Date created: | Last Updated:

Category: Project

Description: People express emotion through vocalizations, facial expressions, and body movements, as well as through abstract forms as in art, architecture, music, and poetry. The structure of these expressions often seems intuitively linked to its meaning: e.g., romantic poetry is written in flowery curlicues, while the logos of death metal bands use spiky script. Here we show that these associations are not arbitrary, but are supported by a supramodal code for emotional arousal. Specifically, the central tendency of the frequency spectrum of a stimulus—its spectral centroid—predicts emotional arousal judgments. We show that the predictive power of the spectral centroid generalizes across sounds, shapes, speech, and human body movements, providing a strong signal that can be used to efficiently estimate an agent’s internal emotional state.