December 9, 2024 • 46 mins
How can we understand music's effect on human brains? Is music universal or does it rely on your experiences? How is music similar to a language? Can music be leveraged to help anxiety, dementia, or Parkinson's disease? What does any of this have to do with Stevie Wonder on the high hat, or the relationship between music and color? Join Eagleman with guest Daniel Levitin -- neuroscientist, musician, and author of This Is Your Brain on Music and I Heard There Was A Secret Chord.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:05):
How can we understand what music is about from the
point of view of neuroscience. Can music be leveraged to
help with anxiety disorders, or with dementia or with Parkinson's disease?
Is music universal or does it have to do with
what you have absorbed in your lifetime? How is music
(00:26):
like a language but one with very particular structure and
therefore high predictability. And what does this have to do
with Stevie Wonder on the High Hat or the relationship
between music and color. Welcome to Inner Cosmos with me
David Eagleman. I'm a neuroscientist and an author at Stanford
(00:47):
and in these episodes we sail deeply into our three
pound universe to understand why and how our lives look
the way they do. Today's episode is about music and
(01:08):
the brain, and this is a topic that has been
requested by several different listeners, so I prioritized making this episode. Music.
I suspect is a popular topic because music can be
so emotive for us and so catchy and so meaningful.
My father, for example, who was the quintessential tough guy
(01:30):
when he would listen to Mozart or Brahms or Beethoven,
he would have tears streaming down his cheeks. And as
a child, I didn't have much understanding of classical music,
but that really caused me to wonder, what is going
on here? Why does this music wafting out of the
radio evoke such strong emotions and perhaps such deep memories
(01:54):
in my father? And as I got older and became
a neuroscientist, I wondered, is there something you unique in
the structure of the human brain that ties music so
closely with our emotional experiences? So I decided to do
an episode on this today, and I realized there was
no one better to ring up than my friend and colleague,
(02:14):
Daniel Leviton. He's the founding Dean of Arts and Humanities
at Minerva University in San Francisco and a professor emeritus
of psychology and neuroscience at McGill University in Montreal. You
may know Dan because he wrote a book called This
Is Your Brain on music, the Science of a Human Obsession,
(02:35):
which became a big New York Times bestseller, and he's
also written four other best selling books, including his latest,
which is called I Heard There Was a Secret Chord.
Music as Medicine. Now, as you may suspect. Dan is
also a very talented musician. He composes music and he's
worked on albums by Blue Oyster Cult and Chris Isaac
(02:56):
and Joe Satriani, among many others. He does this as
an advisory producer a recording engineer. So given his expertise,
I wanted to sit down with Dan to get his
take on music and the brain. Okay, Dan, what goes
on in the brain when you listen to music?
Speaker 2 (03:19):
Yeah, well, it really is quite fascinating. So it begins
with the sound waves impinging on your ear drums. They
wiggle in and out, and all the information you have
about the auditory world comes from molecules vibrating in some medium,
in our case air, it could be underwater, and then
(03:39):
your ear drums just wiggle in and out, and then
your brain has to take that wiggling in and out
and extract from it all the different sounds a bird chirping,
a leaf blower going, the oboe in the symphony as
opposed to the French horns in a crowded room, the
conversation you're trying to listen to in front of you,
as well as that when you're eavesdropping in. It has
(04:01):
to separate all that out. And the way it does
that is that your brain has special processing circuits. I
was going to use the word designed, but of course
they weren't designed, but evolved to do different distinct functions.
One circuit processes the loudness is anything louder, is it
(04:21):
getting softer? And it follows that loudness trajectory, which can
be an important cue as.
Speaker 3 (04:28):
To what's going on. Pitch duration.
Speaker 2 (04:32):
The pitches get in a separate circuit, bound into a
representation of melody and harmony, the durations into a representation
of rhythm and meter, the loudness into accent structure, timbre,
which is the quality that distinguishes your voice from my
voice when we're saying the same thing, or a trumpet
(04:53):
from a piano when they're playing the same note. That's
a combination of spectral temporal information, in other words, pitch
and time and loudness. It all comes together later in
the brain and you just hear that song where later
in the brain is maybe forty milliseconds. It happens so
(05:14):
seamlessly that it just sounds like we're hearing the song,
but we're not. Our brain is hearing the pitch, the rhythm,
the loudness, the timber, and our evidence, David, that this
happens is not just from neuroimaging, but from patients. We
see patients with focal brain damage who suddenly lose their
perception of pitch, but they retain rhythm or vice versa.
Speaker 1 (05:38):
Okay, so it's this terrifically complicated computational process. So why
does it end up feeling so emotional for us?
Speaker 3 (05:46):
Well?
Speaker 2 (05:46):
So I think there are both neurobiological reasons and evolutionary
reasons that are connected. Of course, the simple answer is
it's emotional for us because emotion circuits in the brain
are involved in music processing. And by that I mean,
among others, the well known reward center that you and
(06:09):
I have talked about in our own classes over the
years and with one another, the limbic system, the structures
like the nucleus of Cumban's, the amygdala, the ventral tegmental area.
This is the emotional core part of the reptilian brain
that you know motivates us to move out of the
way of some approaching danger or to signal pleasure when
(06:30):
we're hungry and we finally get a taste of something sweet.
Music activates that same center, and in fact, it was
my lab that was the first to show that our
brain produces dopamine in response to music listening, and later
we showed that our brain produces its own endogenous opioids
in response to music listening, all part of that well
(06:51):
known pleasure network. Now, of course, that raises the question
why why is you know, over ever solutionary timescals did
music hit that emotional center and hear The answer is
by fer Kate. One is that the well known startle response.
(07:11):
You hear a sudden, loud noise and you jump. That's
evolutionarily adaptive because you know, you know, even lizards, snakes,
reptiles have to move out of the way of something
that might step on them or smash them. And that
startle response in humans goes directly from the inner ear
(07:35):
to the cerebellum and the brain stem. Before we even
figure out what the sound is, we startle and that's
connected to emotion centers. And so music, because it's an
auditory stimulus, is hardwired to movement and to emotion.
Speaker 1 (07:51):
So what happens in the brain when you play an instrument.
Speaker 2 (07:54):
Playing instrument is one of the most neuroprotective things we
can do. Glistening activates every area of the brain that
we so far mapped, as does playing an instrument. But
the added advantage of playing an instrument is that it's
active rather than passive, and it involves prediction centers in
the prefrontal cortex, particularly broad An area forty seven, which
(08:19):
is a pattern detector. You know, Venode, Menna and I
at Stanford for thirty years have been looking at this
little sliver of tissue on either side of your between
your top of your ears and your eyeballs broad and
forty seven, and Michael Patritus and I at mcgil have
also looked at it as that part of the human
brain that primates lack that allows us to process temporal patterns,
(08:45):
either in vision, touch or sound. And so in playing
an instrument, you've got to plan what you want it
to sound like. Hopefully you've got some idea of what
you want to come out, and then you've got to
use a feedback loop to listen to what came out
and see how well it matches with what you intended
(09:06):
and that adjust or not. And Broaden forty seven is
a part of that, as well as other prefrontal areas
and temporal areas.
Speaker 1 (09:15):
Before we move on some other questions, I want to
ask what got you into this intersection between music and
the brain.
Speaker 3 (09:23):
It wasn't intentional.
Speaker 2 (09:25):
I had dropped out of college after my sophomore year
to play in a series of bands, and that led
to me becoming a staff producer at Columbia Records in
the eighties and in the nineties, when the music business
seemed to be imploding, a bunch of us who had
entered the business around the same time, figured we needed
(09:46):
a plan B, that this may not be a sustainable career,
and so I went back and finished my bachelor's degree
at Stanford. I just worked in every lab that would
have me. Then I went to graduate school and Oregon
in order to work with Doug Hintsman and Helen Neville
a language specialist, and of course Mike Posner on neuroimaging.
(10:09):
And I was just doing all these things in parallel
and loving it. I mentioned all these names because they
were very important mentors to me, each of them. And
in the third year of my graduate program, Posner, who
was my principal advisor, said, you know you're gonna have
to specialize. What do you want to do when you
(10:32):
grow up. I said, well, I really love all of this.
I love psycholinguistics, I love memory, I love decision making,
and I had worked in all those areas. And he said, well,
you know, you have this background as a musician, and
there's a lot of competition for jobs in those other fields.
There's this emerging field of psychology of music with a
(10:54):
handful of people in it. Maybe if you go into
that field, there'll be a lot of low hanging fruit,
as it were, a lot of studies that have obviously
need to be done that haven't been done yet. And
in addition, although nobody's going to advertise for a music
psychology faculty member, all those other things apply to music psychology,
(11:15):
decision making, how do we decide what we want to
listen to? Cyclel linguistics, psychology of lyrics and music, individual
differences in musical taste, memory for music. So it was
Mike who said preciently that I should brand myself as
a music psychologist. Of course, back in those days, there
were no neuroscience departments, there were no neuroscience programs. You
(11:37):
could not study neuroscience as you and I know it.
