Your Brain on Music: The Neuroscience of Listening and Playing
Your Brain on Music: The Neuroscience of Listening and Playing
Much of what you hear in music isn't in the sound. In eight lectures, find out how your brain builds a note, predicts the next one, supplies the beat and answers its own predictions with a reward, and what playing an instrument does and doesn't do to that machinery. Along the way: a cockatoo who dances, a pianist whose hand stopped obeying, and famous findings that shrank when someone checked.
Sign up free to unlock:
- Resume-where-you-stopped listening
- Request & vote on new courses
- Save courses for later listening
- Get personalized recommendations
Already have an account? Log in
Chapters
Click play to listen, or tap a chapter to read its transcript.
1Introduction
In 2001, ten students at McGill University had their brains scanned while they listened to music. Each brought a piece that reliably gave them chills. One brought Barber's Adagio for Strings. For another student in the study, that same Adagio was the dull piece. One listener got chills, and the other was left cold.
That puzzle points to the idea under this whole course. Much of what you hear in music isn't in the sound. Your brain builds the pitch, predicts the next note and helps supply the beat, and its reward system answers those predictions. We'll see how it does each of those things, and what playing an instrument does to the machinery.
By the end, you'll be able to explain how your brain turns vibrating air into a note, a tune you expect and a beat you feel. You'll be able to say what happens when music gives you chills. And you'll be able to judge the big claims made for music: for practice, for lessons and intelligence, and for brains hurt by Parkinson's disease or a stroke.
We start with listening. The first lecture asks what your brain hears when it hears a note. The second asks whether there's a music center in the brain. The third and fourth follow your brain's predictions, from the next note to the chills. The fifth is about the beat, and why it pulls you to move. Then we turn from listening to what music can change. The sixth lecture asks what practice does to a brain. The seventh asks whether music makes you smarter, and the eighth what music can do for a damaged brain. Along the way, several famous findings shrink when someone checks them.
2The Note That Isn't There
A laptop's little speaker can't make deep sounds. Play a cello through it, and the lowest frequency of its low notes never comes out. That lowest frequency is the one that matches the note's pitch. So what should you hear? By rights, the cello ought to sound wrong. Its notes should jump, or the tune should come apart. Neither happens. The sound gets thinner, and you still hear the right notes in the right order.
A deep voice on the phone does the same trick. The phone doesn't carry the lowest frequency of the voice either, and you still hear exactly how high or low the voice is. You do this every day without noticing.
The frequency that matches the pitch never reaches you, and you hear the pitch anyway. Something in the sound that does arrive is enough. That puzzle is where we begin, and the question behind it is a plain one. What is your brain actually hearing when you hear a note?
We'll answer it in two halves. The first half takes a note apart: what your ear does to it, and a quarrel in the 1840s over a missing frequency. The second half puts it back together: how your brain builds a pitch, and why remembering an exact pitch is more common than naming one.
Start with what reaches you. Each eardrum moves to and fro with tiny, fast changes in air pressure. That's all your brain gets from the air. In a busy restaurant, the voices, the music and the plates all arrive at each eardrum as one single wave. Everything you hear has to be pulled out of that wave.
One of the most influential accounts of what's inside that wave came from a man drawn to sound by its mathematics, not its music. Georg Simon Ohm is the Ohm of Ohm's law, the rule for electric circuits he published in 1827. In the summer of 1843 he published a paper on what a musical note is. In it he used a piece of mathematics from the study of how heat spreads. It breaks any repeating wave into a sum of simple waves, each with a single frequency.
Ohm applied it to the sound of a siren. A siren in his day was an instrument for experiments on sound: a spinning disc with holes in it. Air blown through the holes came out as a fast string of puffs, and the faster the puffs came, the higher the pitch. Ohm concluded that a musical note is a mixture of simple waves. He said the pitch belongs to the lowest of them, and that this lowest wave has to be physically there in the sound. Ohm ran no experiment of his own. He worked entirely from another man's published measurements.
In the 1850s and 1860s, Hermann von Helmholtz carried Ohm's idea into the body. He proposed that the inner ear does Ohm's mathematics for real, splitting every sound into its simple waves. His theory gave each pitch its own place in the ear.
Helmholtz was right about that first step: the ear does split the sound. A long membrane runs through your inner ear, and it works like a prism. A prism spreads light into its colors, and the membrane spreads a sound into its frequencies, each to its own place along the membrane. Every point answers best to one frequency. High frequencies peak near the end where sound comes in, and low ones toward the far end. The membrane changes in thickness and stiffness from one end to the other, so each stretch is tuned a little differently. This has been measured directly, on human ears after death.
Tens of thousands of nerve fibers carry the result to the brain. Each one is wired to a short stretch of the membrane, and is deaf to most of the rest. The layout survives the trip, with the order from low to high kept through the brainstem and right up to the hearing cortex. That's the part of the brain's outer layer where sound first arrives. Call this layout the place map. It's the basic way your hearing is organized, and we'll lean on it from here on.
The fibers carry a second clue as well. They tend to fire at the same point in each cycle of the wave, at least for lower frequencies, so the timing of their signals follows the sound. That gives the brain two things to read: where on the membrane a sound lands, and when the fibers fire.
Now send a musical note through it. A violin playing the orchestra's A makes a wave that repeats 440 times a second. That rate is the note's lowest frequency, called its fundamental. The wave also carries energy at 880, at 1,320 and at 1,760. Those are two, three and four times the fundamental, and the series goes on up from there.
Those higher frequencies are the note's harmonics, stacked above the fundamental at whole-number multiples of it. Their mix is part of what makes a trumpet sound like a trumpet on the same A. On the place map, the violin's A arrives as a row of separate peaks, one for each of the lower harmonics. And yet you hear one note, with one pitch, at 440.
Ohm had the explanation ready. The 440 is in the stack, so the pitch is the 440. His rule was that the pitch is the lowest simple wave, and that it has to be there. It's a tidy rule. The man whose measurements Ohm had borrowed was not convinced.
That man was August Seebeck, an acoustician who worked with sirens. Four months after Ohm's paper, Seebeck replied. He described a siren with three spinning discs. It gave a clear pitch, and yet the frequency of that pitch didn't seem to be in the sound at all.
Seebeck would not accept a theory that went against his own two ears. He asked how else you could "determine what belongs to a tone, if not with the ear." His own idea of a note was broader than Ohm's. For Seebeck, any pattern of puffs that repeats regularly makes a note, whatever the shape of each puff.
Ohm couldn't answer on those terms. In 1844 he replied that he could not check the matter by listening, because "nature has altogether denied me a musical ear." So the man who had just defined a musical note said he couldn't judge one by ear. Ohm never published on sound again, and Seebeck died in 1849.
The verdict came from Helmholtz in 1856, two years after Ohm's death. He reviewed the whole dispute and found for Ohm. In time, he also put Seebeck's pitch down to distortion in the ear. Famous hearing scientists still held that view well into the twentieth century, and Seebeck's observations were almost forgotten.
They shouldn't have been. By the middle of the twentieth century, it was firmly established that a note keeps its pitch with its fundamental gone. You can predict the result before I give it. Go back to the violin's A, and filter out the 440 entirely. The lowest frequency left is 880, twice as high, so on Ohm's rule the pitch should leap up to match it. It doesn't. Listeners still hear the A at 440.
Look at what's left. The remaining harmonics still add up to a wave that repeats 440 times a second. Only the energy at 440 is gone. That's awkward for a theory where each pitch has its own place, because nothing in the sound now sets the matching spot on the membrane moving. This is called the missing fundamental: you hear a note's pitch even when its fundamental has been taken out of the sound. It works when the fundamental is drowned in noise, too.
That's the cello on the laptop, and the voice on the phone. It also makes sense for hearing to work this way. Suppose a pitch vanished every time noise covered its lowest frequency. In a noisy world, it would be little use. You still see a whole cup when a hand hides half of it. In the same way, you still hear a note when its bottom is missing.
So each side of the old quarrel had half the truth. Ohm and Helmholtz were right that your ear takes a note apart and lays its harmonics out along the place map. Seebeck was right that a pitch needs no simple wave of its own, only a pattern that repeats. That leaves the obvious question. Your ear has split the note into pieces, so what reads the repeating rate back out of them?
Helmholtz had an answer, and it deserves a fair hearing. He said distortion in the ear adds a new frequency at the difference between two others. The arithmetic works. The neighboring harmonics 880 and 1,320 differ by exactly 440. So maybe the ear quietly puts the 440 back.
A study aimed at a different theory answers it, I think. It came from Adrianus Houtsma, a pitch researcher at MIT. He played trained musicians two-note melodies, and asked them to name the interval, the musical step from the first note to the second. Each note was made of just two neighboring harmonics. In the notes made of higher pairs, there was nothing at the fundamental. The pitch they heard was the missing fundamental's. The musicians could do it. Then Houtsma split each note between the ears, sending one harmonic to the left ear and the other to the right. Now each ear got a single frequency, and nothing else.
Think about what that does to Helmholtz's answer. Distortion in the ear is real. It's made inside the inner ear when two frequencies reach it together, and here no ear got two. So Helmholtz's account needs the pitch to vanish. Instead, the musicians did about as well as when both harmonics went to both ears. Houtsma published the result in 1972 with a co-author, J. L. Goldstein. Houtsma later wrote that the findings were taken as strong evidence that the pitch of a missing fundamental must come from "some central, neural process." In other words, the brain. Neither ear had the pitch to give.
The same experiment showed which harmonics matter. The musicians did best when the two harmonics were the lowest ones, and worse and worse as the harmonics went higher. Many studies agree that the low harmonics set the pitch. They're also the ones the ear keeps apart. The low harmonics sit far enough apart that each one lands on its own spot. Higher up, the harmonics crowd so close that several share a spot and blur together. So the brain can get a clean reading of each low one.
This, then, is what your brain does with a note. It takes the separate low harmonics and works out the one repetition rate they all fit. The violin's A has low harmonics at 880, 1,320 and 1,760. Every one of them is a whole-number multiple of 440.
The wave they make repeats 440 times a second, and that's the pitch you hear, whether or not the sound has any energy at 440. How the brain reads those harmonics is still argued. It may go by their places on the map, by the timing of the nerve signals, or by both.
A bell shows the brain doing this with a real instrument. A tuned bell rings at several frequencies at once, and the note you hear matches a weak one called the Prime. Houtsma's group took a recording of one bell from a church carillon in Gouda, and shifted its frequencies one at a time. Retuning the Prime left the pitch alone. Shifting three strong frequencies above it moved the pitch. Those three sit at two, three and four times the Prime, like harmonics. The brain treats them as harmonics and hears their fundamental, which happens to sit right on the weak Prime.
Where in the brain this happens is disputed. A team recording from a few marmoset monkeys found neurons at the edge of the hearing cortex that answer to a pitch. They fire whether or not the fundamental is in the sound. One such neuron answered best to 182 times a second. It ignored each of the note's harmonics played on its own. It fired when they played together, with nothing at 182.
Babies are a puzzle of their own. Three-month-olds can pick out a changed note in "Twinkle, Twinkle, Little Star" with every fundamental removed. They were tested on a caregiver's lap in a sound booth. At that age the hearing cortex isn't fully developed, and the team that tested them thinks lower parts of the brain may be enough. But a study that recorded babies' brain waves found the cortex responding to the pitch of a missing fundamental only from four months, not at three. The two infant results don't agree.
Your brain builds a pitch from a pattern. A pitch the brain makes might also be one it keeps. Does it keep it, and can it name what it keeps?
In 1994 Daniel Levitin tested the first question with a shelf of CDs. He set out well-known pop and rock albums in front of forty-six students at Stanford. Each student sang two songs they knew well. For each, they held the album, closed their eyes and sang the song from memory. Pop songs suit this test because most exist in a single recording. People have heard them, maybe hundreds of times, always in the same key.
Levitin compared the first notes they sang with the notes on the record. A pure guess would land on the right note one time in twelve. With two songs, guessing would hit at least once about one time in six. Forty percent of the students hit the record's exact pitch on at least one of their two songs. Their misses bunched around the right note, too, and musical training made no difference.
Six labs later repeated the test with 277 people. The memory showed up again, but weaker, and it varied a lot from lab to lab. A quarter of them hit the exact pitch at least once, against Levitin's 40 percent. That's still above what guessing gives.
Now set that beside absolute pitch, what many people call perfect pitch. Give most musicians one note and its name, and they can name the notes around it by how far away they are. What most of them can't do is name a note heard on its own. That is absolute pitch: naming or singing a note with no other note to go by.
An anonymous letter to a newspaper in 1763 describes the seven-year-old Mozart doing it. He listened from another room and named each note as it was played. He could even name the note of a bell, a clock or a pocket watch as it struck.
It's often said that fewer than one person in ten thousand has absolute pitch. A team of researchers traced that number to a 1955 estimate. Its author had found 123 musicians with some absolute pitch around Chicago, and assumed there must be many more. He seems to have set that guessed number against the four million or so people in the region. Absolute pitch is rare in the general public. Nobody has a reliable count of how rare.
Levitin's idea is that absolute pitch is two abilities. One is pitch memory, holding on to the actual pitch of a sound. The other is naming, tying that pitch to a label like C-sharp. His shelf test suggests the memory is the more common of the two. Diana Deutsch studies music perception and has absolute pitch herself, and she sees the same puzzle. There are only twelve note names to learn, so the mystery is why so few people learn them. She compares the rest of us to people who can see colors perfectly well but can't put names to them.
The naming seems to come most easily to people who learn it young. Deutsch tested first-year students at two music schools. One was the Central Conservatory of Music in Beijing, and the other the Eastman School of Music in the United States. Look only at students who had started lessons at four or five. About 60 percent of them in Beijing passed her test, against 14 percent at Eastman.
The Beijing students all spoke Mandarin, a language in which pitch is part of a word. Say "ma" on a high, level pitch, and it means mother. Say it on a pitch that dips and then rises, and it means horse. Deutsch's hypothesis is that children like these are tying pitches to names from infancy. Others credit early piano lessons, and nobody has settled it.
Some of the naming can be learned later. In a 2025 study, a dozen adult musicians practiced naming notes online for about twenty hours. Every one of them learned to name at least a few notes quickly and reliably. Two learned all twelve, and one of those had taken up music only as an adult. It was a small study, with no comparison group.
