Using Unobtrusive Wearable Sensors to Measure the Physiological Synchrony Between Presenters and Audience Members

Abstract

The widespread adoption of mobile and wearable devices enables new approaches for the unobtrusive and continuous monitoring of humans' behavior, physiological state, interactions and more. Within this line of research, we focus on the physiological synchrony between a presenter and her audience and investigate whether it can be used to characterize the experience of presenters and audience members during presentations. To this end, we collect data from 17 presenters and six audience members during a two-days conference. For 40, unique presenter-audience pairs we gather electrodermal activity (EDA) signals and self-reports on different aspects of the experience: engagement, immersion and enjoyment/satisfaction. For 28 of these pairs, we also collect inter-beat interval (IBI) traces. We then apply seven approaches for measuring the synchrony of physiological signals and we contextualize these measures using metrics derived from the self-reports. We find that physiological synchrony -- measured using the Dynamic Time Warping algorithm -- can be used as a proxy to quantify participants' agreement on self-reported engagement. Our findings can be used to provide automated presenter-audience feedback in a conference setting and may be applicable in other scenarios, including education (teacher-student), arts (performer-audience), or meetings (presenter-audience).