Abstract
Population receptive field (pRF) mapping developed by Dumoulin and Wandell (2008) traditionally uses 100% contrast black and white checkerboard patterns to produce retinotopic maps and receptive field information from across the visual cortex. pRF maps therefore reflect the spatial tuning of the neuronal populations responding to high contrast achromatic stimuli. However, psychophysical and single unit studies show that chromatic pathways tend to be spatially low-pass compared to luminance pathways. To date, there are no reports of fMRI-based pRF techniques being applied to chromatic pathways. In this study, we attempted to use pRF mapping techniques to measure these innate resolution differences in human subjects.
A modified version of Dumoulin and Wandell's ‘drifting bars’ stimulus was used to produce three pRF mapping conditions: L+M+S cone achromatic, L-M cone ‘red/green’ and S-cone isolating ‘yellow/blue’. Isoluminance was set by radiometric calibration of the scanner display system, followed by individual minimum motion settings for each subject (N=5). The bars for each condition contained a randomly-updated (2Hz) spatially pink noise (1/f) carrier so that neurons responsive to all spatial frequencies could, in theory, respond to the stimulus.
We replicated the pRF parameters from Dumoulin and Wandell's paper: pRF size increased with eccentricity and with ascending hierarchy of visual areas.
However, we did not observe any differences for the chromatic conditions. We report the pRF properties obtained for each condition, and discuss our findings in the context of pRF mapping stimulus design. We also consider potential applications of this method for investigating neuronal population tuning in dichromats.