Peter Vuust is a Professor at the Center for Music in the Brain in Aarhus, a jazz musician, and composer. In this conversation , we talk about his recent review in Nature Reviews Neuroscience, how he got to where he is, active inference in music, jazz improvisation, and much more.
BJKS Podcast is a podcast about neuroscience, psychology, and anything vaguely related, hosted by Benjamin James Kuper-Smith. In 2022, episodes will appear irregularly, roughly twice per month. You can find the podcast on all podcasting platforms (e.g., Spotify, Apple/Google Podcasts, etc.).
00:05: How Peter became a jazz musician
04:54: How Peter became professor of neuroscience
08:20: How to combine two different professions practically?
11:50: Start discussing 'Music in the brain'
24:53: How do prediction errors change with familiarty of a piece of music?
38:18: How does moving to the beat (active inference) reduce prediction errors?
46:48: The 3 dynamics in musical synchronisation
55:10: How does Peter compose for improvisation in jazz?
References and links
Heggli, Konvalinka, ..., & Vuust (2021). Transient brain networks underlying interpersonal strategies during synchronized action. Social cognitive and affective neuroscience.
Heggli, Konvalinka, Kringelbach, & Vuust (2019). Musical interaction is influenced by underlying predictive models and musical expertise. Scientific reports.
Heggli, Cabral, ..., & Kringelbach. (2019). A Kuramoto model of self-other integration across interpersonal synchronization strategies. PLoS computational biology.
Morillon, & Baillet (2017). Motor origin of temporal predictions in auditory attention. Proceedings of the National Academy of Sciences.
Rosso, Maes, & Leman (2021). Modality-specific attractor dynamics in dyadic entrainment. Scientific Reports.
Vuust, Heggli, Friston, & Kringelbach (2022). Music in the brain. Nature Reviews Neuroscience.
See the painting with the 'false' line at 7:30 in this talk: https://www.youtube.com/watch?v=hOfGX6KSiX8&t=458s
Stravinsky's Rite of Spring: https://www.youtube.com/watch?v=rP42C-4zL3w
The last part with frequent time signature changes starts at 30:07.
A survivor from Warsaw by Schoenberg: https://www.youtube.com/watch?v=LBNz76YFmEQ
3rd movement of Sinfonia by Berio: https://www.youtube.com/watch?v=9YU-V2C4ryU
Beatles Documentary by Peter Jackson (Get Back): https://www.imdb.com/title/tt9735318/
Blame it on the Boogie, by The Jacksons: https://www.youtube.com/watch?v=nqxVMLVe62U
(This is an automated transcript that will contain many errors)
Benjamin James Kuper-Smith: [00:00:00] Actually the very first, very silly question. How do I pronounce your surname
Peter Vuust: Vuust
Benjamin James Kuper-Smith: Vuust? okay. Vuust, I want to get at least vaguely, correct?
Peter Vuust: That's fine.
Benjamin James Kuper-Smith: So the reason I'm asking this is I kind of want us to ask how you got into music and then how, and how into neuroscience of music and that kind of stuff.
And then I found a video on YouTube that might give a little bit of, a bit of an explanation of background. And it's called, um, five times Vuust is that I'm assuming your family, are all of you musicians?
Peter Vuust: Yeah. Uh, I mean, in my family, we learn to play the piano at the age of six that has been going on for eight years. And I was taught that my siblings were told that my sons have been taught that. So that's the point. I know my father is, um, is a teacher and he, he, he never did. He never really, um, was interested in, uh, in, in knowing our [00:01:00] grades from the school or, you know, he wasn't interested in how we did in school, but he said, one thing, you had to learn how to play the piano.
You learn it. You have to learn how to ski and you had to learn to play golf because those things you cannot learn if you become older. So, so that's, that's basically where I come from.
Benjamin James Kuper-Smith: So I'm assuming you can do all three or did you, did you
Peter Vuust: I can do all three. I would say.
Benjamin James Kuper-Smith: That's right. That's that's good. And then the, I mean, your main instruments, the base, right? So that came later then, or.
Peter Vuust: Yeah, That came later. It was in the beginning. It was an electric bass for many years. And then I realized I was best at playing Jess. And, uh, when I had been paying electric bass in JS for some years, I realized that it actually sounds better if it's an acoustic piece. And I, yeah. So, so, so that took me some, some years to learn to play that, but I entirely PLE acoustic bass nowadays.
So yeah, I, [00:02:00] I played on a little bit more than 100 records and, uh, I'm a professor at the Royal academy of music, so
Benjamin James Kuper-Smith: Yeah, exactly. It was pretty cool. That's uh, just how, yeah. How much music of use out there. I mean, I, um, I wanted to read another paper in preparation, but I had just ended up listening to more of your music
Peter Vuust: no,
Benjamin James Kuper-Smith: Uh, cause just got distracted. Um, but I have read songs that already I have prepared. Um, but uh, yeah, I mean, I actually also played double bass.
Um, but I played it, um, classically, so I never. Where do you play chess? Um, I'm first, like did used like always jazz or
Peter Vuust: No, no, no. I started out being mostly interested in the Beatles. Paul McCartney then, uh, Paul Simon and, uh, the, from, from there, I think, uh, suddenly I was, uh, I was on a tour in the states with a band and we were playing 70 gigs in the states around. so we were driving in this Greyhound bus and [00:03:00] then I started listening to some jazz because I had the feeling that you could learn something from it.
I basically didn't really like it. And then they
Benjamin James Kuper-Smith: how old was we this point?
Peter Vuust: Uh, sorry.
Benjamin James Kuper-Smith: How old were you at this point?
Peter Vuust: Oh, I was actually 22 or 21,
something like that. So, so then I started playing Diaz pretty late. Uh, but then, then I listened to Chett baker, uh, Stan Getz, pat Matheeney, uh, and, uh, you know, different stuff at that point in time. And then suddenly I realized that I started like liking it.
So, and then I realized also that I was actually better at playing that than I was at playing rock, rock and roll for instance. So.
Benjamin James Kuper-Smith: Hmm. Okay. Somebody would have assumed that playing jazz would be harder just technically than rock and roll. Is that just a false assumption?
Peter Vuust: Yeah. I, I always had, uh, this very good ear. So [00:04:00] from early on, I could play anything I heard. So like, I could basically play the whole Beatles forever to repertoire when I was 13, 12, 13, like all the course in beatless because yeah, for some reason, and that is a fantastic skill to have when you play IDs, because it's the most important thing.
