WEBVTT

00:00:00.160 --> 00:00:02.320
This is Evolutionary Psychology of the Podcast.

00:00:02.400 --> 00:00:03.920
I'm David Petrozowski.

00:00:04.240 --> 00:00:05.520
And I'm David Pinsov.

00:00:05.760 --> 00:00:10.560
This is actual evolutionary psychology by actual Evolutionary Psychologist.

00:00:23.200 --> 00:00:25.120
David, did you hear that?

00:00:25.520 --> 00:00:26.320
What what?

00:00:26.400 --> 00:00:29.130
I I Oh well uh the nothing in particular.

00:00:29.290 --> 00:00:34.010
I was just wondering because what who we're talking to today is Greg Bryant, and Greg

00:00:34.010 --> 00:00:35.770
Bryant does all things acoustic.

00:00:36.170 --> 00:00:36.730
That's right.

00:00:36.810 --> 00:00:42.410
Uh we get into vocal communication, music, uh why

00:00:42.730 --> 00:00:48.490
different sounds evoke different emotions in us, whether those connections between sounds

00:00:48.490 --> 00:00:52.810
and emotions are arbitrary, or whether they have sort of interesting evolutionary

00:00:52.900 --> 00:00:53.860
explanations.

00:00:54.580 --> 00:00:59.300
Yeah, I think I think for me the mystery of why music makes us feel anything is was

00:00:59.300 --> 00:01:00.740
partially resolved by this.

00:01:00.820 --> 00:01:04.740
And I feel like Greg did an amazing job um organizing my thinking.

00:01:04.820 --> 00:01:06.340
And I knew about Greg's work.

00:01:06.420 --> 00:01:10.260
Um, but even this conversation for me really helped organize my thinking.

00:01:10.340 --> 00:01:14.340
Um, for example, understanding, yeah, like what's up with music, why do we feel emotions

00:01:14.340 --> 00:01:15.060
around that?

00:01:15.220 --> 00:01:18.020
Um, why is rock music the way it is?

00:01:18.100 --> 00:01:23.990
Um, and how is it related to yeah, um you will learn why electric guitars are

00:01:23.990 --> 00:01:27.510
distorted and what role that distortion is playing.

00:01:27.750 --> 00:01:33.110
Yeah the evolutionary psychology of uh overdrive is should be the title of

00:01:33.670 --> 00:01:38.310
of this episode maybe um uh yeah and then we also talk about Greg's other work.

00:01:38.710 --> 00:01:41.190
We spent a lot of time on music which is fantastic.

00:01:41.350 --> 00:01:45.910
Um but we also do talk about um some of his other work about and we get toward the end we

00:01:45.910 --> 00:01:46.950
talk about humor and laughter.

00:01:47.220 --> 00:01:52.020
Oh yeah, I talked a lot about laughter and um uh how it ties in with encryption theory and

00:01:52.100 --> 00:01:54.260
with my own work on uh mix-ups.

00:01:54.420 --> 00:01:58.020
Uh it was a really fun conversation and I learned a lot and had a great time.

00:01:58.260 --> 00:02:01.620
Yeah, so uh this is in acoustic format.

00:02:01.700 --> 00:02:03.460
So sit back, enjoy and listen.

00:02:03.540 --> 00:02:08.980
And yeah, I've I you'll learn a lot and uh maybe think about what you're doing with your

00:02:08.980 --> 00:02:13.620
ears slightly differently after you do the following with your ears, which is a weird way

00:02:13.620 --> 00:02:19.510
to say that, but I guess I'll enjoy listening to a podcast about sound with your ears.

00:02:21.670 --> 00:02:25.190
We usually do just you know like an introduction before your own, so we'll just get right

00:02:25.190 --> 00:02:25.590
into it.

00:02:25.750 --> 00:02:31.670
Um if evolutionary behavioral sciences were a heist movie, um, you would be the acoustic

00:02:31.670 --> 00:02:34.790
expert, is sort of my representation of you.

00:02:35.350 --> 00:02:38.950
Um uh have you always been an audiophile?

00:02:40.040 --> 00:02:44.200
Yeah, well, I mean it's funny you use the term audiophile because that is has a specific

00:02:44.200 --> 00:02:49.000
meaning, um, meaning that it's people that are into super high-end gear, and there's

00:02:49.000 --> 00:02:54.440
actually can be almost a religious element where you think there are differences in sound

00:02:54.440 --> 00:02:59.640
between the tube amp with the shielded cables, the whatever.

00:02:59.960 --> 00:03:03.850
Yeah, Versus a solid state modeling app that sounds exactly the same.

00:03:04.010 --> 00:03:04.490
And yeah.

00:03:04.890 --> 00:03:05.370
That's the thing.

00:03:05.450 --> 00:03:09.210
So I usually land on that side of the debate because I'm a perceptual psychologist, yeah.

00:03:09.370 --> 00:03:13.850
Um, but I am I mean I am an audiophile and that's I was raised by audiophiles.

00:03:13.930 --> 00:03:19.530
My dad was was into stereos and he was a um actually in the sound um

00:03:20.010 --> 00:03:25.500
in the army corps with uh the uh the uh radio corps, sorry.

00:03:25.900 --> 00:03:26.940
Nice army.

00:03:27.100 --> 00:03:32.460
So he was interested in that kind of stuff, and so we always had nice stereos, and my dad

00:03:32.460 --> 00:03:37.420
was really into music, and my brother got into that that way, and then he does is he's a

00:03:37.420 --> 00:03:42.300
software engineer who works on audio compression, so it's a sound situation in my family,

00:03:42.380 --> 00:03:43.900
so I guess that's where I get it.

00:03:44.060 --> 00:03:49.890
And then I was obviously very into music from a young age, And so yeah, I've always kind of

00:03:49.890 --> 00:03:50.450
been into sound.

00:03:50.530 --> 00:03:55.410
And that was sort of my attraction to um evolution and behavior was the connection between

00:03:55.410 --> 00:04:00.610
language and music and trying to think about, you know, speech communication and vocal

00:04:00.610 --> 00:04:01.570
communication in general.

00:04:01.810 --> 00:04:02.530
Yeah, that's awesome.

00:04:02.610 --> 00:04:07.410
I mean, uh, learn versus innate is a dumb distinction, but I mean it captures something in

00:04:07.410 --> 00:04:11.860
w looking at your family and being like, wow, we're really all into music, have you thought

00:04:11.860 --> 00:04:14.740
about, you know, like is it is it that because it was all around?

00:04:14.820 --> 00:04:16.420
Or or what's your take on those?

00:04:16.580 --> 00:04:16.740
Yeah.

00:04:17.140 --> 00:04:21.380
Yeah, that I mean that probably that must have played some role, right?

00:04:21.540 --> 00:04:26.980
I mean, it's hard to say, though it's funny though, the music thing is they're into music

00:04:26.980 --> 00:04:29.460
listening, but not into music performance.

00:04:29.780 --> 00:04:33.540
So I had to beg my parents to give me any kind of musical instrument.

00:04:33.700 --> 00:04:36.100
And I didn't really get one until I was about fifteen years old.

00:04:36.260 --> 00:04:39.890
I got an electric guitar, which I'm and at the time they were probably like, this is why we

00:04:39.890 --> 00:04:40.930
didn't want to do this.

00:04:42.530 --> 00:04:43.730
Because you were like distortion.

00:04:43.970 --> 00:04:44.610
What was it?

00:04:45.570 --> 00:04:51.490
It was a Hondo strat copy, and then a Sun 20-watt

00:04:51.650 --> 00:04:55.810
amp, which that I wish I still had that thing because Sun is a really cool make and they

00:04:55.810 --> 00:04:57.650
had a really distortion built in.

00:04:57.810 --> 00:05:01.160
But I was into like speed metal and punk rock.

00:05:01.240 --> 00:05:03.320
I mean, it was probably a nightmare for them.

00:05:03.560 --> 00:05:08.280
Um, but I mean, I would have played piano if I had one and I wanted a guitar when I was

00:05:08.280 --> 00:05:08.760
eight years old.

00:05:08.840 --> 00:05:12.280
I actually asked first for a guitar and they just wouldn't give me an instrument because I

00:05:12.360 --> 00:05:15.960
think they you know they had the classic, you know, you're we're gonna buy it for you and

00:05:15.960 --> 00:05:17.400
then it's just gonna sit around.

00:05:18.520 --> 00:05:21.720
And they didn't believe that I was really into it, and I was.

00:05:21.880 --> 00:05:22.200
Right.

00:05:22.520 --> 00:05:26.510
So in that sense, they've they hindered my music development.

00:05:27.070 --> 00:05:27.630
I'm sorry.

00:05:27.710 --> 00:05:28.110
Yeah.

00:05:28.430 --> 00:05:33.550
I probably would, you know, I would have different skills if I had learned m instruments

00:05:33.550 --> 00:05:34.750
when I was eight, you know.

00:05:34.990 --> 00:05:35.150
Yeah.

00:05:35.390 --> 00:05:38.510
Like I mean, in in some ways, maybe it would be better.

00:05:38.830 --> 00:05:40.750
I mean, I don't read music.

00:05:41.070 --> 00:05:46.000
Um, I understand music theory uh you know conceptually, but I can't sight read.

00:05:46.160 --> 00:05:49.280
And so if I look at some sheet music, I can't play it on the guitar.

00:05:49.360 --> 00:05:52.240
Without meaning it would be a lot of work for me to figure that out.

00:05:52.800 --> 00:05:55.840
Um and so that gives you different skills.

00:05:55.920 --> 00:06:01.600
Like, for example, I might have developed perfect pitch if I had learned the notes and the

00:06:01.600 --> 00:06:03.680
theory and the instrument when I was young.

00:06:03.840 --> 00:06:06.240
And now I'll never learn I'll never have perfect pitch.

00:06:06.320 --> 00:06:08.080
Once you're an adult, the evidence is pretty good.

00:06:08.480 --> 00:06:11.810
Once you pass that threshold, you never get perfect pitch.

00:06:11.890 --> 00:06:13.650
So I have good relative pitch.

00:06:14.290 --> 00:06:19.330
Um, but yeah, so there's these certain little skills that you get when you learn early that

00:06:19.330 --> 00:06:21.570
are just obtainable later.

00:06:21.810 --> 00:06:23.970
So yeah, it was kind of an interesting combination.

00:06:24.050 --> 00:06:25.970
They were really into the listening, but not the playing.

00:06:26.290 --> 00:06:26.610
Interesting.

00:06:26.770 --> 00:06:26.930
Yeah.

00:06:27.010 --> 00:06:27.810
Well, this is a PSA.

00:06:28.130 --> 00:06:28.450
Sorry, David.

00:06:28.690 --> 00:06:29.090
PSA.

00:06:29.330 --> 00:06:34.500
If your kid wants to play music, uh my editorial here is let them.

00:06:34.580 --> 00:06:38.100
And if they don't use it, you can always put the thing on reverb and get a good reset.

00:06:38.420 --> 00:06:40.820
Yeah, let them if they want to, but don't push it on them either.

00:06:40.980 --> 00:06:41.540
No, yes, exactly.

00:06:44.420 --> 00:06:47.380
I know so many people that were forced to take piano lessons that now have like an

00:06:47.380 --> 00:06:48.900
aversion, they don't even want to do anything.

00:06:49.300 --> 00:06:50.020
Totally, David.

00:06:50.100 --> 00:06:50.820
Yeah, really good.

00:06:51.060 --> 00:06:51.380
Exactly.

00:06:51.460 --> 00:06:52.100
Just like horror.

00:06:52.180 --> 00:06:57.270
It's it's it's you let them be autonomous and let the and let them let them do overdrive.

00:06:57.350 --> 00:06:59.430
You can give them uh headphones and stuff.

00:06:59.670 --> 00:07:01.750
Uh Greg, can you talk about perfect pitch?

00:07:01.990 --> 00:07:05.110
Cause that's like maybe a nice little hook into content.

00:07:05.350 --> 00:07:08.550
Um because uh you know I've been reading into this too.

00:07:08.630 --> 00:07:12.870
I came across this because I had always had this toy idea that perfect pitch was just

00:07:12.870 --> 00:07:14.710
something people are born with or not.

00:07:14.950 --> 00:07:18.550
Um for people who don't know what it is, can you explain what it is and then can you tell

00:07:18.550 --> 00:07:22.470
us if to extent you know I mean I'm I know this is not your primary research, but like what

00:07:22.470 --> 00:07:24.380
is the state of the art that we know about it?

00:07:24.700 --> 00:07:25.100
Yeah.

00:07:25.340 --> 00:07:30.700
Well, uh the idea of perfect pitch, my understanding, is that when um a person can map a

00:07:30.940 --> 00:07:32.620
tone onto a note.

00:07:33.180 --> 00:07:38.140
So if you hear, you know, a tone of whatever it might be, you can just say that's an A.

00:07:38.460 --> 00:07:38.860
That's right.

00:07:38.940 --> 00:07:44.300
And you if you say like do an A, they can hum it or uh Exactly.

00:07:44.590 --> 00:07:46.990
So they um and they nail it right exactly.

00:07:47.150 --> 00:07:47.790
Yeah, yeah.

00:07:48.030 --> 00:07:52.590
And so that's something that you learn that mapping between the sound and the category

00:07:52.590 --> 00:07:53.310
early.

00:07:53.550 --> 00:07:56.830
And um because obviously the category is culturally constrained.

00:07:56.910 --> 00:08:01.150
There's no way it could be an A to know an A because A is a cultural invention.

00:08:01.390 --> 00:08:02.190
Yeah, right, exactly.

00:08:02.430 --> 00:08:04.510
In German it's like there's another different note.

00:08:04.590 --> 00:08:06.160
I know Annie always talks about that.

00:08:06.480 --> 00:08:09.840
It's like it's a it's it's like a different there's like an H or something.

00:08:10.640 --> 00:08:11.680
I I forget the DS.

00:08:11.840 --> 00:08:16.080
I'm I'm completely a musical, so I'm is it the fun note, the huh?

00:08:18.640 --> 00:08:19.840
Yeah, that's exactly it's right.

00:08:20.080 --> 00:08:22.400
Yeah, good and talk to all of our German uh friends.

00:08:22.560 --> 00:08:23.040
So yeah.

00:08:25.670 --> 00:08:25.990
Yeah.

00:08:26.230 --> 00:08:31.830
So obviously, you know, you have to learn some system, and there's a lot of them out there,

00:08:32.070 --> 00:08:32.950
potentially.

00:08:33.190 --> 00:08:38.630
Um, so if it's learned within a critical period, which I think is it's right in the

00:08:38.630 --> 00:08:39.750
language critical period.

00:08:39.830 --> 00:08:41.750
So it's kind of like before your five.

00:08:41.910 --> 00:08:45.430
And people a lot of people say, I had perfect pitch and I never had music lessons.

00:08:45.510 --> 00:08:50.150
It's like, except that one time, you know, that it always, if you push them, it always, oh,

00:08:50.230 --> 00:08:51.270
there was one little thing.

00:08:51.430 --> 00:08:52.790
It doesn't take much, I think.

00:08:52.870 --> 00:08:54.860
You know, kids soak it up pretty fast.

00:08:55.100 --> 00:09:00.940
Um, once you pass that threshold, I think the evidence is that that you can't learn

00:09:00.940 --> 00:09:01.180
it.

00:09:01.420 --> 00:09:07.100
So you have to have some sort of relative pitch to the know you know the

00:09:07.100 --> 00:09:11.580
difference between it and so you have a reference pitch, then you can say what the note is

00:09:11.660 --> 00:09:12.780
if you're really good at it.

00:09:12.940 --> 00:09:13.660
You can learn that.

00:09:13.900 --> 00:09:15.020
And and can I just interrupt?

00:09:15.180 --> 00:09:19.270
Is it that because I mean, so you know, I I know somebody with perfect pitch, like it's

00:09:19.350 --> 00:09:23.750
when you when they play a note or you do a note, it's it's it would be like asking the most

00:09:23.750 --> 00:09:25.110
people like what color is this?

00:09:25.190 --> 00:09:26.950
It's like, oh my gosh, of course it's red, right?

00:09:27.030 --> 00:09:29.510
It's like it's boring, and I'm just like, I don't I don't know how you do that.

00:09:29.750 --> 00:09:35.350
Um, is it that the lessons create the discrete

00:09:35.350 --> 00:09:37.190
mappings between the notes?

00:09:37.270 --> 00:09:42.870
Or is it just that you need a linguistic um label, but the shelf, the

00:09:42.870 --> 00:09:47.360
cognitive shelf for that um note is already there, if that makes sense.

00:09:47.440 --> 00:09:52.080
Or if that I don't know if I'm making the uh well, I mean the instruments produce there's

00:09:52.080 --> 00:09:52.960
not a lot of variation.

00:09:53.040 --> 00:09:58.080
So I mean if they learned with a bad piano, for example, then that would probably screw the

00:09:58.080 --> 00:09:58.800
mechanism up.

00:09:59.040 --> 00:09:59.360
Okay.

00:09:59.600 --> 00:10:04.080
Um, because then people later would say they'd play A440 and then they learned it like

00:10:04.080 --> 00:10:05.840
A452.

00:10:06.000 --> 00:10:07.920
Oh, I think that's then they'd they would that might.

00:10:08.000 --> 00:10:12.800
I mean, I'd that would be kind of a bad experiment to do to do that.

00:10:13.280 --> 00:10:15.850
But um, but I think that could mess it up.

00:10:16.010 --> 00:10:19.450
So the thing is that people are learning with instruments that are tuned properly, you

00:10:19.450 --> 00:10:22.490
know, to like a regular piano that's tuned reasonably.

00:10:22.570 --> 00:10:24.090
But I mean there's probably some error in there.

00:10:24.330 --> 00:10:24.570
I see.

00:10:24.730 --> 00:10:25.530
So it's the intervals.

00:10:25.690 --> 00:10:31.290
They're actually learning the the the fundamental frequency intervals, or like what counts

00:10:31.290 --> 00:10:35.050
as a step in in a continuous range of frequencies.

00:10:35.450 --> 00:10:38.890
They do, except actually you learn it without reference pitch.

00:10:38.970 --> 00:10:42.490
So you add a context, just one note, you can do it.

00:10:42.620 --> 00:10:46.220
So that means you it's not necessarily about an interval because the interval is two or

00:10:46.220 --> 00:10:46.780
more notes.

00:10:47.020 --> 00:10:47.420
Oh, I see.

00:10:47.500 --> 00:10:48.220
I was using the wrong term.

00:10:48.380 --> 00:10:49.740
I just meant a space.

00:10:49.820 --> 00:10:50.060
Sorry.

00:10:50.140 --> 00:10:51.100
I I should just share it.

00:10:51.180 --> 00:10:52.060
I know nothing about music.

00:10:52.380 --> 00:10:55.260
Yeah, the interval is this a real relationship between two notes.

00:10:55.500 --> 00:10:55.980
Right, yeah.

00:10:56.060 --> 00:10:56.220
Right.

00:10:56.460 --> 00:10:58.780
And um, and so they learn those two.

00:10:59.180 --> 00:11:03.100
Um, and you learn the scales and all that stuff, you learn it in a different way when

00:11:03.100 --> 00:11:03.740
you're young.

00:11:04.140 --> 00:11:08.510
But yeah, so if you but just out of context, they're doing a mapping to some category,

00:11:08.670 --> 00:11:09.470
whatever it might be.

00:11:09.630 --> 00:11:13.630
If it's you know, if you're Western music, then you're learning the the letters.

00:11:13.790 --> 00:11:17.470
Um, but then in another system you're learning it in some other different way, but you're

00:11:17.470 --> 00:11:20.830
mapping the the frequency to the category.

00:11:21.390 --> 00:11:26.830
So I'm I'm curious, so um what the relationship is between our

00:11:26.830 --> 00:11:31.700
representation of sound and the objective reality of vibrational frequencies.

00:11:31.860 --> 00:11:34.980
Because I know we were talking with Nick Graziano about the relationship between our

00:11:34.980 --> 00:11:39.220
representation of color and the objective facts about uh color.

00:11:39.380 --> 00:11:44.260
And you know, we we're going into this this cool idea that like we we view white light as

00:11:44.260 --> 00:11:48.260
like a kind of purity or um absence of contamination, right?

00:11:48.420 --> 00:11:50.260
It's just like there's nothing there, it's pure and bright.

00:11:50.340 --> 00:11:53.300
But it's actually if you look into the physics of it, it's the opposite.

00:11:53.460 --> 00:11:56.420
White light is all the different wavelengths just sort of smooshed together.

00:11:56.580 --> 00:12:01.730
And it's weird that our representation of white doesn't quite fit the objective reality of

00:12:01.730 --> 00:12:01.810
it.

00:12:01.970 --> 00:12:05.570
There are plenty of other ways in which our representation of color doesn't quite fit the

00:12:05.570 --> 00:12:06.850
objective facts of color.

00:12:07.010 --> 00:12:11.890
Like, you know, if I see a red object in different lighting conditions, the frequencies of

00:12:11.890 --> 00:12:14.610
light that are hitting my retina are going to be very different, but I'm still going to see

00:12:14.610 --> 00:12:15.490
it as the same color.

00:12:15.650 --> 00:12:16.850
This is color constancy.

00:12:17.010 --> 00:12:20.930
We see an object that looks the same, even crazy.

00:12:21.250 --> 00:12:25.360
That like the objective frequencies of light are totally different, and yet it seems

00:12:25.360 --> 00:12:29.520
exactly the same from our perspective because color is designed to identify and

00:12:29.520 --> 00:12:33.600
differentiate objects and not to necessarily accurately represent reality.

00:12:33.680 --> 00:12:38.480
So are there similar ways in which our representation of sound is different from the

00:12:38.480 --> 00:12:43.440
objective physics of sound, or is it more accurate than it is with color or or how do how

00:12:43.440 --> 00:12:44.400
do we think about this?

00:12:44.990 --> 00:12:45.950
That's a great question.

00:12:46.030 --> 00:12:49.950
And I don't think I have a great answer, other than I would suspect, I think it's more

00:12:49.950 --> 00:12:55.470
accurate that those properties in the world are there to a more substantial degree.

00:12:55.630 --> 00:12:58.110
Whereas color is reflected off it.

00:12:58.190 --> 00:13:01.390
I mean, some would argue there is no color in the world in a sense, right?

00:13:01.470 --> 00:13:04.030
That we're it's like we're constructing it.

00:13:04.190 --> 00:13:07.070
Um and it does change in context considerably.

00:13:07.230 --> 00:13:12.720
I mean, you could there are contextual effects, I think, in in sound where um you know the

00:13:12.720 --> 00:13:17.680
same sound presented in two contexts would be perceived differently and then represented

00:13:17.680 --> 00:13:17.920
differently.

00:13:18.320 --> 00:13:18.800
Oh, interesting.

00:13:18.880 --> 00:13:21.280
So it's like sort of an analog to color constancy.

00:13:21.680 --> 00:13:23.520
I think there's some similarities there, though.

00:13:23.600 --> 00:13:26.560
I can't really come up with a really great um example exactly.

00:13:26.720 --> 00:13:32.400
But um in like auditory scene analysis, which is the study of how people perceive the

00:13:32.400 --> 00:13:37.280
auditory environment, how you localize a sound, and how you decide that these frequencies

00:13:37.280 --> 00:13:42.590
could should belong in this or belong to this source, and these frequencies belong in this

00:13:42.590 --> 00:13:45.310
source, even though it's coming in as a wash of frequencies.

00:13:45.390 --> 00:13:50.190
We have this coherent separation of sounds that seems like that that sounds coming from

00:13:50.190 --> 00:13:53.550
there and that's coming from there, even though all these frequencies are all around you

00:13:53.630 --> 00:13:54.750
and you have to parse it.

00:13:54.910 --> 00:14:00.350
So, in that sense, you're constructing these representations of things that are probably um

00:14:00.670 --> 00:14:06.110
not completely accurate as a function of what they really are and what those things are

00:14:06.110 --> 00:14:06.750
emitting.

00:14:07.000 --> 00:14:09.400
Um, but I think it's more accurate than color.

00:14:09.560 --> 00:14:11.480
Color seems to me more constructed.

00:14:11.640 --> 00:14:17.080
But that's I'm I'm not gonna bet much on that conjecture.

00:14:17.240 --> 00:14:21.080
But um I mean a fundamental frequency is a fundamental frequency, and I think we're

00:14:21.080 --> 00:14:27.000
detecting that in sources that have it um in a way that is pretty the the mapping is pretty

00:14:27.000 --> 00:14:28.280
isomorphic.

