UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional brain network activity predicts behaviour in schizophrenia Luk, Jessica


Impairment in language association and detriments in executive functioning are characteristic features of schizophrenia. Available external measures are often used to describe the profile of schizophrenia, but whether these measures can reflect underlying neuropathology of schizophrenia remains to be answered. In the current research, we attempt to 1) identify functional brain networks underlying a semantic association task and 2) relate functional brain activity to external measures of behaviour, symptomatology, and neurocognition. A data-driven exploratory method was employed to investigate the functional brain networks underlying a semantic association and word recall task, and how these networks may be impaired in schizophrenia. This revealed three distinct networks: a Language Network (LN), Default Mode Network (DMN), and Volitional Attention/Response Network (VAN/RN). All three networks exhibited the greatest magnitude of activation or deactivation under the distant association, forgotten condition. The activation of the DMN significantly differed between patients and controls, where schizophrenia patients showed a greater sensitivity to recall relative to controls, reflected in an increased magnitude of deactivation of the DMN. One interpretation posits that the distant association and forgotten condition was more difficult, and required a greater shift of available cognitive resources away from the DMN (a task-negative network) to the LN and VAN/RN (task-positive networks). A greater shift in resources towards the LN and VAN/RN may have come at a cost of background processes of encoding and memory, where few resources were available for making word pair memories. Analyses relating brain networks to external measures of cognition showed that, while brain networks were able to significantly predict scanner-based behavioural measures of reaction time, accuracy, and recall, they were unable to predict either symptom or neurocognitive measures. Specifically, a desired balance between networks and brain networks acting in conjunction was able to significantly relate to in-scanner behaviour. Taken together, these studies provide evidence that the functional brain networks may be disordered in schizophrenia, and sheds light on how functional network activation relates to external measures of behaviour and symptoms. The results of these series of studies may have implications for informing decisions for personalized targets in future therapeutic treatments of schizophrenia.

Item Media

Item Citations and Data


Attribution-NonCommercial-NoDerivatives 4.0 International