So, back to the cello on the laptop. Its low notes reach you without their bottom frequency. What does reach you is a row of harmonics. Your inner ear lays them out in order along the place map, and that order is kept all the way to the hearing cortex. The nerve fibers carry their timing, too. From the lowest few, the ones the ear keeps apart, your brain works out how often the whole pattern repeats. That rate is the pitch you hear, and it's still there when the bottom frequency is gone.
The pitch your brain builds is also steady enough to keep. In Levitin's test, many people with no special training kept a song's exact pitch. Naming is a separate step, tying a pitch you hold to a word like A or C-sharp. It seems to come most easily to those who learn it young.
Two things are still open. Nobody knows yet exactly where in the brain the rate is worked out, or whether it's read from places or from timing. And nobody has settled why naming comes so easily to some children. Deutsch links it to speech, to languages where pitch changes what a word means. That raises a larger question, and it's the next lecture's. Does your brain treat music as a thing of its own, or as a kind of language?
If you keep one thing, keep this. The pitch of a note is a conclusion your brain draws from a repeating pattern. The frequency that matches it doesn't even have to be there.
3Is There a Music Center in the Brain?
In 1933 a French neurologist named Théophile Alajouanine began examining the composer Maurice Ravel. The neurologist was musical himself, and he added a piano to his tests.
By the neurologist's account, Ravel's ear seemed sharp. Parts of one of his own pieces were played to him with small mistakes, and he heard them and wanted the music played properly. He tried two notes on the neurologist's piano and could tell it was out of tune.
Then they asked him to play. He got through the opening bars of one of his own pieces almost perfectly, and couldn't finish. He managed only two or three notes of a piece he didn't know. He could barely read music, and writing it down was slow and full of errors. His speech and his writing were failing too. The neurologist's description of the case begins, "Maurice Ravel was struck down by an aphasia." Aphasia is a loss of language after brain damage. Ravel composed nothing more, and he died in December 1937.
Ravel lost words and music together. Was his music just another kind of language, failing along with the rest? Put bluntly, is there a music center in the brain?
Ravel's own case can't settle it. Losing two abilities at once proves nothing about whether they share machinery, because a single injury can damage two separate systems. And nobody knows for sure what Ravel's illness was. An operation on his brain days before he died found no tumor, and his family didn't allow an autopsy.
What his case does show is a split inside music. He lost the making of it: playing it, reading it, writing it down. He kept the hearing of it: recognizing it, judging it, remembering it. These tests were made before modern tests for musical disorders existed. Still, Alajouanine called Ravel's musical thinking "comparatively well preserved." For music against language, we need other evidence.
It comes from two directions, and at first they disagree. People with damaged brains show music and language coming apart. Brain scans of healthy people seem to show them in the same places. We'll take the patients first, then the scans. Next we'll look closer than an ordinary scan can, at populations of neurons that answer almost only to music, and one that answers to song. Last come the researchers who say most of it is shared, even up close.
Two words need keeping apart. Language, here, means words and grammar, which is what aphasia takes away. Speech means the sound of someone talking.
Showing that two abilities run on separate machinery takes a pair of cases. In one person, the first ability is lost and the second is kept. In another person, it's the other way round. That two-way split is called a double dissociation. For music and language, the pair exists.
Isabelle Peretz is a psychologist in Montreal who studies music and the brain. A review by Peretz gathers the cases. In one, an opera singer had brain damage and could no longer sing intervals, the steps between notes. The same singer still spoke with normal expression, and with normal intonation, the rise and fall of the voice. The reverse is common. Many people with aphasia can no longer talk, and yet they can sing a familiar tune, and some can learn a new one. Peretz reads the evidence as showing that the tune survives while the words fail, sung or spoken. Peretz also notes that patients who keep some words when they sing are reported more often among English speakers than French ones. A later lecture comes back to singing and speech.
There's an obvious objection. Most of us are amateurs at music and experts at talking, so damage might just wreck the weaker skill first. Peretz answers with the patients who sing. The objection predicts that talking, the expert skill, should be the one to survive. Yet the common case runs the other way, with talking lost and singing kept.
A team in Finland tested the split in 97 stroke patients. Within three weeks of each stroke, the patients heard pairs of short piano melodies and said whether the two in each pair were the same. In half the pairs, a note or a rhythm had changed. Their language was tested too.
About two-thirds of the patients had amusia, a failure to make sense of music. Forty-five of those had no aphasia. Fifteen other patients showed the reverse. They had aphasia, and they made sense of music normally. A stroke can take music and spare language, or take language and spare music.
Some people show the music side of the split from birth. Congenital amusia is amusia from birth, with no brain injury behind it. It's often called tone deafness. People with it miss a wrong note in a tune they haven't heard before. They can't tell when someone sings out of tune, themselves included. Many can't recognize a familiar song without its words. Yet they follow everyday speech normally, and they recognize people by their voices.
One person born with amusia described it to researchers in London: "When the music finished, the sound was always gone—as though it had never happened." How common is it? Peretz's lab in Montreal tested more than sixteen thousand adults online. By a strict measure, one and a half percent were born with amusia.
You might call yourself tone-deaf because you can't sing in tune. You probably aren't, though. The London group notes that people who say that rarely have amusia. Their trouble is with singing, and their hearing of music tends to be normal.
Damaged brains and brains that were different from the start agree, then. Music and language can come apart, in either direction.
Healthy brains in a scanner seemed to say the opposite. These scanners track blood flow as a stand-in for the activity of neurons. For years, scans showed music lighting up areas known for language, and speech lighting up areas that preferred music.
By 2014 the idea of shared machinery had caught on, and it carried a hope. Shared neurons would mean that training one skill might help the other, and music lessons might help children learn to read. A later lecture weighs that hope. Peretz's team counted the papers on music and language overlapping, year by year, and found the number climbing for a decade. They wrote that much of that work had no direct comparison behind it.
Their example was a brain-scan study of professional musicians. It found a stronger response to rhythm in musicians than in non-musicians, in a region usually tied to grammar. Its authors concluded that musicians seem to handle rhythm with the brain's resources for grammar. The study never tested grammar.
Peretz's answer is that overlap doesn't mean sharing. A brain scan is built of tiny dots. Each dot is the smallest piece of brain the scanner can measure, and it holds hundreds of thousands of neurons. What the scanner reads from a dot is their combined activity.
Say you hang one microphone over a section of a stadium, where home and away fans sit mixed together. The home team scores, and the microphone picks up a roar. The away team scores, and it picks up a roar again. By the microphone's account, the section cheers for both teams. It has no way to tell you that the section holds two crowds, each cheering for its own side.
A dot that lights up for music and for language is that microphone. It might hold one set of neurons doing both jobs. It might hold two sets side by side, each with a job of its own. From the dot alone, the scanner can't tell which.
The old scans had a second problem, and the lab of Evelina Fedorenko at MIT went after it. Most scans averaged many people's brains together. But each person's language areas sit in a slightly different place. Averaging can smear two neighboring areas into one blur, even when they never overlap in any single person. So Fedorenko's group found each person's language areas first, from how they responded to reading sentences. Then they played that person music.
In 2011 the language areas barely responded to music. In four more experiments in 2023, their response to music was generally low, often lower than when nothing was playing at all. Songs drove them no more than the same words spoken. And three people with severe aphasia, who couldn't tell whether a sentence was grammatical, judged melodies well.
Looked at person by person, then, the language areas leave music alone. That fits the patients. But in the brain's hearing areas, in and around the hearing cortex, music and speech still seem to share dots. Both pictures could still be true there, if each dot holds more than one crowd. The way to find out is to look inside the dots.
In 2015 another team at MIT found a way to do it. The study was led by Sam Norman-Haignere, then a postdoctoral researcher there. The team put ten non-musicians in a scanner and played them 165 everyday sounds, two seconds each. Among them were a ringing telephone, footsteps and a car engine starting up, with many kinds of speech and music mixed in. Nobody asked the listeners about music. Their only task was to press a button when a repeated sound came out a little quieter.
Norman-Haignere's team didn't ask each dot whether it preferred music. They treated every dot as a blend of a few underlying populations, and let the mathematics find them. A population, in this sense, is a group of neurons that all answer to the same kinds of sound.
The stadium shows how that can work. One microphone can't tell you there are two crowds. Thousands of microphones can, if each one hangs over a different mix of fans. The home team scores, and some microphones get much louder than others. The away team scores, and a different set gets louder. From those differences, you can work out that there are two crowds, and what each one cheers for.
The team had thousands of dots, each a different blend. The analysis was never told which sound was music. Nobody saw these neurons directly, either. The populations are what the mathematics says must be in the dots.
Six populations accounted for most of what the dots did. Four answered to plain features of sound. Two followed the last lecture's place map, the ear's layout from low sounds to high: one took low sounds and one high. That made a useful check: working blind, the method had found something everyone knew was there. A third took sounds with a clear pitch, and a fourth took sharp, fast changes in a sound.
The last two were new. Speech drove one of them, even in languages the listeners didn't know. Music drove the other. Of the thirty sounds that drove the music population hardest, all but two were music. The two exceptions were wind chimes and a ringtone, which at least have tunes. Measures of the raw sound couldn't explain its taste. Music chopped into tiny pieces and shuffled drove it less.
Why had nobody found a music population before? The stadium suggests a guess. In the dots most tuned to speech, the speech population supplied about seventy percent of the response. Speech had patches almost to itself, and ordinary scans had found them. In the dots most tuned to music, the music population supplied under half. Much of the rest came from the population tuned to pitch. The music crowd never had a section of the stands to itself, so no single dot looked very interested in music.
Had they found the music center, then? A warning came from Josh McDermott, one of the team's senior scientists: "It's inappropriate at this point to conclude that this is the seat of music in the brain." The team had looked mostly at the hearing areas.
Seven years later, Norman-Haignere's team looked closer still. Some people with epilepsy have electrodes placed on the surface of the brain before surgery, to show doctors where their seizures start. They wait in hospital for several days. Fifteen such patients agreed to hear the same 165 sounds. Electrodes on the brain give a much sharper picture than a scanner.
A song makes a neat test, because a song is speech and music at once. Suppose the brain handled a song with a speech population plus a music population. Then a song could draw no more than speech and music would draw added together. The analysis turned up a population that answered almost only to music with a singing voice in it. Every song drew a strong response from it. Speech and instrumental music drew little or none. And a song drew more from it than speech and instrumental music added together. The team thinks this population answers to the combination itself, a voice singing with music. Exactly what it does is still unknown. Norman-Haignere said, "This was a finding we really didn't expect."
The song population came from all fifteen patients' electrodes taken together. Seven single electrodes, in three of the patients, showed the same taste on their own, and the analysis still found it with those seven left out. One of the three patients had no musical training, so this isn't something only musicians have. Norman-Haignere said that to a scanner, the song and music populations are "so close that you can't disentangle them."
Not everyone reads these results as music having machinery of its own. Aniruddh Patel, who studies music and language, offers a different picture. In his view, language and music keep separate stores of knowledge, which is why a stroke can take one and spare the other. But they share a limited pool of resources for fitting each new word or chord into a structure as it arrives. That idea was built to square the patients with the scans.
In 2009 a team including Patel tested it. Ninety-six students in San Diego read sentences on a screen, one chunk at a time, while a chord played under each chunk. Some sentences set a trap, as in an example from Patel's paper: "The attorney advised the defendant was guilty." Most readers first take the defendant as the one being advised. The word "was" forces a rethink, and readers slow down there. They slowed down more when the chord under "was" came from a distant key. A word that was merely odd in meaning, like angry pigs where you'd expect angry dogs, didn't interact with the chords. Nor did a chord played on an unexpected instrument.
Fedorenko's group answers that if such a shared pool exists, it lies outside the language areas, in the brain's general machinery for hard tasks. Some studies have also found subtle trouble with music in people with aphasia, though that evidence is mixed.
The widest look came in 2024, from a team in Marseille. Eighteen people with epilepsy had thin electrodes running deep into the brain, again to find their seizures. Most of them were recorded in their hospital bedrooms. Each heard ten minutes of a spoken story and ten minutes of instrumental music. Between them, the electrodes reached most regions of the brain, well beyond the hearing areas.
The Marseille team called a response shared when a spot answered to the story and the music about equally. About seven in ten of the spots that responded at all were shared. No region of the brain was given over to speech or to music. The team concluded that speech and music mostly run on the same resources.
Put the two studies side by side, and ask each the same two questions. First, the sounds. The Marseille team argue that two-second clips are too short for music, whose phrases run far longer than speech's. Marseille, for its part, used one story and one piece of music. Second, the tool. A reviewer of the Marseille study pointed out that each of its electrodes takes in thousands of neurons. A small population mixed among others is what such a tool can miss. Even Norman-Haignere's single electrodes showed the song taste only weakly. I think the tool matters more than the clips. Short clips were enough to reveal a music population, and one story and one piece of music can't rule it out.
There's one more piece of evidence for sharing, and it comes from people born with amusia. Early tests found they could tell a statement from a question as well as anyone. "He speaks French," with the voice falling at the end, is a statement. The same words with the voice rising are a question. The London group made that rise smaller. With small rises, people with amusia struggled, in speech as well as in plain tones. Only two of the sixteen they tested had noticed any trouble in daily life, probably because faces and context fill the gap. So the music deficit reaches into the pitch of speech. Some of the machinery for pitch, at least, seems to be shared.
The stroke patients point the same way: more than a quarter of the Finnish patients with amusia had aphasia as well.
Here is how I'd put the two sides together. It's my reading. Much of what the brain does with a song, it also does with speech, from taking in the raw sound to tracking its pitch. Patel thinks some of the work of building structure is shared as well. Inside that shared machinery sit small populations that answer almost only to music, and one to singing. The Marseille team allow for exactly that. They write that their results and Norman-Haignere's can be reconciled if separate populations sit inside the same region. In 2015, Peretz and colleagues called the overlap between music and speech "an open question for the field." Nine years later, it still is.
Is there a music center in the brain? Not one you could circle on a scan. I'd say the question gets two answers, because it can be asked from two distances. From a scanner's distance, music shares a neighborhood with speech. Up close, some populations answer almost only to music, and one to song. How close you look decides whether music has a place of its own.
So the patients and the scans only seemed to clash. Strokes and amusia from birth pull music and language apart. Blurry averages and crowded dots push them back together. Look person by person, or inside the dot, and the music population appears, tangled among neurons tuned to pitch. Pitch stays common ground, and perhaps the building of structure does too.
