Convergence and divergence in the afferent projections to cat area 17

Abstract

We have examined the topography of the afferent connections to area 17 in the cat by means of double retrograde label tracing techniques. Injections of two fluorescent retrograde tracers, diamidino yellow and fast blue, were made with variable separations in area 17 and the spatial distributions of the resulting populations of labeled cells examined in afferent cortical areas and subcortical structures. When injections were separated rostrocaudally, the topographic organizations of the projections were characterized quantitatively with two graphic methods: the labeling density curve and the connectivity graph. The labeling density curve measures labeled neuron density in successive rostrocaudal sections, whereas the connectivity graph provides a two-dimensional model of the topography of a given connectivity. The connectivity graph makes it possible to define two parameters that characterize the topography of the connection: the convergence and the divergence. The convergence is defined as the extent of an afferent structure that contains neurons converging on a line normal to the cortical surface in area 17. The divergence is the extent of area 17 that is innervated by neurons contained in an infinitely small region of the afferent structure. The results show that a number of subcortical structures project to area 17 in a nontopographic manner, i.e., that in each of these structures neurons contained in an infinitely small region send projections to the whole of area 17 and that a line normal to the surface of area 17 is innervated by neurons distributed throughout the afferent structure in question. Nontopographic projections are found from the intralaminar nuclei, the ventral mesencephalic tegmental region, the diagonal band of Broca, and the locus coeruleus. All remaining subcortical structures and cortical areas send topographically organized projections to area 17. The extent of the convergence and divergence, however, varies between structures. Only the projection from the A laminae of the LGN was found to approximate a point-to-point projection with a convergence of 0.4 mm and 2 mm in divergence. Much larger convergence and divergence values are found in the projections from the claustrum and the cortical areas. For example, the divergence reaches 20 mm for the projections from area 20 or from the anterior part of the lateral suprasylvian sulcus. Knowing the convergence and divergence values and the retinotopic organizations of area 17 and a number of its afferents, it becomes possible to test whether connections in the visual system link regions representing the same zone of the visual field. This is the case for the geniculostriate projection only and does not apply for any of the other topographically organized afferents to area 17. For example, the region of area 18 containing neurons projecting to a line normal to the surface of area 17 represents 15 degrees of visual angle, whereas the neurons empaled by this line in area 17 represent only 5 degrees of visual angle.