Benjamin James Kuper-Smith: Okay. Um, yes, I mean, I want to talk today mainly about, um, I mean the, the one articles I've read in depth is music in the brain, the nature reviews, neuroscience article, um, maybe if anyone who's new to the podcast, I will put references and links to everything we discussed in the description. And I put that paper there to see, I mean, I am just going to, I'd like to talk about the neuroscience of music and your work in there, but I'm kind of just, just to make the bridge kind of between like the bad graphic stuff you see just mentioned, like, how would you go from there to suddenly be professor of
Peter Vuust: Uh, yeah,
Benjamin James Kuper-Smith: having your own love?
Peter Vuust: that's a, that's a long story. You can see. I, I was [00:05:00] trained as a mathematician, so while I was in my early twenties, I started having a lot of gigs. You know, it, I practiced a lot and I was also, uh, going to the university. I had my friend's degree when I was in, which is a bachelor's degree when I was 20.
And then I started on mathematics. So I have a mathematical degree, you less, a master's degree. And then I also have a degree in music. So this was sort of, my hobby was mainly going to the university and my gig was to. Great. So, so it turned out that way. And then I was hired at the Conservatorium, which is all the Royal academy of music.
And soon I suddenly had a tenureship there. So I wasn't thinking most of it, but I, I had this idea that I wanted to write about the miles Davis quintet from the sixties, because I think their music is so fantastic and interesting. So I wrote a book [00:06:00] on, on, on the poly rhythms in the miles Davis, quintet, you know, for a mathematician, that's not a long stretch.
So, so it's, that was kind of what I was interested in. And I actually had a small, or I had one year of money to do that. I applied for it and I got, uh, I got it from, from one of the councils research councils on my, on my face, basically. So, so I wrote that one and then, uh, I was offered. PSD at a new new center for neuroscience because the director of it was interested in yes, he was an amateur Desmond session and he had read the book and he was thinking, Hmm, and this guy's a mathematician.
Why shouldn't he be able to neuroscience? So, but.
Benjamin James Kuper-Smith: so, but you will a professor of music. And then the other side, like, do you want to do a PhD now?
Peter Vuust: I was an associate professor of music already at that point. And then, uh, this guy says, would do, wants to do a [00:07:00] PhD. And I said, well, that sounds really scary. Uh, let's do it. So, uh, so, uh, and, and then I S I've spent 20 years doing this and, uh, you know, bill from one, I mean, I did took, did my own PhD. Then I got one PhD student that I got two, then suddenly I had 13 PhD student.
And then I thought, well, I might try it for a bigger grant. And then I got this enormous grant and one more. And, uh, yeah, so nowadays we, I think we are 35 at my center, so it's a lot of people. So, uh, luckily nowadays I'm not supervising everybody. So. We are five or we are four very prominent professors, a modern cleanly buck, who is also a professor at Oxford university.
Then there's Peter Keller who was also really well known and has had a Mack max plank professorship, uh, and [00:08:00] is from Sydney, Australia. And, uh, then there's MVR particle who, uh, was, uh, employed in Finland before it would be four. She came to us. She's Italian originally also move very prominent professors.
Benjamin James Kuper-Smith: I mean, maybe one last question then about the, about this side, um, before we get into the neuroscience, um, kind of one question I had in general was kind of like, how would you combine these two things, right? Because music, you have to practice, you have to rehearse, um, justice to play, right? If you want to compose stuff and that kind of stuff or record it, then that's a whole nother thing.
And the research thing, I'm just curious, like on a, on a practical level, how do you do that? Do you have other. Um, like half a day each or is it yeah. How does
Peter Vuust: I mean, yeah. How does that work?
Yeah, it's been it's, it's, it's been a challenge since I was 18 years old and started at the university. Right. And it's still a challenge, but basically I'm very, very good at structuring my child. I get up super early in the morning And then I practice. So [00:09:00] every day, not just every day, every single day, I practice my instrument in the morning before I do anything else.
So, because you can't, you can't, uh, you can't play an instrument at a high level unless you are always in shape. So, uh, so that's just the cost of living like this, I think. And when I was, when I was studying in my twenties from, I was 22 till I was 38, I didn't have, uh, I didn't have a television. So, so that frees up a lot of time.
I mean, at that time we didn't have, uh, you know, Facebook and what you can spend your, or YouTube or whatever. So basically I didn't have any of that kind of, um, input. So I was free in the unit. Play and do stuff and play with others and so forth. So, so it is possible, but it takes a lot of, um, it takes a lot of planning ahead.[00:10:00]
I would see, but nowadays I play around 60 gigs a year. And like you said, we just put out this record, which has had enormous, has been really well received, uh, in the Danish media. Uh, and, um, yeah, so that's, that's kind of a, it's difficult. I would say it's difficult, but it's mostly a matter of really wanting.
Benjamin James Kuper-Smith: And then having the discipline to do it.
Peter Vuust: But, you know, this is also about the brains. You know, our brains is shipped everyday by what we do. So if you teach your brains that you get up in the morning, it may be hard the first week. And if you then say, okay, on the Friday, I didn't, I didn't manage to get up and practice. So then it's, it's suddenly forgot, forgotten, always soon forgotten.
But if you, then, if you can keep on doing it for like three months or four months, then suddenly that's actually what your brains will expect of you. So, um, for [00:11:00] me now, it's, it's so obvious that I, if, if I don't do it one day, it's terrible. I, and actually just today for good reasons, I wasn't able to manage.
So now I have to do it.
in the evenings. If I don't do it in the evening, I'm simply not satisfied. So, yeah.
Benjamin James Kuper-Smith: Yeah, I knew I needed to get it back into that. I started taking piano lessons again recently, so I hadn't taken them for like 10 years and I just paid occasionally. I mean, so piano was probably a maintenance to. And now I'm in this phage where for the last few weeks, I've basically every day is like, oh, I should practice again.
And then I don't get around to it because I just haven't built up a routine or anything like that. Um, yeah,
Peter Vuust: yeah, I can. It's
Benjamin James Kuper-Smith: need to get into that.
Peter Vuust: yeah,
Benjamin James Kuper-Smith: Yeah. Anyway, um, so, uh, I guess the, the article they were talking about is to some extent, I feel like it's kind of an overview of the predictive coding of music model, um, and all that kind of stuff.
So maybe, can you give a brief overview, maybe let's maybe [00:12:00] just start with predictive coding, um, kind of what's predictive
Peter Vuust: I, I.
Benjamin James Kuper-Smith: do we, or.
Peter Vuust: Yeah.
