Influence of luminance gradient reversal on simple cells in feline striate cortex

Publisher Website
Google Scholar
Abstract
The sensitivity of simple cells to luminance gradient reversal in bar stimuli was investigated in the striate cortex of lightly anesthetized cats. The influence of segment(s) of 1 polarity of contrast, partially masking or added end-on to a bar of reversed contrast, was assessed against a stationary textured background (of intermediate average luminance), which was itself without influence on cell behavior. In either configuration, short segments of reversed contrast were suppressive of bar response, to an extent varying with location along the receptive-field axis, but much greater than predictable from length-summation characteristics. Response suppression, even by very short segments, was often total. The effects of longer segments, added end-on to a bar of opposite contrast, depended on the extent of length summation exhibited by each cell. With progressive extension of these segments, some recovery of response occurred in a few simple cells with larger receptive fields, but not in small-field simple cells. The behavior of end-stopped simple cells was comparable in all respects to that of their end-free counterparts, within the length-summation zone; thereafter, invasion of the inhibitory end-zones by bars of either polarity elicited a generalized decline in response. Where responsiveness was restored by adding longer segments of opposite polarity, the response peak shifted to the discharge center appropriate to that polarity. All these results were found with either polarity of contrast, i.e., a light bar plus dark segments or the converse. Similar results were obtained for motion in either direction across the receptive field; and also for flash-presentation, over either the center or flanks (whether on or off) of the receptive field. The effectiveness of contrasting added segments was assessed as a function of location along the receptive-field axis for flash-presented and for moving stimuli. Suppressive effects were greatest over the center of the receptive field, declining progressively in either direction along its axis, but over greater distances than anticipated from length summation. None of the results presented could have been predicted from length-summation characteristics. They are strongly suggestive of gating, rather than linear, antagonistic interactions.