Physiological organization of callosal connections of a visual lateral suprasylvian cortical area in the cat

An extensive representation of the ipsilateral visual field is normally found in the posterior medial lateral suprasylvian area of the cat cortex (PMLS; see Ref. 39). This representation is binocular and can be mediated both by crossed connections originating from the temporal retinas and by callosal connections relaying visual information from one hemisphere to the opposite PMLS (see Ref. 35). In order to study the physiological organization of the callosal connections of PMLS, we recorded responses of PMLS neurons to visual stimuli presented to either the ipsilateral or the contralateral eye in cats with a midsagittal splitting of the optic chiasm. Due to the interruption of all crossed fibers from either nasal or temporal hemiretinas, all visual receptive fields of cortical neurons in this preparation depend on the uncrossed input from the temporal hemiretinas and, therefore, lie in the hemifield opposite to the eye used for stimulation. In eight split-chiasm cats, 81 neurons of 142 (57%) responded to stimulation of either eye; these binocular neurons, therefore, had receptive fields on both sides of the vertical meridian, one in the contralateral visual field (through the ipsilateral eye) and one in the ipsilateral visual field (through the conralateral eye). With a single exception (a neuron that was driven solely through the contralateral eye), all other neurons could be activated only through the ipsilateral eye. PMLS neurons that had receptive fields lying more than 10° away from the vertical meridian of the ipsilateral eye were monocular, i.e., they did not respond to stimulation of the contralateral eye. By contrast, most of the PMLS neurons with ipsilateral eye receptive fields apposed to or lying in the vicinity of the vertical meridian had a matching receptive field in the other eye and the other visual field. Receptive fields of the contralateral eye also abutted or lay very close to the vertical meridian. Monocular neurons tended to be concentrated in the upper part and binocular neurons tended to be concentrated in the lower part of PMLS, but several cases of spatial overlapping between the two types of neurons were observed. The receptive fields in the contralateral eye but not those in the ipsilateral eye disappeared after a posterior callosal section. The abolition of the responses of PMLS neurons to stimulation of the contralateral eye in split-chiasm cats was observed after both acute and chronic callosal sections. PMLS binocularity remained unaffected following a bilateral ablation of the superior colliculi. In cats with a unilateral optic tract section, PMLS neurons in the deafferented hemisphere responded to visual stimuli presented to either eye and their receptive fields, which must have depended on the corpus callosum, straddled, or were apposed to or lay in the close vicinity of the vertical meridian. Taken together with anatomical results suggesting that the callosal connections of PMLS are widespread throughout this cortical area (46), the present physiological results indicate that nonetheless, in each portion of PMLS, callosal connections are limited to those neurons that are concerned with the representation of the vertical meridian of the visual field. Due to the large size of receptive fields of PMLS neurons, the convergence of intrahemispheric and interhemispheric visual afferents on these neurons results in an extensive and homogeneous bilateral representation of the visual field, virtually devoid of discontinuities at the midline. We conclude that even in those visual cortical areas where the retinotopic representation is coarse and the callosal connections appear to be diffuse, the 'vertical meridian rule' determines the relations between the retinotopic map and the interhemispheric projections.