Soc. Neurosci. Abstr. 1993.11


I. Fujita*,1,2 and T. Fujita2, 1PRESTO, Research Development Corporation of Japan, and 2Laboratory for Neural Information Processing, Frontier Research Program, RIKEN, Wako, Saitama 351-01, Japan

Neurons in area TE of the monkey inferotemporal cortex are arranged into columns according to visual features of objects to which they respond (Fujita et al., Nature, 360:343-346,1992). Multiple columns appear to be selective for similar object features, because 2 or 3 separate clusters of cells respond to similar stimuli along a tangential recording penetration. To explore anatomical substrate for this organization, we made extracellular iontophoretic injections of biocytin into area TE of Japanese macaques.
After injections into single or restricted layers (2-3, 3, 3-4, 4, 5 or 6), most retrogradely labeled cells were found above or below the injection site to make a columnar appearance. Labeled cells were distributed through layers 2-6, except that those after layer 6 injection avoided layer 4. In addition to this radial labeling, a smaller number of cells were labeled lateral to the injection site, especially in layers 2 and 3. These cells, mostly pyramidal type, tend to cluster within terminal arbors of axons which radiated horizontally from the injection site
A single injection produced more than 15 patches of terminal arbors in successful cases. The diameter of the patches was 0.5+0.2 mm (range: 0.3-0.9 mm). The mean center-to-center distance between adjacent patches was 0.7 mm (range: 0.5-1.8 mm). The distribution of the patches around the injection site was elongated in a direction parallel to the superior temporal sulcus.
Strong vertical connections between layers may contribute to the shared stimulus selectivity within a column. The results show that intrinsic horizontal axons in area TE arborize in a clustered manner as in the striate and prestriate cortices. Size and spacing of the terminal patches are, however, larger than those reported for V1 and V2, and roughly correspond to those of TE cell clusters with similar selectivity measured in physiological studies. We thus suggest that horizontal axons may link columns with similar stimulus selectivity.