Abstract
Disparity is an efficient and readily available cue for estimating the position in depth of our surroundings and objects of interest. It is also essential for successfully converging our eyes. While a great deal is known about disparity inputs to perception, we aimed here to determine the efficacy of disparity information from different parts of the visual field in driving vergence. Is disparity information equally effective across the visual field? We used a psychophysical approach to investigate the quality of vergence and binocular fusion for different eccentricities. Participants viewed dichoptic pairs of nonius lines, surrounded by binocular fusion locks consisting of annular sectors to drive vergence and binocular fusion. We manipulated the disparity of the fusion locks, effectively varying the vergence load of the task. We also manipulated the fusion lock eccentricity while compensating its size to account for cortical magnification. Participants aligned the centrally-positioned nonius lines in an adjustment task. The relative offset of the two lines was used as a readout for vergence accuracy and precision. We then computed vergence errors by subtracting each nonius offset from the fusion lock disparity. We show that the eccentricity at which the fusion lock stimuli are presented degrades vergence accuracy as well as precision: with fusion locks presented centrally at about 1 degree we observed the smallest vergence errors across all disparities. Moving the fusion lock to the periphery impaired vergence, resulting in significantly higher errors across disparities. Interestingly, the stability of the vergence drive also seems to be affected, with higher variability in the vergence errors observed at higher eccentricities. Similar to the strong bias towards central vision for the perception of disparity, our results quantify a similar bias for vergence control.