00:14:29.150 --> 00:14:29.470
Yeah.

00:14:30.030 --> 00:14:35.310
It I mean it seems like the the selection pressure for auditory scene analysis or knowing

00:14:35.310 --> 00:14:39.390
when certain sounds are coming from a particular source would be you know there'd be huge

00:14:39.390 --> 00:14:40.430
selection pressure on that.

00:14:40.590 --> 00:14:44.910
Like to know, you know, to differentiate the birds from the rustling in the leaves, you

00:14:44.910 --> 00:14:49.390
know, from the potential sound of a predator or someone else sneaking up on you, like being

00:14:49.390 --> 00:14:52.510
able to make those fine grained distinctions probably would have had huge fitness

00:14:52.510 --> 00:14:53.070
implications.

00:14:53.310 --> 00:14:53.470
Yeah.

00:14:53.550 --> 00:14:54.670
So like how big is it?

00:14:54.920 --> 00:14:55.160
Yeah.

00:14:55.240 --> 00:14:55.960
How big is it?

00:14:56.040 --> 00:14:56.440
Yeah.

00:14:56.600 --> 00:14:57.640
Um is it approaching?

00:14:57.720 --> 00:14:59.800
Is it is it uh receding?

00:14:59.960 --> 00:15:02.600
You know, those are really important distinctions, you know, survival relevant

00:15:02.600 --> 00:15:03.240
distinctions.

00:15:03.480 --> 00:15:04.280
Uh yeah.

00:15:04.520 --> 00:15:08.680
Um But there are, I mean, that's you just brought up a good one, like John Newhoff stuff

00:15:08.760 --> 00:15:10.120
with auditory looming, right?

00:15:10.280 --> 00:15:15.240
We overestimate how quickly something will arrive and we don't have that same thing for a

00:15:15.240 --> 00:15:17.080
thing that's moving away from us.

00:15:17.320 --> 00:15:22.600
So there are gonna be various kinds of you know, air management issues and and uh

00:15:22.760 --> 00:15:24.280
distortions and yeah.

00:15:24.590 --> 00:15:29.390
Craig, I wanted to uh I wanted to ask um w either you can pick which one you want to do.

00:15:29.470 --> 00:15:33.950
One is I have this feeling that like vision science has its own conference.

00:15:34.030 --> 00:15:36.990
Do do uh do acous does acoustics have its own science?

00:15:37.150 --> 00:15:38.190
And is it is large?

00:15:38.350 --> 00:15:38.430
Yeah.

00:15:38.830 --> 00:15:42.030
Like I feel like as an intro psych professor, I teach vision.

00:15:42.110 --> 00:15:43.630
I don't teach teach acoustics.

00:15:43.790 --> 00:15:48.270
I feel like acoustics gets the shaft a bit compared to like the other perceptions.

00:15:49.080 --> 00:15:49.480
Totally.

00:15:49.720 --> 00:15:53.320
The the main journal, the one of them, in my opinion, the best journal in the field for

00:15:53.320 --> 00:15:58.680
acoustics, Jazza, Journal of Acoustical Society of America, its impact factor is like 1.4.

00:15:59.000 --> 00:16:02.040
I mean, it's it's not um there's not a lot of work.

00:16:02.200 --> 00:16:05.480
I mean, there they there is a lot of different kinds of stuff, like underwater acoustics,

00:16:05.720 --> 00:16:07.880
and I mean, you know, it's a crazy field.

00:16:08.040 --> 00:16:12.920
And a lot of it is um not psychophysics, a lot of it's just pure, it's just you know,

00:16:13.080 --> 00:16:17.910
engineering and yeah, it's other not psychology or even behavioral science at all.

00:16:18.070 --> 00:16:21.510
But um yeah, no, there is definitely an acoustics field.

00:16:21.590 --> 00:16:23.110
It's way smaller than vision.

00:16:23.270 --> 00:16:27.990
I mean, when you have like acoustics, acoustic perception, sound perception, that's

00:16:27.990 --> 00:16:32.710
definitely uh a lot, you know, there's a lot less work out there, a lot less people doing

00:16:32.710 --> 00:16:32.870
it.

00:16:33.110 --> 00:16:36.070
I I I realize this even because, you know, we've done work together.

00:16:36.150 --> 00:16:40.710
And even I did social categorization work where I started with visual cues like race or or

00:16:40.710 --> 00:16:43.560
team membership visually marked, but I wanted to get into accent.

00:16:43.640 --> 00:16:46.200
And then like starting to make the acoustic stimuli.

00:16:46.360 --> 00:16:50.520
And you know, I consulted with you about how to do it, but I realized like there is very

00:16:50.520 --> 00:16:56.360
much a modality bias in the behavioral sciences, I feel like, at least from what I've seen.

00:16:56.520 --> 00:16:59.240
Um, and it sounds like that might be true.

00:16:59.400 --> 00:17:02.280
And is that because it's or why is that?

00:17:02.600 --> 00:17:04.600
Like I mean, it's a good question.

00:17:04.840 --> 00:17:10.430
Um what I do think is auditory perception and vision perception share this

00:17:10.430 --> 00:17:14.430
feature that they've made a lot of progress in understanding how the brain does it because

00:17:14.430 --> 00:17:17.470
of an inherent functionalist perspective from the beginning.

00:17:17.630 --> 00:17:17.870
Right.

00:17:17.950 --> 00:17:21.870
So people have been studying auditory perception since MLTs.

00:17:22.190 --> 00:17:27.790
And um, so I think we like vision perception, we have actually a better understanding

00:17:27.790 --> 00:17:29.550
relative to other areas of our psychology.

00:17:29.710 --> 00:17:33.390
We understand vision better because I think of the functionalist perspective and being able

00:17:33.390 --> 00:17:37.600
to come up with theories that are testable about how, and then you can look at how it

00:17:37.600 --> 00:17:38.720
happens in the brain.

00:17:38.960 --> 00:17:42.400
Um, so there are auditory perception people out there.

00:17:43.040 --> 00:17:48.000
But um I think one reason, I mean you can say this for vision too, but I think what one

00:17:48.000 --> 00:17:51.920
reason is the technical, there's a technical constraint that you have to really get into

00:17:52.240 --> 00:17:56.880
the um the acoustic theory and you and you also have to um be able to learn how to do

00:17:56.880 --> 00:18:00.480
acoustic analysis, which until recently was actually rather hard.

00:18:00.720 --> 00:18:06.170
And this is a funny thing with Lita, one of her first papers, um she was doing

00:18:06.570 --> 00:18:07.210
analysis.

00:18:07.370 --> 00:18:11.610
And it's really I I love this paper, and it really was a huge, it had a huge influence on

00:18:11.610 --> 00:18:11.770
me.

00:18:11.850 --> 00:18:15.610
It's called Um Acoustic Invariances and the Expression of Emotion and Speech.

00:18:15.690 --> 00:18:21.290
And she it was her honors thesis that she did in as an undergrad, posted JP, of course.

00:18:21.450 --> 00:18:27.440
Um, like all honors thesis is Um and I really love this paper.

00:18:27.520 --> 00:18:32.160
And then when I approached her early on and I was interested in working with them, and I

00:18:32.240 --> 00:18:34.080
she said that that gave her PTSD.

00:18:34.160 --> 00:18:38.960
I mean, it was so hard to do that analysis that she was like, I'm never doing acoustics

00:18:38.960 --> 00:18:39.200
again.

00:18:39.360 --> 00:18:43.040
She hated it because it was really difficult in those days.

00:18:43.280 --> 00:18:45.040
You didn't have software that did it.

00:18:45.120 --> 00:18:48.410
And so you had to actually fire up the oscilloscope.

00:18:48.490 --> 00:18:50.410
And I mean, it was kind of a nightmare.

00:18:50.490 --> 00:18:55.930
And so I think that probably had an impact on how people choosing that as an area of study

00:18:56.010 --> 00:18:59.130
because there was sort of a high learning curve, especially back in the day.

00:18:59.450 --> 00:19:02.810
Now you have software packages that are relatively easy to use.

00:19:02.970 --> 00:19:06.490
If you know how to use R, you can figure out how to do a lot of these things.

00:19:06.570 --> 00:19:10.410
And Pratt is a standalone program that allows you to do acoustic analysis.

00:19:10.810 --> 00:19:14.490
And so more people are doing it, which has the downside of people doing it and they don't

00:19:14.490 --> 00:19:15.530
really know what they're doing.

00:19:16.220 --> 00:19:19.820
So, but more people are doing it because it is accessible.

00:19:19.900 --> 00:19:25.420
And so I I think that um it's probably going to get better, but for some reason a lot of

00:19:25.420 --> 00:19:30.220
people just are not drawn to doing that kind of work where you're doing psychophysics

00:19:30.300 --> 00:19:30.860
essentially.

00:19:31.020 --> 00:19:31.260
Yeah.

00:19:31.420 --> 00:19:32.140
That's what I like.

00:19:32.380 --> 00:19:32.620
Yeah.

00:19:32.700 --> 00:19:36.380
My I always talk about my favorite movie is Jurassic Park and watching I've watched it so

00:19:36.380 --> 00:19:39.740
many times and like what I realized after watching it multiple times is the uh sound

00:19:39.740 --> 00:19:43.580
engineering is so important for highlighting and steel uh Spielberg does this with all of

00:19:43.580 --> 00:19:48.370
his movies is the sound tells you what to look at in the scene and you don't even realize

00:19:48.370 --> 00:19:48.530
it.

00:19:48.610 --> 00:19:52.530
But like, yeah, it's it's just such an interesting thing of like sound is so important.

00:19:53.010 --> 00:19:58.610
And but it f I it's just I I find it funny that it is so um there's so much instinct

00:19:58.610 --> 00:20:03.490
blindness generally, but then even like even among like behavioral scientists who take an

00:20:03.490 --> 00:20:06.770
evolutionary approach, there's like I feel like there's just not enough acoustic work, like

00:20:06.850 --> 00:20:08.050
it's like extra instinct blindness.

00:20:08.130 --> 00:20:08.930
Yeah, no, I agree.

00:20:09.080 --> 00:20:13.080
I mean, vision too, really, there's not that many people that are doing um vision

00:20:13.080 --> 00:20:14.600
perception in the ebb psych.

00:20:15.160 --> 00:20:21.080
I mean, you have people in in doing cognitive psychology and psychophysics that

00:20:21.080 --> 00:20:21.800
do vision.

00:20:21.960 --> 00:20:25.800
Um, you know, like we had one guy in my department where I went to grad school, Bruce

00:20:25.800 --> 00:20:28.520
Bridgman, who was into evolutionary psychology too.

00:20:28.760 --> 00:20:34.250
But um, you know, they're almost nobody in EbSight does that, right?

00:20:34.410 --> 00:20:40.250
I mean, there was um Josh New Yeah did that kind of stuff and then he left

00:20:40.250 --> 00:20:40.650
academia.

00:20:41.050 --> 00:20:44.250
Russell Jackson sort of doing his Russell left academia too.

00:20:45.530 --> 00:20:46.890
Oh yeah, okay, yeah.

00:20:47.530 --> 00:20:50.970
And so and of two, if you go into vision, you're gonna leave.

00:20:51.690 --> 00:20:52.010
Yeah.

00:20:53.130 --> 00:20:53.930
Don't say that.

00:20:54.330 --> 00:20:59.960
Um yeah, so there's there's for some reason uh psychophysics is not really taking

00:20:59.960 --> 00:21:02.360
hold in evolutionary psychology, unfortunately.

00:21:02.440 --> 00:21:04.680
I thought it would have been now, but yeah.

00:21:05.080 --> 00:21:08.200
Um, that that was, and I was gonna give you a choice, but that was the question.

00:21:08.360 --> 00:21:12.200
The other question, like, because this is a big hard question, and maybe we can just pause

00:21:12.200 --> 00:21:12.440
it.

00:21:12.600 --> 00:21:18.570
But my great mystery, and the thing I wanted to ask you more, is uh number

00:21:18.570 --> 00:21:20.170
one, what's up with music?

00:21:20.410 --> 00:21:24.570
And number two, I asked a bunch of people, like I told people you're coming on, and

00:21:24.570 --> 00:21:26.410
everyone wanted to know what your take on music was.

00:21:26.650 --> 00:21:32.010
And number two, what I find a mystery, and David has alluded to this, is number one, what's

00:21:32.010 --> 00:21:32.490
up with music?

00:21:32.570 --> 00:21:37.290
But number two, and equally, uh, why do why are emotions

00:21:37.850 --> 00:21:43.760
hooked up to these weird vibrations in the air, such that a vibration in the air

00:21:43.760 --> 00:21:48.400
can make me feel a certain way and really feel that way in a way that so many other things

00:21:48.400 --> 00:21:48.880
cannot.

00:21:49.040 --> 00:21:50.320
Like that is weird.

00:21:50.560 --> 00:21:55.040
I would never have predicted that like from first principles, at least for me.

00:21:55.200 --> 00:22:01.200
Um, and it's in and it's just um inevit so obvious and it's such a weird thing, I I feel

00:22:01.200 --> 00:22:01.520
like.

00:22:01.910 --> 00:22:02.230
Yeah.

00:22:02.390 --> 00:22:06.070
So the you know, you're saying like vocal emotion communication or something like that.

00:22:06.470 --> 00:22:11.110
No, I mean like if I'm at a concert and there's a symphony in front of me, that symphony

00:22:11.110 --> 00:22:11.590
can connect to the colour.

00:22:11.990 --> 00:22:13.350
It's a powerful experience to hear this loud.

00:22:13.590 --> 00:22:16.710
Yeah, and it's just like that is such a while how how is that?

00:22:16.790 --> 00:22:17.910
Why why isn't evolving?

00:22:24.550 --> 00:22:28.620
Listen, all these different emotions based on the the type of you know the difference

00:22:28.620 --> 00:22:33.340
between minor and major minor being more sad and gloomy like this is like it's so hooked up

00:22:33.340 --> 00:22:39.260
with the emotions like what what yeah yeah so well my take on it is that it's

00:22:39.260 --> 00:22:45.020
it's basically that vocal communication of emotion is underlying all of that

00:22:45.180 --> 00:22:49.260
and then cultural evolution has built I mean this is the simple answer we can get into it

00:22:49.340 --> 00:22:55.170
more detail but that cultural evolution has like built this whole um set of

00:22:55.170 --> 00:23:01.010
systems, technology and practices that have allowed us to make these huge sounds, like huge

00:23:01.010 --> 00:23:04.690
orchestras and giant sounds and movies and everything.

00:23:04.850 --> 00:23:10.770
But that the the underlying sensitivity to it is because we're sensitive to emotions

00:23:10.770 --> 00:23:11.650
and voices.

00:23:11.890 --> 00:23:17.720
And the reason that we are adapted to paying attention to emotion and voices because it's a

00:23:17.720 --> 00:23:20.920
really good mode of communication at a distance in particular, right?

00:23:21.000 --> 00:23:22.360
You don't need to see my face.

00:23:22.520 --> 00:23:28.040
And so I can give you information at a distance and allow you to make judgments about

00:23:28.040 --> 00:23:31.240
whatever situation is going on based on my emotional signal.

00:23:31.640 --> 00:23:36.990
But I really think cultural evolution has then built on top of that this um super

00:23:37.070 --> 00:23:38.270
stimulus, right?

00:23:38.590 --> 00:23:42.990
So we have a sensitivity to the sound frequencies because we have to pay attention to other

00:23:42.990 --> 00:23:44.350
people communicating with us.

00:23:44.510 --> 00:23:46.670
And then it just has been elaborated.

00:23:46.750 --> 00:23:50.030
I mean, if you have a super stimulus, then you're gonna have a super response.

00:23:50.590 --> 00:23:55.310
I mean, that might not be the most satisfying answer, but I think that it's probably some

00:23:55.310 --> 00:23:59.710
version of that that we have the pre ad we have an adaptation for detecting emotions and

00:23:59.710 --> 00:24:00.030
voices.

00:24:00.270 --> 00:24:02.580
Yeah, does that mean voices?

00:24:03.060 --> 00:24:08.980
This might be a dumb question, but like we also have uh words and linguistic labels for

00:24:08.980 --> 00:24:09.460
emotions.

00:24:09.620 --> 00:24:11.940
Like if I'm mad, I could say I'm pissed off.

00:24:12.020 --> 00:24:13.700
If I'm sad, I could say I'm sad.

00:24:14.020 --> 00:24:18.580
But if we just hear the words I'm pissed off or I'm sad, it doesn't have the same emotional

00:24:18.580 --> 00:24:24.190
effect than if I were to hear a symphony playing sad sounding music

00:24:24.430 --> 00:24:30.030
or like a punk rock band like playing angry sounding music with like a distorted guitar.

00:24:30.190 --> 00:24:34.670
Like there's something different about the sound than about the label, even though so like

00:24:34.750 --> 00:24:36.270
so so what's what's different there?

00:24:36.350 --> 00:24:40.270
Like what shouldn't the label be having the same effect if it has the same function to

00:24:40.270 --> 00:24:41.470
inform me about the emotion?

00:24:41.710 --> 00:24:42.190
Right.

00:24:42.430 --> 00:24:44.740
I mean the label is conceptual, right?

00:24:44.820 --> 00:24:49.940
And so you're informing somebody of a concept that then they can figure out might have

00:24:49.940 --> 00:24:52.020
something to do with your internal feeling.

00:24:52.260 --> 00:24:57.780
But the sound is actually directly reflecting your feeling and it's inducing the

00:24:57.780 --> 00:24:58.420
listener.

00:24:58.660 --> 00:25:04.420
And so that's the ancestral um evolved capacity that's as old as

00:25:04.420 --> 00:25:10.260
mammals, where you by producing a fear scream, you induce fear in

00:25:10.260 --> 00:25:12.230
your audience, Right.

00:25:12.550 --> 00:25:17.430
Or even like a s a high pitched string that kind of sounds reminiscent of a s of a scream

00:25:17.590 --> 00:25:18.470
could have the same effect.

00:25:18.630 --> 00:25:22.950
Like the, you know, the the shrieking thing from Psycho that like the violin sounds a lot

00:25:22.950 --> 00:25:23.350
like a scream.

00:25:23.750 --> 00:25:24.150
Exactly.

00:25:24.230 --> 00:25:24.390
Yeah.

00:25:24.470 --> 00:25:24.710
Yeah.

00:25:24.950 --> 00:25:25.350
That's right.

00:25:25.430 --> 00:25:27.670
It's got high frequencies, they're dissonant.

00:25:27.750 --> 00:25:33.190
Um, and so that dissonance is reflective of a highly aroused vocal

00:25:33.190 --> 00:25:36.470
communication system that's blowing it at blowing out the system a little bit, right?

00:25:36.550 --> 00:25:42.500
If I scream loud enough the amplitude is such that the waveforms can't resolve

00:25:42.500 --> 00:25:43.300
in my system.

00:25:43.460 --> 00:25:47.060
And so it's just like a speaker that can't handle the amplitude.

00:25:47.220 --> 00:25:52.340
And so you get this distortion that it that indicates you're overblowing the system.

00:25:52.500 --> 00:25:55.940
And that is going to be associated with high emotion.

00:25:56.180 --> 00:26:02.010
And then when you perceive that as an animal with um, you the ability to

00:26:02.010 --> 00:26:05.850
map that sound onto some sort of physiological state of the sender.

00:26:06.010 --> 00:26:09.850
And then that arouses you to actually have that state to some extent as well.

00:26:10.250 --> 00:26:14.970
So that's a very reductive stance to rock and roll and overdrive in rock and roll.

00:26:15.290 --> 00:26:15.530
Yeah.

00:26:15.690 --> 00:26:19.850
So Paul, Paul and I have a paper on that exact topic that just came out recently.

00:26:20.170 --> 00:26:20.570
Awesome.

00:26:21.380 --> 00:26:22.740
cultural evolution of distortion.

00:26:22.820 --> 00:26:28.180
And yeah, it's I mean, maybe that's a sad for punk rock or whatever, but I mean that's so

00:26:28.340 --> 00:26:29.780
the so the hmm, interesting.

00:26:29.940 --> 00:26:35.140
So when we f hear a scream, we're responding not only to the pitch but also to the

00:26:35.140 --> 00:26:35.540
distortion.

00:26:36.180 --> 00:26:37.140
David, you're getting an exile.

00:26:37.460 --> 00:26:38.900
I think your mic is still on.

00:26:38.980 --> 00:26:39.620
Yeah.

00:26:40.260 --> 00:26:40.820
For some reason.

00:26:41.140 --> 00:26:42.580
Speaking of speaking of audio, yeah.

00:26:42.900 --> 00:26:44.340
We've never had audio issues before.

00:26:46.580 --> 00:26:47.140
Did it go away?

00:26:47.380 --> 00:26:49.060
Oh, I bring audio issues with me, man.

00:26:49.300 --> 00:26:49.460
Okay.

00:26:49.540 --> 00:26:50.100
That's weird.

00:26:51.070 --> 00:26:53.710
Anyway, um, so are we with audio issues?

00:26:53.950 --> 00:26:54.110
Yeah.

00:26:54.350 --> 00:26:58.830
So when we feel afraid at the sound of a scream, are w what is triggering it is

00:26:58.830 --> 00:27:01.630
simultaneously the pitch and the distortion?

00:27:02.030 --> 00:27:02.270
Right.

00:27:02.350 --> 00:27:04.350
And which is a function of the loudness.

00:27:04.590 --> 00:27:04.750
Right.

00:27:04.990 --> 00:27:08.670
Um, but yeah, so there are not what we call nonlinearities.

00:27:08.830 --> 00:27:13.630
Um, this is kind of what in bioacoustics with the the category of different properties of

00:27:13.790 --> 00:27:16.670
sound that are have that uh association with harshness.

00:27:16.830 --> 00:27:22.440
So there are um subharmonics, there's chaos that's basically broadband

00:27:22.600 --> 00:27:23.560
kind of noise.

00:27:23.960 --> 00:27:26.760
That's the one that I think is most used in music.

00:27:26.920 --> 00:27:32.840
But then you also have um other kinds of um um sidebands and other

00:27:32.840 --> 00:27:38.760
little weird phenomena that are oftentimes just um bands of energy that

00:27:38.760 --> 00:27:44.600
are not in the harmonic series, and so they create they can create these sort of dissonant

00:27:44.600 --> 00:27:45.720
kind of sounds.

00:27:45.880 --> 00:27:51.290
Um but the the deterministic chaos is the one that I think is the closest

00:27:51.610 --> 00:27:54.650
to distortion, which is just broadband noise.

00:27:55.050 --> 00:27:56.810
That's a great band name if nobody's done that.

00:27:57.370 --> 00:27:58.010
Broadband noise.

00:27:58.250 --> 00:27:59.370
No deterministic chaos.

00:27:59.530 --> 00:28:00.250
Deterministic chaos.

00:28:01.610 --> 00:28:02.250
So it's interesting.

00:28:02.330 --> 00:28:06.410
So in terms of you know the the evolution of signaling, what we're talking about is an

00:28:06.410 --> 00:28:10.880
indice where it's not necessarily a costly signal, It's just that there's something about

00:28:11.040 --> 00:28:14.880
intrinsic to the production of the signal that makes it a reliable cue of, say,

00:28:14.960 --> 00:28:19.600
physiological arousal, that just as a byproduct of the body amping itself up and pumping

00:28:19.600 --> 00:28:24.160
itself full of energy and tensing muscles and tensing vocal cords, it will, as a byproduct

00:28:24.160 --> 00:28:28.720
of that, uh raise the pitch and the distortion of what's coming out.

00:28:29.920 --> 00:28:31.330
Basically, Yeah.

00:28:31.490 --> 00:28:35.890
And I mean, I think these things um started as cues and then they get ritualized into

00:28:35.890 --> 00:28:36.370
signals.

00:28:36.530 --> 00:28:42.450
And so then those nonlinearity, nonlinearity uh features, nonlinear features, um,

00:28:42.770 --> 00:28:46.050
would um be exaggerated, right?

00:28:46.210 --> 00:28:46.610
Essentially.

00:28:47.090 --> 00:28:51.730
Yeah, and so Greg, this byproduct idea or this elaboration, cultural evolution, uh,

00:28:51.890 --> 00:28:54.380
evolutionary byproduct idea, does it no?

00:28:54.460 --> 00:28:55.420
I I have two questions.

