Are consonant intervals music to their ears? Spontaneous acoustic preferences in a nonhuman primate.

Introduction

Music is among the defining features of human culture, playing a central role in every society known to Western scholars. However, from the standpoint of evolution, music is also one of the most mysterious of human behaviors, as it serves no obvious function that might have driven its evolution. Evolutionary theorists since the time of Darwin have speculated about the adaptive function of music and its evolutionary origins (Darwin, 1871), with little consensus or empirical support.

Recently, however, work on infants and animals has begun to illustrate how empirical evidence might shape theories of music's evolution. In particular, because animals can be tested in the absence of any exposure to music, parallel perceptual abilities in nonhuman animals can help establish whether aspects of our music faculty are innate and therefore candidate products of natural selection. Moreover, as nonhuman animals do not themselves make music, any perceptual effect found in a nonhuman animal cannot be part of an adaptation for music. Music-related experiments on animals are thus poised to play an important role in the debate about the origins of music.

One of the striking and mysterious features of how we experience music and other forms of art is the aesthetic response we often have to what we experience. As is the case for most aspects of music, the function and origins of aesthetic responses are unclear. As a first step in investigating these issues, we studied preferences for relatively simple sounds that lack the complex temporal structure of extended passages of music. Perhaps the best-known example of such acoustic preferences involves harmonic musical intervals. Some combinations of notes tend to sound good, at least to Western listeners, and are termed consonant; others sound bad and are termed dissonant. Pythagoras was the first to note that consonance tends to be generated by pairs of tones whose fundamental frequencies are related by simple integer ratios. Helmholtz later proposed the widely accepted notion that peripheral auditory effects (namely, beating) distinguish consonance and dissonance, and both neurophysiological and behavioral studies in birds, cats, and primates suggest that these peripheral differences are shared across mammals and birds.

These peripheral effects account for the discriminability of consonant and dissonant intervals, but shed little light on the preferences between the two classes of stimuli that are arguably the main reason for their importance in music. Where do such preferences come from? Are they acquired through exposure to music, which perhaps contains more consonant intervals than dissonant ones? Are they part of an adaptation to music? Or might they be a byproduct of some general feature of the auditory system? Experiments in closely related animals, especially primates, can help to clarify these issues.

Humans find some sounds more pleasing than others; such preferences may underlie our enjoyment of music. To gain insight into the evolutionary origins of these preferences, we explored whether they are present in other animals.

We designed a novel method to measure the spontaneous sound preferences of cotton-top tamarins, a species that has been extensively tested for other perceptual abilities. Animals were placed in a V-shaped maze, and their position within the maze controlled their auditory environment. One sound was played when they were in one branch of the maze, and a different sound for the opposite branch; no food was delivered during testing. We used the proportion of time spent in each branch as a measure of preference. The first two experiments were designed as tests of our method. In Experiment 1, we used loud and soft white noise as stimuli; all animals spent most of their time on the side with soft noise. In Experiment 2, tamarins spent more time on the side playing species-specific feeding chirps than on the side playing species-specific distress calls. Together, these two experiments suggest that the method is effective, providing a spontaneous measure of preference. In Experiment 3, however, subjects showed no preference for consonant over dissonant intervals. Finally, tamarins showed no preference in Experiment 4 for a screeching sound (comparable to fingernails on a blackboard) over amplitude-matched white noise. In contrast, humans showed clear preferences for the consonant intervals of Experiment 3 and the white noise of Experiment 4 using the same stimuli and a similar method.

We conclude that tamarins' preferences differ qualitatively from those of humans. The preferences that support our capacity for music may, therefore, be unique among the primates, and could be music-specific adaptations.

Discussions

Preferences for consonance over dissonance are widespread in human adults and have also been demonstrated in human infants. Our results suggest that although such preferences may be innate in humans, they likely have evolved after the divergence point with our primate cousins. It is of course possible that another primate species, more closely related to humans (e.g. chimpanzees), might exhibit more similar acoustic preferences, or that tamarins tested with a different procedure would show a preference. Certain Java sparrows showed preferences for some types of music over others. This preference could conceivably be related to the singing behavior of this species, and it would be interesting to test them with the consonant and dissonant stimuli that we used.

Given the present results, however, we conclude that if humans and nonhuman primates share acoustic preferences for sounds, this capacity evolved more recently than the divergence with New World monkeys such as the cotton-top tamarin (i.e. some 40 million years ago). This conclusion stands in contrast to the many perceptual mechanisms shared between humans and tamarins (and other species as well), particularly with respect to speech perception and presumably also to the discriminability of consonance and dissonance. This contrast raises the possibility that some of the acoustic preferences observed in humans evolved as a specific adaptation for music.