Abstract
A new technique is described to measure the magnitudes and signs of cone inputs to visual neurons. The color of a uniform field is modulated around the circumference of a circle in a plane of cone-excitation space in either the clockwise or counter-clockwise direction. Relative cone weighting determines a cell's preferred vector in the cone-excitation plane; response phase is determined by the angle of the vector and by a phase lag intrinsic to the cell. The average of response phases to clockwise and counter-clockwise modulation cancels the intrinsic phase delay to reveal cone weighting. The method is quick and sensitive compared to methods based on measurements of response amplitudes.
To assess the S-cone input to PC-cells, we used a stimulus modulated in an equiluminant plane (L+M constant) defined by an L/M (constant S) axis and an S only axis. The average luminance and chromaticity of the modulation, i.e., the center of the circle, was equal energy white (W). Temporal frequency was varied between 1.22 and 39.4 Hz. The preferred directions of PC-cells were scattered around 0° (for +L-M cells) and 180° (for +M−L cells) in the equiluminant plane independent of temporal frequency. The scatter can reflect S cone inputs or calibration error or variability in preretinal absorption. As a control, the center of the circle was displaced from W in the +S direction along a tritan line. As the mean level of S-cone adaptation increased, cells' preferred directions converged towards 0° or 180°. This is consistent with a minor degree of S-cone input to some PC-cells.
Supported by EY13112 and EY07556.