00:28:55.580 --> 00:28:59.980
One is is it is the idea that there are design features that it's really about con specific

00:28:59.980 --> 00:29:05.580
vocalization, or is it a broader thing where it could be other animals or or definitely

00:29:05.660 --> 00:29:09.900
could be other animals or even the acoustics of of non-living things?

00:29:10.060 --> 00:29:14.780
Or is it really seem like it's about um living vocalizations?

00:29:15.820 --> 00:29:20.930
Yeah, I mean I I think there probably are there's some general principles of the sound

00:29:20.930 --> 00:29:25.890
features of stuff in the world, you know, that you know if you hear a big sound like of

00:29:25.890 --> 00:29:30.450
trees falling down, exactly that's scary and stop because it's derived from your perception

00:29:30.450 --> 00:29:31.730
of animal vocalizations.

00:29:31.970 --> 00:29:32.130
Right.

00:29:32.290 --> 00:29:37.740
So we get all kinds of kinds of auditory C analysis mechanisms that are um

00:29:38.140 --> 00:29:43.900
but the um but the emotionality but the emotionality of the claim is that it

00:29:43.900 --> 00:29:49.660
seems to have design features that are for the living world vocalizations.

00:29:50.300 --> 00:29:52.540
And and for the other animals as well as people, right?

00:29:52.780 --> 00:29:55.900
So if you interact with other animals, you want to know if you're gonna get bit.

00:29:56.140 --> 00:29:56.380
Yes.

00:29:56.690 --> 00:29:58.210
Order Or whatever.

00:29:58.930 --> 00:30:01.890
If other animals are screaming, that's important information for you as well.

00:30:02.210 --> 00:30:03.650
Yeah, seriously no, not my species.

00:30:03.890 --> 00:30:04.290
I don't care.

00:30:04.450 --> 00:30:04.930
I don't care.

00:30:05.650 --> 00:30:05.810
Yeah.

00:30:06.130 --> 00:30:10.450
I mean, um a big dog that is growling at me is going to scare me as much as almost any

00:30:10.450 --> 00:30:11.010
person would.

00:30:11.090 --> 00:30:11.330
Yeah.

00:30:11.490 --> 00:30:11.890
Yeah.

00:30:12.290 --> 00:30:13.490
In the right context.

00:30:13.650 --> 00:30:13.970
Okay.

00:30:14.130 --> 00:30:18.450
Um though I might I I like dogs so much I might be like, oh come on, it's okay.

00:30:18.530 --> 00:30:24.120
You know but um so we definitely there there's uh actually an interesting uh paper

00:30:24.280 --> 00:30:30.040
um showing that people can detect arousal in vocalizations across the all animal

00:30:30.040 --> 00:30:30.360
kingdom.

00:30:30.600 --> 00:30:36.280
So even like insects make you know, and frogs and um

00:30:36.520 --> 00:30:40.920
birds and I mean all all kinds of animals have the same kinds of properties in their in

00:30:40.920 --> 00:30:45.800
their threat displays, which is th there's noise and it's loud, and people can

00:30:46.310 --> 00:30:49.110
arousal even in very distant animals.

00:30:49.190 --> 00:30:51.830
And so probably all animals have that basic ability.

00:30:52.230 --> 00:30:53.030
Is is interesting.

00:30:53.110 --> 00:30:56.790
So are is the thing that we're picking up on loudness and distortion that the louder and

00:30:56.790 --> 00:31:00.230
more distorted it is, the more we infer physiological arousal?

00:31:00.550 --> 00:31:00.950
Yeah.

00:31:01.190 --> 00:31:04.870
I mean, to that's what we that's what has been shown now.

00:31:04.950 --> 00:31:09.910
I mean, I feel like there's another element that I've been struggling to measure, and other

00:31:09.910 --> 00:31:13.510
people uh nobody's really got a good handle on it, but it's vocal effort.

00:31:13.750 --> 00:31:19.640
And so you can kind of tell where in the range of effort

00:31:19.720 --> 00:31:21.720
are is a person at any given moment.

00:31:21.880 --> 00:31:26.200
You know, so if I'm really even if I'm in a distance and the sound is quiet, you can tell

00:31:26.200 --> 00:31:27.560
I'm screaming my brains out.

00:31:27.720 --> 00:31:28.040
Yeah.

00:31:28.680 --> 00:31:31.720
And it's not the absolute loudness that's giving you the information.

00:31:31.800 --> 00:31:35.240
And so there's something about telling where people are in their range.

00:31:35.320 --> 00:31:36.840
And I think it's true for pitch as well.

00:31:37.080 --> 00:31:41.090
So you can tell where people are landing in their their ordinary pitch range.

00:31:41.170 --> 00:31:43.810
And we're usually more towards the low end, right?

00:31:43.890 --> 00:31:47.090
So I can go much more high than I can go down.

00:31:47.570 --> 00:31:47.810
Right.

00:31:47.970 --> 00:31:52.690
So I mean I could go down a little bit, but I mean um I can go I can get up there.

00:31:54.210 --> 00:31:59.010
And um so we can tell where people are are in their range.

00:31:59.090 --> 00:32:02.850
And so you know people like Dave Feimer and others have done research showing that when

00:32:02.940 --> 00:32:07.580
guys try to be lower pitched than they really are, that people can tell, even though

00:32:07.580 --> 00:32:09.340
they're not that much lower than their baseline.

00:32:09.420 --> 00:32:13.820
So if I'm just watching this all the time, it sounds like I'm trying to be low.

00:32:14.460 --> 00:32:20.060
And so I that w with vocal effort and also pitch effort or whatever you might call it, I

00:32:20.060 --> 00:32:22.140
think is an interesting thing that's really not been explored.

00:32:22.220 --> 00:32:26.700
And I think it's hugely important because how people manipulate that in a given context

00:32:26.700 --> 00:32:32.650
could be important for, you know, figuring out what they're intending and Interesting.

00:32:32.890 --> 00:32:38.650
Can we um so it's it the I I really like the example of hearing a scream from far away

00:32:38.730 --> 00:32:42.250
that we still know it was a scream, even though it sounds quiet to the ear.

00:32:42.330 --> 00:32:42.810
It's sort of like a lot of things.

00:32:45.690 --> 00:32:47.370
Yeah, which is really interesting.

00:32:47.450 --> 00:32:49.450
I mean, it's sort of analogous to distance perception.

00:32:49.530 --> 00:32:53.210
So, like something that's far away, like a big bear that's far away, we can still tell it's

00:32:53.210 --> 00:32:58.100
big, it's a big bear, but but the the actual stimulus hitting our retina is quite small,

00:32:58.260 --> 00:33:02.740
which suggests that we have algorithms for telling us that um yeah, this may seem like it's

00:33:02.740 --> 00:33:04.500
small, but it's actually big and it's far away.

00:33:04.660 --> 00:33:08.020
Or s similarly, this might seem like it's a quiet sound, but it's actually a really loud

00:33:08.020 --> 00:33:09.140
sound coming from far away.

00:33:09.220 --> 00:33:11.060
Like it seems like that would be pretty complicated.

00:33:11.540 --> 00:33:11.780
That's cool.

00:33:11.940 --> 00:33:12.260
Yeah, yeah.

00:33:12.340 --> 00:33:16.100
Like a train, you're like, oh, that sounds huge, but it's like not as loud as like the the

00:33:16.100 --> 00:33:19.460
baby that's a little bird talking or whatever, crying.

00:33:19.540 --> 00:33:19.940
Yeah, yeah.

00:33:20.340 --> 00:33:20.580
Yeah.

00:33:20.740 --> 00:33:24.740
I mean, with vision, you often have context that the sound doesn't provide.

00:33:25.220 --> 00:33:30.990
So if you see a bear, I mean, even if you don't know bears are big, you see a

00:33:30.990 --> 00:33:36.030
thing and you go, I can tell how far it is, I can see the trees, and that thing must be

00:33:36.030 --> 00:33:37.630
pretty big if I can see it from here.

00:33:37.790 --> 00:33:39.470
So you the context helps a lot.

00:33:39.710 --> 00:33:44.830
Whereas if you just saw a dot on a screen, a bear in a distance without any context, you

00:33:44.830 --> 00:33:45.550
wouldn't know.

00:33:45.710 --> 00:33:49.550
But if you heard a sound, no matter what, you'd be able to hear something without context.

00:33:49.710 --> 00:33:50.110
Yeah.

00:33:50.440 --> 00:33:52.520
This is this is anecdotal.

00:33:52.680 --> 00:33:58.360
Um, but people talk about how cellos are very close to the uh frequencies of human

00:33:58.360 --> 00:34:01.640
voices, and that those are that is also very emotionally evocative.

00:34:01.880 --> 00:34:02.920
Number one, is that true?

00:34:03.080 --> 00:34:08.760
And number two, if so, is that evidence for that there is a privileging of like con

00:34:09.080 --> 00:34:11.320
specific uh acoustics in the emotional?

00:34:11.480 --> 00:34:15.690
I would argue almost all instruments are in the range of okay vocal production.

00:34:15.770 --> 00:34:21.770
But DiCello is probably um if you rank them, that would probably be up there pretty

00:34:21.770 --> 00:34:22.090
high.

00:34:22.250 --> 00:34:22.570
Okay.

00:34:22.810 --> 00:34:28.170
Um and so yeah, but most instruments are in are in the voice range.

00:34:28.410 --> 00:34:28.650
Okay.

00:34:28.730 --> 00:34:33.610
So if we if we were aliens, if there's aliens and they do music, it's probably gonna be

00:34:33.690 --> 00:34:37.690
their instruments are going to be in their con specific vocalization range.

00:34:38.120 --> 00:34:39.800
Or that's not even sound.

00:34:40.120 --> 00:34:40.360
Sure.

00:34:42.920 --> 00:34:44.360
Non-sound music, yeah.

00:34:44.840 --> 00:34:48.440
That's interesting that there are a lot of logically possible instruments that we could

00:34:48.440 --> 00:34:51.000
have created, but but we didn't because they're not in the range.

00:34:51.240 --> 00:34:55.080
They don't emit sounds that are in the range of human vocal communication, so we don't find

00:34:55.080 --> 00:34:56.680
them interesting or engaging.

00:34:57.240 --> 00:34:57.800
Exactly.

00:34:57.960 --> 00:35:03.250
I mean we can have um instruments that play, you know, where the normal range is ten K or

00:35:03.250 --> 00:35:04.690
something, which is a super high pitch.

00:35:04.850 --> 00:35:05.410
We can hear it.

00:35:05.490 --> 00:35:07.490
I actually I might not be able to hear that anymore.

00:35:07.650 --> 00:35:11.330
But um, you know, normal people people under forty, whatever.

00:35:12.050 --> 00:35:14.130
Normal people under who are under forty.

00:35:14.690 --> 00:35:15.170
Yeah.

00:35:15.650 --> 00:35:18.050
Um there's a lot of people can hear ten K.

00:35:18.290 --> 00:35:21.170
And so you don't have an instrument that plays at ten K.

00:35:22.210 --> 00:35:22.530
Yeah.

00:35:22.740 --> 00:35:23.380
Oh, that makes sense.

00:35:23.620 --> 00:35:24.420
That's that's super powerful.

00:35:24.740 --> 00:35:26.340
We want it, we want it in that sweet spot.

00:35:26.500 --> 00:35:32.420
And so what would be interesting would be just to see um, you know, show I mean

00:35:32.660 --> 00:35:38.100
all instruments have to play within this the scale too, the normal scale, which is A440 for

00:35:38.100 --> 00:35:41.700
Western music, meaning that the note of A is 440 hertz.

00:35:41.940 --> 00:35:44.020
So that's just right there in the voice range.

00:35:44.100 --> 00:35:48.020
But it'd be interesting to see the distributions of different instruments and then compare

00:35:48.020 --> 00:35:48.660
that to voices.

00:35:48.740 --> 00:35:52.590
And I betcha it would be the voice distribution would look like this, and all the other

00:35:52.590 --> 00:35:54.670
instruments would be all like this clustering around it.

00:35:54.990 --> 00:35:55.390
That's okay.

00:35:55.790 --> 00:35:55.950
Okay.

00:35:56.030 --> 00:36:01.470
So so fear is like you got the distortion, the vocal effort, and the high pitch.

00:36:01.710 --> 00:36:06.750
It seems like anger would be similar components because when you're pissed off, you're also

00:36:06.750 --> 00:36:09.790
physiologically aroused and potentially ready to fight.

00:36:09.950 --> 00:36:14.060
But instead of a high pitch, it would be a low pitch because you want to convey

00:36:14.060 --> 00:36:14.780
formidability.

00:36:14.940 --> 00:36:18.060
Presumably, lower pitches uh convey larger size.

00:36:18.300 --> 00:36:23.180
And so anger would be something like distortion, vocal effort, and low pitch.

00:36:23.260 --> 00:36:24.220
Is that the idea there?

00:36:24.300 --> 00:36:24.620
Yeah.

00:36:25.100 --> 00:36:25.340
Yeah.

00:36:25.420 --> 00:36:29.100
Well, so loud and low actually correlates with physical strength.

00:36:29.340 --> 00:36:34.300
So if you can keep your voice loud and low at the same time, just loudness actually is the

00:36:34.300 --> 00:36:35.820
best predictor of physical strength.

00:36:36.370 --> 00:36:40.770
The louder you can produce a sound is actually correlated with your upper body strength.

00:36:40.930 --> 00:36:44.210
So how much can you bench press or how your hip strength or something like that?

00:36:44.370 --> 00:36:47.650
Whereas pitch doesn't correlate with strength, which is an interesting thing that you would

00:36:47.650 --> 00:36:49.330
think is true, but it's not.

00:36:49.570 --> 00:36:52.530
Um David Putz and I differ on this a little bit.

00:36:52.690 --> 00:36:56.530
We debate about that because he's the only one that actually can show is ever shown that

00:36:56.530 --> 00:36:58.450
there's a relationship between pitch and strength.

00:36:59.020 --> 00:37:03.100
I've tried to do it a bunch, and I can't show that, and other people can't show it.

00:37:03.260 --> 00:37:04.140
So we debate that.

00:37:04.300 --> 00:37:09.340
He had, and now we have a paper actually together that I guess will come out someday that

00:37:09.900 --> 00:37:14.140
is a meta-analysis that is showing at least a small effect of pitch.

00:37:14.220 --> 00:37:17.260
I'm not really a big believer in the pitch strength connection.

00:37:17.740 --> 00:37:19.020
But could it be size?

00:37:19.820 --> 00:37:21.100
Could it be size rather than strength?

00:37:21.180 --> 00:37:25.020
Just like if you have bigger vocal cords, there's a bigger body, it's going to be more

00:37:25.020 --> 00:37:26.140
resounding and lower pitch.

00:37:26.380 --> 00:37:30.470
And so it's, and you know, it's size isn't you know a kind of a cue of formidability, but

00:37:30.470 --> 00:37:31.590
it's not strength per se.

00:37:31.830 --> 00:37:32.630
Would you buy that?

00:37:33.030 --> 00:37:36.630
Um the problem is that vocal folds don't really scale with body size.

00:37:36.870 --> 00:37:37.430
Oh, interesting.

00:37:37.510 --> 00:37:37.670
Okay.

00:37:37.830 --> 00:37:43.030
Is it so you can have somebody who is a little person who has just the same size head and

00:37:43.030 --> 00:37:45.430
vocal folds as a person who's six feet tall.

00:37:45.750 --> 00:37:46.310
Oh, really?

00:37:46.710 --> 00:37:47.670
And so where is this?

00:37:47.830 --> 00:37:50.710
I have this idea like from Honey I Shrunk the Kids, when you shrink something, they go,

00:37:50.870 --> 00:37:52.260
Yeah, that that little tight.

00:37:52.500 --> 00:37:52.660
Yeah.

00:37:53.220 --> 00:37:53.940
Is that a myth?

00:37:54.100 --> 00:37:56.820
Like if you if you shrink something, it doesn't necessarily have to be a lot of if you

00:37:56.820 --> 00:38:00.340
shrink the kids with a magic machine, I think you might get a slightly different sound in

00:38:00.340 --> 00:38:00.900
their voices.

00:38:00.980 --> 00:38:01.140
Yeah.

00:38:01.780 --> 00:38:01.860
Okay.

00:38:02.180 --> 00:38:04.980
But because everything is smaller, including the vocal folds.

00:38:05.380 --> 00:38:08.580
Okay, but the but the vocal folds don't necessarily scale with the rest of the body.

00:38:08.660 --> 00:38:08.820
Okay.

00:38:09.140 --> 00:38:09.540
They don't.

00:38:09.620 --> 00:38:14.470
And um actually the vocal folds are unconstrained by other um morphological features in the

00:38:14.470 --> 00:38:15.110
larynx.

00:38:15.190 --> 00:38:21.190
So you can get huge vocal folds that um are differentiated at at puberty from

00:38:21.190 --> 00:38:27.190
testosterone in in small people who are skinny, short, weak people can have like these

00:38:27.190 --> 00:38:31.830
ridiculous vocal folds potentially because of the actions of testosterone.

00:38:31.990 --> 00:38:35.510
Um so but testosterone doesn't really correlate with strength either.

00:38:35.590 --> 00:38:37.670
So higher testosterone doesn't mean you're stronger.

00:38:37.990 --> 00:38:42.940
Is it is it maybe picking up uh uh on just like adult maleness?

00:38:44.060 --> 00:38:44.460
Yes.

00:38:44.620 --> 00:38:45.580
I think that's right.

00:38:45.820 --> 00:38:46.060
Okay.

00:38:46.220 --> 00:38:49.100
Yep. 'Cause that 'cause that would also be a cue of formidability potential.

00:38:51.020 --> 00:38:53.740
In the grand scheme of b the big picture of humans that yes.

00:38:53.900 --> 00:38:54.140
Yeah.

00:38:54.300 --> 00:38:54.620
Exactly.

00:38:54.780 --> 00:38:57.020
And so that's why you would see sex differences.

00:38:57.180 --> 00:39:01.420
Originally people showed that that low pitch was stronger, but then when you control for

00:39:01.420 --> 00:39:02.940
sex, that effect goes away.

00:39:03.390 --> 00:39:08.990
Um and so but the one the one thing I like about the I it occurred to me that you know

00:39:09.230 --> 00:39:12.830
what happens when people get stronger by training?

00:39:13.150 --> 00:39:14.990
What how does their voice change?

00:39:15.230 --> 00:39:18.990
And and it that that I think thought about that for a minute and I thought, okay, that's

00:39:18.990 --> 00:39:20.510
why a pitch is not a thing.

00:39:20.670 --> 00:39:25.240
When when you lift weights for a year and you get a lot stronger, your voice doesn't go

00:39:25.240 --> 00:39:25.960
down, right?

00:39:26.360 --> 00:39:32.200
But you do get um better ability to push air with the muscles that control the airflow in

00:39:32.200 --> 00:39:32.680
your lungs.

00:39:32.840 --> 00:39:36.760
Those are stronger, and so you can push air out more forcefully, so then you could produce

00:39:36.760 --> 00:39:38.120
a louder vocalization.

00:39:38.440 --> 00:39:42.680
But then when you are produce a louder vocalization, it makes the voice pitch go up because

00:39:42.680 --> 00:39:44.600
that's one of the mechanisms of pitch change.

00:39:44.760 --> 00:39:45.320
There are two.

00:39:45.480 --> 00:39:51.010
One is um tightening the vocal core the vocal folds with these um muscles in the

00:39:51.010 --> 00:39:51.490
larynx.

00:39:51.570 --> 00:39:54.050
And the other one is just pushing air through the system harder.

00:39:54.130 --> 00:39:56.450
And then that makes the vocal folds vibrate faster.

00:39:56.770 --> 00:39:59.970
So when you go up in volume, you go up in pitch.

00:40:00.290 --> 00:40:06.210
And so it seems to be your ability to mediate that increase in pitch when you

00:40:06.450 --> 00:40:11.410
increase the subglottal air pressure, it seems to be associated with actual strength.

00:40:11.570 --> 00:40:14.130
So keeping it loud and low at the same time.

00:40:14.290 --> 00:40:16.530
And then that relates to physical strength.

00:40:16.980 --> 00:40:19.060
Greg, what is format dispersion?

00:40:19.140 --> 00:40:21.140
You've told me for years, I still don't know what it is.

00:40:21.220 --> 00:40:23.140
And does that relate to physical size?

00:40:23.220 --> 00:40:26.500
And is that part of what we're talking about without using the technical language?

00:40:26.660 --> 00:40:30.980
Or is is format dispersion something different than size and strength and loudness?

00:40:31.540 --> 00:40:36.980
Um it it it's not I think there's a there's evidence that it's correlated with strength to

00:40:36.980 --> 00:40:40.020
the extent that body size is correlated with strength.

00:40:40.180 --> 00:40:42.740
Um so it's not giving you any other information.

00:40:44.030 --> 00:40:49.870
Formant dispersion essentially is the average distance between formants in the

00:40:49.870 --> 00:40:50.190
signal.

00:40:50.270 --> 00:40:56.110
And the formant is a resonating frequency that is a function of the shape of the vocal

00:40:56.110 --> 00:40:56.510
tract.

00:40:56.590 --> 00:41:01.230
So you have the source, which is the air coming up from your lungs, it goes through your

00:41:01.230 --> 00:41:06.510
vocal folds, the vocal folds vibrate at a certain rate, and then that determines the source

00:41:06.590 --> 00:41:07.950
fundamental frequency.

00:41:08.110 --> 00:41:13.760
But then that sound goes through the vocal tract, through up your throat and out of your

00:41:13.760 --> 00:41:18.240
mouth, and there's these resonating um cavities in your mouth, in your nose, and in the

00:41:18.240 --> 00:41:19.440
back of your throat.

00:41:19.600 --> 00:41:24.800
And so those resonating chambers cause the signal to have certain

00:41:24.800 --> 00:41:27.360
frequencies be um amplified.

00:41:27.600 --> 00:41:28.160
I see.

00:41:28.400 --> 00:41:30.000
So those are the formants.

00:41:30.320 --> 00:41:36.320
And so the because it's just a physical um calculation of cow of seeing how

00:41:36.430 --> 00:41:41.230
distant are those acoustically, and it's inversely related to the length of the tube.

00:41:41.390 --> 00:41:44.750
So formants that are closer together come from a longer tube.

00:41:44.990 --> 00:41:45.150
Okay.

00:41:45.310 --> 00:41:45.550
Yeah.

00:41:45.630 --> 00:41:50.670
And so just to analogize it in a guitar, like an acoustic guitar, you've got all the

00:41:50.670 --> 00:41:55.630
different uh frequencies or pitches being done by the strings, but then the body itself of

00:41:55.630 --> 00:41:57.550
the guitar is a resonating chamber.

00:41:57.790 --> 00:42:02.750
And that's going to in the and analogously, like basically our vocal cords are like the

00:42:02.800 --> 00:42:07.040
strings of a guitar and then our heads in a weird way are like a guitar body.

00:42:07.200 --> 00:42:12.160
Um I mean the guitar is a l is different enough where it's a little weird, but I think a

00:42:12.160 --> 00:42:13.760
better analogy is like a sax.

00:42:14.400 --> 00:42:14.560
Okay.

00:42:14.880 --> 00:42:18.480
Or an you know a wind instrument where this you actually have a source coming from the

00:42:18.480 --> 00:42:18.800
mouth.

00:42:19.040 --> 00:42:19.360
I see, yeah.

00:42:19.600 --> 00:42:22.480
That's the thing you produce, and then it goes through the body of the instrument.

00:42:22.800 --> 00:42:23.120
Yeah, yeah.

00:42:24.080 --> 00:42:24.240
Cool.

00:42:24.320 --> 00:42:24.560
Yeah.

00:42:24.640 --> 00:42:24.880
Yeah.

00:42:25.090 --> 00:42:25.410
Cool.

00:42:25.730 --> 00:42:25.970
Yeah.

00:42:26.370 --> 00:42:29.250
So if you have like a bigger head, are you gonna make difference?

00:42:29.490 --> 00:42:31.410
Is it gonna that like a lower pitch sound?

00:42:31.490 --> 00:42:34.530
Like a more head, yeah.

00:42:35.250 --> 00:42:35.810
I don't know.

00:42:36.210 --> 00:42:37.250
Let's get some data.

00:42:38.130 --> 00:42:39.090
Get some calipers out.