And I'll make a distinction for our listeners between like
molecular neuroscience, which was done in biology departments where you
only look at a single neuron and you've never even
considered what a thought might be and what we call you.
And I work in the area systems neuroscience, where we're
looking at the big ideas andions of neurons communicate with
(12:01):
other millions in neurons.
Speaker 1 (12:03):
I know that you feel like that the music psychology
world was not particularly good let's say twenty years ago,
so give us a sense of what it was like
at that point when people thought about I'm going to
study the psychology music and what it's like now and
what's changed.
Speaker 2 (12:19):
And there were a bunch of other people who hadn't
studied cognitive psychology but fancy themselves music psychologists, and they
did a bunch of bad studies, the poster child for
that being the Mozart effect study, which purported to show
that listening to Mozart for twenty minutes would make you smarter.
(12:40):
And there have now been literally one hundred studies that
shows that that was just bullshit. It was a poorly
controlled study, It was done by people who had no
experience in human experimental design. They stepped outside their lane,
and so there was a lot of garbage work being done.
What's different now is that we've had twenty years of
(13:02):
people applying to graduate school who knew they wanted to
study music psychology. Some of them went into music psychologists'
labs like mine. Others did what I did. They went
into the lab of a memory person or an attention
or brain imaging person and just used that interest in
(13:23):
music to design studies.
Speaker 1 (13:25):
And so what sort of things have come out in
the last twenty years. Part of this has to do
with the advent of neuroimaging, right.
Speaker 2 (13:33):
Well, that really, that really is what kicked it off, because,
as you know, David, the study of emotion was rather unseemly.
I think it began with the foundation of the first
psychology labs in the world by Vuntenfeckner in Europe. Of course,
William James was always more interested in the esthetics and
artistics side, but it was the behaviorist movement of the
(13:56):
fifties led by BF Skinner, that if something wasn't deservable
and replicable, it wasn't worth study. And so emotions just
seemed too squishy, and music as the language of emotion
seemed like the squishiest of all. What happened with the
first studies of neuroimaging, which Mike Posner was part of
(14:16):
nineteen ninety eight ninety nine. We were able to actually
see pictures of the brain caught in the act of
thinking and remembering and imagining, and that gave it a
biological basis reality. You could replicate brain imaging experiments, and
so first the study of emotion, and then shortly followed
(14:41):
by the study of music with the first neu imaging
studies of music in the early two thousands, that put
it back on the table as something that was worthy
of study and could be done in a rigorous fashion.
Speaker 1 (14:54):
So your latest book, I heard, There was a Secret Chord,
looks at this issue of muse musick as medicine. So
tell us about that.
Speaker 2 (15:03):
That's something that I think most of us have experienced intuitively, certainly,
it goes back tens of thousands of years using music
to treat injury and disease to shamans and faith healers
and indigenous tribes. And it was really in the last
ten years. I would say that the idea that music
(15:27):
had an evidence base for treating injury disease promoting wellness,
helping with mental disorders like depression, post traumatic stress disorder, anxiety.
There's been eight thousand papers in the last two years
alone in peer review journals on medical applications of music,
and so about five years ago I became involved with
(15:51):
the National Institutes of Health and the White House Science Office,
leading various expert panels to figure out what do we
really know and what do we don't know and what
remains to be done. That led to a call for proposals.
The NIH put forty million into music and medicine research
a few years back. So I looked for a book
(16:12):
on music and medicine because I wanted to know what
the state of the art was, and I couldn't find one.
There were a lot of papers, but no books, and
so I ended up writing the book I wanted.
Speaker 1 (16:22):
To read and so give us a sense when it
comes to something like, let's start with dementia, how would
music be useful in the case of something like Alzheimer's.
Speaker 2 (16:34):
Not all dementia is Alzheimer's, of course, and not all
memory loss comes from Alzheimer's. But the most straightforward case
of somebody with profound memory loss, perhaps due to Alzheimer's,
Corsokov's stroke, whatever. They may not recognize where they are,
They may not recognize loved ones. They may not even
(16:55):
recognize themselves in the mirror. This is profoundly We've seen
patients who will walk by a mirror and think they're
talking to someone else, and then they get angry because
the person in the mirror appears to be mocking them
by gesturing the same way they are, and it causes
one of two reactions. Individuals either turn in on themselves,
(17:19):
fold in on unsells because the external world makes no
sense and they become somewhat catatonic, or they become angry, agitated,
and violent and in that case have to be medicated.
They might start beating up on their spouses, not recognizing them.
Music follows a kind of principle that computer science talks about,
(17:43):
which is first in, last out.
Speaker 3 (17:45):
This is a holder for the old models of computer memory.
Speaker 2 (17:48):
The first thing that goes into the memory is the
last thing to come out. And because we've been listening
to music in the womb most of us and through
our childhoods, those memories the most deeply embedded in the
brain and the most resilient and resistant to decay or damage.
And so if we play music from the youth of
(18:10):
an Alzheimer's patient, somebody with profound memory loss, play the
music from say the ages of twelve to fourteen or
sixteen that's preserved in most cases, and it allows them
to profoundly reconnect with a part of themselves they had lost.
It eases them, it comforts them, It triggers memories that
(18:35):
had been buried, and that kind of therapy or intervention
can pull them out of the state they're in and
actually have consequences for days or weeks where they come
alive again.
Speaker 1 (18:49):
It doesn't cure or help the dementia the cognitive loss exactly,
but it triggers memories and pulls them back to state
where they've been. And it can also revivify skills that
someone has. For example, musicians with profound dementia who get
(19:11):
an instrument put in their hand and they go and
play again as though they're young.
Speaker 2 (19:17):
It's really extraordinary. And we saw this play out in
recent years with Glenn Campbell first and then with Tony Bennett,
both of whom had profound memory loss and did not
know where they were. When Glenn did his final tour
with dementia and in the throes of Alzheimer's. He would
sometimes play a song two or three times in a
(19:39):
row because they didn't realize he had just played it,
or he'd forget what song he was supposed to play.
But once the notes, the first few notes happened, he
knew where he was. Say, with Tony Bennett, he could
sing for an hour and a half without stopping once
the music took over. These are what we cognitive scientists
call overlearned. They're not just in memory, but they're in
(20:03):
memory with thousands and thousands of traces overlaid on top
of one another and their procedural memory. Once your vocal
cords and your fingers get going, they kind of take over. Now,
to be clear, the memory is not in your fingers,
although it feels that way. If I were to scoop
your brain outside your head, your fingers would not keep
playing like a chicken with its head cut off. But
(20:24):
those pathways are so profoundly deeply embedded that yeah, you
can keep going even with Alzheimer's, and it gives the
patient a rare act of competence in a world in
which they're otherwise incompetent. It gives them agency in the world.
This can really affect their mood and their quality of
(20:47):
life and way of being in the world.
Speaker 1 (21:04):
Now, one of the things I've talked about on in
previous episode is right Bo's law, which is where older
memories are more secure, they're burned down more deeply than
newer memories. And of course we see this with people
with cogno.
Speaker 2 (21:17):
I thought you were going with Bow's law, like the
bow on a violin, the earliest violin pieces are the
ones that are the most embedded.
Speaker 1 (21:25):
Yeah, nope, Riebo Ribot, Yeah, which is you know, he
was the first I think this is actually the first
rule in neurology, as in the oldest rule.
Speaker 2 (21:35):
Yeah.
Speaker 1 (21:35):
But anyway, he saw that, you know, things from childhood
were remembered by people with let's say dementia, and things
they did last week or last month were not remembered,
so older members more stable. Now, the reason this is
so strange is because nothing else works that way. Institutions,
for example, don't remember their older stuff better than they
(21:56):
remember their newer stuff. But brain works this way.
Speaker 3 (22:01):
That is true.
Speaker 1 (22:02):
So my question is, I mean, when when it comes
to music, presumably most of these great musicians have been
playing since they were a little kids, and these particular
songs are overlearned, as you mentioned. Is this an expression
of simply of Ribo's law, which is that it's an
older memory and that's why they're able to do it?
Or is there something different about music than if I
(22:24):
ask them something about their you know, their their childhood home.
Speaker 3 (22:28):
What a great question.
Speaker 2 (22:29):
Well, there is something different about music to invoke Claude Shannon.
It's a highly organized and structured stimulus like language, and
so in an information theory perspective, which is Shannon, your
ability to predict what will come next in any sequence
(22:54):
defines how structured highly structured it is. Even more so
than language. Music is highly constrained. So I could say
a sentence to you like this, Let's try this, the
pizza was too hot to blank? What comes to mind?
Eat yeah or touch yeah? I would not be likely
(23:15):
to say the pizza was too hot to sleep. Once
I say it, you understand what I meant, and you
understand that I use the correct part of speech. I
put a verb at the end, But it's an unlikely outcome.
