Abstract
Heading perception depends on the ability of different regions of the visual field to extract accurate information about the direction of the visual flow. Hence due to its ability to extract the most accurate information, the central visual field plays a major role in heading estimation. With experience people learn to utilize other regions especially if there is central field loss/impairment. Nevertheless, it is not clear what happens when information in central vision becomes altered or cannot be picked up. In the present study, we examined the effects of gaze-contingent alteration of regions of the visual field on heading. On each trial, one of six different directions of self-motion were simulated (headings ±7.5°, ±5.0° and ±2.5° from the centre of the screen). The simulated defects were analogous to two typical visual field disturbances resulting from macular degeneration, either metamorphopsia or scotomas. Specifically, with a force choice procedure we compared performance with no visual defects to that with five different simulated defects (either 5° or 10° horizontal perturbations, 5° or 10° Gaussian perturbations, or a 10° scotoma). We also looked at three gaze conditions - free viewing, directional viewing and tracking features in the scene. Heading performance was not significantly different in the two environments examined (translation over a plane covered with blue particles or through a forest). Performance declined in the presence of simulated visual defects, as well as when they were instructed to visually track specific scene features. Performance was most accurate for all heading directions during the free view conditions. We conclude that when people are free to direct their gaze in the scene they are able to minimize the impact of simulated central visual field loss/distortion.
Meeting abstract presented at VSS 2015