One skill looks most like music's own: knowing which notes belong in a tune. Peretz points to one candidate for a music-only system, the one that fits pitches into a key. My guess is that Ravel kept that sense. He could still hear small mistakes in his own music while he struggled to play it. People born with amusia miss a wrong note in a tune they've never heard. Most people catch one without a single music lesson. That's where the next lecture begins. How does a brain that has never studied music know which note is wrong?
4Your Brain Is Guessing the Next Note
Sing a scale in your head, the plain one that goes do, re, mi. Climb it one step at a time, and stop on the seventh note, ti. Leave off the last one. You can feel it waiting. The tune leans toward that top do like a sentence missing its last word. You feel that pull whether or not you ever had a music lesson. Nobody sat you down and told you that ti leads to do. And a singer who ended the climb on a note from outside the scale would sound wrong to you.
The last lecture ended on a question about exactly this feeling. How does a brain that has never studied music know which note is wrong? Inside it sits a larger question. Is the wrongness of a wrong note in the sound, or in you? Some chords seem sweet and others sour, as if by nature. Does every ear on earth agree?
We begin with an old theory of musical feeling, and with babies who pick up the odds of sounds without trying. Between them, they explain where expectations come from. A computer program that bets on every next note puts numbers on those expectations. A CD mailed to new parents reveals how early the learning sets. And villages in the Bolivian rainforest give the question its hardest test.
Leonard Meyer had a theory about that pull. In 1956 he published a book called Emotion and Meaning in Music. His idea was that you feel something when a response you were leaning toward gets blocked. The scale that stops on ti is his kind of case. The top note you expect never comes, and what you feel is a small frustration. Meyer argued that the meaning music has within itself comes from expectation too. A note means what it leads you to expect.
Where do the expectations come from? Meyer named two sources. Some come from habits of perception that the mind brings to any pattern. One such habit is to expect a pattern to carry on as it began, so a line that climbs should keep climbing. Other expectations are learned from a particular style of music. Meyer was sure the learned kind mattered, and he doubted any account that left it out. The question under this whole lecture is how much is built in, and how much is learned.
The learned kind now has a name, statistical learning. It means picking up the odds of what you hear just by hearing it, without trying and without knowing you're doing it. Jenny Saffran is a psychologist who studies how babies learn. In 1996 Saffran and her colleagues played eight-month-old babies a stream of nonsense syllables, with no pauses. The syllables made up nonsense words, something like golabu, all run together. Inside a word, one syllable reliably led to the next. Across the join between two words, what came next was less predictable. That was the only clue to where the words were. Two minutes of listening was enough. Afterwards the babies could tell a made-up word from a string of the same syllables that ran across a join. Nobody taught them, and nobody told them when they were right. Three years later, the team ran the same test with short groups of musical tones in place of syllables. The babies learned those too.
Music gives this kind of counting plenty to work on. Take a key. A key is the family of notes a tune is built from, with one note as its home. In Western music, the notes that anchor a key turn up more often than the rest. They also tend to land on strong beats and at the ends of phrases. The do of your scale is the home note, and tunes tend to come to rest on it. Children probably learn keys from these odds. Babies of seven months already link particular pitches with strong beats. Western adults with no training spot sour notes that fall outside the key. Recordings of the brain's electrical activity show it flagging an unexpected chord even while they're busy with something else.
One set of odds can't serve every kind of music, though. The music scholar David Huron argues that we keep a separate set of rules for each style we know. One reviewer of his book gave the reason. Suppose your brain averaged everything you'd ever heard. Then Beethoven would set up much the same expectations as Britney Spears. He doesn't.
Meyer himself tied expectation to probability, in an article the year after his book. What was missing, until fairly recently, was a model that could learn the probabilities the way a listener does. In London, the cognitive scientist Marcus Pearce built a computer program that does. It learns the way Saffran's babies did, by counting. It was fed 903 folk songs and hymn tunes. From them it counted which notes tend to follow which, over stretches of one note, two notes or more. The program gives each possible next note a chance before it arrives.
Those chances give two numbers, and this course will keep both. Treat the program as a gambler, spreading chips across the possible notes. Uncertainty is how thinly the chips are spread before the note arrives. Uncertainty peaks when every note looks as likely as every other. It's low when one note is the obvious bet. Surprise comes after, when the note arrives. It measures how little was staked on the note that came. Uncertainty belongs to the moment before a note. Surprise belongs to the note itself.
Your scale shows both. Stop on ti, and a gambler who knew your music would pile the chips on do, so uncertainty is low. A singer who then lands outside the scale lands where almost nothing was staked. That's high surprise.
The gambler is more than a figure of speech. In one experiment, fifteen people played a real betting game with a hymn tune. Before each note, they staked play money on which of twenty pitches would come next. They kept betting until they got it right. A bigger stake on the right note won more, so their winnings showed how expected each note was. The program was tested on the same tune. Across studies of several kinds, from betting games to brain recordings, its surprise accounts for up to 83 percent of the differences in how strongly listeners expected notes.
The program can also be checked against Meyer's own ear, on a tune from Schubert's Octet. The case turns on an octave. Two notes are an octave apart when one repeats twice as fast as the other. The violin's A at 440 and the A at 880 above it are an octave apart. Western music gives both the same letter. In another book, Explaining Music, Meyer analyzed the tune. In its second bar, the melody moves through the notes of a chord, and the next note should be a high G. It is a G, but an octave too low. Meyer wrote that the continuation was "in the wrong register," the right note at the wrong height. The program has never read Meyer, and it agrees. It gave the high G nearly nineteen chances in a hundred. It gave the low G that Schubert wrote fewer than two.
The brain keeps the same score. One team recorded listeners through electrodes on the scalp while they heard 58 hymn tunes. About a tenth of a second after each note, the brain's electrical response dipped. The more surprising the program found the note, the deeper the dip.
Pearce himself gives the caution that goes with all of this. A program that fits Western listeners shows that counting could explain what they hear. It doesn't show that counting is how they learned it. That takes an experiment that changes the listening itself.
That answers the last lecture's question, at least in outline. A brain that never studied music has been studying all along. It counts what follows what in the music around it, and it bets on every next note. A wrong note is one it gave long odds. What the program can't tell us is when the counting happens, or whether it reaches what sounds pleasant. Does any of right and wrong come with the ear itself?
When does the counting start, and when does it set? On listening tests, the sense of key takes years. Babies of eight months notice when one note in a short tune is changed. To them, a note that breaks the key is just another change. Adults, with or without training, pick out the key-breaking note more easily. North American children do so by age five. By six or seven, they also catch a note that clashes with the chord beneath it. Brain recordings find some sense of harmony earlier still.
Rhythm narrows sooner on these tests, and the psychologists Erin Hannon and Sandra Trehub tracked it in babies. Western dance tunes mostly keep their beats evenly spaced. Many Balkan dance tunes don't. Their bars run to seven quick counts, grouped three, two, two. Earlier work had found that North American adults notice a broken Western rhythm but miss a broken Balkan one. Adults from Bulgaria and Macedonia notice both, and so do North American babies of six months. In 2005 Hannon and Trehub tested babies of twelve months. Each bar of the test tunes got one extra note. In one version the rhythm made room for it. In the other, the extra note broke the rhythm. Place your own bet before the answer. The twelve-month-olds noticed a break in the even, Western kind of rhythm, and not in the Balkan kind. All the test tunes were Balkan folk tunes, some with even rhythms and some with uneven ones. In six months, they had come to hear like North American adults.
Then Hannon and Trehub ran the kind of experiment Pearce's caution calls for. They changed what the babies heard. Two weeks before a lab visit, they mailed families a CD, with instructions and a log sheet. It held about ten minutes of dance music from Macedonia, Bulgaria and Bosnia, and none of the test tunes. Parents were asked to play it every morning and every afternoon, while the baby was alert and content. They were asked not to draw attention to it. After two weeks, the twelve-month-olds noticed a broken Balkan rhythm as well as other babies noticed a broken Western one.
Forty college students got their chance too. Half of them took the same kind of CD home for one or two weeks. They could earn up to ten dollars for recognizing its tunes afterwards. They improved a little, but not enough to beat guessing. Adults in the study even rated the broken Balkan version as closer to the original, because it fit a Western pattern. You might suspect the students just listened less. But how much each one listened made no difference to how much they improved. The researchers add that money was at stake for the students, so they were likely listening on purpose. The babies heard the music in the background.
Hannon and Trehub's reading is that years of Western music leave adults committed to its patterns. A foreign rhythm gets bent to fit a familiar one. A baby's patterns are still loose enough to reset. Two weeks were enough for a baby, and not for an adult. Rhythm and key were measured by different tests, though, so nobody can yet say which truly comes first.
Rhythm is one thing. Surely some notes just sound good together. The followers of Pythagoras thought so, and credited simple numbers. Modern theories look to the ear instead. An octave's two notes repeat at two to one, and a fifth's at three to two, like C with G. Western listeners call pairs like these consonant, meaning pleasing together. They call pairs without such simple ratios dissonant, and C with F-sharp is one of those. From here on, call them sweet pairs and grating pairs. If the liking were built into the ear, every ear should share it. The test needs listeners who have barely heard Western music. The last lecture introduced the MIT scientist Josh McDermott. He names the difficulty: "It's pretty hard to find people who don't have a lot of exposure to Western pop music."
An anthropologist who had long studied the Tsimane asked McDermott to help test how they respond to music. The Tsimane are farmers and foragers in the Bolivian Amazon, and most have heard little Western music. Their music is usually made by one person at a time, so they rarely hear two notes at once. In 2016 the team reported what Tsimane listeners made of sweet and grating pairs. On average, they liked the grating pairs as much as the sweet ones.
Bolivians in a nearby town and in La Paz showed a small preference for the sweet pairs, and Americans the largest. Did the Tsimane misunderstand the question? The checks suggest not. They rated laughter above gasps, like everyone else, and they disliked roughness as much as anyone. Roughness is the fast flutter in loudness when two tones are close in pitch but not the same.
Liking a sound and hearing it are different questions. In August 2019, McDermott's lab went back to ask what the Tsimane hear. They chose villages with no electricity or phone signal. Some were two days' walk from the nearest town, and others two or three days by motor canoe. Each listener wore closed headphones, with a translator alongside. Two notes sounded together. Was it one sound or two? And how much did the listener like it? Thirty-one Tsimane took part, along with twenty-eight Bostonians with no musical training.
Both groups heard the sweet pairs as one sound more often than the grating ones. Both heard the octave as one sound most often of all. Hearing two notes as one is called fusion, and this trip was the fusion study. But only the Bostonians liked the sweet pairs better. The lab also tested a hundred more North Americans online. Even in them, fusion and liking came apart. A person who fused the sweet pairs strongly was no more likely to like them strongly.
Why would sweet pairs fuse? The answer builds on the first lecture. Notes from voices and from instruments like the violin are stacks of harmonics, at whole-number multiples of the lowest frequency. The brain groups frequencies that fit one stack into one sound. That's how it tells two voices apart, in the rainforest as in Boston. Take the violin's A at 440 and the A an octave up, at 880. The harmonics of the higher A run 880, 1,760 and on up. All of them sit in the lower A's stack already. Together the two notes look like a single note. A fifth fits one stack nearly as well, and a grating pair hardly at all.
The octave has one more lesson. Played together, its two notes fused for listeners in both places. Played one after the other, they may not count as the same note. In another study, McDermott's lab asked people to sing back pairs of tones, some far above the range of any voice. Americans tended to land on the same letter, a whole number of octaves away. The Tsimane didn't, not even their most accurate singers. The team tried coaching, with a recorded voice saying "Excellent!", "Good" or "OK" after each try. They left out "Bad", so as not to discourage anyone. It changed little. The lab reads this as a sign that hearing an octave as the same note comes from experience. Singing back isn't the same as hearing, though, and an older study found young French infants treating octaves alike.
The music researcher Imre Lahdelma set out the critics' case against the 2016 study of liking. First, it hadn't tested the octave, the sweetest pair of all. The fusion study did, and the Tsimane still showed no preference for the sweet pairs. Second, the Tsimane did prefer a fifth to one pair, the harshest in Western music. The 2016 paper hadn't discussed that. Third, the critics point to babies. Several studies found infants looking longer toward sweet pairs. A later study found no such preference, only one for music the babies had heard before.
Lahdelma's own complaint was that the 2016 study left out truly rough chords. So Lahdelma's team tested two remote peoples in northwest Pakistan, and listeners in Britain, on chords. Everyone disliked a dense cluster of neighboring notes. Only the British preferred the major chord, a standard sweet chord of three notes.
To my mind, the critics win one point. Every group in these studies dislikes roughness. So far, they haven't won the larger one, that we're born liking the sweet pairs. McDermott's lab suggests why the liking varies: familiarity. On that account, we come to like the pairs we hear most, and Western music is full of sweet ones. The lab has looked for that among the Tsimane themselves. Villagers with more contact with towns, phones and churches showed a small preference for the sweet pairs, and the others none. The gap between the two groups was too small to be sure of, and most of this work comes from one lab.
A brain with no lessons still has an education. Meyer put musical feeling and meaning in expectation, the lean toward a note that may not come. Statistical learning supplies the lean. You absorb which notes follow which, as babies find words in a stream of syllables. A program that counts the same way scores every note twice. Uncertainty is how widely it spreads its bets before a note. Surprise is how little it put on the note that landed. On this evidence, a wrong note is mostly a long shot. Rhythm narrows to your own music within the first year. Key takes until about five, on the tests so far. And the ear's share is smaller than it feels. Fusion followed the same pattern in Boston and the Amazon, and so did a dislike of roughness. Liking the sweet pairs did not. Nor did singing an octave back as the same note, which Americans did and the Tsimane didn't.
The gambler still has to be paid. Huron argues that the brain rewards a correct prediction, so we feel good when we're proved right. If he's right, a good bet should pay out somewhere in the brain. The next lecture follows the payout, into what happens when music gives you chills.
How much of the liking is built in is still argued, and both camps say the chance to settle it is shrinking. Between 2019 and 2021, the Tsimane village of Mara got solar panels and phone signal. The Tsimane hear a sweet pair blend into one sound, as Bostonians do. What may be changing, McDermott's lab thinks, is whether they like it.
