Spatio-temporal judgements of observed actions : Contrasts between first- and third-person perspective after motor priming.

When observing actions, motor simulation processes aid the prediction and understanding of future events. A central issue concerns whether such action simulation serves social functions of interpreting other people, where performance is predicted to be better when third-person perspective (3PP) actions are viewed; or whether it is most beneficial to guide self actions, whereby the first-person perspective (1PP) would be advantageous. We show that in a spatio-temporal judgement task there is an advantage for the prediction of 1PP. However, this is only detected after motor priming whereby participants perform the observed actions prior to making spatio-temporal judgements. The results, firstly, confirm that we draw on our motor experience for the accurate simulation and prediction of action. Secondly, the results suggest that such experience facilitates more accurate state estimation for actions perceived in the 1PP which map more closely onto visual input of self-generated action. More forward prediction error is retained for 3PP viewed actions, which may however have the benefit of compensating for the uncertainty involved in interacting with others.