Models of neuronal subcircuits in the crustacean cardiac ganglion and the mammalian perirhinal cortex

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT REQUEST OF AUTHOR.] This dissertation deals with four related studies that involve simulations of computational models of neurons and neuronal networks at varying levels of complexity and biological realism. First, a framework for developing reduced order firing-rate network models, which incorporates reduced biophysical mechanisms of single cell dynamics and calcium-dependent synaptic learning, was developed with a biophysical network model of the lateral amygdala (LA) as a benchmark. Experiments were performed that demonstrate the utility of the firing-rate network and further explore the properties of the LA for fear conditioning and extinction. Next, a novel model of the crustacean cardiac ganglion (CG) was described. This model was used to explore potential mechanisms of homeostatic regulation. At the single-cell level, it was found that co-regulation specific ion currents were required to preserve function in the CG motorneurons. Then, using a full network model of the CG, it was shown how, due to strong gap junction coupling among the cells of the network, that regulation of potassium currents can modulate whole-network bursting. Finally, a large network model of the mammalian perirhinal cortex (PRC) was developed and studied. The known properties of the PRC were incorporated into a biophysical model that accurately reproduces and predicts the necessary features of the PRC that support the simultaneous performance of novelty recognition and stimulus association, two memory functions in which the PRC is known to be involved.