I, I would say that that the article is, which came out a month ago or something like that. The article, the, I mean, the purpose of the article is of course it is recording, but it's also to give an overview of the whole literature of music and the brain. So I would say it's, it's a, you can read it even if you don't know Buy into the producer coding theory as such there's this and the idea in the article, you can see it's a predictive coding is a framework for understanding how, how music is understood by our brains. And, uh, the whole basis of this is that if you look back in time, you were talking more about representations in the brain that you would like you, you, uh, experienced the world, you experienced different things.
Things come into your senses. [00:13:00] And back in the day, you would say, will that makes a representation in the, in the brain. And I think predictive coding diverged from. Understanding by seeing, well, actually it doesn't come into the brain, the brain already, when it comes in, has a prediction of what is going to happen.
So even at the first note, so if I sing, uh, then you uh, We'll already have a prediction of maybe the tonality, maybe even when is the next note going to come. And when you get the next note, it will create a better prediction. And maybe, uh, so, so the idea is about a behind predictive coding. Is that what the brain is trying to do is that is trying to make sense of the world by sort of reverse engineering and see if this is the input.
What is, what, how does that fit? My model is my model, correct? Given the input put, and it does so like in a hierarchical way. [00:14:00] So, so the different areas of the brain will all have predictions and feed forward prediction, error, and get back, uh, predictions from higher levels in the brain. So that's, that's the idea behind predictive coding.
And in that, in, uh, in a way it's, it's very intuitively correct, because. It must be easier for the brain to, to do the calculation that the brain does all the time. Of course, if it only has to, uh, has to process what is different in the input. So, I mean. if you have a model let's say, and, and music is such a good example of it, that's also a way where we, we think it's worth writing about, because if, if you think about music, if I, for instance, sing dumb bump, bump, bump, bump, You have your brains will already think, oh, key.
This is the, this is a meter. [00:15:00] It's a meter scale and bomb. This is the banality dadadadadada and so forth. And it also has a meter dumb bump, 2, 3, 4, 1, 2, 3, 4 1 2, or whatever it it makes for, for predictions and all these, these predictive models is something that the music that you, that, that then inserts your ears we'll play around with in a way.
So if I see then you think this is the, uh, this is the downbeat and, uh, So, so, so that would be the downbeat. So, and probably a four, four. So if I just got BombBomb, then you have already the downbeat and you probably also think that this is the key of the analogy, but in fact, it could easily be that this was, um, this was actually the tonight.[00:16:00]
that would still be, I had, I couldn't change, but anyway, Yeah.
Benjamin James Kuper-Smith: Yeah.
Peter Vuust: so now you have a different personality, uh, and also BombBomb. So you could even have, so, so if you say BombBomb, you go would go 1, 2, 3, 4 BombBomb. So that's a one, right, but it might even be that it 1, 2, 3, 4 BombBomb BombBomb. . So we have like a predictive framework when we listen to music and we can't get away from it.
You can even take base stable percepts for instance, as for instance, a three against four, if you have a three against 4 1, 2, 3, 4 1 2 4 1 2 3 4 1 2, 1 2. So when you count them off, this is exactly the same acoustical input to, to our brains. But whether you count it off in four or [00:17:00] in three, it gives a completely different, it's a completely different experience.
Benjamin James Kuper-Smith: Yeah, maybe to clarify to the, to the business you were, the, the snapping was the same each
Peter Vuust: This, so this, the snapping is always the same. I'll do it again. So, so the stabbing is the same, but when I changed from counting It off as a four, four meter and March to a waltz, then it suddenly sounds completely different. So I'll try to do it again. 1, 2, 4, 1, 2, 4 1 2 1 2 1 2, 1 2. So Dan , it sounds completely different.
So, so differently. So, so, um,
Benjamin James Kuper-Smith: It just becomes more like a Spanish dance or something. I crawled on the walls.
Peter Vuust: And, and this is, this is, uh, at least when you listen to music, then it's very clear that the brain helps you construct the reality. And this is a [00:18:00] key point in predictive coding. It is that the brain is trying to figure out what is the hidden courses of its input. And the hidden course in this instance is either a four, four meter or three, four meter.
And what it does is that it's trying to calculate the probability that the course that it has figured out like the four, four meter is correct. Given the. And in fact, when you hear X or when you listen to exactly this, a three against form a rhythm, then when we are from the Western world or Germany or Denmark, we it's much easier for us to listen to this as a three, four meter awards than a four, four meter.
Whereas in some African countries like Ghana, for instance, you'll probably, eh, the, the four, four meter would, would be at least equally as, as relevant as the three, four meters. So, so that this tells you [00:19:00] something very deep about how the brain makes sense of the world and what learning is all about, because in a way, what predictive coding then say, Is that what the brain can do when it experiences a conflict.
For instance, like this three against four, there there's a conflict. It might be the four that is the course or the three that is the course. Then it can either change its prediction, which most Westerners would do to this with, oh, they would go, oh, okay. It was actually a three-four meter. And then they will just switch to the three, four meter or we could tap our feet.
Eh, that means that we could act to re sample the environment. And when we do that, That then you could actually keep the four for me that even if it's less easy for you to do then, or the four for me to, even if it's less easy for you to do then the three, four meter. So you re sampled the [00:20:00] environment by moving your body and trying to keep the rhythm.
And so now you have perception and action linked with this predictive coding, uh, principle, uh, to each other. And when you, when this happens, you'll probably also experience some kind of emotion. There'll be some kind of reward, maybe even, uh, and, and you'll also learn stuff over time, because that will mean that you will change your further possibilities of, of how you can predict what is coming.
Benjamin James Kuper-Smith: I have one kind of a big picture question here is, um, so like the predictive coding of music model is that. I mean, it also was listening. It also was kind of reading the article. I felt like this is, I mean, it's, it's kind of in that sense standard predictive coding, right. It's just saying like the context of music is a really good way of looking at it.
Is, is that the way, or does, is it actually something slightly different? Like, is it a different variant of predictive coding or is it [00:21:00] just a really good yeah.
Peter Vuust: I, I hope it's not a different version of
Benjamin James Kuper-Smith: Okay. That's the way it appeared to me, but it wasn't.
Peter Vuust: now. I hope. And I hope, and, and as consistently as one of the authors of this paper, also, I hope he would have, uh, caught us if we were seeing something that wasn't completely true, predictive coding waste, but I think one thing that is slightly, uh, so, so we are also investigating how long can we take, how will does predictive coding fit music in a way?
