Abstract
Interocular contrast gain-control (interocular-CG) was first used to model binocular phase combination (Ding & Sperling 2006), and later modified to explain both phase and contrast combination in binocular vision (Huang et al 2010, Ding et al 2013). In the present study, we demonstrate that, by adding monocular luminance gain-control (monocular-LG), the interocular-CG model can also explain binocular luminance summation. Luminance disks were used as stimuli. The target luminance was either increased by 1, 2, 4, 8, 16 cd/m2 against a darker background (0, 2, 4, 8, 16 cd/m2) or decreased against a lighter background (4, 8, 16 cd/m2). Experimental trials consisted of two stimulus intervals, one with a standard luminance which was identical in the two eyes, and the other with a test luminance presented to both eyes. The interocular ratio of the test luminance varied from trial to trial. The observer’s task was to judge which interval had the higher luminance increment or decrement. Consistent with previous studies, we found that binocular luminance summation behaves very differently depending on the background luminance. Fechner’s paradox was observed on the equal-luminance-increment contour with the dark background (0 cd/m2), while an apparent winner-take-all phenomenon was observed on both increment- and decrement-contours with the light background (16 cd/m2). Using one set of model parameters, 3 parameters for monocular-LG and 3 parameters for interocular-CG, the combined model provides a unified explanation of binocular combination of both luminance increments and decrements over a large range of luminance input (reduced-chi-square = 2.35). Combined with previous studies on phase (Ding & Sperling 2006), contrast matching (Huang et al 2010, Ding et al 2013), contrast discrimination (Ding et al 2013), stereovision (Hou et al 2013), and second-order stimuli (Zhou et al 2014), we conclude that the interocular-CG can provide a unified explanation of binocular combination for multiple binocular tasks.
Meeting abstract presented at VSS 2015