00:42:40.530 --> 00:42:44.610
I mean that it I don't that's a that's I've never really thought about the big head issue,

00:42:44.690 --> 00:42:47.410
but yeah, probably that would have some consequences for sure.

00:42:47.760 --> 00:42:51.120
Um I mean a big head is gonna probably have a longer vocal track for one.

00:42:51.280 --> 00:42:53.360
Um but that's not quite what you're getting at.

00:42:53.440 --> 00:42:58.880
And uh um it's just like a more hollow head, like more space in there to resound, you know.

00:43:01.120 --> 00:43:03.600
Yeah, the format dispersion intelligence hypothesis.

00:43:04.000 --> 00:43:07.440
Oh my god, it's yeah, a small brain, big skull.

00:43:08.720 --> 00:43:11.310
Um Yeah.

00:43:11.630 --> 00:43:14.670
Well, we're we're we're 25 episodes in or something like that.

00:43:14.830 --> 00:43:16.670
And now we're getting into bad episode.

00:43:16.830 --> 00:43:16.990
Okay.

00:43:18.270 --> 00:43:19.230
This is the line.

00:43:19.310 --> 00:43:20.670
All right, let's do it.

00:43:22.270 --> 00:43:23.470
Yeah, nice.

00:43:23.710 --> 00:43:27.230
Um, well, so you can see how format dispersion is related to body size, right?

00:43:27.310 --> 00:43:30.670
It gets it gives you a pretty much an index in the length of the vocal tract, which is

00:43:30.670 --> 00:43:31.790
correlated to body size.

00:43:31.870 --> 00:43:33.630
And this is true across mammals.

00:43:33.710 --> 00:43:38.140
Um, I mean it's not a huge effect, but it's much bigger effect than pitch for in terms of

00:43:38.140 --> 00:43:39.820
relating to strength and body size.

00:43:39.980 --> 00:43:42.700
Well body size even so low pitch isn't really related to body.

00:43:43.020 --> 00:43:46.940
So what what is it sound what is format dispersion sound like if not pitch?

00:43:48.140 --> 00:43:54.140
Um so if if you heard the pitch of a sound, it would you know I mean I could if

00:43:54.140 --> 00:43:58.530
I was prepared I'd be like the timbre of the sound It's like the difference between baby

00:43:58.530 --> 00:44:01.890
cooing because we we did the Greg and I did the stimulant and I had the same question.

00:44:02.050 --> 00:44:07.170
And for me, at least, the intuition is you can imagine a baby like making a low pitch

00:44:07.250 --> 00:44:10.930
sound, but it kind of sounds like cooing, you know, like kind of like if you just imagine

00:44:11.010 --> 00:44:16.530
like a baby kind of making a noise vocalizing, but it's kind of a deep for a baby

00:44:16.530 --> 00:44:22.530
vocalizing, and then imagine like a whale producing, and you kind of get that.

00:44:22.760 --> 00:44:26.600
I don't even know how to articulate it verbally, but it at least it kind of captures like

00:44:26.680 --> 00:44:30.280
imagine what a baby sounds like and what a whale sounds like, but they're both making low

00:44:30.360 --> 00:44:31.480
pitched noises.

00:44:33.080 --> 00:44:35.160
Yeah, though I mean, yeah, no, that's interesting.

00:44:35.240 --> 00:44:38.760
Uh is it like a kind of like um reverbiness?

00:44:39.160 --> 00:44:44.280
I think the the formant I mean, if you're asking what the formants are what they sound like

00:44:44.280 --> 00:44:50.210
relative to the source, think of just the source being something like I mean it's just a

00:44:50.210 --> 00:44:52.050
noise like that, maybe a little cleaner than that.

00:44:52.850 --> 00:44:55.330
It's it's formats being valves.

00:44:55.570 --> 00:45:00.210
So it's like AIO is I'm adjusting the the format.

00:45:00.530 --> 00:45:05.570
Um the I'm adjusting the formats because I'm adjusting the resonating chambers in my mouth

00:45:05.570 --> 00:45:07.730
by moving my tongue around and stuff like that.

00:45:07.890 --> 00:45:12.530
So the formats are like the vowel sounds that think about vowels changing across the

00:45:12.530 --> 00:45:13.410
different vowels.

00:45:13.570 --> 00:45:18.740
That's the action of formats shifting, especially the first Right.

00:45:18.900 --> 00:45:24.180
And then the underlying pitch of the voice, which is just the source sound of, you know,

00:45:24.260 --> 00:45:26.740
think of it as it's not really any vowel.

00:45:26.900 --> 00:45:31.380
I mean, it's probably closest to ah, which is open, like uh the open.

00:45:32.100 --> 00:45:32.260
Right.

00:45:32.340 --> 00:45:37.620
So if I'm singing the same note and I go ah ooh e, like the the change in the vowel

00:45:37.620 --> 00:45:40.980
represents the ch the shift in the in the f format format.

00:45:41.610 --> 00:45:41.930
Right.

00:45:42.090 --> 00:45:42.490
Yep.

00:45:42.730 --> 00:45:43.210
Exactly.

00:45:43.290 --> 00:45:47.770
And so and then, you know, you could you probably inadvertently raise the pitch a little

00:45:47.770 --> 00:45:51.850
bit when you are like there's some sometimes you could there's a perceptual interaction,

00:45:51.930 --> 00:45:55.850
but I mean in theory these things are independent dimensions, the source and the filter.

00:45:55.930 --> 00:46:00.010
Um though E would sound higher pitched, maybe people might make that judgment.

00:46:00.330 --> 00:46:00.650
Okay.

00:46:01.460 --> 00:46:01.860
Cool.

00:46:02.020 --> 00:46:02.340
All right.

00:46:02.420 --> 00:46:04.180
So we got we got fear, we got anger.

00:46:04.340 --> 00:46:07.220
What's going on with sad, lugubrious strings?

00:46:07.540 --> 00:46:07.700
Yeah.

00:46:07.860 --> 00:46:08.500
Or reflection.

00:46:08.580 --> 00:46:11.140
Or why does it make ref uh must reflect too?

00:46:11.300 --> 00:46:14.980
Because I think there's also like it's not just like seeing a snake, it also feels like it

00:46:14.980 --> 00:46:18.260
controls our thoughts to put it too s crudely.

00:46:18.580 --> 00:46:19.460
Um right?

00:46:19.540 --> 00:46:21.300
Like why do I feel reflective?

00:46:21.460 --> 00:46:23.220
Um that's interesting.

00:46:23.390 --> 00:46:29.070
Um, I mean, when you're sadness, you're trying to solicit um assistance at some level.

00:46:29.150 --> 00:46:33.710
If you take like the Ed Hagen-like approach to communicating sadness, which I like, um,

00:46:33.870 --> 00:46:39.870
there's something about you're expressing a a state that is not an immediate

00:46:39.870 --> 00:46:42.830
state about a moment where you need to act quick.

00:46:42.990 --> 00:46:48.180
It's more of a long it's a something about a long-term situation, and you're soliciting in

00:46:48.180 --> 00:46:54.180
some way, or you need whatever you you um need at least um to check on who's your

00:46:54.180 --> 00:46:58.180
ally and and I mean there it depends on the context.

00:46:58.260 --> 00:47:03.780
But so I think that it I like the idea of it in invoking inflection or a reflection because

00:47:03.780 --> 00:47:07.460
you have to think about like what is happening, what is being communicated.

00:47:07.540 --> 00:47:12.340
It's not as simple as something like I'm going to attack you or I'm scared of a thing.

00:47:12.500 --> 00:47:16.790
It's it seems like it's about other more complex long-term phenomena.

00:47:17.110 --> 00:47:17.270
Yeah.

00:47:17.910 --> 00:47:21.590
And I don't want to be um like I we're we're all on this journey together.

00:47:21.670 --> 00:47:23.670
Like I don't want to do a gotcha thing here.

00:47:23.750 --> 00:47:25.350
Um and I'm not doing it in this spirit.

00:47:25.510 --> 00:47:31.350
But like one thing that I I I I'm puzzled by is why there's fluidity with first and third

00:47:31.350 --> 00:47:31.670
person.

00:47:32.310 --> 00:47:38.070
Like, um, like why is it that if I hear sad music, I don't just react as if I'm hearing

00:47:38.070 --> 00:47:40.260
some something else that's sad.

00:47:40.980 --> 00:47:43.220
Why is it that it feels like it's in my head?

00:47:43.300 --> 00:47:44.980
You know, like I think that's the other part of it.

00:47:45.060 --> 00:47:50.660
Is like, why is it almost equally likely that it's almost as if

00:47:50.820 --> 00:47:54.980
this it's coming from me, which which feels right?

00:47:55.060 --> 00:47:58.500
Like, and I don't know if that's again or at the very least, like someone you care about.

00:47:58.580 --> 00:48:02.310
Like you could imagine like your enemy being sad and you not feeling sad, right?

00:48:02.470 --> 00:48:05.110
In fact, you might even feel happy because your enemy is sad, right?

00:48:05.350 --> 00:48:06.950
Yeah, but I'm like a whimpering animal.

00:48:07.030 --> 00:48:09.110
It's not like I'm like, oh, where's the whimpering animal?

00:48:09.270 --> 00:48:10.630
It's like who's sad around me?

00:48:10.710 --> 00:48:15.590
It's like I hear sad music and all of a sudden I'm like starting to reflect on the things

00:48:15.590 --> 00:48:18.150
that either were sad for me or might be sad.

00:48:18.390 --> 00:48:19.510
Why is it in me?

00:48:19.670 --> 00:48:20.230
Yeah.

00:48:20.550 --> 00:48:21.910
I mean that that's a good question.

00:48:21.990 --> 00:48:27.030
I mean uh the first past answer is that it's it's also because of affect induction which is

00:48:27.030 --> 00:48:30.920
what I had mentioned before like the fear scream kind of makes scared, right?

00:48:31.000 --> 00:48:33.480
And because that causes an adaptive reaction.

00:48:33.720 --> 00:48:39.080
And so I think all of emotions are communicated, could have that kind of inducing effect,

00:48:39.240 --> 00:48:39.560
right?

00:48:39.640 --> 00:48:41.480
I mean, it's contagious.

00:48:41.720 --> 00:48:45.960
Um, and so sadness can be contagious, um, as can happiness.

00:48:46.200 --> 00:48:52.120
Laughter is very contagious, um, and um obviously fear and anger are

00:48:52.120 --> 00:48:53.080
contagious as well.

00:48:53.160 --> 00:48:59.110
And so I think what you're describing could be some kind of conceptualization uh or

00:48:59.270 --> 00:49:04.790
conceptual um you know translation of this inducing sadness

00:49:04.790 --> 00:49:08.790
effect that would be my first past answer i have to think about that for a little bit

00:49:08.870 --> 00:49:12.470
though yeah i would sadness is an interesting yeah yeah i would i would also bet that

00:49:12.470 --> 00:49:17.350
sadness would be lower in vocal effort lower in distortion lower in volume definitely

00:49:17.830 --> 00:49:21.670
because in that case you're you what you're signaling is that you're not physiologically or

00:49:21.830 --> 00:49:24.960
that you're depleted down regulation in terms of Yeah, down regulation.

00:49:25.040 --> 00:49:28.800
You're you're defeated, you're tired, you're you're immobilized, you know, you're

00:49:28.800 --> 00:49:30.160
incapacitated, whatever.

00:49:30.320 --> 00:49:34.640
Like when you're in a sad state, your energy reserves are much lower, and and that is what

00:49:34.640 --> 00:49:35.920
is being signaled, presumably.

00:49:36.080 --> 00:49:39.600
Like if you think if I think of like a lot of sad music, it's often very quiet.

00:49:39.840 --> 00:49:41.600
Uh uh Yeah.

00:49:42.080 --> 00:49:42.960
Yeah, exactly.

00:49:43.040 --> 00:49:48.530
And there's I mean, the the the melodic contours are very much like sad vocals, you know,

00:49:48.770 --> 00:49:49.810
sad speech.

00:49:49.970 --> 00:49:55.650
Um, and you were talking earlier about um, you know, w why certain

00:49:55.810 --> 00:49:58.370
um tonality or um how did you put it?

00:49:58.530 --> 00:50:04.530
I mean, there's tonality and scales and other kinds of musical features are associated with

00:50:04.530 --> 00:50:07.170
the emotions in the speech.

00:50:07.810 --> 00:50:13.640
So the when we're speaking in more sad tones relative to anger, for example,

00:50:13.800 --> 00:50:19.320
um, there are different kinds of frequencies that are highlighted that are associated with

00:50:19.320 --> 00:50:21.000
those kinds of melodies.

00:50:21.320 --> 00:50:25.640
And so like all these different kinds of musical properties are you can show have been

00:50:25.640 --> 00:50:28.920
derived directly from the vocal expression of them.

00:50:29.610 --> 00:50:35.210
And so that Dan regulation, the dis descending contours and the melodies.

00:50:35.290 --> 00:50:39.210
And yeah, you don't have evidence of high arousal, so there's not distortion, you know.

00:50:39.290 --> 00:50:41.210
I mean a sad song with distortion ends up.

00:50:41.450 --> 00:50:45.210
I mean, I guess you could have sad punk rock or whatever, and it's that has distortion in

00:50:45.210 --> 00:50:45.610
it some level.

00:50:46.010 --> 00:50:47.770
Kind of sounds like laxymoron, yeah.

00:50:48.490 --> 00:50:49.050
Yeah.

00:50:49.450 --> 00:50:54.560
Like I can't I it's hard I'm having a hard time imagining like a sad electric guitar, you

00:50:54.560 --> 00:50:54.880
know?

00:50:56.000 --> 00:51:00.240
Or a a sad loud like distorted, yeah.

00:51:00.320 --> 00:51:00.400
Yeah.

00:51:00.640 --> 00:51:01.280
Then it's gonna sad.

00:51:01.520 --> 00:51:05.280
Yeah, there's it's definitely I mean you'd have to be pretty good to like get that across

00:51:05.600 --> 00:51:07.840
having th those kinds of characteristics.

00:51:08.160 --> 00:51:12.570
But it's also gonna be relative to other music that the person is producing, and then you

00:51:12.570 --> 00:51:18.570
hear the relative emotion, even though out of context, maybe it wouldn't be taken that way,

00:51:18.810 --> 00:51:19.530
potentially.

00:51:20.570 --> 00:51:20.810
Yeah.

00:51:20.970 --> 00:51:25.450
There's there's another like really mysterious part of it where there's also like a sense

00:51:25.450 --> 00:51:27.850
of triumphant that you hear in music.

00:51:27.930 --> 00:51:32.570
Like when it shifts from minor to major, like it's like, ah, you've achieved something

00:51:32.570 --> 00:51:33.370
momentous.

00:51:33.610 --> 00:51:38.560
Like that is is really mysterious to me because that doesn't seem to relate to vocal effort

00:51:38.560 --> 00:51:42.400
or distortion or anything like that, which would be a valid cue of physiological arousal.

00:51:42.480 --> 00:51:46.960
That's something that's just purely in the tones and the change from uh you know minor to

00:51:46.960 --> 00:51:47.280
major.

00:51:47.360 --> 00:51:49.680
And that I have no idea what's going on.

00:51:50.400 --> 00:51:51.920
Do you have any idea what's going on there?

00:51:52.320 --> 00:51:58.320
I mean I I I think that um b positive negative um and minor major

00:51:58.560 --> 00:52:02.290
have to do with positive and negative emotions in the voice again.

00:52:02.370 --> 00:52:07.330
And so when you're ex extremely happy, then you don't have those nonlinear features to the

00:52:07.330 --> 00:52:10.850
same extent than when you are, I mean, you do a little bit though.

00:52:10.930 --> 00:52:15.730
So people have done things like look at when people win in sports and other things.

00:52:15.810 --> 00:52:17.330
And they're definitely high arousal.

00:52:17.490 --> 00:52:22.930
So some of those nonlinear features are there just by by as a byproduct of the production.

00:52:23.250 --> 00:52:28.370
But you don't emphasize them in the same way as if you're growling at somebody or you are

00:52:28.380 --> 00:52:30.460
threatening them or whatever.

00:52:31.740 --> 00:52:36.220
So the triumphant positive notes are going to be cleaner, right?

00:52:36.540 --> 00:52:41.740
And then more negative songs or

00:52:42.380 --> 00:52:48.060
melodies are going to be more distorted and have different kinds of melodic

00:52:48.060 --> 00:52:50.540
structure that's associated with that negative emotion.

00:52:50.780 --> 00:52:54.010
But I guess I would want to see the evidence for the triumphant thing.

00:52:54.090 --> 00:52:55.690
I I would assume it'd be a little more clean.

00:52:55.770 --> 00:53:01.770
You know, like horns, whatever, and um it's you know, like growling.

00:53:02.250 --> 00:53:02.490
Yeah.

00:53:02.810 --> 00:53:08.730
Is is there I mean, um, you know, if you there's certain traditions that are positive where

00:53:08.730 --> 00:53:13.290
it's a positive triumphant kind of thing and it's big sounds with like noisy features.

00:53:13.610 --> 00:53:17.460
So here's a here's a different, perhaps weirder question.

00:53:17.700 --> 00:53:23.540
If you rewound the tape of evolution um and you replayed it, could it

00:53:23.540 --> 00:53:29.540
be possible that um minor actually would connote happiness and major would connote

00:53:29.540 --> 00:53:29.940
sadness?

00:53:30.020 --> 00:53:31.540
Like, is this just an evolutionary accident?

00:53:31.700 --> 00:53:36.340
Is it arbitrary that these emotions are mapped to these melodies and sound structures?

00:53:36.500 --> 00:53:41.140
Or is there something inherent to these sound structures that connects to a particular

00:53:41.140 --> 00:53:41.860
emotion?

00:53:42.260 --> 00:53:42.500
Right.

00:53:42.710 --> 00:53:43.670
I mean, that's a good question.

00:53:43.830 --> 00:53:48.870
And I think the evidence that I like, this is Dale Perves and Dan Bowling and others have

00:53:48.870 --> 00:53:53.990
done a lot of work looking at how musical features are related to vocal communication.

00:53:54.150 --> 00:53:58.230
And it's a very voice centric approach to music that I like.

00:53:58.550 --> 00:54:04.230
And their argument is that when you are producing emotions

00:54:04.230 --> 00:54:08.790
vocally like sadness or anger, that there um are

00:54:10.280 --> 00:54:16.200
the the relative frequencies end up mapping better to something like

00:54:16.200 --> 00:54:22.120
a dissonant musical key than when you have a positive uh vocal emotions,

00:54:22.200 --> 00:54:26.920
then those energy, those frequencies map better to um a major key.

00:54:27.080 --> 00:54:31.320
Um, so the idea there is that major and minor is universal to some extent.

00:54:31.480 --> 00:54:34.040
And I think there's evidence that not everybody hears it that way.

00:54:34.120 --> 00:54:39.410
And there could be cultural evolution of musical traditions that make that small effect not

00:54:39.410 --> 00:54:41.250
so much an effect anymore.

00:54:41.570 --> 00:54:47.410
But my thinking of it is that it is derived from how

00:54:47.410 --> 00:54:49.890
these kinds of emotions manifest in the voice.

00:54:50.130 --> 00:54:56.050
And so there's really interesting work showing that if you normalize little

00:54:56.210 --> 00:55:01.650
tiny snippets of a voice, the acoustics, and then you convert the

00:55:01.650 --> 00:55:07.620
frequencies to just um uh octaves, So you have um ratios of frequencies

00:55:07.620 --> 00:55:12.500
that start at where the maximum amplitude is in the in the signal, which is used at pretty

00:55:12.500 --> 00:55:13.380
low frequency.

00:55:13.540 --> 00:55:18.340
And then instead of having the x-axis be frequency, you have it just be octaves.

00:55:18.580 --> 00:55:20.820
And then you take an octave slice.

00:55:21.060 --> 00:55:25.860
Um, and if you look at that octave slice, the spectral structure, there are bumps that

00:55:25.860 --> 00:55:28.340
correspond really well to the chromatic scale.

00:55:28.570 --> 00:55:33.370
And so you can, you know, you octave is highlighted, the unison pitch is highlighted, and

00:55:33.370 --> 00:55:35.610
then a fifth is get there's a bump.

00:55:35.690 --> 00:55:39.130
And then all the little notes, there's a really cool figure in this one paper where they

00:55:39.130 --> 00:55:44.410
just show exactly how the little bumps in the spectral structure of a speech snippet is

00:55:44.410 --> 00:55:47.370
related to the chromatic scale and the Western musical scale.

00:55:47.450 --> 00:55:50.330
And then they can also do that same kind of thing for musical scales.

00:55:50.650 --> 00:55:55.210
So it's not just the intervals but the scales themselves actually there's there's certain

00:55:55.210 --> 00:55:59.940
scales um happen more often in cultural evolution than others, right?

00:56:00.020 --> 00:56:05.620
They cluster, and the way they cluster is related to how the spectral features of speech

00:56:05.620 --> 00:56:07.700
map onto the chromatic scale.

00:56:07.860 --> 00:56:13.700
So to me, that's really that that we are really it's a cultural elaboration of this basic

00:56:13.700 --> 00:56:18.500
perceptual sensitivity to voices, and that we can cultural evolution can build so much

00:56:18.500 --> 00:56:24.340
richness and elaboration on just that basic um predisposition we have perceptually to

00:56:24.660 --> 00:56:29.310
that we like to listen to voices, or we have voice selective adaptations, right?

00:56:29.390 --> 00:56:30.590
I mean, that's that's super clear.

00:56:30.830 --> 00:56:31.550
Yeah, that's interesting.

00:56:31.790 --> 00:56:37.310
I mean especially the part where like sp speech corresponds more to it the chromatic scale

00:56:37.310 --> 00:56:40.830
than you know, say a major scale or a pentatonic scale or anything like that.

00:56:40.990 --> 00:56:46.830
Um like jazz to me comes to mind as like really making use of the chromatic

00:56:46.830 --> 00:56:51.950
scale, making use of just like lots of tiny changes in notes and just ascending

00:56:51.950 --> 00:56:54.430
chromatically, you know, half step by half step.

00:56:54.590 --> 00:56:58.480
Um and to me, intuitively, jazz sounds more conversational.

00:56:58.560 --> 00:56:59.920
It sounds like the instruments are talking.

00:57:00.080 --> 00:57:03.600
It sounds like a person, just you know, like if you hear people jamming, it's just like the

00:57:03.600 --> 00:57:07.360
instruments are talking to each other, and it's just sounds like you don't know, more

00:57:09.760 --> 00:57:12.160
rich metaphor that jazz musicians like a lot.

00:57:12.240 --> 00:57:14.560
And they and I've I teach it that way in this one.

00:57:14.640 --> 00:57:16.400
I used to teach a class called group communication.

00:57:16.560 --> 00:57:21.520
I would draw explicit connections between conversational behavior and musical interaction.

00:57:21.680 --> 00:57:26.030
And yeah, they're you're taking turns and I mean there's all these elements they map really

00:57:26.030 --> 00:57:26.350
well.

00:57:26.990 --> 00:57:27.390
Yeah.

00:57:27.550 --> 00:57:33.470
So um I think that we we're we're sensitive to voices and that that's the kinds of sounds

00:57:33.470 --> 00:57:33.950
we want to hear.

00:57:34.030 --> 00:57:38.110
We've we've culturally evolved instruments that are in the same range and that the way

00:57:38.190 --> 00:57:42.830
emotions are expressed are gonna if you would analyze it correctly you're gonna see map

00:57:42.830 --> 00:57:45.070
onto the way we express those things vocally.

00:57:45.230 --> 00:57:50.660
And so that's why we major minor consonants dissonance um descending versus ascending

00:57:50.740 --> 00:57:53.860
melodic contours, they just all seem to map.

00:57:54.100 --> 00:57:58.660
But then you have all these other features that cultural evolution uses to build musical

00:57:58.660 --> 00:58:04.260
traditions that then it's just it's crazy when you think about the music varia variability

00:58:04.260 --> 00:58:09.860
out there and that what basic adaptive problems and it that originally music solved is not

00:58:10.180 --> 00:58:13.940
going to predict all the elaboration we see that culture has produced.

00:58:14.650 --> 00:58:20.090
And you just mentioned music being and solving an or was it a cultural adaptation?

00:58:20.170 --> 00:58:20.970
I mean, what is your take?

00:58:21.050 --> 00:58:26.890
So I mean there's one idea that you know cultural selection or biological selection

00:58:27.210 --> 00:58:32.890
uh made music an adaptation, you know, the big A, the, the A-word, the onerous concept.

