Exploring Different Types of Feedback in Preschooler and Robot Interaction

Abstract

This study considered the feedback of a robot during second language tutoring. Traditionally, robots are programmed to provide feedback as teacher; we propose a robot that acts as a peer to motivate preschoolers during the tutoring. We conducted an experiment with 65 preschoolers (M = 3.6 years) in which the robot varied feedback in three conditions: peer-like (explicit negative), adult-like (explicit positive and implicit negative) and no feedback. The results suggest that feedback did not influence children's engagement (measured via eye-gaze), although children who received peer-like feedback seemed to perform more independently during the learning task (requiring less interventions from the experimenter).