An empirical study of the effect of acoustic-prosodic entrainment on the perceived trustworthiness of conversational avatars

Abstract

Entrainment is the tendency of interlocutors to become more similar to each other in their way of speaking. This phenomenon has been repeatedly documented and is associated with multiple social aspects of human-human conversations. However, there is a dearth of research on the effects of spoken dialogue systems (SDSs) with implemented acoustic-prosodic (dis)entrainment policies. The goal of the present work is to provide further empirical evidence on how acoustic-prosodic (dis)entraining policies affect users. In particular, this article focuses on its effects on users’ trust toward the SDSs. In the experiments reported here we analyze if and how different acoustic-prosodic (dis)entrainment policies affect users’ perception of a system's ability. We collected data from 98 unique participants, all native speakers of Argentine Spanish. Our results suggest that acoustic-prosodic (dis)entrainment in spoken dialogue systems is effectively associated with the way users perceive the capabilities of such systems. Characterizing these effects remains a challenging task. Overall, we observe a positive effect on trust of entrainment on intensity and a negative effect of entrainment on pitch. Estimated effect sizes are far from negligible.