Abstract
Purpose: The perceived direction of moving Type 2 plaids is usually intermediate between the veridical intersection-of-constraints (IOC) direction and the errant vector-sum direction. Our recent analysis of a Bayesian model of motion extraction predicts that when one of a plaid's component gratings gives a more reliable motion signal than the other component grating, perceived direction should be biased toward the grating with the more reliable motion signal. This predicts that perceived direction is either a monotonic or nonmonotonic function of relative reliability, depending on baseline direction percept. Reliability can be manipulated, for example, by contrast or by spatial frequency (SF). We have been accumulating evidence in support of these predictions. Methods: Naïve subjects viewed Type 2 plaids whose component gratings drifted in directions separated by 15 deg and whose speeds differed by a factor of ?1.5. We manipulated duration, aperture diameter, baseline spatial frequency, and baseline contrast. We also manipulated the ratio of the gratings' SF and of their contrast. Subjects pointed an arrow in the direction of perceived drift. Results: When the gratings were of equal SF and equal contrast, perceived direction was close to the IOC direction or vector-sum direction, depending on baseline manipulations, as predicted by the model. When the gratings were of unequal SF or unequal contrast, perceived direction was biased toward the grating with more reliable motion signal, as predicted, whether reliability was manipulated via contrast ratio or SF ratio. Conclusions: The results consolidate confirmations of predictions of a Bayesian computational model of motion extraction that suggests that IOC and vector-sum directions, as well as directions beyond vector-sum, can arise from a motion analysis that weights motion signals according to their reliability. These predictions arise from strictly computational rather than physiological considerations.
Supported by NIH grant EY 013362.