Can Human–Robot Interaction Promote the Same Depth of Social Information Processing as Human–Human Interaction?

Abstract

Recent studies on human–robot interactions have suggested that humanoid robots have considerable potential in social cognition research. However, the authors are not aware of any studies regarding social information processing from human–robot interactions. To address this issue, we considered two types of social interaction tasks (initiating and responding joint attention tasks) and two types of interaction partners (robot and human partners). Distinguishing between these types of joint attention (JA) is important, because they are thought to reflect unique but common constellations of processes in human social cognition and social learning. Thirty-seven participants were recruited (Study 1: 20 participants, Study 2: 17 participants) for the current study, and they conducted a picture recognition social information processing task with either robot or human partners. The results of Study 1 suggested that participants who interacted with a humanoid robot achieved a better recognition memory performance in the initiating JA condition than in the responding JA condition. The results of Study 2 suggested that the human–human and human–robot interactions resulted in no quantifiable differences in recognition memory. We discuss the implications of our results for the utility of humanoid robots in social cognition studies and future research questions on human–robot interactions.