While researchers have learned vast amounts about brain function at a cellular and molecular level over the past several years, we still have a poor understanding of the manner in which these phenomena give rise to behavior, and the way that neurobiological dysfunction creates psychiatric symptoms. This is particularly true for schizophrenia: while a large number of research studies have implicated hippocampus in the etiology of schizophrenia, and several possible neuroanatomic and biochemical abnormalities have been identified in this region, we still do not know how these findings, alone or in combination, lead to the clinical syndrome. Computational modeling is a research tool that allows one to make links between cellular phenomena and clinical behaviors by offering insights at the level of cells or cell ensembles as to how emergent properties arise. We have developed a computer simulation of a subsection of hippocampus CA1 incorporating 452 cells. Each cell is a biologically realistic model of a neuron, featuring an extensive dendritic arborization and Na+, Ca++, K+DR, K*AHp, K+c, and K+A channels, as well as AMPA, NMDA, and GABA synapses. We will use an elaborated version of this computational model as a tool to examine particular hypothesis regarding the hippocampal neuropathology of schizophrenia. Specifically, we will """"""""lesion"""""""" the model in a schizophrenogenic way by separately and in combination (a) altering aspects of the GABA system, (b) disrupting the glutamatergic system, and (c) increasing dopaminergic tone. We will examine two outcome measures: oscillatory activity and performance on context-dependent memory tasks;both of these have clinical correlations with schizophrenia. Finally, we will apply a number medications, including typical and atypical antipsychotics, and examine their effects on system functioning. We expect these studies to 1) offer mechanistic and system-level insights into the neuropathological basis of the illness;2) generate hypotheses that can subsequently be tested experimentally;and 3) identify potentially effective antipsychotic medications. Also, we hope that this study can make a contribution from a methodological point of view: though it involves only one brain area performing a discrete psychological task for a particular psychiatric illness, it is an approach that potentially can be applied to other brain regions and neuropsychiatric diseases.
