Abstract
Contrast sensitivity functions (CSF) are commonly used to characterise the sensitivity of the human visual system at different spatial scales, but little is known how the CSF changes from the mesopic range to a highly photopic range reflecting outdoor illumination levels. The purpose of our study was to further characterise the CSF by measuring the luminance and the chromatic sensitivity for background luminance levels from 0.2 cd/m2 to 7000 cd/m2. Stimuli consisted of Gabor patches of different spatial frequencies, generated using MatLab, varying from 0.5 cpd to 6 cpd andwere displayed on an HDR display generating luminance levels of up to 15000 cd/m^2. Contrast sensitivity functions were measured in three directions in colour space, reflecting early post-receptoral processing stages: an achromatic (L+M) direction, a ‘red-green’ (L/(L-M)) direction, and a ‘lime-violet’ direction (S/(L+M). Thresholds are defined as the distance in cone contrast space at which 84% correct performance was achieved. Within each session, observers were fully adapted to the fixed background luminance (0.2, 2, 20, 200, 2000 or 7000 cd/m2) and on each trial a stimulus of a different frequency (0.5, 1, 2, 4, 6 cpd) and colour (achromatic, red/green, lime/violet) was presented. Our main finding is that the background luminance has a differential effect on luminance compared to chromatic contrast sensitivity. Contrast sensitivity is increasing with background luminance up to around 200 cd/m^2, in particular for medium and high frequencies. Even higher luminance levels (2000 cd/m^2) yield a decrease in luminance contrast sensitivity which is not observed in the chromatic sensitivity curves, or occurs at higher light levels. The differential effect of background luminance on luminance and chromatic contrast sensitivity implies that a local cone contrast adaptation model is not sufficient to account for the observed sensitivity changes.
Acknowledgement: EPSRC EP/P007503