Visual contrast detection by a single channel versus probability summation among channels (Q1117167)

It is shown that with respect to the class of static models of visual contrast detection under very general and at least partially testable conditions contrast-interrelationship functions reported by \textit{J. J. Kulikowski} and \textit{P. E. King-Smith} [Spatial arrangement of line, edge and grating detectors revealed by sub-threshold summation. Vision Res. 13, 1455-1478 (1973)] cannot be interpreted in terms of models of detection by probability summation among independent subsystems. Sufficient and observable conditions for the strict log-concavity/log- convexity of distribution functions are presented.