The development of ocular dominance columns in normal and visually deprived monkeys

Abstract The main purpose of this study was to examine the normal postnatal development of ocular dominance columns in the striate cortex of the macaque monkey and to determine how this developmental process is influenced by monocular lid‐suture. The physiological pattern of ocular dominance was studied in long, tangential electrode penetrations. For anatomical demonstration of the distribution of afferents we relied principally on the transneuronal transport of [ 3 H]proline injected into one eye, and to a lesser extent on the Liesegang silver method. The effects of deprivation on cell size in the lateral geniculate nucleus (LGN) were also studied. Twenty‐six monkeys, divided into 5 groups (A–E), were used. The process of normal columnar development was examined in four monkeys aged from 1 to 6 weeks. At one week, there was both an anatomical and a physiological mixing of left‐ and right‐eye inputs to layer IVC, but the basic columnar pattern was evident, and some small regions were already monccular. At three weeks the columnar pattern resembled that seen in the adult, except for a suggestion that the borders between columns were not so sharply demarcated. By six weeks an adult degree of columnar segregation was established. A series of ten monkeys had monocular suture performed at successively later ages, ranging from 2 days to adult, and they were allowed to survive for a long period. It was found that deprivation begun at any age from birth to about 6 weeks had approximately the same effect: the afferents for the open eye formed greatly expanded columns in layer IVC, and the columns for the closed eye were shrunken and fragmented. In the layers above and below IVC the open eye dominated almost completely. At 10 weeks, closure had only a mild effect on columnar size in layer IVC. With closures at 7–14 months there was no change in the size of columns in layer IVC, but when stained with the Liesegang silver method the deprived‐eye columns were paler than those for the open eye, suggesting a lower density of fibers. These late deprivations still caused a shift of ocular dominance in the upper cortical layers but not such an extreme change as with earlier closures. Lid suture in an adult had no detectable anatomical or physiological effects. Even monocular enucleation, in an adult, failed to induce sprouting of the geniculocortical afferents for the remaining eye. In order to investigate the rate at which monocular deprivation produces its effects, six monkeys were examined after short periods of deprivation in infancy. Eye‐closure from birth to 3 weeks was sufficient to produce the full anatomical and physiological effects. A nine‐day closure at 3 weeks, followed by reopening of the eye, also produced the full effects. A two‐week period of deprivation begun at five weeks of age, however, caused relatively mild columnar expansion in layer IVC, suggesting that late closures may require more time to produce anatomical changes than early closures. The effects of reopening the deprived eye and closing the experienced eye (reverse suture) were studied in three monkeys. The initial lid sutures were performed in the first few days of life. With reverse suture a t 3 weeks the relative sizes of the two sets of columns were reversed. This anatomical reversal was, however, limited to the afferents from the parvocellular laminae of the LGN; the magnocellular afferents remained in the pattern induced by the early deprivation. The layers outside of IVC were strongly dominated by the initially deprived eye. Reverse suture at 6 weeks allowed an anatomical recovery of the parvocellular afferents to a normal columnar size, but not a complete reversal. Again, the magnocellular afferents for the initially deprived eye were not induced t o enlarge their territory. Reverse suture at 1 year of age did not lead to any recovery. One monkey was reared in complete darkness from 3 days of age until 7 weeks of age. This animal, studied autoradiographically, showed a normal columnar pattern in layer IVC. We draw four conclusions from these experiments. (i) Ocular dominance columns are only partially formed a t birth but develop rapidly in the first few weeks of postnatal life. The process of segregation of the left‐ and right‐eye afferents‐in layer IV occurs in the presence or absence of visual experience. (ii) This developmental process may be redirected by early monocular deprivation, causing the segregation of the two sets of afferents into columns of unequal width, (iii) A rearrangement of the afferents can be induced for a short period after their segregation is complete. This is true of both normal and deprived animals. (iv) The eye preference of neurons in the upper and lower layers may be changed even after plasticity in layer‐IVC is no longer detectable.