Local intracortical connections in the cat's visual cortex: Postnatal development and plasticity

1. We have studied the nature, development, and plasticity of local intracortical interactions by examining the visual responses from pairs of cells in the visual cortex of anesthetized and paralyzed adult cats and kittens at postnatal age 4 wk. Simultaneous discharge from nearby cells was analyzed by the cross-correlation method to infer three types of neuronal interactions: monosynaptic excitation, monosynaptic inhibition, and polysynaptic excitation. 2. Among cell pairs that exhibit correlated discharge, the nature of these interactions is similar in kittens and adults. For example, the mean monosynaptic delay is the same for cell pairs in the adult cat and the kitten: 0.8 ms. The primary difference in the distribution of neural interactions is the increased prevalence of inhibitory interactions among adult cat cell pairs compared with kitten cell pairs. However, in both age groups and in nearly all laminae, polysynaptic interactions are the most commonly observed type of interaction. 3. Neural interactions revealed by cross-correlation analysis also were studied with respect to the receptive field properties of the cells. The hierarchical theory proposed by Hubel and Wiesel suggests that monosynaptic excitation should be seen from simple to complex cells. It further suggests that complex cells do not provide monosynaptic excitation either to simple or to other complex cells. However, we find no cases of monosynaptic excitation from simple to complex cells in the adult cat. Moreover, we find explicitly antihierarchical connections, i.e., excitatory connections from complex to simple cells, in both kittens and adults. The excitatory influence of complex cells on simple cell receptive field properties is not incorporated into current models of receptive field structure formation. Our results suggest that complex cells might have an important modulatory role in simple cell discharge. 4. Plasticity of intracortical synaptic connections was studied by examining the dynamics of monosynaptic peaks with cross-correlograms. We examined Hebb's seminal hypothesis regarding synaptic plasticity that the excitatory connections between two neurons should strengthen with simultaneous activity. To test directly whether Hebbian plasticity can be rapidly induced by stimulus-evoked activity, the strength of monosynaptic excitation between pairs of neurons was monitored during periods of visual stimulation lasting ≤2 h. Synaptic strength was computed by summing the cross-correlogram bins containing the monosynaptic peak. Transient increases in monosynaptic peak area are found among cell pairs linked by monosynaptic excitation after 8-15 min of visual stimulation. The duration and amplitude of these changes are similar in adult cats and kittens. In about half of the cell pairs showing such transients, the increases in correlated discharge are not related to variations in the individual discharge rates of the two cells. 5. Changes in the amount of correlated discharge associated with monosynaptic excitation imply that the actual efficacy of the connections between cortical cells can increase during the course of visual stimulation in both adult cats and kittens. However, no sustained increases in correlated discharge, as would be predicted by a strictly Hebbian model of plasticity, are observed. Furthermore, there are no obvious differences between kittens, for which plasticity should be relatively high, and adult cats. These findings suggest that lasting Hebbian changes in synaptic efficacy may depend on additional inputs, such as attention-based modulation that were not present in our preparation. It is also possible that the time course required to cause clear alterations is considerably longer than what we were able to employ with the techniques of this study.