00:58:33.050 --> 00:58:34.010
Um what is your take?

00:58:34.090 --> 00:58:39.620
So I mean there's one idea that you know music evolved for social bonding, for example.

00:58:39.860 --> 00:58:43.860
Um, I think what's your take on that?

00:58:44.100 --> 00:58:44.980
Um yeah.

00:58:45.380 --> 00:58:51.060
So there's there's a really good um BBS um thing that so um Max

00:58:51.220 --> 00:58:57.060
Krasnow and Sam and Ed Hagen and I um wrote a BBS Target article about

00:58:57.060 --> 00:58:57.300
this.

00:58:57.460 --> 00:59:02.170
And then it was, I think, the only time this has happened in BBS where we submitted two

00:59:02.170 --> 00:59:03.130
target articles.

00:59:03.290 --> 00:59:08.890
So we did one and then another group of folks, um, Pat Savage and um

00:59:09.210 --> 00:59:13.610
Tecumseh Fitch and a bunch of folks um produced one that was about social bonding.

00:59:13.770 --> 00:59:18.650
Cause we did this conference in in Paris or outside of Paris many years ago where we were

00:59:18.650 --> 00:59:23.050
talking about the origins of music, and then two camps kind of happened us four, and then

00:59:23.050 --> 00:59:25.290
everybody else that likes social bonding.

00:59:25.520 --> 00:59:27.280
So we decided to do this BBS thing.

00:59:27.360 --> 00:59:29.920
They took both papers and then they solicited commentaries for both.

00:59:30.000 --> 00:59:31.760
And it's this huge, it's like a book.

00:59:32.240 --> 00:59:35.920
And um there we go after the social bonding idea.

00:59:36.080 --> 00:59:40.560
Um, and I don't want to pick on that here too much, but that the the idea is that um

00:59:40.880 --> 00:59:46.160
because they you can show effects of people playing music together, cooperate more and like

00:59:46.160 --> 00:59:50.880
each other more, and there's all these sort of proximate effects, um, they argue that it's

00:59:50.880 --> 00:59:51.520
for that.

00:59:51.910 --> 00:59:56.150
And we make the case it's actually not well suited for that, particularly.

00:59:56.230 --> 00:59:59.670
Just because it has that effect doesn't mean that the adaptation is to cause social

00:59:59.990 --> 01:00:00.470
bonding.

01:00:00.550 --> 01:00:05.670
And we make the case rather that it's better to think of it as credible signaling.

01:00:05.910 --> 01:00:11.430
So when you have a practiced um effect of a

01:00:11.430 --> 01:00:17.420
coalition producing a coherent, obviously choreographed um show that

01:00:17.420 --> 01:00:18.140
it takes practice.

01:00:18.220 --> 01:00:20.540
It's an honest signal of time spent with one another.

01:00:20.620 --> 01:00:24.940
Um it gives an honest indicator of how well you coordinate your actions.

01:00:25.260 --> 01:00:31.020
And and so then you and the the idea is that that could have

01:00:31.020 --> 01:00:37.020
caused selection for the ability to entrain physically with another person, which is a

01:00:37.020 --> 01:00:39.340
human specific um adaptation.

01:00:39.500 --> 01:00:45.350
No other, almost no other mammal can really do it well, um, where they can actually

01:00:45.510 --> 01:00:51.110
readily learn how to entrain to another person with their motor abilities.

01:00:51.350 --> 01:00:53.350
Um that's something that only people can do.

01:00:53.510 --> 01:00:57.110
And so if you're gonna look for a musical adaptation, I think you gotta look for why would

01:00:57.110 --> 01:01:00.070
people be coordinating their muscle movement like that.

01:01:00.640 --> 01:01:02.960
um in a super um fine-grained way.

01:01:03.040 --> 01:01:07.360
So that's the argument that Ed and I had developed a long time ago.

01:01:07.680 --> 01:01:13.600
It's my first paper ever published, arguing that it's you're giving

01:01:13.600 --> 01:01:18.960
an honest signal of coalition quality by coordinating your and music is a vehicle to do

01:01:18.960 --> 01:01:19.280
that.

01:01:20.160 --> 01:01:25.920
And then Sam and Max it's sort of developed this idea about how

01:01:26.160 --> 01:01:32.050
infant caregiver communication caregivers signaling to infants that they um are

01:01:32.050 --> 01:01:37.490
actually honestly have their attention and that that promoted this uh motivation to produce

01:01:37.490 --> 01:01:42.210
melodies that that uh got the infant's attention and that it's hard to do that while doing

01:01:42.210 --> 01:01:42.610
other things.

01:01:42.690 --> 01:01:46.130
If you're singing to your kid you could be doing some other things but you definitely have

01:01:46.210 --> 01:01:48.690
you there you have they have your attention.

01:01:48.930 --> 01:01:54.560
And so it it that's an emotional communication mechanism that is sort of

01:01:54.560 --> 01:01:58.800
exploiting the predisposition for detecting emotions and voices, right?

01:01:58.960 --> 01:02:04.800
Some lullabies sound like friendly, affiliative kinds of vocalizations, but they're just

01:02:04.960 --> 01:02:08.160
um, you know, ritualized, highly ritualized.

01:02:08.320 --> 01:02:13.600
And so their argument was that that could be the origins of melody, is that the initial

01:02:13.600 --> 01:02:17.680
singing between um kit for k caregivers singing to their offspring.

01:02:18.000 --> 01:02:23.050
So we just make the case that these are two examples of credible signaling that could form

01:02:23.050 --> 01:02:26.810
the basis for some adaptations that are actually for music.

01:02:26.970 --> 01:02:32.250
But then most other things that we incorporate into the music faculty, as they say, are

01:02:32.490 --> 01:02:35.050
byproducts of other things that we have.

01:02:35.210 --> 01:02:40.250
So that's my view view: that there are a couple adaptations plus cheesecake, right?

01:02:40.570 --> 01:02:45.620
And Pinker's idea of violatory cheesecake is sort of the hypothesis to beat, and that it's

01:02:45.620 --> 01:02:49.940
culturally cobbled together all these mechanisms that really have solved all these other

01:02:49.940 --> 01:02:50.500
problems.

01:02:50.660 --> 01:02:54.260
Um, and language is also possibly a lot like that.

01:02:54.420 --> 01:02:58.100
Like, what is what specific adaptation is there for language?

01:02:58.180 --> 01:02:59.540
That's a big question.

01:02:59.700 --> 01:03:03.540
Uh, I mean, now it's getting reduced down to recursion, is the only thing.

01:03:03.700 --> 01:03:05.620
Um, even that's questionable.

01:03:05.700 --> 01:03:11.190
Um, language is a whole nother can't affect so I think there's a similar there's an

01:03:11.190 --> 01:03:16.310
analogous sort of argument argument there that you we're cobbling together all these

01:03:16.310 --> 01:03:21.990
abilities and culturally, but then there could be these core adaptive problems that

01:03:22.230 --> 01:03:26.470
that uh should be considered proper in the domain of music and not actual.

01:03:26.870 --> 01:03:32.070
And those sound like they're getting into uh time because you know, music is because the

01:03:32.070 --> 01:03:36.220
emotionality thing, there was no I mean, there was a little bit of a time commo component.

01:03:36.380 --> 01:03:40.380
But there is, of course, this whole thing of like rhythm or prosody.

01:03:40.940 --> 01:03:45.980
Um, and just in in the, you know, somebody can't be a good musician if they don't have good

01:03:45.980 --> 01:03:49.740
rhythm or they don't have a good sense of timing, even if they're if they can if they're

01:03:49.740 --> 01:03:53.980
like, yeah, this pitch is gonna make you feel like this, if they can't do that over time,

01:03:54.140 --> 01:03:55.900
they're not gonna be a good musician.

01:03:56.140 --> 01:04:01.580
Um so, but but it sounds like dancing or like coordinating movement is an example.

01:04:01.900 --> 01:04:07.490
So is that correct that maybe these other uh proper domain-y things of

01:04:08.130 --> 01:04:13.650
coordination and possibly this byproduct of melody um

01:04:13.890 --> 01:04:16.130
is part of where timing comes in.

01:04:16.610 --> 01:04:22.450
Yeah, well, I mean the the timing is we humans are really even the people with the worst

01:04:22.450 --> 01:04:27.410
timing are um a pretty good relatively species, okay.

01:04:27.490 --> 01:04:27.890
Yeah, yeah.

01:04:28.220 --> 01:04:28.700
Definitely.

01:04:28.940 --> 01:04:31.980
You cannot train a dog to play a drum, no matter how many treats.

01:04:32.300 --> 01:04:33.580
Believe me, I've tried cases.

01:04:33.900 --> 01:04:34.540
Yeah.

01:04:35.420 --> 01:04:35.740
Right.

01:04:35.820 --> 01:04:41.420
I mean, people have evidence for bonobos, um, which I mean, you don't see a bonobo in the

01:04:41.420 --> 01:04:46.700
wild doing anything remotely like that, but you can get them to kind of, you know, keep a

01:04:46.700 --> 01:04:48.540
beat along with the thing that they hear.

01:04:48.620 --> 01:04:52.460
That I don't think they've ever shown a bonobo being able to keep a beat without something

01:04:52.460 --> 01:04:53.340
to listen to.

01:04:53.740 --> 01:04:59.590
Um, there's evidence that sea lions can bob their head to a beat you know but it takes a

01:04:59.590 --> 01:05:05.350
lot of training and um and then birds can do it so there's the famous snowball the

01:05:05.350 --> 01:05:09.990
cockatoo who dances and cockatoos are well known to dance and so all these animals are

01:05:09.990 --> 01:05:14.390
vocal sea lions are the exception which is an interesting one but um all these animals are

01:05:14.390 --> 01:05:20.390
vocal learners so um any animal that needs to pay attention to the periodic

01:05:20.580 --> 01:05:26.420
features of some other vocalizer, and then being able to make the same sound,

01:05:26.580 --> 01:05:28.020
and you have to learn how to do it.

01:05:28.100 --> 01:05:30.260
It's not something that's just genetically endowed.

01:05:30.500 --> 01:05:33.140
You actually have to experience the thing to produce it.

01:05:33.300 --> 01:05:39.060
Animals that do that can then keep a beat to some exogenous stimulus

01:05:40.260 --> 01:05:46.130
and then have the the the perception action coupling ability to then move their m muscles

01:05:46.130 --> 01:05:47.490
to the thing that they hear.

01:05:47.730 --> 01:05:50.210
But there's very few animals that actually have that ability.

01:05:50.450 --> 01:05:55.650
Sea lions are interesting because they are not vocal learners, but they are in their clad,

01:05:55.890 --> 01:05:57.090
they're our vocal learners.

01:05:57.170 --> 01:06:01.970
So there's a latent ability in there that's not expressed in that species, but the the

01:06:01.970 --> 01:06:03.170
ability is clearly there.

01:06:03.250 --> 01:06:07.090
And then other related animals can do it, our vocal learners.

01:06:07.410 --> 01:06:13.120
So um I think the evidence is pretty good that vocal learning is important for entrainment.

01:06:13.200 --> 01:06:19.040
And so anybody, any human is at least is is kind of an expert in that sense, but then to be

01:06:19.040 --> 01:06:23.360
really good at rhythm is then you're tapping into like the best people.

01:06:23.840 --> 01:06:24.080
Right.

01:06:24.160 --> 01:06:29.280
Or maybe the emotional expression of like of like the timing of uh David, I know you're

01:06:29.280 --> 01:06:32.560
gonna say, is there any evidence of um emotion in music linkage?

01:06:32.640 --> 01:06:38.450
So like are really good musicians better at linking emotions and sound

01:06:38.610 --> 01:06:40.130
or emotions in general?

01:06:40.450 --> 01:06:40.850
Yes.

01:06:41.010 --> 01:06:41.330
Yes.

01:06:41.490 --> 01:06:46.770
So trained musicians are better at identifying emotion categories in music and in

01:06:46.770 --> 01:06:47.650
voices.

01:06:48.850 --> 01:06:51.730
Not much, but there is an effect for sure.

01:06:52.130 --> 01:06:52.450
Cool.

01:06:52.610 --> 01:06:55.170
But then trained musicians are better than a lot of things.

01:06:55.330 --> 01:06:59.940
You know, it's like there's there's so many variables like when you're a trained musician,

01:07:00.020 --> 01:07:01.300
that's a pretty privileged situation.

01:07:01.460 --> 01:07:01.780
Sure.

01:07:01.940 --> 01:07:04.100
So you probably have a lot of advantages.

01:07:04.260 --> 01:07:04.420
Yeah.

01:07:04.500 --> 01:07:10.260
Um, and also um people you learn music, mu just learning music enhances other cognitive

01:07:10.260 --> 01:07:10.980
abilities.

01:07:11.140 --> 01:07:13.540
Um, it's sort of like exercises for the brain.

01:07:13.620 --> 01:07:18.820
And so um, if you're doing music, that's good for your brain and it activates all all areas

01:07:18.820 --> 01:07:22.020
of the brain, music triggers, unlike most other things.

01:07:22.970 --> 01:07:23.770
Interesting.

01:07:24.010 --> 01:07:29.930
Uh why do I remember so vividly and connect so deeply with the music I heard

01:07:29.930 --> 01:07:32.410
during my teen years, maybe early twenties?

01:07:32.570 --> 01:07:35.690
Like I feel like everyone has that experience where they really connect with music at that

01:07:35.690 --> 01:07:35.930
age.

01:07:36.330 --> 01:07:38.970
Uh and like all their favorite stuff is comes from that era.

01:07:39.130 --> 01:07:42.890
Like, you know, my dad still listens to the the Beatles, like he's listened to that growing

01:07:42.890 --> 01:07:46.250
up growing up, and uh like all those old bands from the sixties.

01:07:46.570 --> 01:07:50.010
Um like what there's a strong age effect there.

01:07:50.330 --> 01:07:50.890
Huge.

01:07:51.340 --> 01:07:53.980
How do you have any sense of why that is?

01:07:54.460 --> 01:07:55.020
Yeah.

01:07:55.500 --> 01:08:00.300
I mean, I think this is it fits really well into the coalition signaling argument because

01:08:01.820 --> 01:08:03.500
think of music as cultural capital.

01:08:03.580 --> 01:08:06.140
You're learning the music of your culture.

01:08:06.620 --> 01:08:10.700
You have to learn the songs, you have to learn the scale, the musical system, right?

01:08:11.020 --> 01:08:12.380
And you learn that like language.

01:08:12.540 --> 01:08:18.540
So as soon as you learn the musical system of your music, then you start being

01:08:18.610 --> 01:08:20.930
unable to make distinctions in other systems.

01:08:21.010 --> 01:08:26.130
And so you can't like Western music listeners can't hear that the sitar is out of tune, for

01:08:26.130 --> 01:08:26.690
example.

01:08:26.850 --> 01:08:31.970
Um there's kind of the famous story of Robbie Shankar tuning up his sitar for an audience

01:08:32.050 --> 01:08:38.050
and he finished tuning, but it clapped and then and George mortified and and uh and then

01:08:38.050 --> 01:08:42.370
he said something like Well if you like that I hope you really like the songs.

01:08:43.970 --> 01:08:49.840
So um we we are like language, we close ourselves off to other languages once

01:08:49.840 --> 01:08:52.080
we learn the distinctions that are important in our language.

01:08:52.160 --> 01:08:55.280
So think of that as like the musical capital that you've learned.

01:08:55.360 --> 01:08:56.880
I like the capital idea.

01:08:57.040 --> 01:09:02.720
And so um during puberty, you're sort of there's I think like an imprinting process

01:09:02.880 --> 01:09:08.030
where the culture a lot of cultural phenomena, including t whatever it is happens to be in

01:09:08.030 --> 01:09:08.430
your culture.

01:09:08.510 --> 01:09:14.110
In our case, it's like TV shows and music and movies and um the th sayings

01:09:14.350 --> 01:09:16.190
and and accents or whatever.

01:09:16.510 --> 01:09:17.470
And clothes, yeah.

01:09:21.390 --> 01:09:27.150
Um y or memory systems are um there I think there's probably could be adaptations

01:09:27.150 --> 01:09:30.940
there that um are sort of imprinting those memories.

01:09:31.020 --> 01:09:34.460
So you remember the lyrics, you remember the songs in ways that you don't remember things

01:09:34.460 --> 01:09:34.700
later.

01:09:34.780 --> 01:09:36.940
And it's not just because you're getting old and you can't remember.

01:09:37.020 --> 01:09:40.300
It's there's something specific about that puberty old time.

01:09:41.100 --> 01:09:46.140
And so I think it's in an ancestral environment, that's just the way that you adopt the

01:09:46.140 --> 01:09:49.100
culture that you're from and then you can honestly signal you're from it.

01:09:49.260 --> 01:09:53.580
Because if you come in as an outsider, as an adult, you won't have that effect.

01:09:53.740 --> 01:09:57.350
And so it would be easy to detect that you're an outsider.

01:09:57.430 --> 01:10:00.550
And similar to like um David Dave's work on accents, right?

01:10:00.710 --> 01:10:05.910
There's a certain thing about you learn accents, and it's hard to actually go into a group

01:10:05.910 --> 01:10:09.750
and pretend you're from there and have the accent.

01:10:09.830 --> 01:10:14.390
Like there's I don't think I could ever go to uh some village in England and pretend that

01:10:14.390 --> 01:10:18.870
I'm no matter how long I live there, they would always know I was not one of them.

01:10:19.110 --> 01:10:20.760
And so I think music's the same way.

01:10:20.840 --> 01:10:25.400
And that's that's the effect you're describing is basically that process of encoding the

01:10:25.400 --> 01:10:27.320
musical culture that you grew up in.

01:10:27.560 --> 01:10:31.080
And so most people are like, music stopped being good after that, right?

01:10:31.480 --> 01:10:36.280
It's just like I predict like what when people think music started sucking is like when

01:10:36.280 --> 01:10:37.480
they were 30 or whatever.

01:10:37.560 --> 01:10:37.880
Yeah.

01:10:38.680 --> 01:10:39.880
Has there been work on them?

01:10:40.040 --> 01:10:41.400
Has there been work on exception?

01:10:41.720 --> 01:10:43.640
I'm always looking for new music, but the what?

01:10:43.960 --> 01:10:45.880
Has there been empirical work on that?

01:10:46.770 --> 01:10:50.610
very little actually there was a there was um um

01:10:52.770 --> 01:10:57.410
i i shouldn't say that there i was gonna talk about an ehb paper it didn't make it but um

01:10:57.810 --> 01:11:03.570
yeah not enough work on that i've always been fascinated with that idea and it's in

01:11:03.730 --> 01:11:07.490
some ways some aspects of that are quite tractable i mean you just ask people about what

01:11:07.490 --> 01:11:12.580
music they like yeah how old they are Showing showing the tendency would be easy.

01:11:12.980 --> 01:11:14.980
If you if you didn't show it, you fucked up.

01:11:15.060 --> 01:11:16.500
I mean it's like Yeah.

01:11:16.660 --> 01:11:17.780
It's in the Simpsons.

01:11:17.940 --> 01:11:21.780
Uh so Homer Simpson, there's a classic quote from the 90s Simpsons that rock achieved

01:11:21.860 --> 01:11:23.380
perfection in 1974, right?

01:11:23.460 --> 01:11:25.220
Like when they're listening to music in the car.

01:11:25.300 --> 01:11:25.860
Yeah.

01:11:27.300 --> 01:11:32.890
So um, but then I think what is a more difficult thing is to try to measure um

01:11:33.130 --> 01:11:38.890
how that might impact other aspects of your judgment about other individuals and other

01:11:38.890 --> 01:11:40.410
social effects and I don't know.

01:11:40.570 --> 01:11:44.010
But yeah, uh that's a really interesting and super robust effect in my opinion.

01:11:44.170 --> 01:11:44.410
Yeah.

01:11:45.370 --> 01:11:46.010
Yeah, totally.

01:11:46.410 --> 01:11:49.290
I have a weird thing where I'm thinking I'm echoing again.

01:11:49.450 --> 01:11:49.690
Jeez.

01:11:49.850 --> 01:11:53.930
Um I have a weird thing where in addition to memor you know, remembering the songs of that

01:11:53.930 --> 01:11:59.220
era very vividly, I have a really vivid memory of all the different film scores from that

01:11:59.220 --> 01:11:59.940
time in my life.

01:12:00.100 --> 01:12:04.260
If you just name one movie from like the late nineties, early two thousands, I'll be able

01:12:04.260 --> 01:12:06.500
to like tell you the hum the film score to it.

01:12:06.660 --> 01:12:11.300
Like even rent like speed, I remember the melody from speed, like from Keanu Reeves.

01:12:11.460 --> 01:12:12.820
Like I so it's a weird thing.

01:12:13.060 --> 01:12:16.500
I I don't know if that's a common thing, but like I have a really good memory for film

01:12:16.500 --> 01:12:20.340
scores, particularly from that time of my life, even if I saw the movie like maybe once or

01:12:20.340 --> 01:12:20.660
twice.

01:12:20.740 --> 01:12:21.940
Like Interesting.

01:12:22.100 --> 01:12:25.220
Yeah, I mean we and you you're obviously pretty into film, right?

01:12:25.300 --> 01:12:27.610
So then that's an attractor for you, right?

01:12:27.850 --> 01:12:29.610
Um, and then that example's interesting.

01:12:29.690 --> 01:12:33.370
So I'm older than you, and so that is not a period that I know, right?

01:12:33.530 --> 01:12:36.970
I'm I wouldn't watching as many movies back then, and I certainly could not.

01:12:37.050 --> 01:12:38.090
I've seen speed maybe twice.

01:12:38.250 --> 01:12:42.330
And that's not true, but there's no way I couldn't even identify in a forced choice,

01:12:42.410 --> 01:12:42.730
probably.

01:12:42.890 --> 01:12:43.370
Yeah.

01:12:45.210 --> 01:12:45.850
Yeah.

01:12:46.170 --> 01:12:47.130
It's funny.

01:12:47.370 --> 01:12:53.370
Um, I I could talk about music the whole time, but uh, but I I know there's a lot

01:12:53.370 --> 01:12:59.320
of things that you've studied um in the acoustic modality that people are either sensitive

01:12:59.320 --> 01:13:03.000
to, um, that maybe may be surprising for people.

01:13:03.560 --> 01:13:09.480
Um are there are there ones that uh you think are cool

01:13:09.480 --> 01:13:10.200
or exciting?

01:13:10.360 --> 01:13:14.360
I mean they're all cool and exciting, but are there particular ones that either you've done

01:13:14.360 --> 01:13:17.160
in the past or that you're working on now?

01:13:17.750 --> 01:13:19.510
You mean areas of research or topics?

01:13:19.990 --> 01:13:20.710
Yeah, yeah, yeah, yeah.

01:13:20.790 --> 01:13:22.310
I mean things you're doing or colleagues.

01:13:22.470 --> 01:13:25.990
But yeah, I mean I mean because I don't want to I don't want to have people just leave with

01:13:25.990 --> 01:13:29.030
the impression that you're doing just music.

01:13:29.190 --> 01:13:32.630
Um that's probably the the thing I've done the least empirical work.

01:13:32.710 --> 01:13:33.590
That's all I think.

01:13:33.910 --> 01:13:38.070
So I've done so yeah I do have some I have a couple papers on music perception stuff.

01:13:38.470 --> 01:13:42.310
Yeah and I think it's fascinating but I do want to like you know I I want to be able to

01:13:42.310 --> 01:13:46.100
talk a bit about what you've what else you've done empirically.

01:13:46.260 --> 01:13:51.220
Um so can you give us, I don't know what the best like, is there a story of like have you

01:13:51.220 --> 01:13:54.180
come around to like this this is the thing I'm focusing on?

01:13:54.340 --> 01:13:57.140
Or is it just that there's like this you're a kid in a candy shop.

01:13:57.220 --> 01:13:58.340
There's so much to do.

01:13:58.580 --> 01:14:02.500
You have the lucky position of being an acoustic adaptationist and you're like, we're gonna

01:14:02.500 --> 01:14:03.700
do this and this and this and this.

01:14:03.860 --> 01:14:04.020
Yeah.

01:14:04.260 --> 01:14:04.500
Yeah.

01:14:04.660 --> 01:14:10.650
I mean, I guess I came into the whole thing interested in how people

01:14:10.650 --> 01:14:12.810
use their voice to convey emotions.

