Receptive field size and magnification have been studied in striate cortex of awake, behaving rhesus monkeys at visual eccentricities in the range of 5-160 min. The major findings that emerge are (1) magnification in the foveola achieves values in the range of 30 mm/deg, (2) mean field size is not proportional to inverse magnification in contrast with previous reports, and (3) the product, magnification X aggregate field size, is greater in central vision than in peripheral vision. Thus, a point of light projected onto foveal retina is "seen" by larger numbers of striate cortical cells than a point of light projected onto peripheral retina. Implications of these findings for visual localization and two-point discrimination are discussed.

1. The activity of single cortical cells in area 17 of anesthetized and unanesthetized cats was recorded in response to prolonged stimulation with moving stimuli. 2. Under the appropriate conditions, all cells observed showed a progressive response decrement during the stimulation period, regardless of cell classification, i.e., simple, complex, or hypercomplex. 3. The observed response decrement was shown to be largely cortical in origin and could be adequately described with an exponential function of the form R = Rf +(R1-Rf)e-t/T. Time constants derived from such calculations yielded values ranging from 1.92 to 12.45 s under conditions of optimal-stimulation. 4. Most cells showed poststimulation effects, usually a brief period of reduced responsiveness that recovered exponentially. Recovery was essentially complete in about 5-35 s. 5. The degree to which stimuli were effective at inducing response was shown to have significant effects on the magnitude of the response decrement. 6. Several cells showed neural patterns of response and recovery that suggested the operation of intracortical inhibitory mechanisms. 7. A simple two-process model that adequately describes the behavior of all the studied cells is presented. 8. Because the properties of the cells studied correlate well with human psychophysical measures of contour and movement adaptation and recovery, a causal relationship to similar neural mechanisms in humans is suggested.

No abstract

Color-tuning curves were obtained for 218 cells in the foveal striate cortex of behaving macaques. Each cell was tested with its optimal spatial stimulus. Test colors (14 interference filters, 4 Wratten filters, and white) were matched for human photopic luminosity and presented at luminance levels sufficient to induce vigorous responding from most cells. One hundred eighty-four cells were selected for further analysis on the basis of a color-tuning index. Of these, 130 with tuning curves that correlated well (0.9 or better) with other tuning curves were studied in detail. Individual cells were found with peak responses to every color tested. Sixty-three tuning curves fell into the six largest cross-correlation groups, containing 15, 14, 12, 9, 7, and 6 cells, with mean tuning-curve peaks at 450, 656, 656, 506, 577, and 506 nm, respectively. Cross-correlation groups having the same peak location (656 nm, 506 nm) were distinguishable on the basis of tuning-curve width. Response patterns, cone input estimates, and comparison with human psychophysics suggest that two of these cell groups function as an opponent pair processing the colors red and green. Two other cell groups process the colors blue and yellow but show less well-developed opponency. Microdrive depth readings, correlated with histological lesion sites, show these "red," "green," "blue," and "yellow" cells to be most common in layer 4 of the striate cortex.

Microelectrode penetrations nearly normal to the layers of foveal striate cortex in awake, behaving monkeys reveal a shift in orientation preference between cells in the upper and lower layers. Mean shift size for 57 penetrations is 54.8 degrees, with 70% of the penetrations showing shifts of 45-90 degrees. Marking lesions localize the shift to the border between layers 4C and 5. The data are suggestive of inhibition between the upper and lower layers within an "orientation column".