Impact of Conversational Robot that Worrying Gestures Behaviors on Human Impression

Abstract

As English proficiency needs rise with globalization, there is increasing interest in using con-(breakpoint)versational robots for language learning. However, typical robots lack natural disfluencies like pauses and gestures. For beginner-level English learners in particular, this lack of human-like behavior can be in-(breakpoint)timidating and detrimental to sustaining engagement during crucial repetitive practice sessions. Without relatable disfluencies, learners may struggle to view the robot as an approachable conversational partner. To create a less intimidating presence for novice learners, the present study aims to enhance the anthro-(breakpoint)pomorphic quality of robot interactions. Specifically, it investigates the robot’s ability to strategically introduce conversational pauses and perform self-adaptor gestures such as fidgeting at predetermined in-(breakpoint)tervals during scripted conversational scenarios. The experimental evaluation revealed that approximately 90% of participants perceived a a heightened sense of human-likeness when the robot exhibited these paus-(breakpoint)ing and worrying gesture behaviors. These results suggest that strategically incorporating naturalistic disfluencies and human-like self-adaptor gestures into a conversational robot’s non-verbal repertoire can increase its perceived anthropomorphism.