Abstract
Purpose: A point light walker can be effectively masked by the motion of spatially randomized dots that have the same motion components as the walker. However, little is known about the temporal fidelity with which the visual system processes biological motion. We sought to address this issue by determining the effect of temporal incoherence between the point light walker and the masking dots. Method: Participants detected point light walkers embedded in spatially randomized masks. The motion of the mask dots was either synchronized with the point light walker or non-synchronized by varying amounts of the step-cycle. The task was performed for upright and inverted point light walkers. Results: Detectability (as measured by d') followed an inverted double-U function. Specifically, the most effective masks were those synchronized with the point light walker and those out of synchrony by half of a step-cycle. The least effective masks were those out of synchrony by one-quarter or three-quarters of a step-cycle. Inverted walkers were more difficult to detect, but had the same basic inverted double-U function. Additional experiments showed that masks begin to lose their effectiveness when out of synchrony by more than 1/25 of a step-cycle. This was true for naïve and non-naïve participants and for upright and inverted walkers. Conclusions: Overall, temporal incoherence between mask and point light walker has a large effect on the detectability of a point light walker within a mask. The data suggest the swing of arms and legs are critical in determining the effectiveness of the mask, although the mask and point light walker can be slightly out of synchrony before detectability of the walker increases.