Music is even more constrained because there are only twelve
notes in our scale, and there are I wouldn't call
them laws of music theory, but customs of music theory.
(23:38):
And there are rhythmic rules or customs.
Speaker 3 (23:41):
And so once you.
Speaker 2 (23:43):
Get going on a piece of music, its own structure
constrains what the possible completions are, making it easier to
remember and then easier to recollect, easier to store and
easier to retrieve, and then moreover, as its own internal tempo.
Once the beat is going, it's carrying you along, whether
(24:06):
you're ready to go along with it or not. And
so you're going to fill those slots with what needs
to go there or your best approximation for it. And
when you look at performing musicians, like typically the ones
that work holiday ins on Friday night lounge bands and stuff,
they might know two thousand songs where I would use
the word no in quotes, they probably don't know every
(24:30):
single note and every single rhythm.
Speaker 3 (24:32):
But they can approximate it.
Speaker 2 (24:34):
They can improvise and estimate it so that you don't
really notice the difference.
Speaker 1 (24:38):
Well, that's actually a good segue into Parkinson's disease.
Speaker 2 (24:42):
How is music used there? So in all these cases
of music is medicine. We're not talking about like a
music module in the brain or a music medicine prescription
that's straightforward, because music is doing different things in different
parts of the brain, different aspects of the music are
doing it, and Parkinson's is I'm so glad you brought
(25:04):
this up. It's actually the best case for understanding this
differentiation in Parkinson's disease. At some point, most patients will
experience difficulty walking movement disorders in general, but walking in particular,
and it's because the disease degrade circuits in the basal
ganglia that are required to maintain a smooth and steady
(25:27):
gait and to orchestrate the movements of one foot has
to go after the other and you have to put
it down at a certain time where you end up
with both feet in the air at the same time,
and that's not good for walking. And the circuits that
allow you to walk rely on an internal intrinsic timer
in the brain, a clock, and that's what gets degraded.
(25:47):
If you listen to music that has the tempo of
your natural gait, you have neurons, neuronal clusters that were
not damaged that synchronize to that pulse, and then they
can act as an external clock that allows you to
walk smoothly and continuously.
Speaker 1 (26:07):
What else besides dementia and Parkinson's, where else do we
see therapeutic effects?
Speaker 2 (26:13):
Well, I think one of the big ones is an anxiety.
Dentists figured this out a long time ago. They play
you what's supposed to be relaxing music to reduce your anxiety,
which reduces swelling and inflammation.
Speaker 1 (26:28):
And why does that work?
Speaker 2 (26:30):
Some music we find to be relaxing, some we find
to be stimulating, some we find to be inspiring. And
the difficulty is, there's no one music that will do
those things for everybody. In fact, there's no one song
everybody likes. There's no one song everybody hates. It's subjective,
(26:50):
like your taste for food. You know, why doesn't everybody
like Indian food? I love food, Not everybody does. You
have our own taste. It seems as though we have
an esthetic module each of us that governs things like
our taste and colors and people and tastes and music.
And we actually showed in a PNAS paper that a
(27:12):
person's preferences for certain musical combinations is correlated with their
preferences for certain color combinations, which seemed completely crazy to me.
It just seems so far fetched. But there's this underlying
aesthetic module that we can't account for yet.
Speaker 1 (27:29):
What was the argument there, did you guys forward a hypothesis?
I read no?
Speaker 2 (27:33):
Oh well, I mean there was some hand waving. Yes,
there's an aesthetics module, and yeah, I mean to some extent,
it had to do with consonants and dissonance. Do you
like to see contrasting colors and contrasting chords? Do you
like to see things that are more consonant and harmonious?
But sharp edges and the metaphorical sharp edges and music.
(27:56):
But apart from that, it was pretty speculative.
Speaker 1 (27:59):
So let me double click on this issue about individual differences,
because one thing that's clear is across the population, some
people don't really like music that much. Other people love music,
it's a big part of their lives. But on one
on the spectrum, you just it's sort of meaningless to
many people. So how do you interpret that?
Speaker 2 (28:15):
Well, I just look at this as as a necessity
of Darwinian theory, which is that we can't all be alike,
or we'd you know, genetically or behaviorally or we would
all be wiped out by a single opportunistic virus. So,
you know, a cornerstone of Darwinian theory, as you teach it,
as I teach it, is descent with modification and random mutation.
(28:39):
And so you know, most random mutations end up being unobservable.
Some of them we see a phenotypic variation, that is
a behavioral variation. And in that case, yeah, ten percent
of the population probably don't like music, and they don't
understand why the rest of us spend so much money
and time on it. Ten percent of the population probably
(29:00):
don't like chocolate. I find that so impossible to believe,
but they don't. And then there are you know, some
percentage of the population don't like sex. They tend not
to pass that on through reproduction, but that trait, but
it's the way it works.
Speaker 1 (29:16):
How do you think music fits into the story of
evolution and human evolution in particular.
Speaker 2 (29:24):
So Stephen Pinker famously threw down a gauntlet in nineteen
ninety seven when he said that he thinks that music
has nothing to do with evolution, that it was just
sort of a byproduct of other things that we developed,
like language. He called music auditory cheesecake, And what he
(29:45):
was saying was that, well, we didn't really evolve to
like cheesecake. We evolved to like sweets and fats because
in the very small amounts they were available across evolutionary
time periods to it was adaptive to seek them out. Now,
if you get cheesecake, it'll just you know, you know,
(30:06):
spike your blood sugar levels and be bad for your health.
Speaker 1 (30:10):
I'm curious if you have a different view on it,
whether there's a role in evolution for music.
Speaker 2 (30:15):
We do have some data and I'd like to share
that with you, our and our listeners. So to begin with,
I mean, just as a resource, I would mention Stephen
Mithen's book The Singing Neanderthals, where he makes the case
that music was a proto language that preceded, you know,
linguistic language speech. I would say the evidence that we
(30:36):
have from neuroanatomy is that, from the work that Vanode
Menon and I have done in others, those circuits that
are engaged with music, listening and performing are phylogenetically older
than the speech circuits. And that's the reason why Gabby
Giffords was able to recover speech after she was shot
in the head and lost the ability to speak. It's
(30:58):
called melodic intonation therapy. We can take somebody who's lost
speech ephasic expressive aphasia and teach them to sing what
they need to communicate.
Speaker 1 (31:08):
So give us an example of that. What Gabby Gifference does.
Speaker 3 (31:11):
Well.
Speaker 2 (31:11):
She could not speak, but she might have been taught
things like I need a glass of water, show me
to the bathroom, I'm ready for bed now. She could
sing those things perfectly, and through neuroplasticity, the brain rewired
itself by passing those damaged circuits using the intact music circuits.
(31:32):
They are phylogenetically, that is, evolutionarily older. That's one piece
of evidence. Another is if we look at contemporary hunter
gatherer preliterate societies or all around the world that have
been cut off from Western civilization, and we make the assumption, David,
that they're living life now pretty much as they have
for ten or twenty thousand years, and all of them
(31:55):
use music for a number of things, not just a
single thing, but the emerging ideas that music evolved as
many things did, not for a single to solve a
single adaptive problem, but multiple problems, and one of them
was how do you encode knowledge in a pre literate society.
We've only had a written language for five thousand years.
(32:16):
We've been on the planet ten or twenty times. As
long as that, we still had to remember things like, oh,
don't go over that hill there, because my grandfather went
over there and they killed him because they're very vicious
and warring, and so you know, don't go there. And
when this water well runs dry, this is a route
to the other well, the supplementary one. Don't eat that
(32:39):
plant unless you boil it in this particular way. This
kind of knowledge is embedded in song in pre literate
hunter gatherer tribes and probably has been for a long time.
Mothers soothing their infants to imprint their infant on their
voice so that if they become separated, the infant will
know the sound the mother's voice. So we've got knowledge,
(33:02):
we've got bonding between mother and infant. We have social bonding.
Singing around a campfire to ward off a neighboring tribe
or predators as a way to defuse interpersonal tensions within
a tribe and to protect you from outside invaders.
Speaker 3 (33:18):
Lots of different uses.
Speaker 1 (33:22):
Let me double click on this for a second. Do
you feel that music is universal in terms of when
you compare across culture around the world, or are there
important differences locally both.
Speaker 2 (33:35):
Music is a cultural universal. There is no known culture
now or any time in the past that lacked it.
In David Huron's words, music is marked by its ubiquity
and its antiquity, and there are huge local variations. One
thing in common is we all have the octave, which
(33:57):
is defined by frequencies that are double or having of
one another hundred hertz, two hundred hertz, fifty hertz, all
the same. We perceive them as perceptually very similar um
similar notes. We give them the same name. In Western system,
there's middle C and there's high C, and there's low C,
things like that.
Speaker 3 (34:18):
But the way we divide up the.
Speaker 2 (34:19):
Octave into pieces is different across cultures. We divide the
octave up into twelve pieces, and we tend to use
only five or seven of the notes at a time.
