Investigating the role of pitch contours in the expression of emotion: Do prosody and music share a code for communicating emotion

There is a long history of attempts to explain why music is perceived as expressive of emotion. Many of the acoustic parameters that are used to communicate emotion in music are also used in human vocal expressions of emotion, suggesting that the acoustic code for communicating emotion may be shared across domains. One of the most important cues for identifying valence in music is the relationship between pitches. For instance, the relationship between two fundamental frequencies may determine the perceived affective valence; the musical interval known as a major third (a frequency ratio of 5:4) generally conveys positive valence, whereas the minor third (a ratio of 6:5) generally conveys sadness. The possibility that these associations are also present in the pitch contours of human vocal expressions has been virtually unexplored. A series of investigations was designed to examine whether the pitch contours of human vocal expressions vary systematically according to the emotional state of the speaker. Bisyllabic speech samples conveying happiness, anger, pleasantness, and sadness were recorded from nine actresses. The semantic content of the vocalizations was controlled across the four emotions. Acoustic analyses of the pitch contours revealed that the relationship between the two salient pitches of the sad speech samples tended to approximate a minor third, which is consistent with the emotional associations in the domain of music. Other patterns were observed for the other emotions. The speech samples were rated for perceived emotion, and the use of various acoustic parameters as cues for the identification of emotion was modeled using regression analysis. The minor third was the most reliable cue for identifying sadness. Other experiments assessed the emotional perception of pitch contours across domains. The results suggest that there are correspondences across domains in the use of pitch contours to encode and decode emotion. These findings support the theory that human vocal expressions and music share an acoustic code for communicating emotion.