Abstract
Navigating through complex, dynamic environments requires people to select actions and guide locomotion relative to moving objects. When people intercept moving targets on foot, their behavior is consistent with a constant bearing angle (CBA) strategy, according to which interception is achieved by moving so as to keep the target at a fixed bearing angle. However, the CBA model does not explain how direction and speed of locomotion are coordinated during interception, nor does it account for the ability to perceive whether interception is even possible given the person's movement capabilities. In this study, we introduce an alternative to the CBA model that offers a starting point for addressing these limitations and accounts for findings from previous studies that were difficult to reconcile with the CBA model. The new model is based on eyeheight-scaled information that specifies the locomotor speed (in units of eyeheights per second) needed to intercept the target as a function of the direction in which one intends to move. In Experiment 1, we tested the predictions of the new model by manipulating eyeheight, which should affect walking trajectories if subjects rely on eyeheight-scaled information but not if they use a CBA strategy. Subjects walked in a virtual environment viewed through a head-mounted display to intercept targets that appeared on the left or right side and moved inward. Target speed and trajectory varied across trials. Analyses focused on walking speed and direction in the control and reduced-EH conditions. In Experiment 2, we tested another prediction of the new model–that the perceived walking speed needed to intercept the target should be affected by manipulations of visual self-motion information. This prediction was tested by manipulating subjects' movement relative to the stationary background in the virtual environment without affecting their movement relative to the moving target.