Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.730529
Title: Computational insights into the architecture and operation of the rat head direction cell system
Author: Page, Hector
ISNI:       0000 0004 6497 9932
Awarding Body: University of Oxford
Current Institution: University of Oxford
Date of Award: 2015
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Restricted access.
Access from Institution:
Abstract:
Head direction (HD) cells signal HD in the horizontal plane in a body-external (allothetic) frame of reference, and are a critical neural component of spatial cognition. This thesis uses computational modelling to investigate their properties. HD cell preferred firing directions are influenced depending on the discrepancy between sources of directional information, with allothetic information dominating up to a critical discrepancy and body-internal (idiothetic) information dominating after this. A computational mechanism is outlined which is capable of replicating these discrepancy-dependent changes in preferred firing direction. This mechanism depends on dynamic changes in HD cell responses to allothetic information and is shown to be affected by salience, defined as the degree to which a landmark has been previously experienced as a stable source of directional information. This model is then built on by providing a mechanism explaining how the preferred firing directions of any two given HD cells maintain their angular separation following idiothetic-allothetic discrepancies, a property known as isomapping. The accuracy of HD system updating solely via idiothetic information (path integration) in examined a model system. Two major limiting factors are found: the time neurons take to respond to stimulation, and the presence of direct reciprocal connectivity between HD cells. Finally a neural network model, based on known physiology, which does not incorporate direct HD to HD connectivity is proposed. This model is demonstrated to be capable of extremely accurate path integration.
Supervisor: Stringer, Simon Sponsor: Oxford Foundation for Theoretical Neuroscience and Artificial Intelligence
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.730529  DOI: Not available
Share: