hide
Free keywords:
-
Abstract:
How do we control our bodily movements when socially interacting with others? Research on online motor control provides evidence that task relevant visual information is used for guiding corrective movements of ongoing motor actions. In social interactions observers have been shown to use their own motor system for predicting the outcome of another person's action (direct matching hypothesis) and it has been suggested that this information is used for the online control of their social interactions such as when giving someone a high five. Because only human but not non-human (e.g. robot) movements can be simulated within the observer's motor system, the human-likeness of the interaction partner should affect both the planning and online control of movement execution. We examined this hypothesis by investigating the effect of human-likeness of the interaction partner on motor planning and online motor control during natural social interactions. To this end, we employed a novel virtual reality paradigm in which participants naturally interacted with a life-sized virtual avatar. While 14 participants interacted with a human avatar, another 14 participants interacted with a robot avatar. All participants were instructed to give a high-five to the avatar. To test for online motor control we randomly perturbed the avatar's hand trajectories during participants' motor execution. Importantly, human and robot looking avatars were executing identical movements. We used optical tracking to track participants' hand positions. The analysis of hand trajectories showed that participants were faster in carrying out the high-five movements with humans than with robots suggesting that the human-likeness of the interaction partner indeed affected motor planning. However, there was little evidence for a substantial effect of the human-likeness on online motor control. Taken together the results indicate that the human-likeness of the interaction partner influences motor planning but not online motor control.