain on Musicknow other NYC phone numbers and are familiar with them—you don’thave to remember the area code as three individual numerals, rather,you remember it as a single unit: 212. Likewise, you may know that LosAngeles is 213, Atlanta is 404, or that the country code for England is 44.The reason that chunking is important is because our brains have limitson how much information they can actively keep track of. There isno practical limit to long-term memory that we know of, but workingmemory—the contents of our present awareness—is severely limited,generally to nine pieces of information. Encoding a North Americanphone number as the area code (one unit of information) plus seven digitshelps us to avoid that limit. Chess players also employ chunking, rememberingboard configurations in terms of groups of pieces arrangedin standard, easy-to-name patterns.Musicians also use chunking in several ways. First, they tend to encodein memory an entire chord, rather than the individual notes ofthe chord; they remember “C major 7” rather than the individual tonesC - E - G - B, and they remember the rule for constructing chords, so thatthey can create those four tones on the spot from just one memory entry.Second, musicians tend to encode sequences of chords, rather than isolatedchords. “Plagal cadence,” “aeolian cadence,” “twelve-bar minorblues with a V-I turnaround,” or “rhythm changes” are shorthand labelsthat musicians use to describe sequences of varying lengths. Havingstored the information about what these labels mean allows the musicianto recall big chunks of information from a single memory entry.Third, we obtain knowledge as listeners about stylistic norms, and asplayers about how to produce these norms. Musicians know how to takea song and apply this knowledge—schemas again—to make the songsound like salsa, or grunge, or disco, or heavy metal; each genre and erahas stylistic tics or characteristic rhythmic, timbral, or harmonic elementsthat define it. We can encode those in memory holistically, andthen retrieve these features all at once.These three forms of chunking are what Richard Parncutt uses whenhe sits at the piano to play thousands of songs. He also knows enoughmusic theory and is acquainted enough with different styles and genres

What Makes a Musician? 215that he can fake his way through a passage he doesn’t really know, justas an actor might substitute words that aren’t in the script if she momentarilyforgets her lines. If Richard is unsure of a note or chord, he’llreplace it with one that is stylistically plausible.Identification memory—the ability that most of us have to identifypieces of music that we’ve heard before—is similar to memory for faces,photos, even tastes and smells, and there is individual variability, withsome people simply being better than others; it is also domain specific,with some people—like my classmate Carla—being especially good atmusic, while others excel in other sensory domains. Being able to rapidlyretrieve a familiar piece of music from memory is one skill, but beingable to then quickly and effortlessly attach a label to it, such as thesong title, artist, and year of recording (which Carla could do) involves aseparate cortical network, which we now believe involves the planumtemporale (a structure associated with absolute pitch) and regions ofthe inferior prefrontal cortex that are known to be required for attachingverbal labels to sensory impressions. Why some people are better at thisthan others is still unknown, but it may result from an innate or hardwiredpredisposition in the way their brains formed, and this in turn mayhave a partial genetic basis.When learning sequences of notes in a new musical piece, musicianssometimes have to resort to the brute-force approach that most of ustook as children in learning new sequences of sounds, such as the alphabet,the U.S. Pledge of Allegiance, or the Lord’s Prayer: We simply doeverything we can to memorize the information by repeating it over andover again. But this rote memorization is greatly facilitated by a hierarchicalorganization of the material. Certain words in a text or notes in amusical piece (as we saw in Chapter 4) are more important than othersstructurally, and we organize our learning around them. This sort ofplain old memorization is what musicians do when they learn the musclemovements necessary to play a particular piece; it is part of the reasonthat musicians like Kenny can’t start playing on just any note, but tend togo to the beginnings of meaningful units, the beginnings of their hierarchicallyorganized chunks.

216 This Is Your Brain on Music* * *Being an expert musician thus take many forms: dexterity at playingan instrument, emotional communication, creativity, and special mentalstructures for remembering music. Being an expert listener, which mostof us are by age six, involves having incorporated the grammar of ourmusical culture into mental schemas that allow us to form musical expectations,the heart of the aesthetic experience in music. How all thesevarious forms of expertise are acquired is still a neuroscientific mystery.The emerging consensus, however, is that musical expertise is not onething, but involves many components, and not all musical experts will beendowed with these different components equally—some, like IrvingBerlin, may lack what most of us would even consider a fundamental aspectof musicianship, being able to play an instrument well. It seems unlikelyfrom what we now know that musical expertise is wholly differentfrom expertise in other domains. Although music certainly uses brainstructures and neural circuits that other activities don’t, the process of becominga musical expert—whether a composer or performer—requiresmany of the same personality traits as becoming an expert in other domains,especially diligence, patience, motivation, and plain old-fashionedstick-to-it-iveness.Becoming a famous musician is another matter entirely, and may nothave as much to do with intrinsic factors or ability as with charisma, opportunity,and luck. An essential point bears repeating, however: All ofus are expert musical listeners, able to make quite subtle determinationsof what we like and don’t like, even when we’re unable to articulate thereasons why. Science does have something to say about why we like themusic we do, and that story is another interesting facet of the interplaybetween neurons and notes.

8. My Favorite ThingsWhy Do We Like the Music We Like?You wake from a deep sleep and open your eyes. It’s dark. The distantregular beating at the periphery of your hearing is still there. Yourub your eyes with your hands, but you can’t make out any shapes orforms. Time passes, but how long? Half an hour? One hour? Then youhear a different but recognizable sound—an amorphous, moving, wigglysound with fast beating, a pounding that you can feel in your feet. Thesounds start and stop without definition. Gradually building up and dyingdown, they weave together with no clear beginnings or endings.These familiar sounds are comforting, you’ve heard them before. As youlisten, you have a vague notion of what will come next, and it does, evenas the sounds remain remote and muddled, as though you’re listeningunderwater.Inside the womb, surrounded by amniotic fluid, the fetus hears sounds.It hears the heartbeat of its mother, at times speeding up, at other timesslowing down. And the fetus hears music, as was recently discovered byAlexandra Lamont of Keele University in the UK. She found that, a yearafter they are born, children recognize and prefer music they were exposedto in the womb. The auditory system of the fetus is fully functionalabout twenty weeks after conception. In Lamont’s experiment, mothersplayed a single piece of music to their babies repeatedly during the final

218 This Is Your Brain on Musicthree months of gestation. Of course, the babies were also hearing—through the waterlike filtering of the amniotic fluid in the womb—all ofthe sounds of their mothers’ daily life, including other music, conversations,and environmental noises. But one particular piece was singledout for each baby to hear on a regular basis. The singled-out pieces includedclassical (Mozart, Vivaldi), Top 40 (Five, Backstreet Boys), reggae(UB40, Ken Boothe) and world beat (Spirits of Nature). After birth,the mothers were not allowed to play the experimental song to their infants.Then, one year later, Lamont played babies the music that they hadheard in the womb, along with another piece of music chosen to bematched for style and tempo. For example, a baby who had heard UB40’sreggae track “Many Rivers to Cross” heard that piece again, a year later,along with “Stop Loving You” by the reggae artist Freddie McGregor. Lamontthen determined which one the babies preferred.How do you know which of two stimuli a preverbal infant prefers?Most infant researchers use a technique known as the conditioned headturningprocedure, developed by Robert Fantz in the 1960s, and refinedby John Columbo, Anne Fernald, the late Peter Jusczyk, and their colleagues.Two loudspeakers are set up in the laboratory and the infant isplaced (usually on his mother’s lap) between the speakers. When the infantlooks at one speaker, it starts to play music or some other sound,and when he looks at the other speaker, it starts to play different musicor a different sound. The infant quickly learns that he can control whatis playing by where he is looking; he learns, that is, that the conditions ofthe experiment are under his control. The experimenters make sure thatthey counterbalance (randomize) the location that the different stimulicome from; that is, half the time the stimulus under study comes fromone speaker and half the time it comes from the other. When Lamont didthis with the infants in her study, she found that they tended to looklonger at the speaker that was playing music they had heard in the wombthan at the speaker playing the novel music, confirming that they preferredthe music to which they had the prenatal exposure. A controlgroup of one-year-olds who had not heard any of the music before

My Favorite Things 219showed no preference, confirming that there was nothing about the musicitself that caused these results. Lamont also found that, all things beingequal, the young infant prefers fast, upbeat music to slow music.These findings contradict the long-standing notion of childhoodamnesia—that we can’t have any veridical memories before around theage of five. Many people claim to have memories from early childhoodaround age two and three, but it is difficult to know whether these aretrue memories of the original event, or rather, memory of someone tellingus about the event later. The young child’s brain is still undeveloped, functionalspecialization of the brain isn’t complete, and neural pathways arestill in the process of being made. The child’s mind is trying to assimilateas much information as possible in as short a time as possible; there aretypically large gaps in the child’s understanding, awareness, or memoryfor events because he hasn’t yet learned how to distinguish importantevents from unimportant ones, or to encode experience systematically.Thus, the young child is a prime candidate for suggestion, and could unwittinglyencode, as his own, stories that were told to him about himself.It appears that for music even prenatal experience is encoded in memory,and can be accessed in the absence of language or explicit awarenessof the memory.A study made the newspapers and morning talk shows several years ago,claiming that listening to Mozart for ten minutes a day made you smarter(“the Mozart Effect”). Specifically, music listening, it was claimed, canimprove your performance on spatial-reasoning tasks given immediatelyafter the listening session (which some journalists thought implied mathematicalability as well). U.S. congressmen were passing resolutions, thegovernor of Georgia appropriated funds to buy a Mozart CD for everynewborn baby Georgian. Most scientists found ourselves in an uncomfortableposition. Although we do believe intuitively that music can enhanceother cognitive skills, and although we would all like to see moregovernmental funding for school music programs, the actual study thatclaimed this contained many scientific flaws. The study was claiming

220 This Is Your Brain on Musicsome of the right things but for the wrong reasons. Personally, I found allthe hubbub a bit offensive because the implication was that musicshould not be studied in and of itself, or for its own right, but only if itcould help people to do better on other, “more important” things. Thinkhow absurd this would sound if we turned it inside out. If I claimed thatstudying mathematics helped musical ability, would policy makers startpumping money into math for that reason? Music has often been thepoor stepchild of public schools, the first program to get cut when thereare funding problems, and people frequently try to justify it in terms ofits collateral benefits, rather than letting music exist for its own rewards.The problem with the “music makes you smarter” study turned out tobe straightforward: The experimental controls were inadequate, and thetiny difference in spatial ability between the two groups, according to researchby Bill Thompson, Glenn Schellenberg, and others, all turned onthe choice of a control task. Compared to sitting in a room and doingnothing, music listening looked pretty good. But if subjects in the controltask were given the slightest mental stimulation—hearing a book ontape, reading, etc.—there was no advantage for music listening. Anotherproblem with the study was that there was no plausible mechanism proposedby which this might work—how could music listening increasespatial performance?Glenn Schellenberg has pointed out the importance of distinguishingshort-term from long-term effects of music. The Mozart Effect referredto immediate benefits, but other research has revealed long-term effectsof musical activity. Music listening enhances or changes certain neuralcircuits, including the density of dendritic connections in the primary auditorycortex. The Harvard neuroscientist Gottfried Schlaug has shownthat the front portion of the corpus callosum—the mass of fibers connectingthe two cerebral hemispheres—is significantly larger in musiciansthan nonmusicians, and particularly for musicians who began theirtraining early. This reinforces the notion that musical operations becomebilateral with increased training, as musicians coordinate and recruitneural structures in both the left and right hemispheres.Several studies have found microstructural changes in the cerebel-

My Favorite Things 221lum after the acquisition of motor skills, such as are acquired by musicians,including an increased number and density of synapses. Schlaugfound that musicians tended to have larger cerebellums than nonmusicians,and an increased concentration of gray matter; gray matter is thatpart of the brain that contains the cell bodies, axons, and dendrites, andis understood to be responsible for information processing, as opposedto white matter, which is responsible for information transmission.Whether these structural changes in the brain translate to enhancedabilities in nonmusical domains has not been proven, but music listeningand music therapy have been shown to help people overcome a broadrange of psychological and physical problems. But, to return to a morefruitful line of inquiry regarding musical taste ...Lamont’s results areimportant because they show that the prenatal and newborn brain areable to store memories and retrieve them over long periods of time.More practically, the results indicate that the environment—even whenmediated by amniotic fluid and by the womb—can affect a child’s developmentand preferences. So the seeds of musical preference are sown inthe womb, but there must be more to the story than that, or childrenwould simply gravitate toward the music their mothers like, or that playsin Lamaze classes. What we can say is that musical preferences are influenced,but not determined, by what we hear in the womb. There alsois an extended period of acculturation, during which the infant takes inthe music of the culture she is born into. There were reports a few yearsago that prior to becoming used to the music of a foreign (to us) culture,all infants prefer Western music to other musics, regardless of their cultureor race. These findings were not corroborated, but rather, it wasfound that infants do show a preference for consonance over dissonance.Appreciating dissonance comes later in life, and people differ inhow much dissonance they can tolerate.There is probably a neural basis for this. Consonant intervals and dissonantintervals are processed via separate mechanisms in the auditorycortex. Recent results from studying the electrophysiological responsesof humans and monkeys to sensory dissonance (that is, chords thatsound dissonant by virtue of their frequency ratios, not due to any har-

222 This Is Your Brain on Musicmonic or musical context) show that neurons in the primary auditorycortex—the first level of cortical processing for sound—synchronizetheir firing rates during dissonant chords, but not during consonantchords. Why that would create a preference for consonance is not yetclear.We do know a bit about the infant’s auditory world. Although infantears are fully functioning four months before birth, the developing brainrequires months or years to reach full auditory processing capacity. Infantsrecognize transpositions of pitch and of time (tempo changes), indicatingthey are capable of relational processing, something that eventhe most advanced computers still can’t do very well. Jenny Saffran ofthe University of Wisconsin and Laurel Trainor of McMaster Universityhave gathered evidence that infants can also attend to absolute-pitch cuesif the task requires it, suggesting a cognitive flexibility previously unknown:Infants can employ different modes of processing—presumablymediated by different neural circuits—depending on what will best helpthem to solve the problem at hand.Trehub, Dowling, and others have shown that contour is the mostsalient musical feature for infants, who can detect contour similaritiesand differences even across thirty seconds of retention. Recall that contourrefers to the pattern of musical pitch in a melody—the sequence ofups and downs that the melody takes—regardless of the size of the interval.Someone attending to contour exclusively would encode only thatthe melody goes up, for example, but not by how much. Infants’ sensitivityto musical contour parallels their sensitivity to linguistic contours—which separate questions from exclamations, for example, and which arepart of what linguists call prosody. Fernald and Trehub have documentedthe ways in which parents speak differently to infants than toolder children and adults, and this holds across cultures. The resultingmanner of speaking uses a slower tempo, an extended pitch range, and ahigher overall pitch level.Mothers (and to a lesser extent, fathers) do this quite naturally withoutany explicit instruction to do so, using an exaggerated intonationthat the researchers call infant-directed speech or motherese. We be-

My Favorite Things 223lieve that motherese helps to call the babies’ attention to the mother’svoice, and helps to distinguish words within the sentence. Instead of saying,as we would to an adult, “This is a ball,” motherese would entailsomething like, “Seeeeee?” (with the pitch of the eee’s going up to theend of the sentence). “See the BAAAAAALLLLLL?” (with the pitch coveringan extended range and going up again at the end of the word ball).In such utterances, the contour is a signal that the mother is asking aquestion or making a statement, and by exaggerating the differences betweenup and down contours, the mother calls attention to them. In effect,the mother is creating a prototype for a question and a prototype fora declaration, and ensuring that the prototypes are easily distinguishable.When a mother gives an exclamatory scold, quite naturally—andagain without explicit training—she is likely to create a third type ofprototypical utterance, one that is short and clipped, without much pitchvariation: “No!” (pause) “No! Bad!” (pause) “I said no!” Babies seem tocome hardwired with an ability to detect and track contour, preferentially,over specific pitch intervals.Trehub also showed that infants are more able to encode consonantintervals such as perfect fourth and perfect fifth than dissonant ones, likethe tritone. Trehub found that the unequal steps of our scale make it easierto process intervals even early in infancy. She and her colleaguesplayed nine-month-olds the regular seven-note major scale and two scalesshe invented. For one of these invented scales, she divided the octaveinto eleven equal-space steps and then selected seven tones that madeone- and two-step patterns, and for the other she divided the octave intoseven equal steps. The infants’ task was to detect a mistuned tone. Adultsperformed well with the major scale, but poorly with both of the artificial,never-before-heard scales. In contrast, the infants did equally well onboth unequally tuned scales and on the equally tuned ones. From priorwork, it is believed that nine-month-olds have not yet incorporated amental schema for the major scale, so this suggests a general processingadvantage for unequal steps, something our major scale has.In other words, our brains and the musical scales we use seem tohave coevolved. It is no accident that we have the funny, asymmetric

224 This Is Your Brain on Musicarrangement of notes in the major scale: It is easier to learn melodieswith this arrangement, which is a result of the physics of sound production(via the overtone series we visited earlier); the set of tones we usein our major scale are very close in pitch to the tones that constitutethe overtone series. Very early in childhood, most children start to spontaneouslyvocalize, and these early vocalizations can sound a lot likesinging. Babies explore the range of their voices, and begin to explorephonetic production, in response to the sounds they are bringing in fromthe world around them. The more music they hear, the more likely theyare to include pitch and rhythmic variations in their spontaneous vocalizations.Young children start to show a preference for the music of their cultureby age two, around the same time they begin to develop specializedspeech processing. At first, children tend to like simple songs, wheresimple means music that has clearly defined themes (as opposed to, say,four-part counterpoint) and chord progressions that resolve in directand easily predictable ways. As they mature, children start to tire of easilypredictable music and search for music that holds more challenge.According to Mike Posner, the frontal lobes and the anterior cingulate—a structure just behind the frontal lobes that directs attention—are notfully formed in children, leading to an inability to pay attention to severalthings at once; children show difficulty attending to one stimulus whendistracters are present. This accounts for why children under the age ofeight or so have so much difficulty singing “rounds” like “Row, Row, RowYour Boat.” Their attentional system—specifically the network that connectsthe cingulate gyrus (the larger structure within which the anteriorcingulate sits) and the orbitofrontal regions of the brain—cannot adequatelyfilter out unwanted or distracting stimuli. Children who have notyet reached the developmental stage of being able to exclude irrelevantauditory information face a world of great sonic complexity with allsounds coming in as a sensory barrage. They may try to follow the partof the song that their group is supposed to be singing, only to be distractedand tripped up by the competing parts in the round. Posner has

My Favorite Things 225shown that certain exercises adapted from attention and concentrationgames used by NASA can help accelerate the development of the child’sattentional ability.The developmental trajectory, in children, of first preferring simpleand then more complex songs is a generalization, of course; not all childrenlike music in the first place, and some children develop a taste formusic that is off the beaten path, oftentimes through pure serendipity. Ibecame fascinated with big band and swing music when I was eight,around the time my grandfather gave me his collection of 78 rpm recordsfrom the World War II era. I was initially attracted by novelty songs, suchas “The Syncopated Clock,” “Would You Like to Swing on a Star,” “TheTeddy Bear’s Picnic,” and “Bibbidy Bobbidy Boo”—songs that weremade for children. But sufficient exposure to the relatively exotic chordpatterns and voicings of Frank de Vol’s and Leroy Anderson’s orchestrasbecame part of my mental wiring, and I soon found myself listening to allkinds of jazz; the children’s jazz opened the neural doors to make jazz ingeneral palatable and understandable.Researchers point to the teen years as the turning point for musicalpreferences. It is around the age of ten or eleven that most children takeon music as a real interest, even those children who didn’t express suchan interest in music earlier. As adults, the music we tend to be nostalgicfor, the music that feels like it is “our” music, corresponds to the musicwe heard during these years. One of the first signs of Alzheimer’s disease(a disease characterized by changes in nerve cells and neurotransmitterlevels, as well as destruction of synapses) in older adults is memory loss.As the disease progresses, memory loss becomes more profound. Yetmany of these old-timers can still remember how to sing the songs theyheard when they were fourteen. Why fourteen? Part of the reason weremember songs from our teenage years is because those years weretimes of self-discovery, and as a consequence, they were emotionallycharged; in general, we tend to remember things that have an emotionalcomponent because our amygdala and neurotransmitters act in concertto “tag” the memories as something important. Part of the reason also

226 This Is Your Brain on Musichas to do with neural maturation and pruning; it is around fourteenthat the wiring of our musical brains is approaching adultlike levels ofcompletion.There doesn’t seem to be a cutoff point for acquiring new tastes inmusic, but most people have formed their tastes by the age of eighteenor twenty. Why this is so is not clear, but several studies have found it tobe the case. Part of the reason may be that in general, people tend to becomeless open to new experiences as they age. During our teenageyears, we begin to discover that there exists a world of different ideas,different cultures, different people. We experiment with the idea that wedon’t have to limit our life’s course, our personalities, or our decisions towhat we were taught by our parents, or to the way we were brought up.We also seek out different kinds of music. In Western culture in particular,the choice of music has important social consequences. We listen tothe music that our friends listen to. Particularly when we are young, andin search of our identity, we form bonds or social groups with peoplewhom we want to be like, or whom we believe we have something incommon with. As a way of externalizing the bond, we dress alike, shareactivities, and listen to the same music. Our group listens to this kind ofmusic, those people listen to that kind of music. This ties into the evolutionaryidea of music as a vehicle for social bonding and societal cohesion.Music and musical preferences become a mark of personal andgroup identity and of distinction.To some degree, we might say that personality characteristics are associatedwith, or predictive of, the kind of music that people like. But toa large degree, it is determined by more or less chance factors: whereyou went to school, who you hung out with, what music they happenedto be listening to. When I lived in northern California as a kid, CreedenceClearwater Revival was huge—they were from just down the road. WhenI moved to southern California, CCR’s brand of quasi-cowboy, countryhickmusic didn’t fit in well with the surfer/Hollywood culture that embracedthe Beach Boys and more theatrical performance artists likeDavid Bowie.