5Why Music Gives You Chills
Some people hear music perfectly well and get little pleasure from it. They can tell you that a piece is meant to be sad, or happy. They don't feel much of it. Robert Zatorre is a neuroscientist at McGill University in Montreal, and he helped describe them. He told the CBC that many of them try to hide it. His example was a party with music blaring. They'd go along with it, he said, and "try to pretend that they liked it as much as everyone else did."
Most people get something from music that these listeners don't, and sometimes it shows on the skin. A favorite piece reaches a certain moment, and a shiver runs down your spine, or you get goosebumps. These are chills. For a given listener, they often come at the same spot in the same piece. Not everyone gets them.
The people at the party hear the same notes you do. Something happens in you that doesn't happen in them. What happens in your brain when music gives you chills? And why should a run of notes feel good at all? Food keeps you alive, and money buys food. Music's use, if it has one, is far less obvious.
The last lecture left a gambler waiting to be paid. It argued that your brain bets on every next note, by odds it picked up just from listening. David Huron, the music scholar from that lecture, argues that the brain rewards a correct prediction. If he's right, something in the brain should be doing the paying. This lecture goes looking for it.
The search takes four steps. The first catches a chill in a brain scanner, to see what answers it. The second winds back fifteen seconds, to something that starts before the chill does. The third pushes on that machinery, with drugs and with the people from the party. The fourth asks what the machinery is paying for, and whether pleasure is even the right word for what it does.
A chill is handy for science because you can time it, and the body shows it. Researchers treat a chill as a marker: a burst of the body's arousal at a peak of pleasure. Chills aren't required. In one scan, pleasure without chills moved the same brain regions, only less.
This is the study the course began with. In 2001 Anne Blood and Zatorre scanned ten students at McGill, using a machine that tracked blood flow in the brain. All ten were trained musicians, chosen because certain pieces gave them chills reliably. Each brought one classical piece. Among them were Rachmaninoff's Third Piano Concerto and Barber's Adagio for Strings. Each student's chill piece doubled as another student's dull one, so only the listener changed. The Barber that gave one student chills was dull to the student who brought the Rachmaninoff.
With their own pieces, the students reported chills in 77 percent of the scans. With somebody else's, never. Blood flow rose in several regions as the chills grew stronger, and fell in others, one of them tied to fear. Among the regions that rose were parts of the reward system. It's an ancient circuit. Its job is to make an animal repeat what keeps it and its kind alive, such as eating, sex and caring for young. In people it answers to money as well. A small region near its center is called the nucleus accumbens. I'll call it the reward hub. One region that rose sat right where the hub is, though this scanner was too coarse to be sure it was the hub. Blood and Zatorre noted that food, sex and drugs like cocaine work the same regions, though that says nothing about how strongly.
But blood flow can't name the chemical at work. The reward system runs partly on dopamine, a chemical messenger. Dopamine released there can make an animal more likely to repeat what it just did. Valorie Salimpoor worked with Zatorre at McGill, and she set out to see whether music releases it. She screened people who answered advertisements. Each sent in pieces that gave them chills, from classical and jazz to punk, techno and tango. She kept the ones whose chills could be counted on. Their chills had to come at the same point every time, and not because the music brought back a memory.
Her scan used a radioactive tracer that competes with dopamine for the same docking points on nerve cells. Released dopamine crowds the tracer out, so a weaker tracer signal means more dopamine. The scan can only see the deep core of the reward system, where the hub sits. There, their own music released more dopamine than neutral music did. The biggest changes were in the hub and in the region just above it, called the caudate. Two people's scans were discarded, so the result rests on eight. A run of notes, then, can reach the chemistry that food and money reach.
Salimpoor's next question was about timing. Her team scanned the same people again, with a machine that follows blood flow from moment to moment. This one can't see dopamine. Each listener pressed a button whenever a chill came. The team watched only the spots that had released dopamine in the first scan. They asked what those spots were doing in the run-up, the fifteen seconds before each chill. That window was set in advance, from earlier work showing that the body's arousal starts climbing about then.
The obvious guess is that the reward system is busiest at the chill itself. The hub did peak at the chill. But the caudate rose earlier, and more, during the run-up, then dropped back once the chill arrived. The hub rose in the run-up too, and kept climbing. The first scan fit this pattern. People who had more chills, and so more run-ups, had released more dopamine in the caudate. People whose chills were stronger had released more in the hub. Part of the reward system works while the moment is still on its way.
There's a reason a reward system would get to work before the reward. Much of its job is prediction. It learns which signals come before something good, and those signals become rewarding in themselves. Its response also grows when an outcome turns out better than expected. Salimpoor's team suggests that the notes leading up to a peak can work that way. They tell the brain the best part is near.
The team called the two stages anticipation and the peak. They suggested the two might match what reward researchers call wanting and liking. Wanting is the pull toward a reward that's on its way. Liking is the pleasure when it arrives. The team pointed out that composers and performers work the anticipation hard. They hold back what you're waiting for, with an unexpected note or a slower tempo, to make you want it more. In the last lecture's terms, the caudate works while the chips are on the table, and the hub when the moment arrives.
These listeners knew their pieces well, though, so they knew a chill was coming. Their run-up might come from memory of one piece, or from a feel for how music in general goes. The team said later experiments would have to tell the two apart.
So far, then, the reward system answers to music. Dopamine is released in it, and part of it works in the run-up. All of this is watching, though. A system that lights up with the pleasure might only be along for the ride.
To show the reward system makes a difference, you have to push on it. A group in Barcelona did that in a drug study, with Zatorre on the team. Healthy volunteers came to a drug-research unit at a Barcelona hospital three times, at least a week apart. Each time they swallowed a single capsule. One held levodopa, a drug that nerve cells turn into dopamine, with a second drug that slows its breakdown outside the brain. One held risperidone, which blocks some of the places where dopamine docks. One held milk sugar. Neither the volunteers nor the researchers knew which was which. An hour later the volunteers heard pop songs and favorites of their own, rating their pleasure with buttons, from none up to chills. They could then buy each pop song with their own money, at one of four prices from nothing to a euro and 29 cents.
Some volunteers got chills even on the milk sugar. Compared with the sugar, those listeners spent 65 percent more time in chills on levodopa, and 43 percent less on risperidone. Their skin's response at moments of high pleasure rose and fell the same way, and so did their bids. They offered more for the songs on levodopa than on risperidone. The firmest result was the gap between the two drugs. Their other judgments held steady. The drugs didn't change how happy or sad the music seemed, or how calm or rousing, or how familiar.
These were single doses, given to healthy volunteers for a study. They say nothing about people who take these medicines for an illness. Risperidone also blocks receptors for serotonin and other messengers. So I'd hold open the chance that part of the drop on risperidone wasn't dopamine's doing. Still, the authors say no study before had changed dopamine directly and shown musical pleasure following. Earlier work had mostly watched. This one pushed, and the pleasure moved. The ratings people gave after each song moved the same way, though too weakly to be sure of.
The people from the party come next, and a 2013 study from Zatorre's lab shows why they matter. People heard music they had never heard before, and could bid to buy it. The hub was busier for the pieces people bid more for. The brain's hearing areas were busy for every piece alike. These are the areas in and around the hearing cortex. What changed was how closely they rose and fell with the hub. For pieces worth more to someone, the two moved together more closely. Zatorre and Salimpoor think the hearing areas hold templates, stored patterns of how music tends to go, built from a lifetime of listening. They suggest that some of that knowledge comes from the statistical learning of the last lecture, the brain's silent counting of what follows what. In their view, the hub sets a value on what those templates predict.
That line from hearing to reward suggests a test. Somewhere there should be people in whom it's weak. The Barcelona group found them. They gave a questionnaire about music and pleasure to more than a thousand university students, and picked ten of the lowest scorers. They compared them with ten average listeners and ten keen ones. The low scorers weren't depressed, and they enjoyed other things normally. They did as well as anyone on a test for amusia, the failure to make sense of music from the second lecture. They could tell happy music from sad, scary or peaceful music. In the other listeners, heart rate and skin response climbed with their pleasure. Zatorre said the low scorers' responses "more or less flatlined." In a game for money, their bodies responded like anyone's. The group named the condition specific musical anhedonia. Anhedonia is a loss of pleasure, and here the loss is for music alone.
It's found in people who are otherwise well, and a year later the low scorers' ratings held. That's different from losing pleasure in music you used to love. The US National Institute of Mental Health lists "loss of interest or pleasure in hobbies and activities" among the signs of depression.
Later the group scanned fifteen people of each kind. In all three groups, the hearing areas answered music the same way. In the low scorers, the reward hub answered music weakly, and a gambling game normally. Their right-side hearing areas also moved less closely with the hub than in the keenest listeners. Their hearing works, and their reward system works. What seems weak is the line between them.
One caution covers these studies. Most are small, from eight people to forty-five. All of them come from one circle of labs working together, in Montreal and Barcelona. Other labs have also found the reward system answering to music, but the dopamine and anhedonia work is that circle's. Together they point to a reward system that matters to the pleasure, working in step with the hearing areas.
What is the reward system paying for? Here the last lecture's program comes back, with its two numbers. Uncertainty is how thinly the program spreads its chips before a note. Surprise is how little it staked on the note that came. An old idea says people like things best at a middling level of complexity. Too simple bores, and too tangled baffles. On that idea, enjoyment plotted against complexity makes a hump, an upside-down U, with a sweet spot at the top. A 2017 review of 57 studies counted most of them as compatible with some form of the hump. Benjamin Gold and colleagues at McGill read the same review more harshly. By their count only 15 showed the hump, and each study had measured complexity its own way, so they called the evidence mixed.
Gold's team used the program instead. Zatorre was on the team, and so was Marcus Pearce, the cognitive scientist in London who built it. The program scored real melodies, played one note at a time on a computer flute, for their average uncertainty and average surprise. Listeners rated how much they liked each melody they didn't recognize. Enjoyment rose and then fell. People liked best the melodies in between, and mostly liked the predictable ones more than the least predictable.
A separate team, which also included Pearce, took the idea down to single chords. Its copy of the program learned the chords of hits from the Billboard Hot 100. The team took chord sequences from those songs and stripped out the melody and rhythm. They moved them all into C major, so nobody could recognize a song. Listeners rated each chord's pleasantness on a slider as it played. Pleasure depended on uncertainty and surprise together. A surprising chord pleased when the next chord had been easy to call, as near the end of a section. An unsurprising chord pleased when the next one had been hard to call, as in an unusual sequence.
In the gambler's terms, people enjoyed a confident bet that lost. The chips were piled on one chord, and a different one came. They also enjoyed an uncertain bet that came in. The chips were spread thin, and one of the likelier chords came. Gold's team read their melody results as a reward for learning. Confirming a shaky guess teaches you something. So does losing a confident one, if the chord that came still makes sense. A dead certainty teaches nothing, and neither does noise. The last lecture's guess was that the brain rewards a correct prediction. On Gold's reading, it also rewards a mistake you can learn from.
The chord study had a second half, in a scanner. The pattern of pleasure showed up in the hearing areas, and in regions tied to emotion and memory. The reward hub didn't follow surprise at all. It followed only uncertainty, working harder when the next chord was hard to call. The team read that as the hub supplying the wanting, the pull toward whatever will settle the uncertainty. That moves the hub to the other side of Salimpoor's two stages. Her team had suggested its peak might be the liking. This reading makes it the wanting, and the reading has a history.
In 1989 the psychologist Kent Berridge and his team expected that wiping out nearly all of a rat's dopamine would dull its pleasure in sugar. They read that pleasure from the rat's face. Rats relax their faces and move their tongues in a rhythm at a sweet taste, and gape at a bitter one. The rats with almost no dopamine stopped eating on their own altogether. Yet their faces at a taste of sugar were completely normal. Berridge concluded that dopamine drives wanting, not liking. On their account, the liking comes from tiny hotspots worked by other chemicals. Opioids are among them, the brain's own heroin-like signals. In a rat, the hotspot in the reward hub is about a tenth of the hub. The rest of the hub can drive wanting but not liking. On that account, the hub's peak at a chill could be wanting at its height. Berridge wrote that in the 1990s this was "a lonely scientific position." He says few researchers who study reward still claim dopamine is the pleasure.
The Barcelona drug study was built to test exactly this. It counted the pleasure buttons as liking and the bids as wanting. Suppose dopamine drove only wanting. Then the drugs should have moved the bids and left the buttons alone. Both moved. The Barcelona team still allows another reading. Dopamine may be a step the pleasure can't do without, while opioids further along make the pleasure itself. Studies that blocked opioids have dimmed musical pleasure in some listeners. One of them dimmed unpleasant music too, which muddies it, and two later studies didn't repeat the effect on pleasure. In 2025 a team in Finland scanned fifteen women listening to music they had chosen for pleasure. More chills went with signs of more opioid release in the hub. Comparing music with silence overall, though, the authors couldn't tell whether opioid activity rose or fell. I find the firm part narrower than either side's claim. Dopamine changes how much people enjoy music, and how much they want more of it. Nobody can yet say whether dopamine is the enjoying itself.
Why does music give you chills? A chill marks a moment when the reward system that handles food and money answers your brain's bets about the music. Part of that system starts work in the seconds before the big moment, while the bet is still riding. Its hub peaks as the moment lands. In a study, turning dopamine up or down moved both the time spent in chills and the price people would pay. With music nobody had heard before, the hub still tracked what people would pay. Their liking still followed the program's odds. So memory of one piece can't be the whole story. The people at the party hear all of it. In them, the hearing areas and the hub barely work in step.
And why should a run of notes feel good at all? Gold's team offers an answer: the brain rewards learning, and music offers plenty of predictions to confirm and to break. By people's own ratings, music pleases most when you can bet on it and still be beaten by it. What dopamine adds is still argued. It may carry some of the pleasure itself, or only the wanting that comes before it. I suspect the two are hard to separate in music, where so much of the pleasure lies in what's coming.
The bets in this lecture were all about which note or chord comes next. Music also sets up bets about when. Even the 2001 chills scan caught activity in parts of the brain that handle movement. Why does a beat make you want to move, even when you're sitting still?
