Abstract
It has been shown that disparity can be used as a token for visual search (possibly pre-attentively). However, there has been no systematic investigation of the distribution of attentional resources across the disparity dimension. Here we evaluated whether position in depth, relative to the screen plane, influences attentional allocation. We conducted two experiments using the same visual search task but with different stimuli. In the first experiment, stimuli consisted of four simple geometric shapes and in the second experiment the stimuli consisted of four orientated lines enclosed by a circle. In both cases, the stimuli were arranged in an annulus about a central fixation marker. On each trial, observers indicated whether the target was present or not within the annular array. The distractor number was varied randomly on each trial (2, 4, 6, or 8) and the target was present on half of the trials. On all trials, one element had a disparity offset by 10 arcmin relative to the others. On half of target present trials the target was in the disparate location, on the remainder it was presented at the distractor disparity. Trials were further subdivided such that on equal numbers of trials the disparate element was on or off the plane of the screen. We measured search time, analysing only trials on which observers responded correctly. Both experiments showed that when the target was the disparate item, reaction time was significantly faster when the target was off the screen plane compared to at the screen plane. This was true for a range of crossed and uncrossed disparities. We conclude that there exists a selective attentional bias for stimuli lying off the screen plane. These data are the first evidence of a disparity-selective attentional bias that is not mediated by relative disparity.
Meeting abstract presented at VSS 2012