My Favorite Things 227Also, our brains are developing and forming new connections at anexplosive rate throughout adolescence, but this slows down substantiallyafter our teenage years, the formative phase when our neural circuitsbecome structured out of our experiences. This process applies tothe music we hear; new music becomes assimilated within the frameworkof the music we were listening to during this critical period. Weknow that there are critical periods for acquiring new skills, such as language.If a child doesn’t learn language by the age of six or so (whether afirst or a second language), the child will never learn to speak with theeffortlessness that characterizes most native speakers of a language.Music and mathematics have an extended window, but not an unlimitedone: If a student hasn’t had music lessons or mathematical training priorto about age twenty, he can still learn these subjects, but only with greatdifficulty, and it’s likely that he will never “speak” math or music likesomeone who learned them early. This is because of the biologicalcourse for synaptic growth. The brain’s synapses are programmed togrow for a number of years, making new connections. After that time,there is a shift toward pruning, to get rid of unneeded connections.Neuroplasticity is the ability of the brain to reorganize itself. Althoughin the last five years there have been some impressive demonstrationsof brain reorganization that used to be thought impossible, theamount of reorganization that can occur in most adults is vastly less thancan occur in children and adolescents.Of course, there are individual differences. Just as some people canheal broken bones or skin cuts faster than others, so, too, can somepeople forge new connections more easily than others. Generally, betweenthe ages of eight and fourteen, pruning starts to occur in thefrontal lobes, the seat of higher thought and reasoning, planning, and impulsecontrol. Myelination starts to ramp up during this time. Myelin is afatty substance that coats the axons, speeding up synaptic transmission.(This is why as children get older, generally, problem solving becomesmore rapid and they are able to solve more complex problems.) Myelinationof the whole brain is generally completed by age twenty. Multiple

228 This Is Your Brain on Musicsclerosis is one of several degenerative diseases that can affect themyelin sheath surrounding the neurons.The balance between simplicity and complexity in music also informsour preferences. Scientific studies of like and dislike across a variety ofaesthetic domains—painting, poetry, dance, and music—have shownthat an orderly relationship exists between the complexity of an artisticwork and how much we like it. Of course, complexity is an entirely subjectiveconcept. In order for the notion to make any sense, we have to allowfor the idea that what seems impenetrably complex to Stanley mightfall right in the “sweet spot” of preference for Oliver. Similarly, what oneperson finds insipid and hideously simple, another person might find difficultto understand, based on differences in background, experience,understanding, and cognitive schemas.In a sense, schemas are everything. They frame our understanding;they’re the system into which we place the elements and interpretationsof an aesthetic object. Schemas inform our cognitive models and expectations.With one schema, Mahler’s Fifth is perfectly interpretable, evenupon hearing it for the first time: It is a symphony, it follows symphonicform with four movements; it contains a main theme and subthemes, andrepetitions of the theme; the themes are manifested through orchestralinstruments, as opposed to African talking drums or fuzz bass. Those familiarwith Mahler’s Fourth will recognize that the Fifth opens with avariation on that same theme, and even at the same pitch. Those well acquaintedwith Mahler’s work will recognize that the composer includesquotations from three of his own songs. Musically educated listeners willbe aware that most symphonies from Haydn to Brahms and Brucknertypically begin and end in the same key. Mahler flouts this conventionwith his Fifth, moving from C-sharp minor to A minor and finally endingin D major. If you had not learned to hold in your mind a sense of key asthe symphony develops, or if you did not have a sense of the normal trajectoryof a symphony, this would be meaningless; but for the seasonedlistener, this flouting of convention brings a rewarding surprise, a violationof expectations, especially when such key changes are done skill-

My Favorite Things 229fully so as not to be jarring. Lacking a proper symphonic schema, or ifthe listener holds another schema, perhaps that of an aficionado of Indianragas, Mahler’s Fifth is nonsensical or perhaps rambling, one musicalidea melding amorphously into the next, with no boundaries, nobeginnings or endings that appear as part of a coherent whole. Theschema frames our perception, our cognitive processing, and ultimatelyour experience.When a musical piece is too simple we tend not to like it, findingit trivial. When it is too complex, we tend not to like it, finding itunpredictable—we don’t perceive it to be grounded in anything familiar.Music, or any art form for that matter, has to strike the right balance betweensimplicity and complexity in order for us to like it. Simplicity andcomplexity relate to familiarity, and familiarity is just another word fora schema.It is important in science, of course, to define our terms. What is “toosimple” or “too complex”? An operational definition is that we find apiece too simple when we find it trivially predictable, similar to somethingwe have experienced before, and without the slightest challenge.By analogy, consider the game tic-tac-toe. Young children find it endlesslyfascinating, because it has many features that contribute to interestat their level of cognitive ability: It has clearly defined rules that anychild can easily articulate; it has an element of surprise in that the playernever knows for sure exactly what her opponent will do next; the gameis dynamic, in that one’s own next move is influenced by what one’s opponentdid; when the game will end, who will win, or whether it will bea draw is undetermined, yet there is an outer limit of nine moves. That indeterminacyleads to tension and expectations, and the tension is finallyreleased when the game is over.As the child develops increasing cognitive sophistication, she eventuallylearns strategies—the person who moves second cannot win againsta competent player; the best the second player can hope for is a draw.When the sequence of moves and the end point of the game become predictable,tic-tac-toe loses its appeal. Of course, adults can still enjoy

230 This Is Your Brain on Musicplaying the game with children, but we enjoy seeing the pleasure on thechild’s face and we enjoy the process—spread out over several years—of the child learning to unlock the mysteries of the game as her braindevelops.To many adults, Raffi and Barney the Dinosaur are the musical equivalentsof tic-tac-toe. When music is too predictable, the outcome too certain,and the “move” from one note or chord to the next contains noelement of surprise, we find the music unchallenging and simplistic. Asthe music is playing (particularly if you’re engaged with focused attention),your brain is thinking ahead to what the different possibilities forthe next note are, where the music is going, its trajectory, its intendeddirection, and its ultimate end point. The composer has to lull us intoa state of trust and security; we have to allow him to take us on a harmonicjourney; he has to give us enough little rewards—completions ofexpectations—that we feel a sense of order and a sense of place.Say you’re hitchhiking from Davis, California, to San Francisco. Youwant the person who picks you up to take the normal route, Highway 80.You might be willing to tolerate a few shortcuts, especially if the driver isfriendly, believable, and is up-front about what he’s doing. (“I’m just goingto cut over here on Zamora Road to avoid some construction on thefreeway.”) But if the driver takes you out on back roads with no explanation,and you reach a point where you no longer see any landmarks,your sense of safety is sure to be violated. Of course, different people,with different personality types, react differently to such unanticipatedjourneys, musical or vehicular. Some react with sheer panic (“ThatStravinsky is going to kill me!”) and some react with a sense of adventureat the thrill of discovery (“Coltrane is doing something weird here, butwhat the hell, it won’t hurt me to stick around awhile longer, I can takecare of my harmonic self and find my way back to musical reality if Ihave to”).To continue the analogy with games, some games have such a complicatedset of rules that the average person doesn’t have the patience tolearn them. The possibilities for what can happen on any given turn are

My Favorite Things 231too numerous or too unpredictable (to the novice) to contemplate. Butan inability to predict what will happen next is not always a sign that agame holds eventual interest if only one sticks with it long enough. Agame may have a completely unpredictable course no matter how muchpractice you have with it—many board games simply involve rolling thedice and waiting to see what happens to you. Chutes and Ladders andCandy Land are like this. Children enjoy the sense of surprise, but adultscan find the game tedious because, although no one can predict exactlywhat will happen (the game is a function of the random throw of thedice), the outcome has no structure whatsoever, and moreover, there isno amount of skill on the part of the player that can influence the courseof the game.Music that involves too many chord changes, or unfamiliar structure,can lead many listeners straight to the nearest exit, or to the “skip” buttonon their music players. Some games, such as Go, Axiom, or Zendoare sufficiently complicated or opaque to the novice that many peoplegive up before getting very far: The structure presents a steep learningcurve, and the novice can’t be sure that the time invested will be worthit. Many of us have the same experience with unfamiliar music, or unfamiliarmusical forms. People may tell you that Schönberg is brilliant, orthat Tricky is the next Prince, but if you can’t figure out what is going onin the first minute or so of one of their pieces, you may find yourselfwondering if the payoff will justify the effort you spend trying to sort itall out. We tell ourselves that if we only listen to it enough times, we maybegin to understand it and to like it as much as our friends do. Yet, we recallother times in our lives when we invested hours of prime listeningtime in an artist and never arrived at the point where we “got it.” Tryingto appreciate new music can be like contemplating a new friendship inthat it takes time, and sometimes there is nothing you can do to speed itup. At a neural level, we need to be able to find a few landmarks in orderto invoke a cognitive schema. If we hear a piece of radically new musicenough times, some of that piece will eventually become encoded in ourbrains and we will develop landmarks. If the composer is skillful, those

232 This Is Your Brain on Musicparts of the piece that become our landmarks will be the very ones thatthe composer intended they should be; his knowledge of compositionand human perception and memory will have allowed him to create certain“hooks” in the music that will eventually stand out in our minds.Structural processing is one source of difficulty in appreciating a newpiece of music. Not understanding symphonic form, or the sonata form,or the AABA structure of a jazz standard, is the music-listening equivalentof driving on a highway with no road signs: You never know whereyou are or when you’ll arrive at your destination (or even at an interimspot that is not your destination, but one that provides an orienting landmark).For example, many people just don’t “get” jazz; they say that itsounds like an unstructured, crazy, and formless improvisation, a musicalcompetition to squeeze as many notes as possible into as smalla space as possible. There are more than a half-dozen subgenres ofwhat people collectively call “jazz”: Dixieland, boogie-woogie, big band,swing, bebop, “straight-ahead,” acid-jazz, fusion, metaphysical, and soon. “Straight-ahead,” or “classic jazz,” as it is sometimes called, is moreor less the standard form of jazz, analogous to the sonata or the symphonyin classical music, or what a typical song by the Beatles or BillyJoel or the Temptations is to rock music.In classic jazz, the artist begins by playing the main theme of the song;often a well-known one from Broadway, or one that has already been ahit for someone else; such songs are called “standards,” and they include“As Time Goes By,” “My Funny Valentine,” and “All of Me.” The artistruns through the complete form of the song once—typically two versesand the chorus (otherwise known as a “refrain”), followed by anotherverse. The chorus is the part of a song that repeats regularly throughout;the verses are what change. We call this form AABA, where the letter Arepresents the verse and the letter B represents the chorus. AABA meanswe play verse-verse-chorus-verse. Many other variations are possible, ofcourse. Some songs have a C section, called the bridge.The term chorus is used to mean not just the second section of thesong, but also one run through the entire form. In other words, runningthrough the AABA portion of a song once is called “playing one chorus.”

My Favorite Things 233When I play jazz, if someone says, “Play the chorus,” or, “Let’s go over thechorus” (using the word the), we all assume he means a section of thesong. If, instead someone says, “Let’s run through one chorus,” or, “Let’sdo a couple of choruses,” we know he means the entire form.“Blue Moon” (Frank Sinatra, Billie Holiday) is an example of a songwith AABA form. A jazz artist may play around with the rhythm or feel ofthe song, and may embellish the melody. After playing through the formof the song once, what jazz musicians refer to as “the head,” the differentmembers of the ensemble take turns improvising new music over thechord progression and form of the original song. Each musician playsthrough one or more choruses and then the next musician takes over atthe beginning of the head. During the improvisations, some artists stickclose to the original melody, some add ever distant and exotic harmonicdepartures. When everyone has had a chance to improvise, the band returnsto the head, playing it more or less straight, and then they’re done.The improvisations can go on for many minutes—it is not uncommon fora jazz rendition of a two- or three-minute song to stretch out to ten to fifteenminutes. There is also a typical order to how the musicians taketurns: The horns go first, followed by the piano and/or guitar, followed bythe bass player. Sometimes the drummer also improvises, and he wouldtypically follow the bass. Sometimes the musicians also share part of achorus—each musician playing four or eight measures, and then handingoff the solo to another musician, a sort of musical relay race.To the newbie, the whole thing may seem chaotic. Yet, simply knowingthat the improvisation takes place over the original chords and formof the song can make a big difference in orienting the neophyte to wherein the song the players are. I often advise new listeners to jazz to simplyhum the main tune in their mind once the improvisation begins—this iswhat the improvisers themselves are often doing—and that enriches theexperience considerably.Each musical genre has its own set of rules and its own form. Themore we listen, the more those rules become instantiated in memory.Unfamiliarity with the structure can lead to frustration or a simple lackof appreciation. Knowing a genre or style is to effectively have a cate-

234 This Is Your Brain on Musicgory built around it, and to be able to categorize new songs as being eithermembers or nonmembers of that category—or in some cases, as“partial” or “fuzzy” members of the category, members subject to certainexceptions.The orderly relationship between complexity and liking is referred to asthe inverted-U function because of the way a graph would be drawn thatrelates these two factors. Imagine a graph in which the x-axis is howcomplex a piece of music is (to you) and the y-axis is how much you likeit. At the bottom left of this graph, close to the origin, there would be apoint for music that is very simple and your reaction being that you don’tlike it. As the music increases in complexity, your liking increases as well.The two variables follow each other for quite a while on the graph—increased complexity yields increased liking, until you cross some personalthreshold and go from disliking the piece intensely to actuallyliking it quite a bit. But at some point as we increase complexity, the musicbecomes too complex, and your liking for it begins to decrease. Nowmore complexity in the music leads to less and less liking, until youcross another threshold and you no longer like the music at all. Too complexand you absolutely hate the music. The shape of such a graph wouldmake an inverted U or an inverted V.The inverted-U hypothesis is not meant to imply that the only reasonyou might like or dislike a piece of music is because of its simplicity orcomplexity. Rather, it is intended to account for this variable. The elementsof music can themselves form a barrier to appreciation of a newpiece of music. Obviously, if music is too loud or too soft, this can beproblematic. But even the dynamic range of a piece—the disparity betweenthe loudest and softest parts—can cause some people to reject it.This can be especially true for people who use music to regulate theirmood in a specific way. Someone who wants music to calm her down, orsomeone else who wants music to pep him up for a workout, is probablynot going to want to hear a musical piece that runs the loudness gamutall the way from very soft to very loud, or emotionally from sad to exhilarating(as does Mahler’s Fifth, for example). The dynamic range as

My Favorite Things 235well as the emotional range is simply too wide, and may create a barrierto entry.Pitch can also play into preference. Some people can’t stand thethumping low beats of modern hip-hop, others can’t stand what they describeas the high-pitched whininess of violins. Part of this may be a matterof physiology; literally, different ears may transmit different parts ofthe frequency spectrum, causing some sounds to appear pleasant andothers aversive. There may also exist psychological associations, bothpositive and negative, to various instruments.Rhythm and rhythmic patterns influence our ability to appreciate agiven musical genre or piece. Many musicians are drawn to Latin musicbecause of the complexity of the rhythms. To an outsider, it all justsounds “Latin,” but to someone who can make out the nuances of whena certain beat is strong relative to other beats, Latin music is a wholeworld of interesting complexity: bossa nova, samba, rhumba, beguine,mambo, merengue, tango—each is a completely distinct and identifiablestyle of music. Some people genuinely enjoy Latin music and Latinrhythms without being able to tell them apart, of course, but others findthe rhythms too complicated and unpredictable, and this is a turnoff tothem. I’ve found that if I teach one or two Latin rhythms to listeners, theycome to appreciate them; it is all a question of grounding and having aschema. For other listeners, rhythms that are too simple are the dealbreakerfor a style of music. The typical complaint of my parents’ generationabout rock and roll, apart from how loud it seemed to them, wasthat it all had the same beat.Timbre is another barrier for many people and its influence is almostcertainly increasing, as I argued in Chapter 1. The first time I heard JohnLennon or Donald Fagen sing, I thought the voices unimaginably strange.I didn’t want to like them. Something kept me going back to listen,though—perhaps it was the strangeness—and they wound up being twoof my favorite voices; voices that now have gone beyond familiar and approachwhat I can only call intimate; I feel as though these voices havebecome incorporated into who I am. And at a neural level, they have.Having listened to thousands of hours of both these singers, and tens of

236 This Is Your Brain on Musicthousands of playings of their songs, my brain has developed circuitrythat can pick out their voices from among thousands of others, evenwhen they sing something I’ve never heard them sing before. My brainhas encoded every vocal nuance and every timbral flourish, so that if Ihear an alternate version of one of their songs—as we do on the JohnLennon Collection of demo versions of his albums—I can immediatelyrecognize the ways in which this performance deviates from the one Ihave stored in the neural pathways of my long-term memory.As with other sorts of preferences, our musical preferences are also influencedby what we’ve experienced before, and whether the outcome ofthat experience was positive or negative. If you had a negative experienceonce with pumpkin—say, for example, it made you sick to yourstomach—you are likely to be wary of future excursions into pumpkingustation. If you’ve had only a few, but largely positive, encounters withbroccoli, you might be willing to try a new broccoli recipe, perhaps broccolisoup, even if you’ve never had it before. The one positive experiencebegets others.The types of sounds, rhythms, and musical textures we find pleasingare generally extensions of previous positive experiences we’ve hadwith music in our lives. This is because hearing a song that you like is alot like having any other pleasant sensory experience—eating chocolate,fresh-picked raspberries, smelling coffee in the morning, seeing a workof art or the peaceful face of someone you love who is sleeping. We takepleasure in the sensory experience, and find comfort in its familiarityand the safety that familiarity brings. I can look at a ripe raspberry, smellit, and anticipate that it will taste good and that the experience will besafe—I won’t get sick. If I’ve never seen a loganberry before, there areenough points in common with the raspberry that I can take the chancein eating it and anticipate that it will be safe.Safety plays a role for a lot of us in choosing music. To a certain extent,we surrender to music when we listen to it—we allow ourselves totrust the composers and musicians with a part of our hearts and our spirits;we let the music take us somewhere outside of ourselves. Many of us

My Favorite Things 237feel that great music connects us to something larger than our own existence,to other people, or to God. Even when music doesn’t transport usto an emotional place that is transcendent, music can change our mood.We might be understandably reluctant, then, to let down our guard, todrop our emotional defenses, for just anyone. We will do so if the musiciansand composer make us feel safe. We want to know that our vulnerabilityis not going to be exploited. This is part of the reason why somany people can’t listen to Wagner. Due to his pernicious anti-Semitism,the sheer vulgarity of his mind (as Oliver Sacks describes it), and hismusic’s association with the Nazi regime, some people don’t feel safe listeningto his music. Wagner has always disturbed me profoundly, and notjust his music, but also the idea of listening to it. I feel reluctant to giveinto the seduction of music created by so disturbed a mind and so dangerous(or impenetrably hard) a heart as his, for fear that I might developsome of the same ugly thoughts. When I listen to the music of a greatcomposer I feel that I am, in some sense, becoming one with him, or lettinga part of him inside me. I also find this disturbing with popular music,because surely some of the purveyors of pop are crude, sexist, racist,or all three.This sense of vulnerability and surrender is no more prevalent thanwith rock and popular music in the past forty years. This accounts forthe fandom that surrounds popular musicians—the Grateful Dead, theDave Matthews Band, Phish, Neil Young, Joni Mitchell, the Beatles,R.E.M., Ani DiFranco. We allow them to control our emotions and evenour politics—to lift us up, to bring us down, to comfort us, to inspire us.We let them into our living rooms and bedrooms when no one else isaround. We let them into our ears, directly, through earbuds and headphones,when we’re not communicating with anybody else in the world.It is unusual to let oneself become so vulnerable with a total stranger.Most of us have some kind of protection that prevents us from blurtingout every thought and feeling that comes across our minds. When someoneasks us, “How’re ya doin’?” we say, “Fine,” even if we’re depressedabout a fight we just had at home, or suffering a minor physical ailment.My grandfather used to say that the definition of a bore is someone who

238 This Is Your Brain on Musicwhen you ask him “How are you?” actually tells you. Even with closefriends, there are some things we simply keep hidden—digestive andbowel-related problems, for example, or feelings of self-doubt. One ofthe reasons that we’re willing to make ourselves vulnerable to our favoritemusicians is that they often make themselves vulnerable to us (orthey convey vulnerability through their art—the distinction betweenwhether they are actually vulnerable or merely representing it artisticallyis not important for now).The power of art is that it can connect us to one another, and to largertruths about what it means to be alive and what it means to be human.When Neil Young singsOld man look at my life, I’m a lot like you were. . . .Live alone in a paradise that makes me think of two.we feel for the man who wrote the song. I may not live in a paradise, butI can empathize with a man who may have some material success but noone to share it with, a man who feels he has “gained the world but losthis soul,” as George Harrison once sang, quoting at once the gospel accordingto Mark and Mahatma Gandhi.Or when Bruce Springsteen sings “Back in Your Arms” about losinglove, we resonate to a similar theme, by a poet with a similar “everyman”persona to Neil Young’s. And when we consider how much Springsteenhas—the adoration of millions of people worldwide, and millions ofdollars—it becomes all the more tragic that he cannot have the onewoman he wants.We hear vulnerability in unlikely places and it brings us closer to theartist. David Byrne (of the Talking Heads) is generally known for his abstract,arty lyrics, with a touch of the cerebral. In his solo performanceof “Lilies of the Valley,” he sings about being alone and scared. Part ofour appreciation for this lyric is enhanced by knowing something aboutthe artist, or at least the artist’s persona, as an eccentric intellectual, whorarely revealed something as raw and transparent as being afraid.Connections to the artist or what the artist stands for can thus be part

My Favorite Things 239of our musical preferences. Johnny Cash cultivated an outlaw image,and also showed his compassion for prison inmates by performing manyconcerts in prisons. Prisoners may like Johnny Cash’s music—or growto like it—because of what the artist stands for, quite apart from anystrictly musical considerations. But fans will only go so far to follow theirheroes, as Dylan learned at the Newport Folk Festival. Johnny Cashcould sing about wanting to leave prison without alienating his audience,but if he had said that he liked visiting prisons because it helped him appreciatehis own freedom, he would no doubt have crossed a line fromcompassion to gloating, and his inmate audience would have understandablyturned on him.Preferences begin with exposure and each of us has our own “adventuresomeness”quotient for how far out of our musical safety zone weare willing to go at any given time. Some of us are more open to experimentationthan others in all aspects of our lives, including music; and atvarious times in our life we may seek or avoid experimentation. Generally,the times when we find ourselves bored are those when we seeknew experiences. As Internet radio and personal music players are becomingmore popular, I think that we will be seeing personalized musicstations in the next few years, in which everyone can have his or her ownpersonal radio station, controlled by computer algorithms that play us amixture of music we already know and like and a mixture of music wedon’t know but we are likely to enjoy. I think it will be important thatwhatever form this technology takes, listeners should have an “adverturesomeness”knob they can turn that will control the mix of old andnew, or the mix of how far out the new music is from what they usuallylisten to. This is something that is highly variable from person to person,and even, within one person, from one time of day to the next.Our music listening creates schemas for musical genres and forms,even when we are only listening passively, and not attempting to analyzethe music. By an early age, we know what the legal moves are in themusic of our culture. For many, our future likes and dislikes will be aconsequence of the types of cognitive schemas we formed for musicthrough childhood listening. This isn’t meant to imply that the music

240 This Is Your Brain on Musicwe listen to as children will necessarily determine our musical tastes forthe rest of our lives; many people are exposed to or study music of differentcultures and styles and become acculturated to them, learningtheir schemas as well. The point is that our early exposure is often ourmost profound, and becomes the foundation for further musical understanding.Musical preferences also have a large social component based on ourknowledge of the singer or musician, on our knowledge of what our familyand friends like, and knowledge of what the music stands for. Historically,and particularly evolutionarily, music has been involved withsocial activities. This may explain why the most common form of musicalexpression, from the Psalms of David to Tin Pan Alley to contemporarymusic, is the love song, and why for most of us, love songs seem tobe among our favorite things.

