The indigenous tribe helping scientists study singing

We’re thinking about the variety of ways our brains interpret and respond to music. To start us off, we need to know what the term music actually describes, something that proves more difficult than you might have assumed. To help, James Tytko caught up with Malinda McPherson of the University of California, San Diego, while she was paying Cambridge a visit...

James - We're here at Churchill College, and I'm delighted to be joined by Malinda McPherson who studies auditory cognition. So you're a scientist, Malinda, but you came at this discipline from a background in music. Can you tell me a bit about that?

Malinda - Thank you so much for having me. I started playing piano when I was around six years old and, when I was eight, I picked up the violin and the viola. And my entire life since then I have been a musician; I still play regularly. I think it inspired me, when I was starting out in university, to have this broader interest in just how we hear and how we make sense of the sonic world around us. And so while I don't play professionally (I just play for fun, now) it definitely does inspire some of my scientific questions and keeps me in wonder at how we actually can hear this complicated world around us.

James - We're gathered here to talk broadly about the mind and music and how the two relate. Physiologically, people might be aware that sound travels in waves, it hits our ears, vibrates through the various anatomy in there, reaches our brains. But what I'm interested in for this conversation is how our brain interprets certain sounds as musical. What can you tell me about that?

Malinda - That's really where the most interesting questions lie: how our brains make sense of this pretty chaotic world around us most of the time. And one of the really interesting things about music is we often feel like we know it when we hear it. There has been some recent neuroscientific evidence suggesting that there are actually separate neural pathways in our brain that process sounds like speech and sounds like music, but when we try to define music, we actually run into some really interesting problems. Most definitions of music are along the lines of 'organised sound' or 'sound organised in time.' If you think about that definition just a little bit harder, that can describe speech, it can describe almost any other sort of sound that's made by humans. Even tapping your fingers on a table: that's organised sound, but would we consider that music? Maybe not. There's some really beautiful auditory illusions that can also demonstrate that we can hear the same exact sound and in some cases perceive it as speech and in other cases perceive it as music.

James - You mentioned an illusion and it strikes me as one I might like to demonstrate on our audience. Can you explain how it works?

Malinda - So this illusion is called the speech to song illusion, and it was discovered by Diana Deutsch who's a professor emerita at UC San Diego in the US and what actually happened, the story behind this, which she tells quite delightfully in some places, is she was actually recording some audio for a book and she had it on loop. She walked away to the kitchen to grab some tea and all of a sudden she started hearing somebody singing, and it was her own audio. What happens is, if you listen to this sentence that she says multiple times, eventually it starts to sound like she's singing. What's amazing is that the first time you hear it, it sounds like she's speaking, and then at some point after repetitions, almost everybody, it doesn't work on absolutely everybody but it works on many people, a couple repetitions in, 5, 10 repetitions, you start to hear her as singing.

Diana - The sounds as they appear to you are not only different from those that are really present, but they sometimes behave so strangely as to seem quite impossible. But they sometimes behave so strangely. They sometimes behave so strangely. Sometimes behave so strangely. Sometimes behave so strangely. Sometimes behave so strangely. Sometimes behave so strangely…

Malinda - And then if you listen to the full sentence back, it'll sound like she's speaking. And halfway through the sentence, she just breaks out into song as she reaches this passage that you've heard on repeat.

Diana - The sounds as they appear to you are not only different from those that are really present, but they sometimes behave so strangely as to seem quite impossible.

Malinda - And so it's a beautiful example of the fact that even the exact same sound wave, the exact same sound, hitting your ear and hitting your brain, will sound like speech in one context and music in another. And it really beautifully demonstrates that music is in the ear of the beholder. The amazing thing about this illusion is, once you hear it, you will never unhear it. Anytime you hear this sentence again, you'll hear her singing.

James - That was my question. Once I listen to this, are my ears contaminated forever? Or perhaps after a span of time I would go back to hearing the speech?

Malinda - I'll tell you, I think I first heard this illusion about 10 years ago and every time I hear the sentence it sounds like she's bursting into song. So I wouldn't call it a contamination, I'd call this a pretty good illusion..

