Specialization and integration in brain networks underlying cognitive control in healthy individuals and patients with schizophrenia

© 2013 Dr. Ian Herbert HardingCognitive control lies at the foundation of dynamic and adaptive human behaviour. Through the flexible top-down regulation of lower-order processes, cognitive control operations serve to direct the perceptual, motor, and other cognitive resources of the brain in response to ever changing environmental demands and behavioural goals. These abilities, including cognitive interference resolution and working memory operations, rely on a common set of brain regions located within the prefrontal and parietal association cortices, together forming the frontoparietal control network. The component regions of this network are variously responsible for encoding and updating goal and context representations, signalling motivational salience, monitoring action-outcomes, and discriminating amongst ambiguous perceptual information and behavioural contingencies. Meaningful and coherent behaviour is dependent both on information processing within each of these regions (specialization) and the amalgamation of function across the network (integration). Although the frontoparietal control network is well defined and has been widely investigated, little is yet known about how it operates when faced with multiple concurrent control demands, as would be expected in real-world environments. Moreover, the shared and unique nature of connectivity patterns within this common brain network across different cognitive control processes is currently unknown. In schizophrenia patients, dysfunction in cognitive control abilities is endemic and is thought to lie at the core of the significant disability faced by patients suffering from the illness. Current theories variously propose that abnormalities in the integrity and efficiency of neural functioning, as well as in the normal integration of activity within the frontoparietal control network may underlie these deficits. This thesis presents a series of experiments exploring the activation and connectivity patterns defining the frontoparietal network as a function of distinct cognitive control demands. Functional magnetic resonance imaging (fMRI) data was acquired during performance of a novel cognitive paradigm in which cognitive interference and working memory demands were manipulated in a factorial manner. Functional activations and effective connectivity were assessed using statistical parametric mapping (SPM) and dynamic causal modelling (DCM) techniques, respectively. Investigations were first undertaken in a group of healthy adults, and followed thereafter by a characterization of differences evident in a cohort of patients suffering from schizophrenia. Taken together, the frontoparietal system was found to be highly adaptable, widely interconnected, and characterized by both common and unique dynamics in response to different cognitive control demands. These characteristics were generally shared in patients with schizophrenia, although distinct decrements in inter-regional interactions within the prefrontal cortex were observed. The outcomes of this work serve to build upon, and in some cases challenge, current mechanistic models of cognitive functioning and pathophysiological processes, and inform compelling future research directions in the fields of cognitive neuroscience and psychiatry