Abstract
A variety of creatures ranging from predators in the wild to humans on the playing field repeatedly demonstrate the ability to intercept moving targets. Behavioral studies suggest that humans intercept moving targets by maintaining a constant bearing angle; that is, by keeping the target in a fixed exocentric direction. In the majority of these studies, the targets moved along linear paths at constant speeds. By comparison, targets in the real world often change directions and speeds in ways that range from completely predictable to completely unpredictable. Our experiment was designed to investigate the “in between” case in which targets change speeds in a semi-predictable manner. Subjects sat in front of a large projection screen and watched computer generated displays that simulated linear self-motion over a textured ground plane. Simulated speed was controlled by adjusting a foot pedal, the position of which was mapped onto speed according to a first-order lag. A spherical target approached the subject's path from one of three angles. Between 2.5 and 3.25 s after the trial began, target speed changed by an amount that was taken from a normal distribution of speed changes. The mean of the distribution was positive such that target speed usually increased, but occasionally decreased. Subjects' behavior was compared to that of an ideal pursuer, who moved in such a way as to maximize the likelihood of being able to intercept the target by anticipating the most likely change in target speed. Subjects' behavior was similar to that of the ideal pursuer, indicating that they adopted a strategy that allowed them to perform at near optimal levels. Differences between behavior of subjects and the ideal pursuer were also found under some conditions. Our findings provide insight into the strategies that humans adopt to deal with uncertainty in realistic interception tasks.