Cortical Double-Opponent Cells in Color Perception: Perceptual Scaling and Chromatic Visual Evoked Potentials.

Valerie Nunez,Robert M Shapley,James Gordon

doi:10.1177/2041669517752715

Valerie Nunez, Robert M Shapley + Show 1 more

Open Access

https://doi.org/10.1177/2041669517752715

Copy DOI

Journal: i-Perception	Publication Date: Jan 1, 2018
Citations: 18	License type: cc-by

Affiliation: New York University, Hunter College

Abstract

In the early visual cortex V1, there are currently only two known neural substrates for color perception: single-opponent and double-opponent cells. Our aim was to explore the relative contributions of these neurons to color perception. We measured the perceptual scaling of color saturation for equiluminant color checkerboard patterns (designed to stimulate double-opponent neurons preferentially) and uniformly colored squares (designed to stimulate only single-opponent neurons) at several cone contrasts. The spatially integrative responses of single-opponent neurons would produce the same response magnitude for checkerboards as for uniform squares of the same space-averaged cone contrast. However, perceived saturation of color checkerboards was higher than for the corresponding squares. The perceptual results therefore imply that double-opponent cells are involved in color perception of patterns. We also measured the chromatic visual evoked potential (cVEP) produced by the same stimuli; checkerboard cVEPs were much larger than those for corresponding squares, implying that double-opponent cells also contribute to the cVEP response. The total Fourier power of the cVEP grew sublinearly with cone contrast. However, the 6-Hz Fourier component’s power grew linearly with contrast-like saturation perception. This may also indicate that cortical coding of color depends on response dynamics.

Highlights

The primary visual cortex V1 is a bottleneck for color perception in the cortex; color processes occurring later in the cortex are based on the responses from the neural substrates for color perception in V1
The color of the red checks in the high-frequency checkerboard scaled in saturation fairly linearly with cone contrast
For the range of cone contrasts used, there is no evidence of the ratings hitting a response ceiling with increasing cone contrast; the perceived saturation grew monotonically and directly proportional to the cone contrast

Summary

Introduction

The primary visual cortex V1 is a bottleneck for color perception in the cortex; color processes occurring later in the cortex are based on the responses from the neural substrates for color perception in V1. Previous research established that color-responsive neurons in the primary visual cortex, V1, of macaque monkeys can be assigned to colorpreferring and color-luminance cell classes (Johnson, Hawken, & Shapley, 2001). Later work showed that color-preferring cells were mostly single-opponent cells and color-luminance cells were mostly double-opponent cells (for a review, see Shapley & Hawken, 2011). As a consequence of this formative work, we know that all color-responsive neurons in V1 are divided into just two known groups that mediate color: single- and doubleopponent neurons. Given that in V1, the only color-responsive neurons are single- and double-opponent cells, we sought to answer the fundamental question, what is the relative contribution of each class of cells to color perception (cf different opinions published previously: Shapley, Hawken, & Johnson, 2014; Solomon & Lennie, 2007)?

Objectives

Methods

Results