01:14:13.290 --> 01:14:16.170
And then the and and through music.

01:14:16.650 --> 01:14:20.570
So what what are the musical properties of speech is kind of a thing to think about.

01:14:20.650 --> 01:14:22.490
And so I got interested in prosody.

01:14:22.730 --> 01:14:24.650
That's sort of the interface between language and music.

01:14:24.730 --> 01:14:30.490
So prosody is this pitch and loudness and rhythm and some spectral features that are

01:14:31.130 --> 01:14:36.970
acting on top of utterances that give you information that's beyond what's in the language.

01:14:37.360 --> 01:14:40.800
And um I think it's a super powerful mechanism.

01:14:40.960 --> 01:14:45.360
I mean, you can interpret a lot with from prosody without hearing any of the words.

01:14:45.440 --> 01:14:50.560
You can content mask speech and people are really good at identifying all sorts of rich

01:14:50.560 --> 01:14:53.120
dimensions about the speech signal just based on the melody.

01:14:53.280 --> 01:14:57.120
So I get that's what originally drove drew me into this whole area of research.

01:14:57.200 --> 01:15:03.150
And then Lita's paper um that I mentioned before was um you know focusing on

01:15:03.150 --> 01:15:07.870
that, looking at um how you know there's these certain invariances and acoustic structure

01:15:07.870 --> 01:15:10.750
across different emotions that are not about the language.

01:15:10.990 --> 01:15:16.670
And um in 1983, that was kind of a novel idea, kind of, and so um, yeah,

01:15:16.830 --> 01:15:19.870
so I got into that and then I started studying spontaneous speech.

01:15:20.030 --> 01:15:23.550
And when I got my PhD in it's in cognitive psychology and psycholinguistics was the

01:15:23.550 --> 01:15:24.830
emphasis in the program.

01:15:25.950 --> 01:15:27.870
One of them, one of the big things.

01:15:28.030 --> 01:15:33.640
And so I started working with my advisor Jeannie Foxtree, who studies spontaneous speech,

01:15:33.720 --> 01:15:35.080
not from an evolutionary perspective.

01:15:35.160 --> 01:15:36.520
That was an issue, actually.

01:15:37.480 --> 01:15:43.080
But so we kind of settled on this topic of verbal irony and

01:15:43.320 --> 01:15:48.280
the idea of whether when people produce verbal irony spontaneously like sarcasm, are they

01:15:48.280 --> 01:15:52.040
doing some particular thing with their voice to express that indirect meaning?

01:15:52.410 --> 01:15:57.530
or is it an illusion, which she thought it might be an illusion because she studied what um

01:15:57.610 --> 01:15:59.130
syntactic disambiguation?

01:15:59.210 --> 01:16:04.650
So people think that when you have a syntactically ambiguous utterance, then um when you

01:16:04.650 --> 01:16:08.890
say it one way versus another in context, people think they say it differently.

01:16:09.050 --> 01:16:11.370
But if you look acoustically, they don't.

01:16:11.530 --> 01:16:16.940
And so if I say a man eating fish, And people are like, and then you might think of a man

01:16:16.940 --> 01:16:19.820
eating a fish, or you might think of a shark, right?

01:16:19.980 --> 01:16:25.100
And people assume that you say it differently when it's one thing versus another.

01:16:25.180 --> 01:16:26.460
But if you look, they don't really.

01:16:26.540 --> 01:16:28.700
And so we kind of hear something that's not really there.

01:16:28.860 --> 01:16:30.620
So she thought this was true for sarcasm.

01:16:30.700 --> 01:16:34.780
And I didn't believe I thought sarcasm actually did have some special thing that was

01:16:34.780 --> 01:16:37.260
helping people understand that it was sarcasm.

01:16:37.500 --> 01:16:39.580
And it turned out we were kind of both right.

01:16:39.660 --> 01:16:41.020
There, it's less than you think.

01:16:41.100 --> 01:16:45.950
It's it's pretty ambiguous, but there are certain things that people do when they're

01:16:45.950 --> 01:16:49.790
speaking ironically that are, I think, helping listeners understand it.

01:16:49.870 --> 01:16:51.230
So people slow down.

01:16:51.550 --> 01:16:55.470
For example, that's the one most robust effect is a speed change.

01:16:55.790 --> 01:17:00.910
And so when I was studying irony, and that was sort of the way I could get into

01:17:00.910 --> 01:17:04.830
experimentally looking at the effect of prosody on people's language processing.

01:17:04.990 --> 01:17:07.150
But then I noticed people were laughing a lot.

01:17:07.310 --> 01:17:08.750
And that was the signal.

01:17:08.980 --> 01:17:12.820
And so then that that got me going into like, well, where's laughter coming from?

01:17:12.980 --> 01:17:18.500
And then it there's there was uh some existing um literature on laughter as a play

01:17:18.500 --> 01:17:20.820
vocalization, which I was like, yes, that is it right there.

01:17:20.900 --> 01:17:25.860
And and that's now still I'm doing a lot of work like that, um, looking at laughter as a

01:17:25.860 --> 01:17:28.580
play vocalization and non-human animals including.

01:17:28.820 --> 01:17:32.660
But um that's how I got into laughter is from this irony stuff.

01:17:32.820 --> 01:17:38.570
And then I realized laughter is this really understudied and super tractable vocalization

01:17:38.570 --> 01:17:42.330
that gives you a window into the evolution of vocal communication that's really nice

01:17:42.490 --> 01:17:45.290
because it's homologous across all mammals.

01:17:45.610 --> 01:17:50.410
But it's incorporated into our linguistic, you know, into speech.

01:17:51.050 --> 01:17:56.810
And then it also is a good example of a vocalization that is produced by the

01:17:56.810 --> 01:18:02.580
vocal motion system, which is a separate evolutionarily conserved system that is

01:18:02.580 --> 01:18:07.540
used to produce crying, laughs, pain shrieks, copulation calls.

01:18:07.860 --> 01:18:13.460
And then there's the speech system, which is a different control system for

01:18:13.460 --> 01:18:16.980
articulators in the larynx and the mouth for speech.

01:18:17.140 --> 01:18:23.070
And it's a different system, and only one can actually control the vocal apparatus of the

01:18:23.070 --> 01:18:23.230
time.

01:18:23.390 --> 01:18:27.470
So that's why crying children and laughing people can't talk, right?

01:18:27.630 --> 01:18:31.630
Because that system is hijacking the vocal production mechanism.

01:18:31.710 --> 01:18:34.350
And so the speech system can't get in there.

01:18:34.510 --> 01:18:37.870
And that's why when you're talking and you're laughing while you're talking, then you're

01:18:37.870 --> 01:18:43.870
giving evidence that you're laughing in from the volitional system because it's actually

01:18:43.870 --> 01:18:44.910
happening with speech.

01:18:45.150 --> 01:18:47.950
Otherwise, if you're really laughing, you have to stop talking.

01:18:48.260 --> 01:18:54.260
So laughter is this really beautiful vocalization to study by

01:18:54.260 --> 01:18:56.340
looking at how these two systems interact.

01:18:56.420 --> 01:19:02.260
When we're unique in um all animals for having two vocal production systems like that, um,

01:19:02.420 --> 01:19:06.100
this volitional system is super important for language and speech.

01:19:06.180 --> 01:19:10.980
And then turns out that I think one of the targets of selection for that volitional system

01:19:10.980 --> 01:19:14.490
is emulating the production of the vocal emotion system.

01:19:14.570 --> 01:19:19.450
So we do fake laughs, fake cries, fake orgasms, fake pain streaks, or whatever.

01:19:19.690 --> 01:19:25.370
And then that creates this whole extra level of dynamics if you're thinking about

01:19:25.530 --> 01:19:31.450
like game theoretically, and like how you're gonna understand the equilibria that

01:19:31.450 --> 01:19:32.650
emerge in communication systems.

01:19:32.730 --> 01:19:36.250
And when now we have this added level of volitional control that just makes it way more

01:19:36.250 --> 01:19:36.810
complicated.

01:19:36.970 --> 01:19:39.530
So you could potentially be faking almost anything.

01:19:39.850 --> 01:19:40.170
Interesting.

01:19:40.890 --> 01:19:44.400
So laughter is a totally cool vocalization to study for that reason.

01:19:44.720 --> 01:19:48.400
And um now I'm studying, I've been getting more into bioacoustics.

01:19:48.480 --> 01:19:54.320
So um my grad student um is the director of a Gibbon Conservation Center in Santa Clarita,

01:19:54.400 --> 01:19:55.920
and so she's got 40 gibbons.

01:19:56.160 --> 01:20:00.560
And so I've been getting into gibbon and gibbons are interesting because they do a lot of

01:20:00.560 --> 01:20:06.530
dueting, so they do territorial um um displays,

01:20:06.690 --> 01:20:11.010
you know, mating pair duets, and they give information that this is our territory, and you

01:20:11.010 --> 01:20:15.810
can get inform you can glean like how long they've been together, and there's other aspects

01:20:15.810 --> 01:20:19.170
of their duetting that give you important information about the ecology.

01:20:19.490 --> 01:20:23.250
And they produce these other vocals, I mean the crazy vocalizations, these gibbons.

01:20:23.490 --> 01:20:27.970
And then I have a postdoc who is studying dog play vocalizations.

01:20:28.050 --> 01:20:33.500
And this is more connected to laughter, where looking at dogs, and we just went and did

01:20:33.500 --> 01:20:35.180
some foxes last week.

01:20:35.340 --> 01:20:37.740
There's some Russian domesticated foxes down in San Diego.

01:20:37.820 --> 01:20:43.100
So I've been getting more into this non-human animal vocalizations and trying to connect

01:20:43.100 --> 01:20:48.060
those to things that we think might be true for human laughter and showing that, for

01:20:48.060 --> 01:20:53.980
example, domestication in play vocalizations and canids could be informative regarding the

01:20:53.980 --> 01:20:56.750
domestication of the play vocalizations in humans.

01:20:56.910 --> 01:21:02.270
And as a domesticated species, we've now um our a lot of our vocal behavior has changed, in

01:21:02.270 --> 01:21:07.150
part because we have this new speech system that allows us all this flexibility.

01:21:07.790 --> 01:21:13.390
So laughter is just just seems to always pop up and it's

01:21:14.030 --> 01:21:15.550
it's fun too.

01:21:16.270 --> 01:21:16.510
Yeah.

01:21:19.620 --> 01:21:25.060
Um do you have any ideas about I mean I have ideas about why laughter is contagious and why

01:21:25.060 --> 01:21:29.620
hearing laughter makes you reciprocate it and why we you know you laugh and I laugh and you

01:21:29.620 --> 01:21:33.220
laugh and I laugh and we you know it's it it's this almost this kind of duet.

01:21:33.380 --> 01:21:34.340
Um I have ideas on that.

01:21:34.420 --> 01:21:36.340
I'm curious if you have any thoughts on that.

01:21:36.790 --> 01:21:37.110
Yeah.

01:21:37.270 --> 01:21:42.310
Well, I mean, I've heard, I mean, I I know your the theory, which I think is overlaps

01:21:42.310 --> 01:21:43.590
hugely with encryption.

01:21:43.750 --> 01:21:43.910
Yeah.

01:21:44.070 --> 01:21:49.830
And that's always been my approach is that laughter is um a signal of of decryption

01:21:49.910 --> 01:21:50.390
essentially.

01:21:50.470 --> 01:21:56.470
And then so you're giving covert um a covert signal of I have the information that you're

01:21:56.470 --> 01:21:58.790
referring to to explicitly state it.

01:21:59.020 --> 01:22:04.620
I know your thing is about mix-ups in particular, which we we could I think we've talked

01:22:04.620 --> 01:22:05.100
about this a bit.

01:22:05.180 --> 01:22:06.460
We'll probably can talk about it more.

01:22:06.620 --> 01:22:11.500
I mean, you might say mix-ups are the higher the superordinate category, and I'd say

01:22:11.500 --> 01:22:12.060
encryption is.

01:22:12.140 --> 01:22:13.660
And so I mean we go back.

01:22:14.140 --> 01:22:16.460
But those are highly compatible ideas.

01:22:16.620 --> 01:22:16.780
Yeah.

01:22:16.940 --> 01:22:20.620
Um, so we're definitely on the same um wavelength on that one.

01:22:20.780 --> 01:22:26.700
But yeah, so laughter, I think, often functions as a way to signal to others that you have

01:22:26.700 --> 01:22:29.570
information without explicitly stating it.

01:22:29.810 --> 01:22:34.370
Um I also think that there's good evidence that laughter, co-laughter between multiple

01:22:34.370 --> 01:22:39.650
individuals can form a chorus signal that alert outsiders or give not alert necessarily,

01:22:39.730 --> 01:22:44.850
maybe sometimes, but more of like inform outsiders about the coalition strength of the

01:22:44.850 --> 01:22:45.490
people laughing.

01:22:45.570 --> 01:22:50.500
Like you can detect how well for one, I have empirical evidence that cross-culturally you

01:22:50.500 --> 01:22:54.900
can detect how if people are friends or strangers in one second of co-laughter.

01:22:55.060 --> 01:22:57.860
Um, and it doesn't work like that with speech exactly.

01:22:58.020 --> 01:23:04.020
And so, but I think that um there could be some exaggerated features of laughter

01:23:04.180 --> 01:23:09.860
can be explained as being a signal for wide broadcast um and giving information about

01:23:09.860 --> 01:23:15.110
coalition strength that's beyond just friends or strangers, but you can actually get pretty

01:23:15.110 --> 01:23:19.830
rich information about the constitution of a group by just paying attention for a minute on

01:23:19.830 --> 01:23:22.390
who's laughing and and how loud.

01:23:22.550 --> 01:23:28.150
And so I think laughter is doing stuff that's outside of the immediate group that the

01:23:28.150 --> 01:23:29.750
laughter is occurring within.

01:23:30.790 --> 01:23:31.750
Yeah, no, that's interesting.

01:23:31.830 --> 01:23:37.760
I mean, my my thinking on the topic is that um i if we mix up our meanings of a

01:23:37.760 --> 01:23:41.680
phrase, like for instance, you're confused about what whether hook up with means meet up

01:23:41.680 --> 01:23:44.560
with or have sex with, and you say you hooked up with my mother, right?

01:23:44.720 --> 01:23:50.000
Um if uh so that would be a pretty costly mix-up, you know, if you mix up those two

01:23:50.000 --> 01:23:51.680
meanings of of hookup with.

01:23:51.760 --> 01:23:56.240
Um and we uh we need some way of diffusing the costs of that mix-up so there's no grudge,

01:23:56.320 --> 01:23:58.010
it doesn't damage our relationship.

01:23:58.090 --> 01:24:01.770
And we not only need to diffuse those costs, we need to create common knowledge of the fact

01:24:01.770 --> 01:24:03.130
that those costs have been diffused.

01:24:03.370 --> 01:24:06.810
Like I need to know that you know, that I know, that there was a mix-up, and we're not

01:24:06.810 --> 01:24:09.130
taking it seriously as actual communication, right?

01:24:09.210 --> 01:24:13.050
And I think that is is part of the function of laughter and why laughter is contagious.

01:24:13.210 --> 01:24:17.770
You want to reciprocate the laughter to to create common knowledge such that I send you the

01:24:17.770 --> 01:24:21.740
message and you send me a receipt of the message, and I send you a receipt of your receipt

01:24:21.740 --> 01:24:23.020
of the message, and so on.

01:24:23.100 --> 01:24:26.060
It's sort of like you know, Stephen Pinker talks about this in his new book, sort of like

01:24:26.060 --> 01:24:30.540
uh linking two electronic devices, or like you pairing two advice devices with the

01:24:30.540 --> 01:24:30.940
Bluetooth.

01:24:31.180 --> 01:24:35.420
Often they will send a message, and then the the the recipient system will send a receipt

01:24:35.420 --> 01:24:41.260
of the message, and then the other system will send a receipt of the of the receipt, and so

01:24:41.260 --> 01:24:44.510
on, on and on until they establish a connection.

01:24:44.590 --> 01:24:48.590
And I think laughter is pretty similar and it's a way of creating uh common knowledge and

01:24:48.590 --> 01:24:53.070
sort of ascending the almost infinite recursion of n thoughts about thoughts about

01:24:53.070 --> 01:24:53.630
thoughts, right?

01:24:53.790 --> 01:24:58.990
Um so I think that's that's what's going on with with contagious uh laughter and why we we

01:24:58.990 --> 01:25:00.590
feel the need to reciprocate it.

01:25:00.830 --> 01:25:06.830
Um and uh yeah I'm curious number one whether cool um yeah I'll I should I just put

01:25:07.230 --> 01:25:07.790
out the preprint.

01:25:07.870 --> 01:25:09.550
Maybe I'll send it to you uh when when we're done.

01:25:09.630 --> 01:25:12.400
Um yeah but Yeah, you've been sitting on that for a while, huh?

01:25:12.640 --> 01:25:13.840
Yeah, I've been sitting on that for a while.

01:25:14.000 --> 01:25:16.000
Finally finished the damn preprint.

01:25:16.240 --> 01:25:17.280
Uh yeah.

01:25:17.440 --> 01:25:21.200
Uh I had to work through all the the math is was really a pain in the ass.

01:25:21.280 --> 01:25:24.240
Uh um so that that took me a while to really finalize that.

01:25:24.400 --> 01:25:29.600
But um yeah, uh I'm curious number one if you you find that idea plausible and and number

01:25:29.600 --> 01:25:35.210
two, uh I I'm really curious about this uh distinction between fake and real

01:25:35.210 --> 01:25:40.330
laughter, and whether you think there's some kind of evolutionary arms race between fake

01:25:40.330 --> 01:25:43.290
laughters and fake laughter f detectors.

01:25:43.450 --> 01:25:47.610
Um if that's behind this and whether that applies to other kinds of vocal signals as well.

01:25:48.410 --> 01:25:54.330
Yeah, I think uh that it applies to all vocal signals in a sense that are um

01:25:54.680 --> 01:25:56.360
produced by both systems, right?

01:25:56.440 --> 01:26:00.760
Which is constrained by what the vocal emotion system produces, which is only nonverbal

01:26:00.760 --> 01:26:01.480
vocalizations.

01:26:01.640 --> 01:26:07.320
So yeah, I think it's the Athena activists and Naya have a paper that kind of started this

01:26:07.480 --> 01:26:13.080
the spontaneous volitional laughter idea, and we talk about the arms race and that um

01:26:13.240 --> 01:26:17.400
clearly there is going to be selection to detect whether people are producing something

01:26:17.400 --> 01:26:22.410
volitionally that gives you information about what their potential intentions are.

01:26:22.570 --> 01:26:27.610
Now, how I think a lot of times volitional laughter is also just polite laughter that

01:26:27.610 --> 01:26:31.370
functions in a conversation and um is like a conversational marker.

01:26:31.450 --> 01:26:32.730
So it's not deceptive.

01:26:32.890 --> 01:26:35.690
It's not always going to be deceptive when I volitionally laugh.

01:26:35.850 --> 01:26:40.090
It's I can I can give like a little signal that I understand you, sort of like what the

01:26:40.090 --> 01:26:45.530
function would be in a for a real laugh, but I'm not emotionally triggered to do it.

01:26:45.740 --> 01:26:51.500
Instead, I just do it as a signal that is not honest in the same way, but I

01:26:51.500 --> 01:26:56.860
mean then you could talk about the honesty there is that I think a complicated issue.

01:26:57.020 --> 01:27:03.020
But so it's not always a manipulative thing when we produce a volitional version

01:27:03.020 --> 01:27:05.580
of something that the vocal motion system does.

01:27:05.950 --> 01:27:10.830
Um I'm trying what was the first thing that you um whether you buy the idea that we're

01:27:10.830 --> 01:27:15.310
diffusing the costs of like uh uh either a miscommunication or or a faux pas or something

01:27:15.310 --> 01:27:19.550
like that and signaling that I like basically co-opting the play signal to to signal that

01:27:19.630 --> 01:27:23.390
you know we're not taking this seriously, just as the play signal often communicates, you

01:27:23.390 --> 01:27:26.990
know, this is not real aggression, I'm not actually chasing you or trying to hurt you.

01:27:27.150 --> 01:27:28.270
This is just pretend.

01:27:28.430 --> 01:27:32.990
I think that w uh evolution co-opted the play signal uh when we are mixed up with one

01:27:32.990 --> 01:27:37.660
another or when we have a miscommunication to say to signal that we're not mad at each

01:27:37.660 --> 01:27:41.580
other, this isn't yes, uh we're not ta treating this as real communication, we're sort of

01:27:41.580 --> 01:27:43.660
treating it as a play uh interaction instead.

01:27:43.900 --> 01:27:44.300
Yeah, yeah.

01:27:44.460 --> 01:27:45.420
No, I think that's right.

01:27:45.500 --> 01:27:48.220
And so the mix-up thing I think is totally consistent with that.

01:27:48.460 --> 01:27:52.620
But I I also think there could be there are other reasons we laugh that have to do with

01:27:52.620 --> 01:27:53.660
social cohesion.

01:27:53.820 --> 01:27:58.550
So if you're with an old friend, you're not really signaling about mix-ups anymore, right?

01:27:58.710 --> 01:28:03.830
Um, if somebody I've known for 40 years, um, I think that even though it might be derived

01:28:03.830 --> 01:28:07.270
from that at some level, and you might be able to make an argument that it's sort of like

01:28:07.350 --> 01:28:09.590
why do people laugh when they're by themselves?

01:28:09.750 --> 01:28:13.350
It's not obviously a different function, it's that there, it's as if there's another person

01:28:13.350 --> 01:28:13.590
there.

01:28:13.750 --> 01:28:19.430
So laughing with an old friend could be as if there's a mis a mix-up that we don't have.

01:28:19.670 --> 01:28:24.020
But I think it could also be that you are there, there's sort of endogenous reward when

01:28:24.020 --> 01:28:28.340
you're laughing together so there is a social bonding effect that can happen and so you

01:28:28.340 --> 01:28:34.100
might I I I think about like jokes between familiar people there the the mix-up angle I I'm

01:28:34.100 --> 01:28:37.940
gonna think of some good examples for you though and see how you answer them.

01:28:38.100 --> 01:28:42.660
I don't have them offhand but um I think there are examples of times where people are

01:28:42.660 --> 01:28:47.780
laughing a lot very motivated to laugh and there's not a mix up but um I have to think of

01:28:47.780 --> 01:28:52.190
some good ones because I I know what you'll you'll be able to figure out the mix up if you

01:28:52.190 --> 01:28:52.910
dig deep enough.

01:28:52.990 --> 01:28:55.070
I which is a good thing about the theory.

01:28:55.230 --> 01:28:55.470
Yeah.

01:28:55.630 --> 01:29:01.150
Um, but I think there's other things about just mutually recognizing preferences, um,

01:29:01.390 --> 01:29:06.750
you know, or in some in some cases, like you might be critical of something that you both

01:29:06.750 --> 01:29:06.990
know.

01:29:07.150 --> 01:29:11.150
I know your view, and you know my view, and we each know that we know that.

01:29:11.310 --> 01:29:14.590
Um, and so there's no need to signal that exactly anymore.

01:29:14.670 --> 01:29:20.120
But there's still like has it that has a pleasurable um element to it that causes us to

01:29:20.120 --> 01:29:21.960
still engage in the signaling system.

01:29:22.120 --> 01:29:26.600
And so it could be that it came from a mix-up and now it's taken on this other kind of

01:29:26.600 --> 01:29:31.960
function, which also could be a motivating proximate mechanism for generating a signal that

01:29:31.960 --> 01:29:35.320
other people can hear that gives information about our coalition status.

01:29:35.480 --> 01:29:40.520
So I guess that and that's consistent with your idea that it's not just about those people

01:29:40.520 --> 01:29:44.870
figuring out a mix-up, but and like I mentioned before, it's it could be about signaling

01:29:44.870 --> 01:29:48.150
something to other people outside of the group, which is why it's so loud.

01:29:48.310 --> 01:29:48.550
Right?

01:29:48.710 --> 01:29:52.070
Laughter is really ridiculously loud for what it should be doing.

01:29:52.310 --> 01:29:57.030
A better covert signal for the mix-up would be much quieter, like a play vocalization in

01:29:57.030 --> 01:29:58.230
most non-human animals.

01:29:58.310 --> 01:29:59.110
They're very quiet.

