The Evolution of Nonhuman Primate Vocalizations: Effects of Phylogeny, Body Weight, and Social Context

Abstract

E. S. Morton proposed that, in birds and mammals, individuals tend to produce low-frequency atonal vocalizations in highly aggressive situations, whereas they typically produce high-frequency tonal vocalizations during nonaggressive or fearful situations. This hypothesis, referred to as the "motivation-structural (MS) rules," is based on two assumptions: the frequency of a vocalization is negatively correlated with body weight, and large animals are dominant over smaller animals, and thus aggressive vocalizations tend to have a lower pitch than fearful vocalizations. The relationship between body weight and frequency is examined using data on 36 nonhuman primate species representing 23 genera and 474 vocalizations. Results show that there is a statistically significant negative correlation between body weight and frequency: larger species produce relatively lower-pitched vocalizations than smaller species. A test of Morton's MS rules provided overall support for the predicted relationship between motivational state and frequency (i.e., high-frequency calls were produced by fearful individuals, and low-frequency calls were produced by aggressive individuals) but no support for the expected relationship between motivational state and tonality. However, the motivational state-frequency pairing was confounded by the fact that some taxonomic groups (Platyrrhini and Catarrhini) showed a much stronger level of association than other groups (Prosimii and Hominoidea). In summary, therefore, the nonhuman primate data provide only partial support for MS rules. At least three factors may have influenced the outcome of the current test. First, in some species, motivational state may be more closely associated with other acoustic parameters than absolute frequency and tonality. Second, the acoustic structure of nonhuman primate vocalizations is, at least in some cases, more closely associated with an external referent than with the caller's internal state. And third, features of the species-typical habitat have had direct selective effects on signal structure, optimizing for effective propagation through the environment.