9. The Music InstinctEvolution’s #1 HitWhere did music come from? The study of the evolutionary originsof music has a distinguished history, dating back to Darwin himself,who believed that it developed through natural selection as part ofhuman or paleohuman mating rituals. I believe that the scientific evidencesupports this idea as well, but not everyone agrees. After decadesof only scattered work on the topic, in 1997 interest was suddenly focusedon a challenge issued by the cognitive psychologist and cognitivescientist Steven Pinker.There are about 250 people worldwide who study music perceptionand cognition as a primary research focus. As with most scientific disciplines,we hold conferences once a year. In 1997, the conference washeld at MIT, and Steven Pinker was invited to give the opening address.Pinker had just completed How the Mind Works, an important largescalework that explains and synthesizes the major principles of cognitivescience, but he had not yet found popular notoriety. “Language isclearly an evolutionary adaptation,” he told us during his keynote speech.“The cognitive mechanisms that we, as cognitive psychologists and cognitivescientists, study, mechanisms such as memory, attention, categorization,and decision making, all have a clear evolutionary purpose.” Heexplained that, once in a while, we find a behavior or attribute in an

242 This Is Your Brain on Musicorganism that lacks any clear evolutionary basis; this occurs when evolutionaryforces propagate an adaptation for a particular reason, andsomething else comes along for the ride, what Stephen Jay Gould calleda spandrel, borrowing the term from architecture. In architecture, a designermight plan for a dome to be held up by four arches. There will necessarilybe a space between the arches, not because it was planned for,but because it is a by-product of the design. Birds evolved feathers tokeep warm, but they coopted the feathers for another purpose—flying.This is a spandrel.Many spandrels are put to such good use that it is hard to know afterthe fact whether they were adaptations or not. The space betweenarches in a building became a place where artists painted angelsand other decorations. The spandrel—a by-product of the architects’design—became one of the most beautiful parts of a building. Pinker arguedthat language is an adaptation and music is its spandrel. Among thecognitive operations that humans perform, music is the least interestingto study because it is merely a by-product, he went on, an evolutionaryaccident piggybacking on language.“Music is auditory cheesecake,” he said dismissively. “It just happensto tickle several important parts of the brain in a highly pleasurable way,as cheesecake tickles the palate.” Humans didn’t evolve a liking forcheesecake, but we did evolve a liking for fats and sugars, which were inshort supply during our evolutionary history. Humans evolved a neuralmechanism that caused our reward centers to fire when eating sugarsand fats because in the small quantities they were available, they werebeneficial to our well-being.Most activities that are important for survival of the species, such aseating and sex, are also pleasurable; our brains evolved mechanisms toreward and encourage these behaviors. But we can learn to short-circuitthe original activities and tap directly into these reward systems. We caneat foods that have no nutritive value and we can have sex without procreating;we can take heroin, which exploits the normal pleasure receptorsin the brain; none of these is adaptive, but the pleasure centers inour limbic system don’t know the difference. Humans, then, discovered

The Music Instinct 243that cheesecake just happens to push pleasure buttons for fat and sugar,Pinker explained, and music is simply a pleasure-seeking behavior thatexploits one or more existing pleasure channels that evolved to reinforcean adaptive behavior, presumably linguistic communication.“Music,” Pinker lectured us, “pushes buttons for language ability(with which music overlaps in several ways); it pushes buttons in the auditorycortex, the system that responds to the emotional signals in a humanvoice crying or cooing, and the motor control system that injectsrhythm into the muscles when walking or dancing.”“As far as biological cause and effect are concerned,” Pinker wrote inThe Language Instinct (and paraphrased in the talk he gave to us), “musicis useless. It shows no signs of design for attaining a goal such as longlife, grandchildren, or accurate perception and prediction of the world.Compared with language, vision, social reasoning, and physical knowhow,music could vanish from our species and the rest of our lifestylewould be virtually unchanged.”When a brilliant and respected scientist such as Pinker makes a controversialclaim, the scientific community takes notice, and it caused meand many of my colleagues to reevaluate a position on the evolutionarybasis of music that we had taken for granted, without questioning.Pinker got us thinking. And a little research showed that he is not theonly theorist to deride music’s evolutionary origins. The cosmologistJohn Barrow said that music has no role in survival of the species, andpsychologist Dan Sperber called music “an evolutionary parasite.” Sperberbelieves that we evolved a cognitive capacity to process complexsound patterns that vary in pitch and duration, and that this communicativeability first arose in primitive, prelinguistic humans. Music, accordingto Sperber, developed parasitically to exploit this capacity thathad evolved for true communication. Ian Cross of Cambridge Universitysums up: “For Pinker, Sperber, and Barrow, music exists simply becauseof the pleasure that it affords; its basis is purely hedonic.”I happen to think that Pinker is wrong, but I’ll let the evidence speakfor itself. Let me back up first a hundred and fifty years to Charles Darwin.The catchphrase most of us are taught in school, “survival of the

244 This Is Your Brain on Musicfittest” (unfortunately propagated by the British philosopher HerbertSpencer), is an oversimplification of evolution. The theory of evolutionrests on several assumptions. First, all of our phenotypic attributes (ourappearance, physiological attributes, and some behaviors) are encodedin our genes, which are passed from one generation to the next. Genestell our body how to make proteins, which generate our phenotypiccharacteristics. The action of genes is specific to the cells in which theyreside; a given gene may contain information that is useful or not usefuldepending on the cell in question—cells in your eye don’t need to growskin, for example. Our genotype (particular sequence of DNA) gives riseto our phenotype (particular physical characteristics). So to sum up:Many of the ways in which members of a species differ from one anotherare encoded in the genes, and these are passed on through reproduction.The second assumption of evolutionary theory is that there exists betweenus some natural genetic variability. Third, when we mate, our geneticmaterial combines to form a new being, 50 percent of whosegenetic material comes from each parent. Finally, due to spontaneous errors,mistakes or mutations sometimes occur that may be passed on tothe next generation.The genes that exist in you today (with the exception of a small numberthat may have mutated) are those that reproduced successfully inthe past. Each of us is a victor in a genetic arms race; many genes thatfailed to reproduce successfully died out, leaving no descendants. Everyonealive today is composed of genes that won a long-lasting, large-scalegenetic competition. “Survival of the fittest” is an oversimplification becauseit leads to the distorted view that genes that confer a survival advantagein their host organism are those that will win the genetic race.But living a long life, however happy and productive, does not pass ongenes. An organism needs to reproduce to pass on its genes. The nameof the evolutionary game is to reproduce at all costs, and to see thatone’s offspring live to do the same, and for their offspring to live longenough to do the same, and so on.If an organism lives long enough to reproduce, and if its children arehearty and protected so that they can do the same, there is no com-

The Music Instinct 245pelling evolutionary reason for the organism to live a long time. Someavian species and spiders die during or after sexual mating. The postmatingyears do not confer any advantage to the survival of the organism’sgenes unless it is able to use that time to protect its offspring,secure resources for them, or help them to find mates. Thus, two thingslead to genes’ being “successful”: (1) the organism is able to successfullymate, passing its genes on, and (2) its offspring are able to survive in orderto do the same.Darwin recognized this implication of his theory of natural selectionand came up with the idea of sexual selection. Because an organismmust reproduce to pass its genes on, qualities that will attract a mateshould eventually become encoded in the genome. If a square jaw andoutsized biceps are attractive features for a man to have (in the eyes ofpotential mates), men with those features will reproduce more successfullythan their narrow-jawed, scrawny-armed competitors. The squarejaw,large-bicep genes will then become more plentiful. Offspring alsoneed to be protected from the elements, from predators, from disease,and to be given food and other resources so that they can reproduce.Thus, a gene that promotes nurturing behavior postcopulation couldalso spread throughout the population, to the extent that the offspring ofpeople with the nurturing gene fare better, as a group, in the competitionfor resources and mates.Might music play a role in sexual selection? Darwin thought so. In TheDescent of Man he wrote, “I conclude that musical notes and rhythmwere first acquired by the male or female progenitors of mankind for thesake of charming the opposite sex. Thus musical tones became firmlyassociated with some of the strongest passions an animal is capableof feeling, and are consequently used instinctively. . . .” In seeking mates,our innate drive is to find—either consciously or unconsciously—someone who is biologically and sexually fit, someone who will provideus with children who are likely to be healthy and able to attract mates oftheir own. Music may indicate biological and sexual fitness, serving toattract mates.Darwin believed that music preceded speech as a means of courtship,

246 This Is Your Brain on Musicequating music with the peacock’s tail. In his theory of sexual selection,Darwin posited the emergence of features that served no direct survivalpurpose other than to make oneself (and hence one’s genes) attractive.The cognitive psychologist Geoffrey Miller has connected this notionwith the role that music plays in contemporary society. Jimi Hendrix had“sexual liaisons with hundreds of groupies, maintained parallel longtermrelationships with at least two women, and fathered at least threechildren in the United States, Germany, and Sweden. Under ancestralconditions before birth control, he would have fathered many more,”Miller writes. Robert Plant, the lead singer of Led Zeppelin, recalls hisexperience with their big concert tours in the seventies:“I was on my way to love. Always. Whatever road I took, the car washeading for one of the greatest sexual encounters I’ve ever had.”The number of sexual partners for rock stars can be hundreds oftimes what a normal male has, and for the top rock stars, such as MickJagger, physical appearance doesn’t seem to be an issue.During sexual courtship, animals often advertise the quality of theirgenes, bodies, and minds, in order to attract the best possible mate. Manyhuman-specific behaviors (such as conversation, music production, artisticability, and humor) may have evolved principally to advertise intelligenceduring courtship. Miller suggests that under the conditions thatwould have existed throughout most of our evolutionary history in whichmusic and dance were completely intertwined, musicianship/danceshipwould have been a sign of sexual fitness on two fronts. First, anyone whocould sing and dance was advertising to potential mates his stamina andoverall good health, physical and mental. Second, anyone who had becomeexpert or accomplished in music and dance was advertising that hehad enough food and sturdy enough shelter that he could afford to wastevaluable time on developing a purely unnecessary skill. This is the argumentof the peacock’s beautiful tail: The size of the peacock’s tail correlateswith the bird’s age, health, and overall fitness. The colorful tailsignals that the healthy peacock has metabolism to waste, he is so fit, sotogether, so wealthy (in terms of resources) that he has extra resourcesto put into something that is purely for display and aesthetic purposes.

The Music Instinct 247In contemporary society, we see this with rich people who build elaboratehouses or drive hundred-thousand-dollar cars. The sexual selectionmessage is clear: Choose me. I have so much food and so manyresources that I can afford to squander them on these luxury items. It isno accident that many men living at or near the poverty line in the U.S.buy old Cadillacs and Lincolns—impractical, high-status vehicles thatunconsciously signal their owner’s sexual fitness. This can also be seenas the origin of bling, the tendency for men to wear gaudy jewelry. Thatthe yearning for and purchasing of cars and jewelry peaks in men duringadolescence, when they are most sexually potent, serves the theory. Musicmaking, because it involves an array of physical and mental skills,would be an overt display of health, and to the extent that someone hadtime to develop his musicianship, the argument goes, it would indicateresource wealth.In contemporary society, interest in music also peaks during adolescence,further bolstering the sexual-selection aspects of music. Far morenineteen-year-olds are starting bands and trying to get their hands onnew music than are forty-year-olds, even though the forty-year-olds havehad more time to develop their musicianship and preferences. “Musicevolved and continues to function as a courtship display, mostly broadcastby young males to attract females,” Miller argues.Music as a sexual fitness display is not so farfetched an idea when werealize the form that hunting took in some hunter-gatherer societies.Some protohumans would rely on persistence hunting—hurling spears,rocks, and other projectiles at their prey, then chasing the prey for hoursuntil the animal dropped from injury and exhaustion. If dancing in pasthunter-gatherer societies was anything like what we observe in contemporaryones, it typically extended for many hours, requiring great aerobiceffort. As a display of a male’s fitness to take part in or lead a hunt, suchtribal dancing would be an excellent indicator. Most tribal dancing includesrepeated high-stepping, stomping, and jumping using the largest,most energy-hungry muscles of the body. Many mental illnesses are nowknown to undermine the ability to dance or to perform rhythmically—schizophrenia and Parkinson’s, to name just two—and so the sort of

248 This Is Your Brain on Musicrhythmic dancing and music making that have characterized most musicacross the ages serves as a warranty of physical and mental fitness, perhapseven a warranty of reliability and conscientiousness (because, as wesaw in Chapter 7, expertise requires a particular kind of mental focus).Another possibility is that evolution selected creativity in general as amarker of sexual fitness. Improvisation and novelty in a combined music/danceperformance would indicate the cognitive flexibility of thedancer, signaling his potential for cunning and strategizing while on thehunt. The material wealth of a male suitor has long been consideredamong the most compelling attractors to females, who assume that itwill increase the likelihood of their offspring having ample food, shelter,and protection. (Protection accrues to the wealthy because they canmarshal support of other community members in exchange for food orsymbolic tokens of wealth such as jewelry or cash.) If wealth is the nameof the dating game, then music would seem relatively unimportant. ButMiller and his colleague Martie Haselton at UCLA have shown that creativitytrumps wealth, at least in human females. Their hypothesis is thatwhile wealth may predict who will make a good dad (for child rearing),creativity may better predict who will furnish the best genes (for childfathering).In a clever study, women at various stages of their normal menstrualcycle—some during their peak of fertility, others at their minimum of fertilityand others in between—were asked to rate the attractiveness of potentialmates based on written vignettes describing fictional males. Atypical vignette described a man who was an artist, and who displayedgreat creative intelligence in his work, but who was poor due to badluck. An alternate vignette described a man who had average creative intelligence,but happened to be wealthy due to good luck. All the vignetteswere designed to make clear that each man’s creativity was afunction of his traits and attributes (and thus, endogenous, genetic, andheritable) while each man’s financial state was largely accidental (andthus exogenous and not heritable).The results showed that when they were at their peak fertility, women

The Music Instinct 249preferred the creative but poor artist to the not creative but rich man asa short-term mate, or for a brief sexual encounter. At other times duringtheir cycle, women did not show such preferences. It is important tobear in mind that preferences are to a large degree hardwired and noteasily overpowered by conscious cognitions; the fact that women todaycan avoid pregnancy through almost foolproof birth control is a conceptso new in our species as to have no influence on any innate preferences.The men (and women) who might make the best caregivers are not necessarilythose who can contribute the best genes to potential offspring.People don’t always marry those to whom they are the most sexually attracted,and 50 percent of people of both sexes report to having extramaritalaffairs. Far more women want to sleep with rock stars and athletesthan to marry them. In short, the best fathers (in the biological sense)don’t always make the best dads (for child rearing). This may account forwhy, according to a recent European study, 10 percent of mothers reportedthat their children were being raised by men who falsely believedthe children were their own. Although today reproduction may not be themotive, it is difficult to separate out innate, evolutionarily derived preferencesfor mating partners from our societally and culturally inducedtastes in sexual partners.For musicologist David Huron of Ohio State, the key question forthe evolutionary basis is what advantage might be conferred on individualswho exhibit musical behaviors, versus those who do not. If musicis a nonadaptive pleasure-seeking behavior—the auditory cheesecakeargument—we would expect it not to last very long in evolutionary time.Huron writes, “Heroin users tend to neglect their health and are knownto have high mortality rates. Furthermore, heroin users make poor parents;they tend to neglect their offspring.” Neglecting one’s health andthe health of one’s children is a surefire way to reduce the probability ofone’s genes being passed on to future generations. First, if music wasnonadaptive, then music lovers should be at some evolutionary or survivaldisadvantage. Second, music shouldn’t have been around very long.Any activity that has low adaptive value is unlikely to be practiced for

250 This Is Your Brain on Musicvery long in the species’s history, or to occupy a significant portion of anindividual’s time and energy.All the available evidence is that music can’t be merely auditorycheesecake; it has been around a very long time in our species. Musicalinstruments are among the oldest human-made artifacts we have found.The Slovenian bone flute, dated at fifty thousand years ago, which wasmade from the femur of a now-extinct European bear, is a prime example.Music predates agriculture in the history of our species. We can say,conservatively, that there is no tangible evidence that language precededmusic. In fact, the physical evidence suggests the contrary. Music is nodoubt older than the fifty-thousand-year-old bone flute, because fluteswere unlikely the first instruments. Various percussion instruments, includingdrums, shakers, and rattles were likely to have been in use forthousands of years before flutes—we see this in contemporary huntergatherersocieties, and from the record of European invaders reportingon what they found in native American cultures. The archaeologicalrecord shows an uninterrupted record of music making everywhere wefind humans, and in every era. And, of course, singing most probably predatedflutes as well.To restate the summary principle of evolutionary biology, “Geneticmutations that enhance one’s likelihood to live long enough to reproducebecome adaptations.” The best estimates are that it takes a minimumof fifty thousand years for an adaptation to show up in the humangenome. This is called evolutionary lag—the time lag between when anadaptation first appears in a small proportion of individuals and when itbecomes widely distributed in the population. When behavioral geneticistsand evolutionary psychologists look for an evolutionary explanationfor our behaviors or appearance, they consider what evolutionaryproblem was being addressed by the adaptation in question. But due toevolutionary lag, the adaptation in question would have been a responseto conditions as they were at least fifty thousand years ago, not as theyare today. Our hunter-gatherer ancestors had a very different lifestylethan anyone who is reading this book, with different priorities and pressures.Many of the problems we face today—cancers, heart disease,

The Music Instinct 251maybe even the high divorce rate—have come to torment us because ourbodies and our brains were designed to handle life the way it was for usfifty thousand years ago. Fifty thousand years from now in the year 52,006(give or take a few millennia), our species may finally have evolved tohandle life the way it is now, with overcrowded cities, air and water pollution,video games, polyester, glazed doughnuts, and a gross imbalancein the distribution of resources worldwide. We may evolve mental mechanismsthat allow us to live in close quarters without feeling a loss of privacy,and physiological mechanisms to process carbon monoxide,radioactive waste, and refined sugar, and we may learn to use resourcesthat today are unusable.When we ask about the evolutionary basis for music, it does no goodto think about Britney or Bach. We have to think what music was likearound fifty thousand years ago. The instruments recovered from archeologicalsites can help us understand what our ancestors used to makemusic, and what kinds of melodies they listened to. Cave paintings, paintingson stoneware, and other pictorial artifacts can tell us somethingabout the role that music played in daily life. We can also study contemporarysocieties that have been cut off from civilization as we know it,groups of people who are living hunter-gatherer lifestyles that have remainedunchanged for thousands of years. One striking find is that inevery society of which we’re aware, music and dance are inseparable.The arguments against music as an adaptation consider music only asdisembodied sound, and moreover, as performed by an expert class foran audience. But it is only in the last five hundred years that music hasbecome a spectator activity—the thought of a musical concert in whicha class of “experts” performed for an appreciative audience was virtuallyunknown throughout our history as a species. And it has only been in thelast hundred years or so that the ties between musical sound and humanmovement have been minimized. The embodied nature of music, the indivisibilityof movement and sound, the anthropologist John Blackingwrites, characterizes music across cultures and across times. Most of uswould be shocked if audience members at a symphonic concert got outof their chairs and clapped their hands, whooped, hollered, and danced

252 This Is Your Brain on Musicas is de rigueur at a James Brown concert. But the reaction to JamesBrown is certainly closer to our true nature. The polite listening response,in which music has become an entirely cerebral experience(even music’s emotions are meant, in the classical tradition, to be felt internallyand not to cause a physical outburst) is counter to our evolutionaryhistory. Children often show the reaction that is true to ournature: Even at classical music concerts they sway and shout and generallyparticipate when they feel like it. We have to train them to behave“civilized.”When a behavior or trait is widely distributed across members of aspecies, we take it to be encoded in the genome (regardless of whetherit was an adaptation or a spandrel). Blacking argues that the universaldistribution of music-making ability in African societies suggests that“musical ability [is] a general characteristic of the human species, ratherthan a rare talent.” More important, Cross writes that “musical abilitycannot be defined solely in terms of productive competence”; virtuallyevery member of our own society is capable of listening to and hence ofunderstanding music.Apart from these facts about music’s ubiquity, history, and anatomy, itis important to understand how and why music was selected. Darwinproposed the sexual-selection hypothesis, which has been advancedmore recently by Miller and others. Additional possibilities have beenargued as well. One is social bonding and cohesion. Collective musicmaking may encourage social cohesions—humans are social animals,and music may have historically served to promote feelings of group togethernessand synchrony, and may have been an exercise for other socialacts such as turn-taking behaviors. Singing around the ancientcampfire might have been a way to stay awake, to ward off predators,and to develop social coordination and social cooperation within thegroup. Humans need social linkages to make society work, and music isone of them.An intriguing line of evidence for the social-bonding basis of musiccomes from my work with Ursula Bellugi on individuals with mental dis-

The Music Instinct 253orders such as Williams syndrome (WS) and autism spectrum disorders(ASD). As we saw in Chapter 6, WS is genetic in origin, and causes abnormalneuronal and cognitive development, resulting in intellectual impairment.People with WS, in spite of their overall mental impairment,are particularly good at music, and they’re particularly social.A contrast is people with ASD, many of whom also suffer from intellectualimpairment. It remains a controversial issue whether ASD has agenetic basis or not. A marker of ASD is the inability to empathize withothers, to understand emotions or emotional communication, particularlyemotions in others. People with ASD can certainly become angryand upset, they are not robots. But their ability to “read” the emotions ofothers is significantly impaired, and this typically extends to their utterinability to appreciate the aesthetic qualities of art and music. Althoughsome people with ASD play music, and some of them have reached ahigh level of technical proficiency, they do not report being emotionallymoved by music. Rather, the preliminary and largely anecdotal evidenceis that they are attracted to the structure of music. Temple Grandin, aprofessor who is autistic, has written that she finds music “pretty” butthat in general, she just “doesn’t get it” or understand why people reactto it the way that they do.With WS and ASD, we have two complementary syndromes. On theone hand we have a population who are highly social, gregarious, andhighly musical. On the other, we have a population who are highly antisocialand not very musical. The putative link between music and socialbonding is strengthened by complementary cases such as these, whatneuroscientists call a double dissociation. The argument is that theremay be a cluster of genes that influences both outgoingness and musicality.If this were true, we would expect to find that deviations in oneability co-occur with deviations in the other, as we do in WS and ASD.The brains of people with WS and ASD also, as we might expect, revealcomplementary impairments. Allan Reiss has shown that the neocerebellum,the newest part of the cerebellum, is larger than normal inWS, and smaller than normal in ASD. Because we already know the im-

254 This Is Your Brain on Musicportant role played by the cerebellum in music cognition, this is not surprising.Some as yet unidentified genetic abnormality appears to cause,either directly or indirectly, the neural dismorphology in WS, and we presumein ASD as well. This, in turn, leads to abnormal development ofmusical behaviors that in one case are enhanced and the other are diminished.Because of the complex and interactive nature of genes, it is certainthat there are other genetic correlates to sociability and musicality thatgo beyond the cerebellum. The geneticist Julie Korenberg has speculatedthat there exists a cluster of genes that are related to outgoingnessversus inhibitedness, and that people with WS lack some of the normalinhibition genes that the rest of us have, causing their musical behaviorsto be more uninhibited; for over a decade anecdotal reports, on CBSNews’s 60 Minutes, in a movie narrated by Oliver Sacks on Williams, andin a host of newspaper articles, have claimed that people with WS aremore fully engaged with—immersed in—music than most people. Myown laboratory has provided neural evidence for this point. We scannedthe brains of individuals with WS while they listened to music, and foundthey were using a vastly larger set of neural structures than everyoneelse does. Activation in their amygdala and cerebellum—the emotionalcenters of the brain—was significantly stronger than in “normals.” Everywherewe looked, we found stronger neural activation, and more widespreadneural activation. Their brains were humming.A third argument in favor of music’s primacy in human (and protohuman)evolution is that music evolved because it promoted cognitivedevelopment. Music may be the activity that prepared our pre-humanancestors for speech communication and for the very cognitive, representationalflexibility necessary to become humans. Singing and instrumentalactivities might have helped our species to refine motor skills,paving the way for the development of the exquisitely fine muscle controlrequired for vocal or signed speech. Because music is a complex activity,Trehub suggests that it may help prepare the developing infant forits mental life ahead. It shares many of the features of speech and it may

The Music Instinct 255form a way of “practicing” speech perception in a separate context. Nohuman has ever learned language by memorization. Babies don’t simplymemorize every word and sentence they’ve ever heard; rather, they learnrules and apply them in perceiving and generating new speech. Onepiece of evidence for this is empirical; the other is logical. The empiricalevidence comes from what linguists call overextension: Children justlearning the rules of language often apply them logically, but incorrectly.We see this most clearly in the case of irregular verb conjugations and irregularplurals in English. The developing brain is primed to make newneural connections and to prune away old ones that are not useful or accurate,and its mission is to instantiate rules insofar as possible. This iswhy we hear young children say, “He goed to the store,” instead of “Hewent to the store.” They are applying a logical rule: Most English verbs inpast tense take an -ed ending: play/played, talk /talked, touch /touched.Reasonable application of the rule leads to overextensions such as buyed,swimmed, and eated. In fact, intelligent children are more likely to makethese mistakes and to make them sooner during the course of their development,because they have a more sophisticated rule-generating system.Because many, many children make these speech errors and few adultsdo, this is evidence that children are not simply mimicking what theyhear, but rather, their brains are developing theories and rules aboutspeech that they then apply.The second piece of evidence that children don’t simply memorizelanguage is logical: All of us speak sentences that we’ve never heard before.We can form an infinite number of sentences to express thoughtsand ideas that we have neither expressed before nor heard expressedbefore—that is, language is generative. Children must learn the grammaticalrules for generating unique sentences to become competentspeakers of their language. A trivial example of how the number of sentencesin human language is infinite is that for any sentence you give me,I can always add “I don’t believe” to the beginning of it, and make a newsentence. “I like beer” becomes “I don’t believe I like beer.” “Mary saysshe likes beer” becomes “I don’t believe Mary says she likes beer.” Even“I don’t believe Mary says she likes beer” becomes “I don’t believe I don’t

