Flexible prediction of target motion with internal representation in chase behavior

Abstract

Skilled interception behavior often relies on accurate predictions of external objects because there is a large delay in our sensorimotor systems. To deal with the sensorimotor delay, the brain needs to predicts future states of the target based on the current state available, but it is not well understood how we predict the motion of interactive targets such as evasive opponents. Here we conduct one-on-one chase and escape task and estimated the predictive manner of pursuer by analyzing the response behavior to a sudden directional change of evader. In addition, we verified that a prediction as estimated is computationally feasible using neural network models. These suggest that the predictive mechanisms that humans use to compensate for sensorimotor delays during pursuit are more sophisticated than previously thought and that the internal representation acquired through prior experience in similar situations may be useful even in predicting the motion of unknown opponent.