Spatial organization of the peripheral input to area 1 cell columns. II. The forelimb representation achieved by a mosaic of segregates.

The view (advanced in the previous paper) that the topographic organization in the forelimb region of area 1 of Macaca fascicularis monkeys should be regarded as a mosaic of discrete units--segregates--is evaluated. It is found that in all cortical layers the RFs sampled within a single segregate possess a wide variety of sizes and configurations, and occupy a wide variety of positions on the skin relative to the segregate RF center (the latter is a small skin area common to RFs of all neurons in the segregate). This within-segregate RF variability is structured so that the position of RFs of neurons sampled from different sectors of a segregate exhibits little, if any, systematic shift. The skin area that provides sensory input to any given area 1 segregate (estimated by the aggregate of the RFs sampled from that segregate) is extensive. This 'segregate RF', however, is not homogenous: i.e. central regions of the segregate RF are included in the RFs of a higher fraction of the neurons in the segregate than are peripheral regions. Segregate RFs appear particularly extensive when their size is compared with a relatively small shift in skin position that takes place when one shifts from one segregate to the next. Consequently, the skin areas that provide the sensory inputs to neighboring segregates overlap to a very large degree; and even fairly remote segregates in area 1 can receive a substantial common input. The arrangement of segregate RFs in area 1 is, in general, somatotopic. Nevertheless, the local relationships that are obtained among different segregates can deviate significantly from a strictly somatotopic pattern. The area 1 topographic organization detected in this study appears to differ substantially from that described by other investigators. A more detailed analysis suggests that most of the major differences between this and the previous descriptions of area 1 organization may largely be attributable to the different experimental conditions employed, and that the results of this study and those described by workers using different mapping methods are, in fact, generally compatible. Finally, it is suggested that the mosaic pattern of the topographic organization detected in area 1 may reflect the bundled nature of its afferent input.