The Impact of a Social Robot's Attributions for Success and Failure in a Teachable Agent Framework

Teachable agents foster student learning by employing the learning by teaching paradigm. Since social factors influence learning from this paradigm, understanding which social behaviors a teachable agent should embody is an important first step for designing such an agent. Here, we focus on the impact of causal attributions made by a teachable agent. To obtain data on student perceptions of agent attributions, we conducted a study involving students interacting with a social robot that made attributions to ability and effort, and to the student, itself, or both. We analyzed data from semi-structured interviews to understand how different attributions influence student perceptions, and discuss design opportunities for manipulating these attributions to improve student motivation.