- Title
- A multi-modal computational study of cognitive and emotional dysfunction in psychosis
- Creator
- Koussis, Nikitas Chris
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2024
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Psychotic disorders are debilitating conditions involving periods of delusions, hallucinations and disordered thinking, known as psychosis. Psychotic disorders, such as schizophrenia (SCZ), can also manifest in disruptions to cognitive performance leading to emotional and social difficulties. Despite advances in the remediation of delusions and hallucinations, understanding this constellation of illnesses is lacking, particularly regarding the cognitive and emotional deficits. This thesis proposes a transdiagnostic and multi-modal approach, with a focus on SCZ, to the cognitive control system (CCS) and the striatum during affective response inhibition to understand cognitive and emotional deficits in psychotic disorders. This thesis investigates the neural responses of the Affective Go/No-Go task in functional MRI (fMRI) in a transdiagnostic cohort of 30 individuals with early psychosis (EPs) and matched controls. The thesis develops multi-disciplinary knowledge from neuroscience, physics and mathematics that contributes to the understanding of psychotic symptoms, cognitive, social and behavioural deficits in psychotic disorders. Recommendations are made for future work, as well as translational applications of the findings. This thesis identifies the CCS and striatal regions involved in response inhibition, discovering that EPs have increased fMRI signal in posterior insula (PI), a region responsible for emotional processing, integrating salient information with the anterior insula (AI). There are no significant differences in the accuracy and reaction time of EPs and matched controls. Using a modelling framework known as Dynamic Causal Modelling (DCM) it is shown that to uphold performance, EPs recruit more top-down control from the CCS on AI-PI connectivity. The amount of top-down control is significantly associated with negative symptoms, indicating impacts on emotional conflict-resolution driven by cognitive control regions. Via their influence on AI-PI emotional integration, such processes may lead to the negative symptoms of psychosis, including dysfunctional emotion processing. Expansion of this state-based finding to large, distributed systems involved in cognitive function such as cortico-striatal communication is then undertaken. This system is necessary for many behavioural processes, including cognitive control. This system’s expression during Affective No-Go is studied with psychophysiological interactions (PPI). PPI is then combined with the “functional connectopic mapping” (FCM) approach, deriving maps of spatial transitions (gradients) of cortical-striatal function during cognitive control. For statistical inference on these smooth spatial patterns (or gradients), this thesis develops a method drawing on mathematics and physics approaches to smooth maps that employ geometric basis sets (eigenmodes). A novel method uses the geometry of the brain to decompose a pattern of activity into these eigenmodes. This representation is resampled to provide surrogate patterns. It is shown that this method called Eigenstrapping provides improved statistical control of false positives and broadens the generalisability in cortical and subcortical brain maps over existing methods. Combining eigenstrapping and the PPI+FCM approach, this thesis next reports lateralisation of a cognitive control gradient in EPs. This expresses as a left/right asymmetry of cortico-striatal communication. The maximum and minimum scores (denoting the degree of asymmetry) in the cognitive control gradient are correlated with higher psychotic symptoms. This indicates a distributed and lateralised process of aberrant cognitive control impacting the ability of the striatum to perform response inhibition. Therefore, it is speculated that for the EPs to retain performance of the task, they recruit top-down cognitive control in the right hemisphere of the brain. Notably, this was the same hemisphere that the DCM study demonstrated aberrant CCS modulation of insula connectivity. By integrating network connectivity findings and maps of distributed function in a transdiagnostic EP cohort, this thesis reports that the functional connectivity of the CCS and the striatum are deeply implicated in early psychotic disorders. Given the transdiagnostic nature of these findings, it is possible that this is a state marker of psychotic illness. Future work could determine the link of these findings to neurobiology such as neurotransmitters implicated in psychotic disorders (e.g., dopamine and glutamate). Several potential clinical outcomes of these multi-modal imaging and computational approaches are recommended: Therapeutic interventions involving the CCS-striatal system could lead to remediation of both positive and negative symptoms.
- Subject
- schizophrenia; negative symptoms; positive symptoms; cognitive control; prediction error; striatum; manifold learning
- Identifier
- http://hdl.handle.net/1959.13/1512407
- Identifier
- uon:56613
- Rights
- Copyright 2024 Nikitas Chris Koussis
- Language
- eng
- Full Text
- Hits: 331
- Visitors: 340
- Downloads: 11
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Thesis | 20 MB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Abstract | 318 KB | Adobe Acrobat PDF | View Details Download |