Abstract
To catch a ball, one must move one's hand to the right place at the right time, with the palm oriented roughly orthogonal to the ball's direction of motion. Predictive models of catching are based on information about when the ball will pass within reach, and where it will be and in what direction it will be traveling when it reaches that point. Recent studies suggest that both passing distance and direction of motion-in-depth are overestimated, leading some to conclude that hand movements are not guided by predictive information. However, viewing conditions were restricted and eye, head, and body movements were constrained in these studies, and the perceptual judgment tasks used to measure perceived passing distance and direction of motion-in-depth were unlike natural catching tasks. In this study, we used a natural catching task to measure perceived passing distance and direction of motion-in-depth. Subjects wore a stereoscopic head-mounted display. Head and hand position and orientation were tracked by a motion tracking system. Virtual balls were projected into the air along a parabolic path and passed within reach of the subject. Passing distance and approach angle were manipulated independently. Subjects were instructed to reach out and catch the virtual balls with their right hand, orienting the hand so that the palm was orthogonal to the ball's direction of motion. Analyses focused on the effects of passing distance and approach angle on hand position and orientation. The implications of these findings for predictive models of catching will be discussed.