Abstract
Purpose: To measure the relative contribution of short (S-), middle (M-), and long (L-) wavelength sensitive cones to luminance as gauged with a minimum motion task. Methods: For a given achromatic intensity I, an annular, rotary motion display (subtending the central two degrees of visual angle) was used to find colored lights that were equiluminant to I. The motion display was designed so that, in the context of this display and using a high temporal frequency, a given light would yield ambiguous first-order motion if and only if it was equiluminant to I. Each of ten (4 female) observers was tested using lights of 20 different hues saturated to the maximum extent afforded by our monitor and varying in intensity. For each hue, a 1-up-1-down staircase was used to concentrate observations in the neighborhood of the light that was equiluminant to I. Results: For each observer, the 20 lights estimated to be equiluminant to I projected within measurement error to a plane in the space spanned by the (Stockman-Sharpe) S-, M- and L-cone fundamentals. The axis perpendicular to this equiluminant plane was taken to be the motion luminance axis. This motion luminance axis was then expressed as a linear combination of the (normalized) S-, M- and L-cone sensitivity functions. For three of our observers (all male), the S-cone contribution to motion luminance was negative (as reported by Ripamonti et al., 2009 for flicker luminance) but small. However, for the other seven observers, the S-cone contribution to motion luminance was positive. Wide variations were seen across observers in the relative contributions to motion luminance of the S-, M- and L-cones, from (0.1 : 0.01 : 9.8) to (0.2 : 2.3 : 7.3), in the most extreme cases for a particular I, but in all cases the L-cone contribution was dominant.
Meeting abstract presented at VSS 2013