The patterns that we make once we've divided the octave,
the way we combine them, either sequentially or simultaneously melodies
and chords are different. And that's why Chinese opera and
(34:41):
the music of Sub Saharan Africa sounds so different to us.
It's based on different customs and a different system. And
it's not the case that our music is better or
that if only you went into the Amazon and played
the indigenous people their mozart, they would feel that they
had suddenly heard from God himself. It doesn't work that way.
(35:04):
We've done the experiment.
Speaker 1 (35:06):
What is their impression of it?
Speaker 3 (35:08):
Eh? Meh? Yeah.
Speaker 1 (35:10):
Presumably whatever you've grown up with culturally affects your enjoyment
of what you're hearing, right.
Speaker 3 (35:18):
Very much.
Speaker 2 (35:18):
So. Yeah, we imprint on the music we're raised on,
and so we implicitly learn the grammar of our music
the way we implicitly learn the grammar of our language.
As Chomsky had said, here's a little test. Hang on
a second. Okay for the listener, Dan has des grabbed
(35:39):
his guitar, So we all know implicitly, uh scales, even
if we don't realize we do. Now you're expecting another note.
You're probably expecting this one.
Speaker 1 (35:57):
Thank God about that? Yeah, otherwise.
Speaker 3 (36:06):
That doesn't sound bad. But that sounds quite not quite right.
Speaker 2 (36:15):
And we know chords. We know that if I go
it wants to go somewhere.
Speaker 3 (36:28):
I can't just stop I've got to come back to.
Speaker 2 (36:32):
Or or.
Speaker 3 (36:37):
There's some options, but not unlimited options, right.
Speaker 1 (36:42):
And I've talked in many episodes about how fundamentally the
brain is a prediction machine. It's just it is trying
to guess ahead at what's going on. It's got an
internal model of what's coming next. So in those cases,
let's say, with the chords that you were playing, is
it the case that I had certain predictions because of
the culture I've grown up in. To what extent would
(37:02):
that be universal about what chorn comes next?
Speaker 2 (37:04):
Absolutely, and something I very much appreciate about you, because
this is not something that all cognitive scientists cognitive scientists
talk about, but you and I, I think, have been
big proponents for this idea that if the brain is
if the brain is nothing else, it is a giant
prediction machine. That is its job. It's just based on
(37:25):
what we've heard over and over and over again. And
the job of the composer and the performer is to
reward those expectations just enough of the time that you
feel like you're following along with the story as it unfolds.
But they have to surprise you just enough of the
time that you've learned something, or you get that little
(37:47):
hit of oh, isn't that interesting? As much as I
was trying to predict what comes next, they came up
with something even better than I could have imagined.
Speaker 1 (37:57):
Just like with everything in life, there's a spectrum between
novelty and familiarity and all the sweet spot is in
between that. Somewhere on that note, what is the reason
(38:22):
that we care for rhythm so much? Do you feel
as I do that it has to do with predictability,
where the brain says, oh, now I know what's coming next.
There's the next beat, in the next beat, and there's
something very satisfying about having some structure of prediction, and
then there's surprises thrown on top of that.
Speaker 2 (38:42):
It's because the brain's a prediction machine that rhythm is
meaningful to us. There are populations of neurons that fire
in synchrony with the beat, with the tempo that sets
us up for movement. And I mean that metaphor and literally,
music is possibly the only art form that makes you
(39:05):
want to wiggle your body in response to it. People
aren't standing in front of the Mona Lisa and dancing.
Speaker 3 (39:10):
Although they do, they.
Speaker 2 (39:12):
Seem weird and they might get kicked out of the louver,
but we dance to music because we can't help it.
And in the neuroimaging studies I've done all of them,
where we ask people to stay still, we still see
activity in their premotor cortex and their motor cortex that
they're trying to suppress. And then on the metaphorical movement, yes,
rhythm is important because it's telling us that there is
(39:35):
more to come and we want to know what that is.
It sets up a narrative momentum and when a musician
could play around with that rhythm. If you listen to
what Stevie Wonder does and the opening to Superstition, I
can't replicate this exactly, but he's playing around on the
high hat. He's setting up a beat. The high hat's
that little symbol and ordinarily a normal drummer would just go,
(40:02):
but Stevie doesn't goes. He's playing around. No, two times
are the same, and we may not notice it because
he keeps the underlying pulse there, but it is ear
Candy Man. It is just so interesting for the brain
(40:23):
to take all that in, even at a subliminal level,
that there's so much going on, and not only is
he changing in the rhythm. He's moving his stick around
so he gets the bell of the symbol, he gets
the edge, he gets the middle, he gets different sounds
out of it.
Speaker 1 (40:38):
So this is a good segue into something I've been
wanting to ask you, which is, what is your take
on artificial intelligence and music the future of And there
are two ways to think about this, of course. One
is AI composing music. Another is AI finding the perfect
music for you. Maybe there are other ways to think
about it as well, But tell us your take on
(40:59):
the future.
Speaker 2 (41:00):
AI composing music. I look at it this way. I
have a friend, a novelist, Gail Jones, wonderful novelist, who
says that AI music now is like those artificial flowers
at a holiday inn. You walk in the lobby, you
see this big display and you go, wow, isn't that nice?
And you get up close and you realize they have
(41:21):
no nice odor and they're plastic, And so AI music
at a distance probably sounds just fine. And it's already
crept into advertising on social media. It's being used in
the background. It's sonic wallpaper. It's kind of like the
painting in the bedroom of that same holiday in room
it's like, Okay, it's covering the wall. It's kind of nice,
(41:43):
but I'm not going to sit there and stare at
it for hours. I think that you and I subscribe
to Dan Dennett's functionalism, the idea that in theory, all
of our thoughts, hopes, desires, and beliefs, all of our
mental activity can be redo to brain activity, and that
brain activity can be characterized by patterns of neural firings
(42:05):
and connections if you could replicate a human brain and
add in the right amount of random factors. I think
in theory, yes, AI music could be great. I don't
like the thought of it, but I mean that's the reality.
But for now, I don't think AI music is going
to be a threat to real music and real feeling.
(42:25):
But you nailed it on the head. Where I think
AI can be the most use is helping us to
find music that we like. Most of us listen to
a couple of thousand songs maximum, over and over and again,
and occasionally let in a few new ones, and those
can be comforting and rewarding, but they can dig deep
(42:47):
neural ruts such that we get tired of them and
we want something new. There are now two hundred million
songs across the streaming services, with a one hundred thousand
new ones being uploaded every day. So how do you
find what you like? The major streaming services have recommendation systems.
(43:08):
They don't work particularly well for me, But AI in
principle can extract hundreds of features latent features for music
and build a multi dimensional model, a higher dimensional manifold
of musical structure and DNA that might have one hundred
and fifty orthogonal dimensions, and set up a kind of
(43:31):
universe of music where things that are where each song
is like a planet, and planets that are near each
other are going to be similar, not just structurally, but
in an emotional space where they're likely to cause a
similar emotional reaction in a given listener, knowing what your
own personal space is, right, It's not going to be
(43:52):
something that is entirely objective and prescriptive and one one
hundred and fifty dimensional map for everybody. They'll have to
be to maps for each person, because your tastes are
different than mine. As my grandfather used to say, if
everybody liked the same things, they'd all want to get
with your grandma, I'm going.
Speaker 1 (44:12):
To steal that line if you don't mind, when I'm
a grandfather. What have I not asked you that I
should ask you?
Speaker 2 (44:19):
I wanted to ask you of the many successful books
and highly regarded books you've written, did you write them
because you wanted to read them and you couldn't find them?
Or did you write them because you just thought that
you had a different take on something than others?
Speaker 1 (44:35):
Oh? No, I actually know precisely why I write my
books and who I'm writing them for, which is I'm
writing them for the younger version of me that didn't
know those particular facts when he was younger, but would
have loved that. That's who I'm always writing for.
Speaker 2 (44:52):
You know, Joni Mitchell taught me about something about songwriting
that was really transformative and my songs I've been writing
since I I was eighteen, but since Joni told me
this in two thousand and five, I not only did
my current songs get better, but I went back to
the old ones and started rewriting them. And what she
said was, you don't write a song because you figured
(45:13):
something out. You write it in order to figure something out.
Speaker 1 (45:24):
That was my friend and colleague Dan Levitton, neuroscientist, cognitive psychologist, writer, musician,
and record producer. So what has emerged from a couple
of decades of the study of music in the brain
is that music can't be understood simply as different pitches
hitting the ear. The sounds of music trigger a neural
(45:47):
hurricane of spikes on the inside of the skull, and
this correlates with the pitch in rhythm and timbre, but
it jins up a silent neural symphony, and that activity
is closely tied in with our emotions and it pulls
strings to our memories. We don't just hear music, we
(46:08):
actually resonate with it neurally. So the brain doesn't just
process music, it magnifies it, which shows us once again
that the ultimate instrument is not the piano or the guitar,
but the one between our ears. Go to eagleman dot
(46:29):
com slash podcast for more information and to find for
How can we understand what music is about from the
point of view of neuroscience. Can music be leveraged to
help with anxiety disorders, or with dementia or with Parkinson's disease?
