Abstract
Perceptual organization is critical for parsing incoming sensory information, and an important aspect of this challenge is to detect contours that represent edges of independent objects in the real world. During every day functioning, eye movements are frequently made to bring a subset of the visual field into the fovea, which offers improved spatial resolution. However, eye movements also change which specific neurons in the early visual system are activated. In this study, we investigate how this tradeoff is resolved by measuring eye movements during a contour integration task and how they impact performance. Participants performed a forced-choice discrimination task which required them to find an egg-shaped contour embedded in distractors and make a speeded response about its facing direction (left or right). The difficulty of this task was varied by jittering the orientations of the individual Gabor patches that made up the target contour. Orientations of the distractor Gabor patches were randomly chosen. To assess differential strategies, eye movement data were collected during this search task. The results show the expected response time pattern, such that greater contour jitter led to slower performance. Interestingly, dwell time before the initiation of the first saccade significantly correlated with subsequent behavior. Participants with longer dwell times made greater amplitude first saccades and required fewer eye movements to respond correctly. These results offer some perspective on how individuals may tradeoff between exploratory saccadic behavior in favor of extended perceptual processing which may benefit contour integration.
Meeting abstract presented at VSS 2018