6The Beat Is in Your Head
In 2008, a cockatoo named Snowball took part in an experiment at a bird shelter in Indiana. He was set on the back of an armchair, and a song he knew came on: "Everybody", by the Backstreet Boys. A computer had sped it up or slowed it down without changing its pitch. The question was whether he would stay with the beat. His two owners stayed in the room but didn't dance, so that he couldn't copy them. One of them was Irena Schulz, who ran the shelter. In the first two sessions, she bobbed her head to the beat anyway. It was slight, and she didn't mean to.
Her head raises this lecture's question. Why does a beat pull at a body that's meant to keep still?
Snowball was the one the scientists were watching. Moving to a beat was thought to be something only humans do. Dogs have lived with people and their music for thousands of years, and you don't see them do it. But the person in that room is a puzzle too. She was meant to keep still beside a song with a beat, and her head kept time without her. You've probably caught a foot of your own doing the same.
The last two lectures were about predicting which note comes next. A beat raises a different prediction. It's about when.
I'll break the question into four smaller ones. Is the beat in the sound, or in you? Babies bounced to a drum rhythm answer that one. Then which part of the brain keeps it? Adults lying still in a scanner show us. How well do the famous findings about the beat hold up? Two of them shrank when bigger studies checked. And who else can keep a beat? Snowball comes back for that, with a sea lion named Ronan.
A beat seems to live in the music, in the drums and the bass. Researchers define it as a steady pulse that you perceive in music. It's what you'd tap your foot to. You also group the beats. Feel them in twos, one-two, one-two, and you have a march. Feel them in threes, one-two-three, and you have a waltz. That grouping is called meter. Usually the music marks it for you, with louder or longer notes on the strong beats.
But the beat doesn't have to be in the sound at all. Play a run of identical clicks at an even pace. People can hear every second one as stronger, or every third. Nothing in the clicks is stronger. A beat can even fall on silence. In syncopation, a note lands just before the beat, and nothing sounds on the beat itself. You still feel the beat there.
Two Canadian psychologists asked whether the body helps decide which grouping you hear. They played seven-month-old babies a drum rhythm for two minutes. It had no accents at all, so it could be felt as a march or as a waltz. Half the babies were bounced on every second beat as it played. The other half were bounced on every third. Each baby heard two versions of the rhythm, one with louder beats in twos and one with louder beats in threes. The babies controlled how long each version played. They listened longer to the one that matched their bouncing.
Was it something they saw? A second group was blindfolded while they were bounced, and they showed the same preference. A third group sat still and watched an adult bounce, in twos or in threes. What would watching do? Nothing. Those babies preferred neither version, and neither did babies who were never bounced. Their own bodies had to move. Adults showed it too. Those who bounced along in twos later picked out the march version as the one they'd heard. Those who bounced in threes picked the waltz.
The two psychologists put it in a line: "We hear the melody in music, but we feel the beat." The babies hadn't moved themselves. They had been moved, so the psychologists propose that the sense of balance and motion did the work. The beat is partly something you add, then, and the way your body moves helps decide which one you hear. Something in your brain must be doing the adding.
The adding happens somewhere you might not expect: in the parts of the brain that plan and time movement. Call them the movement system. Jessica Grahn and Matthew Brett put 27 people in a scanner in Cambridge, England, the blood-flow kind from the second lecture. Fourteen of the 27 were musicians, and the rest had no musical training.
They heard short rhythms made of plain tones. Each rhythm played three times, and they said whether the third was different. They were told to keep their whole body still, apart from pressing the answer button. The rhythms came in three kinds. In one, the notes were spaced so that a steady beat came through. In another, the very same note lengths were shuffled, so no steady beat emerged. The third used lengths too awkward for any beat to fit. Rhythms with a beat are easier to remember, so the team tuned the task until people scored about the same on all three. A harder task couldn't then explain a difference in the brain.
Every kind of rhythm woke up the movement system. But the rhythms with a beat drove two parts of it harder. One was the basal ganglia, the brain's deep movement-timing machinery. The other was the supplementary motor area, a planning area in the brain's outer layer. The beat made the same difference in musicians and in people with no training. Meanwhile, the area that becomes active when the body moves stayed quiet. The listeners had kept still. The planners were busy, and nothing was moving.
The deep timing machinery does seem to matter for the beat. Parkinson's disease damages the basal ganglia. In a later study, Grahn and Brett found that people with Parkinson's were worse than healthy volunteers at telling apart rhythms with a beat. With the other rhythms, they weren't reliably worse.
A scanner is too slow to follow each beat. Recordings of the brain's magnetic signals can. A team led by Takako Fujioka in Toronto recorded twelve people as they watched a movie of their choice, with steady tones playing. They weren't asked to listen for anything. The recordings followed a fast brain wave called the beta wave. It has long been tied to movement: it dips when you plan or make a movement, and recovers afterwards.
In these still, distracted viewers, the beta wave dipped just after every tone. Then it climbed back, and peaked just before the next tone arrived. Faster tones brought the climb sooner. With random tones, the climb lost its timing. The wave was getting ready for a beat it could predict. It rose and fell in movement areas too, including the planning area.
These listeners weren't moving, or even paying attention. So a later review of this work calls the movement system's part in hearing a beat "obligatory." It suggests that may be why music makes people want to move. As I see it, this finishes something the third lecture started. There, the brain predicted which note would come next. Here it predicts when the next beat will come, and it does that with the movement system.
So far, two pieces fit. You add the beat, and the way your body moves helps decide which beat you hear. The movement system keeps that beat, and runs a little ahead of it, while the body stays still. What we haven't asked is how well the famous evidence holds up, and what makes one rhythm pull at you harder than another.
The first famous finding seemed to show a march or a waltz inside the brain. It came in 2011, from Sylvie Nozaradan and her colleagues in Brussels. They recorded the brainwaves of eight volunteers, through electrodes on the scalp. Each heard a pulsing tone, a little more than twice a second. They imagined the pulses in twos, as a march, or in threes, as a waltz. They learned to do it by tapping and counting aloud. Then they had to keep it going in their heads, sitting still.
The sound never changed, and the brainwaves rose and fell with its pulse. Imagining a march added something the sound didn't have: a second, slower wave, at half the pulse's rate. Imagining a waltz added one at a third of the rate. On the team's reading, the meter itself could be seen in the brainwaves.
Eight people is a small study for a claim that size. A psychology journal later invited labs to repeat it, under a plan written and reviewed before any data came in. Every lab watched its volunteers for movement. One volunteer was dropped for moving rhythmically. In the end, results from 152 people in thirteen labs counted.
The plan set its tests in advance. By those tests, the extra waves came out at most a third of their 2011 size. That was too small to tell from no effect at all, so the study failed to replicate. A second analysis, not in the plan, put all 152 people together. It found a small effect, in the expected direction. Seeing it took nearly twenty times as many people as the first study. And the extra wave didn't predict who could tell where their imagined beats fell. The team's verdict is careful: it may be "premature to conclude that there is no such effect."
That takes away the most vivid picture of an imagined beat: the waltz in your head, showing up in your brainwaves. It doesn't touch the bounced babies, who were judged by how long they listened, though that study was small too. And the scanner result on the movement system has turned up in several studies, even when people were listening for changes in loudness or pitch instead.
The second finding was about the pull itself. Researchers call the pleasurable urge to move to music groove, and music with groove is often full of syncopation, notes that land just before the beat. The music researcher Maria Witek and her colleagues ran an online survey. Sixty-six people rated fifty funk drum breaks for how much each made them want to move.
The ratings rose with syncopation, peaked in the middle, and fell again. It was the upside-down U from the last lecture, where enjoyment rose and fell with complexity, with groove at the sweet spot. Witek's team argued that in the middle, the beat is clear enough to predict, and the off-beat notes can still surprise. They called syncopation "an invitation to the body" to move with the beat.
A decade later, a team in Switzerland tested the sweet spot again. Witek was one of its authors, testing her own finding. This time every drum pattern was copied from a real record by a famous drummer, and no hump appeared.
So the team went back to Witek's ratings. Fourteen of her fifty breaks had been composed by the experimenters, to reach the far ends of syncopation. The made-up breaks got low ratings, and they sat at the two extremes of syncopation, the least and the most. They were mainly what bent the line into a hump. Without them, the hump was still there, but it did almost nothing to predict which breaks people rated highly. The Swiss team argues that listeners can probably hear when drumming doesn't sound real, and rate it down for that.
That doesn't end the argument. Six earlier studies had found the full hump, and three found part of it. The Swiss team says that most of them, like Witek's, mixed real patterns in the middle with made-up ones at the ends. But one study made up every pattern, at every level, and still found the hump. That suggests made-up drumming isn't the whole story. And the Swiss patterns all came from great drummers, with none at the extremes, where a drop-off would show. I'd score the sweet spot for groove as unproven, not disproven. The Swiss study did find two routes to the urge to move. Listeners heard simpler patterns as more regular, and regularity made them want to move. They heard complex patterns as more interesting, and that made them want to move too.
Both checks teach the same caution. A first study of eight people, or of sixty-six raters online, can find a pattern that a much bigger study can't. The beta-wave studies are small too, so I'd hold them more loosely than the scans.
Which brains can keep a beat at all? Two abilities are involved: hearing a beat, and moving in step with it. Moving in step means matching the beat's speed and landing on it. Researchers call that entrainment. People land on the beat, or a hair ahead of it. That means predicting it, since the movement has to start before the sound arrives. Fujioka's team suggests that the beta wave's climb helps plan exactly that kind of movement.
Hearing a beat comes first. A team in Budapest played a rock drum pattern to sleeping babies two or three days old, with their mothers beside them. Now and then, the first beat of the pattern was left out, and the babies' brains responded as if something had gone wrong. A review pointed out a snag: dropping that beat also dropped drums, so the babies might just have noticed fewer sounds.
So the same team tried a cleaner test. Now a drum hit was sometimes made much quieter, either on the beat or between beats, so the change was the same sound either way. Each sequence came in two forms, one with steady timing and one with the gaps jumbled. The babies' brains reacted more to a quiet hit on the beat than between beats, but only when the timing was steady. The order of the sounds was the same in both forms. So the babies weren't just learning which sound follows which, the statistical learning of the third lecture. Two macaque monkeys given the same test sensed the steady timing, but showed no sign of the beat. On these two studies, both from one lab, babies track a beat within days of birth. Which rhythms sound normal is learned later, as the third lecture showed with babies of twelve months.
Moving in step comes much later. Another team filmed 120 babies, from five months to two years old, on a parent's lap. The babies moved more to music than to speech, and faster to faster music. But the team found no sign of them landing on the beat. In tests like these, that seems to wait until about preschool age, and it goes on improving for years.
Snowball's test asked whether a bird was capable of entrainment: matching the speed and landing on the beat. Aniruddh Patel ran it. In the second lecture, he argued that music and language share some of their processing resources. Someone had e-mailed him a video of Snowball. "This is much more than just a cute pet trick," he recalls saying. His team made eleven versions of the song, at different speeds.
In more than half his dancing trials, at most of the speeds, Snowball had stretches of head bobs locked to the beat. A check against chance made luck very unlikely. In the study, nobody trained him or rewarded him with food. His early history is unknown, and a previous owner had danced with him. And the owner's bobbing head? The trials with her bobbing and the trials with her still came out about the same. A little over half had locked stretches either way. He wasn't copying her.
Patel's original explanation was that keeping a beat rides on vocal learning, the ability to learn and imitate new sounds. Copying a sound takes tight links between hearing and movement. He argued that keeping a beat does too. People and parrots learn sounds. Dogs, cats and monkeys don't. In early studies, macaques that learned to tap along landed about a quarter of a second after each beat. That's faster than a plain reaction, so they predict a little, but not the way people do. With other training they got closer, and later work suggests monkeys can learn to anticipate a steady metronome.
Then there's Ronan, a California sea lion at a university marine lab in Santa Cruz. Sea lions show no sign of learning sounds in the wild. On Patel's rule, she shouldn't keep a beat. At three, Ronan was trained to bob her head in time to a metronome, with fish as rewards. She soon kept time to speeds she had never heard, and later to music.
In 2021, Patel answered that sea lions' vocal skills haven't been tested with modern methods. He pointed out that some seals, which are related to sea lions, are skilled sound-imitators. He also noted that Ronan's timing slid around the beat as the tempo changed, unlike a person's. And he revised his rule rather than dropping it. Vocal learning comes in degrees, he now argues, and more of it should mean more humanlike beat keeping.
At fifteen, Ronan was tested again, beside ten students. They chopped an arm down in time to the same snare drum she bobbed to. Her timing varied less than any student's. Some students beat her on one measure or another, but none on all of them. And her timing no longer slid with the tempo. At the end, the students were thanked. Ronan got a toy filled with fish and ice.
Ronan is one animal, tested by her own trainers on a plain, steady beat, beside students trying an arm chop for the first time. Her trainers call the vocal-learning idea "a stretch". They think any complex brain may have the basic equipment, and that things like motivation, control of the body and practice decide who uses it. The argument is open.
Why did Irena Schulz's head move? On the best evidence, her brain was keeping the beat with its movement system. The basal ganglia and the supplementary motor area work harder for a rhythm with a beat, even in a body that's still. A wave tied to movement climbs toward each beat before it arrives. That's the explanation these researchers favor, though it isn't proven. Even the experiments had to fight the pull. A head bobbed in the bird shelter, and a volunteer in the thirteen labs was dropped for moving rhythmically. My own gloss goes one step further. Sitting still to music means holding back a movement your brain is already preparing. Others read the same activity as the brain simulating movement to keep time, with nothing held back at all.
The beat, in the end, is a joint product. The sound offers the timing, and you supply the grouping, with your body's movement able to tip it toward a march or a waltz. Landing a hair ahead of a beat is the mark of a brain predicting it. The most striking evidence shrank when thirteen labs checked it: a waltz imagined into the brainwaves. The sweet spot for groove is now in dispute too. And keeping a beat isn't only human, though which brains can do it is not settled.
Ronan also points ahead. At fifteen she keeps time better than she did at three. The next lecture asks what years of practice do to a human brain, and whether it matters when you start.
7What Practice Does to a Brain
One afternoon, at ten past two, Clive Wearing wrote "This time properly awake" in his journal. At 2:14 he wrote "this time finally awake." At 2:35 it was "this time completely awake." In March 1985, the virus that usually causes cold sores had reached his brain. He was in his mid-forties, a conductor and an expert on early music. The infection left him with a memory that lasted only seconds. It also wiped out almost his entire past.
