An explanation of perisaccadic compression of visual space.

A large number of experiments show that perisaccadic flash mislocalization can vary according to the spatial location of the flash relative to the saccade, especially in the presence of background stimuli. The temporal attributes of this mislocalization suggest that, around the time of a saccade, a transient compression of visual space occurs. The present study offers a model to account for such compression. A basic aspect of the model is that the mislocalization is a consequence of flash retinal signal persistence interacting with an extraretinal signal. Of central importance, however, the model suggests that the extraretinal signal is different when a saccade occurs in the dark from when a saccade occurs with background stimuli. In the dark, the extraretinal signal begins to change with little or no time difference from one retinal locus to another, resulting in little or no compression. However, in the presence of background stimuli, the extraretinal signal begins at considerably different times across the retina, giving rise to a large amount of compression.