Listening to the Ensemble of Story-Telling Performance and Audience Reactions

In vaudeville settings, story-telling performers and their audiences mutually depend on each other. While the audience often smiles in response to incongruent lines and interpreted gestures, the performer sometimes delivers the punch lines only after audience-initiated smiles and movements. However, these temporal structures are too complex to capture directly. In this study, I have proposed a rigid method of formulation and a method of listening to the implicitly organized temporal patterns between the performance and audience reactions. A professional Rakugo storyteller performed live in a laboratory in front of 20 audience members aged 16 to 67 (M = 40.6, SD = 16.4). The qualitative observation of motions and voices were converted to quantitative intensities by using computer programs. I also detected where each audience member concentrated his or her attention, by focusing on eyeblinks, and assigned the identical musical pitch to the reactions of the same person. As a result of mixing these data sequentially along with the time series, an actual musical score was composed. This music provided rich data in which we could easily recognize coordination patterns in the vaudeville setting as a temporal gestalt. The synchronization of reactions among audience members was represented as forming a chord and time-delayed coordination between performer and audience, which sounded like interchanges. It is suggested that the musical perspective introduced in this study could lead to the development of assessment techniques for a variety of orators that regularly speak in front of audiences.