Conference Paper Published

Abstract

Musical training enhances emotion recognition based on speech prosody, but it remains unclear whether it also improves emotion recognition through semantics and enhances the capacity to regulate emotional recognition when prosodic and semantic cues conflict. To address these questions, this study compared 23 well-trained musicians with 20 non-musicians using the Stroop paradigm. They were presented with disyllabic Mandarin words that were either positive or negative in semantics and spoken in positive (angry) or negative (happy) prosody. They were asked to identify either the prosodic (the prosodic task) or semantic (the semantic task) emotion of the words, while ignoring the other dimension. The effects of musical training were observed in reaction time. Musicians were faster than nonmusicians in the perception of prosodic emotions. Surprisingly, musicians were also faster in the perception of semantic emotion. Besides, musicians performed better in both congruent and incongruent conditions. However, no significant interaction was found between group and congruency, indicating comparable levels of regulation between musicians and non-musicians. The findings suggested that musical training is associated with cross-domain benefits to emotion recognition in speech, and music and emotional processing in speech might engage shared resources.

Keywords