And, and one thing that I think is at least something that you start thinking about when you, when, when, when you look at is at music in the predictive coding way, is. What music does is that it basically creates prediction error. That Brene has a hard time explaining a week. [00:22:00] So in the predictive quality setting, it's, I mean, standard predictive coding says, well, what the brain is trying to do all the time is that it's trying to, uh, minimize prediction error either by changing the, uh, prediction or, or trying to, uh, to re sample the environment by acting or acting, and then reassembling the environment.
So that's basically what, uh, what the brain has to do. It has to try to minimize prediction errors so that it can free up its energy for other stuff, But in music and in art, what art really does is that he creates prediction error. That is hard to get with. I have this example of, uh, you know, it's a piece of art from that I saw at the art museum in Omaha, and I took a picture of it, uh, with my, my, my camera.
And [00:23:00] then what it has, it's basically one line, but then there's these. And annoying bumps on this straight line. And, uh, and then, uh, there's some colors, but it's, it's very, very simple. It's actually the Ukrainian flag. So, so usually when I do lectures, I ask people to look at this, uh, this, uh, picture and then say, what are you looking at?
And then nowadays they say the Ukrainian flag, but, but you know, basically they also
look at this
Benjamin James Kuper-Smith: or something, right?
Peter Vuust: exact exactly, exactly. And, you know, it's so annoying. They're these two annoying bumps on this straight line. And that's of course our eyes would Like uh, the easiest thing for, uh, for our eyes would be to follow the, the straight line is so when there's something that doesn't fit the straight line, which is our prediction, uh, on normal prediction, because we have so many strays straight lines, then we actually, we are, we are caught up with looking at these two [00:24:00] annoying bumps and you can't get rid of it because it's, it's, it's schematic knowledge.
That straight lines lines are straight. Right. If they're, if they start out being straight, they should end up, uh, being straight that's. That's kind of the prediction we would have and the same thing in music when I go that, uh, oh, Uh, so no matter how many times you've listened to this, you still think, okay, that's an odd note that that's not how it should be.
That's schematic expectations. And there are so many in music, and this is what we learned by statistical learning when we are small kids and we learn the different schemes, that music is from our culture. Right. So, yeah.
Benjamin James Kuper-Smith: Yeah, I mean, do you, so this is something I've been wondering about since reading your article, which is, I mean, so you talk about, especially the precision way to prediction errors so that the, and this kind of thing. So prediction errors, [00:25:00] depending on how confident you are on them, it can be weighted strangled as strongly, that kind of stuff.
And I've been wondering, like, if you, like, if you know a music piece really well, do you, do you still have a prediction error or not? Or like, how does that work? Because you know exactly what's going to happen. Or if it's something there's a syncopation and a piece of something like that, like, and you know, it's going to come, I don't know.
Do you think you still have one or is it, or is their average is weighted very low or, yeah.
Peter Vuust: I'll bring it back to the straight line. So you could probably look at this picture every day for the rest of your life. And still, when you look at it, you're thinking, why couldn't he just have drawn a straight line? I mean, this is, this is. And, and, uh, the way to explain this, I think David Warren does it better than I can, but what he sees is that, and this is special for music that you can listen to a piece of music for so many times, for some reason.
Right. But the thing is that what the music does is, is that it plays around with our schematic predictions and our short-term [00:26:00] predictions and our critical predictions. And even though your vertical predictions, the predict predictions that you get from listening to piece many, many times, you know, of course, but the letter that, uh, wa uh, I mean, now, you know, I'm doing it, but still part of your brain, the schematic brain was it.
Now it should have been that the data. Well, right. So,
Benjamin James Kuper-Smith: it, basically my head. Yeah. It's like, he's not going to do it this time again.
Peter Vuust: Yeah. Yeah, exactly. Yeah. I mean, and
Benjamin James Kuper-Smith: I'm not going to fall for this twice.
Peter Vuust: now you're not going to fall, but of course, when you have a piece of music for instance, and you listen to it 10 times, then there's one thing that I have no, I noticed already when I was a kid, the records that I didn't like at first were the records that I liked the most in the ending, because probably my guess is.
Because it's so nicely done the music, you have exactly the right amount of [00:27:00] prediction error, according to the schematic and short-term predictions. So that once your vertical predictions are satisfied. So, you know, okay, now I don't have to worry so much about that anymore. Then you, you can actually start really appreciating the interplay with the other predictions.
So in the beginning, like, like some, some songs just strike you momentarily. Oh, oh, I already liked to listen to this, this music, and then you listen to them 10 times and you think, okay, that it wasn't that interesting. Cause now your, your vertical predictions have, have been satisfied. And then, uh, uh, some of the other stuff, uh, might not be that interesting.
So, so I think it's this interplay between the different types of memory systems that we have in the brain.
Benjamin James Kuper-Smith: Yeah. Like the example, listening to music for the first time. I mean, for me, so when I said I had lots of classical music, I [00:28:00] especially like a lot of the kind of 20th century classical music. And I remember still the first time I listened to Rite of spring. And I first, I was like, I don't know, what do I like this?
And it's kind of weird. I, don't not sure where they understand what's going on or something like that. And then, you know, you, listen, I listen to it more and more because I felt there was something then now I really like it. But the funny thing there to me is, I mean, I'm probably losing like half of the audience now.
Uh, but that's just a specific example,
Peter Vuust: Nobody said really good example, actually.
Benjamin James Kuper-Smith: And I'm especially curious about the last section, because the last section there is this like super irregular rhythm where it goes like, you know, 5, 7, 3, 8,
Peter Vuust: Yeah.
Benjamin James Kuper-Smith: like, oh,
Peter Vuust: data, all this
Benjamin James Kuper-Smith: and yeah, I've listened to it so many times and I still have, I'm always wrong with predicting what the next note is going to be. And when it's going to come, I think I'm really curious, like, yeah. I mean, I have, I guess I have predictions there. I clearly have predictions. They're just always wrong, even though
Peter Vuust: They are, they're always kind of wrong. Even Stravinsky's own predictions were wrong, right? Because he started out, uh, he, [00:29:00] when he wrote it the first time he wrote it as a 3 8, 5, 8, 7, 8, and so forth in, in changing, uh, meters. But then he found out that he, the only way he couldn't conducted was in four, four, and then he actually made a new version where he had written it in four, four.
So there you have a very good example because for my, when I listened to the first time, I was just thinking, well, I guess it's some kind of 12, eight, or something like that. Um, but, but th the idea is, I think is that you can infect, you could take different viewpoints, you could change your predictions all the time.
