Abstract
Mislocalization of a perisaccadic flash can vary according to the physical location of the flash relative to the saccade, especially when the saccade occurs in the presence of a continuous background stimulus (e.g., Ross et al., 1997; Lappe et al., 2000). For example, mislocalization tends to be in the saccade direction if the flash is near the saccade starting point, whereas the mislocalization is opposite to the saccade direction if the flash is beyond the saccade endpoint. These findings have been interpreted as reflecting a neural process concerned specifically with a transient compression of visual space. I have proposed a model showing that perisaccadic flash mislocalization comes from flash retinal signal persistence interacting with a post-saccadic monotonic extraretinal (exR) signal (Pola, 2004). This model simulates mislocalization when a saccade occurs in the dark, a visual circumstance in which there is little or no indication of compression. In contrast, the prior studies (above) involving a background, insofar as they suggest the existence of compression, raise the possibility that a background modifies the exR signal. I have used my model to explore what sort of exR signal modification might occur. The model suggests that a continuous background stimulus does not result in a process concerned specifically with compression. Instead, the background may simply have the effect of varying the exR signal onset time across the retina, e.g., the onset begins before the saccade for retinal loci corresponding to a flash near the saccade starting point, whereas the onset begins after the saccade for retinal loci corresponding to a flash beyond the saccade endpoint. This variation in onset time gives rise to many of the features of background-contingent flash mislocalization.