The role of body and tool-based information in joint action coordination

Influential theories (eg common coding theory) suggest that visual body information about the interaction partner is critical when two people physically interact (joint action). Evidence supporting this idea comes from studies investigating computer mediated joint actions. Surprisingly, this hypothesis has not been examined in a real-life joint action task. To test this hypothesis, participants cooperatively played table tennis in complete darkness with the ball, net, and table illuminated with fluorescent markings (Experiment 1). In different conditions we added different types of visual information by placing glowing point markers on the players' body (body information) and/or marking the paddle's rim with fluorescent paint (tool information). We found a significantly reduced error rate when body and/or tool information was available. In Experiment 2 the same types of visual information were available to only one player at a time. The error rate was measured in every condition for each player separately. We found improved performance when the player saw his/her own body or the other player's tool but not when he/she saw his/her own tool or the other player's body. These findings challenge the view that information about the body of an interaction partner is critical in a joint table tennis task