Abstract
The perception of color and spatial frequency varies across the visual field. The neural mechanisms that explain effects of eccentricity on visual perception have been investigated in humans using imaging techniques such as fMRI and single-unit recording in monkeys. To connect these two approaches and to test the extent to which neural mechanisms in macaque correspond to the human case, we collected 208 runs of fMRI data in rhesus macaques while the animals viewed carefully calibrated visual stimuli similar to those used in human studies. fMRI data were collected following intravenous injection of a MION-contrast agent, in two rhesus macaques while they viewed retinotopic stimuli or slowly drifting vertically oriented gratings (0.75 cycles / sec). The gratings varied in spatial frequency, color (defined with a cone-opponent color space), and contrast. The fMRI responses increased with color contrast across all visual areas, with maximal contrast eliciting a 1.5%-2.5% greater signal increase than minimal contrast when compared to gray. Color contrast sensitivity decreased with eccentricity for L-M color gratings, with the L-M contrast response function significantly reduced in the periphery of V2 and V3 (p<0.01). ‘Daylight-axis’ colors, varying from blue to orange along an intermediate direction in cone-opponent color space, elicited stronger responses across cortical areas compared to ‘anti-daylight-axis’ colors, varying from lime-green to magenta. There was higher fMRI signal for low than high spatial frequency stimuli in all areas, this difference became more pronounced with greater eccentricity. These results provide a quantitative basis for comparing cortical responses between humans and rhesus macaques, important due to the macaque’s common use as a model of the human visual system. Our results are consistent with human psychophysical and fMRI data, suggesting that the neural mechanisms underlying changes in perception across the visual field are conserved across the two species.