Interpreting the Intensity of Vocal Emotions Across Cultures.

Abstract

Intensity is a fundamental dimension of emotions that affects their perception. However, theoretical and empirical studies on intensity, particularly in the vocal domain, remain limited. Furthermore, research on the effects of emotional dimensions (e.g., arousal, valence, and basicness) on intensity ratings remains sparse. This study investigates cross-cultural intensity ratings using the Demo (Dutch) and Koremo (Korean) corpora, each based on eight actors vocalizing eight emotions by pronouncing a pseudo-sentence. Both corpora were rated on intensity by Dutch and Korean listeners. The first goal is to examine the presence of in-group bias in intensity ratings across all responses and correct responses. The second goal is to assess the relative contributions of arousal, valence, and basicness to intensity ratings across all responses and correct responses. To achieve these goals, we conducted an intensity-rating experiment in which all participants rated the intensity of each emotion they perceived on a 4-point scale (1 = low intensity; 4 = high intensity). Participants consisted of 31 native Dutch listeners and 24 native Korean listeners, none of whom had prior knowledge of the other group's culture or language. Our results corroborate earlier findings and shed new light on intensity ratings of vocal emotions. First, contrary to previous findings, we did not find an in-group bias in intensity ratings, such that neither listener group gave higher ratings to emotions produced in their native language than in the unknown language. Second, intensity ratings were higher for high-arousal than for low-arousal emotions, for negative than for positive emotions, and for basic than for non-basic emotions. Notably, intensity ratings are more strongly correlated with arousal and basicness than valence, supporting earlier findings that high-arousal emotions are characterized by increased intensity. Despite the significant effects of arousal, valence, and basicness on intensity ratings, these dimensions do not yield a successful dichotomy of emotions in terms of intensity, since some emotions violate the general patterns of intensity ratings based on these three dimensions. Additionally, intensity ratings were higher for correct than incorrect responses. Together, these findings contribute to a better understanding of the role of intensity in vocal emotion across cultures. Instead of an in-group bias, listeners tended to rate specific emotions as more intense, even when expressed in a non-native language. Although intensity ratings were affected by arousal, valence, and basicness, none of these dimensions provided a strict dichotomy of the eight emotions in intensity, since intensity ratings for certain emotions cannot be reliably predicted from the general patterns. Together, these findings contribute to a better understanding of the role of intensity in vocal emotion across cultures.