Decoding Individual differences and musical preference via music-induced movement

Yudhik Agrawal,Emily Carlson,Petri Toiviainen,Vinoo Alluri

doi:10.1038/s41598-022-06466-3

Abstract

Movement is a universal response to music, with dance often taking place in social settings. Although previous work has suggested that socially relevant information, such as personality and gender, are encoded in dance movement, the generalizability of previous work is limited. The current study aims to decode dancers’ gender, personality traits, and music preference from music-induced movements. We propose a method that predicts such individual difference from free dance movements, and demonstrate the robustness of the proposed method by using two data sets collected using different musical stimuli. In addition, we introduce a novel measure to explore the relative importance of different joints in predicting individual differences. Results demonstrated near perfect classification of gender, and notably high prediction of personality and music preferences. Furthermore, learned models demonstrated generalizability across datasets highlighting the importance of certain joints in intrinsic movement patterns specific to individual differences. Results further support theories of embodied music cognition and the role of bodily movement in musical experiences by demonstrating the influence of gender, personality, and music preferences on embodied responses to heard music.

Highlights

Movement is a universal response to music, with dance often taking place in social settings
Vuong and Atkinson[6] found that observers made reliable judgements about personality from gait cues, these judgements did not always align with the self-reported personalities of the walkers. While it is not yet clear exactly which features of human movement allow for such information to be decoded, computational analysis of complex movement offers a way to explore how information about individual differences are encoded in subtle ways that make it difficult to identify with the naked eye
Gait is probably the most common means of studying individual characteristics as they relate to features of bodily movement, a paradigm in which participants perform free, spontaneous dance movements offers the potential advantage of greater individual variability of movement as well as theoretical connections to a range of psychological and social functions

Summary

Introduction

Movement is a universal response to music, with dance often taking place in social settings. Being able to identify an individual from limited perceptual information has clear evolutionary advantages, in the uniquely social context of early human cultures, where identifying group members and non-members could be necessary for survival[3] Along these same lines, it could be considered adaptive for other information, such as gender, mood state, or individual characteristics such as personality, to be encoded in and identifiable from a person’s bodily movement. Vuong and Atkinson[6] found that observers made reliable judgements about personality from gait cues, these judgements did not always align with the self-reported personalities of the walkers While it is not yet clear exactly which features of human movement allow for such information to be decoded, computational analysis of complex movement offers a way to explore how information about individual differences are encoded in subtle ways (e.g., how different joints move in relation to another in a particular dimension) that make it difficult to identify with the naked eye. Christensen, Cila-Conde and Gomila[17] have proposed six different neural and bio-behavioral functions of human dance, including communication, self-intimation, and social cohesion

Objectives

Methods

Results

Conclusion