Is music universal or does it have to do with
what you have absorbed in your lifetime? How is music
(00:26):
like a language but one with very particular structure and
therefore high predictability. And what does this have to do
with Stevie Wonder on the High Hat or the relationship
between music and color. Welcome to Inner Cosmos with me
David Eagleman. I'm a neuroscientist and an author at Stanford
(00:47):
and in these episodes we sail deeply into our three
pound universe to understand why and how our lives look
the way they do. Today's episode is about music and
(01:08):
the brain, and this is a topic that has been
requested by several different listeners, so I prioritized making this episode. Music.
I suspect is a popular topic because music can be
so emotive for us and so catchy and so meaningful.
My father, for example, who was the quintessential tough guy
(01:30):
when he would listen to Mozart or Brahms or Beethoven,
he would have tears streaming down his cheeks. And as
a child, I didn't have much understanding of classical music,
but that really caused me to wonder, what is going
on here? Why does this music wafting out of the
radio evoke such strong emotions and perhaps such deep memories
(01:54):
in my father? And as I got older and became
a neuroscientist, I wondered, is there something you unique in
the structure of the human brain that ties music so
closely with our emotional experiences? So I decided to do
an episode on this today, and I realized there was
no one better to ring up than my friend and colleague,
(02:14):
Daniel Leviton. He's the founding Dean of Arts and Humanities
at Minerva University in San Francisco and a professor emeritus
of psychology and neuroscience at McGill University in Montreal. You
may know Dan because he wrote a book called This
Is Your Brain on music, the Science of a Human Obsession,
(02:35):
which became a big New York Times bestseller, and he's
also written four other best selling books, including his latest,
which is called I Heard There Was a Secret Chord.
Music as Medicine. Now, as you may suspect. Dan is
also a very talented musician. He composes music and he's
worked on albums by Blue Oyster Cult and Chris Isaac
(02:56):
and Joe Satriani, among many others. He does this as
an advisory producer a recording engineer. So given his expertise,
I wanted to sit down with Dan to get his
take on music and the brain. Okay, Dan, what goes
on in the brain when you listen to music?
Speaker 2 (03:19):
Yeah, well, it really is quite fascinating. So it begins
with the sound waves impinging on your ear drums. They
wiggle in and out, and all the information you have
about the auditory world comes from molecules vibrating in some medium,
in our case air, it could be underwater, and then
(03:39):
your ear drums just wiggle in and out, and then
your brain has to take that wiggling in and out
and extract from it all the different sounds a bird chirping,
a leaf blower going, the oboe in the symphony as
opposed to the French horns in a crowded room, the
conversation you're trying to listen to in front of you,
as well as that when you're eavesdropping in. It has
(04:01):
to separate all that out. And the way it does
that is that your brain has special processing circuits. I
was going to use the word designed, but of course
they weren't designed, but evolved to do different distinct functions.
One circuit processes the loudness is anything louder, is it
(04:21):
getting softer? And it follows that loudness trajectory, which can
be an important cue as.
Speaker 3 (04:28):
To what's going on. Pitch duration.
Speaker 2 (04:32):
The pitches get in a separate circuit, bound into a
representation of melody and harmony, the durations into a representation
of rhythm and meter, the loudness into accent structure, timbre,
which is the quality that distinguishes your voice from my
voice when we're saying the same thing, or a trumpet
(04:53):
from a piano when they're playing the same note. That's
a combination of spectral temporal information, in other words, pitch
and time and loudness. It all comes together later in
the brain and you just hear that song where later
in the brain is maybe forty milliseconds. It happens so
(05:14):
seamlessly that it just sounds like we're hearing the song,
but we're not. Our brain is hearing the pitch, the rhythm,
the loudness, the timber, and our evidence, David, that this
happens is not just from neuroimaging, but from patients. We
see patients with focal brain damage who suddenly lose their
perception of pitch, but they retain rhythm or vice versa.
Speaker 1 (05:38):
Okay, so it's this terrifically complicated computational process. So why
does it end up feeling so emotional for us?
Speaker 3 (05:46):
Well?
Speaker 2 (05:46):
So I think there are both neurobiological reasons and evolutionary
reasons that are connected. Of course, the simple answer is
it's emotional for us because emotion circuits in the brain
are involved in music processing. And by that I mean,
among others, the well known reward center that you and
(06:09):
I have talked about in our own classes over the
years and with one another, the limbic system, the structures
like the nucleus of Cumban's, the amygdala, the ventral tegmental area.
This is the emotional core part of the reptilian brain
that you know motivates us to move out of the
way of some approaching danger or to signal pleasure when
(06:30):
we're hungry and we finally get a taste of something sweet.
Music activates that same center, and in fact, it was
my lab that was the first to show that our
brain produces dopamine in response to music listening, and later
we showed that our brain produces its own endogenous opioids
in response to music listening, all part of that well
(06:51):
known pleasure network. Now, of course, that raises the question
why why is you know, over ever solutionary timescals did
music hit that emotional center and hear The answer is
by fer Kate. One is that the well known startle response.
(07:11):
You hear a sudden, loud noise and you jump. That's
evolutionarily adaptive because you know, you know, even lizards, snakes,
reptiles have to move out of the way of something
that might step on them or smash them. And that
startle response in humans goes directly from the inner ear
(07:35):
to the cerebellum and the brain stem. Before we even
figure out what the sound is, we startle and that's
connected to emotion centers. And so music, because it's an
auditory stimulus, is hardwired to movement and to emotion.
Speaker 1 (07:51):
So what happens in the brain when you play an instrument.
Speaker 2 (07:54):
Playing instrument is one of the most neuroprotective things we
can do. Glistening activates every area of the brain that
we so far mapped, as does playing an instrument. But
the added advantage of playing an instrument is that it's
active rather than passive, and it involves prediction centers in
the prefrontal cortex, particularly broad An area forty seven, which
(08:19):
is a pattern detector. You know, Venode, Menna and I
at Stanford for thirty years have been looking at this
little sliver of tissue on either side of your between
your top of your ears and your eyeballs broad and
forty seven, and Michael Patritus and I at mcgil have
also looked at it as that part of the human
brain that primates lack that allows us to process temporal patterns,
(08:45):
either in vision, touch or sound. And so in playing
an instrument, you've got to plan what you want it
to sound like. Hopefully you've got some idea of what
you want to come out, and then you've got to
use a feedback loop to listen to what came out
and see how well it matches with what you intended
(09:06):
and that adjust or not. And Broaden forty seven is
a part of that, as well as other prefrontal areas
and temporal areas.
Speaker 1 (09:15):
Before we move on some other questions, I want to
ask what got you into this intersection between music and
the brain.
Speaker 3 (09:23):
It wasn't intentional.
Speaker 2 (09:25):
I had dropped out of college after my sophomore year
to play in a series of bands, and that led
to me becoming a staff producer at Columbia Records in
the eighties and in the nineties, when the music business
seemed to be imploding, a bunch of us who had
entered the business around the same time, figured we needed
(09:46):
a plan B, that this may not be a sustainable career,
and so I went back and finished my bachelor's degree
at Stanford. I just worked in every lab that would
have me. Then I went to graduate school and Oregon
in order to work with Doug Hintsman and Helen Neville
a language specialist, and of course Mike Posner on neuroimaging.
(10:09):
And I was just doing all these things in parallel
and loving it. I mentioned all these names because they
were very important mentors to me, each of them. And
in the third year of my graduate program, Posner, who
was my principal advisor, said, you know you're gonna have
to specialize. What do you want to do when you
(10:32):
grow up. I said, well, I really love all of this.
I love psycholinguistics, I love memory, I love decision making,
and I had worked in all those areas. And he said, well,
you know, you have this background as a musician, and
there's a lot of competition for jobs in those other fields.
There's this emerging field of psychology of music with a
(10:54):
handful of people in it. Maybe if you go into
that field, there'll be a lot of low hanging fruit,
as it were, a lot of studies that have obviously
need to be done that haven't been done yet. And
in addition, although nobody's going to advertise for a music
psychology faculty member, all those other things apply to music psychology,
(11:15):
decision making, how do we decide what we want to
listen to? Cyclel linguistics, psychology of lyrics and music, individual
differences in musical taste, memory for music. So it was
Mike who said preciently that I should brand myself as
a music psychologist. Of course, back in those days, there
were no neuroscience departments, there were no neuroscience programs. You
(11:37):
could not study neuroscience as you and I know it.
And I'll make a distinction for our listeners between like
molecular neuroscience, which was done in biology departments where you
only look at a single neuron and you've never even
considered what a thought might be and what we call you.
And I work in the area systems neuroscience, where we're
looking at the big ideas andions of neurons communicate with
(12:01):
other millions in neurons.
