Abstract
The existence of neurons that selectively respond to the intermediate hues (hues off the cardinal axes of cone-opponent color space) in human early visual areas is still under debate. Kuriki et al. (2015) provided direct evidence for such neurons in human V1 to V4, using phase encoding mapping and a hue-selective adaptation paradigm in fMRI. Using similar stimuli as Kuriki et al. (2015), we now examine hue selectivity employing steady-state visual evoked potentials (SSVEPs). All colors used in this study were chosen from the equiluminant plane in the cone-opponent color space. The origin of this color space was Equal Energy White (EEW, 30 cd/m2): colors along the vertical axis only differ from the origin in S-cone response and those along the horizontal axis in L-cone (and incidentally in M-cone) response. Participants observed a 6 deg checkerboard pattern for 25 sec each trial. Half of the tiles (0.8 x 0.8 deg each) of the checkerboard were uniformly filled with the background hue (EEW), and the rest of them with the test hue. During the trial, the test hue smoothly changed at 24 s/cycle rate while the tiling pattern alternated at 5 Hz. To control participants' attentional state, they performed a simple Go/No-Go task at fixation. There were three color-contrast conditions; standard (full contrast, with ∆L = 8 %, ∆S = 80 %), half (4/40 %), and quarter (2/20 %). SSVEP amplitudes increased with color-contrast, suggesting that it reliably reflects hue-selective activity. Average SSVEP amplitude elicited by intermediate hues was just as large as to the cardinal hues and the circular profile of SSVEP amplitude along hue was neither point nor line symmetric, which supports Kuriki et al. (2015). We conclude that our approach using SSVEP is a valid technique to assess hue selectivity in the human brain.
Meeting abstract presented at VSS 2018