256 This Is Your Brain on Musicbelieve Mary says she likes beer.” Although a sentence like this is awkward,it doesn’t alter the fact that it expresses a new idea. For languageto be generative, children must not be learning by rote. Music is also generative.For every musical phrase I hear, I can always add a note to thebeginning, end, or middle to generate a new musical phrase.Cosmides and Tooby argue that music’s function in the developingchild is to help prepare its mind for a number of complex cognitive andsocial activities, exercising the brain so that it will be ready for the demandsplaced on it by language and social interaction. The fact that musiclacks specific referents makes it a safe symbol system for expressingmood and feelings in a nonconfrontational manner. Music processinghelps infants to prepare for language; it may pave the way to linguisticprosody, even before the child’s developing brain is ready to processphonetics. Music for the developing brain is a form of play, an exercisethat invokes higher-level integrative processes that nurture exploratorycompetence, preparing the child to eventually explore generative languagedevelopment through babbling, and ultimately more complex linguisticand paralinguistic productions.Mother-infant interactions involving music almost always entail bothsinging and rhythmic movement, such as rocking or caressing. This appearsto be culturally universal. During the first six months or so of life,as I showed in Chapter 7, the infant brain is unable to clearly distinguishthe source of sensory inputs; vision, hearing, and touch meld into aunitary perceptual representation. The regions of the brain that willeventually become the auditory cortex, the sensory cortex, and the visualcortex are functionally undifferentiated, and inputs from the varioussensory receptors may connect to many different parts of the brain,pending pruning that will occur later in life. As Simon Baron-Cohen hasdescribed it, with all this sensory cross talk, the infant lives in a state ofcomplete psychedelic splendor (without the aid of drugs).Cross explicitly acknowledges that what music has become, today,with the influence of time and culture, is not necessarily what it was fiftythousand years ago, nor should we expect it to be. But considering ancientmusic’s character does account for why so many of us are literally

The Music Instinct 257moved by rhythm; by almost all accounts the music of our distant ancestorswas heavily rhythmic. Rhythm stirs our bodies. Tonality and melodystir our brains. The coming together of rhythm and melody bridges ourcerebellum (the motor control, primitive little brain) and our cerebralcortex (the most evolved, most human part of our brain). This is howRavel’s Bolero, Charlie Parker’s “Koko,” or the Rolling Stones’ “HonkyTonk Women” inspire us and move us, both metaphorically and physically,exquisite unions of time and melodic space. It is why rock, metal,and hip-hop music are the most popular musical genres in the world, andhave been for the past twenty years. Mitch Miller, the head talent scoutfor Columbia Records, famously said in the early sixties that rock-androllmusic was a fad that would soon die. Now, in 2006, there is no signof it slowing down. Classical music as most of us think of it—say, from1575 to 1950, from Monteverdi to Bach to Stravinsky, Rachmaninoff, andso on—is no longer being written. Contemporary composers in musicconservatories are not creating this sort of music as a rule, but rather,they are writing what many refer to as twentieth-century (now twentyfirst-century)art music. And so we have Philip Glass and John Cage andmore recent, lesser-known composers whose music is rarely performedby our symphony orchestras. When Copeland and Bernstein were composing,orchestras played their works and the public enjoyed them. Thisseems to be less and less the case in the past forty years. Contemporary“classical” music is practiced mostly in universities; it is listened to by almostno one; it deconstructs harmony, melody, and rhythm, renderingthem all but unrecognizable; it is a purely intellectual exercise, and savefor the rare avant-garde ballet company, no one dances to it either.A fourth argument for music as an adaptation comes from otherspecies. If we can show that other species use music for similar purposes,this presents a strong evolutionary argument. It is especiallyimportant, however, not to anthropomorphize animal behaviors, interpretingthem only from our own cultural perspective. What sounds to uslike music or a song may be serving, in animals, a very different functionfor them than it does for us. When we see a dog rolling around in freshsummer grass, with that very doglike grin on his face, we think, “Spike

258 This Is Your Brain on Musicmust be really happy.” We are interpreting the rolling-on-the-grass behaviorin terms of what we know about our own species, without stoppingto consider that it might mean something different to Spike and tohis species. Human children roll around in the grass, do somersaults andcartwheels, when they are happy. Male dogs roll around in the grasswhen they find a particularly pungent smell there, preferably from a recentlydead animal, and they cover their fur with it to make other dogsthink that they are skilled hunters. Similarly, birdsong that sounds joyfulto us is not necessarily intended that way by the bird-singer, or interpretedthat way by the bird-listener.Yet of all the calls of other species, birdsong holds a special positionof awe and intrigue. Who among us hasn’t sat and listened to a songbirdon a spring morning and found the beauty, the melody, the structure ofit enticing? Aristotle and Mozart were among those who did; they consideredthe songs of a bird to be every bit as musical as the compositionsof humans. But why do we write and perform music? Are our motivationsany different from those of the animals?Birds, whales, gibbons, frogs, and other species use vocalizations fora variety of purposes. Chimpanzees and prairie dogs have alert calls tocaution their fellows about an approaching predator, and the calls arespecific to the predator. Chimps use one vocalization to signal an approachingeagle (alerting their primate pals to hide underneath something)and another to broadcast the incursion of a snake (alerting theirfriends to climb a tree). Male birds use their vocalizations to establishterritory; robins and crows reserve a particular call to warn of predatorssuch as dogs and cats.Other animal vocalizations are more clearly related to courtship. Insongbirds, it is generally the male of the species that sings, and for somespecies, the larger the repertoire, the more likely he is to attract a mate.Yes, for a female songbird, size matters; it indicates male-bird intellectand, by extension, a source of potentially good bird genes. This wasshown in a study that played different songs over loudspeakers to femalebirds. The birds ovulated more quickly in the presence of a large birdsongrepertoire than in the presence of a small one. Some male songbirds will

The Music Instinct 259sing their courtship song until they drop dead from exhaustion. Linguistspoint to the generative nature of human music, the ability we have to createnew songs out of components, in an almost limitless fashion. This isnot a uniquely human trait either. Several bird species generate theirsongs from a repertoire of basic sounds, creating new melodies and variationson them, and the male who sings the most elaborate songs is typicallythe one who is most successful at mating. Music’s function in sexualselection thus has an analogue in other species.Music’s evolutionary origin is established because it is present acrossall humans (meeting the biologists’ criterion of being widespread in aspecies); it has been around a long time (refuting the notion that it ismerely audio cheesecake); it involves specialized brain structures, includingdedicated memory systems that can remain functional when othermemory systems fail (when a physical brain system develops across allhumans, we assume that it has an evolutionary basis); and it is analogousto music making in other species. Rhythmic sequences optimally excite recurrentneural networks in mammalian brains, including feedback loopsamong the motor cortex, the cerebellum, and the frontal regions. Tonalsystems, pitch transitions, and chords scaffold on certain properties of theauditory system that were themselves products of the physical world, ofthe inherent nature of vibrating objects. Our auditory system develops inways that play on the relation between scales and the overtone series. Musicalnovelty attracts attention and overcomes boredom, increasing memorability.Darwin’s theory of natural selection was revolutionized by the discoveryof the gene, specifically Watson and Crick’s discovery of the structure ofDNA. Perhaps we are witnessing another revolution in the aspect of evolutionthat depends on social behavior, on culture.Undoubtedly one of the most cited discoveries in neuroscience in thepast twenty years was of mirror neurons in the primate brain. GiacomoRizzolatti, Leonardo Fogassi, and Vittorio Gallese were studying the thebrain mechanisms responsible for movements such as reaching andgrasping in monkeys. They read the output from a single neuron in the

260 This Is Your Brain on Musicmonkey’s brain as it reached for pieces of food. At one point, Fogassireached for a banana, and the monkey’s neuron—one that had alreadybeen associated with movement—started to fire. “How could this happen,when the monkey did not move?” Rizzolatti recalls thinking. “Atfirst we thought it to be a flaw in our measuring or maybe equipment failure,but everything checked out OK and the reaction was repeated as werepeated the movement.” A decade of work since then has establishedthat primates, some birds, and humans have mirror neurons, neuronsthat fire both when performing an action and when observing someoneelse performing that action.The purpose of mirror neurons is presumably to train and prepare theorganism to make movements it has not made before. We’ve found mirrorneurons in Broca’s area, a part of the brain intimately involved inspeaking, and in learning to speak. Mirror neurons may explain an oldmystery of how it is that infants learn to imitate the faces that parentsmake at them. It may also explain why musical rhythm moves us, bothemotionally and physically. We don’t yet have solid evidence, but someneuroscientists speculate that our mirror neurons may be firing when wesee or hear musicians perform, as our brain tries to figure out how thosesounds are being created, in preparation for being able to mirror or echothem back as part of a signaling system. Many musicians can play back amusical part on their instruments after they’ve heard it only once. Mirrorneurons are likely involved in this ability.Genes are what pass protein recipes between individuals and acrossgenerations. Maybe mirror neurons, now in concert with sheet music, CDs,and iPods, will turn out to be the fundamental messengers of music acrossindividuals and generations, enabling that special kind of evolution—cultural evolution—through which develop our beliefs, obsessions, and allof art.For many solitary species, the ability to ritualize certain aspects of fitnessin a courtship display makes sense, because a potential mate pairmay only see each other for a few minutes. But in highly social societieslike ours, why would you need to demonstrate fitness through such ahighly stylized and symbolic means as dancing and singing? Humans live

The Music Instinct 261in social groups and have ample opportunities to observe one another ina variety of situations and over long periods of time. Why would musicbe needed to show fitness? Primates are highly social, living in groups,forming complex long-term relationships that involve social strategies.Hominid courtship was probably a long-term affair. Music, particularlymemorable music, would insinuate itself into the mind of a potentialmate, leading her to think about her suitor even when he was out on along hunt, and predisposing her toward him when he returned. The multiplereinforcing cues of a good song—rhythm, melody, contour—causemusic to stick in our heads. That is the reason that many ancient myths,epics, and even the Old Testament were set to music in preparation forbeing passed down by oral tradition across the generations. As a tool foractivation of specific thoughts, music is not as good as language. As atool for arousing feelings and emotions, music is better than language.The combination of the two—as best exemplified in a love song—is thebest courtship display of all.

APPENDIX AThis Is Your Brain on MusicMusic processing is distributed throughout the brain. The figures onthe following two pages show the brain’s major computationalcenters for music. The first illustration is a view of the brain from theside. The front of the brain is to the left. The second illustration showsthe inside of the brain from the same point of view as the first illustration.These figures are based on illustrations by Mark Tramo published inScience in 2001, but are redrawn and include newer information.

264 Appendix A

Appendix A 265

APPENDIX BChords and HarmonyWithin the key of C, the only legal chords are chords built off of thenotes of the C major scale. This causes some chords to be majorand some minor, because of the unequal spacing of tones in the scale. Tobuild the standard three-note chord—a triadic chord—we start on any ofthe tones of the C major scale, skip one, and then use the next, then skipone again and use the next one after that. The first chord in C major,then, comes from the notes C-E-G, and because the first interval formed,between C and E, is a major third, we call this chord a major chord, andin particular, a C major chord. The next one we build in a similar fashionis composed of D-F-A. Because the interval between D and F is a minorthird, this chord is called a D minor chord. Remember, major chords andminor chords have a very different sound. Even though most nonmusicianscan’t name a chord on hearing it, or label it as major or minor, ifthey hear a major and minor chord back to back they can tell the difference.And their brains can certainly tell the difference—a number ofstudies have shown that nonmusicians produce different physiologicalresponses to major versus minor chords, and major versus minor keys.In the major scale, considering the triadic chords constructed in thestandard way I’ve just described, three are major (on the first, fourth,

268 Appendix Band fifth scale degrees), three are minor (on the second, third, and sixthdegrees) and one is called a diminished chord (on the seventh scale degree)and is made up of two intervals of a minor third. The reason we saythat we’re in the key of C major, even though there are three minorchords in the key, is because the root chord—the chord that the musicpoints to, the one that feels like “home”—is C major.Generally, composers use chords to set a mood. The use of chordsand the way they are strung together is called harmony. Another, perhapsbetter-known use of the word harmony is to indicate when two ormore singers or instrumentalists are playing together and they’re notplaying the same notes, but conceptually this is the same idea. Somechord sequences are used more than others, and can become typical of aparticular genre. For example, the blues is defined by a particular chordsequence: a major chord on the first scale degree (written I major) followedby a major chord on the fourth scale degree (written IV major) followedby I major again, then V major, optionally to IV major, then backto I major. This is the standard blues progression, found in songs such as“Crossroads” (Robert Johnson, later covered by Cream), “Sweet Sixteen”by B. B. King, and “I Hear You Knockin’” (as recorded by SmileyLewis, Big Joe Turner, Screamin’ Jay Hawkins, and Dave Edmunds). Theblues progression—either verbatim or with some variations—is the basisfor rock and roll music, and is found in thousands of songs including“Tutti Frutti” by Little Richard, “Rock and Roll Music” by Chuck Berry,“Kansas City,” by Wilbert Marrison, “Rock and Roll” by Led Zeppelin, “JetAirliner” by the Steve Miller Band (which is surprisingly similar to “Crossroads”),and “Get Back” by the Beatles. Jazz artists such as Miles Davisand progressive rock artists like Steely Dan have written dozens of songsthat are inspired by this progression, with their own creative ways ofsubstituting more exotic chords for the standard three; but they are stillblues progressions, even when dressed up in fancier chords.Bebop music leaned heavily on a particular progression originallywritten by George Gershwin for the song “I’ve Got Rhythm.” In the keyof C, the basic chords would be:

Appendix B 269C major–A minor–D minor–G7–C major–A minor–D minor–G7C major–C7–F major–F minor–C major–G7–C majorC major–A minor–D minor–G7–C major–A minor–D minor–G7C major–C7–F–F minor–C major–G7–C majorThe 7 next to a note name indicates a tetrad—a four-note chord—thatis simply a major chord with a fourth note added on top; the top note isa minor third above the third note of the chord. The chord G7 is called either“G seven” or “G dominant seven.” Once we start using tetrads insteadof triads for chords, a great deal of rich tonal variation is possible.Rock and blues tend to use only the dominant seven, but there are twoother types of “seven” chords in common use, each conveying a differentemotional flavor. “Tin Man” and “Sister Golden Hair” by the group Americause the major seven chord to give them their characteristic sound (amajor triad with a major third on top, rather than the minor third of thechord we’re calling the dominant seven); “The Thrill Is Gone” by B. B.King uses minor seven chords throughout (a minor triad with a minorthird on top).The dominant seven chord occurs naturally—that is, diatonically—when it starts on the fifth degree of the major scale. In the key of C, then,G7 can be constructed by playing all white notes. The dominant sevencontains that formerly banned interval, the tritone, and it is the onlychord in a key that does. The tritone is harmonically the most unstableinterval we have in Western music, and so it carries with it a very strongperceptual urge to resolve. Because the dominant seven chord also containsthe most unstable scale tone—the seventh degree (B in the key ofC)—the chord “wants to” resolve back to C, the root. It is for this reasonthat the dominant seven chord built on the fifth degree of a major scale—the V7 chord, or G7 in the key of C—is the most typical, standard, andclichéd chord right before a composition ends on its root. In otherwords, the combination of G7 to C major (or their equivalents in otherkeys) gives us the single most unstable chord followed by the singlemost stable chord; it gives us the maximum feeling of tension and reso-

270 Appendix Blution that we can have. At the end of some of Beethoven’s symphonies,when the ending seems to go on and on and on, what the maestro is doingis giving us that two-chord progression over and over and over againuntil the piece finally resolves on the root.

BIBLIOGRAPHIC NOTESThe following are some of the many articles and books that I have consulted.The list is by no means complete, but represents additional sources that aremost relevant to the points made in this book. This book was written for the nonspecialistand not for my colleagues, and so I have tried to simplify topics withoutoversimplifying them. A more complete and detailed account of the brainand music can be found in these readings, and in the readings cited in them.Some of the works cited below are written for the specialist researcher. I haveused an asterisk (*) to indicate the more technical readings. Most of the markedentries are primary sources, and a few are graduate-level textbooks.IntroductionChurchland, P. M. 1986. Matter and Consciousness. Cambridge: MIT Press.In the passage on mankind’s curiosity having solved many of the greatestscientific mysteries, I have borrowed liberally from the introduction tothis excellent and inspiring work on the philosophy of mind.*Cosmides, L., and J. Tooby. 1989. Evolutionary psychology and the generationof culture, Part I. Case study: A computational theory of social exchange. Ethologyand Sociobiology 10: 51–97.An excellent introduction to the field of evolutionary psychology by twoleading scholars.*Deaner, R. O., and C. L. Nunn. 1999. How quickly do brains catch up with bodies?A comparative method for detecting evolutionary lag. Proceedings of BiologicalSciences 266 (1420):687–694.A recent scholarly article on the topic of evolutionary lag, the notion that

272 Bibliographic Notesour bodies and minds are at present equipped to deal with the worldand living conditions as they were fifty thousand years ago, due to theamount of time it takes for adaptations to become encoded in the humangenome.Levitin, D. J. 2001. Paul Simon: The Grammy Interview. Grammy September,42–46.Source of the Paul Simon quote about listening for sound.*Miller, G. F. 2000. Evolution of human music through sexual selection. In TheOrigins of Music, edited by N. L. Wallin, B. Merker, and S. Brown. Cambridge:MIT Press.Written by another leader in the field of evolutionary psychology, this articlediscusses many of the ideas discussed in Chapter 9, which are mentionedonly briefly in Chapter 1.Pareles, J., and P. Romanowski, eds. 1983. The Rolling Stone Encyclopedia ofRock & Roll. New York: Summit Books.Adam and the Ants get eight column inches plus a photo in this edition,U2—already well known with three albums and the hit “New Year’sDay”—get only four inches, and no photo.*Pribram, K. H. 1980. Mind, brain, and consciousness: the organization of competenceand conduct. In The Psychobiology of Consciousness, edited by J. M. D.Davidson, R.J. New York: Plenum.*———. 1982. Brain mechanism in music: prolegomena for a theory of the meaningof meaning. In Music, Mind, and Brain, edited by M. Clynes. New York:Plenum.Pribram taught his course from a collection of articles and notes that hehad compiled. These were two of the papers that we read.Sapolsky, R. M. 1998. Why Zebras Don’t Get Ulcers, 3rd ed. New York: Henry Holtand Company.An excellent book and a fun read on the science of stress, and the reasonsthat modern humans suffer from stress; the idea of “evolutionary lag” thatI introduce more fully in Chapter 9 is dealt with very well in this book.*Shepard, R. N. 1987. Toward a Universal Law of Generalization for psychologicalscience. Science 237 (4820):1317–1323.*———. 1992. The perceptual organization of colors: an adaptation to regularitiesof the terrestrial world? In The Adapted Mind: Evolutionary Psychologyand the Generation of Culture, edited by J. H. Barkow, L. Cosmides, andJ. Tooby. New York: Oxford University Press.*———. 1995. Mental universals: Toward a twenty-first century science of mind.In The Science of the Mind: 2001 and Beyond, edited by R. L. Solso and D. W.Massaro. New York: Oxford University Press.Three papers by Shepard in which he discusses the evolution of mind.

Bibliographic Notes 273Tooby, J., and L. Cosmides. 2002. Toward mapping the evolved functional organizationof mind and brain. In Foundations of Cognitive Psychology, edited byD. J. Levitin. Cambridge: MIT Press.Another paper by these two leaders in evolutionary psychology, perhapsthe more general of the two papers I’ve listed here.Chapter 1*Balzano, G. J. 1986. What are musical pitch and timbre? Music Perception 3(3):297–314.A scientific article on the issues involved in pitch and timbre research.Berkeley, G. 1734/2004. A Treatise Concerning the Principles of Human Knowledge.Whitefish, Mont.: Kessinger Publishing Company.The famous question—if a tree falls in the forest and no one is there tohear it, does it make a sound?—was first posed by the theologian andphilosopher George Berkeley, bishop of Cloyne, in this work.*Bharucha, J. J. 2002. Neural nets, temporal composites, and tonality. In Foundationsof Cognitive Psychology: Core Readings, edited by D. J. Levitin. Cambridge:MIT Press.Neural networks for chord recognition.*Boulanger, R. 2000. The C-Sound Book: Perspectives in Software Synthesis,Sound Design, Signal Processing, and Programming. Cambridge: MIT Press.An introduction to the most widely used software sound synthesis program/system.The best book I know of for people who want to learn toprogram computers to make music and create timbres of their ownchoosing.Burns, E. M. 1999. Intervals, scales, and tuning. In Psychology of Music, editedby D. Deutsch. San Diego: Academic Press.On the origin of scales, relationships among tones, nature of intervals andscales.*Chowning, J. 1973. The synthesis of complex audio spectra by means of frequencymodulation. Journal of the Audio Engineering Society 21:526–534.FM synthesis, as eventually manifested in the Yamaha DX synthesizers,was first described in this professional journal.Clayson, A. 2002. Edgard Varèse. London: Sanctuary Publishing, Ltd.Source of the quotation “Music is organized sound.”Dennett, Daniel C. 2005. Show me the science. The New York Times, August 28.Source of the quotation “Heat is not made of tiny hot things.”Doyle, P. 2005. Echo & Reverb: Fabricating Space in Popular Music Recording,1900–1960. Middletown, Conn.

274 Bibliographic NotesAn expansive, scholarly survey of the recording industry’s fascinationwith space and creating artificial ambiences.Dwyer, T. 1971. Composing with Tape Recorders: Musique Concrète. New York:Oxford University Press.For background information on the musique concrète of Schaeffer,Dhomon, Normandeau, and others.*Grey, J. M. 1975. An exploration of musical timbre using computer-based techniquesfor analysis, synthesis, and perceptual scaling. Ph.D. Thesis, Music,Center for Computer Research in Music and Acoustics, Stanford University,Stanford, Calif.The most influential paper on modern approaches to the study of timbre.*Janata, P. 1997. Electrophysiological studies of auditory contexts. DissertationAbstracts International: Section B: The Sciences and Engineering, Universityof Oregon.Contains the experiments showing that the inferior colliculus of the barnowl restores the missing fundamental.*Krumhansl, C. L. 1990. Cognitive Foundations of Musical Pitch. New York: OxfordUniversity Press.*———. 1991. Music psychology: Tonal structures in perception and memory.Annual Review of Psychology 42:277–303.*———. 2000. Rhythm and pitch in music cognition. Psychological Bulletin 126(1):159–179.*———. 2002. Music: A link between cognition and emotion. Current Directionsin Psychological Science 11 (2):45–50.Krumhansl is one of the leading scientists working in music perceptionand cognition; these articles, and the monograph, provide foundations ofthe field, and in particular, the notion of tonal hierarchies, the dimensionalityof pitch, and the mental representation of pitch.*Kubovy, M. 1981. Integral and separable dimensions and the theory of indispensableattributes. In Perceptual Organization, edited by M. Kubovy and J. Pomerantz.Hillsdale, N.J.: Erlbaum.Source for the notion of separable dimensions in music.Levitin, D. J. 2002. Memory for musical attributes. In Foundations of CognitivePsychology: Core Readings, edited by D. J. Levitin. Cambridge: MIT Press.Source for the listing of eight different perceptual attributes of a sound.*McAdams, S., J. W. Beauchamp, and S. Meneguzzi. 1999. Discrimination of musicalinstrument sounds resynthesized with simplified spectrotemporal parameters.Journal of the Acoustical Society of America 105 (2):882–897.

