Search for: All records

In this paper, we argue in favor of creating robots that both teach and learn. We propose a methodology for building robots that can learn a skill from an expert, perform the skill independently or collaboratively with the expert, and then teach the same skill to a novice. This requires combining insights from learning from demonstration, human-robot collaboration, and intelligent tutoring systems to develop knowledge representations that can be shared across all three components. As a case study for our methodology, we developed a glockenspiel-playing robot. The robot begins as a novice, learns how to play musical harmonies from an expert, collaborates with the expert to complete harmonies, and then teaches the harmonies to novice users. This methodology allows for new evaluation metrics that provide a thorough understanding of how well the robot has learned and enables a robot to act as an efficient facilitator for teaching across temporal and geographic separation. 

In this experiment, we investigated how a robot’s violation of several social norms influences human engagement with and perception of that robot. Each participant in our study (n = 80) played 30 rounds of rock-paper-scissors with a robot. In the three experimental conditions, the robot violated a social norm by cheating, cursing, or insulting the participant during gameplay. In the control condition, the robot conducted a non-norm violating behavior by stretching its hand. During the game, we found that participants had strong emotional reactions to all three social norm violations. However, participants spoke more words to the robot only after it cheated. After the game, participants were more likely to describe the robot as an agent only if they were in the cheating condition. These results imply that while social norm violations do elicit strong immediate reactions, only cheating elicits a significantly stronger prolonged perception of agency.