We have previously described aspects of neurotic mental pathology, in terms of its relation to memory function and proposed a neural network model, whereby neurotic behavior is described as an associative memory process. Modules corresponding to sensory and symbolic memories interact, representing unconscious and conscious mental processes. Our main contribution with respect to current work regarding machine models of consciousness is to propose a neuronal associative memory mechanism that describes conscious and unconscious memory activity, based on Freud's description of neurosis, which illustrates how access to symbolization mechanisms is essential for consciousness. We modeled memory access dynamics by the Boltzmann machine (BM) and Generalized Simulated Annealing (GSA) and investigated the effects of using both on network associativity.