Target recovery in visual backward masking: no clear explanation in sight

Theories of visual backward masking have generally ignored demonstrations of the perceptual recovery of a masked target when the masking stimulus (M1) is followed by a second masking stimulus (M2). This failure to address recovery may reflect a belief that recovery effects are relatively small and inconsistent. The present article reports a series of experiments, using single-letter targets, a patterned M1 and a light-flash M2, which indicate that recovery can be a robust and substantial phenomenon. It is observable under a variety of conditions, and the magnitude of recovery matches the masking effect itself. A two-choice discrimination task was employed to test an explanation for recovery that attributes the effect to inhibition from the transient response to the light flash on the sustained response to the patterned mask. That explanation calls for the shape of the function relating the discriminability of the patterned mask to the onset asynchrony of M1 and M2 to be U-shaped in form. The data, however, reveal a monotonic masking function, suggesting that only sustained, intrachannel interactions are involved in the recovery of the target percept. Two conflicting explanations for recovery are discussed, one based upon inhibition of M1 responses, the other upon enhancement of target features. Thus, while recovery is demonstrated to be very real, its explanation remains elusive.