Abstract
The perceived shape of a pattern (target) can be masked by that of a subsequent pattern (mask). The purpose of the present work is to elucidate the nature of this shape masking. Perceived deformation of a target radial frequency (RF) pattern (radius 1.1deg) was measured when followed by an RF mask (radius 1.6 deg) of high amplitude (16 × detection threshold). The effect reaches a maximum at a stimulus onset asynchrony (SOA) of 80–100 ms, where thresholds are elevated by a factor of 16. Various configurations in which target and mask are separated by this SOA were tested (the mask is termed primary here). Conditions in which a second mask is interleaved in time between the target and primary mask, lead to smaller threshold elevations and reveal that the onset of the mask that appears first in time determines the magnitude of masking. Configurations in which apparent motion is possible between the target and both masks lead to large threshold elevations (factors of 16–20) and demonstrate that both masks contribute to the effect. The magnitude of these effects is much larger than that predicted by a combination of spatial lateral interactions and apparent motion between target and mask. Results suggest that shape, apparent motion, and stimulus onset play an interactive role in masking, and that target shape is not determined at initial onset, but rather, is extrapolated (postdictively) after a window of 80–100ms.
This work was supported by NSERC grants to FW (#OP0007551) and HRW (# OP227224 ) and by a CIHR Postdoctoral Fellowship to CH.