Emotion-driven encoding of music preference and personality in dance

Thirty rhythmic music excerpts were presented to 60 individuals. Dance movements to each excerpt were recorded using an optical motion-capture system, preference for each excerpt recorded on a 5-point Likert scale, and personality assessed using the 44-item version of the Big Five Inventory. From the movement data, a large number of postural, kinematic and kinetic features were extracted, a subset of which were chosen for further analysis using sequential backward elimination with variance inflation factor (VIF) selection. Multivariate analyses revealed significant effects on these 11 features of both preference and personality, as well as a number of interactions between the two. As regards preference, a U-shaped curvilinear relationship between excerpt preference and amount of movement was identified, hypothesized to relate to the role of emotional arousal in guiding music preference and dance moves. As regards personality, a different pattern of movement characteristics was associated with each of the Big Five dimensions, broadly supporting previous work.