Speaker 1 (12:03):
I know that you feel like that the music psychology
world was not particularly good let's say twenty years ago,
so give us a sense of what it was like
at that point when people thought about I'm going to
study the psychology music and what it's like now and
what's changed.
Speaker 2 (12:19):
And there were a bunch of other people who hadn't
studied cognitive psychology but fancy themselves music psychologists, and they
did a bunch of bad studies, the poster child for
that being the Mozart effect study, which purported to show
that listening to Mozart for twenty minutes would make you smarter.
(12:40):
And there have now been literally one hundred studies that
shows that that was just bullshit. It was a poorly
controlled study, It was done by people who had no
experience in human experimental design. They stepped outside their lane,
and so there was a lot of garbage work being done.
What's different now is that we've had twenty years of
(13:02):
people applying to graduate school who knew they wanted to
study music psychology. Some of them went into music psychologists'
labs like mine. Others did what I did. They went
into the lab of a memory person or an attention
or brain imaging person and just used that interest in
(13:23):
music to design studies.
Speaker 1 (13:25):
And so what sort of things have come out in
the last twenty years. Part of this has to do
with the advent of neuroimaging, right.
Speaker 2 (13:33):
Well, that really, that really is what kicked it off, because,
as you know, David, the study of emotion was rather unseemly.
I think it began with the foundation of the first
psychology labs in the world by Vuntenfeckner in Europe. Of course,
William James was always more interested in the esthetics and
artistics side, but it was the behaviorist movement of the
(13:56):
fifties led by BF Skinner, that if something wasn't deservable
and replicable, it wasn't worth study. And so emotions just
seemed too squishy, and music as the language of emotion
seemed like the squishiest of all. What happened with the
first studies of neuroimaging, which Mike Posner was part of
(14:16):
nineteen ninety eight ninety nine. We were able to actually
see pictures of the brain caught in the act of
thinking and remembering and imagining, and that gave it a
biological basis reality. You could replicate brain imaging experiments, and
so first the study of emotion, and then shortly followed
(14:41):
by the study of music with the first neu imaging
studies of music in the early two thousands, that put
it back on the table as something that was worthy
of study and could be done in a rigorous fashion.
Speaker 1 (14:54):
So your latest book, I heard, There was a Secret Chord,
looks at this issue of muse musick as medicine. So
tell us about that.
Speaker 2 (15:03):
That's something that I think most of us have experienced intuitively, certainly,
it goes back tens of thousands of years using music
to treat injury and disease to shamans and faith healers
and indigenous tribes. And it was really in the last
ten years. I would say that the idea that music
(15:27):
had an evidence base for treating injury disease promoting wellness,
helping with mental disorders like depression, post traumatic stress disorder, anxiety.
There's been eight thousand papers in the last two years
alone in peer review journals on medical applications of music,
and so about five years ago I became involved with
(15:51):
the National Institutes of Health and the White House Science Office,
leading various expert panels to figure out what do we
really know and what do we don't know and what
remains to be done. That led to a call for proposals.
The NIH put forty million into music and medicine research
a few years back. So I looked for a book
(16:12):
on music and medicine because I wanted to know what
the state of the art was, and I couldn't find one.
There were a lot of papers, but no books, and
so I ended up writing the book I wanted.
Speaker 1 (16:22):
To read and so give us a sense when it
comes to something like, let's start with dementia, how would
music be useful in the case of something like Alzheimer's.
Speaker 2 (16:34):
Not all dementia is Alzheimer's, of course, and not all
memory loss comes from Alzheimer's. But the most straightforward case
of somebody with profound memory loss, perhaps due to Alzheimer's,
Corsokov's stroke, whatever. They may not recognize where they are,
They may not recognize loved ones. They may not even
(16:55):
recognize themselves in the mirror. This is profoundly We've seen
patients who will walk by a mirror and think they're
talking to someone else, and then they get angry because
the person in the mirror appears to be mocking them
by gesturing the same way they are, and it causes
one of two reactions. Individuals either turn in on themselves,
(17:19):
fold in on unsells because the external world makes no
sense and they become somewhat catatonic, or they become angry, agitated,
and violent and in that case have to be medicated.
They might start beating up on their spouses, not recognizing them.
Music follows a kind of principle that computer science talks about,
(17:43):
which is first in, last out.
Speaker 3 (17:45):
This is a holder for the old models of computer memory.
Speaker 2 (17:48):
The first thing that goes into the memory is the
last thing to come out. And because we've been listening
to music in the womb most of us and through
our childhoods, those memories the most deeply embedded in the
brain and the most resilient and resistant to decay or damage.
And so if we play music from the youth of
(18:10):
an Alzheimer's patient, somebody with profound memory loss, play the
music from say the ages of twelve to fourteen or
sixteen that's preserved in most cases, and it allows them
to profoundly reconnect with a part of themselves they had lost.
It eases them, it comforts them, It triggers memories that
(18:35):
had been buried, and that kind of therapy or intervention
can pull them out of the state they're in and
actually have consequences for days or weeks where they come
alive again.
Speaker 1 (18:49):
It doesn't cure or help the dementia the cognitive loss exactly,
but it triggers memories and pulls them back to state
where they've been. And it can also revivify skills that
someone has. For example, musicians with profound dementia who get
(19:11):
an instrument put in their hand and they go and
play again as though they're young.
Speaker 2 (19:17):
It's really extraordinary. And we saw this play out in
recent years with Glenn Campbell first and then with Tony Bennett,
both of whom had profound memory loss and did not
know where they were. When Glenn did his final tour
with dementia and in the throes of Alzheimer's. He would
sometimes play a song two or three times in a
(19:39):
row because they didn't realize he had just played it,
or he'd forget what song he was supposed to play.
But once the notes, the first few notes happened, he
knew where he was. Say, with Tony Bennett, he could
sing for an hour and a half without stopping once
the music took over. These are what we cognitive scientists
call overlearned. They're not just in memory, but they're in
(20:03):
memory with thousands and thousands of traces overlaid on top
of one another and their procedural memory. Once your vocal
cords and your fingers get going, they kind of take over. Now,
to be clear, the memory is not in your fingers,
although it feels that way. If I were to scoop
your brain outside your head, your fingers would not keep
playing like a chicken with its head cut off. But
(20:24):
those pathways are so profoundly deeply embedded that yeah, you
can keep going even with Alzheimer's, and it gives the
patient a rare act of competence in a world in
which they're otherwise incompetent. It gives them agency in the world.
This can really affect their mood and their quality of
(20:47):
life and way of being in the world.
Speaker 1 (21:04):
Now, one of the things I've talked about on in
previous episode is right Bo's law, which is where older
memories are more secure, they're burned down more deeply than
newer memories. And of course we see this with people
with cogno.
Speaker 2 (21:17):
I thought you were going with Bow's law, like the
bow on a violin, the earliest violin pieces are the
ones that are the most embedded.
Speaker 1 (21:25):
Yeah, nope, Riebo Ribot, Yeah, which is you know, he
was the first I think this is actually the first
rule in neurology, as in the oldest rule.
Speaker 2 (21:35):
Yeah.
Speaker 1 (21:35):
But anyway, he saw that, you know, things from childhood
were remembered by people with let's say dementia, and things
they did last week or last month were not remembered,
so older members more stable. Now, the reason this is
so strange is because nothing else works that way. Institutions,
for example, don't remember their older stuff better than they
(21:56):
remember their newer stuff. But brain works this way.
Speaker 3 (22:01):
That is true.
Speaker 1 (22:02):
So my question is, I mean, when when it comes
to music, presumably most of these great musicians have been
playing since they were a little kids, and these particular
songs are overlearned, as you mentioned. Is this an expression
of simply of Ribo's law, which is that it's an
older memory and that's why they're able to do it?
Or is there something different about music than if I
(22:24):
ask them something about their you know, their their childhood home.
Speaker 3 (22:28):
What a great question.
Speaker 2 (22:29):
Well, there is something different about music to invoke Claude Shannon.
It's a highly organized and structured stimulus like language, and
so in an information theory perspective, which is Shannon, your
ability to predict what will come next in any sequence
(22:54):
defines how structured highly structured it is. Even more so
than language. Music is highly constrained. So I could say
a sentence to you like this, Let's try this, the
pizza was too hot to blank? What comes to mind?
Eat yeah or touch yeah? I would not be likely
(23:15):
to say the pizza was too hot to sleep. Once
I say it, you understand what I meant, and you
understand that I use the correct part of speech. I
put a verb at the end, But it's an unlikely outcome.
Music is even more constrained because there are only twelve
notes in our scale, and there are I wouldn't call
them laws of music theory, but customs of music theory.
(23:38):
And there are rhythmic rules or customs.
Speaker 3 (23:41):
And so once you.
Speaker 2 (23:43):
Get going on a piece of music, its own structure
constrains what the possible completions are, making it easier to
remember and then easier to recollect, easier to store and
easier to retrieve, and then moreover, as its own internal tempo.
