Influence of Robot Roles on Self-Review

Abstract

In recent years, a considerable number of studies have been made on learning with social robots as partner for learners. In these studies, robots are often assigned their roles in advance such as tutor, peer, or learners themselves. On the other hand, the roles assigned to the robots could give learners a cognitive bias to their own thoughts, emotions, attitudes, interaction ways, etc., related to learning. However, such role influence has not been sufficiently explored so far. In this work, we investigate this in the context where learners self-review their own presentation with a social robot. We have conducted a case study with the presentation robot we developed in our previous work. The robot reproduced learners’ presentation with two roles that were their peer and themselves. The results suggest that the role of robot as peer brings about more effective self-review in regard to reducing uncomfortableness in self-review and promoting awareness of points to be improved in presentation than the role as learners themselves.