Volume 94, No.3, May-June 2008

Duke Magazine-In Search of Music's Biological Roots by Ker Than
Music in speech. Frequency ratios between the first two formants or areas of high energy
Figure 2. Music in speech. Frequency ratios between the first two formants or areas of high energy produced when people make certain vowel sounds. For example, when you say the “o” (a) sound in “bod,” the frequency ratio between the first two formants (F1 and F2) matches a major sixth —the distance between C and A, as indicated on the piano keys. The vertical axis of the graphs shows loudness represented in decibels. The horizontal axis shows frequencies in hertz.
Purves Lab

Purves thinks that human speech can explain more than just relative consonance. It might also hold the key to explaining music's most mysterious property, the one that makes it enchanting even to people with no musical training.

Much of music's power lies in its ability to communicate without words, to speak directly to our emotions. Melodies can etch themselves into our brains and remain for a lifetime. Songs can break your heart, or help to mend it. "We're all familiar with the fact that music has an emotional impact," says Purves. "That's one of the reasons we like it. It generates different emotional responses, and producing those responses is clearly the goal of a lot of musical compositions."

In particular, music composed in major scales sounds bright, spirited, and happy, while minor scale music tends to sound sad, lugubrious, and dark. Musicians have known about and used these relationships for centuries to great effect, but there is no consensus about why major and minor tone combinations evoke the emotions that they do.

In a study currently under way, the researchers are testing the hypothesis that when people talk in a happy way, the formant relationships in their vowels correspond with major keys. And when they talk in a bored, neutral, or sad way, their formant relationships are minor. "Whenever we've heard happy speech, we've tended to hear major-scale tonal ratios," Purves says. "Whenever we've heard sad speech, minor tones tend to be involved.

"We have thus been making those associations since the day we were born. Perhaps when we hear music in a major scale, we unconsciously associate it with happy speech and tend to have that emotional response, and vice versa for music in a minor scale."

Dale Purves' musical research is generally consistent with other work on human perception he has conducted in an attempt to understand vision (see Duke Magazine, July-August, 2000). Purves has long argued that when we see, our brain is not so much analyzing the present as it is constructing a perception based on past experiences.

Dissatisfied with conventional explanations for how vision works, Purves hypothesized that the properties of vision must somehow be shaped by the world-that through evolution, vision, and perception in general, humans must have adapted to the environment we live in. "We need to understand the environment in which we have to make our living, or we won't be making that living for very long," Purves says.

The evidence that vision works in this counterintuitive way is most apparent in visual illusions, in which discord exists between how people perceive the world and how the world really is. A good example is the standard "brightness contrast" illusion found in psychology textbooks. When pictures of two identically shaded tiles are placed against different shades of gray, people see the tile on the dark background as lighter than the tile on the lighter background. Many scientists explain these illusions as perceptual errors made by an otherwise well-functioning visual system. Purves hypothesized "that they were not in fact mistakes, but correct perceptions if you understood what the visual system is actually trying to do," Schwartz says.

Purves' alternative explanation is based on a long-recognized problem with the visual and other sensory systems, including auditory. Any aspect of a given sensory stimulus, such as the amount of light coming to the eye from its surface in the tile example, can arise from an infinite number of real-world scenarios. For example, our eyes receive exactly the same physical stimulus from a highly reflective surface in weak lighting and a dull surface in stronger lighting.

So how does the brain distinguish between the two real-world scenarios and respond appropriately? Purves and his collaborators argue that the visual part of the brain generates perceptions on the basis of what a given stimulus-such as an image on the retina-has signified in the past.

According to this view, humans and other visual animals do not see the world as it really is. They see it through the filters of their sophisticated sense organs and brains, and they also see it through the distorted lens of experience, both their own and that of their species. Understood in this way, visual illusions are not perceptual errors on the part of our visual system, but correct perceptual decisions made in unnatural settings. "What are termed perceptual errors or illusions are in fact evidence of just how sophisticated the visual system is," Schwartz says.

Purves' musical research extends his theory to the auditory system because, here too, experience plays a critical role: Our exposure to speech has shaped our preferences for the kinds of sounds that we like to hear.

A major implication of his research is that music is not an abstract phenomenon explained by mathematical formulas, neither is the human love of music a cosmic coincidence begging for a mystical explanation. It is a wondrous byproduct of evolution.

"Pythagoras wanted to explain music in mathematical ratios. That just doesn't work," Purves says.

"Music is far more complex than Pythagoras. The reason doesn't have to do with mathematics. It has to do with biology."

• return to page one of this article.