Abstract
Background: When interpreting the shape of 3D surfaces the visual system assumes that illumination is mostly from above and slightly from the left. Visual search is parallel for below-lit targets amidst top-lit distractors, but not vice versa. However, visual search for oddly-illuminated singletons is serial when the 3D pose of cube targets and distractors is randomized. We asked whether the direction of lighting of randomly posed 3D cubes would influence their brightness (and/or perceived intensity of illumination). Method: An array of nine Lambertian cubes was stereoscopically rendered in low ambient illumination against a dark background. Individual cubes varied in their 3D pose, but all possessed identical triplets of visible faces. The cubes were illuminated from one of four directions: top-left (+60o elevation; -90o azimuth); top-right (+60o; +90o); bottom-left (-60o; -90o); and bottom-right (-60o; +90o). Illumination intensity (arbitrary units) ranged from 1-15 (linear steps). "Standard" cubes were illuminated from top-left at intensity 8; comparison cubes were illuminated from all four possible directions, and appeared in either the left or right visual fields. Using the method of adjustment (N=20, 13 male) we determined comparison cube illumination levels required to establish subjective equality with the standard cubes as a function of comparison cube visual field, illumination elevation, and illumination azimuth. Results: Cubes appeared significantly brighter in the left visual field (p=.008), and when illuminated from below (p<.001). There was a trend for right-lit cubes to appear brighter (p=.069). The enhanced brightness of surfaces lit from below was greatest when also lit from the right (p=.001). Conclusions: Surfaces lit from below appear brighter (more illuminated) than identical surfaces lit from above, due perhaps to long-term adaptation to downward lighting. This effect is modulated by azimuth, being strongest for illumination from the right. Brightness is amplified in the left visual field, presumably via attentional enhancement.
Meeting abstract presented at VSS 2013