 | Up a level |
Group by: Item Type | No Grouping
Jump to: Thesis
Number of items: 1.
Thesis
A sense of place: a model of synaptic plasticity in the hippocampus
http://jb4-2.eprints-hosting.org/2263
Artificial synaptic plasticity is a programming approach used in artificial neural simulations to replicate the change in efficacy between two synapses observed in biological neurons. This biological synaptic plasticity is thought to enable neurons to control the connections between them. This control is then thought to lead to complex behaviour such as path integration. This stochastic process of activity dependent biological synaptic plasticity forces groups of neurons to operate together. The operation of large subsets of neurons underlies the cognition and memory formation in animals, without which life could not flourish.<br/><br/>The most well studied region in the brain for synaptic plasticity is the hippocampus.This region was the first to display both long term potentiation, as well as long term depression. It has also been implicated in memory retention and has been shown to display spatial tuning. Furthermore, the discovery of place cells, and the more recent discovery of grid cells has created a surge of interest in the region. Entirely plausible models for grid field, place field, and memory formation have been suggested. The hippocampus could very well be the first brain region to be understood which does not represent purely sensory input.<br/><br/>This thesis applies the rules of activity dependent synaptic plasticity to the hippocampus by modelling the region in silicon. This model focuses on the head direction cells, grid cells, and place cells. The head direction cells are generated using rotational inputs. The grid cells are then generated using both head direction input and forward motion inputs. Finally, the place cells are created using grid cell inputs. To facilitate the construction of this model, a simulator has also been created.
Peyman Askari
http://jb4-2.eprints-hosting.org/2263
Artificial synaptic plasticity is a programming approach used in artificial neural simulations to replicate the change in efficacy between two synapses observed in biological neurons. This biological synaptic plasticity is thought to enable neurons to control the connections between them. This control is then thought to lead to complex behaviour such as path integration. This stochastic process of activity dependent biological synaptic plasticity forces groups of neurons to operate together. The operation of large subsets of neurons underlies the cognition and memory formation in animals, without which life could not flourish.<br/><br/>The most well studied region in the brain for synaptic plasticity is the hippocampus.This region was the first to display both long term potentiation, as well as long term depression. It has also been implicated in memory retention and has been shown to display spatial tuning. Furthermore, the discovery of place cells, and the more recent discovery of grid cells has created a surge of interest in the region. Entirely plausible models for grid field, place field, and memory formation have been suggested. The hippocampus could very well be the first brain region to be understood which does not represent purely sensory input.<br/><br/>This thesis applies the rules of activity dependent synaptic plasticity to the hippocampus by modelling the region in silicon. This model focuses on the head direction cells, grid cells, and place cells. The head direction cells are generated using rotational inputs. The grid cells are then generated using both head direction input and forward motion inputs. Finally, the place cells are created using grid cell inputs. To facilitate the construction of this model, a simulator has also been created.
Peyman Askari