Bibliographic Notes 275McAdams, S., and E. Bigand. 1993. Introduction to auditory cognition. In Thinkingin Sound: The Cognitive Psychology of Audition, edited by S. McAdams andE. Bigand. Oxford: Clarendon Press.*McAdams, S., and J. Cunible. 1992. Perception of timbral analogies. PhilosophicalTransactions of the Royal Society of London, B 336:383–389.*McAdams, S., S. Winsberg, S. Donnadieu, and G. De Soete. 1995. Perceptualscaling of synthesized musical timbres: Common dimensions, specificities, andlatent subject classes. Psychological Research/Psychologische Forschung 58(3):177–192.McAdams is the leading researcher in the world studying timbre, andthese four papers provide an overview of what we currently know abouttimbre perception.Newton, I. 1730/1952. Opticks: or, A Treatise of the Reflections, Refractions, Inflections,and Colours of Light. New York: Dover.Source for Newton’s observation that light waves are not themselvescolored.*Oxenham, A. J., J. G. W. Bernstein, and H. Penagos. 2004. Correct tonotopic representationis necessary for complex pitch perception. Proceedings of the NationalAcademy of Sciences 101:1421–1425.On tonotopic representations of pitch in the auditory system.Palmer, S. E. 2000. Vision: From Photons to Phenomenology. Cambridge: MITPress.An excellent introduction to cognitive science and vision science, at theundergraduate level. Full disclosure: Palmer and I are collaborators, andI made some contributions to this book. Source for the different attributesof visual stimuli.Pierce, J. R. 1992. The Science of Musical Sound, revised ed. San Francisco:W. H. Freeman.Excellent source for the educated layperson who wants to understandthe physics of sound, overtones, scales, etc. Full disclosure: Pierce wasmy teacher and friend when he was alive.Rossing, T. D. 1990. The Science of Sound, 2nd ed. Reading, Mass.: Addison-Wesley Publishing.Another excellent source for the physics of sound, overtones, scales, andso on, appropriate for undergraduates.Schaeffer, Pierre. 1967. La musique concrète. Paris: Presses Universitairesde France.———. 1968. Traité des objets musicaux. Paris: Le Seuil.The principles of musique concrète are introduced in the first work, andSchaeffer’s masterpiece on the theory of sound in the second. Unfortunately,no English translation yet exists.

276 Bibliographic NotesSchmeling, P. 2005. Berklee Music Theory Book 1. Boston: Berklee Press.I learned music theory at Berklee College, and this is the first volume intheir set. Suitable for self-teaching, this covers all the basics.*Schroeder, M. R. 1962. Natural sounding artificial reverberation. Journal of theAudio Engineering Society 10 (3):219–233.The seminal article on the creation of artificial reverberation.Scorsese, Martin. 2005. No Direction Home. USA: Paramount.Source of the reports of Bob Dylan being booed at the Newport Folk Festival.Sethares, W. A. 1997. Tuning, Timbre, Spectrum, Scale. London: Springer.A rigorous introduction to the physics of music and musical sounds.*Shamma, S., and D. Klein. 2000. The case of the missing pitch templates: Howharmonic templates emerge in the early auditory system. Journal of the AcousticalSociety of America 107 (5):2631–2644.*Shamma, S. A. 2004. Topographic organization is essential for pitch perception.Proceedings of the National Academy of Sciences 101:1114–1115.On tonotopic representations of pitch in the auditory system.*Smith, J. O., III. 1992. Physical modeling using digital waveguides. ComputerMusic Journal 16 (4):74–91.The article that introduced wave guide synthesis.Surmani, A., K. F. Surmani, and M. Manus. 2004. Essentials of Music Theory: AComplete Self-Study Course for All Musicians. Van Nuys, Calif.: Alfred PublishingCompany.Another excellent self-teaching guide to music theory.Taylor, C. 1992. Exploring Music: The Science and Technology of Tones andTunes. Bristol: Institute of Physics Publishing.Another excellent college-level text on the physics of sound.Trehhub, S. E. 2003. Musical predispositions in infancy. In The Cognitive Neuroscienceof Music, edited by I. Perets and R. J. Zatorre. Oxford: Oxford UniversityPress.*Västfjäll, D., P. Larsson, and M. Kleiner. 2002. Emotional and auditory virtual environments:Affect-based judgments of music reproduced with virtual reverberationtimes. CyberPsychology & Behavior 5 (1):19–32.A recent scholarly article on the effect of reverberation on emotional response.Chapter 2*Bregman, A. S. 1990. Auditory Scene Analysis. Cambridge: MIT Press.The definitive work on general auditory grouping principles.

Bibliographic Notes 277Clarke, E. F. 1999. Rhythm and timing in music. In The Psychology of Music, editedby D. Deutsch. San Diego: Academic Press.An undergraduate-level article on the psychology of time perception inmusic, and the source for the Eric Clarke quote.*Ehrenfels, C. von. 1890/1988. On “Gestalt qualities.” In Foundations of GestaltTheory, edited by B. Smith. Munich: Philosophia Verlag.On the founding of Gestalt psychology and the Gestalt psychologists’ interestin melody.Elias, L. J., and D. M. Saucier. 2006. Neuropsychology: Clinical and ExperimentalFoundations. Boston: Pearson.Textbook for introducing fundamental concepts of neuroanatomy andthe functions of different brain regions.*Fishman, Y. I., D. H. Reser, J. C. Arezzo, and M. Steinschneider. 2000. Complextone processing in primary auditory cortex of the awake monkey. I. Neural ensemblecorrelates of roughness. Journal of the Acoustical Society of America108:235–246.The physiological basis of consonance and dissonance perception.Gilmore, Mikal. 2005. Lennon lives forever: Twenty-five years after his death, hismusic and message endure. Rolling Stone, December 15.Source of the John Lennon quote.Helmholtz, H. L. F. 1885/1954. On the Sensations of Tone, 2nd revised ed. NewYork: Dover.Unconscious inference.Lerdahl, Fred. 1983. A Generative Theory of Tonal Music. Cambridge: MIT Press.The most influential statement of auditory grouping principles in music.*Levitin, D. J., and P. R. Cook. 1996. Memory for musical tempo: Additional evidencethat auditory memory is absolute. Perception and Psychophysics58:927–935.This is the article mentioned in the text, in which Cook and I asked peopleto sing their favorite rock songs, and they reproduced the tempo withvery high accuracy.Luce, R. D. 1993. Sound and Hearing: A Conceptual Introduction. Hillsdale,N.J.: Erlbaum.Textbook on the ear and hearing, including physiology of the ear, loudness,pitch perception, etc.*Mesulam, M.-M. 1985. Principles of Behavioral Neurology. Philadelphia: F. A.Davis Company.Advanced, graduate textbook for introducing fundamental concepts ofneuroanatomy and the functions of different brain regions.Moore, B. C. J. 1982. An Introduction to the Psychology of Hearing, 2nd ed.London: Academic Press.

278 Bibliographic Notes———. 2003. An Introduction to the Psychology of Hearing, 5th ed. Amsterdam:Academic Press.Textbooks on the ear and hearing, including physiology of the ear, loudness,pitch perception, etc.Palmer, S. E. 2002. Organizing objects and scenes. In Foundations of CognitivePsychology: Core readings, edited by D. J. Levitin. Cambridge: MIT Press.On the Gestalt principles of visual grouping.Stevens, S. S., and F. Warshofsky. 1965. Sound and Hearing, edited by R. Dubos,H. Margenau, C. P. Snow. Life Science Library. New York: Time Incorporated.A good introduction to the principles of hearing and auditory perceptionfor the general reader.*Tramo, M. J., P. A. Cariani, B. Delgutte, and L. D. Braida. 2003. Neurobiology ofharmony perception. In The Cognitive Neuroscience of Music, edited by I.Peretz and R. J. Zatorre. New York: Oxford University Press.The physiological basis of consonance and dissonance perception.Yost, W. A. 1994. Fundamentals of Hearing: An Introduction, 3rd ed. San Diego:Academic Press, Inc.Textbook on hearing, pitch, and loudness perception.Zimbardo, P. G., and R. J. Gerrig. 2002. Perception. In Foundations of CognitivePsychology, edited by D. J. Levitin. Cambridge: MIT Press.The Gestalt principles of grouping.Chapter 3Bregman, A. S. 1990. Auditory Scene Analysis. Cambridge: MIT Press.Streaming by timbre and other auditory grouping principles. My analogyabout the eardrum as a pillowcase stretched over a bucket borrows liberallyfrom a different analogy Bregman proposes in this book.*Chomsky, N. 1957. Syntactic Structures. The Hague, Netherlands: Mouton.About the innateness of a language capacity in the human brain.Crick, F. H. C. 1995. The Astonishing Hypothesis: The Scientific Search for theSoul. New York: Touchstone/Simon & Schuster.The idea that all of human behavior can be explained by the activity of thebrain and neurons.Dennett, D. C. 1991. Consciousness Explained. Boston: Little, Brown and Company.On the illusions of conscious experience, and brains updating information.———. 2002. Can machines think? In Foundations of Cognitive Psychology:Core Readings, edited by D. J. Levitin. Cambridge: MIT Press.———. 2002. Where am I? In Foundations of Cognitive Psychology: Core Readings,edited by D. J. Levitin. Cambridge: MIT Press.

Bibliographic Notes 279These two articles address foundational issues of the brain as computerand the philosophical idea of functionalism; “Can Machines Think?” alsosummarizes the Turing test for intelligence, and its strengths and weaknesses.*Friston, K. J. 2005. Models of brain function in neuroimaging. Annual Review ofPsychology 56:57–87.A technical overview on research methods for the analysis of brain imagingdata by one of the inventors of SPM, a widely used statistical packagefor fMRI data.Gazzaniga, M. S., R. B. Ivry, and G. Mangun. 1998. Cognitive Neuroscience. NewYork: Norton.Functional divisions of the brain; basic divisions into lobes, major anatomicallandmarks; undergraduate text.Gertz, S. D., and R. Tadmor. 1996. Liebman’s Neuroanatomy Made Easy andUnderstandable, 5th ed. Gaithersburg, Md.: Aspen.An introduction to neuroanatomy and major brain regions.Gregory, R. L. 1986. Odd Perceptions. London: Routledge.On perception as inference.*Griffiths, T. D., S. Uppenkamp, I. Johnsrude, O. Josephs, and R. D. Patterson.2001. Encoding of the temporal regularity of sound in the human brainstem. NatureNeuroscience 4 (6):633–637.*Griffiths, T. D., and J. D. Warren. 2002. The planum temporale as a computationalhub. Trends in Neuroscience 25 (7):348–353.Recent work on sound processing in the brain from Griffiths, one of themost esteemed researchers of the current generation of brain scientistsstudying auditory processes.*Hickok, G., B. Buchsbaum, C. Humphries, and T. Muftuler. 2003. Auditorymotorinteraction revealed by fMRI: Speech, music, and working memory in areaSpt. Journal of Cognitive Neuroscience 15 (5):673–682.A primary source for music activation in a brain region at the posteriorSylvian fissure at the parietal-temporal boundary.*Janata, P., J. L. Birk, J. D. Van Horn, M. Leman, B. Tillmann, and J. J. Bharucha.2002. The cortical topography of tonal structures underlying Western music. Science298:2167–2170.*Janata, P., and S. T. Grafton. 2003. Swinging in the brain: Shared neural substratesfor behaviors related to sequencing and music. Nature Neuroscience 6(7):682–687.*Johnsrude, I. S., V. B. Penhune, and R. J. Zatorre. 2000. Functional specificity inthe right human auditory cortex for perceiving pitch direction. Brain Res CognBrain Res 123:155–163.

280 Bibliographic Notes*Knosche, T. R., C. Neuhaus, J. Haueisen, K. Alter, B. Maess, O. Witte, and A. D.Friederici. 2005. Perception of phrase structure in music. Human Brain Mapping24 (4):259–273.*Koelsch, S., E. Kasper, D. Sammler, K. Schulze, T. Gunter, and A. D. Friederici.2004. Music, language and meaning: brain signatures of semantic processing.Nature Neuroscience 7 (3):302–307.*Koelsch, S., E. Schröger, and T. C. Gunter. 2002. Music matters: Preattentive musicalityof the human brain. Psychophysiology 39 (1):38–48.*Kuriki, S., N. Isahai, T. Hasimoto, F. Takeuchi, and Y. Hirata. 2000. Music andlanguage: Brain activities in processing melody and words. Paper read at 12th InternationalConference on Biomagnetism.Primary sources on the neuroanatomy of music perception and cognition.Levitin, D. J. 1996. High-fidelity music: Imagine listening from inside the guitar.The New York Times, December 15.———. 1996. The modern art of studio recording. Audio, September, 46–52.On modern recording techniques and the illusions they create.———. 2002. Experimental design in psychological research. In Foundations ofCognitive Psychology: Core Readings, edited by D. J. Levitin. Cambridge: MIT Press.On experimental design and what is a “good” experiment.*Levitin, D. J., and V. Menon. 2003. Musical structure is processed in “language”areas of the brain: A possible role for Brodmann Area 47 in temporal coherence.NeuroImage 20 (4):2142–2152.The first research article using fMRI to show that temporal structure andtemporal coherence in music is processed in the same brain region thatdoes so for spoken and signed languages.*McClelland, J. L., D. E. Rumelhart, and G. E. Hinton. 2002. The appeal of paralleldistributed processing. In Foundations of Cognitive Psychology: Core Readings,edited by D. J. Levitin. Cambridge: MIT Press.The brain as a parallel processing machine.Palmer, S. 2002. Visual awareness. In Foundations of Cognitive Psychology:Core Readings, edited by D. J. Levitin. Cambridge: MIT Press.The philosophical foundations of modern cognitive science, dualism, andmaterialism.*Parsons, L. M. 2001. Exploring the functional neuroanatomy of music performance,perception, and comprehension. In I. Peretz and R. J. Zatorre, Eds., BiologicalFoundations of Music, Annals of the New York Academy of Sciences,Vol. 930, pp. 211–230.*Patel, A. D., and E. Balaban. 2004. Human auditory cortical dynamics duringperception of long acoustic sequences: Phase tracking of carrier frequency bythe auditory steady-state response. Cerebral Cortex 14 (1):35–46.

Bibliographic Notes 281*Patel, A. D. 2003. Language, music, syntax, and the brain. Nature Neuroscience6 (7):674–681.*Patel, A. D., and E. Balaban. 2000. Temporal patterns of human cortical activityreflect tone sequence structure. Nature 404:80–84.*Peretz, I. 2000. Music cognition in the brain of the majority: Autonomy and fractionationof the music recognition system. In The Handbook of Cognitive Neuropsychology,edited by B. Rapp. Hove, U.K.: Psychology Press.*Peretz, I. 2000. Music perception and recognition. In The Handbook of CognitiveNeuropsychology, edited by B. Rapp. Hove, U.K.: Psychology Press.*Peretz, I., and M. Coltheart. 2003. Modularity of music processing. NatureNeuroscience 6 (7):688–691.*Peretz, I., and L. Gagnon. 1999. Dissociation between recognition and emotionaljudgements for melodies. Neurocase 5:21–30.*Peretz, I., and R. J. Zatorre, eds. 2003. The Cognitive Neuroscience of Music.New York: Oxford.Primary sources on the neuroanatomy of music perception and cognition.Pinker, S. 1997. How The Mind Works. New York: W. W. Norton.Pinker claims here that music is an evolutionary accident.*Posner, M. I. 1980. Orienting of attention. Quarterly Journal of ExperimentalPsychology 32:3–25.The Posner Cueing Paradigm.Posner, M. I., and D. J. Levitin. 1997. Imaging the future. In The Science of theMind: The 21st Century. Cambridge: MIT Press.A more complete explanation of the bias that Posner and I have againstsimple “mental cartography” for its own sake.Ramachandran, V. S. 2004. A Brief Tour of Human Consciousness: From ImpostorPoodles to Purple Numbers. New York: Pi Press.Consciousness and our naive intuitions about it.*Rock, I. 1983. The Logic of Perception. Cambridge: MIT Press.Perception as a logical process and as constructive.*Schmahmann, J. D., ed. 1997. The Cerebellum and Cognition. San Diego: AcademicPress.On the cerebellum’s role in emotional regulation.Searle, J. R. 2002. Minds, brains, and programs. In Foundations of CognitivePsychology: Core Readings, edited by D. J. Levitin. Cambridge: MIT Press.The brain as a computer; this is one of the most discussed, argued, andcited articles in modern philosophy of mind.

282 Bibliographic Notes*Sergent, J. 1993. Mapping the musician brain. Human Brain Mapping 1:20–38.One of the first neuroimaging reports of music and the brain, still widelycited and referred to.Shepard, R. N. 1990. Mind Sights: Original Visual Illusions, Ambiguities, andOther Anomalies, with a Commentary on the Play of Mind in Perception andArt. New York: W. H. Freeman.Source of the “Turning the Tables” illusion.*Steinke, W. R., and L. L. Cuddy. 2001. Dissociations among functional subsystemsgoverning melody recognition after right hemisphere damage. CognitiveNeuroscience 18 (5):411–437.*Tillmann, B., P. Janata, and J. J. Bharucha. 2003. Activation of the inferiorfrontal cortex in musical priming. Cognitive Brain Research 16:145–161.Primary sources on the neuroanatomy of music perception and cognition.*Warren, R. M. 1970. Perceptual restoration of missing speech sounds. Science,January 23, 392–393.Source of the example of auditory “filling in” or perceptual completion.Weinberger, N. M. 2004. Music and the Brain. Scientific American (November2004):89–95.*Zatorre, R. J., and P. Belin. 2001. Spectral and temporal processing in human auditorycortex. Cerebral Cortex 11:946–953.*Zatorre, R. J., P. Belin, and V. B. Penhune. 2002. Structure and function of auditorycortex: Music and speech. Trends in Cognitive Sciences 6 (1):37–46.Primary sources on the neuroanatomy of music perception and cognition.Chapter 4*Bartlett, F. C. 1932. Remembering: A Study in Experimental and Social Psychology.London: Cambridge University Press.On schemas.*Bavelier, D., C. Brozinsky, A. Tomann, T. Mitchell, H. Neville, and G. Liu. 2001. Impactof early deafness and early exposure to sign language on the cerebral organizationfor motion processing. The Journal of Neuroscience 21 (22):8931–8942.*Bavelier, D., D. P. Corina, and H. J. Neville. 1998. Brain and language: A perspectivefrom sign language. Neuron 21:275–278.The neuroanatomy of sign language.*Bever, T. G., and Chiarell, R. J. 1974. Cerebral dominance in musicians and nonmusicians.Science 185 (4150):537–539.A seminal paper on hemispheric specialization for music.

Bibliographic Notes 283*Bharucha, J. J. 1987. Music cognition and perceptual facilitation—a connectionistframework. Music Perception 5 (1):1–30.*———. 1991. Pitch, harmony, and neural nets: A psychological perspective. InMusic and Connectionism, edited by P. M. Todd and D. G. Loy. Cambridge: MITPress.*Bharucha, J. J., and P. M. Todd. 1989. Modeling the perception of tonal structurewith neural nets. Computer Music Journal 13 (4):44–53.*Bharucha, J. J. 1992. Tonality and learnability. In Cognitive Bases of MusicalCommunication, edited by M. R. Jones and S. Holleran. Washington, D.C: AmericanPsychological Association.On musical schemas.*Binder, J., and C. J. Price. 2001. Functional neuroimaging of language. In Handbookof Functional Neuroimaging of Cognition, edited by A. Cabeza andA. Kingston.*Binder, J. R., E. Liebenthal, E. T. Possing, D. A. Medler, and B. D. Ward. 2004.Neural correlates of sensory and decision processes in auditory object identification.Nature Neuroscience 7 (3):295–301.*Bookheimer, S. Y. 2002. Functional MRI of language: New approaches to understandingthe cortical organization of semantic processing. Annual Review ofNeuroscience 25:151–188.The functional neuroanatomy of speech.Cook, P. R. 2005. The deceptive cadence as a parlor trick. Princeton, N.J., Montreal,Que., November 30.Personal communication from Perry Cook, who described the deceptivecadence this way in an e-mail to me.*Cowan, W. M., T. C. Südhof, and C. F. Stevens, eds. 2001. Synapses. Baltimore:Johns Hopkins University Press.In-depth information on synapses, the synaptic cleft, and synaptic transmission.*Dibben, N. 1999. The perception of structural stability in atonal music: the influenceof salience, stability, horizontal motion, pitch commonality, and dissonance.Music Perception 16 (3):265–24.On atonal music, such as that by Schönberg described in this chapter.*Franceries, X., B. Doyon, N. Chauveau, B. Rigaud, P. Celsis, and J.-P. Morucci.2003. Solution of Poisson’s equation in a volume conductor using resistor meshmodels: Application to event related potential imaging. Journal of AppliedPhysics 93 (6):3578–3588.The inverse Poisson problem of localization with EEG.

284 Bibliographic NotesFromkin, V., and R. Rodman. 1993. An Introduction to Language, 5th ed. FortWorth, Tex.: Harcourt Brace Jovanovich College Publishers.The basics of psycholinguistics, phonemes, word formation.*Gazzaniga, M. S. 2000. The New Cognitive Neurosciences, 2nd ed. Cambridge:MIT Press.Foundations of neuroscience.Gernsbacher, M. A., and M. P. Kaschak. 2003. Neuroimaging studies of languageproduction and comprehension. Annual Review of Psychology 54:91–114.A recent review of studies of the neuroanatomical basis for language.*Hickok, G., B. Buchsbaum, C. Humphries, and T. Muftuler. 2003. Auditory-motorinteraction revealed by fMRI: Speech, music, and working memory in area Spt.Journal of Cognitive Neuroscience 15 (5):673–682.*Hickok, G., and Poeppel, D. 2000. Towards a functional neuroanatomy ofspeech perception. Trends in Cognitive Sciences 4 (4):131–138.The neuroanatomical basis for speech and music.Holland, B. 1981. A man who sees what others hear. The New York Times, November19.An article about Arthur Lintgen, the man who can read record grooves.He can only read them for music that he knows, and only for classical musicpost-Beethoven.*Huettel, S. A., A. W. Song, and G. McCarthy. 2003. Functional Magnetic ResonanceImaging. Sunderland, Mass.: Sinauer Associates, Inc.On the theory behind fMRI.*Ivry, R. B., and L. C. Robertson. 1997. The Two Sides of Perception. Cambridge:MIT Press.On hemispheric specialization.*Johnsrude, I. S., V. B. Penhune, and R. J. Zatorre. 2000. Functional specificity inthe right human auditory cortex for perceiving pitch direction. Brain Res CognBrain Res 123:155–163.*Johnsrude, I. S., R. J. Zatorre, B. A. Milner, and A. C. Evans. 1997. Left-hemispherespecialization for the processing of acoustic transients. NeuroReport 8:1761–1765.The neuroanatomy of speech and music.*Kandel, E. R., J. H. Schwartz, and T. M. Jessell. 2000. Principles of Neural Science,4th ed. New York: McGraw-Hill.Foundations of neuroscience, cowritten by Nobel Laureate Eric Kandel.This is a widely used text in medical schools and graduate neuroscienceprograms.*Knosche, T. R., C. Neuhaus, J. Haueisen, K. Alter, B. Maess, O. Witte, and A. D.Friederici. 2005. Perception of phrase structure in music. Human Brain Mapping24 (4):259–273.

Bibliographic Notes 285*Koelsch, S., T. C. Gunter, D. Y. v. Cramon, S. Zysset, G. Lohmann, and A. D.Friederici. 2002. Bach speaks: A cortical “language-network” serves the processingof music. NeuroImage 17:956–966.*Koelsch, S., E. Kasper, D. Sammler, K. Schulze, T. Gunter, and A. D. Friederici.2004. Music, language, and meaning: Brain signatures of semantic processing.Nature Neuroscience 7 (3):302–307.*Koelsch, S., B. Maess, and A. D. Friederici. 2000. Musical syntax is processed inthe area of Broca: an MEG study. NeuroImage 11 (5):56.Articles on musical structure by Koelsch, Friederici, and their colleagues.Kosslyn, S. M., and O. Koenig. 1992. Wet Mind: The New Cognitive Neuroscience.New York: Free Press.A general audience’s introduction to cognitive neuroscience.*Krumhansl, C. L. 1990. Cognitive Foundations of Musical Pitch. New York: OxfordUniversity Press.On the dimensionality of pitch.*Lerdahl, F. 1989. Atonal prolongational structure. Contemporary Music Review3 (2).On atonal music, such as that of Schönberg.*Levitin, D. J., and V. Menon. 2003. Musical structure is processed in “language”areas of the brain: A possible role for Brodmann Area 47 in temporal coherence.NeuroImage 20 (4):2142–2152.*———. 2005. The neural locus of temporal structure and expectancies in music:Evidence from functional neuroimaging at 3 Tesla. Music Perception 22(3):563–575.The neuroanatomy of musical structure.*Maess, B., S. Koelsch, T. C. Gunter, and A. D. Friederici. 2001. Musical syntax isprocessed in Broca’s area: An MEG study. Nature Neuroscience 4 (5):540–545.The neuroanatomy of musical structure.*Marin, O. S. M. 1982. Neurological aspects of music perception and performance.In The Psychology of Music, edited by D. Deutsch. New York: Academic Press.Loss of musical function due to lesions.*Martin, R. C. 2003. Language processing: Functional organization and neuroanatomicalbasis. Annual Review of Psychology 54:55–89.The neuroanatomy of speech perception.McClelland, J. L., D. E. Rumelhart, and G. E. Hinton. 2002. The Appeal of ParallelDistributed Processing. In Foundations of Cognitive Psychology: Core Readings,edited by D. J. Levitin. Cambridge: MIT Press.On schemas.

