The aim of this thesis was to test predictive coding as a model of cortical organization and function using a specific brain response, the mismatch negativity (MMN), and a novel tool for connectivity analysis, dynamic causal modelling (DCM). Predictive coding models state that the brain perceives and makes inferences about the world by recursively updating predictions about sensory input. Thus, perception would result from comparing bottom-up input from the environment with top-down predictions. The generation of the MMN, an event- related response elicited by violations in the regularity of a structured auditory sequence, has been discussed extensively in the literature. This thesis discusses the generation of the MMN in the light of predictive coding, in other words, the MMN could reflect prediction error, occurring whenever the current input does not match a previously learnt rule. This interpretation is tested using DCM, a methodological approach which assumes the activity in one cortical area is caused by the activity in another cortical area. In brief, this thesis assesses the validity of DCM, shows the usefulness of DCM in explaining how cortical activity is expressed at the scalp level and exploits the potential of DCM for testing hierarchical models underlying the MMN. The first part of this thesis is concerned with technical issues and establishing the validity of DCM. The second part addresses hierarchical cortical organization in MMN generation, plausible network models or mechanisms underlying the MMN, and finally, the effect of repetition or learning on the connectivity parameters of the causal model.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:631800 |
| Date | January 2008 |
| Creators | Figueiredo Garrido, M. Isabel |
| Publisher | University College London (University of London) |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://discovery.ucl.ac.uk/1444174/ |
Page generated in 0.0017 seconds