01:29:59.270 --> 01:30:01.750
They're not broadcasting that they're playing to everybody.

01:30:01.910 --> 01:30:05.270
They're only, and this is why they're hard to measure because you got to really get in

01:30:05.270 --> 01:30:05.910
there to record it.

01:30:05.990 --> 01:30:10.390
And it's hard to tell who produces it because they're barely, you know, and they're barely

01:30:10.390 --> 01:30:11.350
moving your mouth.

01:30:11.510 --> 01:30:16.080
And so these are very dis they're um conspiratorial whispers, if you want to put it in the

01:30:16.080 --> 01:30:19.680
Krebs and Dawkins sort of language, um play vocalizations are.

01:30:19.760 --> 01:30:24.160
But but human laughter and some other species, only a few, are loud.

01:30:24.480 --> 01:30:30.160
So that suggests another function that's beyond the dyadic interaction.

01:30:30.400 --> 01:30:30.640
Yeah.

01:30:31.360 --> 01:30:35.520
So I would just say there's other reasons we laugh, but I I like the mix-up idea explains a

01:30:35.520 --> 01:30:36.080
lot, I think.

01:30:36.400 --> 01:30:37.200
Yeah, that makes sense.

01:30:37.280 --> 01:30:40.800
And yeah, uh if you come up with any stumpers, uh I'm I'm happy to hear them.

01:30:40.960 --> 01:30:43.450
Yeah, I'm gonna please please do.

01:30:44.250 --> 01:30:44.810
Science and action.

01:30:45.050 --> 01:30:45.610
Science and action.

01:30:46.250 --> 01:30:52.250
Yeah, no, my thinking with the the loudness is that um we coordinate and cooperate in much

01:30:52.250 --> 01:30:53.210
larger groups.

01:30:53.290 --> 01:30:58.490
And if we're suffering a mix-up, uh the the people who are mixed up not only need to know

01:30:58.490 --> 01:31:02.330
it, but you know, all of their friends and associates and anyone who's nearby would benefit

01:31:02.330 --> 01:31:06.410
from knowing it as well because we coordinate as a group our our social conventions are

01:31:06.410 --> 01:31:09.920
community wide and not uh a function of you know two or three people.

01:31:10.080 --> 01:31:14.000
You know, we we share the same language as a community, and if there's a miscommunication,

01:31:14.080 --> 01:31:18.400
like we're all part playing that coordination game together of which sounds correspond to

01:31:18.400 --> 01:31:19.360
which things in the world.

01:31:19.520 --> 01:31:24.560
And insofar as there's a mix up between you know two people about which sounds correspond

01:31:24.560 --> 01:31:28.240
to which things, then the entire linguistic community has an interest in becoming aware of

01:31:28.240 --> 01:31:31.200
that and avoiding that kind of miscommunication as well.

01:31:31.440 --> 01:31:36.950
So that would be my speculation for why um No, you should check out

01:31:37.590 --> 01:31:43.270
Tom Flampson has the empirical part of well, there's the first encryption paper, which was

01:31:43.270 --> 01:31:48.630
Tom and Clark did um that had some experiments about how much prior knowledge do you have

01:31:48.710 --> 01:31:52.710
and how that affects humor or whatever, um, supporting the encryption idea.

01:31:52.790 --> 01:31:57.030
But then he Tom has um other cross-cultural data from the people he was working with in

01:31:57.030 --> 01:31:58.870
Brazil, uh, who's the name of the group?

01:31:58.950 --> 01:31:59.750
I can't remember.

01:31:59.990 --> 01:32:05.620
And um, and he has these examples of in communities, large the the whole

01:32:05.620 --> 01:32:10.580
village is meeting and then people make these certain kinds of jokes that then cause people

01:32:10.580 --> 01:32:10.980
to laugh.

01:32:11.060 --> 01:32:15.380
And he was analyzing like who was laughing at what and how that related to their social

01:32:15.380 --> 01:32:15.860
network.

01:32:16.020 --> 01:32:20.740
And um I don't know if you if you've seen that paper, but I think that would be useful for

01:32:20.740 --> 01:32:21.780
you to talk.

01:32:21.860 --> 01:32:24.580
I mean, I think your theory would fit well with that.

01:32:24.820 --> 01:32:25.450
Yeah, totally.

01:32:25.610 --> 01:32:30.170
I mean the other the other idea I had for why laughter varies in volume and why it might be

01:32:30.170 --> 01:32:34.970
louder in some contexts is if there's a larger cost to be diffused, you want to be much

01:32:34.970 --> 01:32:36.810
more certain that that cost has been diffused.

01:32:37.050 --> 01:32:41.290
And so you want to laugh louder and longer to really make sure that we both know that the

01:32:41.290 --> 01:32:44.650
other one knows that the other one knows that this that you actually didn't sleep with my

01:32:44.650 --> 01:32:45.450
mother, for example.

01:32:45.610 --> 01:32:50.250
You know, like the so the costlier the mix up is, the louder and longer we should laugh to

01:32:50.250 --> 01:32:54.180
really make sure that those costs have been so hooking up with the guy's a taco bell is not

01:32:54.180 --> 01:32:55.300
as bad as hooking up with your mother.

01:32:56.020 --> 01:32:59.140
Yeah, but I think it's evolved from example that you like.

01:32:59.540 --> 01:32:59.860
Exactly.

01:33:00.180 --> 01:33:04.420
His workshopping the the his workshopping I've been developing my material yeah through

01:33:04.420 --> 01:33:04.820
traveling.

01:33:06.820 --> 01:33:10.180
I mean that's that seems like a pretty testable prediction right there.

01:33:10.420 --> 01:33:10.660
Yeah.

01:33:10.740 --> 01:33:15.530
Yeah absolutely that that would be a cool way to work out you know what you know, have a

01:33:15.530 --> 01:33:18.730
good convincing argument about what the costs are, which seems it seems like a hard

01:33:18.730 --> 01:33:19.050
problem.

01:33:19.130 --> 01:33:19.610
Uh yeah.

01:33:19.770 --> 01:33:24.490
And it seems like intuitive it fits my personal experience and and a lot of like anecdotes

01:33:24.490 --> 01:33:28.810
of like people when people really give a big belly laugh, it's usually at like kind of like

01:33:28.810 --> 01:33:34.730
a like a dark joke or like a like a really like uh spicy, like costly mix up

01:33:34.810 --> 01:33:39.420
that that, you know, um we I it seems intuitively right to me.

01:33:39.660 --> 01:33:40.140
I don't know.

01:33:40.300 --> 01:33:40.540
Yeah.

01:33:40.940 --> 01:33:46.300
I mean, the the idea is similar to encryption in multiple ways, but one way is that people

01:33:46.460 --> 01:33:51.260
will have examples that at first pass seems to violate the thing, and then you can come up

01:33:51.260 --> 01:33:52.860
with actually that's encryption.

01:33:53.020 --> 01:33:54.460
And then you can show how it's encryption.

01:33:54.540 --> 01:33:58.140
It's not obvious immediately how it's encryption or how it's a mix-up.

01:33:58.540 --> 01:34:01.660
And so I think something to work out there is like what is the difference?

01:34:01.820 --> 01:34:07.100
Like are there encryptions that aren't mix-ups and are that but there's no mix-ups that

01:34:07.100 --> 01:34:08.300
aren't encryption.

01:34:09.930 --> 01:34:11.690
Um yeah.

01:34:11.850 --> 01:34:13.850
Uh yeah, I guess that's true.

01:34:13.930 --> 01:34:14.090
Yeah.

01:34:14.170 --> 01:34:19.690
There there are no mix-ups that aren't encryption in some sense, and that you need to be uh

01:34:19.930 --> 01:34:23.930
privy to the coordination game being played in order to understand that there has been a

01:34:23.930 --> 01:34:24.250
mix up.

01:34:24.570 --> 01:34:24.730
Right?

01:34:24.810 --> 01:34:28.410
You need to speak that language, you need to be in that culture to know what was mixed up

01:34:28.410 --> 01:34:28.810
with what.

01:34:28.890 --> 01:34:31.050
And if you're not a member of that culture, you won't know.

01:34:31.210 --> 01:34:37.060
Um so yeah, I guess in that sense, all mix-ups are in encryption are encrypted uh, you

01:34:37.060 --> 01:34:39.940
know, among the people who are insiders to that culture.

01:34:40.100 --> 01:34:40.420
Yeah.

01:34:40.580 --> 01:34:40.740
Right.

01:34:40.980 --> 01:34:46.100
Would would would a would one let me just push on that though, and I'm the bystander here.

01:34:46.260 --> 01:34:52.180
Um, if somebody is like imagining like a new culture and they don't know, and the I'm just

01:34:52.180 --> 01:34:57.860
imagining like this trope of like some westerner comes to like this very like belligerent

01:34:58.100 --> 01:35:03.380
culture where like they might kill the person and they're like the per the the westerner is

01:35:03.380 --> 01:35:08.850
sort of like says something which is inappropriate to the context, like oh nice weather

01:35:08.930 --> 01:35:10.370
we're having, a kind of laugh.

01:35:10.530 --> 01:35:15.090
You know, it's sort of a signal of like, I hope we're on the same page.

01:35:15.250 --> 01:35:17.890
Um, but they are laughing as sort of a gauge.

01:35:17.970 --> 01:35:23.890
Um, I mean, maybe that's a claim of like they're trying to probe for in um encryption, or

01:35:23.890 --> 01:35:28.610
that that's a affiliative, don't we have some common understanding?

01:35:28.690 --> 01:35:34.640
You know, um, some I don't know, maybe that's a a bad uh No, I mean I think that that fits

01:35:34.640 --> 01:35:36.400
within David's idea for sure.

01:35:38.080 --> 01:35:43.600
That you're you're you're trying to um exploit their their understanding of the signal,

01:35:43.680 --> 01:35:48.080
even if the con they they we have mutual knowledge that the con we don't have the common

01:35:48.080 --> 01:35:53.520
ground to for the language to handle it, but then I'm signaling as if we do as a good

01:35:53.520 --> 01:35:53.920
thing.

01:35:54.050 --> 01:35:58.050
Like I'm trying now, it sounded like you're being insincere in your laugh.

01:35:58.210 --> 01:36:00.610
If you had a bad fake laugh, that might be costly, right?

01:36:00.690 --> 01:36:05.090
You're if you say something inappropriate and you're like or whatever, then you know that

01:36:05.090 --> 01:36:07.570
might be worse than not producing a laugh.

01:36:07.730 --> 01:36:09.490
So you better make it a good one.

01:36:10.210 --> 01:36:10.450
Yeah.

01:36:10.690 --> 01:36:11.010
Yeah.

01:36:11.170 --> 01:36:11.650
Uh yeah.

01:36:12.050 --> 01:36:16.500
The more I think about humor and laughter, the more I realize just like how unfathomably

01:36:16.500 --> 01:36:17.220
complicated it is.

01:36:17.380 --> 01:36:21.380
Like every joke is just like you could write a whole book dissecting why it's funny.

01:36:22.500 --> 01:36:23.140
And they do.

01:36:23.380 --> 01:36:23.620
Yeah.

01:36:24.100 --> 01:36:27.620
And and just like the dove into the humor literature, I'm sure you have.

01:36:28.180 --> 01:36:28.660
Oh, I'm sure.

01:36:28.740 --> 01:36:28.980
Yeah.

01:36:29.060 --> 01:36:33.460
But even even beyond the funniness, there are like really complicated game theoretic

01:36:33.460 --> 01:36:37.650
constraints about like the incentives of different uh actors in the social network and

01:36:37.650 --> 01:36:41.410
whether they should laugh or whether they shouldn't or whether how loudly they should laugh

01:36:41.570 --> 01:36:45.970
or like what their interests are in in terms of whether they want those costs to be

01:36:45.970 --> 01:36:50.530
diffused or not, uh whether they should try to fake laughing if they don't get the joke,

01:36:50.690 --> 01:36:56.610
like really complicated considerations that I feel like with cost for whatever decision you

01:36:56.610 --> 01:37:02.050
make and then there um it it opens up a whole window of of costs and benefits that um hard

01:37:02.050 --> 01:37:03.250
to go back on.

01:37:03.840 --> 01:37:04.160
Yeah.

01:37:04.320 --> 01:37:04.640
Yeah.

01:37:04.720 --> 01:37:10.000
I I think the um the volitional laughter adds so I mean volitional production of anything

01:37:10.160 --> 01:37:12.640
really adds so much complication there for sure.

01:37:12.880 --> 01:37:17.120
And there was something else I was gonna say about the laughter signaling, and then it I

01:37:17.280 --> 01:37:20.880
forgot, but I'll I'll give you a second because I just can't imagine I'm the humor

01:37:20.880 --> 01:37:24.000
literature I imagine has no jokes in it whatsoever.

01:37:24.720 --> 01:37:26.720
Like this I imagine like jokes are so bad.

01:37:26.800 --> 01:37:31.040
I I've generally yeah, I'm sure like I imagine like there's an entire book on humor, like

01:37:31.120 --> 01:37:33.830
the book, the blurbs in the back, like I didn't laugh once, right?

01:37:33.910 --> 01:37:34.550
Like something like that.

01:37:34.790 --> 01:37:35.430
Yeah, totally.

01:37:35.830 --> 01:37:36.230
Yeah, yeah.

01:37:37.750 --> 01:37:41.270
My my page, the preprint, uh has some genuinely funny funny jokes.

01:37:41.750 --> 01:37:45.830
At least according to me, well, it has like, you know, all the movies that I grew up with,

01:37:45.910 --> 01:37:49.110
all the comedies, like, you know, Dumb and Dumber and stuff like that, uh, Seinfeld

01:37:49.350 --> 01:37:50.470
Seinfeld jokes.

01:37:50.550 --> 01:37:54.630
Um so I I like to think that the humor in there is is is funnier than the average.

01:37:54.950 --> 01:37:58.470
And then David, what you can do is you can be a testimonial, you can say this is

01:37:58.470 --> 01:37:59.590
scientifically funny.

01:37:59.670 --> 01:38:05.040
You could become that you're the you're the official if you don't recognize this, you are

01:38:05.120 --> 01:38:07.440
yeah, you know, technically have the bad sense of humor.

01:38:07.680 --> 01:38:07.920
Exactly.

01:38:08.000 --> 01:38:09.600
Yes, this is empirically funny.

01:38:09.680 --> 01:38:12.960
Yes, wonderful way to abuse my scientific credit credibility.

01:38:15.520 --> 01:38:16.880
Actually, that dad joke was funny.

01:38:16.960 --> 01:38:18.720
Yeah, you're you're actually mistaken.

01:38:21.200 --> 01:38:24.000
Well, and that's the thing about the both of these ideas.

01:38:24.080 --> 01:38:29.040
Um, the the mix-up and the encryption is that it explains why jokes are not universally

01:38:29.040 --> 01:38:29.790
funny, right?

01:38:29.870 --> 01:38:34.110
I mean, it's all context dependent because it, you know, it depends what we're talking

01:38:34.110 --> 01:38:35.390
about and what the rules are.

01:38:35.550 --> 01:38:41.310
And um, and so yeah, no other theory of humor really has been able to

01:38:41.550 --> 01:38:42.670
point at that quite a bit.

01:38:42.750 --> 01:38:47.150
There are a lot of theories of humor out there that are approximate and capture some

01:38:47.470 --> 01:38:48.990
subgroup of jokes.

01:38:49.070 --> 01:38:54.190
And it's also the emphasis on jokes is is I think is big they've gone too far.

01:38:54.350 --> 01:39:00.260
The people that study humor um are thinking about the joke as sort of a proto humor.

01:39:00.580 --> 01:39:04.500
I that just that's not the way I mean, I think most people that study humor would realize

01:39:04.500 --> 01:39:09.460
it emerged first in conversational interactions that are context specific and that the joke

01:39:09.460 --> 01:39:10.420
is a trope.

01:39:10.580 --> 01:39:14.340
Um but it seems like there's way more emphas like punchlines and all that kind of stuff.

01:39:14.420 --> 01:39:19.060
I mean yeah comic timing too is a super rich Yeah.

01:39:19.220 --> 01:39:24.890
Um that fascinating I'm I'm sure there's tons of like things you could do with that.

01:39:24.970 --> 01:39:28.890
I doubt you know people have done a fraction of of the experiments and and and studies that

01:39:28.890 --> 01:39:30.410
you could do studying comic timing.

01:39:30.490 --> 01:39:33.850
But like I mean my I I think it were oh let's go ahead.

01:39:34.250 --> 01:39:37.130
Yeah no most thing about it is it's just like micro mix ups.

01:39:37.210 --> 01:39:41.210
Like we have these conventions with like you know the timing of our speech, the speed, you

01:39:41.210 --> 01:39:43.450
know how much we pause, you know, vocal intonation.

01:39:43.530 --> 01:39:47.290
We just talked about, you know, how, you know, that your tone of voice uh communicates so

01:39:47.290 --> 01:39:47.450
much.

01:39:47.530 --> 01:39:48.890
I mean you could have mix-ups there.

01:39:48.970 --> 01:39:54.800
Like you if I you know exaggerate um my my voice into kind of a s overly sing-songgy way,

01:39:54.960 --> 01:39:55.840
like that can be kind of funny.

01:39:55.920 --> 01:39:59.840
I'm like kind of mixed up about what the conventions are about how much I should change my

01:39:59.840 --> 01:40:04.560
tone of voice, you know, um, or if I you speak too loudly or too slowly, or if I pause in

01:40:04.560 --> 01:40:05.360
an awkward way.

01:40:05.520 --> 01:40:10.960
Like I think these these kinds of considerations factor into what we call comic timing, as

01:40:10.960 --> 01:40:16.570
it's layering in uh micro mix-ups about the uh the pacing and body language and

01:40:16.890 --> 01:40:21.050
intonation of the delivery of the joke uh on top of the content of the joke itself, which

01:40:21.050 --> 01:40:23.290
accentuates the the the comedy of it.

01:40:24.090 --> 01:40:27.690
And it's really uh the resolution of the timing is is incredible.

01:40:27.850 --> 01:40:32.170
I mean, if you just wait an extra hundred milliseconds, it matters.

01:40:32.410 --> 01:40:37.960
And um so I've done some work looking at how um timing can affect pragmatic

01:40:37.960 --> 01:40:41.480
inferences in quick interactions of just answering questions.

01:40:41.720 --> 01:40:47.640
Um and you know, if you wait just a little extra in how you answer a question, it

01:40:47.640 --> 01:40:52.200
has all these implications for what kinds of implicatures you're communicating, um, what

01:40:52.200 --> 01:40:54.440
kinds of things the listener might draw from it.

01:40:54.760 --> 01:41:00.680
I use an example in my class of um if you say um just in a normal greeting,

01:41:00.760 --> 01:41:01.880
like, hey, how are you doing?

01:41:01.960 --> 01:41:05.160
And so if um like ask me how I'm doing.

01:41:05.480 --> 01:41:06.440
How are you doing?

01:41:08.250 --> 01:41:08.890
Good.

01:41:09.370 --> 01:41:11.130
Now ask again.

01:41:11.450 --> 01:41:12.490
How you doing?

01:41:12.810 --> 01:41:13.450
Good.

01:41:14.410 --> 01:41:15.770
And then ask again.

01:41:16.010 --> 01:41:17.210
How are you doing?

01:41:17.610 --> 01:41:18.250
Good.

01:41:18.810 --> 01:41:19.210
Right?

01:41:19.290 --> 01:41:21.370
So that seemed to be kind of a normal.

01:41:21.690 --> 01:41:27.450
If you go too fast, it's there's all there's much different inferences you draw than and if

01:41:27.450 --> 01:41:29.850
I wait too long, then there's other complete.

01:41:30.010 --> 01:41:34.730
And the difference between those is like maybe was a 200 milliseconds, you know, a fifth of

01:41:34.730 --> 01:41:35.210
a second.

01:41:35.370 --> 01:41:35.530
Yeah.

01:41:35.770 --> 01:41:41.740
And so I think that basic principle of timing and pragmatics is then a particular

01:41:41.740 --> 01:41:43.820
feature in comic timing that's important.

01:41:44.140 --> 01:41:44.540
Totally.

01:41:44.620 --> 01:41:48.940
And if you're just a little bit off of what the the social convention is, it can be very

01:41:48.940 --> 01:41:49.100
funny.

01:41:49.180 --> 01:41:50.700
Like ask me how I'm doing.

01:41:51.020 --> 01:41:52.300
How you doing?

01:41:53.980 --> 01:41:54.540
See, that was funny.

01:41:56.060 --> 01:41:57.660
Like I didn't even wait for you to finish.

01:41:57.900 --> 01:41:58.620
So yeah.

01:41:59.820 --> 01:42:01.020
Yeah, there's so much there.

01:42:01.260 --> 01:42:07.230
Um I also have another pitch for you to work on differentiating um a hostile or e evil

01:42:07.230 --> 01:42:10.510
laugh from a friendly or convivial laugh.

01:42:10.830 --> 01:42:13.310
I feel like that could be really interesting to look at.

01:42:13.390 --> 01:42:18.750
Um I mean my my my thinking about why laughter can sometimes be hostile or creepy is that

01:42:19.150 --> 01:42:23.150
uh mirth or amusement evolved to you know diffuse the cost of mix-ups.

01:42:23.230 --> 01:42:27.460
And in order to do that, it needs to deactivate uh regions of the brain that process costs.

01:42:27.540 --> 01:42:29.060
So deactivate negative emotions.

01:42:29.140 --> 01:42:31.460
So i uh the same thing goes with if you're playing.

01:42:31.540 --> 01:42:34.180
If you're having rough and tumble play, you want to deactivate your fears so you're not

01:42:34.180 --> 01:42:35.940
actually afraid for your life.

01:42:36.020 --> 01:42:39.380
You know, if you want to deactivate your anger so you're not like you know vengeful and

01:42:39.380 --> 01:42:41.380
retaliating against the person who's play fighting with you.

01:42:41.540 --> 01:42:44.660
But in in the same way, if if we're trying to diffuse the costs of mix-ups, we want to

01:42:44.660 --> 01:42:49.540
deactivate those negative emotions, deactivate fear, anger, sadness, um, you know, uh w

01:42:49.700 --> 01:42:50.500
whatever.

01:42:50.740 --> 01:42:55.890
And uh the problem is though, uh, as soon as you have that machinery in place, um, that

01:42:55.890 --> 01:42:58.930
makes you very threatening to people who are not sharing your mirth.

01:42:59.250 --> 01:43:03.810
So if I'm laughing and mirthful, that means that my fear has been deactivated and uh and I

01:43:03.810 --> 01:43:07.010
could just attack you without fear of uh you know, retaliation.

01:43:08.210 --> 01:43:10.530
Um and my empathy might even be deactivated.

01:43:10.610 --> 01:43:14.450
If you're crying in pain or sad about something, like I'm not gonna feel empathy for you

01:43:14.530 --> 01:43:15.010
because I'm laughing.

01:43:15.090 --> 01:43:15.650
I'm mirthful.

01:43:15.730 --> 01:43:18.050
It's my those parts of my brain have been deactivated.

01:43:18.220 --> 01:43:22.220
So for that reason, I think mirth can often signal that I have a lack of fear and that I

01:43:22.220 --> 01:43:25.420
have a lack of empathy for your suffering, and that makes me really dangerous.

01:43:25.580 --> 01:43:31.420
Um and so uh I I I would be I would be uh interested in in in maybe

01:43:31.660 --> 01:43:35.900
uh you know evil laughter having a different signature, or maybe it maybe it doesn't.

01:43:35.980 --> 01:43:36.300
I don't know.

01:43:36.380 --> 01:43:38.940
Maybe we just prefer evil laughter.

01:43:39.020 --> 01:43:43.210
Yeah, I think my most recent paper on laughter, it looks at that this issue, um, among

01:43:43.210 --> 01:43:47.690
other things, but looking at um whether people can distinguish between different laugh

01:43:47.690 --> 01:43:49.770
types that we've had people produce.

01:43:49.930 --> 01:43:55.690
So we create contexts where you know malicious laughter and um um co amusement

01:43:55.850 --> 01:43:58.250
and sarcastic and all these different things.

01:43:58.410 --> 01:44:04.330
And it turns out that um it's really the dimensional view of emotion is the best way to

01:44:04.410 --> 01:44:05.380
to predict it.

01:44:05.540 --> 01:44:07.060
And in particular, valence.

01:44:07.220 --> 01:44:10.580
So it seems that LAF does work on the dimension of valence.

