Avoiding accidents at the champagne reception: A study of joint lifting and balancing

Using a lifting and balancing task we contrasted two alternative views of planning joint actions: one postulating that joint action involves distinct predictions for self and other, the other postulating that joint action involves coordinated plans between the co-actors and the reuse of bimanual models. We compared compensatory movements required to keep a tray balanced between a simultaneous joint action condition where two participants lifted glasses from each other's tray and a serial joint action taking where they took turn lifting. Simultaneous joint action made it easier to keep the tray balanced than serial joint action. Thus, in keeping with the bimanual view, predicting the timing of one's own lifting action helped actors compensating for another's lifting action. These results raise the possibility that simultaneous joint actions do not necessarily require distinguishing between one's own and the co-actor's contributions to the action plan and may potentially afford an agent-neutral stance.