Once the beat is going, it's carrying you along, whether
(24:06):
you're ready to go along with it or not. And
so you're going to fill those slots with what needs
to go there or your best approximation for it. And
when you look at performing musicians, like typically the ones
that work holiday ins on Friday night lounge bands and stuff,
they might know two thousand songs where I would use
the word no in quotes, they probably don't know every
(24:30):
single note and every single rhythm.
Speaker 3 (24:32):
But they can approximate it.
Speaker 2 (24:34):
They can improvise and estimate it so that you don't
really notice the difference.
Speaker 1 (24:38):
Well, that's actually a good segue into Parkinson's disease.
Speaker 2 (24:42):
How is music used there? So in all these cases
of music is medicine. We're not talking about like a
music module in the brain or a music medicine prescription
that's straightforward, because music is doing different things in different
parts of the brain, different aspects of the music are
doing it, and Parkinson's is I'm so glad you brought
(25:04):
this up. It's actually the best case for understanding this
differentiation in Parkinson's disease. At some point, most patients will
experience difficulty walking movement disorders in general, but walking in particular,
and it's because the disease degrade circuits in the basal
ganglia that are required to maintain a smooth and steady
(25:27):
gait and to orchestrate the movements of one foot has
to go after the other and you have to put
it down at a certain time where you end up
with both feet in the air at the same time,
and that's not good for walking. And the circuits that
allow you to walk rely on an internal intrinsic timer
in the brain, a clock, and that's what gets degraded.
(25:47):
If you listen to music that has the tempo of
your natural gait, you have neurons, neuronal clusters that were
not damaged that synchronize to that pulse, and then they
can act as an external clock that allows you to
walk smoothly and continuously.
Speaker 1 (26:07):
What else besides dementia and Parkinson's, where else do we
see therapeutic effects?
Speaker 2 (26:13):
Well, I think one of the big ones is an anxiety.
Dentists figured this out a long time ago. They play
you what's supposed to be relaxing music to reduce your anxiety,
which reduces swelling and inflammation.
Speaker 1 (26:28):
And why does that work?
Speaker 2 (26:30):
Some music we find to be relaxing, some we find
to be stimulating, some we find to be inspiring. And
the difficulty is, there's no one music that will do
those things for everybody. In fact, there's no one song
everybody likes. There's no one song everybody hates. It's subjective,
(26:50):
like your taste for food. You know, why doesn't everybody
like Indian food? I love food, Not everybody does. You
have our own taste. It seems as though we have
an esthetic module each of us that governs things like
our taste and colors and people and tastes and music.
And we actually showed in a PNAS paper that a
(27:12):
person's preferences for certain musical combinations is correlated with their
preferences for certain color combinations, which seemed completely crazy to me.
It just seems so far fetched. But there's this underlying
aesthetic module that we can't account for yet.
Speaker 1 (27:29):
What was the argument there, did you guys forward a hypothesis?
I read no?
Speaker 2 (27:33):
Oh well, I mean there was some hand waving. Yes,
there's an aesthetics module, and yeah, I mean to some extent,
it had to do with consonants and dissonance. Do you
like to see contrasting colors and contrasting chords? Do you
like to see things that are more consonant and harmonious?
But sharp edges and the metaphorical sharp edges and music.
(27:56):
But apart from that, it was pretty speculative.
Speaker 1 (27:59):
So let me double click on this issue about individual differences,
because one thing that's clear is across the population, some
people don't really like music that much. Other people love music,
it's a big part of their lives. But on one
on the spectrum, you just it's sort of meaningless to
many people. So how do you interpret that?
Speaker 2 (28:15):
Well, I just look at this as as a necessity
of Darwinian theory, which is that we can't all be alike,
or we'd you know, genetically or behaviorally or we would
all be wiped out by a single opportunistic virus. So,
you know, a cornerstone of Darwinian theory, as you teach it,
as I teach it, is descent with modification and random mutation.
(28:39):
And so you know, most random mutations end up being unobservable.
Some of them we see a phenotypic variation, that is
a behavioral variation. And in that case, yeah, ten percent
of the population probably don't like music, and they don't
understand why the rest of us spend so much money
and time on it. Ten percent of the population probably
(29:00):
don't like chocolate. I find that so impossible to believe,
but they don't. And then there are you know, some
percentage of the population don't like sex. They tend not
to pass that on through reproduction, but that trait, but
it's the way it works.
Speaker 1 (29:16):
How do you think music fits into the story of
evolution and human evolution in particular.
Speaker 2 (29:24):
So Stephen Pinker famously threw down a gauntlet in nineteen
ninety seven when he said that he thinks that music
has nothing to do with evolution, that it was just
sort of a byproduct of other things that we developed,
like language. He called music auditory cheesecake, And what he
(29:45):
was saying was that, well, we didn't really evolve to
like cheesecake. We evolved to like sweets and fats because
in the very small amounts they were available across evolutionary
time periods to it was adaptive to seek them out. Now,
if you get cheesecake, it'll just you know, you know,
(30:06):
spike your blood sugar levels and be bad for your health.
Speaker 1 (30:10):
I'm curious if you have a different view on it,
whether there's a role in evolution for music.
Speaker 2 (30:15):
We do have some data and I'd like to share
that with you, our and our listeners. So to begin with,
I mean, just as a resource, I would mention Stephen
Mithen's book The Singing Neanderthals, where he makes the case
that music was a proto language that preceded, you know,
linguistic language speech. I would say the evidence that we
(30:36):
have from neuroanatomy is that, from the work that Vanode
Menon and I have done in others, those circuits that
are engaged with music, listening and performing are phylogenetically older
than the speech circuits. And that's the reason why Gabby
Giffords was able to recover speech after she was shot
in the head and lost the ability to speak. It's
(30:58):
called melodic intonation therapy. We can take somebody who's lost
speech ephasic expressive aphasia and teach them to sing what
they need to communicate.
Speaker 1 (31:08):
So give us an example of that. What Gabby Gifference does.
Speaker 3 (31:11):
Well.
Speaker 2 (31:11):
She could not speak, but she might have been taught
things like I need a glass of water, show me
to the bathroom, I'm ready for bed now. She could
sing those things perfectly, and through neuroplasticity, the brain rewired
itself by passing those damaged circuits using the intact music circuits.
(31:32):
They are phylogenetically, that is, evolutionarily older. That's one piece
of evidence. Another is if we look at contemporary hunter
gatherer preliterate societies or all around the world that have
been cut off from Western civilization, and we make the assumption, David,
that they're living life now pretty much as they have
for ten or twenty thousand years, and all of them
(31:55):
use music for a number of things, not just a
single thing, but the emerging ideas that music evolved as
many things did, not for a single to solve a
single adaptive problem, but multiple problems, and one of them
was how do you encode knowledge in a pre literate society.
We've only had a written language for five thousand years.
(32:16):
We've been on the planet ten or twenty times. As
long as that, we still had to remember things like, oh,
don't go over that hill there, because my grandfather went
over there and they killed him because they're very vicious
and warring, and so you know, don't go there. And
when this water well runs dry, this is a route
to the other well, the supplementary one. Don't eat that
(32:39):
plant unless you boil it in this particular way. This
kind of knowledge is embedded in song in pre literate
hunter gatherer tribes and probably has been for a long time.
Mothers soothing their infants to imprint their infant on their
voice so that if they become separated, the infant will
know the sound the mother's voice. So we've got knowledge,
(33:02):
we've got bonding between mother and infant. We have social bonding.
Singing around a campfire to ward off a neighboring tribe
or predators as a way to defuse interpersonal tensions within
a tribe and to protect you from outside invaders.
Speaker 3 (33:18):
Lots of different uses.
Speaker 1 (33:22):
Let me double click on this for a second. Do
you feel that music is universal in terms of when
you compare across culture around the world, or are there
important differences locally both.
Speaker 2 (33:35):
Music is a cultural universal. There is no known culture
now or any time in the past that lacked it.
In David Huron's words, music is marked by its ubiquity
and its antiquity, and there are huge local variations. One
thing in common is we all have the octave, which
(33:57):
is defined by frequencies that are double or having of
one another hundred hertz, two hundred hertz, fifty hertz, all
the same. We perceive them as perceptually very similar um
similar notes. We give them the same name. In Western system,
there's middle C and there's high C, and there's low C,
things like that.
Speaker 3 (34:18):
But the way we divide up the.
Speaker 2 (34:19):
Octave into pieces is different across cultures. We divide the
octave up into twelve pieces, and we tend to use
only five or seven of the notes at a time.
The patterns that we make once we've divided the octave,
the way we combine them, either sequentially or simultaneously melodies
and chords are different. And that's why Chinese opera and
(34:41):
the music of Sub Saharan Africa sounds so different to us.
It's based on different customs and a different system. And
it's not the case that our music is better or
that if only you went into the Amazon and played
the indigenous people their mozart, they would feel that they
had suddenly heard from God himself. It doesn't work that way.
