The existing literature on symptoms of thought disorder in schizophrenia does not present a clear picture of the underlying mechanisms involved in these symptoms. This thesis focused on a specific subset of theoretically defined symptoms, namely, discourse coherence disturbance which include symptoms which reflect an inability to maintain a discourse plan. This thesis comprises three studies employing cognitive or physiological measures in the attempt to discover the physiological and cognitive

more »

... mechanisms underlying symptoms of discourse coherence disturbance in schizophrenia. The first study was a series of case studies investigating cognitive performance in patients with schizophrenia exhibiting symptoms of discourse coherence disturbance. Eight patients with a DSM-TV and ICD-10 diagnosis of schizophrenia exhibiting varying degrees of discourse coherence disturbance were compared to a group of nineteen comparison volunteers. Based on four current theories of discourse coherence disturbance, a range of cognitive areas, including short and long-term memory, semantic priming, contextual naming, working memory, verbal fluency, error-monitoring, and inhibition were explored. Patients with discourse coherence disturbance differed from controls and patients without discourse coherence disturbance on measures of central executive processes of working memory, verbal fluency, and error-monitoring. These results were consistent with McGrath's (1991) theory that discourse coherence disturbance is due to a disconnection between thought and action reflected in disconnectivity between frontal and temporoparietal regions. The second and third studies were designed to further investigate the predictions of McGrath's theory. The second study explored the role of the dorsolateral prefrontal cortex in patients with discourse coherence disturbance. Five patients with schizophrenia with varying degrees of discourse coherence disturbance and sixteen healthy comparison volunteers participated in an event-related fMRI study exploring activation in the dorsolateral prefrontal cortex during a working m e m o r y task. This study showed that although patients with discourse coherence disturbance performed poorly on tasks typically associated with activation in the dorsolateral prefrontal cortex, such as working memory tasks, they showed normal activation of the dorsolateral prefrontal cortex while performing these tasks. The third study aimed to explore the interaction between the dorsolateral prefrontal cortex and other brain regions, as well as exploring the nature of the error-monitoring deficit in patients with discourse coherence disturbance. Six patients with discourse coherence disturbance, nine patients without discourse coherence disturbance, and fifteen healthy comparison volunteers participated in an ERP testing session investigating error negativity (ERN) in relation to performance on a reaction time task. The results of this study showed that patients with discourse coherence disturbance are able to detect errors, but are unable to correct them which is consistent with McGrath's (1991) hypothesised disconnection between thought and action. Further, these patients showed a similar ERN profile to patients with lesions of the dorsolateral prefrontal cortex. These results suggest that impaired interaction between the anterior cingulate and the dorsolateral prefrontal cortex may account for symptoms and cognitive deficits found in patients with discourse coherence disturbance. Given the role of the anterior cingulate in modulating fronto-temporal interactions, it is proposed that there is a disruption to dorsolateral prefrontal networks resulting from impaired communication between the anterior cingulate and dorsolateral prefrontal cortex.