Summary Some combinations of musical notes are consonant (pleasant), while others are dissonant (unpleasant), a distinction central to music. Explanations of consonance in terms of acoustics, auditory neuroscience, and enculturation have been debated for centuries [1-12]. We utilized individual differences to distinguish the candidate theories. We measured preferences for musical chords as well as nonmusical sounds that isolated particular acoustic factors – specifically, the beating and the harmonic relationships between frequency components, two factors that have long been thought to potentially underlie consonance [2, 3, 10, 13-20]. Listeners preferred stimuli without beats and with harmonic spectra, but across over 250 subjects, only the preference for harmonic spectra was consistently correlated with preferences for consonant over dissonant chords. Harmonicity preferences were also correlated with the number of years subjects had spent playing a musical instrument, suggesting that exposure to music amplifies preferences for harmonic frequencies because of their musical importance. Harmonic spectra are prominent features of natural sounds, and our results indicate they also underlie the perception of consonance.

Melodies, speech, and other stimuli that vary in pitch are processed largely in terms of the relative pitch differences between sounds. Relative representations permit recognition of pitch patterns despite variations in overall pitch level between instruments or speakers. A key component of relative pitch is the sequence of pitch increases and decreases from note to note, known as the melodic contour. Here we report that contour representations are also produced by patterns in loudness and brightness (an aspect of timbre). The representations of contours in different dimensions evidently have much in common, as contours in one dimension can be readily recognized in other dimensions. Moreover, contours in loudness and brightness are nearly as useful as pitch contours for recognizing familiar melodies that are normally conveyed via pitch. Our results indicate that relative representations via contour extraction are a general feature of the auditory system, and may have a common central locus.

Despite its central role in music, the origins of the preference for consonant over dissonant intervals remains controversial. We applied an individual differences approach to this question. Subjects were first tested for their aversion to pure-tone beats, and to inharmonic spectra of single complex tones. Consistent individual differences were observed in the magnitude of subjects’ aversion to beats and inharmonicity, but the two effects were uncorrelated across subjects, suggesting independent factors. Subjects then rated the pleasantness of musical intervals and three-note chords. If particular acoustic factors determine consonance, stronger sensitivity to those factors should predict stronger musical preferences. In each of two cohorts, only the preference for harmonic spectra was consistently correlated with preferences for consonant over dissonant chords and intervals. Although the strength of a subject’s consonance preference was also correlated with their musical experience, so was their harmonicity preference, which explained most of the effect of experience on consonance ratings. The results suggest that much of consonance perception is rooted in the basic acoustic property of harmonicity. Harmonicity is evidently predictive of important structures in Western music, and with exposure to music, listeners learn to attend to it. [Work supported by NIH grant R01DC05216.]