Search for a command to run...
In witnessing face-to-face conversation, observers perceive authentic communication according to the social contingency of nonverbal feedback cues ('back-channeling') by non-speaking interactors. The current study investigated the generality of this function by focusing on nonverbal communication in musical improvisation. A perceptual experiment was conducted to test whether observers can reliably identify genuine versus fake (mismatched) duos from musicians' nonverbal cues, and how this judgement is affected by observers' musical background and rhythm perception skill. Twenty-four musicians were recruited to perform duo improvisations, which included solo episodes, in two styles: standard jazz (where rhythm is based on a regular pulse) or free improvisation (where rhythm is non-pulsed). The improvisations were recorded using a motion capture system to generate 16 ten-second point-light displays (with audio) of the soloist and the silent non-soloing musician ('back-channeler'). Sixteen further displays were created by splicing soloists with back-channelers from different duos. Participants (N = 60) with various musical backgrounds were asked to rate the point-light displays as either real or fake. Results indicated that participants were sensitive to the real/fake distinction in the free improvisation condition independently of musical experience. Individual differences in rhythm perception skill did not account for performance in the free condition, but were positively correlated with accuracy in the standard jazz condition. These findings suggest that the perception of back-channeling in free improvisation is not dependent on music-specific skills but is a general ability. The findings invite further study of the links between interpersonal dynamics in conversation and musical interaction.