286 Bibliographic NotesMeyer, L. B. 2001. Music and emotion: distinctions and uncertainties. In Musicand Emotion: Theory and Research, edited by P. N. Juslin and J. A. Sloboda. Oxfordand New York: Oxford University Press.Meyer, Leonard B. 1956. Emotion and Meaning in Music. Chicago: University ofChicago Press.———. 1994. Music, the Arts, and Ideas: Patterns and Predictions in Twentieth-Century Culture. Chicago: University of Chicago Press.On musical style, repetition, gap-fill, and expectations.*Milner, B. 1962. Laterality effects in audition. In Interhemispheric Effects andCerebral Dominance, edited by V. Mountcastle. Baltimore: Johns Hopkins Press.Laterality in hearing.*Narmour, E. 1992. The Analysis and Cognition of Melodic Complexity: TheImplication-Realization Model. Chicago: University of Chicago Press.*———. 1999. Hierarchical expectation and musical style. In The Psychology ofMusic, edited by D. Deutsch. San Diego: Academic Press.On musical style, repetition, gap-fill, and expectations.*Niedermeyer, E., and F. L. Da Silva. 2005. Electroencephalography: Basic Principles,Clinical Applications, and Related Fields, 5th ed. Philadephia: Lippincott,Williams & Wilkins.An introduction to EEG (advanced, technical, not for the faint of heart).*Panksepp, J., ed. 2002. Textbook of Biological Psychiatry. Hoboken, N.J.: Wiley.On SSRIs, seratonin, dopamine, and neurochemistry.*Patel, A. D. 2003. Language, music, syntax and the brain. Nature Neuroscience 6(7):674–681.The neuroanatomy of musical structure; this paper introduces the SSIRH.*Penhune, V. B., R. J. Zatorre, J. D. MacDonald, and A. C. Evans. 1996. Interhemisphericanatomical differences in human primary auditory cortex: Probabilisticmapping and volume measurement from magnetic resonance scans.Cerebral Cortex 6:661–672.*Peretz, I., R. Kolinsky, M. J. Tramo, R. Labrecque, C. Hublet, G. Demeurisse, andS. Belleville. 1994. Functional dissociations following bilateral lesions of auditorycortex. Brain 117:1283–1301.*Perry, D. W., R. J. Zatorre, M. Petrides, B. Alivisatos, E. Meyer, and A. C. Evans.1999. Localization of cerebral activity during simple singing. NeuroReport10:3979–3984.The neuroanatomy of music processing.*Petitto, L. A., R. J. Zatorre, K. Gauna, E. J. Nikelski, D. Dostie, and A. C. Evans.2000. Speech-like cerebral activity in profoundly deaf people processing signed

Bibliographic Notes 287languages: Implications for the neural basis of human language. Proceedings ofthe National Academy of Sciences 97 (25):13961–13966.The neuroanatomy of sign language.Posner, M. I. 1973. Cognition: An Introduction. Edited by J. L. E. Bourne andL. Berkowitz, 1st ed. Basic Psychological Concepts Series. Glenview, Ill.: Scott,Foresman and Company.———. 1986. Chronometric Explorations of Mind: The Third Paul M. Fitts Lectures,Delivered at the University of Michigan, September 1976. New York: OxfordUniversity Press.On mental codes.Posner, M. I., and M. E. Raichle. 1994. Images of Mind. New York: ScientificAmerican Library.A general-reader introduction to neuroimaging.Rosen, C. 1975. Arnold Schoenberg. Chicago: University of Chicago Press.On the composer, atonal and twelve-tone music.*Russell, G. S., K. J. Eriksen, P. Poolman, P. Luu, and D. Tucker. 2005. Geodesicphotogrammetry for localizing sensor positions in dense-array EEG. ClinicalNeuropsychology 116:1130–1140.The inverse Poisson problem in EEG localization.Samson, S., and R. J. Zatorre. 1991. Recognition memory for text and melody ofsongs after unilateral temporal lobe lesion: Evidence for dual encoding. Journalof Experimental Psychology: Learning, Memory, and Cognition 17 (4):793–804.———. 1994. Contribution of the right temporal lobe to musical timbre discrimination.Neuropsychologia 32:231–240.Neuroanatomy of music and speech perception.Schank, R. C., and R. P. Abelson. 1977. Scripts, plans, goals, and understanding.Hillsdale, N.J.: Lawrence Erlbaum Associates.Seminal work on schemas.*Shepard, R. N. 1964. Circularity in judgments of relative pitch. Journal of TheAcoustical Society of America 36 (12):2346–2353.*———. 1982. Geometrical approximations to the structure of musical pitch.Psychological Review 89 (4):305–333.*———. 1982. Structural representations of musical pitch. In Psychology of Music,edited by D. Deutsch. San Diego: Academic Press.The dimensionality of pitch.Squire, L. R., F. E. Bloom, S. K. McConnell, J. L. Roberts, N. C. Spitzer, and M. J. Zigmond,eds. 2003. Fundamental Neuroscience, 2nd ed. San Diego: Academic Press.Basic neuroscience text.

288 Bibliographic Notes*Temple, E., R. A. Poldrack, A. Protopapas, S. S. Nagarajan, T. Salz, P. Tallal,M. M. Merzenich, and J. D. E. Gabrieli. 2000. Disruption of the neural response torapid acoustic stimuli in dyslexia: Evidence from functional MRI. Proceedings ofthe National Academy of Sciences 97 (25):13907–13912.Functional neuroanatomy of speech.*Tramo, M. J., J. J. Bharucha, and F. E. Musiek. 1990. Music perception and cognitionfollowing bilateral lesions of auditory cortex. Journal of Cognitive Neuroscience2:195–212.*Zatorre, R. J. 1985. Discrimination and recognition of tonal melodies after unilateralcerebral excisions. Neuropsychologia 23 (1):31–41.*———. 1998. Functional specialization of human auditory cortex for musicalprocessing. Brain 121 (Part 10):1817–1818.*Zatorre, R. J., P. Belin, and V. B. Penhune. 2002. Structure and function of auditorycortex: Music and speech. Trends in Cognitive Sciences 6 (1):37–46.*Zatorre, R. J., A. C. Evans, E. Meyer, and A. Gjedde. 1992. Lateralization of phoneticand pitch discrimination in speech processing. Science 256 (5058):846–849.*Zatorre, R. J., and S. Samson. 1991. Role of the right temporal neocortex in retentionof pitch in auditory short-term memory. Brain (114):2403–2417.Studies of the neuroanatomy of speech and music, and of the effect oflesions.Chapter 5Bjork, E. L., and R. A. Bjork, eds. 1996. Memory, Handbook of Perception andCognition, 2nd ed. San Diego: Academic Press.General text on memory for the researcher.Cook, P. R., ed. 1999. Music, Cognition, and Computerized Sound: An Introductionto Psychoacoustics. Cambridge: MIT Press.This book consists of the lectures that I attended as an undergraduate inthe course I mention, taught by Pierce, Chowning, Mathews, Shepard,and others.*Dannenberg, R. B., B. Thom, and D. Watson. 1997. A machine learning approachto musical style recognition. Paper read at International Computer Music Conference,September. Thessoloniki, Greece.A source article about music fingerprinting.Dowling, W. J., and D. L. Harwood. 1986. Music Cognition. San Diego: AcademicPress.On the recognition of melodies in spite of transformations.Gazzaniga, M. S., R. B. Ivry, and G. R. Mangun. 1998. Cognitive Neuroscience:The Biology of the Mind. New York: W. W. Norton.Contains a summary of Gazzaniga’s split-brain studies.

Bibliographic Notes 289*Goldinger, S. D. 1996. Words and voices: Episodic traces in spoken word identificationand recognition memory. Journal of Experimental Psychology: Learning,Memory, and Cognition 22 (5):1166–1183.*———. 1998. Echoes of echoes? An episodic theory of lexical access. PsychologicalReview 105 (2):251–279.Source articles on multiple-trace memory theory.Guenther, R. K. 2002. Memory. In Foundations of Cognitive Psychology: CoreReadings, edited by D. J. Levitin. Cambridge: MIT Press.An overview of the record-keeping vs. constructivist theories of memory.*Haitsma, J., and T. Kalker. 2003. A highly robust audio fingerprinting systemwith an efficient search strategy. Journal of New Music Research 32 (2):211–221.Another source article on audio fingerprinting.*Halpern, A. R. 1988. Mental scanning in auditory imagery for songs. Journal ofExperimental Psychology: Learning, Memory, and Cognition 143:434–443.Source for the discussion in this chapter about the ability to scan musicin our heads.*———. 1989. Memory for the absolute pitch of familiar songs. Memory andCognition 17 (5):572–581.This article was the inspiration for my 1994 study.*Heider, E. R. 1972. Universals in color naming and memory. Journal of ExperimentalPsychology 93 (1):10–20.Under Eleanor Rosch’s married name, a foundational work on categorization.*Hintzman, D. H. 1986. “Schema abstraction” in a multiple-trace memory model.Psychological Review 93 (4):411–428.Hintzman’s MINERVA model is discussed in the context of multiple-tracememory models.*Hintzman, D. L., R. A. Block, and N. R. Inskeep. 1972. Memory for mode of input.Journal of Verbal Learning and Verbal Behavior 11:741–749.Source for the study of fonts that I discuss.*Ishai, A., L. G. Ungerleider, and J. V. Haxby. 2000. Distributed neural systems forthe generation of visual images. Neuron 28:979–990.Source for the work on categorical separation in the brain.*Janata, P. 1997. Electrophysiological studies of auditory contexts. DissertationAbstracts International: Section B: The Sciences and Engineering, University ofOregon.This contains the report of imagining a piece of music bearing a nearlyidentical EEG signature to actually hearing a piece of music.

290 Bibliographic Notes*Levitin, D. J. 1994. Absolute memory for musical pitch: Evidence from the productionof learned melodies. Perception and Psychophysics 56 (4):414–423.This is the source article reporting my study of people singing their favoriterock and pop songs at or near the correct key.*———. 1999. Absolute pitch: Self-reference and human memory. InternationalJournal of Computing Anticipatory Systems.An overview of absolute-pitch research.*———. 1999. Memory for musical attributes. In Music, Cognition and ComputerizedSound: An Introduction to Psychoacoustics, edited by P. R. Cook.Cambridge: MIT Press.Description of my study with tuning forks and memory for pitch.———. 2001. Paul Simon: The Grammy interview. Grammy, September, 42–46.Source of the Paul Simon comment about listening for timbres.*Levitin, D. J., and P. R. Cook. 1996. Memory for musical tempo: Additional evidencethat auditory memory is absolute. Perception and Psychophysics 58:927–935.Source of my study on memory for the tempo of a song.*Levitin, D. J., and S. E. Rogers. 2005. Pitch perception: Coding, categories, andcontroversies. Trends in Cognitive Sciences 9 (1):26–33.Review of absolute-pitch research.*Levitin, D. J., and R. J. Zatorre. 2003. On the nature of early training and absolutepitch: A reply to Brown, Sachs, Cammuso and Foldstein. Music Perception21 (1):105–110.A technical note about problems with absolute-pitch research.Loftus, E. 1979/1996. Eyewitness Testimony. Cambridge: Harvard University Press.Source of the experiments on memory distortions.Luria, A. R. 1968. The Mind of a Mnemonist. New York: Basic Books.Source of the story about the patient with hypermnesia.McClelland, J. L., D. E. Rumelhart, and G. E. Hinton. 2002. The appeal of paralleldistributed processing. In Foundations of Cognitive Psychology: Core Readings,edited by D. J. Levitin. Cambridge: MIT Press.Seminal article on parallel distributed processing (PDP) models, otherwiseknown as “neural networks,” computer simulations of brain activity.*McNab, R. J., L. A. Smith, I. H. Witten, C. L. Henderson, and S. J. Cunningham.1996. Towards the digital music library: tune retrieval from acoustic input. Proceedingsof the First ACM International Conference on Digital Libraries:11–18.Music fingerprinting overview.*Parkin, A. J. 1993. Memory: Phenomena, Experiment and Theory. Oxford, UK:Blackwell.Textbook on memory.

Bibliographic Notes 291*Peretz, I., and R. J. Zatorre. 2005. Brain organization for music processing. AnnualReview of Psychology 56:89–114.Review of neuroanatomical foundations of music perception.*Pope, S. T., F. Holm, and A. Kouznetsov. 2004. Feature extraction and databasedesign for music software. Paper read at International Computer Music Conferencein Miami.On music fingerprinting.*Posner, M. I., and S. W. Keele. 1968. On the genesis of abstract ideas. Journal ofExperimental Psychology 77:353–363.*———. 1970. Retention of abstract ideas. Journal of Experimental Psychology83:304–308.Source for the experiments described that showed prototypes might bestored in memory.*Rosch, E. 1977. Human categorization. In Advances in Crosscultural Psychology,edited by N. Warren. London: Academic Press.*———. 1978. Principles of categorization. In Cognition and Categorization,edited by E. Rosch and B. B. Lloyd. Hillsdale, N.J.: Erlbaum.*Rosch, E., and C. B. Mervis. 1975. Family resemblances: Studies in the internalstructure of categories. Cognitive Psychology 7:573–605.*Rosch, E., C. B. Mervis, W. D. Gray, D. M. Johnson, and P. Boyes-Braem. 1976.Basic objects in natural categories. Cognitive Psychology 8:382–439.Source articles on Rosch’s prototype theory.*Schellenberg, E. G., P. Iverson, and M. C. McKinnon. 1999. Name that tune: Identifyingfamiliar recordings from brief excerpts. Psychonomic Bulletin & Review6 (4):641–646.Source for the study described of people naming songs based on timbralcues.Smith, E. E., and D. L. Medin. 1981. Categories and concepts. Cambridge: HarvardUniversity Press.Smith, E., and D. L. Medin. 2002. The exemplar view. In Foundations of CognitivePsychology: Core Readings, edited by D. J. Levitin. Cambridge: MIT Press.On the exemplar view, as an alternative to Rosch’s prototype theory.*Squire, L. R. 1987. Memory and Brain. New York: Oxford University Press.Textbook on memory.*Takeuchi, A. H., and S. H. Hulse. 1993. Absolute pitch. Psychological Bulletin113 (2):345–361.*Ward, W. D. 1999. Absolute Pitch. In The Psychology of Music, edited byD. Deutsch. San Diego: Academic Press.Overviews of absolute pitch.

292 Bibliographic Notes*White, B. W. 1960. Recognition of distorted melodies. American Journal ofPsychology 73:100–107.Source for the experiments on how music can be recognized under transpositionand other transformations.Wittgenstein, L. 1953. Philosophical Investigations. New York: Macmillan.Source for Wittgenstein’s writings about “What is a game?” and family resemblance.Chapter 6*Desain, P., and H. Honing. 1999. Computational models of beat induction: Therule-based approach. Journal of New Music Research 28 (1):29–42.This paper discusses some of the algorithms the authors used in the foottappingshow I wrote about.*Aitkin, L. M., and J. Boyd. 1978. Acoustic input to lateral pontine nuclei. HearingResearch 1 (1):67–77.Physiology of the auditory pathway, low-level.*Barnes, R., and M. R. Jones. 2000. Expectancy, attention, and time. CognitivePsychology 41 (3):254–311.An example of Mari Reiss Jones’s work on time and timing in music.Crick, F. 1988. What Mad Pursuit: A Personal View of Scientific Discovery. NewYork: Basic Books.Source for the quote about Crick’s early years as a scientist.Crick, F. H. C. 1995. The Astonishing Hypothesis: The Scientific Search for theSoul. New York: Touchstone/Simon & Schuster.Source for Crick’s discussion of reductionism.*Friston, K. J. 1994. Functional and effective connectivity in neuroimaging: asynthesis. Human Brain Mapping 2:56–68.The article on functional connectivity that helped Menon to create theanalyses we needed for our paper on musical emotion and the nucleus accumbens.*Gallistel, C. R. 1989. The Organization of Learning. Cambridge: MIT Press.An example of Randy Gallistel’s work.*Goldstein, A. 1980. Thrills in response to music and other stimuli. PhysiologicalPsychology 8 (1):126–129.The study that showed that naloxone can block musical emotion.*Grabow, J. D., M. J. Ebersold, and J. W. Albers. 1975. Summated auditoryevoked potentials in cerebellum and inferior colliculus in young rat. MayoClinic Proceedings 50 (2):57–68.Physiology and connections of the cerebellum.

Bibliographic Notes 293*Holinger, D. P., U. Bellugi, D. L. Mills, J. R. Korenberg, A. L. Reiss, G. F. Sherman,and A. M. Galaburda. In press. Relative sparing of primary auditory cortex inWilliams syndrome. Brain Research.The article that Ursula told Crick about.*Hopfield, J. J. 1982. Neural networks and physical systems with emergent collectivecomputational abilities. Proceedings of National Academy of Sciences 79(8):2554–2558.The first statement of Hopfield nets, a form of neural network model.*Huang, C., and G. Liu. 1990. Organization of the auditory area in the posteriorcerebellar vermis of the cat. Experimental Brain Research 81 (2):377–383.*Huang, C.-M., G. Liu, and R. Huang. 1982. Projections from the cochlear nucleusto the cerebellum. Brain Research 244:1–8.*Ivry, R. B., and R. E. Hazeltine. 1995. Perception and production of temporal intervalsacross a range of durations: Evidence for a common timing mechanism.Journal of Experimental Psychology: Human Perception and Performance 21(1):3–18.Papers on the physiology, anatomy, and connectivity of the cerebellumand lower auditory areas.*Jastreboff, P. J. 1981. Cerebellar interaction with the acoustic reflex. Acta NeurobiologiaeExperimentalis 41 (3):279–298.Source for information on the acoustic “startle” reflex.*Jones, M. R. 1987. Dynamic pattern structure in music: recent theory and research.Perception & Psychophysics 41:621–634.*Jones, M. R., and M. Boltz. 1989. Dynamic attending and responses to time. PsychologicalReview 96:459–491.Examples of Jones’s work on timing and music.*Keele, S. W., and R. Ivry. 1990. Does the cerebellum provide a common computationfor diverse tasks—A timing hypothesis. Annals of The New York Academyof Sciences 608:179–211.Example of Ivry’s work on timing and the cerebellum.*Large, E. W., and M. R. Jones. 1995. The time course of recognition of novelmelodies. Perception and Psychophysics 57 (2):136–149.*———. 1999. The dynamics of attending: How people track time-varyingevents. Psychological Review 106 (1):119–159.More examples of Jones’s work on timing and music.*Lee, L. 2003. A report of the functional connectivity workshop, Düsseldorf 2002.NeuroImage 19:457–465.One of the papers Menon read to create the analyses we needed for ournucleus accumbens study.

294 Bibliographic Notes*Levitin, D. J., and U. Bellugi. 1998. Musical abilities in individuals with Williamssyndrome. Music Perception 15 (4):357–389.*Levitin, D. J., K. Cole, M. Chiles, Z. Lai, A. Lincoln, and U. Bellugi. 2004. Characterizingthe musical phenotype in individuals with Williams syndrome. ChildNeuropsychology 10 (4):223–247.Information on Williams syndrome and two studies of their musical abilities.*Levitin, D. J., and V. Menon. 2003. Musical structure is processed in “language”areas of the brain: A possible role for Brodmann Area 47 in temporal coherence.NeuroImage 20 (4):2142–2152.*———. 2005. The neural locus of temporal structure and expectancies in music:Evidence from functional neuroimaging at 3 Tesla. Music Perception 22(3):563–575.*Levitin, D. J., V. Menon, J. E. Schmitt, S. Eliez, C. D. White, G. H. Glover, J. Kadis,J. R. Korenberg, U. Bellugi, and A. L. Reiss. 2003. Neural correlates of auditoryperception in Williams syndrome: An fMRI study. NeuroImage 18 (1):74–82.Studies that showed cerebellar activations to music listening.*Loeser, J. D., R. J. Lemire, and E. C. Alvord. 1972. Development of folia in humancerebellar vermis. Anatomical Record 173 (1):109–113.Background on cerebellar physiology.*Menon, V., and D. J. Levitin. 2005. The rewards of music listening: Response andphysiological connectivity of the mesolimbic system. NeuroImage 28 (1):175–184.The paper in which we showed the involvement of the nucleus accumbensand the brain’s reward system in music listening.*Merzenich, M. M., W. M. Jenkins, P. Johnston, C. Schreiner, S. L. Miller, andP. Tallal. 1996. Temporal processing deficits of language-learning impaired childrenameliorated by training. Science 271:77–81.Paper showing that dyslexia may be caused by a timing deficit in children’sauditory systems.*Middleton, F. A., and P. L. Strick. 1994. Anatomical evidence for cerebellar andbasal ganglia involvement in higher cognitive function. Science 266 (5184):458–461.*Penhune, V. B., R. J. Zatorre, and A. C. Evans. 1998. Cerebellar contributions tomotor timing: A PET study of auditory and visual rhythm reproduction. Journalof Cognitive Neuroscience 10 (6):752–765.*Schmahmann, J. D. 1991. An emerging concept—the cerebellar contribution tohigher function. Archives of Neurology 48 (11):1178–1187.*Schmahmann, Jeremy D., ed. 1997. The Cerebellum and Cognition, InternationalReview of Neurobiology, v. 41. San Diego: Academic Press.

Bibliographic Notes 295*Schmahmann, S. D., and J. C. Sherman. 1988. The cerebellar cognitive affectivesyndrome. Brain and Cognition 121:561–579.Background information on the cerebellum, function, and anatomy.*Tallal, P., S. L. Miller, G. Bedi, G. Byma, X. Wang, S. S. Nagarajan, C. Schreiner,W. M. Jenkins, and M. M. Merzenich. 1996. Language comprehension in languagelearningimpaired children improved with acoustically modified speech. Science271:81–84.Paper showing that dyslexia may be caused by a timing deficit in children’sauditory systems.*Ullman, S. 1996. High-level Vision: Object Recognition and Visual Cognition.Cambridge: MIT Press.On the architecture of the visual system.*Weinberger, N. M. 1999. Music and the auditory system. In The Psychology ofMusic, edited by D. Deutsch. San Diego: Academic Press.On the physiology and connectivity of the music/auditory system.Chapter 7*Abbie, A. A. 1934. The projection of the forebrain on the pons and cerebellum.Proceedings of the Royal Society of London (Biological Sciences) 115:504–522.Source of the quote about the cerebellum being involved in art.*Chi, Michelene T. H., Robert Glaser, and Marshall J. Farr, eds. 1988. The Natureof Expertise. Hillsdale, N.J.: Lawrence Erlbaum Associates.Psychological studies of expertise, including chess players.*Elbert, T., C. Pantev, C. Wienbruch, B. Rockstroh, and E. Taub. 1995. Increasedcortical representation of the fingers of the left hand in string players. Science270 (5234):305–307.Source for the cortical changes associated with playing violin.*Ericsson, K. A., and J. Smith, eds. 1991. Toward a General Theory of Expertise:Prospects and Limits. New York: Cambridge University Press.Psychological studies of expertise, including chess players.*Gobet, F., P. C. R. Lane, S. Croker, P. C. H. Cheng, G. Jones, I. Oliver, J. M. Pine.2001. Chunking mechanisms in human learning. Trends in Cognitive Sciences5:236–243.On chunking for memory.*Hayes, J. R. 1985. Three problems in teaching general skills. In Thinking andLearning Skills: Research and Open Questions, edited by S. F. Chipman, J. W.Segal, and R. Glaser. Hillsdale, N.J.: Erlbaum.Source for the study that argued that Mozart’s early works were nothighly regarded, and refutation of the claim that Mozart didn’t need tenthousand hours like everyone else to become an expert.

