Abstract
During binocular rivalry, incompatible dichoptic images compete for perceptual awareness. Dichoptic images do not compete for awareness independently when rivalry is instigated at multiple locations of the visual field. This inter-dependence is thought to be the result of perceptual grouping. We use the inter-dependence in perceptual dominance to disentangle the effect of different cues on grouping during rivalry. Specifically, the simultaneous dominance of spatially separated competing images can be based on similarities in their image-content as well as which eye they are presented to.
We presented spatially separated rival images such that similar images were either presented to the same or to different eyes. Since rivalry competition can be resolved at multiple stages along the visual processing hierarchy we also tested whether the influence of different cues depends on the type of images used. In two experiments, we used a perceptual tracking paradigm to measure the durations of perceiving either identical versus different orientation (Exp_1) or perceiving parts of a face versus parts of a plaid (Exp_2).
For rivaling orientations (Exp_1), we found eye-of-origin to be the strongest cue for grouping. In addition, image-content is also a cue for grouping, as images with identical orientations were perceptually dominant together longer than images with dissimilar orientations, even when they were presented to different eyes. However, in Exp_2, when faces were in binocular competition with plaids, the influence of the cues reversed. That is, in this experiment image-content was the stronger cue for grouping although eye-of-origin still influenced the dominant percept.
We are able to directly compare these different contributions that originate at different levels of the visual processing hierarchy. Overall, our results show that eye-of-origin and image-content cues can independently affect grouping during rivalry and that their relative contributions can further be affected by the semantic content of the competing images.
Meeting abstract presented at VSS 2012