His music survived. He was still in hospital when his wife found out. She held up a score and began to sing one of its lines, and he joined in on the tenor part. He could still play the organ, hands and feet together. In the summer of 2005, the neurologist and writer Oliver Sacks came to visit. Two volumes of Bach's preludes and fugues sat on Clive's piano, and Sacks asked him to play one. Clive told him he'd never played a single one of them. Then he played the Prelude in E major, and said, "I remember this one."
Sacks explains it with two kinds of memory. One holds the events of your life, and that's the kind the virus destroyed. The other holds how to do things. It runs through other parts of the brain, and it can survive damage that wipes out the first. Clive didn't remember ever playing that prelude. His hands did. Years at the keyboard had left something behind that wasn't a memory of practicing. What was it, and where?
The last lecture's sea lion kept time better at fifteen than at three. This lecture asks the same of people. What do years of practice do to a human brain? And does it matter when you start?
A hand will carry us through it. A string player's left hand comes first, and what years of playing do to its place in the brain. Next come children at a keyboard, scanned before their first lesson and again fifteen months later, to see whether the practice made the difference. Then comes the case that starting young counts for more. Last is a famous pianist whose right hand stopped obeying him.
Practice leaves its mark on maps. You've met one already: the first lecture's place map, which lays out sounds from low to high, in the ear and again in the hearing cortex. The brain keeps a map of the body too, for touch. Each part of the body has its own patch in the brain's outer layer. The fingers lie in order, from thumb to little finger.
Thomas Elbert worked in the psychology department at the University of Konstanz, in Germany. In 1995 his group measured the touch map in string players. On a violin or a cello, the left hand does the fine work. Its four fingers press the strings, and the thumb grips the neck, shifting a little now and then. The right hand holds the bow, and its fingers move much less on their own. The team tested nine string players, along with six people who had never played a string instrument. An air-driven device pressed lightly on the thumb and, in separate runs, on the little finger, a thousand times each. Recordings of the brain's magnetic signals showed where on the touch map each press arrived, and how strongly.
In the string players, a touch on the left hand drew a stronger response than in the others. The team read that as more of the map given over to those fingers. The little finger had gained the most. The thumb is used less, and it had gained less. And the right hand, the bow hand, was no different from the non-players'. The map had grown where the practice landed.
Why would a map grow? The answer these researchers give goes back to Donald Hebb, a Canadian psychologist, in 1949. Hebb proposed a rule: when one nerve cell keeps helping to make another fire, the link between them grows stronger. The usual slogan for it is "Cells that fire together, wire together." It seems to have been coined around 1992 by neuroscientists studying how vision develops, though papers sometimes credit it to Hebb.
Experiments on monkeys show the rule redrawing the touch map. Some monkeys were trained to hold two fingertips against a spinning disc with a rough surface. Each time, they were rewarded with banana-flavored chips. After about six hundred of these touches, the researchers mapped the hand. The patches for those two fingertips had grown. Timing matters as well as use. In monkeys whose two fingers were joined by surgery, the two patches merged into one. In people whose joined fingers were separated by surgery, the patches moved apart. Signals that keep arriving together may end up sharing a patch, and signals that arrive apart may keep separate ones. And it seems to take touches that matter to the animal. Touch that it merely receives seems to change little.
The mark shows up beyond touch. A group in Boston scanned eighty men: professional keyboard players, amateurs who practiced about half as much, and men who had never played. In areas for movement and for hearing, and in one used for reading music at sight, the professionals had the most grey matter. The amateurs came next, then the men who had never played. Grey matter is brain tissue, and in rats trained in acrobatics, it gains connections between nerve cells.
Hearing changes too. A team in Münster led by Christo Pantev recorded the hearing cortex of musicians and non-musicians as they heard piano notes and pure tones. A pure tone is a plain sound that no musician practices with. The musicians' responses to piano notes were about a quarter larger than the non-musicians'. For pure tones, the two groups were alike. Among the musicians, how much music they had only listened to made no detectable difference. The team concluded that it seems to take active practice.
Does that mean the practice built those brains? Elbert's team raised the other possibility themselves. People born with a bigger map for the left hand's fingers might make better string players. They'd be more likely to keep at it. The Boston group said the same of its eighty men. They wrote that some of the differences could be inborn. Every comparison of grown musicians runs into this problem, because musicians chose to become musicians. To tell what practice does from what people bring to it, you have to see a brain before the practice starts.
The Boston group set out to do that with children. The study followed thirty-one children, about six years old, from public schools around Boston. None had had music lessons. Fifteen were starting weekly half-hour private lessons on a keyboard. The other sixteen had only a weekly music class at school, with songs, drums and bells. Every child had a brain scan at the start, and another fifteen months later. Every child also took a finger test, pressing number keys in a sequence like five-two-four-three-five, as fast and as accurately as they could. And every child heard pairs of short tunes and said whether the two were the same.
At the start, the two groups' brains showed no differences. Fifteen months later, they did. Several places had grown more in the lesson group, and three of them matter here. One was the hand's patch in the area that becomes active when the body moves. The patch that grew was on the right side, which serves the left hand. The second place was a stretch of the bridge between the brain's left and right halves. The bridge is the band of nerve fibers that joins them. The third was a patch of hearing cortex on the right side, though the evidence for that one was weaker.
The growth also tracked the skills. The hand patch and the bridge grew most in the children whose left hands improved most on the finger test. The hearing patch grew most in those who improved most on the tunes. That's a step past the comparisons of grown musicians. The team concluded that the differences seen in adult musicians are more likely the product of training.
It still wasn't a clean experiment, because the families chose the lessons. The lesson children came from homes where the parents had more education. They were also about five months older. The results the team reports already allow for both. Glenn Schellenberg studies what music lessons do for children, and he reads the study more coldly. The project began with seventy children, and only thirty-one ended up in the brain analysis. Back then, before any lessons, the children about to start had scored higher on all seven tests they took. No single gap was large, but all seven pointed the same way. He argues that children who take lessons already differ from other children, in ability, family background and personality. He suggests that a head start like that could have grown over fifteen months until it showed in the scans. The team's own case rests on the scans, which matched at the start. I'm persuaded the study narrows the gap between training and talent. It doesn't close it.
On this evidence, practice leaves its mark where the work lands, and the children suggest it does more than reveal a brain that was already there. Two questions remain. Does the age you start count on its own? And can the same rule go too far?
Elbert's string players held a second surprise, about age. You might expect the map to follow the hours of practice. It didn't. The size of the little finger's response tracked the age a player had started, and not the hours. Those who began before about twelve had the strongest responses. Those who began after thirteen had weaker ones, though still stronger than the non-players'. In Pantev's study of piano notes, the dividing line fell at about nine. An effect of age like this would point to a sensitive period. That's a window in childhood when what you do leaves an unusually lasting mark on the brain.
A team in Montreal spotted a snag in that kind of evidence. Early starters usually end up with more years of training than late ones. Any difference might come from the extra years.
So the Montreal team matched them. It tested thirty-six highly trained musicians, half of whom had started before seven and half after. The two groups had the same years of lessons, the same years of playing and the same hours of practice each week. A third group had barely played at all. Everyone tapped a key in time with a pattern of short and long flashes on a screen. Everyone also had a scan that traces the brain's wiring, its bundles of nerve fibers.
The early starters kept time with the flashes more precisely. In them, one stretch of the bridge between the brain's halves was better connected. Its fibers link the two sides' areas for touch and movement. The later someone had started, the weaker that stretch. What about the late starters, with all their years of training? In that stretch of the bridge, they looked no different from the people who had barely played.
Why would age matter? Playing an instrument takes two hands working together. The two halves of the brain coordinate the hands across that bridge. That stretch of the bridge also changes a great deal between the ages of six and eight. The early starters began their training while it was still developing. The team's idea is that early training lays down wiring that later practice builds on.
These were still adults who had started at different ages, for reasons nobody controlled. The authors say only an experiment that assigned children at random, or followed them for years, could test a sensitive period directly. The studies don't agree on an age, either. Elbert's fingers pointed to about twelve, and Pantev's piano notes to about nine. The Montreal bridge pointed to seven, and each was measured a different way. And all of these studies are small. The largest had eighty people, and most had a few dozen or fewer. My reading is that an early start probably counts for something. How much, and up to what age, no study here can say.
The same rule may have a darker side. Leon Fleisher was an American pianist who took over his family's piano at the age of four. At sixteen he made his debut at Carnegie Hall. In 1965 he was due to tour the Soviet Union with the Cleveland Orchestra. By then his right hand was failing him. Its fourth and fifth fingers curled under, and he couldn't stop them. He later wrote of "a feeling that my fingers weren't doing what I wanted them to do." In his case, practicing harder only made it worse. He didn't go on the tour, and he began cancelling concerts. He was thirty-six. Eventually the problem got a name: focal dystonia.
Focal dystonia is a painless loss of control over a skilled movement that used to go well. In musicians, fingers start moving together against the player's will. It tends to strike the movements a musician practices most, in the hand that does the hardest work. Pianists and guitarists usually lose control in the right hand, and players of bowed strings in the left. For pianists it's most often the fourth and fifth fingers. A 2013 review suggests the reason: those two fingers carry the melody. For string players, too, it's most often the fourth and fifth, and the little finger is the one whose patch grew most in Elbert's string players. But it can strike any finger. Estimates run from one musician in two hundred to about one professional in a hundred.
Elbert's group suspected the rule itself. A professional musician moves several fingers together, fast and over and over, for hours a day across many years. Monkeys made to repeat the same grasp for a long time developed trouble moving their hands. On their touch maps, the borders between the fingers broke down. In healthy volunteers, repeated touches to several fingertips at once made their patches begin to merge. The volunteers started mistaking which finger had been touched.
Elbert's group went looking for that merging in musicians with dystonia. In 1998 they measured the maps of seven professionals with the condition, among them guitarists, a pianist and an organist. They also measured healthy musicians who played the same instruments, and people who didn't play. The air-driven device pressed each fingertip in turn. The fingers' patches for the affected hand sat closer together than in the people who didn't play. They were still in order, from thumb to little finger. But they had run together. The healthy musicians' patches were spaced normally. In four of the seven, the map for the other hand had drawn together too.
That fits the rule, but one round of measurements can't say which came first. Merged patches might cause the dystonia. The dystonia might merge the patches. Or a third thing might cause both, perhaps a fault in the basal ganglia, the deep timing machinery from the last lecture. A 2015 review of the studies, most of middling or poor quality, judges intensive playing to be one cause, along with an inherited tendency. Plenty of musicians presumably practice as hard and never get it. Elbert's group raises the possibility that those who do had patches prone to merging from the start.
The rule also suggested a treatment. In a small study, Elbert's group had musicians practice with a therapist while splints held some of their other fingers still. That way the affected finger had to move independently of its neighbors. The pianists and guitarists improved, judged by how they played their own pieces, and their maps looked more normal afterwards. The wind players went through the same routine and didn't improve much. The team counted them as, in effect, a placebo group: going through the routine wasn't enough on its own. In a later study of a related training, eight of eleven musicians improved after a week of daily practice. Other treatments include retraining at slow speeds, and injections of botulinum toxin. The 2013 review is plain about the evidence. The splinting and similar retraining rest on case reports and small case series. Slow-speed retraining comes from a larger series of 145 musicians, and after two years about a quarter had better control. Treatment brings relief to some musicians, and some get back to their old level. But symptoms rarely go away completely, and reviews disagree about how many musicians keep their careers.
Fleisher spent the next decades teaching, conducting and playing music written for the left hand alone. In the 1990s, some control slowly came back to his right hand, after treatments that included Botox injections. One man's recovery can't tell us what helped. In 2003 he made a triumphant return to Carnegie Hall, and in 2004 he released an album called Two Hands.
A rule about timing runs through most of this lecture. In the slogan's words, cells that fire together wire together, at least when the activity matters. So years of practice redraw the brain's maps, and the fingers and sounds you work hardest get the most room. A string player's busy left fingers take more of the touch map than a non-player's do. In a few musicians whose hands stopped obeying, the fingers' patches had run together. That may be the same rule overshooting, though which came first is unknown, and it's one suspect among several.
Grown musicians can't show what caused their differences, because they chose to play. Children scanned before their first lesson came closer, though their families chose too. Starting before about seven goes with a better-connected bridge between the brain's halves, even with the years of practice matched, but that is still a comparison.
The Boston children also took tests of reasoning and vocabulary. Whether lessons reach any further than the hands and ears is the next lecture's question. Put plainly, does music make you smarter?
The rule doesn't need you to remember the lesson. Clive Wearing's hands kept the memory for how to play. Sacks places that kind of memory in older, deeper parts of the brain, among them the basal ganglia. A memory researcher wrote to Sacks that every case of amnesia is different, so Clive's is his own. That kind of memory can still take in new things. Clive can make coffee, finding the cups, the milk and the sugar, though he can't say where they're kept. What he can't do is practice on his own. He has no memory of yesterday's session, so the idea of practicing never comes to him. Someone has to put the music in front of him.
8Does Music Make You Smarter?
In January 1998, the governor of Georgia asked his state's legislature for 105,000 dollars. Zell Miller wanted every baby born in the state to go home with a tape or a CD of classical music. Many legislators balked. In the end they didn't have to pay. Sony offered the recordings free. It had a big factory in the state. Hospitals were to hand them to new parents along with the diapers. The album was called Build Your Baby's Brain Through the Power of Music. That May, Florida passed a law telling day-care centers paid by the state to play classical music every day.
Miller's evidence was an article in Time magazine. It cited a study said to show that listening to Mozart briefly lifted college students' IQ scores. He hadn't consulted the researchers. He also said this: "No one questions that listening to music at a very early age affects the spatial, temporal reasoning that underlies math and engineering and even chess."
Quite a few people were about to question it. Does music make you smarter? The last lecture showed years of practice redrawing the brain's maps for the hands and the ears. This one asks whether music reaches anything further. By smarter I mean reasoning, vocabulary and marks at school, and above all IQ, the score on a standard intelligence test.
The evidence comes from four comparisons, each fairer than the one before. Mozart against silence, and then against a story. Lessons, measured on the skills they train and on skills far from music. Children who chose lessons against children given them by lottery, and against their own twins. And last, two big reviews of all the trials against each other, with the gains close to music that hold up. One question runs through all four. Compared with what?
