Selective function of speaker gaze before and during questions

Abstract

This paper presents a method of selecting the answerer from audiences for a museum guide robot. First, we observed and videotaped scenes when a human guide asks visitors questions in a gallery talk to engage visitors. Based on the interaction analysis, we have found that the human guide selects the appropriate answerer by distributing his/her gaze towards visitors and observing visitors' gaze responses during the pre-question phase. Then, we performed the experiments that a robot distributed its gaze towards visitors to select an answerer and analyzed visitors' responses. From the experiments, we have found that the visitors who are asked questions by the robot feel embarrassed when they have no prior knowledge about the questions and the visitor's gaze before and during the question play an important role to avoid being asked questions. Based on these findings we have developed a function for a guide robot to select the answerer by observing visitors' gaze responses.