Heshu Yin¹, Joseph R Busch¹, Vijay Singh¹; ¹Haverford College

The perceived lightness of an object depends on its surrounding context. We measured lightness discrimination thresholds for judging the lightness of two gray patches when the patches were viewed against backgrounds of varying lightness. In each trial of a 2AFC task, human observers (N=3 observers) viewed two square patches (1.68 degree visual angle) presented on larger square backgrounds (5.57 degree visual angle) for 500ms. The patches and the backgrounds were presented simultaneously side by side (centers 5.57 degree apart) on a monitor. One patch had a fixed standard lightness, while the lightness of the comparison square patch was varied. Observers reported the patch that they perceive to be lighter with a controller. Between trials, the lightness of the comparison patch and the lightness of its background were changed. Percent comparison chosen data was collected for 11 comparison lightness values for each standard lightness (30 trials each). The resulting psychometric function was fit with a cumulative normal function to calculate the point of subjective equality (PSE) and 76% discrimination threshold. Psychometric data was collected for three standard lightness levels and three comparison background lightness levels. We develop a signal detection theory that includes the effect of background lightness on the perceived lightness of an object. The model predicts that an additive effect would shift the psychometric function, changing the PSE without altering the slope. On the other hand, a multiplicative effect would change both the PSE and the slope of the psychometric function. Our experimental data shows that for a fixed standard lightness, the PSE and thresholds changed proportionally with the lightness of the comparison background, supporting a multiplicative gain mechanism.