The Mozart study came out in the journal Nature in 1993. Its first author was the psychologist Frances Rauscher. College students listened for ten minutes to a Mozart sonata for two pianos, and then took a test of spatial reasoning. They did better than they had after sitting in silence, or after relaxation instructions on tape. The gain came to the equivalent of eight or nine IQ points. It lasted ten to fifteen minutes. The finding became known as the Mozart effect.
Look at what was measured: one kind of skill, in grown students, for about a quarter of an hour. Critics later attacked the idea that Mozart makes people smarter, and Rauscher's answer was short. "We made no such claim," she wrote. In her account, the effect was limited to spatial tasks that involve picturing shapes and putting steps in order.
The claim grew as it was retold, and two researchers at Stanford measured how. They found 478 articles about the Mozart effect in America's fifty biggest newspapers, from 1993 to 2002.
In 1994, eight articles in ten mentioned college students. After 2000, about three in ten did.
Babies went the other way. No article mentioned them in 1995, and more than half did by 1999. In the researchers' words, there was "no scientific research whatsoever" linking music and intelligence in infants.
Then came the checks. In 1999 a psychologist at Harvard pooled sixteen studies. Against silence, the average gain came to 1.4 IQ points, a small fraction of the original eight or nine. In the same issue of Nature, three labs that had tried to repeat the finding reported no effect at all. "A requiem may therefore be in order," they wrote. Rauscher replied that other labs had found the effect. She compared it to baking: some people can't get bread to rise, and that doesn't disprove yeast.
The last lecture introduced Glenn Schellenberg, who studies what music lessons do for children. In 1999 he and Kristin Nantais tested where the lift came from. Their students sat in a soundproofed booth with headphones on. Each listened for ten minutes, then did a paper-folding test on a computer. A drawing showed a sheet of paper folded and cut. The student picked which of five pictures showed it opened out.
In the first experiment, each student had ten minutes of music one day and ten minutes of silence on another. They did better after the music, and Schubert worked as well as Mozart. Compared with what, though? Ten minutes of silence is dull. So in the second experiment, the silence was swapped for a story. Each student heard the Mozart one day. On another day they heard a Stephen King short story, read aloud, for the same length of time. After which did they score higher? On average, neither. Then the team asked each student which they had preferred, and about half chose each. Those who preferred Mozart scored higher after Mozart. Those who preferred the story did a little better after the story.
The two concluded that the Mozart effect "has nothing to do with Mozart in particular." They noted that every lab that had found it had compared Mozart with silence, a relaxation tape or repetitive music. In Schellenberg's account, Mozart beats those because it leaves people in a better mood and more alert. That helps on a hard spatial puzzle. Set against something people enjoyed as much, the music added nothing.
Lessons are a different kind of claim. They mean months of practice. Could that reach beyond music? Psychologists call it transfer when you get better at one thing because you practiced another. Near transfer goes to something closely related. The Boston team from the last lecture, who followed children through keyboard lessons, gives an example, fingers trained on an instrument getting quicker at typing. Far transfer goes to something with little in common. The team's example is reading rhythms in music helping a child with fractions. The claim that music makes you smarter is a claim of far transfer.
Far transfer is the hard one. Lab studies going back to 1901 find it much rarer than near transfer. A skill carries over most when two tasks share a great deal. In one large review, memory training improved other memory tests for a while, and did nothing measurable for reasoning. In a study Fernand Gobet co-wrote, chess masters were asked to remember positions, or find the best move, from openings they don't usually play. They did drastically worse. So even at chess, skill carries over only in part. Asking piano lessons to carry over to vocabulary is asking a great deal more.
The Boston children were six-year-olds, scanned before and after fifteen months in which about half had keyboard lessons. They took tests of both kinds. The near ones were a finger test on number keys, and telling short tunes apart. The far ones included vocabulary, reasoning puzzles and building patterns from blocks. By the end, the lesson children had improved more than the others on the near tests. On the far ones, they improved no more. The gains landed on the skills the lessons trained.
Mozart, then, lifts the mood for a few minutes, and lessons sharpen what they train. Neither explains a stubborn fact. In study after study, people who have had music lessons score higher on intelligence tests than people who haven't. Comparisons of musicians with non-musicians find a gap four or five times the size of the typical gain in trials of lessons. Why do the people who took lessons score higher?
The answer begins with who takes lessons. The last lecture gave Schellenberg's view: children who take lessons already differ from other children, in ability, family background and personality. They tend to be more curious, for one, and more drawn to the arts. Lessons cost money, and they need parents who back them. In one study, about a quarter of eleven-year-olds whose families were above the poverty line took private music lessons. Among children living in poverty, one in twenty-five did. Schellenberg puts it bluntly: musicians are "as much born as they are made."
The children who start lessons are often ahead before the first one. The last lecture's Boston children scored higher on all seven tests before any lessons, though no single gap was large. A team at the University of Granada, in Spain, looked across fourteen studies in which children or their families had chosen music. Most of those studies were too small to see a gap, so they reported their groups as evenly matched. Pooled, the gap showed. The children who had chosen music had a head start.
There are two ways round a head start. One is a lottery, in which chance decides which children get music lessons. The other is to compare twins. A lottery works best with one more feature. The children who don't win lessons should get something else, a comparison class. It should be new and engaging, and taught the same way, but without music. Then the music is the only difference.
Schellenberg ran an early lottery trial of music lessons himself, in Toronto. A local newspaper advertised free weekly arts lessons for six-year-olds. Families had to agree to take part whichever group their child landed in. Then 144 children were drawn by lottery into four groups. One group learned keyboard. One learned singing by the Kodály method, which teaches with songs, hand signs and clapping. One took drama, which has rehearsing, memorizing scripts and expressing emotion, but no music. And one waited a year for keyboard lessons. The lessons ran for thirty-six weeks at the Royal Conservatory of Music, in classes of six. Most of the children took a full IQ test before the lessons, and again after the year was over. The testers didn't know which group a child was in.
What happened to the children with no lessons? Their IQ went up too, by about four points. Every group's did. That year the children started first grade, and going to school raises IQ. The question is which groups rose most. The two music groups gained seven points on average, and the other two groups 4.3. Schellenberg titled the paper "Music Lessons Enhance IQ." The drama children gained something else. By their parents' ratings, their social skills improved, while the music children's didn't change.
Compared with what, though? Against the children who waited, the music lead was clear. Against drama alone, keyboard was about one point ahead, and singing about two and a half. Some ways of analyzing the data called that gap real, and others didn't. The music children had also barely practiced between lessons, about ten to fifteen minutes a week.
In 2013, a much larger trial in English primary schools tried to repeat the result. It split 909 children of about the same age by lottery, within their classes. A third learned violin or cello, a third learned singing, and a third took drama, once a week for 32 weeks. By the end of the year, the music groups had gained no more than the drama group, in math or in literacy. The difference was close to zero, even for children from poor families. The violins and cellos were meant to go home for practice. In most schools, they didn't.
Looking back on his own trial, Schellenberg called its results "weak, without successful replication for almost 15 years."
One objection is fair. Maybe a year is too short, and six-year-olds don't practice enough. The Granada team argues just that. Twins speak to it, because the comparison isn't a year of lessons. It's a twin who practiced a great deal against one who didn't. Researchers in Sweden studied more than 10,500 twins. Across all of them, those who had practiced more music had somewhat higher IQs. Then the team compared each twin with the other. They explain that this controls for genes and for the home the twins shared. Compared with what? This time, with their own twin. Was the practiced twin's IQ higher? It wasn't. The team's reading is that the same genes nudge people toward both, toward practicing and toward a higher IQ. They write that their findings strongly suggest the link between practice and IQ isn't one of cause and effect. In my view, that doesn't rule out a small effect from years of lessons. But it suggests that much of the famous gap comes from what children bring to the first lesson.
What do lessons add on their own? Two big reviews have pooled the trials, and they disagree. The first is by Giovanni Sala and Fernand Gobet, of the chess study. Across 54 studies, the music children gained a little more than the others. But the gain sat in the weaker studies, those with neither a lottery nor a comparison class. Where there was either one, the gain was about zero. They point to the two largest trials in the field, England's among them, and both found nothing. Their conclusion was blunt: "since there is no phenomenon, there is nothing to explain."
The Granada team answered with a review titled "Please don't stop the music." Its point was that "music lessons" had meant very different things. Most of Sala and Gobet's lottery studies weren't lessons on an instrument. They were group music classes, or computer programs that train musical skills. The Granada team argues that an instrument asks more of a child: two hands working together, and nimble fingers. It also leans on memory and attention. So they pooled only trials of instrument lessons, and found a small gain. It held up in the lottery studies. They also set their instrument trials beside the other music classes from Sala and Gobet's review. In lottery studies, the instruments kept a small gain, and the other classes showed no reliable gain.
How big is the gain? The team puts it at about three IQ points. Pick one child at random from the lesson groups, and one from the comparison groups. By their count, the child with lessons comes out ahead 57 times in a hundred, where chance alone would give 50. Every such figure is an estimate, with a range it could really fall in. In the few lottery studies that had a comparison class, the best guess was still a small gain, but its range reached down to zero. The Granada team writes that three points "probably makes very little difference in daily life." They add that longer programs might do more, and that poorer children seem to gain more, though England's trial saw no such thing.
Strip away the sharp words, and the two sides agree on more than they dispute. Both accept that children who choose music start ahead. Both put what lessons add, in a fair trial, somewhere between nothing and about three IQ points. The fight is over whether it's nothing. My own judgment is that the Granada team is right to keep instruments apart, though that's contested. But on their own numbers, the gain is small, and it rests on few trials.
That is about as far as the evidence on IQ goes. A near claim is a different matter. It asks music to travel only a short way, to a skill that uses the same machinery. The second lecture raised a related hope, that music lessons might help children learn to read. Shared machinery offers a route: music training might sharpen how the brain encodes speech, and that might help reading.
Aniruddh Patel built a theory of how that could work. The second lecture met him arguing that music and language share some of their processing resources. He calls the theory OPERA, from the first letters of five conditions. The first is overlap. Music and speech must share brain machinery for some feature of sound, such as the pitch of a voice or a note. The second is precision. A note one small step out of place can be a wrong note. Speech is more forgiving. Mandarin speakers in a quiet room understood sentences even with the rise and fall of pitch flattened out. So music asks more of the shared machinery than speech does. The other three conditions are about the practice. It has to be enjoyed, repeated often, and listened to closely. Patel argues that when all five are met, the shared machinery sharpens beyond what speech needs, and speech gains too.
One study set out to catch a near gain as it happened. A team at Northwestern University worked with Harmony Project, a free music program for children in Los Angeles neighborhoods marked as gang-reduction zones. Forty-four children from its waiting list were split into two groups, mostly by lottery. Some began lessons at once, and the rest a year later. Once a year, each child heard two syllables, bah and gah, six thousand times each. Electrodes on the scalp recorded the brain's response. The question was how sharply it told the two syllables apart. One year of lessons left it no sharper. Two years sharpened it, and children who had spent more hours on their instruments had gained more. The team credits two of Patel's conditions: lessons that trained the children's attention on sounds, repeated over two years. It notes that the gains were small, and that it had no comparison class. Any engaging program might have done as much.
Schellenberg's review grants one gain near to music. In his reading, the most reliable results come from training built around rhythm. In one trial, children aged eight to eleven with dyslexia had seven months of either rhythm-focused music lessons or painting lessons. The music group improved more at hearing the separate sounds inside words, and at reading aloud. Even so, a computer program built for those sounds has also worked, so the music itself may not be needed. And the Boston children's keyboard lessons, for one, didn't improve their test of those sounds.
Does music make you smarter? Listening doesn't, beyond a pleasant few minutes. Lessons are another matter. They reliably make children better at music, with quicker fingers and a sharper ear. The famous IQ gap between children who take lessons and children who don't is largely about who takes them. Compared with what? That question did most of the work. Each time the comparison got fairer, the gain got smaller. A story replaced the silence, and Mozart's lift was gone. Drama replaced a year of waiting, and the music lead shrank to between one and two and a half points. A twin replaced a stranger, and the link between practice and IQ disappeared. Only the review of instrument lessons kept a gain through the fairer trials, about three IQ points. Even those authors say three points probably makes little difference in daily life.
Closer to the sound itself, the evidence is kinder, if thinner. Rhythm training seems to help children hear the sounds inside words. In one small study, two years of lessons sharpened how children's brains told bah from gah. Notice where lessons did best: close to the sound, on skills that use the same machinery as music. The next lecture asks what that narrower reach can do for a brain hurt by Parkinson's disease or a stroke.
Schellenberg practiced the piano every morning before school, from the age of five until he was sixteen. He has no doubt it shaped his development, and even his brain. What he doubts is that the same years would do the same for another child. In one sentence, Schellenberg sums up the research: "Music training improves musical skills, which should be enough."
9Can Music Help a Damaged Brain?
A music therapist in Houston sings "Happy Birthday" and stops just short of the last word. Her patient sings it. The patient was Gabrielle Giffords, a congresswoman from Arizona. She had been shot in the head in 2011, and the wound had taken away her ability to talk. Her music therapy began with songs like that one.
Over time she learned to say ordinary phrases in a sing-song voice. Songs gave way to chants, and chants to spoken phrases with the rhythm of speech. In November 2011, ABC News aired footage of her therapy. It shows her crying with frustration when she tries to talk and can't. It also shows her singing fluently, and looking joyful. ABC reported that she had relearned how to talk, and gave music therapy part of the credit.
Her story suggests that a song can carry words past damage that blocks speech. Can music help a damaged brain? And if it can, which part of the music does the helping, and how much? The last lecture found music lessons doing best on skills that share music's machinery, like hearing the separate sounds inside words. The fifth lecture found another piece of that machinery: your brain keeps a beat with its movement system, the parts that plan and time movement.
We'll come back to Giffords at the end. Music's help is clearest in walking, so we'll start there and judge her story against it. In trials, people with Parkinson's disease who trained with a beat walked a little better, on average. In trials after a stroke, too, walking to a beat sped people up a little, on average. Then we'll follow the beat from the feet to the mouth, into singing therapies like hers, and ask which part of a song carries the words. The last part is about why a story like hers comes out bigger than the trials. One thing should be clear from the start. Everything here comes from studies of patients, and none of it is a treatment plan for anyone.
