摘要：We explore dance video clip stimuli as a means to test human observers' accuracy in detecting genuine emotional expressivity in full-body movements. Stimuli of every-day-type full-body expressions of emotions usually use culturally very recognizable actions (e.g. fist shaking for anger, etc). However, expressive dance movement stimuli can be created to contain fully abstract movements. The expressivity results from subtle variations in the body movements of the expressor, and emotions cannot be recognised by observers via particular actions (e.g. fist shaking, etc).Forty-one participants watched and rated 24 pairs of short dance videos –from a published normalised dance stimuli library– in randomised order (N = 48). Of each carefully matched pair, one version of the full-body movement sequence had been danced to be emotionally genuinely expressive (clip a), while the other version of the same sequence (clip b) had been danced –while technically correct– without any emotional expressivity. Participants rated (i) expressivity (to test their accuracy; block 1), and (ii) how much they liked each movement (an implicit measure to test their emotional response (“liking”); block 2).Participants rated clips that were intended to be expressive as more expressive (part 1: expressivity ratings), and liked those expressive clips more than the non-expressive clips (part 2: liking ratings). Besides, their galvanic skin response differed, depending on the category of clips they were watching (expressive vs. non-expressive), and this relationship was modulated by interceptive accuracy and arts experience. Results are discussed in relation to the Body Precision Hypothesis and the Hypothesis of Constructed Emotion.
