Abstract
Humans intercept a moving target by walking so as to achieve a constant target-heading angle. Fajen & Warren (submitted) found that participants in the open field do so by changing their direction of travel, at a constant walking speed. They proposed a dynamical model in which turning is controlled by nulling change in the target-heading angle. Chardenon, et al. (submitted) found that participants on a treadmill do so by changing their walking speed, when direction of travel is held constant. They proposed a control law in which speed is varied to maintain a constant target-heading angle. Here we investigate how direction and speed control are integrated by testing fast-moving targets in the open field. One hypothesis is that speed and direction co-vary to achieve a constant target-heading angle; alternatively, direction may be regulated alone until conditions demand an increase in speed. Participants walked in the VENLab, a 12m × 12m virtual environment with a head-mounted display (60 deg H × 40 deg V) and a sonic/inertial tracking system (latency 50 ms). The target appeared in three initial positions (−30? left, 0? straight ahead, 30? right) at a distance of 2 m, and moved on four possible trajectories (0 in walking direction, ±30, ±90, ±120 to the left or right) at two constant speeds (0.9, 1.5 m/s). We analyse participants' paths and the time series of speed and target-heading angle in order to determine the conditions under which direction and speed are changed.
Funded by NIH EY10923, NSF LIS IRI-9720327