Effect of Expressive Lights on Human Perception and Interpretation of Functional Robot

Abstract

Because appearance-constrained robots lack expressiveness, human users often find it hard to understand their behavior and intentions. To address this, expressive lights are considered to be an effective means for such robots to communicate with people. However, existing studies mainly focus on specific tasks or goals, leaving the knowledge of how expressive lights affect people's perception still unknown. In this pilot study, we investigate such a question by using a Roomba robot. We designed two light expressions, namely, green and low-intensity (GL) and red and high-intensity (RH). We used open-ended questions to evaluate people's perception and interpretation of the robot, which showed different light expressions as a way to communicate. Our findings reveal that simple light expressions can allow people to construct rich and complex interpretations of a robot's behavior, and such interpretations are heavily biased by the design of expressive lights.