Abstract
In the retinal blind spot, where the optic nerve exits the eye, visual input is always missing. Perceptual filling-in occurs at the blind spot and sometimes with artificial scotomata. Moving spatial patterns, but not static ones, lead to filling-in of spatiotemporal information in the blind spot (Maus & Whitney, 2016). However, it is unclear if the same occurs for artificial scotomata of the same size and eccentricity. Filling-in of artificial gaps improves with increasing eccentricity (De Weerd et al., 1998). Therefore, strong perceptual filling-in at the blind spot could be largely due to its peripheral location or because of a special filling-in mechanism. Here we compared perceptual filling-in of spatiotemporal patterns across the blind spot, occlusion and artificial scotomata (same size and eccentricity) in the fellow eye. In a psychophysical experiment, subjects viewed sinusoidal gratings at 5 spatial frequencies, 0.25-0.45 cycles/degree (cpd), and 5 gap conditions: control (no gap), blind spot, occlusion (an oval covered the center of the bar), and two deletion conditions (center of bar erased, with sharp or fuzzy edge). Participants responded whether bars from each condition or a standard control bar with no gaps at 0.3 cpd had "more stripes overall". We calculated the Points of Subjective Equality (PSEs) for each condition. Smaller PSEs mean that more stripes were perceived in the gap, indicating filling-in. We found that PSEs for the blind spot were lower than for other gaps, indicating better pattern filling-in across the blind spot. There was no difference in filling-in for occlusion vs deletion conditions. Comparing occlusion/deletion PSEs to the veridical amount of bars presented, we found that subjects somewhat overestimated the number of stripes (PSEs lower than predicted from accurate perception), suggesting some weak perceptual filling-in of occlusion and deletion. Additionally, the amount of filling-in (PSEs) was not related to the occluder size.
Meeting abstract presented at VSS 2018