Modelling Effective Connectivity in Functional Magnetic Resonance Imaging Data by State Space Models

This dissertation develops a new approach using state space models for effective connectivity analysis. The proposed approach integrates both activation and connectivity analysis into one single model. It takes the temporal information in the fMRI data into consideration explicitly and can be used to model the relationship among multiple region-of-interests in the brain. Dynamic characteristics of connectivity can also be investigated. The proposed approach is very flexible and can be used to test for commonalities and differences in the connectivity patterns across subjects. An empirical application for testing an attentional control network from a fMRI experiment is presented. The results provide partial support that higher order control (middle frontal gyrus, MFG) influences the two sites of control (lingual gyrus, LG, and middle occipital gyrus, MOG, in this experiment). There is also strong evidence indicating heterogeneity in the subjects' connectivity between LG and MFG. Moreover, the strength of the connectivity between LG and MFG, and between MOG and MFG show decreasing trend during the experiment.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Psychology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Ho, Moon-Ho Ringo

Contributors dc:contributor

• Hubert, Lawrence J.

• Biology, Neuroscience

Language dc:language

eng

Identifier

(MiAaPQ)AAI3101864

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/82046