Treating robots as social beings - A matter of personal preconceptions or interpersonal alignment?

Abstract

More and more robots with social capabilities like speaking or perceiving users are being developed. However, only little is known about the question if these systems are actually treated in a social manner by their users. In particular, there is a lack of data about how users treat robots when other users are present. Therefore, this paper presents an analysis of users' linguistic behavior in interaction with a social robot. It intends to answer the questions of what users' verbal behavior tells us about their perception of the system and how completing a task in a group with another user influences their behavior towards a robot. In contrast to previous work, the paper does not entirely support the hypothesis that the users' linguistic behavior is mainly determined by their preconceptions. We found that rather the interpersonal alignment between the participants is the main factor influencing the behavior in the interaction.