In Parkinson's disease, walking can stall. A person means to walk, and for a moment the feet stop going forward. That's called freezing. Somewhere between twenty-five and sixty percent of people with Parkinson's have it, usually after some years of the disease.
Oliver Sacks described an extreme case of freezing. Sacks was the neurologist and writer who visited Clive Wearing, the musician whose memory lasted only seconds, in the sixth lecture. His patient Edith T. had been a music teacher. She didn't have ordinary Parkinson's disease. She was one of the patients in his book Awakenings, all survivors of the encephalitis lethargica epidemic. It had left them with an extremely severe form of parkinsonism.
She told Sacks that her movements had become "wooden, mechanical." In her own word, she had been "unmusicked" by the illness. Often she froze, utterly still, "like a still photo, a frozen frame," until music came to her. The music played in her head, not in the room. She called it "catchy tunes, rhythmic tunes, the sort I loved to dance to." Then she could move again, freely. In Sacks's telling, a companion could do the same. With someone walking beside her, she walked perfectly. The moment they stopped, she stopped too.
That's one writer's account of one extreme case, and it isn't a technique. What it shows is a pattern. A body that can't set itself going can still be set going by a rhythm, even one in the head, or by a person walking alongside.
The fifth lecture found two parts of the movement system working harder for a beat: the basal ganglia, the deep timing machinery, and the supplementary motor area, a planning area. Parkinson's damages the basal ganglia, and people with Parkinson's are worse at telling apart rhythms with a beat.
An outside signal that helps a person start a step and keep going is called a cue. A 2018 review of walking to a beat in Parkinson's describes those two areas as the brain's own, inner cueing system. In Parkinson's, that inner cue, the brain's own clock for movement, works poorly. Steps get shorter, and walking slows.
Why would a beat help, then, if the disease damages the machinery that keeps one? The music researcher Michael Thaut began studying walking to a beat in the early 1990s. On his account, a steady beat from outside gives the brain a second clock. I'd add that the brain doesn't have to keep that clock running. A metronome marks every beat by itself. Thaut argues that the steady gap between beats tells the brain how long each step should take. So at any point in a step, the brain knows how much time is left, and it can plan the movement to fit.
Thaut's is one account among several. The 2018 review lists others, and says the benefit may have several causes. It also guesses that movements cued from outside may take another pathway around the damaged circuit, and it presents that as speculation. Nobody yet knows how the outside clock gets to the legs.
How much does it help? The clearest single test here is a trial in three European cities that took 153 people with Parkinson's. A lottery decided who trained first and who waited. A therapist came to each patient's home nine times over three weeks. Each patient wore a device built for the trial. It could give a beep in an earpiece, a flashing light on a pair of glasses, or a pulsing buzz under a wristband. Patients tried all three and trained with the one they liked. Two-thirds chose the beep, and a third the buzz. Almost no one chose the light. They practiced landing each heel on the rhythm. Their usual medicines carried on unchanged. Then testers measured them at home, without the device. The testers were never told which group a patient was in.
On average, the patients walked five centimeters a second faster after the training, with steps four centimeters longer. Their balance improved, and freezing eased a little in those who froze at least once a week. They felt more confident about everyday activities, but the activities themselves and their quality of life didn't change. The gains were there when the training ended, and had shrunk considerably six weeks later. The authors call the benefit "small and specific." They couldn't rule out that some of it came from a therapist's attention. Still, they conclude that cueing training may be a useful addition to the treatment of walking problems in Parkinson's.
A 2018 review of fifty studies found the same pattern. Forty-six of them reported gains, but the gains came mainly where people trained with a beat for weeks, not where they simply walked to one. A later review that kept only trials with a lottery found strides about five centimeters longer. A beat from outside, it seems, can do some of the timing that the brain's own clock has lost. With weeks of practice, it buys a few centimeters of stride, on average.
After a stroke, walking is often slow, lopsided and tiring. A team from several American centers tested a system that turns pop music into a walking coach. Each walker wore a headset and a sensor on each shoe. The songs came from many styles, rock and roll, pop and disco among them, each picked for a strong beat. The system set the tempo to the walker's own pace. It nudged the tempo up only while the walker kept in step and walked evenly. Steps that drifted off the beat brought in a louder rhythm track.
A lottery split the patients into two groups. One walked with the system for half an hour, three times a week for five weeks. The other did the same walking without it. The patients were about eight years past their strokes, on average, and all of them already walked without help.
Over the five weeks, the music walkers sped up by fourteen centimeters a second, on average. The others sped up by six. The trial's own bar for a gain that matters clinically was sixteen centimeters a second. The music group's average fell short of it. Still, four in ten of the music walkers cleared it, against one in eight of the others. The company that makes the system paid for the trial, and several authors work for it or advise it. Everyone knew who had the music, and nobody was tested again afterwards.
A 2017 review of music after brain injury pooled nine trials of walking to a beat after stroke. After training, the people who walked to a beat were faster than the comparison groups by about nineteen centimeters a second, on average. In the walking-coach trial, that gap was about eight. But the reviewers rated the evidence on speed as moderate in quality, and they judged nearly all the trials in their review at high risk of bias.
Does the music matter for walking, or only the beat? No trial in this lecture set the two against each other directly. The 2017 review compared trials that used music with trials that used a bare beat, like a metronome's. In walking speed, trials that used music showed a gap over the comparison groups nearly twice as large. The reviewers say only that music may work better, and call for trials that set the two against each other. They note that music can motivate exercise. For walking, the beat is the part known to help, and the music may add something on top.
Walking is where the evidence is clearest. On average, in trials, a steady beat helps people with Parkinson's or after a stroke walk a little better while they train. Words are a stiffer test. Talking is movement too, and the mouth, the breath and the voice have to work together. Can a beat time that?
The second lecture met people with aphasia, a loss of language after brain damage. Some could no longer talk, and yet could sing a familiar tune. In the reading of Isabelle Peretz, the Montreal psychologist from that lecture, the words usually fail, sung or spoken. But some patients have been reported singing words too. In 1973 that observation became a treatment, first described in three patients. It's called melodic intonation therapy.
A patient sings short, everyday phrases on just two notes. The higher note goes on the syllables that speech would stress. The patient taps each syllable with the left hand, at about one syllable a second. Sessions are one-on-one with a therapist, who starts with a word like water and works up to asking for a cup of coffee. The therapy's inventors thought singing would bring in undamaged areas on the brain's right side, to make up for damaged language areas on the left. Its supporters wrote in 2008 that this hadn't been proven.
Gottfried Schlaug was a neurologist at Harvard Medical School. In 2008 his team compared the therapy with plain spoken practice in two men who could say very little after a stroke. Each spent an hour and a half a day, five days a week, with the same therapist. One sang and tapped. The other spoke the same phrases and didn't tap. Both improved, and the man who sang improved more. Schlaug's team suggested that the tapping might pace speech the way a metronome paces other movements. But the comparison took away the tune and the tapping at once, so it couldn't say which mattered.
Benjamin Stahl and his colleagues in Leipzig tested the tune against the rhythm. They worked with seventeen stroke patients in Berlin who understood much of what they heard and could say little. Each patient sang or spoke along with a recorded voice, over a wooden click. In one condition they sang. In another, they spoke the same words to the beat. In a third, the click switched to a waltz's grouping of three and slipped half a beat off the voice, so it no longer kept time with the words. Every syllable lasted just as long in all three. Tapping wasn't allowed. Some of the words came from a well-known German nursery rhyme. Some were everyday stock phrases, and some were strange, poetic lines that nobody would say.
Which did more for the words, the tune or the beat? In that test, singing added nothing. The patients got a little over half their syllables right whether they sang or spoke to the beat. The mismatched click did cost syllables, and the cost fell on the nine patients with more damage in the deep timing machinery. In the eight with less damage there, it made no such difference. The words mattered too. In every single patient, the stock phrases came out better than the strange lines. The team titled its paper "Rhythm in disguise." It suggests that benefits long credited to singing may have their roots in rhythm.
That was one round of testing, with seventeen people. It measured what they could say then, not what months of therapy can do, and it doesn't show the therapy is useless. It leaves two of the therapy's ingredients standing: well-practiced phrases and a beat. What it questions is the tune. One earlier study had found that singing along did help. Stahl's team notes that singing slowed the syllables down in that study, and slower syllables may have done the work.
Here the feet and the mouth meet. The deep timing machinery turned up in both. In Parkinson's, its damage is thought to be part of why walking can lean on a beat from outside. In Stahl's patients, those with damage there seemed to lean on one for speaking. I'm the one drawing that link. The two illnesses damage that machinery in different ways, and Stahl's result rests on seventeen people tested once. But the link suggests the beat does one job in both places. It times a movement the brain can no longer time well.
What happened to Giffords isn't a study of this therapy, but it works the way a case report does. That's an account of one patient, or a few, treated and watched, with nobody untreated to compare. A 2022 review from Vienna and Berlin set the singing therapy's case reports against its trials. Stahl was one of its authors. Of its twenty-two studies, only three were trials that used a lottery. On the same kind of test, the case reports found gains 5.7 times as large as the trials did.
In the trials, the gain was small to moderate, uncertain enough to include zero, and mostly in repeating phrases. For everyday communication, this one technique's trials showed negligible progress, though only two small trials measured it. The same review notes that other speech therapy methods, used together, have shown moderate gains in communication in large trials.
Why would the case reports run so large? Their gains were biggest in patients treated soon after a stroke, and smaller the longer the wait. The authors point to recovery that comes on its own. In the months after a stroke, a brain gets back some language with or without therapy. Researchers call that spontaneous recovery. A case report counts it as the therapy's. That leaves a case report unable to measure the therapy. It doesn't make therapy pointless.
A therapist's attention is a second worry. The Parkinson's trial couldn't rule it out, and a case report can't separate it from the therapy either. The review's authors still see a use for the singing therapy. They write that for someone with severe aphasia, a set of trained phrases for everyday needs may bring a substantial gain in quality of life.
And Giffords? One recovery has nothing to measure itself against, so nobody can say how much of hers the music did. Her therapist thought that her love of music may have helped her through the hard work. In 2021 Giffords gave PBS an interview she had prepared for over several weeks with her speech pathologist. In it she said: "The words are there in my brain. I just can't get them out."
So music can help a damaged brain, a little, mostly through its beat, and mostly while people practice. For walking, the trials point the same way, and the gains are counted in centimeters. For words, the tune mattered less than its champions thought, at least in one careful test. Rhythm and well-worn phrases seem to carry more. And a single story runs ahead of the trials, because it can't tell the therapy from the recovery a brain makes on its own.
How long the gains last is still open. The trials here either didn't look, or found the gains shrinking within weeks. So is the question of how much music adds to a bare beat. These are research results, not advice for any one person. For that, the people to ask are a person's own physical therapist or speech therapist.
For most of this course, your brain has been the one making the music. It builds a pitch from a stack of harmonics, bets on the next note and helps supply the beat. Here a damaged brain borrows its beat instead. In the largest trial in this lecture, nobody played any music at all. The patients trained to a beep, and a third of them preferred a buzz on the wrist. What helped them, a little, was the beat.
10Conclusion
Here is what the course built, in the order it built it.
The pitch of a note is a conclusion your brain draws from a repeating pattern of harmonics. The frequency that matches it doesn't even have to be in the sound.
From a scanner's distance, music shares a neighborhood with speech. Up close, some populations seem to answer almost only to music, and one to singing. A stroke can take music and spare language, or the reverse.
A wrong note is mostly a long shot, by odds your brain picked up from the music you grew up with. The ear's own share is smaller than it feels.
A chill marks a moment when the reward system that handles food and money answers those bets, and part of it starts work early. Changing dopamine changed how much people enjoyed music, and what they'd pay for it.
The beat is partly something you supply. The leading explanation, still unproven, is that your brain's movement system keeps it. It runs a little ahead of the beat, even while you sit still.
Practice redraws the brain's maps where the work lands, and an early start probably counts for something.
Lessons make children better at music. The famous gap in IQ is mostly about who takes lessons.
Where music helps a damaged brain, it helps a little, mostly through the beat, and mostly while people practice.
The course began with two students and Barber's Adagio. The same notes reached both. What differed was in the listeners. Each had brought a piece that reliably gave them chills. Parts of the reward system rose as those chills grew stronger. Much of music happens in you.
Plenty is still open. Whether dopamine is the pleasure itself, or only the wanting, is disputed. So is which animals can keep a beat, and how much an early start matters. Several famous findings shrank when someone checked them, among them the Mozart effect and the imagined waltz in the brainwaves. What survived is smaller than the first reports.
Sources & References
This course draws from the following sources. Visit them for additional depth.
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗cga.ct.gov — 98 R 0888 ↗webpage
- 🔗libra.unine.ch — Download ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗deutsch.ucsd.edu — Pages ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗newyorker.com — The Abyss ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗nbcnews.com — Wbna45322728 ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗appstate.edu — Mozart N ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗research.gold.ac.uk — Qjep ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗ebi.ac.uk — Search ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗cns.nyu.edu — Pitch ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗medscimonit.com — 883470 ↗webpage
- 🔗nimh.nih.gov — Depression ↗webpage
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗
- 🔗frontiersin.org — Full ↗webpage
- 🔗nature.com — 33918 ↗webpage
- 🔗
- 🔗
- 🔗
Listeners Also Played
Terraforming Mars: Can We Really Make the Red Planet Habitable?
A clear-eyed tour of one of the boldest ideas in science: turning Mars into a world humans could live on. We trace terraforming from pulp science fiction to peer-reviewed research, weigh the hard physics against the latest engineering and biology, and ask what's actually achievable in our lifetimes versus what stays a multi-century dream.
The Biology of Aging: What Science Actually Knows About Why We Get Old
A rigorous yet accessible deep-dive into the cellular and molecular science of aging — from evolutionary theories and the Hayflick limit, to telomeres, epigenetic clocks, cellular senescence, and the mTOR pathway — separating well-supported science from longevity hype.
The Human Gut: Microbiome Science and What It Actually Means for Your Health
A rigorous but accessible tour through the science of the gut microbiome — what it is, how it develops, what disrupts it, and what the current research genuinely supports versus what's hype. You'll come away with a clear-eyed framework for evaluating probiotic claims, dietary advice, and the emerging gut-brain connection.
Want a course that doesn't exist yet? Request one →