Neurobiological mechanisms of hallucinations in schizophrenia

Abstract

All perception is a construct of the brain. Yet occasionally, sensory constructions emerge without origin in the physical world and are experienced as hallucinations. Hallucinations occur transdiagnostically, cross-culturally, and in all sensory modalities. They are common in people with schizophrenia, presenting in 60-80% of patients. Despite over 20 years of active neuroimaging research on hallucinations, the neural systems supporting these anomalous perceptual experiences remain disputed. This dissertation investigates the neurobiology of hallucinations, integrating research across structural and functional magnetic resonance imaging (MRI) to elucidate how hallucinations, chiefly in the context of schizophrenia, are supported by the brain, drawing on MRI indices of neurodevelopment. I introduce the phenomenon of hallucinations and motivate the utility of MRI for studying hallucinations. Considering their prevalence in other medical conditions, I conduct a meta-analysis and systematic review of the structural brain basis of hallucinations across diagnoses, primarily schizophrenia spectrum disorders and Parkinson’s disease. This illustrated distinct neuroanatomical organizations of grey matter associated with hallucinations that occur in neurodevelopmental compared to neurodegenerative disorders, which I hypothesise constitute at least two distinct mechanisms. Focussing on the neurodevelopmental mechanism characterized by fronto-temporal and insular grey matter reductions, I turn to the contribution of cortical sulcation, a product of second and third trimester neurodevelopmental processes, which has been robustly implicated in schizophrenia pathology, and, more recently, in hallucinations. Sulcal patterns derived from structural MRI provide a proxy in adulthood for early brain development. I studied two independent datasets of patients with schizophrenia who underwent clinical assessment and 3T MRI from the United Kingdom and Shanghai, China, stratified into those with and without hallucinations, and healthy controls from Shanghai. I first replicate the finding that left hemisphere paracingulate sulcus (PCS) length is reduced in patients who experience hallucinations, then demonstrate similar associations for superior temporal sulcus depth. Length and depth alterations occurred with focal deviations in sulcal geometry. The interindividual and interhemispheric variability of the PCS necessitated the development of semi-automated methods to characterize its morphology and validation to a manual protocol. I used structural covariance networks of the local gyrification index to investigate how specific sulcal deviations relate to global neurodevelopmental coordination, demonstrating that hallucinations correspond to increased covariance within and between salience and auditory networks. Hypothesizing structure-function relationships, I analyse resting-state functional MRI data from the same datasets described, finding significant interactions between PCS length and hallucinations status, but no main effects. There were no effects of hallucination status on salience and auditory network connectivity or in graph theoretical measures of connectivity, suggesting that resting-state connectivity is not a trait marker for hallucinations. Together, the discovery of neurodevelopmental alterations contributing to hallucinations provides mechanistic insight into the pathological consequences of prenatal origins. The interaction of sulcal alterations and hallucination status are associated with connectivity, which may have a role in the pathophysiology of hallucinations. I provide clear predictions and recommendations for future research.