296 Bibliographic NotesHowe, M. J. A., J. W. Davidson, and J. A. Sloboda. 1998. Innate talents: Reality ormyth? Behavioral & Brain Sciences 21 (3):399–442.One of my favorite articles, although I don’t agree with everything in it; anoverview of the “talent is a myth” viewpoint.Levitin, D. J. 1982. Unpublished conversation with Neil Young, Woodside, CA.———. 1996. Interview: A Conversation with Joni Mitchell. Grammy, Spring, 26–32.———. 1996. Stevie Wonder: Conversation in the Key of Life. Grammy, Summer,14–25.———. 1998. Still Creative After All These Years: A Conversation with Paul Simon.Grammy, February, 16–19, 46.———. 2000. A conversation with Joni Mitchell. In The Joni Mitchell Companion:Four Decades of Commentary, edited by S. Luftig. New York: Schirmer Books.———. 2001. Paul Simon: The Grammy Interview. Grammy, September, 42–46.———. 2004. Unpublished conversation with Joni Mitchell, December, Los Angeles,CA.Sources for the anecdotes and quotations from these musicians aboutmusical expertise.MacArthur, P. (1999). JazzHouston Web site. http:www.jazzhouston.com/forum/messages.jsp?key=352&page=7&pKey=1&fpage=1&total=588.Source of the quote about Rubinstein’s mistakes.*Sloboda, J. A. 1991. Musical expertise. In Toward a General Theory of Expertise,edited by K. A. Ericcson and J. Smith. New York: Cambridge University Press.Overview of issues and findings in musical expertise literature.Tellegen, Auke, David Lykken, Thomas Bouchard, Kimerly Wilcox, Nancy Segal,and Stephen Rich. 1988. Personality similarity in twins reared apart and together.Journal of Personality and Social Psychology 54 (6):1031–1039.The Minnesota Twins study.*Vines, B. W., C. Krumhansl, M. M. Wanderley, and D. Levitin. In press. Crossmodalinteractions in the perception of musical performance. Cognition.Source of the study about musician gestures conveying emotion.Chapter 8*Berlyne, D. E. 1971. Aesthetics and Psychobiology. New York: Appleton-Century-Crofts.On the “inverted-U” hypothesis of musical liking.*Gaser, C., and G. Schlaug. 2003. Gray matter differences between musicians andnonmusicians. Annals of the New York Academy of Sciences 999:514–517.Differences between the brains of musicians and nonmusicians.

Bibliographic Notes 297*Husain, G., W. F. Thompson, and E. G. Schellenberg. 2002. Effects of musicaltempo and mode on arousal, mood, and spatial abilities. Music Perception 20(2):151–171.The “Mozart Effect” explained.*Hutchinson, S., L. H. Lee, N. Gaab, and G. Schlaug. 2003. Cerebellar volume ofmusicians. Cerebral Cortex 13:943–949.Differences between the brains of musicians and nonmusicians.*Lamont, A. M. 2001. Infants’ preferences for familiar and unfamiliar music: Asocio-cultural study. Paper read at Society for Music Perception and Cognition,August 9, 2001, at Kingston, Ont.On infants’ prenatal musical experience.*Lee, D. J., Y. Chen, and G. Schlaug. 2003. Corpus callosum: musician and gendereffects. NeuroReport 14:205–209.Differences between the brains of musicians and nonmusicians.*Rauscher, F. H., G. L. Shaw, and K. N. Ky. 1993. Music and spatial task performance.Nature 365:611.The original report of the “Mozart Effect.”*Saffran, J. R. 2003. Absolute pitch in infancy and adulthood: the role of tonalstructure. Developmental Science 6 (1):35–47.On the use of absolute pitch cues by infants.*Schellenberg, E. G. 2003. Does exposure to music have beneficial side effects?In The Cognitive Neuroscience of Music, edited by I. Peretz and R. J. Zatorre.New York: Oxford University Press.*Thompson, W. F., E. G. Schellenberg, and G. Husain. 2001. Arousal, mood, andthe Mozart Effect. Psychological Science 12 (3):248–251.The “Mozart Effect” explained.*Trainor, L. J., L. Wu, and C. D. Tsang. 2004. Long-term memory for music: Infantsremember tempo and timbre. Developmental Science 7 (3):289–296.On the use of absolute-pitch cues by infants.*Trehub, S. E. 2003. The developmental origins of musicality. Nature Neuroscience6 (7):669–673.*———. 2003. Musical predispositions in infancy. In The Cognitive Neuroscienceof Music, edited by I. Peretz and R. J. Zatorre. Oxford: Oxford University Press.On early infant musical experience.Chapter 9Barrow, J. D. 1995. The Artful Universe. Oxford, UK: Clarendon Press.“Music has no role in survival of the species.”

298 Bibliographic NotesBlacking, J. 1995. Music, Culture, and Experience. Chicago: University ofChicago Press.“The embodied nature of music, the indivisibility of movement andsound, characterizes music across cultures and across time.”Buss, D. M., M. G. Haselton, T. K. Shackelford, A. L. Bleske, and J. C. Wakefield.2002. Adaptations, exaptations, and spandrels. In Foundations of CognitivePsychology: Core Readings, edited by D. J. Levitin. Cambridge: MIT Press.I’ve intentionally avoided making a distinction between two types of evolutionaryby-products, spandrels and exaptations, in order to simplify thepresentation in this chapter, and I’ve used the term spandrels for bothtypes of evolutionary by-products. Because Gould himself did not use theterms consistently through his writings, and because the main point is notcompromised by glossing over this distinction, I present a simplified explanationhere, and I don’t think that readers will suffer any loss of understanding.Buss, et al., discuss this distinction and others, based on thework of Stephen Jay Gould cited below.*Cosmides, L. 1989. The logic of social exchange: Has natural selection shapedhow humans reason? Cognition 31:187–276.*Cosmides, L., and J. Tooby. 1989. Evolutionary psychology and the generationof culture, Part II. Case Study: A computational theory of social exchange. Ethologyand Sociobiology 10:51–97.Perspectives of evolutionary psychology on cognition as adaptation.Cross, I. 2001. Music, cognition, culture, and evolution. Annals of the New YorkAcademy of Sciences 930:28–42.———. 2001. Music, mind and evolution. Psychology of Music 29 (1):95–102.———. 2003. Music and biocultural evolution. In The Cultural Study of Music:A Critical Introduction, edited by M. Clayton, T. Herbert and R. Middleton. NewYork: Routledge.———. 2003. Music and evolution: Consequences and causes. Comparative MusicReview 22 (3):79–89.———. 2004. Music and meaning, ambiguity and evolution. In Musical Communications,edited by D. Miell, R. MacDonald and D. Hargraves.The sources for Cross’s arguments as articulated in this chapter.Darwin, C. 1871/2004. The Descent of Man and Selection in Relation to Sex.New York: Penguin Classics.The source for the ideas Darwin had about music, sexual selection, andadaptation. “I conclude that musical notes and rhythm were first acquiredby the male or female progenitors of mankind for the sake of charmingthe opposite sex. Thus musical tones became firmly associated with someof the strongest passions an animal is capable of feeling, and are consequentlyused instinctively. ...”

Bibliographic Notes 299*Deaner, R. O., and C. L. Nunn. 1999. How quickly do brains catch up with bodies?A comparative method for detecting evolutionary lag. Proceedings of theRoyal Society of London B 266 (1420):687–694.On evolutionary lag.Gleason, J. B. 2004. The Development of Language, 6th ed. Boston: Allyn & Bacon.Undergraduate text on the development of language ability.*Gould, S. J. 1991. Exaptation: A crucial tool for evolutionary psychology. Journalof Social Issues 47:43–65.Gould’s explication of different kinds of evolutionary by-products.Huron, D. 2001. Is music an evolutionary adaptation? In Biological Foundationsof Music.Huron’s response to Pinker (1997); the idea of comparing autism toWilliams syndrome for an argument about the link between musicalityand sociability first appeared here.*Miller, G. F. 1999. Sexual selection for cultural displays. In The Evolution ofCulture, edited by R. Dunbar, C. Knight and C. Power. Edinburgh: EdinburghUniversity Press.*———. 2000. Evolution of human music through sexual selection. In The Originsof Music, edited by N. L. Wallin, B. Merker and S. Brown. Cambridge: MIT Press.———. 2001. Aesthetic fitness: How sexual selection shaped artistic virtuosityas a fitness indicator and aesthetic preferences as mate choice criteria. Bulletinof Psychology and the Arts 2 (1):20–25.*Miller, G. F., and M. G. Haselton. In Press. Women’s fertility across the cycleincreases the short-term attractiveness of creative intelligence compared towealth. Human Nature.Source articles for Miller’s view on music as sexual fitness display.Pinker, S. 1997. How the Mind Works. New York: W. W. Norton.Source of Pinker’s “auditory cheesecake” analogy.Sapolsky, R. M. Why Zebras Don’t Get Ulcers, 3rd ed. 1998. New York: Henry Holtand Company.On evolutionary lag.Sperber, D. 1996. Explaining Culture. Oxford, UK: Blackwell.Music as an evolutionary parasite.*Tooby, J., and L. Cosmides. 2002. Toward mapping the evolved functional organizationof mind and brain. In Foundations of Cognitive Psychology, edited byD. J. Levitin. Cambridge: MIT Press.Another work by these evolutionary psychologists on cognition as adaptation.

300 Bibliographic NotesTurk, I. Mousterian Bone Flute. Znanstvenoraziskovalni Center Sazu 1997 [citedDecember 1, 2005. Available from http:www.uvi.si/eng/slovenia/backgroundinformation/neanderthal-flute/.]The original report on the discovery of the Slovenian bone flute.*Wallin, N. L. 1991. Biomusicology: Neurophysiological, Neuropsychological, andEvolutionary Perspectives on the Origins and Purposes of Music. Stuyvesant,N.Y.: Pendragon Press.*Wallin, N. L., B. Merker, and S. Brown, eds. 2001. The Origins of Music. Cambridge:MIT Press.Further reading on the evolutionary origins of music.

ACKNOWLEDGMENTSI would like to thank all the people who helped me to learn what I know aboutmusic and the brain. For teaching me how to make records, I am indebted to theengineers Leslie Ann Jones, Ken Kessie, Maureen Droney, Wayne Lewis, JeffreyNorman, Bob Misbach, Mark Needham, Paul Mandl, Ricky Sanchez, Fred Catero,Dave Frazer, Oliver di Cicco, Stacey Baird, Marc Senasac, and the producersNarada Michael Walden, Sandy Pearlman, and Randy Jackson; and for giving methe chance to, Howie Klein, Seymour Stein, Michelle Zarin, David Rubinson,Brian Rohan, Susan Skaggs, Dave Wellhausen, Norm Kerner, and Joel Jaffe. Fortheir musical inspiration and time spent in conversation I am grateful to StevieWonder, Paul Simon, John Fogerty, Lindsey Buckingham, Carlos Santana, kdlang, George Martin, Geoff Emerick, Mitchell Froom, Phil Ramone, RogerNichols, George Massenburg, Cher, Linda Ronstadt, Peter Asher, Julia Fordham,Rodney Crowell, Rosanne Cash, Guy Clark, and Donald Fagen. For teaching meabout cognitive psychology and neuroscience, Susan Carey, Roger Shepard,Mike Posner, Doug Hintzman, and Helen Neville. I am grateful to my collaborators,Ursula Bellugi and Vinod Menon, who have given me an exciting and rewardingsecond career as a scientist, and to my close colleagues Steve McAdams,Evan Balaban, Perry Cook, Bill Thompson, and Lew Goldberg. My students andpostdoctoral fellows have been an additional source of pride and inspiration,and helped with their comments on drafts of this book: Bradley Vines, CatherineGuastavino, Susan Rogers, Anjali Bhatara, Theo Koulis, Eve-Marie Quintin,Ioana Dalca, Anna Tirovolas, and Andrew Schaaf. Jeff Mogil, Evan Balaban,Vinod Menon, and Len Blum provided valuable comments on portions of themanuscript. Still, any errors are my own. My dear friends Michael Brook and JeffKimball have helped me throughout the writing of this book in many ways, withtheir conversation, questions, support, and musical insights. My department

302 Acknowledgmentschair, Keith Franklin, and the dean of the Schulich School of Music, DonMcLean, have provided me with an enviably productive and supportive intellectualenvironment within which to work.I would also like to thank my editor at Dutton, Jeff Galas, for his guidanceand support through every step of turning these ideas into a book, for his hundredsof suggestions and excellent advice, and Stephen Morrow at Dutton for hishelpful contributions in editing the manuscript; without Jeff and Stephen, thisbook would not have existed. Thank you both.The subtitle for Chapter 3 is taken from the excellent book edited by R. Steinbergand published by Springer-Verlag.And thank you to my favorite pieces of music: Beethoven’s Sixth Symphony;“Joanne” by Michael Nesmith; “Sweet Georgia Brown” by Chet Atkins and LennyBreau; and “The End” by the Beatles.

INDEXNote: Page numbers in italics refer to illustrations or charts.A440, 32–33Abbie, Andrew Arthur, 206Abdul, Paula, 57, 168absolute pitch, 145–50and infants, 222and melody, 30neural basis for, 27, 191and tone deafness, 184value changes in, 25AC/DC, 57–58, 166Acoustical Society of America, 18Adam and the Ants, 5adaptation, 7–8, 99, 250, 252adolescents, 225–27, 247advertising, 9Aerosmith, 58affect, 178, 187. See also emotion“All Along the Watchtower,” 49Allman Brothers, 111“All of Me,” 232Alzheimer’s disease, 225amplitude, 15, 67–68, 78amygdala, 265and cerebellum, 171and emotion, 85, 185, 225and expressivity in performance,206and memory, 163and mental disorders, 254responding to stimuli, 89“Anarchy in the U.K.,” 49Anderson, Leroy, 225animals, 29, 90, 95, 257–58anterior cingulate, 224antidepressants, 121antiquity of music, 5–6anxiety, 180appearance (physical), 197–98appreciation of music, 109Arab music, 37area MT, 181Aristotle, 136, 137, 139, 141, 258Armstrong, Louis, 144, 208artists, 4–5, 238–39associations with music, 36–37“As Time Goes By,” 232The Astonishing Hypothesis (Crick),175“At a Darktown Cakewalk,” 56attack, 47, 51–52attention, 76, 79, 194, 224–25audience expertise, 6–7, 206, 216auditory cortex, 84, 87, 89, 187, 191, 264auditory systemanatomy, 100–101auditory-code, 119–20and cerebellum, 180, 182, 183and neural processing of music, 101–2,128, 187and perceptual completion, 99physiology of hearing, 22, 27and simultaneous onsets of sounds,77–78startle, 181augmented fourth (tritone), 13, 72, 223Austin Lounge Lizards, 145

304 Indexautism spectrum disorders (ASD), 253–54avant-garde music, 14BA44, 187BA47, 187“Ba Ba Black Sheep,” 60, 61, 62babies. See infancy and childhoodBach, Johann Sebastian, 14, 79, 144backbeat, 64–65, 111–12“Back in Black,” 57–58, 165–66“Back in Your Arms,” 238Balint’s syndrome, 184Baron-Cohen, Simon, 256baroque music, 33Barrow, John, 243bars, 62basal ganglia, 59, 187bass guitars, 209, 210–11Beach Boys, 226beat, 57, 59–63, 166, 169–71. See alsorhythm“Beat It,” 138Beatleson The Ed Sullivan Show, 200and EMI, 126fans of, 237followers of, 5influence on author, 200–201musical significance of, 49timbral qualities in albums, 105, 152use of expectations, 110–11, 115use of keys, 70–71use of synthesizers, 46“Be-Bop-A-Lula,” 153Beethoven, Ludwig van, 65, 116–17, 165,205, 208Bell, Alexander Graham, 67Bellugi, Ursula, 174–75, 176, 180, 182,252–53Bennett, Max, 209Berkeley, George, 22Berle, Milton, 56Berlin, Irving, 204, 216Bernstein, Leonard, 56, 205, 257Berry, Chuck, 64“Bibbidy Bobbidy Boo,” 225Billboard, 189“Billie Jean,” 58, 168binding problem, 183–84birds and birdsongs, 258–59Blacking, John, 251Blood, Anne, 185“Blue Moon,” 233Blues music, 36, 37, 111Bolero, 52, 125, 257bottom-up processing, 101–2, 103Bouchard, Thomas, 196bowed instruments, 51Bowie, David, 37, 226brain. See also specific anatomicalstructuresanatomy, 82–83damage to, 9, 82–83, 85evolution of, 8–9and mind, 81–82, 91–93, 95musical activity in, 83–84organization of, 121–22parallel processing of brains, 86–87brain stem, 55, 72, 84, 206“brainstorming” stage, 5Bregman, Albert, 74, 76, 99Brendel, Alfred, 205bridges, 232Broca’s area, 84, 124, 260Brodmann areas, 89, 127Brown, James, 252Brubeck, Dave, 66Buckingham, Lindsey, 53“Bum-Diddle-De-Um-Bum, That’s It!,” 56Burns, Ed, 147–48Byrne, David, 238cadence, deceptive, 109–10Cage, John, 14, 257call-and-response patterns, 167canonical versions of music, 148Carey, Susan, 93caring and skills acquisition, 193–94Carlos, Walter/Wendy, 46Carpenters, 111, 138Cash, Johnny, 239Castellengo, Michelle, 52categorization, 136–45. See also memoryconstructivist theory, 131, 133, 134, 136,145, 153, 155and evolution, 142–43exemplar theory, 155, 157–58, 160and memory, 145, 155prototypes in categories, 140–41, 143–45,155–56, 157–58, 223record-keeping theory, 131, 135, 136, 145,153, 155, 160Catholic Church, 13“Cathy’s Clown,” 153celebrity, 207cellos, 28Center for Computer Research in Musicand Acoustics (CCRMA), 47, 48cerebellar vermis, 85, 89cerebellum, 264, 265and auditory system, 180, 182, 183effect of music on, 220–21, 257and emotion, 171, 174, 178–80, 183, 187and expressivity in performance, 206and frontal lobes, 185function, 83

Index 305and listening to music, 84, 89and memory, 59and mental disorders, 253–54and meter, 66and performing music, 55, 59and timing, 170–71, 174, 178cerebral cortex, 257cerebrum, 170“Chain Lightning,” 110charisma of performers, 207, 216Cheap Trick, 5chess, 212–13, 214children. See infancy and childhood“China Girl,” 37Chinese music, 36Chomsky, Noam, 107, 174Chopin, Frédéric, 65, 104chordsand cadence, 109–10chord progression, 17, 71and consonance and dissonance, 71–73defining, 38–39and expectations for, 123and harmony, 267–70memory for, 214root of, 209–10schemas for, 115chorus, 232–33Chowning, John, 47–48, 145chromatic scale, 34chunking, 213–14Churchland, Paul, 4cingulate gyrus, 224“The Circle Game,” 210circle of fifths, 72Clapton, Eric, 49, 50, 207, 208clarinets, 44Clarke, Eric, 65classical music, 16, 168, 251–52, 257Clinton, Bill, 203cochlear nuclei, 84cognitive development, 254–56cognitive neuroscience, 93–95, 121, 183–84cognitive psychology, 93, 104, 117Cold Spring Harbor Laboratory, 171–74, 185color, 21, 22, 112Coltrane, John, 110, 208Columbo, John, 218complexity, 234composersand expectations in music, 64, 109, 110and keys, 70and meter, 165–66use of note length, 90use of timbre, 52, 90computers, 117–19, 130–31, 169–70concerts, 69consciousness, 175, 184consonance, tonal, 71–73, 221–22, 223constructive process, 103constructivist theory of memory, 131–36,145, 153, 155, 160context, 155–56, 157contour, 15, 168, 222–23Cooder, Ry, 208Cook, Perry, 58–59, 110, 145, 170Copeland, Aaron, 257Copeland, Stewart, 157corpus callosum, 220, 265Cosmides, Leda, 8, 256country music, 38creativity, 248Creedence Clearwater Revival, 111,226Crick, Francisauthor’s introduction to, 177–78on career in sciences, 175–76, 207on cognitive neuroscience, 183–84on connections, 171, 184–85, 188DNA discovery, 259Crosby, David, 208Cross, Ian, 243, 252, 256cymbals, 52dancing, 17, 247–48Dani tribe of New Guinea, 140–41“Dark Side of the Moon,” 145Darwinian theory, 8, 241, 243–50, 252,259Dave Matthews Band, 237Davidson, Jane, 190–91Davis, Miles, 17–18, 110, 115, 207deafness, 127decibels, 67, 68–69declarative knowledge, 36defining music, 13–14Dennett, Daniel, 21, 92, 96Depeche Mode, 14depression, 180Desain, Peter, 169–70Descartes, René, 81The Descent of Man (Darwin), 245De Vol, Frank, 225Dhomont, Francis, 14Diabolus in musica, 13DiFranco, Ani, 237dissonance, tonal, 71–73, 221–22, 223divertimenti, 78–79Dixieland, 114Doors, 38dopamine, 121, 185, 186, 187, 194dorsalateral prefrontal cortex, 89dorsal cochlear nucleus, 72dorsal temporal lobes, 160–61double-basses, 26Dowling, Jay, 145, 222

306 Indexdrums, 59, 167–68dualism, 81Dylan, Bob, 13dynamic range compression, 67, 68Eagles, 58, 71, 104–5earplugs, 69ear worms, 151echo, 16, 106, 153. See also reverberationechoic memory, 151Edelman, Gerald, 59education, musical, 189–90, 194, 207–8Ehrenfels, Christian von, 73–74“eighties sound” in popular music, 48“Eine Kleine Nachtmusik,” 166Elbert, Thomas, 191electroencephalograms (EEG), 123–24,150Emerson, Lake and Palmer, 46emotionand amygdala, 185, 225and cerebellum, 83, 171, 174, 178–79, 183,187in classical music, 168effect of music on, 187, 234–35, 261evolution of, 178–79and expectations in music, 109and expertise, 204–6and groove, 188and loudness, 69and memory, 225and metrical extraction, 168–69neural basis for, 85, 89, 106, 185and pitch, 25–26, 28in Songs for Swinging Lovers,189and syncopation, 63and tempo, 58and timbre, 52and Williams syndrome (WS), 183environmental influences, 195–96, 198–99,203equal tempered scale, 50eras, 115, 152–53Ericsson, Anders, 192Everly Brothers, 153“Every Breath You Take,” 52, 56–57evolutionadaptation, 7–8, 99, 250, 252and categorization, 142–43and cognitive development, 254–57Darwinian theory, 8, 241, 243–50, 252,259of emotions, 178–81of language, 241–42, 243, 250, 254–55of musical preferences, 242–51, 254in other species, 257–58and perception, 99, 104and sexual selection, 244–50, 252, 258–59, 260–61and social cohesion, 252–54evolutionary psychology, 8exemplar theory, 155, 158, 160expectationsof learned musical systems, 112for meter, 165–66and musical preferences, 229–31for pitch, 70and processing music, 102for rhythm, 111–12studying, 123–24violations of, 64, 90–91, 110–17, 166,168–69, 187experiments, 94–95expertisein audience, 6–7, 206, 216defining, 192, 216and expressivity, 204–7and musical memory, 211–15and nature/nurture debate, 195–203and practice, 191–94study of, 190–91and talent, 190–92and technical prowess, 204, 207, 216Fagen, Donald, 110, 235failure and success, 202–3Fantasy-Impromptu in C-sharp Minor, op.66, 104Fantz, Robert, 218feature integration and extraction,101Ferguson, Jim, 6–7Fernald, Anne, 218fetuses, 217–19Fifth Symphony of Beethoven, 165films, 9, 23first degree (tonic), 37Fitzgerald, Ella, 144five-note (pentatonic) scale, 36flats, 31–32Fleetwood, Mick, 156Fleetwood Mac, 156, 166, 203flux, 47, 52FM synthesis, 47, 48Fogassi, Leonardo, 259–60“A Foggy Day,” 144folk music, 61form in music, 106“For No One,” 70–71, 1104/4 time, 61–62Franklin, Aretha, 144frequencyA440, 32–33fundamental frequencies, 40–41and grouping, 79

