Abstract
Hofer et al. (Journal of Vision, 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. We present a Bayesian calculation that models the data as a natural consequence of information loss arising from chromatic sampling: although the retina as a whole is trichromatic, it is monochromatic at the scale of single cones. The calculation starts with the simulated responses of the individual L-, M-, and S-cones actually present in the cone mosaic and uses these to estimate the trichromatic L-, M-, and S-cone signals that were present at every image location. The calculation incorporates precise measurements of the optics and chromatic topography of the mosaic in individual observers, as well as the spatio-chromatic statistics of natural images. We carefully simulated the experimental procedures of Hofer et al., and predicted the color name on each simulated trial from the average chromaticity of the LMS image estimated by our calculation. There were no free parameters to describe variation between observers. None-the-less, the striking individual variation in the percentage of spots named white emerged naturally as a consequence of the measured individual variation in the relative numbers and arrangement of L-, M- and S-cones. The model also makes testable predictions for experiments that may soon be feasible, including how color naming should vary with spot size and with the local structure of the cone mosaic surrounding the presented spot.
Supported by NIH EY10016.