Abstract
The contrast discrimination threshold decreases (facilitation) and then increases (masking) as pedestal contrast increases. If another pattern (mask) is substituted for the contrast pedestal, both facilitation and masking usually decrease as the patterns become dissimilar. Masking occurs over a wider range than facilitation. When contrast discrimination in measured in the presence of a dissimilar mask of fixed contrast, thresholds are higher at low pedestal contrasts. They decrease as pedestal contrast increases, sometimes crossing the baseline contrast discrimination function and then increase at high pedestal contrasts. These effects are found for masks that vary in spatial frequency, temporal frequency, orientation, and phase. They are also found for lateral masking, forward and backward masking, masking by noise and contrast adaption. Many of these effects are accounted for by a model (Foley, 1994) in which the detecting mechanism receives both excitatory and divisive inhibitory inputs. Their net effect is to produce an S-shaped response function. A fixed context pattern adds an increment to the baseline divisive inhibitory signal that changes the form of the response function. One effect of this change is to shift the steep region of the function to higher contrasts, thus improving contrast discrimination in this range. Elaborations of the model are required to account for some effects. Alternative models in which induced noise is substituted for divisive inhibition, sometimes account for discrimination data as well as the divisive inhibition model. However, experiments in which perceived contrast is matched show that suppression underlies these effects.
Meeting abstract presented at OSA Fall Vision 2012