So that would be, there would be just a shifting with every data that, that data. So now we will want to 1, 2, 1, 2, 3, 1, 2, or whatever it goes. So you changed your meter all the time, but you can also just go 1, 2, 3, 4 . And so, I mean, [00:30:00] yeah, so, so, uh, that wasn't 12 eight, but, but in just, uh, Uh, in, in principle, this, I think is such an interesting thing where you have a sort of base stable, um, based stable perception.
And very importantly, he didn't, he didn't, he did not only do this. He also had this IE flat, uh, flat, uh,
made it, try it on top of, yeah. So,
Benjamin James Kuper-Smith: flat
Peter Vuust: Yeah.
so, so it's, it's really, uh, it's, it's really dissonant, but because it's so dissonant. Almost, it's almost good that it's brought a dissonant on the harmonic and the rhythmic level because, uh, yeah, I think w one wouldn't go without the other.
Then there was B, there would be two big discrepancy between the rhythmic and the, uh, harmonic layer in the music. It's a [00:31:00] wonderful, I wish I actually remember it precisely how the rhythm went, but Yeah.
Benjamin James Kuper-Smith: Yeah. Yeah. Mean just, um, what kind of one question? I mean, so one thing that you already alluded to earlier, Different cultures and these kinds of things. And, um, I'm just curious, like to what extent I just, that kind of the, the, the, the edges of what we might define as music. So, especially when you look at 20th century music, there's a lot of it that at some point, people go, that's not really music and it will become increasingly philosophy philosophical or something if like Stockhausen or whatever.
And you mentioned in the article, you mentioned Schonebeck briefly, I mean, is that like a completely new system of music that you have to learn? So you understand what the predictions on that system, or kind of, how does that work from a predictive coding of music kind of perspective?
Peter Vuust: Yeah. I, I, Yeah.
it's, it's hard to see. I think that, um, I think Shoeneberg deliberately wanted to get out of all these predictable patterns and what. I think what [00:32:00] he actually did when you do an analysis of it, he makes it always even more unpredictable than if you had done it randomly. So he avoids any kind of prediction predictions by making the it's so unpredictable.
And that gives you. It, it gives you a texture, a prediction texture that, that, that is in a way also has it all his own predictability in the, I mean, if this is really hard to, it's hard to explain, but, but, but you know, the 12 tone system where you can't repeat, uh, you have to do all the notes before you, you, you, you, you hear them again.
And then you do these differ in versions and all the stuff that you can do, uh, do with it. So it's sort of a mathematical system, but to my year, when you have you, when you've listened to it, some, a couple of times, I really find it extremely nice because it doesn't have the normal, it stays away from [00:33:00] the, you know, it stays away from all those kinds of predictions and that has a big virtue to it.
So, uh, I think what he's trying to do is to. Yeah. he's, he's making music that is very hard for most people to grasp, but, but I think that there's still some kind of prediction, uh, uh, predictive coding that he, that he plays around with. Yeah.
Benjamin James Kuper-Smith: Yeah. I mean, I find it. Yeah, I've heard it also. I mean, I have to like, I, even though I like. Transaction classical music, the 12 tones and Syrian music from like bullying or something and just never got into, I never quite got it. I don't know, because maybe it wasn't explained to me when they were like, looked into that clearly.
Um, but I do find it interesting that there's kind of some music that you have to, I would just say, I guess kind of, we grew up with like a particular music system and if you grew up with that system, then it can be really hard to listen to other stuff and still find it. Interesting. I mean, [00:34:00] I find like it's, for me, it's very broad in terms of other cultures, but yes, for some, I like 12, 10 music, something that's I guess, intentionally so different that I never really got into
Peter Vuust: But he, of course you have to listen to it many times to the same piece of music. So honestly, honestly, my, my repertoire is super, uh, what can you see slim? But the things that I've been listening to over listening to many times and EDS, it's the same thing for me with the I'm a big fan of John Colton.
But his last records or the ones from, from the mid sixties, their soap just before his, his death. Uh, they're so hard to listen to because, but when, when you do it, when you do it.
enough, like one record you, you, you say, okay, I really want to listen to this. Then you'll be radical. When w if you can stipulate your vertical expertise long enough, you get an extreme amount of beauty in the other stuff.
[00:35:00] So, so, yeah, it's, it's hard, but I think you have to allow yourself to only for instance, no one piece of Schoenberg or, uh, you know, yeah. Or the others also.
Benjamin James Kuper-Smith: I have to admit like the what's it called, like survivor from Warsaw or something like that. That's one that I think I could like to me, that the whole system makes sense. It's just that it makes
Peter Vuust: Yeah.
Benjamin James Kuper-Smith: sense, but, um, yeah, I dunno. The piano musical something. Yeah. Maybe I guess I just have to listen to a piece of often, so actually like, know what it is.
And then I dunno, I mean, in school I took music very, um, like to an advanced level and we discussed and that's the piece that the first time I listened to like, uh, one of the movements, it's like, I have no idea what's going on. Like, it just seemed like random. I was like, I dunno, you know, uh, but then we discussed it and then afterwards, like, oh, this is actually really good.
Once you understand what's going on. It just, it takes a lot of kind of education to get to the level where you can [00:36:00] actually appreciate it. But
Peter Vuust: I have to say, I don't have a full overview of classical music. It's like these, I know these pieces very well. No, Mahler's fifth. I know. Of course, a lot of the all jazz musicians love, uh, love buck for some reason. Uh, so, so, so I know of course I played the cello suites basically every day. So, or parts of it.
Of course you don't play all in all of it, but, but, so, so, uh, so I wouldn't, I always was mostly interested in the French composers, SETI, uh, uh, uh,
Benjamin James Kuper-Smith: with
Peter Vuust: Uh, Debussy avail, uh, and also, uh, and I forget the name of this who was more like Schoenberg, who had these fantastic ideas where he had the same pitch should have the same note, length and so forth.
He had this whole system of, of, of stuff anyway, doesn't matter. So it's, it's not my spell speciality. Uh, yeah.
Benjamin James Kuper-Smith: Yeah. I just find that interesting case, because I guess [00:37:00] at some point you get to the point. Yeah. I wonder whether people still consider it music or not. Um, so like,
Peter Vuust: But one thing that I think is super interesting is that I used to teach, teach music theory at the Royal academy. And I still feel to find an, a theory, theoretical examples, Samuel, where I couldn't use peoples as an example, it would have been, become a little bit boring if I, if I only had that. But it's so interesting how much they were actually, how broad they were able to proudly.
They were able to expand their musical interests over those seven, seven years that they made records. It's still, it's still baffles me. I have to see.
Benjamin James Kuper-Smith: also that it's such a short period of time in which they made with that music. Yeah.
