Modeling the hippocampus : finely controlled memory storage using spiking neurons

Abstract

The hippocampus, an area in the temporal lobe of the mammalian brain, participates in the storage of personal memories and life events. As such traumatic memories and the consequent symptoms of post-traumatic stress are thought to be stored or at least processedin the hippocampus. While a fundamental understanding of a traumatic memory is still elusive, studying the physiology and functional properties of the hippocampus are anessential first step. Towards that goal, I developed a detailed computational model of the hippocampus. The model included the important effects of the neuromodulator Acetylcholine that switches the hippocampal network between the memory encoding state and the memory retrieval state. In the first study, I examined the mechanisms for controlling runaway excitation in the model. The results indicated different mechanisms for controlling runaway excitation in the memory encoding state as opposed to the memory retrieval state of the circuit. These findings produced the first functionally-based categorization of seizures in animals and humans, and may inspire specific treatments for these types of seizures. The second study examined the underpinnings of the rhythmic activity of the hippocampus. These oscillations in the theta range (4-12 Hz) are theorize to play a major role in the memory functions and in processing sequences of events and actions in both place and time. We found the generation of theta rhythmic activity to be best described as a product of multiple interacting generators. Importantly, we found differences in theta generation depending on the functional state of the hippocampus. Finally, the third study detailed the rules of the complex interactions between these multiple theta generators in the circuit. Our results shed more light on the role of specific components in the hippocampal circuit to maintain its function in both health and disease states.