TABLE OF CONTENTS In previous chapters we analyzed a simplified single cell theory of synaptic plasticity. The actual cortical network is very complex. It includes different cell types, intra-cortical interactions, and recurrent collaterals. Most of the input, both excitatory and inhibitory, to cortical cells in visual cortex arises from other cortical cells and not from thalamo-cortical projections. Therefore, theories of synaptic plasticity must address the effect of network dynamics. Networks of many interacting neurons are simulated in detail in Chapter 8. In this chapter we present two complementary forms of analysis that can help elucidate some of the effects of the cortical network, without resorting to complex simulations.
The cells of visual cortex form a network in which inhibitory and excitatory cells receive input from the lateral geniculate nucleus (LGN), from each other and feedback from higher cortical levels. Such a network is a highly complex, non-linear, system. Some of the key questions concern the necessary levels of connectivity, the importance of instantaneous signaling, stochastic behavior as well as the approach to equilibrium. For example, the amount of connectivity is of significance both from a theoretical point of view as well as for the possible embodiment of neural networks in electronic circuitry.
Interacting cortical networks can have two distinct effects on the outcome of synaptic plasticity. Such networks can: (1) alter the development of single cell properties. (2) effect the organization of single cell properties across the cortical map. Our analysis concentrates on addressing the former issue. Analysis in the previous...
