Understanding the Appearance of Small Spot Colors

Hofer et al. (Journal of Vision, vol. 5, no. 5, 2005) report that observers provide a wide range of color names in response to very small monochromatic spots. Here “very small” means spots with a retinal size comparable to that of a single cone (achieved through the use of adaptive optics), and “wide range” includes the term white. In this talk, I argue that a Bayesian calculation designed to estimate the L-, M-, and S-cone signals that were present in the stimulus predicts qualitative features of Hofer et al.’s data. In particular, the calculation predicts that the fraction of spots named white should depend systematically on the arrangement of L-, M-, and S-cones in individual observers’ retinae, in a manner consistent with the experimental data. The calculation combines the (noisy) responses of every cone in the mosaic with prior information about the spatio- chromatic statistics of natural images. It is implemented for actual measurements of the mosaic arrangement of the 5 individual observers of Hofer et al. and for simulations of the small spot presentations through the adaptive optics apparatus.