Abstract
Purpose: A point on the rim of a rolling wheel traces a cycloid. We ask how accurately subjects perceive the speed of a moving stimulus that travels along a cycloidal path. Methods: Stimuli were moving white spots (0.3 deg, 5 cd/m2) on a dark background. Test stimuli were spots moving as though attached to the rim, to the axle, or on a spoke of an invisible wheel of 3.7 ‖ dia. The angular velocity of the wheel varied from 2 to 10 rad/sec. Subjects compared the forward speed of the invisible wheel (the speed of the linear component of the motion) to that of a comparison stimulus seen simultaneously. The comparison spot moved linearly in the opposite direction. Test stimuli comprised one to three dots, each of which was attached to a point on the invisible wheel. An interleaved 2AFC staircase method was used to determine the comparison stimulus speed that appeared to match the forward speed of the invisible wheel. Direction of motion (left/right) and spatial position (upper/lower) of test and comparison stimuli were randomized on each trial. The distance traveled and starting location of each stimulus varied pseudorandomly. Results: The perceived speed of the wheel depended on the spot configuration. When one spot was on the rim, the speed of the linear component of its motion was overestimated. When one spot was placed on the wheel's axle and one on the rim, velocity was consistently underestimated. If two spots were placed on the rim 180 deg apart, speed was veridically matched. Conclusion: The perceived speed of a complex motion such as a cycloid may depend upon a center-surround motion-encoding mechanism that integrates motion within its center, while motion in the surround is inhibitory. From a practical point of view, reflectors on the wheel of a bicycle should be positioned opposite one another.