Abstract
Impaired cognition is a hallmark of schizophrenia. These impairments have been consistently linked to functional capacity and other related factors, such as insight and symptom severity. Recently it has been shown that cognitive impairments are largely unresponsive to pharmacological therapy and cognitive training has limited efficacy. Thus, to better understand the nature of these impairments, other important constructs should be considered. As I investigated in this thesis, the roles ofbrain structure, metacognition, cognitive, insight, and clinical insight are potentially important factors that should be considered.
The primary aim of this thesis was to investigate how brain structural parameters relate to measures of cognitive performance in patients with schizophrenia (SZH), as well other related constructs, namely insight, and metacognition.To achieve this, I analysed data from various relevant neuroimaging datasets, patients with first-episode psychosis (FEP) were also studied. In chapter 2, I investigated whether symptom severity and cognitive performance have differential brain structural (cortical thickness) underpinnings in SZH. Results revealed robust relationships between cognitive domains and structure within specific brain regions, including links between better cognitive performance and thicker cortex in transverse temporal and frontal regions.. This was independent of symptom severity. In chapter 3, I tested gray matter volume correlates of cognitive performance and clinical insight in SZH. Again, results showed significant relationships between brain volumes and cognition, but not with clinical insight. Better performance in executive function was associated with larger gray matter volume in bilateral DLPFC, bilateral VLPFC, bilateral middle temporal, left superior temporal, and right inferior temporal regions, none of these relationships were observed in healthy controls (CON). Chapter 4 investigated cognitive clusters in SZH and their siblings, and relationships between brain volume and cluster assignment. Firstly, I identified three distinct cognitive clusters based on age- and gender-adjusted cognitive domain scores in a sample of SZH, CON, and siblings of both groups. These were a ‘neuropsychologically normal’ cluster, a ‘cognitively impaired cluster’, and an ‘intermediate cluster’. Not surprisingly, the majority of SZH were assigned to the cognitively impaired cluster while most CON were assigned to the neuropsychologically normal cluster. Greater right inferior temporal volume distinguished the normal cluster from the more impaired clusters. Importantly, the observed brain volume differences between SZH and controls disappeared after adjustment for cluster assignment. In chapter 5, I explored brain structural correlates of metacognition in FEP. I found that metacognitive accuracy correlated with hippocampal volume and brain microstructural indices in FEP (but not controls). In the last chapter, I tested brain structural correlates of functional capacity in FEP. I observed that poorer functional capacity correlated with reduced superior frontal volume and lower fractional anisotropy in the left inferior longitudinal fasciculus. Overall, these results enhance our understanding of the brain structural correlates of neurocognition, metacognition, cognitive insight, and clinical insight in SZH and FEP. Thus, the results reported here may help to inform diagnosis and treatment and pave the way for future longitudinal research to better understand the underlying neural mechanisms of these dynamic phenomena.