Universal and Non-universal Features of Musical Pitch Perception Revealed by Singing
Date Posted:
September 19, 2019
Date Recorded:
September 16, 2019
CBMM Speaker(s):
Josh McDermott Speaker(s):
Nori Jacoby
All Captioned Videos Publication Releases
Description:
Published in Current Biology (09/2019)
[MUSIC PLAYING] [PLAYING FLUTE]
JOSH MCDERMOTT: Perception and cognition result from some sort of complex interplay between our genes and our experience. But in humans, the vast majority of experiments that have been conducted are on members of Western societies because that's where most of the scientists are. The experiences of those kinds of people are relatively homogeneous. We don't know a whole lot about how much variation is possible that might be driven by experiential variation.
NORI JACOBY: Now one domain in which this is extremely prominent is music. Music is present in all known cultures. And this suggests some kind of a biological basis. However, music is also very different from one place to another.
JOSH MCDERMOTT: Now one of the key ingredients of music is pitch. And pitch perception has been extremely well documented in Western listeners. And there's a bunch of salient properties that we now know about. So one of those properties is the fact that pitch seems to have an upper and a lower limit.
NORI JACOBY: Now even though we can hear frequencies that are as high as 20 kilohertz if our hearing is very good, musical pitch perception tends to break down at about 4 kilohertz, which is around the highest tone of the piano.
Now if we play melody within the range of the piano, something like that--
[PLAYS "TWINKLE, TWINKLE LITTLE STAR"]
--it's easy to recognize it. But if using a synthesizer, I will play you a melody outside the range of the piano, higher than 4 kilohertz. Then it will be harder to recognize the melody and the sense of pitch will be less salient.
[PLAYS "TWINKLE, TWINKLE LITTLE STAR"]
So there's something different about the pitch information we can extract from low frequencies--
[PLAYS PIANO]
--compared with high frequencies.
[PLAYS PIANO]
JOSH MCDERMOTT: Now one explanation for that is that there is a phenomenon that happens in the ear known as phase locking. So the action potentials, the spikes that are fired in response to sound, are temporally very precise. But there's a biophysical limitation to the ion channels in the ear that causes phase locking to break down when the frequencies get sufficiently high, typically above about 4 kilohertz in the animal species in which this has been measured. So one classical explanation for the upper limit of pitch is that there is a biological limit on the information that's coming out of the ear.
NORI JACOBY: Now on the other hand, in Western music, there is also a limit to the pitches that are being used. For example, in the Western piano, the highest tone is about 4 kilohertz.
[PLAYS TONE]
JOSH MCDERMOTT: Now, of course, it could be that Western instruments are adapted to the limits of pitch perception. But the other possibility is that our pitch perception is actually adapted to what we hear in Western music. And pitch perception is not as good at very high frequencies because we just don't ever have to extract pitch information from frequencies that are that high. So if you just test Western participants, it's always going to be ambiguous. Now another very salient property of pitch in Western music is that notes that are separated by octaves are treated as equivalent.
NORI JACOBY: A basic feature of Western music is so-called octave equivalence, where tones separate by an octave--
[PLAYS TONES]
--are the same. This is familiar to us from using the same notes' name, for example here, C--
[PLAYS Cs]
--for tones at different registers.
[PLAYS Cs]
Here we have a Western piano. And the pattern of black and white keys are organized in octaves. There's also tons of octaves in Western harmony.
[PLAYS CHORDS]
What's really fascinating is that octaves are also tightly linked to acoustics. So for example, if two strings have the same tension and one is twice as long as the other, then this tone will sound an octave below.
[PLAYS OCTAVE]
JOSH MCDERMOTT: In order to get some insight into the potential role of biological constraints and experience with Western music in these kinds of phenomena, there's a society known as the Tsimane that we've been conducting experiments on for the last eight years.
NORI JACOBY: These villages were geographically remote and from the local cities. And we had to travel there by taking a canoe ride or a Cessna plane or a few hours in a truck. And we found participants that have different exposure compared with the participant in the modern US city.
JOSH MCDERMOTT: They don't have electricity or running water. They'll make occasional trips into nearby towns to trade, but that's pretty much the extent of their contact with the Western world.
NORI JACOBY: The experiment was very simple. We played tones that are well above the singing range, something like that--
[PLAYS HIGH NOTES]
--to the participant. And participant reproduced these tones, singing comfortable in their singing range. So for these tones--
[PLAYS HIGH NOTES]
--the sang something like (SINGING) ha ha.
JOSH MCDERMOTT: So when we do these experiments on Westerners, we find what you would expect, namely that reproductions are good when the frequencies are in sort of the classical range of musical pitch and then deteriorate when the frequencies get too high.
NORI JACOBY: The pitch range of Tsimane music is more limited compared with Western music. So they don't have a piano. And when we've gone to these villages and asked them to show us their instruments and play them to us, we realized that the upper limit that we've ever seen produced is usually under 2 kilohertz.
JOSH MCDERMOTT: All right. So if we really think that the limits of pitch perception are determined by the limits of what you hear in music, then we might expect, if anything, that the upper limit of pitch would be lower in the Tsimane.
[PLAYS HIGH PITCHES ON COMPUTER]
[? AUDIENCE: ?] [SINGS PITCHES]
[PLAYS HIGH PITCHES]
[? AUDIENCE: ?] [SINGS PITCHES]
[PLAYS HIGH PITCHES]
AUDIENCE: [SINGS PITCHES]
JOSH MCDERMOTT: However, when we do these experiments and we measure the accuracy of the reproductions, instead we actually find that the dependence on frequency is actually remarkably similar. So as you can see in the graph, their overall accuracy is a little bit lower. But the dependence of accuracy on frequency is actually quite similar to what you see in Westerners.
So that suggests that there is some non-musical constraint that's actually driving this phenomenon, potentially some biological limit on the upper limit of phase locking. Now to assess octave equivalence, we can take the exact same experiment and analyze it in a different way.
NORI JACOBY: We found that US participants try to also create the relation, an octave relation, between the tone that they heard and the tone that they sang back.
JOSH MCDERMOTT: So as to replicate what we call the chroma, or the letter name of the note. So if you play them an A in an upper register, they'll tend to sing back in A in their singing range.
[PLAYS NOTES]
[? AUDIENCE: ?] [SINGS PITCHES]
[PLAYS NOTES]
[? AUDIENCE: ?] [SINGS PITCHES]
JOSH MCDERMOTT: So this phenomenon is extremely pronounced in Western musician participants. It's reduced but still very salient in Western non-musicians. And the question is, what was going to happen in the Tsimane. And as you can see in this graph, chroma matching is absent in the Tsimane. So even though they accurately reproduce the direction and to some extent the interval between the two notes, the absolute pitches that they produce are not related to the absolute pitches of the stimulus.
NORI JACOBY: This suggests that octave equivalence is culturally contingent.
JOSH MCDERMOTT: So these experiments have given us new insights into the origins of different aspects of musical pitch perception. And to me what's most exciting about them is that they reveal dissociations between different parts of pitch perception that are pretty tightly linked in Westerners. And so I think they really suggest that pitch perception is a lot more interesting than you might expect. There's a whole bunch of different facets to it. And those can be teased apart if you respect the diversity of human experience.
Associated Research Module:
have an interactive transcript feature enabled, which appears below the video when playing. Viewers can search for keywords in the video or click on any word in the transcript to jump to that point in the video. When searching, a dark bar with white vertical lines appears below the video frame. Each white line is an occurrence of the searched term and can be clicked on to jump to that spot in the video.
