The size and number of plaid blobs mediate the misperception of type-II plaid direction.

The misperceived direction of type-II plaids has posed a problem for the intersection of constraints (IOC) model of two-dimensional motion perception. Alais et al. (1994, Vision Research, 34, 1823-1834) examined the perceived direction of type-II plaids and concluded that in addition to the direction signalled by the IOC process, a monocular mechanism signalling the motion of plaid features (blobs) is also involved in plaid perception. It was shown that the prominence of this monocular signal in plaid direction judgements depended on several variables, and the notion of blob "optimality" was introduced. This explained the more veridical direction of "optimal" blob plaids in terms of their more effectively activating the proposed feature-sensitive motion mechanism. One distinction between "optimal" and "non-optimal" blob plaids is their different component spatial frequencies, which necessarily entails a difference in the number and size of the blobs and thus raises potential confounds, since both the nature of the blobs and the components differ, which might affect the postulated blob mechanism and/or the IOC process. In the present paper, by offsetting changes in spatial frequency with changes in aperture size so that blob number is held constant, we examine whether differences in sheer blob number or size can alter perceived type-II plaid direction. The results reveal effects of both blob number and blob size, and their implications for the underlying mechanism are considered. Alternative accounts of the results in terms of the IOC model or revisions of it cannot explain the data. Comparison of monocular and binocular conditions adds further systematic evidence in support of the monocularity of the feature-sensitive motion mechanism.