Age-dependent prediction of visible differences in displayed images

20 Mar 2018

A comprehensive computational model of the human visual system has been developed to predict just-noticeable differences between a pair of stimuli. The stimuli are provided as high-dynamic-range, multi-channel images, containing accurate spectral and photometric description of the scene. In this work we extend such a model to predict the effect of age on the sensitivity of the visual system. The extensions are based on the existing models of disability glare, aging of the crystalline lens and reduced pupil size with age (senile miosis). We confirm findings from other studies that those optical effects cannot fully explain the effect of aging, and we attribute the remaining drop of sensitivity to neural changes. The complete model, including an empirical neural component, can well explain the differences in contrast sensitivity between old and young observers.