James - Fair enough. It's interesting you mention that it is such a universal and fundamental part of being human. Why is that? What is the evolutionary function of music?

Malinda - You know, this is an area of active debate in the field and it's an area of active study as well. There are many theories for why we have music. Because we find it in every culture, it does seem like it's something that has had evolutionary pressures on it, that it's an evolved part of human behaviour. Most of the theories revolve around the idea that it's used for social cohesion. Think about if you're in a crowded room and everyone is speaking at the same time: speech loses its meaning if everyone's talking over each other. You have to have this conversation back and forth when you're speaking. Music on the other hand often gains by the participation of more people. You can have a group all doing the same thing together in a coordinated fashion in music, and you can't really do that in the same way in speech. So many of the theories about why we have music and why it's so important and why it's found in every culture have to do with the fact that it seems to have this social function. So whether it's the function of a mother communicating with their baby and trying to sing a song to get them to feel a certain way, or if you're at a social function like a wedding or a funeral or these sort of key moments in our lives, there's always music that brings us together. And that seems to be maybe not fully universal, but pretty close to a universal function of music. So we can think of it in this sort of functional capacity.

James - You're someone who's obviously deeply appreciative of the western classical music tradition. Would a different human race have developed a musical tradition that sounds just so utterly removed from what we've come to appreciate as music today?

Malinda - So anyone who's listened to music from around the world just knows how strikingly different music sounds depending on where it's from. But the amazing thing is that there actually are some seeming universals. There is some structure in music around the world that seems to be shared. I actually study musical harmony, and I do a little bit of work cross-culturally examining how harmony is perceived. And we find that there actually is a little bit of similarity in music and music perception around the world. People do seem to perceive musical intervals in similar ways and people do seem to default to musical ratios. So the ratios between intervals are low integers. In our music we'll have an octave relationship and that's a one to two integer ratio and frequency, or a perfect fifth. So that's the beginning of the melody, twinkle, twinkle little star. That's a two to three integer ratio. And that actually relates to the structure of many natural sounds we hear. Those integer ratios are actually found within the structure of my voice as I speak. And so it does seem that there are these potential universals of perception and perceptual biases that all humans will share. And it seems like that can bias the production and music around the world to make music not similar necessarily, but built on some of the same building blocks because we're all biased in the same ways because we all have the same auditory systems, we all have the same types of brains to process that sound and that leads many musical systems around the world to actually have the same sorts of biases and structure, even though that can then end up sounding quite different and people can have very different preferences for music.

James - And that was the subject of your research recently in Bolivia to try and drill down into those universal qualities. What was that like? What did you learn?

Malinda - So I've been working in Bolivia for the past about five years. I worked with an indigenous population called The Tsimane' who live in a rural area of the Amazon. They are a really wonderful group to work with for many reasons, but one of the reasons that makes them so fascinating to us from a musical perception perspective is that, as far as we can tell, they don't have group musical performance. So I've talked about music as this social cohesion, the possibility that that's why we have it, but for them music is more individual. People sing alone, they don't necessarily sing in groups. People won't just sing by themselves off lonely, they will sing in front of other people, but they don't have what we would consider musical harmony as part of their music. And so we've been looking at the phenomenon of musical harmony, specifically consonance and dissonance, where in Western music we have these intervals that sound really pleasant. And then other intervals that sound unpleasant or dissonant, we can just measure preferences in people, anyone off the street in the UK and the US where I've done work, and they'll show this preference. But if we test Tsimane' with the same type of experiment and ask them whether they like these consonate versus dissonant intervals, they don't show a preference. They don't prefer one to the other. But we've done some experiments to suggest they do hear the difference. They know that these two types of intervals are different and, mathematically, consonate intervals are related by lower integer ratios compared to dissonant intervals, but it seems like they have the same perceptual system that we do as you would expect, they hear these sounds in the same way, but their musical culture is different. And so they don't have the aesthetic connotations that we have from listening to the music that we've listened to.

James - Thank you to Malinda McPherson from the University of California San Diego for sharing that illusion along with the rest of her insights.