Abstract
We examined with fMRI (Siemens MAGNATOM Sonata, 1.5T) the proposal by Lu et al. (PNAS, 1999) that L- and M- receptor-mediated motion perception at isoluminance is due to a tracking-based motion processing mechanism. According to this theory, isoluminant gratings appear to move when differences in the relative salience of the alternating stripes of the moving grating produce features that can be tracked over time. Operationally, relative salience can be manipulated by varying the saturation of a greenish grating that is modulated isoluminantly in spatial antiphase with a 1.0 contrast red grating. We measured the BOLD response to monocularly viewed, eccentrically placed (5.0 deg. nasal visual field), Gaussian windowed (sigma × = 1.25 deg.; sigma y = 2.5 deg.), 0.5 cycle/deg. horizontal gratings that moved either up or down (3.75 Hz) or were stationary. We compared BOLD responses for moving vs. stationary low-, medium- and high-green- saturation gratings and for a luminance-modulated red/black control grating. Isoluminance for each saturation condition was determined in situ by motion photometry. All four motion conditions led to enhanced BOLD responses in contralateral V1 and bilateral V5+, relative to those obtained from their stationary stimulus counterparts. We also obtained a saturation-dependent variation in relative BOLD response in V5+: response was greater for low- or high- saturation than for moderate- saturation isoluminant gratings. These variations in motion-induced V5+ activation at isoluminance parallel the psychophysical findings by Lu et al. under similar isoluminant motion conditions.
Supported in part by NSF BCS-007-9277 to DTL and GOA 2000/11 to GAO.