Peter Vuust: Did you watch the, The Peter Jackson movie
Benjamin James Kuper-Smith: The documentary that
Peter Vuust: in a document? Yeah.
Benjamin James Kuper-Smith: I haven't seen.
Peter Vuust: It's, it's mind blowing. I watched the whole nine hours in one stretch or we, we played a little bit, we were [00:38:00] three, three musicians who sat down and watched it and we played a little bit in the, in salute. We could, we simply had to play, but Yeah.
that was, that was amazing.
Benjamin James Kuper-Smith: Yeah. Um, anyway, yeah. I mean, one thing, one example I've got to sound funny is the, the use, the example here of syncopation and, uh, and the, the, what's it called again now? Uh, blame it on the boogie by the Jacksons.
Peter Vuust: Ah, yeah,
Benjamin James Kuper-Smith: uh, it's funny. I like you wrote, like, it's difficult not to tap your feet. And I just saw like the blaming, I blame it on the big, by the Jacksons.
I didn't know the song, so I listened to it. And so I started like tapping to the beach and then read like, it's difficult to not do that as I go. Yeah, you're right.
Peter Vuust: Yeah. yeah. Sorry. I should, I suits mentioned that there is a little caveat to that or an extra thing to that because it's also of course why the, the lyrics goes. I just go. I just can't. I just can't. I just can't control my feet. That's why we wrote it. It's a little bit of a joke, but
Benjamin James Kuper-Smith: Yeah. I mean, it's, it's a perfect example. Yeah. [00:39:00] Yeah. Maybe just about the, so there's one part that didn't quite understand, and this, this relates to the, to this part about, I guess you use the example of syncopation to explain the idea how active inference works in this kind of predictive cutting of music sense.
And I didn't quite, I'm not sure I quite bought the explanation or I didn't, maybe just didn't quite understand it. Yeah. I guess. Yeah. Maybe can you, can you maybe explain that in a bit more detail because I didn't quite understand how you exactly reduce your prediction errors by reinforcing the beat.
Peter Vuust: Oh, yeah, that's, that's a good that we, this discuss this we'll call. And the thing is that if you, if you go back and think about the three against four, then the brain has this prediction or one of the predictions steady. It has that there are other predictions, but one of the predictions is tap. To the beat and the beat is [00:40:00] probably thinking about it in the brain.
It's probably, it's probably mediated by the, by the Mosa system so that you can see that also from, from studies, from, uh, uh, from, uh, uh, , uh, who did these studies, where he could show that the beat prediction were actually sent to the audit sort of courses from the mortar system. So if you think about it like that, you have a beat in your brain, this is sort of the beat.
and this is your prediction, your motor prediction.
And then you have something in a prediction error from the auditory courses is because what comes in is for instance, a single patient that, that, that, uh, that, uh, that, So that doesn't go along with a 1, 2, 3, 4. So, so now of course you could, you, your brain could start to think, okay, is it the. Do I have the right, um, your motor system would, would say, okay, do [00:41:00] I really have the right prediction?
Cause it it's at odds with what, what I listen to what I hear. And then, uh, in order to try to, to establish the prediction, you might tap your feet and that would actually give you some feedback from your body. And maybe also some auditory feedback. That's what you can hear when people dance. Do you have also a lot of audit sort of feedback from that old you might clap while, while the, that is also one where to, to your body I'll pop your head or whatever we do when we listen to to music.
So that reinforces your, your prediction. Of course, it's still challenged because it's the same, but it's an era that is coming in, but at least, you know that, and you also know this from studies, tapping studies is when you, your produce your tapping, then your predictions are also more precise when you produce your, uh, the rhythm, for instance.
So that's, that's basically idea
and, and yeah.
Benjamin James Kuper-Smith: sorry. I mean, I guess the, [00:42:00] yeah, the confusion, I guess, was that. How does, um, reinforcing the beat, reduced a prediction ever of the syncopation because the syncopation is still the same. Right. But is it just that you're reinforcing the framework within which interpreting
Peter Vuust: I don't think it's you, you have sort of standard. It's not necessarily that the brain is capable of producing the prediction error, but this is its task. So this is what it's trying to do. So it's trying to say, well, okay, this was simply wrong. Here is the right place for the beat to be in. Okay.
It was still wrong, but it's the process of reducing prediction error. And that's also probably what makes it pleasurable because at least the way I think about it is that that it's, it's, it's, uh, the process of reducing prediction, error that, that actually make where you actually, how can I explain this?
This is a long, really long explanation, but the process of reducing prediction error is something that is good for you following. [00:43:00] Because you will, you will become better at producing the beat. And that means that you, uh, you will have more precise predictions by being challenged and then have to produce the beat.
For instance, that's basically what we do when we learn this, of course, uh, not exquisite learning, it's implicit learning, but if we, if I were to, to, when I practice, for instance, I often do I place it, press on my base. I've done that for many years where I play, uh, you know, walking bass. That's what all bass players practice
But then I practice to have the three against four in my. Which makes it?
harder, but that actually reinforces my, my sense of the meter. So, so, uh, so, and, and, you know, learning is actually pleasurable to a certain amount when you, when you hit the sweet spot between what is really too hard for you to understand, and [00:44:00] to too easy for you to understand.
So this is where, and we know that from monkey studies, from, uh, uh, from Schultz, uh, monkey studies, that, that, that in there, there is sort of this sweet spot in between where you, where you are, where the, where the monkeys are more rewarded. Uh, so, so I think this is the way it, it, it, it, it, it, uh, it is meant to, to be honest, I'm sure the car could say this much more eloquently, but then you might also lose, you know, you lose exactly what he was saying, but,
Benjamin James Kuper-Smith: Yeah. Yeah. I mean, I had that once because I did a master's project in his lab and then he explained something to me and he's like, does that make sense? It's like, yeah,
Peter Vuust: it's
Benjamin James Kuper-Smith: sure it does, but not to me.
Peter Vuust: no, you know, it's, it's poetry, it's poetry when he speaks. So I'm just trying to, you know, put it a little bit more down to earth so that yeah.
Benjamin James Kuper-Smith: Yeah. By the way, I hope that, so you have a few examples from funk music in this. I hope that these were examples that he [00:45:00] came up with. I would really love if he was just really into funk music.
Peter Vuust: Uh,
I, I don't think I have to say I wrote most of that article
and, and the, the, The others were, were mostly, um, were very, very important in discussing and commenting on the stuff. So, but I, I don't know why the it's it's especially funk music because I'm not, I'm not particularly interested in front of Lucy.
