Abstract
We have shown that, in the mesopic range, glare reduces the perceived brightness of a foveal test patch in a non-linear fashion (Colombo, Barraza, Issolio, LR&T, 32(2), 2000). Here we report the effects of glare measured under a variety of different contrast conditions obtained by modifying the background luminance.
The experiment was carried out using a reference test patch of 4 deg in diameter and a uniform luminance of 0.5 cd/m2, and a similar comparison patch with a luminance value in an interval of six possibilities. Both patchs was presented on a background of 14×18.5 deg whose luminance was varied in order to obtain contrasts between 1 and −3 with the reference luminance (Lr-Lb/Lr). The reference test patch was presented under conditions of transient glare produced by shining a bright light located 10 deg away from the line of sight, resulting an illuminace of 60 lx between the two eyes .
Subjects had to compare brightness of the two patchs displayed sequentially. The observer reported which field was brighter. A forced choice paradigm using the method of constant stimuli was adopted to determine the luminance that produced equal perceived brightness.
Results show an asymmetry in behaviour for the two polarities of contrast. Over most of the range there is a darkness induction, as in previous results. However, for the smallest negative contrasts there is a brightness induction.
This variation is produced by a trade-off between two effects: first, intraocular scattered light raises the retinal illuminance of the test patch causing brightness induction. Second, the background retinal illuminance reduces the apparent brightness of the test patch. The final sign of the perceptual effect (brightness or darkness induction) depends on the relative weights of these two effects.