Detecting motion trajectories: How do perception and action use visual information?

Abstract

Whenever a person moves to intercept an object, they engage in a complex set of predictions, about the object’s trajectory, and about the set of motions required to intercept it. However, the way that people use perceptual information to intercept rapidly moving objects is currently not well understood. This is because the problem is multifaceted, as there are delays in receptor transduction, neural conduction, processing and muscle activation. There is considerable as to how the two systems interact, there is some evidence that they do (Watamaniuk & Heinen, 2003). In order to assess the differences between trajectory prediction for perceptual judgments and pointing movements we examined participants using the same stimulus, a moving random dot cinematogram (Watamaniuk & Heinen, 1999; Williams & Sekuler, 1984), which was manipulated across conditions. We used a within subjects repeated measures design to compare participants’ performance on two tasks, a perceptual (two alternative forced-choice) task and a pointing task (N = 6). For both tasks we assessed participants’ precision in extrapolating the trajectory of the cinematogram, as well as their response latency. If the two systems use the same visual information, we would expect that precision for each task changes similarly across the conditions. We found similar patterns of error for both tasks, with lower durations and higher bandwidth motion signals displaying greater directional error. This provides further insight into how we use visual information to guide movement. In particular, it provides insight as to how differences in motion perception affects interceptive movements.