Implications of Robot Actions for Human Perception. How Do We Represent Actions of the Observed Robots?

Abstract

Social robotics aims at developing robots that are to assist humans in their daily lives. To achieve this aim, robots must act in a comprehensible and intuitive manner for humans. That is, humans should be able to cognitively represent robot actions easily, in terms of action goals and means to achieve them. This yields a question of how actions are represented in general. Based on ideomotor theories (Greenwald Psychol Rev 77:73–99, 1970) and accounts postulating common code between action and perception (Hommel et al. Behav Brain Sci 24:849–878, 2001) as well as empirical evidence (Wykowska et al. J Exp Psychol 35:1755–1769, 2009), we argue that action and perception domains are tightly linked in the human brain. The aim of the present study was to examine if robot actions would be represented similarly, and in consequence, elicit similar perceptual effects, as representing human actions. Our results showed that indeed robot actions elicited perceptual effects of the same kind as human actions, arguing in favor of that humans are capable of representing robot actions in a similar manner as human actions. Future research will aim at examining how much these representations depend on physical properties of the robot actor and its behavior.