It's just a very, yeah, I guess it's, it's instructive. We also have a uh, I think I have, uh, yeah.
Benjamin James Kuper-Smith: If you mentioned these eight notes.
Peter Vuust: almost to a, you know, almost too down to earth, but.
Benjamin James Kuper-Smith: But you don't have a beach was example, right. If I remember correctly.
Peter Vuust: I do have a beetle stat in the first figure. It's we're certain
Pepper's lonely hearts club band.
Benjamin James Kuper-Smith: about that.
Peter Vuust: Yeah,
Benjamin James Kuper-Smith: So you did
Peter Vuust: yeah. I did get them in. And that was the hardest part about getting it [00:46:00] published. You know, it took six months to get the royalties or to get, to get the copyrights for, for these three bars of beetles.
It was so annoying. It was so
annoying just for the transcripts, which I did myself by the
Benjamin James Kuper-Smith: Yeah. Obviously I assumed so, so they, okay. But did they, did those, like nature publishing have to pay
Peter Vuust: yeah,
probably a by though.
Benjamin James Kuper-Smith: actually owned the Beatles copyright. Wasn't
Peter Vuust: You and I, he brought them back. McCartney brought the back. I
Benjamin James Kuper-Smith: Uh, okay,
Peter Vuust: so uh,
Benjamin James Kuper-Smith: that's good. He's getting,
Peter Vuust: yeah, I think he was good. He was so annoyed about it. I think because he was, I don't know. That's
Benjamin James Kuper-Smith: yeah, that's a different story. Um, maybe the last section of your, uh, of the article is something that is also slightly more religious to actually what I do, because it's about kind of this interpersonal synchronization or rather, um, interestingly. Dyadic interactions, but for strategic decision-making. [00:47:00] So, you know, not, I don't care.
Well, in my recent, at least I ignore like all of the complicated voice and face stuff and all this kind of stuff and just go to the strategic stuff. Um, but I guess the last part is kind of about the social interactions and, uh, yeah, I guess kind of how music can also help, um, study those kinds of areas.
And, um, I, I quite like the, kind of, some of the finger tapping experiments and the kind of three dynamics that emerge from that. Can you maybe talk a little bit about that and.
Peter Vuust: Oh, yeah, I loved them also. It was something that we started a long time ago when I suggested it to one of the other PhD students before I had my own center in a, it was another person's PhDs to study Ivana. kind of a link or. It was so simple and there was, it was so difficult to model. She's an engineer, really a really clever engineer.
And it turned out to be pretty difficult. The idea [00:48:00] is you put two people in separate rooms, you ask them to tap their fingers and they can hear each other. In, in some of the instances, in some instances they can only hear themselves, but then you, um, they tapped, they tapped together. And when they attached together, uh, they are told to synchronize and keep the rhythm.
So keep the temple, that's the only instruction. But then it turns out that at this very, uh, low level, they will be, most of them will be so called hyper follow. So if one goes faster on one beat, the other one will go faster on the next beat, but then the first one will go slower. So they sort of oscillate between, and this was a, and that was actually most of the participants that did this with. and.
this is not something that you are aware of because it's, it's like you go Bing, Bing, Bing, Bing, Bing, and you try to keep the temple, right. And you, you try to, but, but you, you you're really [00:49:00] fast at oscillating in this sense. And, uh, what we have then done is that we have tried to have people tapping with different models, predictive models.
And, and when you, when you do that, you can see that the modest in at first you become not very well synchronized, but then after, uh, some bars you, you synchronize again. So, so having different models either means that you. Converting in the models, which probably is what happens, that, that you change models because you, for some reason you can feel that, well, this model wasn't, it wasn't a three against four.
It was probably just the four or whatever. So, so we can see this at the tablet level level, uh, that you have these interactions, uh, minimalistic interactions. But then what happens when you, when you scan the brains or you use EEG to measure, uh, uh, mascot,
potentials, escalation scent and [00:50:00] so forth, then what you can see is suddenly that there are many things that you can see, but one of the things is that you can see that you have sort of an isolation in the brain that, that goes with your, with your own tapping, but you also have isolations that monitor the other person's.
So the way we did this, and I can't, I can't take any credit of this because this was actually, uh, we'll add in Henley and Matea are also from, uh, EPM in, in Ghent. So what they suggested was, or it was probably material who suggested this was to take two metronomes people what's happening to their own metronomes and there were slightly apart. So, in the beginning they will be very close together. Eh, they would sort of go in antiphase and then after 32, it's a reasons they would be back again. Right. And, and what you can then see is that there's actually a [00:51:00] pizza bursts oscillation. It's sort of underlying, uh, the isolation. So. like 20 Hertz. So it's pretty fast, but it's like, when you tap you, it goes like this, the brain would go something like this.
So there's sort of an underlying lesson that gives you this slower oscillations, that, that, that, that is consistent with your tapping. But what you couldn't then see is that if you're told to keep your own rhythm, but you hear the other, or you see the other, then there's another, there there's also an oscillation that goes to the other at the same time, not as strong as your own.
So you're monitoring Turing the other person. And that means that this super low level musical interaction, which is, I mean, we do that in music all the time. Right. We keep rhythm with each other and sometimes they drift a little bit apart and then we know this guy's a little bit apart, but we have to stick to our own beat.
Right. [00:52:00] So, so this is actually subserved by Newell oscillations in the brain that when it, So, our own tapping, but also predict the peace of the other person and in the predictive coding framework. What I find interesting here is that when you have, uh, when you are playing together, then in the beginning, you might be in different worlds.
When, if you do freed, yes. Let's take the it to the complete and you start the F you in the beginning, you do free. I actually did it. Uh, two days ago I was playing a gig, uh, where we, we, we play some free desks, you know, completely free. Like we didn't know what was to happen. So we asked the
Benjamin James Kuper-Smith: how does, oh yeah, sorry. Yeah. Maybe in general, just free to, does it, is it like, do you have some sort of thing you start when you say that stops at this key? Or do you say, I dunno, it was just like someone starts and
Peter Vuust: sometimes, sometimes there's also a theme sometimes there's no. So you just play or sometimes it [00:53:00] happens when you're weak, if you play a standout or a song. And then in the end of the song, suddenly it goes into some kind of free direction. There are many different ways, but, but, uh, and of course fee is not completely free.
You always bound by by stuff. But, uh, but, but anyway, what we did in this instance was that we asked the audience for three notes and unfortunately they gave us a, what was it? They gave us, they gave us, uh, E.
Benjamin James Kuper-Smith: Okay.
