The corticogeniculate fibers from V1 and close to 90 percent of the back-projecting fibers from V2 to V1 originate within layer 6 of the respective cortices. We propose to first characterize how the amplitude, phase, and latencies of the extracellular responses of these neurons depend on the orientation, spatial and temporal frequency, contrast, motion, color, disparity, size, shape and position of visual stimuli. Back-projecting neurons in V1 and V2 will be identified and axonal conduction times determined by antidromic stimulation of their fibers within the LGN and V1 respectively. These results will be used to determine if the response properties of neurons in the back-projecting pathway(s) differ from those of other layer 6 cells and whether the back-projecting neurons subserve one or more than one of the functional subsystems that have been proposed to encode information selectively for shape, color, motion an depth discrimination. Stimuli as have been determined in the first set of studies to strongly activate back-projecting neurons in V2 and V1, together with adequate control stimuli, will then be used to probe the effects of visual stimulation of back-projecting pathways on the response properties of neurons in V1 and LGN respectively. In this second set of experiments we propose to determine how the response properties of neurons in V1 and LGN when stimulated by appropriate test stimuli within their classical receptive fields are modified by the presentation of the second set of stimuli, i.e., stimuli which strongly activate back-projecting neurons, to region outside their classical fields. Such stimuli will be presented both monocularly and dichoptically. For all LGN neurons and for monocularly responsive neurons in V1, we will also dichoptically stimulate regions representing the classical receptive field for the non-dominant eye. We will also alter the temporal onset of the surround stimuli with respect to the onset of stimulation to the classical receptive field at the lower level so as to determine the latency of putative feedback effects. For neurons in the LGN, putative corticogeniculate effects will be confirmed as such when such effects are abolished by cryogenic blockade of activity in the retinotopically corresponding region of V1. For neurons in V1, putative feedback effects from V2 will be confirmed as such when such effects are abolished by ablation of retinotopically corresponding regions of V2 by subpial aspiration. From these data we expect to delineate the functional characteristics of processing from V2 to V1 and from V1 to LGN and thus better understand the role of these back- projecting systems in visual processing.