(35:04):
We've done the experiment.
Speaker 1 (35:06):
What is their impression of it?
Speaker 3 (35:08):
Eh? Meh? Yeah.
Speaker 1 (35:10):
Presumably whatever you've grown up with culturally affects your enjoyment
of what you're hearing, right.
Speaker 3 (35:18):
Very much.
Speaker 2 (35:18):
So. Yeah, we imprint on the music we're raised on,
and so we implicitly learn the grammar of our music
the way we implicitly learn the grammar of our language.
As Chomsky had said, here's a little test. Hang on
a second. Okay for the listener, Dan has des grabbed
(35:39):
his guitar, So we all know implicitly, uh scales, even
if we don't realize we do. Now you're expecting another note.
You're probably expecting this one.
Speaker 1 (35:57):
Thank God about that? Yeah, otherwise.
Speaker 3 (36:06):
That doesn't sound bad. But that sounds quite not quite right.
Speaker 2 (36:15):
And we know chords. We know that if I go
it wants to go somewhere.
Speaker 3 (36:28):
I can't just stop I've got to come back to.
Speaker 2 (36:32):
Or or.
Speaker 3 (36:37):
There's some options, but not unlimited options, right.
Speaker 1 (36:42):
And I've talked in many episodes about how fundamentally the
brain is a prediction machine. It's just it is trying
to guess ahead at what's going on. It's got an
internal model of what's coming next. So in those cases,
let's say, with the chords that you were playing, is
it the case that I had certain predictions because of
the culture I've grown up in. To what extent would
(37:02):
that be universal about what chorn comes next?
Speaker 2 (37:04):
Absolutely, and something I very much appreciate about you, because
this is not something that all cognitive scientists cognitive scientists
talk about, but you and I, I think, have been
big proponents for this idea that if the brain is
if the brain is nothing else, it is a giant
prediction machine. That is its job. It's just based on
(37:25):
what we've heard over and over and over again. And
the job of the composer and the performer is to
reward those expectations just enough of the time that you
feel like you're following along with the story as it unfolds.
But they have to surprise you just enough of the
time that you've learned something, or you get that little
(37:47):
hit of oh, isn't that interesting? As much as I
was trying to predict what comes next, they came up
with something even better than I could have imagined.
Speaker 1 (37:57):
Just like with everything in life, there's a spectrum between
novelty and familiarity and all the sweet spot is in
between that. Somewhere on that note, what is the reason
(38:22):
that we care for rhythm so much? Do you feel
as I do that it has to do with predictability,
where the brain says, oh, now I know what's coming next.
There's the next beat, in the next beat, and there's
something very satisfying about having some structure of prediction, and
then there's surprises thrown on top of that.
Speaker 2 (38:42):
It's because the brain's a prediction machine that rhythm is
meaningful to us. There are populations of neurons that fire
in synchrony with the beat, with the tempo that sets
us up for movement. And I mean that metaphor and literally,
music is possibly the only art form that makes you
(39:05):
want to wiggle your body in response to it. People
aren't standing in front of the Mona Lisa and dancing.
Speaker 3 (39:10):
Although they do, they.
Speaker 2 (39:12):
Seem weird and they might get kicked out of the louver,
but we dance to music because we can't help it.
And in the neuroimaging studies I've done all of them,
where we ask people to stay still, we still see
activity in their premotor cortex and their motor cortex that
they're trying to suppress. And then on the metaphorical movement, yes,
rhythm is important because it's telling us that there is
(39:35):
more to come and we want to know what that is.
It sets up a narrative momentum and when a musician
could play around with that rhythm. If you listen to
what Stevie Wonder does and the opening to Superstition, I
can't replicate this exactly, but he's playing around on the
high hat. He's setting up a beat. The high hat's
that little symbol and ordinarily a normal drummer would just go,
(40:02):
but Stevie doesn't goes. He's playing around. No, two times
are the same, and we may not notice it because
he keeps the underlying pulse there, but it is ear
Candy Man. It is just so interesting for the brain
(40:23):
to take all that in, even at a subliminal level,
that there's so much going on, and not only is
he changing in the rhythm. He's moving his stick around
so he gets the bell of the symbol, he gets
the edge, he gets the middle, he gets different sounds
out of it.
Speaker 1 (40:38):
So this is a good segue into something I've been
wanting to ask you, which is, what is your take
on artificial intelligence and music the future of And there
are two ways to think about this, of course. One
is AI composing music. Another is AI finding the perfect
music for you. Maybe there are other ways to think
about it as well, But tell us your take on
(40:59):
the future.
Speaker 2 (41:00):
AI composing music. I look at it this way. I
have a friend, a novelist, Gail Jones, wonderful novelist, who
says that AI music now is like those artificial flowers
at a holiday inn. You walk in the lobby, you
see this big display and you go, wow, isn't that nice?
And you get up close and you realize they have
(41:21):
no nice odor and they're plastic, And so AI music
at a distance probably sounds just fine. And it's already
crept into advertising on social media. It's being used in
the background. It's sonic wallpaper. It's kind of like the
painting in the bedroom of that same holiday in room
it's like, Okay, it's covering the wall. It's kind of nice,
(41:43):
but I'm not going to sit there and stare at
it for hours. I think that you and I subscribe
to Dan Dennett's functionalism, the idea that in theory, all
of our thoughts, hopes, desires, and beliefs, all of our
mental activity can be redo to brain activity, and that
brain activity can be characterized by patterns of neural firings
(42:05):
and connections if you could replicate a human brain and
add in the right amount of random factors. I think
in theory, yes, AI music could be great. I don't
like the thought of it, but I mean that's the reality.
But for now, I don't think AI music is going
to be a threat to real music and real feeling.
(42:25):
But you nailed it on the head. Where I think
AI can be the most use is helping us to
find music that we like. Most of us listen to
a couple of thousand songs maximum, over and over and again,
and occasionally let in a few new ones, and those
can be comforting and rewarding, but they can dig deep
(42:47):
neural ruts such that we get tired of them and
we want something new. There are now two hundred million
songs across the streaming services, with a one hundred thousand
new ones being uploaded every day. So how do you
find what you like? The major streaming services have recommendation systems.
(43:08):
They don't work particularly well for me, But AI in
principle can extract hundreds of features latent features for music
and build a multi dimensional model, a higher dimensional manifold
of musical structure and DNA that might have one hundred
and fifty orthogonal dimensions, and set up a kind of
(43:31):
universe of music where things that are where each song
is like a planet, and planets that are near each
other are going to be similar, not just structurally, but
in an emotional space where they're likely to cause a
similar emotional reaction in a given listener, knowing what your
own personal space is, right, It's not going to be
(43:52):
something that is entirely objective and prescriptive and one one
hundred and fifty dimensional map for everybody. They'll have to
be to maps for each person, because your tastes are
different than mine. As my grandfather used to say, if
everybody liked the same things, they'd all want to get
with your grandma, I'm going.
Speaker 1 (44:12):
To steal that line if you don't mind, when I'm
a grandfather. What have I not asked you that I
should ask you?
Speaker 2 (44:19):
I wanted to ask you of the many successful books
and highly regarded books you've written, did you write them
because you wanted to read them and you couldn't find them?
Or did you write them because you just thought that
you had a different take on something than others?
Speaker 1 (44:35):
Oh? No, I actually know precisely why I write my
books and who I'm writing them for, which is I'm
writing them for the younger version of me that didn't
know those particular facts when he was younger, but would
have loved that. That's who I'm always writing for.
Speaker 2 (44:52):
You know, Joni Mitchell taught me about something about songwriting
that was really transformative and my songs I've been writing
since I I was eighteen, but since Joni told me
this in two thousand and five, I not only did
my current songs get better, but I went back to
the old ones and started rewriting them. And what she
said was, you don't write a song because you figured
(45:13):
something out. You write it in order to figure something out.
Speaker 1 (45:24):
That was my friend and colleague Dan Levitton, neuroscientist, cognitive psychologist, writer, musician,
and record producer. So what has emerged from a couple
of decades of the study of music in the brain
is that music can't be understood simply as different pitches
hitting the ear. The sounds of music trigger a neural
(45:47):
hurricane of spikes on the inside of the skull, and
this correlates with the pitch in rhythm and timbre, but
it jins up a silent neural symphony, and that activity
is closely tied in with our emotions and it pulls
strings to our memories. We don't just hear music, we
(46:08):
actually resonate with it neurally. So the brain doesn't just
process music, it magnifies it, which shows us once again
that the ultimate instrument is not the piano or the guitar,
but the one between our ears. Go to eagleman dot
(46:29):
com slash podcast for more information and to find for
Host
David Eagleman
Popular Podcasts
40s and Free Agents: NFL Draft Season
Daniel Jeremiah of Move the Sticks and Gregg Rosenthal of NFL Daily join forces to break down every team's needs this offseason.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
The Bobby Bones Show
Listen to 'The Bobby Bones Show' by downloading the daily full replay.