01:44:10.660 --> 01:44:15.060
So you can tell positive and negative, but we mix up positive categories at a high rate and

01:44:15.060 --> 01:44:17.940
we mix up negative categories at a high rate.

01:44:18.020 --> 01:44:20.980
And also with the machine learning algorithm, does the same kind of mix ups.

01:44:21.060 --> 01:44:24.500
And so we can see what the acoustic features are that are associated with these different

01:44:24.500 --> 01:44:30.290
laugh types, and then how those acoustic features predict it its valence almost completely,

01:44:30.450 --> 01:44:31.890
a little bit of arousal too.

01:44:32.050 --> 01:44:37.170
So the idea of a malicious laugh, like the whatever, is really a kind of a cultural

01:44:37.170 --> 01:44:41.090
invention that doesn't exist in ordinary discourse, usually.

01:44:41.410 --> 01:44:47.330
But I think what you're saying is right in that a context where you're laughing and you

01:44:47.330 --> 01:44:51.410
know you know the other person's not laughing, and they know that you know they're not

01:44:51.410 --> 01:44:57.140
laughing, and then you might hike up a little more negative um sort of energy in there to

01:44:57.140 --> 01:45:00.740
make it sound a little darker because laughter could be very intimidating in certain

01:45:00.740 --> 01:45:01.540
contexts.

01:45:01.700 --> 01:45:04.100
Um it's not always positive, obviously.

01:45:04.260 --> 01:45:07.060
And so in fact, it could be linked to any emotion.

01:45:07.300 --> 01:45:12.740
But um, yeah, so I agree that that there could be situations like what you described where

01:45:12.740 --> 01:45:17.780
the laughter is really communicating something about your lack of empathy or your lack of

01:45:17.780 --> 01:45:20.660
concern for the other, and it's not about correcting a mistake.

01:45:20.820 --> 01:45:22.340
Um same way.

01:45:22.550 --> 01:45:26.230
And but there's no categorical evil laugh.

01:45:26.630 --> 01:45:32.550
Um I think it's, I mean, other than in in movies that are you have to

01:45:32.550 --> 01:45:33.830
practice your evil cackles.

01:45:34.150 --> 01:45:34.310
Yeah.

01:45:34.470 --> 01:45:35.910
Nobody does that seriously, right?

01:45:35.990 --> 01:45:38.710
If you hear that, they're making they're doing the trope.

01:45:39.190 --> 01:45:39.510
Yeah.

01:45:40.070 --> 01:45:40.630
Interesting.

01:45:40.870 --> 01:45:44.630
But I'm sure it plays on, you know, like this, like I think it's what you're both saying,

01:45:44.710 --> 01:45:49.270
is like, you know, I mean, the tr a trope would be like you're a kid getting picked on, and

01:45:49.350 --> 01:45:52.060
and the kids who are picking on you are laughing, right?

01:45:52.140 --> 01:45:53.420
It's exactly what David's saying.

01:45:53.500 --> 01:45:56.220
I mean, that's the proper domain of this trope, right?

01:45:56.380 --> 01:45:56.540
Yeah.

01:45:56.700 --> 01:46:00.380
It's not that the that laughter is the cackle, but it feels like it.

01:46:00.460 --> 01:46:05.020
And then the trope, the the cackle sort of plays on that, yeah, elaborately.

01:46:05.260 --> 01:46:05.420
Yeah.

01:46:05.500 --> 01:46:07.340
I mean, I wonder how universal it is.

01:46:07.420 --> 01:46:09.340
It probably would be recognized pretty widely.

01:46:09.500 --> 01:46:13.580
I mean, I imagine I always do the Schwarz test because I've been there and I can imagine

01:46:13.740 --> 01:46:14.300
how they'd react.

01:46:14.380 --> 01:46:18.050
And it's like, what would a Schwarr person, how would they react to like a prototypical

01:46:18.130 --> 01:46:19.650
malicious laugh like from the movies?

01:46:19.730 --> 01:46:24.610
And would they understand in a forced choice, is this happy or is this is this e bad well,

01:46:24.690 --> 01:46:25.810
it is happy in a way.

01:46:25.890 --> 01:46:31.250
That's part of the meaning that's part of the talking about they're happy that it's bad for

01:46:31.250 --> 01:46:31.410
you.

01:46:31.570 --> 01:46:31.730
Yeah.

01:46:32.130 --> 01:46:35.570
Like why why does everybody in your culture monologue when they're and tell the

01:46:35.570 --> 01:46:38.610
protagonists what they're about to do so that they can foil the phone.

01:46:39.650 --> 01:46:40.290
Yeah.

01:46:44.200 --> 01:46:45.320
Yeah, it might.

01:46:45.400 --> 01:46:50.840
I mean you could ask, like, you know, what is the moral character of the person laughing?

01:46:50.920 --> 01:46:53.640
And like they might be able to say, like, that's that's a bad dude.

01:46:54.040 --> 01:46:56.920
He's gonna do some, he's he's gonna, you know, hurt some people.

01:46:57.160 --> 01:46:58.680
Yeah, he's happy doing bad.

01:46:59.320 --> 01:47:02.760
Although these days they might be like, Oh, isn't that the third uh Spider Man movie or

01:47:02.760 --> 01:47:03.160
something, right?

01:47:05.400 --> 01:47:09.960
Well, right, yeah, that those kinds of experiments are becoming harder and harder to do,

01:47:10.040 --> 01:47:12.440
yeah, to really get an isolated group.

01:47:12.680 --> 01:47:13.320
Yeah.

01:47:13.690 --> 01:47:14.010
Yeah.

01:47:14.250 --> 01:47:18.410
Even then, I mean, when I was down there 20 years ago, I mean, you know, I think somebody

01:47:18.410 --> 01:47:20.410
had a Madonna poster inside their hut.

01:47:20.490 --> 01:47:21.530
I mean, you know.

01:47:21.770 --> 01:47:22.010
Yeah.

01:47:22.170 --> 01:47:22.890
Subjectively.

01:47:23.050 --> 01:47:26.570
That's just the Chris von Rune thing where somebody who was like two days upriver and he

01:47:26.570 --> 01:47:29.610
was like, Did you hear did you hear Michael Jackson just died?

01:47:30.570 --> 01:47:32.410
And Chris von Rune was like, What?

01:47:32.570 --> 01:47:35.850
And we actually checked because we had him on and I verified because he told me this when

01:47:35.850 --> 01:47:36.730
we were in grad school.

01:47:36.890 --> 01:47:40.720
It turns out he Michael Jackson had died and it just turned it just They knew before him.

01:47:40.880 --> 01:47:45.680
He they knew before him he had he actually left the um he had been on the boat and during

01:47:45.680 --> 01:47:49.120
that time Michael Jackson had died and it just transferred by word of mouth faster than

01:47:49.120 --> 01:47:50.000
Chris got there.

01:47:50.160 --> 01:47:55.200
Um so he got told some pop culture from oh two days upriver or something like that.

01:47:55.520 --> 01:47:56.640
That's the state of the world now.

01:47:56.800 --> 01:47:57.040
Yeah.

01:47:57.200 --> 01:47:58.080
And that was a while ago.

01:47:58.160 --> 01:47:58.320
Yeah.

01:47:58.560 --> 01:47:59.760
Yeah, yeah, totally.

01:48:00.240 --> 01:48:02.960
Uh Greg, we are getting toward the end.

01:48:03.040 --> 01:48:06.640
We always have the question of should we have asked you anything?

01:48:09.530 --> 01:48:13.530
Uh I can't really think of anything offhand.

01:48:13.610 --> 01:48:16.170
It seems like you guys have covered it a lot.

01:48:16.490 --> 01:48:22.090
Um you know, I guess um you I guess you didn't really ask me about the

01:48:22.090 --> 01:48:23.610
non-human animal stuff.

01:48:23.690 --> 01:48:28.410
Um, but I mean I think there's an in uh sort of implied thing about the evolutionary

01:48:28.410 --> 01:48:34.300
history of these things is gonna um it's it's gonna be the case that a lot of the

01:48:34.300 --> 01:48:38.460
phenomena we see in human vocalizations are gonna be in non-human animal vocalizations.

01:48:38.540 --> 01:48:42.300
But yeah, is there something important or like that you think people it would be

01:48:42.300 --> 01:48:44.940
interesting to know about the non-human stuff?

01:48:45.660 --> 01:48:49.820
I mean, I guess uh when I think about it in terms of play vocalizations, I think the

01:48:49.980 --> 01:48:55.750
interesting phenomena there is that the adaptive problem of signaling

01:48:55.750 --> 01:49:00.470
something about play is present, and then how it manifests itself in any given species is

01:49:00.470 --> 01:49:03.430
constrained by the communicative systems of that species.

01:49:03.590 --> 01:49:08.150
So play vocalizations have very different sounds across animals because they're

01:49:08.150 --> 01:49:10.470
incorporated into the vocal repertoire of those animals.

01:49:10.630 --> 01:49:14.390
And so the level of analysis there is more appropriate to think about the adaptive problem

01:49:14.390 --> 01:49:15.510
that's being solved.

01:49:15.750 --> 01:49:21.300
And I think that know, that's where the adaptationist program is key.

01:49:21.460 --> 01:49:26.020
And maybe a lot of people that study that stuff in non-human animals aren't always

01:49:26.020 --> 01:49:26.980
adaptationists.

01:49:27.940 --> 01:49:33.620
So you might miss um homologous features because you're not thinking about the adaptive

01:49:33.620 --> 01:49:34.260
problem enough.

01:49:34.580 --> 01:49:39.940
Um I used to say laughter was just a homologous across apes, and that was sort of what

01:49:39.940 --> 01:49:42.830
people had suggested before I started thinking about this.

01:49:42.990 --> 01:49:47.630
The other people that had like Van Hoof and Provine and others, they would all really talk

01:49:47.630 --> 01:49:49.790
about um primates.

01:49:50.110 --> 01:49:51.950
And now I've become convinced.

01:49:52.030 --> 01:49:56.910
I have this recent paper I did with um anthropologist grad student Sasha Winkler looking at

01:49:56.910 --> 01:50:01.550
play vo like every single instance of play vocalizations noted in the literature that we

01:50:01.550 --> 01:50:06.800
could find, 65 species and all over the place, mammals mostly, but birds as well.

01:50:06.960 --> 01:50:11.840
And so the birds, I would say, are that's convergent evolution, but with all the mammals, I

01:50:11.840 --> 01:50:14.080
think those play vocalizations are homologous.

01:50:14.240 --> 01:50:19.280
And I used to think that maybe like um play vocalization in rats saying rats laugh, which

01:50:19.280 --> 01:50:20.560
they do quite a bit.

01:50:20.720 --> 01:50:25.440
Um, I used to think that was an analogy that that was convergent evolution and it wasn't

01:50:25.440 --> 01:50:29.920
homologous, but now I've been convinced that it's homologous, but it's just different, so

01:50:29.920 --> 01:50:35.230
different, because rats are vocalizing in these high frequencies, and we people didn't even

01:50:35.230 --> 01:50:39.550
know they were doing it until they just by accident, like 25 years ago.

01:50:39.950 --> 01:50:45.470
Um, but the rat laugh is at 50 kilohertz, which is well beyond what we can hear.

01:50:45.870 --> 01:50:50.350
Um, and so but I think it's a homologous behavior, even though it's radically different in

01:50:50.350 --> 01:50:52.590
terms of the acoustic properties.

01:50:52.910 --> 01:50:57.580
So I guess laughter might be like a hundred million years old.

01:50:57.900 --> 01:50:58.620
Exactly.

01:50:58.860 --> 01:50:59.180
Definitely.

01:50:59.900 --> 01:51:03.580
Now, whether we call it laughter or not, I mean, do we want to call what humans do

01:51:03.660 --> 01:51:05.580
something unique, which I think it is unique?

01:51:05.820 --> 01:51:10.220
Um, you know, some people say chimps laugh, and then that's fair because they're close.

01:51:10.380 --> 01:51:12.860
Um, but then do dogs laugh?

01:51:13.020 --> 01:51:17.580
And you know, I think you know, you get the risk of anthropomorphizing the behavior.

01:51:17.900 --> 01:51:22.530
Uh I prefer to certainly play vocalizations are you know probably a hundred billion years

01:51:22.530 --> 01:51:23.250
old at least.

01:51:23.490 --> 01:51:23.810
Yes.

01:51:23.890 --> 01:51:24.130
Yeah.

01:51:24.370 --> 01:51:25.010
Definitely.

01:51:25.250 --> 01:51:25.490
Yeah.

01:51:25.650 --> 01:51:26.290
Yeah.

01:51:26.610 --> 01:51:30.530
So that's a nice little behavior to study in humans when it's something that's that

01:51:30.530 --> 01:51:35.090
ancient, you know, and then it gets incorporated into even things like the conversational,

01:51:35.170 --> 01:51:40.370
like we're playing when in this in this conversation we're having right now, we're playing.

01:51:40.530 --> 01:51:40.770
Right.

01:51:40.850 --> 01:51:42.530
There's all kinds of elements of play in there.

01:51:42.610 --> 01:51:43.330
And yep.

01:51:43.490 --> 01:51:48.200
And so it it's it's occurring at the highest level of communication, you know, it's in and

01:51:48.200 --> 01:51:51.800
getting incorporated into their repertoire.

01:51:52.040 --> 01:51:52.360
Yeah.

01:51:52.520 --> 01:51:53.160
Yeah.

01:51:54.040 --> 01:51:57.000
Uh do you you have music instruments behind you?

01:51:57.080 --> 01:52:03.000
Do you want to either plug any of your uh existing music or are you gonna after we start

01:52:03.080 --> 01:52:03.160
with it?

01:52:03.320 --> 01:52:06.280
So we mean plug it as in like promote it or plug it as in like plug.

01:52:06.680 --> 01:52:08.280
Well, actually, I meant they were both questions.

01:52:08.360 --> 01:52:10.600
I was gonna say promote, but then I there's a five year old.

01:52:11.800 --> 01:52:12.200
Yeah, exactly.

01:52:12.280 --> 01:52:13.530
It was a it was a mix up, Yeah.

01:52:13.850 --> 01:52:14.810
We will plug your work.

01:52:14.970 --> 01:52:18.330
Uh Greg has some cool music uh by the name of G.

01:52:18.490 --> 01:52:19.210
Bryant.

01:52:19.290 --> 01:52:22.650
Uh he's got an album called Soft Assembly of a Die Hard.

01:52:22.970 --> 01:52:25.050
Um I was just checking it out before we start talking.

01:52:25.130 --> 01:52:25.770
It's good stuff.

01:52:25.930 --> 01:52:31.850
So and if you're listening, if you're listening, do check out the YouTube um uh for

01:52:31.850 --> 01:52:37.770
at least a moment to check out Greg's bass and banjo and um uh hollow body electric

01:52:37.850 --> 01:52:43.420
and there's a strat on the floor and or on the floor and then there's a a lap stop guitar

01:52:43.660 --> 01:52:44.860
right in the corner.

01:52:45.020 --> 01:52:45.660
Yeah.

01:52:46.220 --> 01:52:46.700
Yeah.

01:52:46.940 --> 01:52:47.260
Awesome.

01:52:47.740 --> 01:52:49.100
All kinds of goodies.

01:52:49.900 --> 01:52:54.140
It's uh I'm not too crazy with the guitars.

01:52:54.300 --> 01:52:58.700
I only have really one of everything that uh of its kind, mostly.

01:52:58.860 --> 01:52:59.260
Yeah.

01:52:59.500 --> 01:53:02.010
Um I have a couple basses, I guess.

01:53:02.250 --> 01:53:06.010
But yeah, I'm not like one of these people that has like a whole row of electric guitars

01:53:06.090 --> 01:53:07.210
that are all functionally the same.

01:53:07.290 --> 01:53:09.290
You know, it's just like this one's uh 1970.

01:53:09.610 --> 01:53:10.010
Yeah, yeah.

01:53:10.330 --> 01:53:10.570
Yeah.

01:53:11.370 --> 01:53:13.290
But that's because you still play it, right?

01:53:13.370 --> 01:53:14.730
You you bought it and you play it.

01:53:14.810 --> 01:53:16.330
So you play them all, yes.

01:53:16.730 --> 01:53:17.770
Your parents were wrong.

01:53:18.090 --> 01:53:19.610
And I can I sell stuff too.

01:53:19.690 --> 01:53:21.370
I I unload equipment.

01:53:21.760 --> 01:53:22.240
Very good.

01:53:22.320 --> 01:53:22.640
Very good.

01:53:23.200 --> 01:53:24.960
Well, Greg, uh super cool.

01:53:25.040 --> 01:53:28.080
Um, really encourage everybody to check out uh Greg's work.

01:53:28.160 --> 01:53:33.280
Um, if especially if you're interested in anything acoustic modality, he is the that that

01:53:33.360 --> 01:53:37.040
that is his um he's the like the getaway driver of a heist movie.

01:53:37.120 --> 01:53:42.000
Greg is the um the audio modality um for for this neck of science.

01:53:42.080 --> 01:53:43.520
So uh it's super cool, Greg.

01:53:43.600 --> 01:53:48.770
I always love coming across your work and it's um yeah, it's I always like whenever I see

01:53:48.770 --> 01:53:52.450
your name on a paper, I'm always excited because I know I'm gonna learn something new and

01:53:52.450 --> 01:53:55.810
crazy about nice compliment.

01:53:56.050 --> 01:53:56.930
Well, it's been fun.

01:53:57.010 --> 01:53:57.890
Thanks for inviting me.

01:53:58.050 --> 01:54:03.410
And um yeah, I'm enjoying um the other I've been watching some of these and um I'm enjoying

01:54:03.410 --> 01:54:03.490
them.

01:54:03.650 --> 01:54:04.530
They're good conversations.

01:54:04.610 --> 01:54:05.650
So keep it up.

01:54:05.890 --> 01:54:06.210
Awesome.

01:54:06.290 --> 01:54:06.450
Thanks.

01:54:06.690 --> 01:54:07.170
Thank you.

01:54:10.610 --> 01:54:16.540
David, it's funny to me that uh well, I guess if you listen to loud music, your your

01:54:16.540 --> 01:54:21.580
ears get tired, but just listening to something, you don't feel like your ears get tired.

01:54:21.980 --> 01:54:26.140
Yeah, if you're just listening to speech or to an interview, like city noise, like if

01:54:26.140 --> 01:54:31.020
you're overwhelmed, like you can have audio overstimulation if there's like too much music.

01:54:31.900 --> 01:54:32.300
That is true.

01:54:33.180 --> 01:54:34.620
You need a break, you need a bit of silence.

01:54:34.860 --> 01:54:37.100
Yeah, but it doesn't feel like it's in your ears.

01:54:38.030 --> 01:54:42.190
No, yeah, I wonder it's kind of an interesting psychological question, actually.

01:54:42.270 --> 01:54:45.950
Like, w what is the function of sonic tiredness?

01:54:46.270 --> 01:54:48.750
Yeah, or perceived effort seems to be unique.

01:54:48.990 --> 01:54:54.030
There's certain aspects of perceived effort across different sensation or perception

01:54:54.030 --> 01:54:54.350
systems.

01:54:54.670 --> 01:54:57.390
I mean it might be it might be the opportunity cost thing.

01:54:57.640 --> 01:55:01.720
That like you know, you can only listen or focus on certain sounds to the exclusion of

01:55:01.720 --> 01:55:01.960
others.

01:55:02.200 --> 01:55:07.720
And if you're just like listening to this particular type of sound, you're eventually like

01:55:07.800 --> 01:55:11.320
the opportunity costs get pretty big if like you're not getting any more information from

01:55:11.320 --> 01:55:12.280
that sound stream.

01:55:12.440 --> 01:55:12.920
I don't know.

01:55:13.240 --> 01:55:19.150
Uh but so but it is the idea that like if you're just hearing one thing, Uh then

01:55:19.550 --> 01:55:23.310
there's not competition for your attention, it's less fatiguing.

01:55:23.630 --> 01:55:24.190
Yeah.

01:55:25.870 --> 01:55:26.270
Maybe.

01:55:26.350 --> 01:55:26.670
I don't know.

01:55:26.910 --> 01:55:27.950
Maybe I'm just bullshitting.

01:55:28.030 --> 01:55:28.350
I don't know.

01:55:28.510 --> 01:55:28.670
Yeah.

01:55:28.750 --> 01:55:34.750
Well, a PSA is um if you have um after moving from Germany to

01:55:34.750 --> 01:55:40.110
the US, ever all this loud music and TVs in restaurants and third places.

01:55:40.750 --> 01:55:42.590
I hate loud restaurants, loud bars.

01:55:42.830 --> 01:55:43.310
Oh my god.

01:55:43.720 --> 01:55:46.360
There's a as David says, there's a loneliness epidemic.

01:55:46.520 --> 01:55:47.960
Please just turn it all off.

01:55:48.200 --> 01:55:50.520
Like let people just have conversations, please.

01:55:50.760 --> 01:55:51.560
Don't get me started.

01:55:51.640 --> 01:55:51.880
Ugh.

01:55:51.960 --> 01:55:56.040
And people have parties at these loud bars and you can't even hear anyone talking.

01:55:56.120 --> 01:55:59.080
Like what uh it drives me uh insane.

01:55:59.400 --> 01:55:59.560
Yeah.

01:55:59.960 --> 01:56:04.360
So just just turn everything off and let people interact.

01:56:04.760 --> 01:56:05.320
Yeah.

01:56:06.040 --> 01:56:06.840
Could not agree more.

01:56:07.400 --> 01:56:08.120
Yeah, great.

01:56:08.280 --> 01:56:13.000
Um, I should end here, but but yeah, uh, please do Patreon.

01:56:14.310 --> 01:56:18.470
I mean that was very reductive, but yeah, we we have a Patreon account if you would like to

01:56:18.470 --> 01:56:20.390
um support the podcast.

01:56:20.550 --> 01:56:22.390
Um please visit that.

01:56:22.630 --> 01:56:24.070
What else can people do, David?

01:56:24.390 --> 01:56:26.150
Uh they can leave us a kind review.

01:56:26.230 --> 01:56:29.030
Uh they can say nice things about us on social media.

01:56:29.190 --> 01:56:32.550
They can send us an email if they have any ideas for topics or guests.

01:56:32.790 --> 01:56:33.670
That's pretty much it.

01:56:33.910 --> 01:56:34.790
Yeah that's great.

01:56:34.870 --> 01:56:38.790
And um yeah I just it was really nice to talk to Greg for two hours.

01:56:38.950 --> 01:56:44.660
I wish we could um Oh, I don't know, play out his his

01:56:44.660 --> 01:56:45.620
music, but you know what?

01:56:45.700 --> 01:56:50.260
I I I I don't know if you know this, David, but I've been playing the full um theme song at

01:56:50.260 --> 01:56:52.500
the end of the episodes because it's just so lovely.

01:56:52.820 --> 01:56:53.860
Oh, thank you.

01:56:54.020 --> 01:56:54.500
That's nice.

01:56:54.980 --> 01:56:57.460
There's not enough Glockenspiel in life.

01:56:57.860 --> 01:56:59.060
I could not agree more.

01:56:59.140 --> 01:57:00.820
I'm a big fan of the Glockenspiel.

01:57:01.060 --> 01:57:05.150
Yeah, so sit back, relax, and enjoy the next 30 thirty some sorry, I cut you off about

01:57:05.150 --> 01:57:05.870
Glockenspiels.

01:57:05.950 --> 01:57:06.510
I don't want to cut it down.

01:57:06.670 --> 01:57:10.350
No, I was just gonna say it had a big moment in like the early two thousands, I feel like

01:57:10.350 --> 01:57:11.790
with the Harry Potter theme, you know.

01:57:12.030 --> 01:57:13.390
And then it just kinda disappeared.

01:57:13.470 --> 01:57:14.510
Like what the what's the deal?

01:57:14.750 --> 01:57:15.390
Bring it back.

01:57:15.790 --> 01:57:16.590
It's a beautiful instrument.

01:57:16.670 --> 01:57:22.510
So so that's our second rant here is uh turn everything off in these

01:57:22.510 --> 01:57:26.110
third places and bring back the Glockenspiel.

01:57:26.340 --> 01:57:26.660
Yep.

01:57:26.900 --> 01:57:27.940
Bring it back.

01:57:28.180 --> 01:57:28.340
Yeah.

01:57:28.580 --> 01:57:29.460
See you next week.