Perception and prediction of simple object interactions

Abstract

For humans, it is useful to be able to visually detect an object's physical properties. One potentially important source of information is the way the object moves and interacts with other objects in the environment. Here, we use computer simulations of a virtual ball bouncing on a horizontal plane to study the correspondence between our ability to estimate the ball's elasticity and to predict its future path. Three experiments were conducted to address (1) perception of the ball's elasticity, (2) interaction with the ball, and (3) prediction of its trajectory. The results suggest that different strategies and information sources are used for passive perception versus actively predicting future behavior.