Peter Vuust: so, so, so no, it was pretty boring, but so, and in the beginning, you're far apart, you're three musicians you're playing and you have, you have your own stuff going, but what you need to do is sort of harmonize your predictions.
So suddenly you get into some kind of predictive framework, it might be a tenacity, it might be a meter. It might, it might be another of these things, but still have an idea, oh, this is where we're going. [00:54:00] So you harmonize your predictions. And this is of course the extreme example, but it happens in all kinds of music.
You can also hear it, hear it when he listened to, for instance, a symphony orchestra. Sometimes they're not really together in the beginning and then suddenly. Th there, there they are. But so what you in principle doing is that you are harmonizing predictions at different levels, the radical, the dynamic predictions and the short-term predictions.
And this tapping together is an ultra minimalistic experimental setup where you can use to, to try to, to manipulate, for instance, the relationship between the two persons, like the social relationship between two persons. So, so this is, this is the whole field is going in that direction at the moment, I think.
And, uh, and we, we are, we are part of that, of course, but it's very.
Benjamin James Kuper-Smith: Yeah. Yeah. I really liked how. Know, I really love if you have a very, [00:55:00] very simple experiment and you can already tell so much with it. And I think that's a great example of it. Uh, maybe my last question was we're talking about, uh, jazz and compositions. It was a really interesting sentence in this article, um, about just compositions that I'd never thought about that much because I, you know, I don't, I never really played jazz per se.
And you, you wrote, um, in jazz, one of the most important purposes of compositions is to serve as a framework for sodas to impro improvise on. And I thought that was really interesting because, you know, I come from a classical perspective. It's pretty much like these other notes play them. And, um, so I'm just curious, like how does, I mean, I don't know whether this is a difficult question, but like how do you compose something if you, if the goal is not to yeah.
To not lay out the notes, but rather to provide this kind of framework, um, into which people can insert their own kind of ideas, but it's a very different way of
Peter Vuust: What a wonderful question. Now I've made seven records in my own name, name, [00:56:00] and for all of those records, I had a band in mind when I wrote the compositions. So when, uh, when you w when you, uh, write jazz compositions, there are lots of things you need to think about. There's one thing we need to think of.
How groovy is it? How groovy do I want this to be? What do I need in my set? So you can't have a record where you just play ballots, you can, but, and people do that. And that's also nice, but, but in a way you have all these constraints that you have to think about beforehand. Then I think about my current band is last chance on from, from Sweden is fantastic piano player.
And I know he's, he's an extraordinary player. He can play anything and he really likes likes it. If, if it has some. Either has some really groove to it. So it's not just like a thing to getting things, but it has like a base a[00:57:00]
So it has some kind of funky groove to it. He likes that, but he also likes it. If it has a lot of changing how. So that he, because that really challenges what he can do. So when you ride your conversation, even though you need a melody in some course, and some, uh, some groove, or kind of at least an idea of how, what the drama and the pace should be playing, then you still need to think about what do the different musicians that are going to play this?
What, what, what will will make them be able to express or give them most freedom to express their own personalities and, and how so? It's a really good question. And I can tell you that now I, I just composed 50 new songs that we have to record when I've stopped being tired, or I'm still tired from this record we just did, which was really fantastic.
but also exhausting. So, [00:58:00] so, so, uh, we have to, to do them and for those songs, I have really clear ideas. About why I made them in this way. So there's a negotiation that has to take place, but it's not different from, from Mossad when, I mean, I'm not comparing myself to, so don't worry. But, uh, but I'm just saying that that is not different.
He, of course knew he needs to have a symphony. Now I need to write a symphony or now I need to write a piano concerto because it has to be used in this occasion. And who can, what can they play? What can these guys actually play? What would be difficult for them to play? Because you know, also the symphony orchestra is they don't lie.
I mean, the violin is they don't like just to go E that. I mean, they want some food. So, so, so this is not different from that. Uh, but, but of course, when you have to, the thing, is that when you then make a jazz composition, After I've [00:59:00] done it. And, you know, the, the course and the, the harmonies and the melody fit together.
But then I tried to play solo myself on it. And many times I mega reduced score, uh, for the, for playing solo on. So if I might have, for instance, a C minor court and an F minor, and a, maybe a, an E flat seven chord in the first Bartlet, then I might just reduce it to one C minor chord for that whole bar, because that then, then it gives them more freedom to improvise on without losing of course the sense that it's still the same song.
So, so yeah, that's, that's a, that's part of that equation. Yeah.
Benjamin James Kuper-Smith: Oh, cool. Yeah. I never thought about that. It's yeah. The, I like the comparison to someone like Mozart, because you do have, like, this is the orchestra of the court that I'm, that pays me to do this things and these other musicians and this, you know, they have this many of these instruments and yeah. What can they do?
And, yeah. [01:00:00] Okay. Um, I guess I've, I've run through my questions. Uh, I don't know if there's anything else you'd like to add otherwise.
Peter Vuust: I learned lots of stuff I want to add, but I would encourage people to read the article. I don't know how you it's. It's interesting for me to meet somebody who has actually read the paper. Uh, but I have the feeling we try to really make it assessable. So readable. So that, I mean, maybe the predictive coding part is a little bit rough.
That's, that's tough to get through if you have never heard about predictive coding, but I think that the parts possible, uh, you know, we have melody, we have harmony, we have rhythm. Then we have groove, we have a musical interaction. We have a little bit about, uh, improvise session and, and, and also pleasure and also musicianship.
So there are lots of things where you can actually get the short story of what happened the last 25 [01:01:00] years, I guess, in, in neuroscience. And, and, you know, when you write for nature neuroscience, they have a whole team of people who will just keep on, on asking you questions do really mean this is what you actually mean this, and then you say, oh, I never thought about that.
It's actually good question. And then you have to refine it. So I would say it has really been, there's nothing in that article. I think that I don't believe in, or is an, is a mistake or something. So if you disagree, that's really good because I like this agreement for the scientific purpose of it. But these at least the things that we wanted to explore.
Benjamin James Kuper-Smith: Yeah. And I agree like also what you mentioned right at the beginning, um, this is not just about predictive coding. Um, you know, you do summarize a lot of other music that you don't need. The predictive coding. Full time for the studies to make sense. Yeah. And I don't know. I mean, I may be not the best person to judge how readable it is given that I have, in fact gonna forgive [01:02:00] coding and music, but, uh, yeah, I've, I find it very, I guess, easy and accessible to read.
So I can, at least from my perspective that
Peter Vuust: So the two of us, we are, we completely agree that this is a very easy to read
Benjamin James Kuper-Smith: exactly. Exactly.