Index 307of light waves, 21low frequencies, 22and notes, 28–29and overtones, 40, 77perception of, 26–27and physiology of hearing, 27and pitch, 15, 19, 20–25, 24, 32–33“Frère Jacques,” 61Friederici, Angela, 124, 126Friston, Karl, 186frontal cortex, 127, 185frontal lobesand cerebellum, 185development of, 224and expressivity in performance, 206function, 83, 125and listening to music, 84and musical structure, 124and performing music, 55, 84and processing music, 102–3pruning of, 227functional and effective connectivityanalysis, 186functionalism, 92functional MRI (fMRI), 126–27, 159,185–86fundamental frequencies, 40–44Funeral March, 61Gabriel, Peter, 67, 167Gage, Phineas, 83Galaburda, Albert, 182Gallese, Vittorio, 259–60Gallistel, Randy, 173games, 137–38gap fill, 115–16Gazzaniga, Michael, 133“Gee, Officer Krupke,” 56genetics, 195–203, 244–50genres, 115, 138–39, 141–42, 145, 233Gershwin, George, 208Gestalt psychologists, 73–74, 96, 131, 134,158Getz, Stan, 52Ghost in the Machine (The Police), 112Gilmour, David, 106glass, breaking, 23Glass, Philip, 257glissandos, 37Gogh, Vincent van, 203Goldinger, Stephen, 138, 160Goldstein, Avram, 185Gould, Stephen Jay, 242Grandin, Temple, 253Grant, Hugh, 199Grateful Dead, 237gray matter, 221“Great Gate of Kiev,” 37Gregory, Richard, 99groove, 166–68, 188grouping, 73–79, 96guitars, 13, 200–202, 208–11Hale, Charles, 56half notes, 61Halpern, Andrea, 147–48, 153Hammerstein, Oscar, 65Hanks, Tom, 199“Happy Birthday,” 147, 148harmonics, 42, 44harmony, 17, 40–41, 70, 211, 267–70Harrison, George, 238Hartford, John, 163Haselton, Martie, 248Haydn, Joseph, 90–91, 110, 144Hayes, John, 195hearing, 22, 27. See also auditory system“Heartbreak Hotel,” 153heavy metal music, 67, 111, 138–39, 165Helfgott, David, 208Helmholtz, Hermann von, 75, 77, 99, 103hemispheric specialization, 121–22Hendrix, Jimi, 49, 53, 166, 246“Here’s That Rainy Day,” 52Hermann, Bernard, 37Hertz (measurement), 19Hertz, Heinrich, 19hierarchical encoding of music, 154, 215high fidelity, 68high-hat cymbal, 167Hintzman, Douglas, 134, 138, 150, 160hip-hop, 235hippocampus, 265and expressivity in performance, 206and listening to music, 84, 89, 161and memory, 82–83, 161, 163and processing of music, 128Holiday, Billie, 36, 233Holly, Buddy, 62–64Honing, Henkjan, 169–70“Honky Tonk Women,” 165, 257Hopfield, John, 173Horowitz, Vladimir, 204“Hotel California,” 58, 71“Hot Fun in the Summertime,” 29“Hound Dog,” 111Howe, Michael, 190–91Huron, David, 249hyperrealities, 106“Hypnotized,” 166Idle, Eric, 152illusions, 97–99, 103, 104, 106“I’m On Fire,” 167improvisation, 232, 233, 248Indian music, 37

308 Indexinfancy and childhoodattentional abilities, 224–25auditory systems in, 222and contour, 222–23and hemispheric specialization, 123and language acquisition, 255–56and musical memory, 217–19, 221and music lessons, 189–90, 194neuroplasticity, 39, 107, 227and preferences in music, 217–19, 221,224, 239–40schema development, 114singing to infants, 9, 256synesthetic phase of, 125and talent, 191vocalizations in, 224inferior frontal cortex, 84, 180, 215inharmonic overtones, 42–43“Instant Karma,” 64–65, 153instrumentation and categorization,145instruments, musicalancient artifacts, 250, 251and attack, 51–52cognitive requirements for playing,55emotional expression, 52frequencies, 22–23, 24and grouping, 76overtones, 44timbral fingerprints, 44–45intelligence, effect of music on, 219–21intervals, 29–31, 30, 71–73, 145, 223inverted-U hypothesis, 234Ionian mode (major scales), 34–35, 36, 37,72, 223–24, 267–68Isley Brothers, 155, 166isomorphic representation of world, 95–96,117Ivry, Richard, 173, 174, 185“I Want You (She’s So Heavy),” 110Jackendoff, Ray, 74–75Jackson, Michael, 58, 138, 168Jackson, Randy, 111–12Jagger, Mick, 246“Jailhouse Rock,” 60–61, 62James, Rick, 166Janata, Petr, 41, 150jazz, 144, 232–33Jobim, Antonio Carlos, 71“Jolene,” 38Jones, Leslie Ann, 3Jones, Mari Reiss, 173Jusczyk, Peter, 218Kamakiriad (Fagen), 110Kaniza figure, 103Keele, Steve, 143–44, 145, 173Kemp, Martin, 4key, 16, 70Kind of Blue (Davis), 18King, B. B., 205–6Kinks, 111Klein, Larry, 209Koelsch, Stefan, 124, 126Koffka, Kurt, 73–74Köhler, Wolfgang, 73–74“Koko,” 257Kosinsky, Jerzy, 203Kottke, Leo, 208Krumhansl, Carol, 37–38“Lady Jane,” 111“Lady Madonna,” 105Lamont, Alexandra, 217–18, 221languageand cerebellum, 185evolution of, 241–42, 243, 250, 254–55language acquisition, 222–23, 227,255–56language centers of the brain, 84, 85,122–23, 124–28and oral tradition, 261The Language Instinct (Pinker), 243lateral cerebellum, 89Latin music, 235learning theory, 193Led Zeppelin, 33, 138–39, 201–02Lee, Lester, 56left-handedness, 121–22left hemisphere, 8, 121–23, 127, 132–33, 169,220Leiber, Jerry, 60leitmotiv, 26length of songs, 115Lennon, John, 64, 153, 235–36Lerdahl, Fred, 74“Light My Fire,” 38“Lilies of the Valley,” 238listening to music, 83–84, 150–51“Little Red Corvette,” 49Little Richard, 49lobotomy, 83Locatelli, Pietro Antonio, 79Locke, John, 97Loftus, Elizabeth, 132logic of perception, 103London Symphony Orchestra, 145“Long Tall Sally,” 49“Lookin’ Out My Back Door,” 111Lortat-Jacob, Bernard, 104loudnessdefining, 15, 20, 67–69and grouping, 78and meter, 16

Index 309neural basis, 69and overtones, 44love songs, 240, 261Lykken, David, 196lyrics, 63, 64magnetic resonance imaging machine(MRI), 126–27Mahler, Gustav, 228–29major chords, 38major scale (Ionian mode), 34–35, 36, 37,72, 223–24, 267–68Mann, Aimee, 199mapping the brain, 94“Mary Had a Little Lamb,” 56mathematics, 227Mathews, Max, 48, 145McCarthy, Joe, 56McClelland, Jay, 159–60McVie, John, 156measures, 62Medin, Douglas, 155, 158, 160melodydefining, 16expectations of, 91, 115–16and harmony, 17and intervals, 30leitmotiv, 26perception of, 131, 169and pitch, 25and rhythm, 257and transposition, 73–74memory, 134–35. See also categorizationaccessing, 161–62accuracy of, 131–33activated by music, 188and caring, 193–94and categorization, 145, 155and chunking, 213–14cues, 162and emotion, 225and exemplar theory, 158and frontal lobes, 83hierarchical encoding of music, 154,215identification memory, 215from infancy, 217–19, 221and listening to music, 150–51multiple-trace memory models, 158–59,160, 161–62muscle memory, 147–48for music, 147–54, 211–15and musical ability, 202and neural network, 88rote memorization, 215and schemas, 114–15strength of, 193for tempo, 58–59theories on, 131, 134, 135, 136, 145, 153,155, 160and tune recognition, 131of voices, 134–36Menon, Vinod, 126, 171, 180, 185–86Mercury, Freddie, 139“Merrie Melody” cartoons, 37Merzenich, Mike, 173mesolimbic system, 187Metallica, 114, 139meterin classical music, 168common meters, 65–67defining, 16, 56, 59–61and loudness, 69neural basis for, 59, 66Metheny, Pat, 106metrical extraction, 168–69Meyer, Leonard, 143The Mickey Mouse Club, 56microphones, 2, 105“microtuning,” 37midbrain, 185A Midsummer Night’s Dream(Shakespeare), 143Miller, Geoffrey, 8, 246, 248Miller, George, 103Miller, Mitch, 257mind and brain, 81–82, 91–93,95MINERVA model, 160Mingus, Charles, 209Minnesota twins registry, 196minor chords, 38minor scale, 35mirror neurons, 259–60“Mission: Impossible,” 66, 67mistakes made in music, 204Mitchell, Joni, 142, 208–11, 237Mitchell, Mitch, 166modulation, 70Monaco, Jimmie, 56“Money,” 67Monty Python, 152motion pictures, 9, 23motivation, 187, 194motor cortex, 55, 82, 84, 89, 264movement and motor skillsand cerebellum, 170development of, 254and emotion, 171, 178–79and expressivity in performance, 206and musical development, 191, 202and parietal lobe, 83Mozart, Wolfgang Amadeus, 60, 78–79,194–95, 258Mozart Effect, 219–20multiple sclerosis, 227–28

310 Indexmultiple-trace memory models, 158–59,160, 161–62muscle memory, 147–48music, defining, 13–14musical syntax, 124music education, 189–90, 194, 207–8musicians, neuroanatomy of, 220–21music industry, 7musicologists, 18music theory, 37music therapy, 221Mussorgsky, Modest Petrovich, 37, 208myelination, 227“My Favorite Things,” 65“My Funny Valentine,” 232nalaxone, 185Narmour, Eugene, 115natural instruments, 45nature/nurture debate, 195–99Needham, Mark, 2Neisser, Ulrich, 103Nelson, Ricky, 153neocerebellum, 253neural codes, 119–20neural network of the brain, 85–90, 120–21and expressivity in performance, 205–6function of, 94mirror neurons, 260and musical expectations, 123–26pruning of, 107, 227redundancy, 181neuroanatomy, 180–81. See also specificanatomical structuresneuroplasticity, 85, 227neuroscience, 117, 120, 140neurotransmitters, 94, 120–21Nevison, Ron, 176–77Newton, Isaac, 21New Wave music, 48Ninth Symphony of Beethoven, 116–17Norman, Jeffrey, 3Normandeau, Robert, 14notation, 62notes. See also tonedefining, 14–15durations, 61–62, 65note names, 28–29, 31, 32and variety in music, 86nucleus accumbens (NAc), 89, 121, 185–86,187–88, 265Nutcracker ballet, 36, 52obsessive-compulsive disorder (OCD),151occipital cortex, 185octaves, 29, 31, 72, 116“Ode to Joy,” 116–17“Ohio,” 167“One After 909,” 213“One Note Samba,” 71“One of These Nights,” 104–5“One Way Out,” 111orbitofrontal regions of the brain, 132–33,180, 224orchestras, 76organs, 45–46overtones, 40–47, 51, 72, 77Page, Jimmy, 202, 209parallel processing in brains, 86–87, 159–60parietal lobes, 83Parker, Charlie, 257Parkinson’s disease, 170Parncutt, Richard, 211, 212, 214–15pars orbitalis, 127Parton, Dolly, 38passive exposure to music, 35Pastorius, Jaco, 209, 211Patel, Ani, 125–26“Pathétique” Sonata of Beethoven, 116, 117,213peacocks, 246Pearlman, Sandy, 3, 4perception, sensory. See sensory perceptionperceptual completion, 98–99, 103percussion instruments, 42, 51Peretz, Isabelle, 169, 184perfect fourth and fifth interval, 31, 72, 223performance of music, 6–7, 55, 59, 84,205–6peripheral nervous system, 120Persian music, 37Peter and the Wolf (Prokofiev), 26Phish, 237phonemes, 128phonogenic quality of musicians,207phonograph records, 119phrase structure, 115, 189physiology of hearing, 27, 235pianos, 23, 26, 31–32, 42Picasso, Pablo, 17piccolos, 24, 26pickup notes, 63Pierce, John R., 48–50, 76, 145Pinker, Steven, 104, 241–43Pink Floyd, 38, 46, 67, 106, 145pipe organs, 45–46pitchA440, 32–33absolute pitch, 25, 27, 30, 145–50, 151,184defining, 15, 18–19, 20–21dimensions of, 112–13dissonance in, 13

Index 311and emotion, 25–26and expectations, 168and frequency, 15, 19, 20–25, 24, 32–33and grouping, 79and guitars, 208–11and harmony, 17and hearing, 27, 101and infants, 222, 223low and high, 19–20, 22and melody, 25and musical memory, 153–54and musical preferences, 235neural basis for, 89, 128overtones, 41–43perception of, 26–27, 29, 41–42proportional changes in, 33–34as psychophysical fiction, 146relative pitch, 25–26, 27, 30, 33and rhythm, 70–73and scales, 27–28and tune recognition, 131and vibration, 39–40and Western music, 50Plant, Robert, 246planum temporale, 191, 215“Please Mr. Postman,” 111Police, 56, 111–12, 156–57polyphony, 13Ponzo illusion, 97popular music, 61, 110, 115, 148, 237Posner, Michaelon attention systems of children, 224–25on Janata’s research, 41on memory, 143–44, 145, 150on mind and brain, 92–93Posner Cueing Paradigm, 92practicing music, 192, 193, 194preferences, musicalin adolescents, 225–27in children, 217–19, 221, 224, 239–40and complexity, 234and cultural bias, 221, 224and evolution, 242–51, 254and expectations in music, 229–31neural basis for, 221–24, 228, 231–32and pitch, 235and prior experiences, 236role of safety, 236–39and schemas, 228–29prefrontal cortex, 264Presley, Elvis, 60–61, 111, 153Pretenders, 167Pribram, Karl, 4“Pride and Joy,” 111Prince, 49producing career of author, 3Prokofiev, Sergey Sergeyevich, 26prosodic cue, 25prototypes in categories, 140–41, 143–45,155–56, 157–58, 223Prozac, 121Psycho, 37psychological issues, effect of music on,221pulse of music, 165–66, 168“Purple Haze,” 166quarter notes, 61Queen, 65, 139Quintina in Sardinian a capella vocal music,104Rachmaninoff, Sergey Vasilyevich, 115Raffi, 139rage, 179–80Ramachandran, V. S., 96Ramones, 83Ravel, Maurice, 52, 53, 125, 257receptors, 120recognition of music, 129–30, 133–34recordings of music, 3, 68, 105, 119, 152–53record-keeping theory of memory, 131,135–36, 145, 153, 155, 160records, 119Redding, Otis, 144redundancy, 181“Refuge of the Roads,” 208reggae music, 111Reinhardt, Django, 201Reinhold, Judge, 199Reiss, Allan, 183, 253relational theory of memory, 131relationships between musical elements, 17R.E.M., 237remembering music, 150–51. See alsomemoryrepetition, 163Repp, Bruno, 173reptilian brain, 170. See also cerebellum“Respect,” 144restoration of the missing fundamental,40–41reverberation, 15–16, 105, 106, 153reviews of musical performances, 18“Revolution 9,” 141–42Revolver (Beatles), 110reward, 187, 242–43rhythm. See also tempodefining, 15, 55–57and evolution, 257and expectations, 111, 169and loudness, 69and meter, 16and metrical extraction, 169and mirror neurons, 260and musical ability, 202

312 Indexrhythm (cont.)and musical preferences, 235, 236neural basis for, 59, 84and pitch, 70–73schemas of, 115and variety in music, 86right-handedness, 121–22right hemisphere, 8, 121–23, 127, 169, 220right temporal lobes, 169Rizzolatti, Giacomo, 259–60Rock, Irvin, 99“Rock and Roll Music,” 64rock musicbackbeat, 64canonical versions, 148chords, 38fans of, 237and loudness, 69and melody, 16and meter, 165–66and musical preferences, 235representative sample of, 49–50standards in, 110and timbre, 50, 76Rodgers, Richard, 65Rolling Stone Encyclopedia of Rock, 5Rolling Stones, 52, 111, 145, 165, 257Rollins, Sonny, 55“Roll Over Beethoven,” 49root of the scale, 34, 35Rosch, Eleanor, 137, 139–41, 143–44, 155,157Ross, Brian, 155Rossini, Gioacchino Antonio, 56rounds, singing, 224–25“Roxanne,” 52, 116Rubinstein, Arthur, 204, 205Rumelhardt, David, 159–60“The Rustle of Spring,” 104Sacks, Oliver, 125, 237Saffran, Jenny, 222“Salisbury Hill,” 67Sapolsky, Robert, 1, 10“Satisfaction,” 52Scaggs, Boz, 177scalesappeal of, 169and categorization, 145chromatic scale, 34defining, 27–29distinguishing between, 35–36equal tempered scale, 50expectations of, 112five-note (pentatonic) scale, 36major scale (Ionian mode), 34–35, 36, 37,72, 223–24, 267–68minor scale, 35and pitch, 27–28root of the scale, 34, 35and schemas, 114, 117and tones, 37–38of Western music, 28, 34, 36Schaeffer, Pierre, 14, 50–51Schellenberg, Glenn, 151–52, 220schemas, 113–17, 168, 214, 228–29schizophrenia, 180Schlaug, Gottfried, 191, 220–21Schmahmann, Jeremy, 171, 174, 179Schönberg, Arnold Franz Walter, 70, 112Schwarzenegger, Arnold, 200scientists and artists, 4–5Scriabin, Aleksandr Nikolayevich, 53Segovia, Andrés, 200, 201selective serotonin reuptake inhibitors(SSRIs), 121semitones, 30, 30–31sensory cortex, 84, 88, 264sensory perceptionand illusions, 97–99, 103, 104, 106isomorphic representation of world,95–96neural basis for, 99–107and startle reactions, 181visual illusions, 96–99serotonin, 121Sex Pistols, 49, 50sexual selection, 244–50, 252, 258–59,260–61sham rage, 179–80Shapiro, Dan, 56shared syntactic integration resourcehypothesis (SSIRH), 125–26sharps, 31–32“shave-and-a-haircut, two bits,” 56“Shave and a Haircut—Shampoo,” 56Shearing, George, 105“Sheep,” 38Shepard, Rogeron categorization, 142on evolution, 8as instructor, 145on memory, 134on perception, 97, 99, 103on pitch, 147Shiffrin, Lalo, 66short-term (“echoic”) memory, 151“Shout,” 166sign language, 127Simon, Herbert, 103Simon, Paul, 2, 53, 152, 207simultaneous onsets of sounds, 77–78Sinatra, Frank, 144, 189, 208, 233Sindig, Christian, 104skill, emphasis on, 7“Sledgehammer,” 167

Index 313Sloboda, John, 190–91, 192Smith, Edward, 155, 158, 160Smith, Julius, 47social variables, 197–98, 252–53“Somewhere Over the Rainbow,” 29, 116Songs for Swinging Lovers (Sinatra), 189Sotho villagers of South Africa, 6–7soundscape, 152–53sound waves, 21–22Sousa, John Philip, 65spandrels, 242, 252spatial location, 15, 78, 83special effects, 106Spencer, Herbert, 244Sperber, Dan, 243spinal cord, 120“Spirits in the Material World,” 112Springsteen, Bruce, 167, 238“Stairway to Heaven,” 139, 201“The Stars and Stripes Forever,” 65startle responses, 181–82“Stayin’ Alive,” 166steady state, 51steams, 76Steely Dan, 110, 115Sting, 52, 111–12, 116, 156–57, 207Stoller, Mike, 60“Straight Up,” 57, 58, 168streaming by timbre, 99stream segregation, 104stringed instruments, 28structure in music, 124–28and illusion, 106and memory, 213and musical ability, 202and musical preferences, 231–33, 233–34and neural processing of music, 186,187styles, 114, 115. See also genressuccess and failure, 202–3Summers, Andy, 157“Super Freak,” 166superior temporal gyrus, 89superior temporal sulcus, 89“Superstition,” 32, 167“Surprise Symphony,” 90–91suspense, 90–91synapses, 120“The Syncopated Clock,” 225syncopation, 63syntax, musical, 124–25synthesizers, 45–48tactus, 57, 62–63. See also beat“Take Five,” 66talent, 190–92, 252Tallal, Paula, 173, 185tape recordings, 3Tchaikovsky, Pyotr Ilich, 36, 52, 67, 205“The Teddy Bear’s Picnic,” 225“Teenage Lobotomy,” 83tempo. See also rhythmand categorization, 145defining, 15, 55–58and expectations, 169and infants, 222and musical memory, 150, 151, 153–54neural basis for, 59variation in, 59temporal lobesand expressivity in performance, 206function, 83, 125and metrical extraction, 169and music semantics, 124and neural processing of music, 128responding to stimuli, 89temporal positioning, 78tension and schematic violations, 116ten-thousand-hours theory, 193, 194“That’ll Be the Day,” 62–64themes, variations on, 144Thompson, William Forde, 5, 2203/4 time, 65, 66timbreand auditory-code readers, 119–20defining, 15, 18, 43–46dimensions of, 47–53of electric guitars, 13and expectations, 168expression through, 26and grouping, 78importance of, 50and musical preferences, 235neural basis for, 89recognition of, 146, 151–52in rock music, 50, 76soundscape, 152–53timing, 106, 170–71, 174, 178, 188, 202tone, 14–15, 16, 37, 44, 51, 145. See alsowhole stepstone deafness, 184tonic (first degree), 37Tooby, John, 8, 256top-down processing, 102–3training, musical, 207–8Trainor, Laurel, 222transposition, 73–74, 145, 160, 222Trehub, S. E., 222, 223tritone (augmented fourth), 13, 72, 223trombones, 28trumpets, 24, 44–45“Tryin’ to Do Something to Get YourAttention,” 163tubas, 24, 26tuning, 28, 32–33“Turning the Tables” illusion, 97

314 Index“Twinkle, Twinkle Little Star,” 60twins studies, 196–99“Twist and Shout,” 155U2, 5ubiquity of music, 5–6unconscious inference, 103Ungerleider, Leslie, 159unison interval, 72Van Halen (group), 111Van Halen, Eddie, 138, 208Varèse, Edgard, 14Vaughan, Stevie Ray, 111ventral striatum, 185vermis, 180vibration, 39–42. See also frequencyVincent, Gene, 153Vines, Bradley, 174, 206violins, 24, 44–45, 235vision, 140–41visual art, 17visual cortex, 84, 181, 264vocabulary of music, 10, 18, 19. See alsolanguagevoices, 24, 29, 43, 134–36, 235–36vulnerability, 236–39Wagner, Richard, 237“Wake Up Little Susie,” 153“Walk This Way,” 58waltz time, 59–60, 65, 66Wanderley, Marcelo, 206Ward, Dixon, 146, 148Waring, Clive, 125Warner Bros., 37Warren, Richard, 99Watson, Doc, 199Watson, James, 171, 259Watts, Alan, 140wave guide synthesis, 47wavelengths, 112Wernicke’s area, 82, 84Wertheimer, Max, 73–74Western musickeys in, 70meter of, 59–60note durations, 61–62preferences for, 221scales of, 28, 34, 36schemas of, 114and social consequences, 226West Side Story, 56What Mad Pursuit (Crick), 175–76White, Benjamin, 133–34, 145White, Norman, 10white matter, 221The Who, 69whole notes, 61whole steps, 30, 31Williams syndrome (WS), 182–83, 212,253–54William Tell Overture, 56wind instruments, 51The Wisdom of Insecurity (Watts), 140Wittgenstein, Ludwig, 137–38, 139Wonder, Stevie, 32, 53, 166, 167, 205, 208“Wonderful Tonight,” 49, 50woodwind instruments, 31“Would You Like to Swing on a Star,”225Wundt, Wilhelm, 77–78Yamaha DX9 and DX7, 48Yes, 145“Yesterday,” 110, 115yodelers, 79Young, Neil, 142, 207, 237, 238“You Really Got Me,” 111Zappa, Frank, 166Zarin, Michelle, 176Zatorre, Robert, 160, 169, 185Zoloft, 121

file0_1490

Create successful ePaper yourself

Delete template?

Save as template?