Dawes, Christopher
(2023)
Contributing factors to cognitive deficits in psychometric schizotypy: implications for schizophrenia.
PhD thesis, University of Nottingham.
Abstract
Chapter summaries
Chapter 1
Literature Review
Schizophrenia is a mental health disorder in which symptoms include hallucinations and delusions, flattened affect and reduced emotional expressiveness, and disorganised thoughts and speech. Patients commonly present reduced functional ability such as poorer quality of life, social integration, and vocational success. Much of the research into explanations for poorer functioning has pointed to cognition, specifically neurocognition (non-social cognition) and social cognition. However, together these processes explain only around 10% of the variance in functioning, with interventions based on these approaches having relatively moderate success. Recently, it has been proposed other factors may both affect cognitive task performance and the translation of ability to functioning. This Chapter reviews the current understanding of some of these factors, the influence of metacognition (‘thinking about thinking’), negative affect (emotions), amotivation and confounds of clinical status. Currently, our understanding of these factors relates to schizophrenia, especially metacognition, is incomplete.
Chapter 2
The influence of metacognition, negative affect, and motivation of neurocognitive task performance in non-clinical psychometric schizotypy.
The first experimental Chapter of this thesis aimed to assess the influence of metacognition, negative affect, amotivation and confounds of clinical status on neurocognition (non-social cognition). To do so, two neurocognitive tasks were chosen: Probabilistic Reversal Learning (PRL) and attentional set-shifting. While PRL is a broad measure of neurocognition that includes visual learning, working memory and attention, set-shifting is a specific assessment of effectively shifting attention from one visual dimension (e.g., colour) to another (e.g., shape). The influence of metacognition, negative affect, and amotivation were investigated by assessing these traits psychometrically and controlling for their influence when predicting performance. People varying in psychometric schizotypy (‘psychosis-proneness’) were assessed to mitigate the effects of clinical confounds. The results found that positive schizotypy (‘hallucinations and delusions’) was associated with poorer performance on the PRL task, specifically to punishing stimuli. Similarly, set-shifting performance was impaired. In contrast, disorganised schizotypy predicted improved performance on both tasks, with the reasons behind this being unclear. None of the potential mediators explained this relationship, but several methodological limitations of both the tasks and psychometric assessments meant this finding was not conclusive. Consequently, three additional experiments were designed to investigate this further in subsequent chapters. The first follow-up experiment adapted both of these neurocognitive tasks to also assess behavioural measures of metacognition. The second experiment extended this adaptation to social cognition to understand whether deficits were domain-specific or generalisable. Finally, the third experiment also adapted the PRL task into an immersive Virtual Reality task; assessing the influence of motivation.
Chapter 3
Metacognitive adaptations of neurocognitive tasks in psychometric schizotypy
The experimental design of the tasks in chapter 2 meant that it was unclear whether poorer performance was caused by cognition or metacognition. Moreover, self-reported metacognition scores were suggested to be too subjective, and so behavioural measures were needed. This chapter adapted these tasks by expanding the response dimensions to include accuracy judgements (cognition) and both confidence and acting on knowledge (both behavioural metacognition). For the metacognitive PRL task (M-PRL), positive schizotypy was again associated with poorer learning of punishing stimuli, supporting deficits in learning. Additionally, these participants also acted on this faulty knowledge and were overconfident when they believed they were incorrect, also supporting deficits in metacognition. However, there were no performance differences in the metacognitive set-shifting task and neither negative affect nor amotivation explained performance differences. These highlighted that while self-reported metacognition may not influence cognitive performance, behavioural assessments may be more powerful tools to assess it.
Chapter 4
High Schizotypy Predicts Emotion Recognition Deficits, but perhaps not poor real-world functioning.
This chapter expanded the current research of chapter 3 into the social cognitive domain of emotion recognition. Participants were presented with video clips of actors representing one of 14 different emotions through facial expression, body language, and nonsense syllables that participants were tasked to recognise. Negative schizotypy predicted poorer emotion recognition of negative emotions which was mediated by poorer metacognitive processes (‘I must control my thoughts at all times’). As these effects are specific to negative emotions, this may be explained by these faces producing negative internal states that participants feel they must control – leading to greater distractibility and avoidance. However, the behavioural metacognitive assessments found that those high in negative schizotypy did not act on this faulty knowledge; illustrating impaired social cognition that is not transferred to poor performance through intact metacognition. Consistent with chapter 2, disorganised schizotypy traits predicted improved performance to negative emotions, although the reasons for this are also unclear. There was also a trend-level association between positive schizotypy and negative emotion recognition, which is potentially consistent with poorer performance to punishing neurocognitive stimuli presented in chapter 2 and chapter 3. No measure predicted recognition of positive emotions and neither negative affect nor motivation mediated performance differences. While the effects of positive and disorganised schizotypy appear to span neurocognition and social cognition, negative schizotypy appears specific to social cognition.
Chapter 5
Can immersive Virtual Reality adaptations of cognitive tasks mitigate deficits in psychometric schizotypy? A 2D vs. Virtual Reality comparison
Chapter 2 found that motivation did not mediate performance deficits in schizotypy. However, the psychometric scale used was unvalidated, task-specific, and returned low internal consistency scores; meaning the influence of motivation was unclear. The current chapter instead manipulated motivation by adapting the PRL task of Chapter 2 into an immersive Virtual Reality assessment (VR-PRL). This adaptation was designed following open-ended participant feedback from Chapter 2, including greater task instruction clarity and contextualisation of task aims. This new sample did not differ in levels of schizotypy, negative affect, or overall task performance, but both pre-task and post-task motivation were much higher in the current study. All associations between schizotypy and performance, including the association between positive schizotypy and poorer learning of punishing cues, were mitigated. It was unclear if motivation was the underlying cause of change as other factors, such as increased comprehension and realism, may have impacted this relationship. Ultimately, it was suggested the VR adaptation may have reduced task-related anxiety, which subsequently meant punishing stimuli were no longer considered threatening.
Chapter 6
General discussion
The final Chapter discussed trends throughout these studies and implications for future research. Firstly, as both neurocognitive and social cognitive deficits were replicated in sub-clinical schizotypy, this provided indirect evidence that deficits in patients are not solely due to confounds of patient status (e.g., medication side effects). Across all tasks, these associations were specifically due to poorer performance towards negative stimuli only, which was suggested to be due to these stimuli evoking negative internal states, thereafter causing distraction or inattention. In contrast, disorganised schizotypy consistently predicted improved performance to negative stimuli, and while the reasons for this association could not be determined, it highlights that schizotypy should not be considered pathological. A potential explanation for these findings came from the lack of association to performance in the VR-task, which may have reduced task-related anxiety. Specifically, participants may not have perceived the stimuli as negative to the same extent, as the perceived ‘cost’ of incorrect performance was reduced by greater enjoyment. This may also be consistent with the findings of the M-PRL, which specifically pointed to deficits in perceived negative stimuli. Moreover, the gamification of cognitive tasks may have benefits beyond enjoyment and willingness to return, including more accurate assessments of applied cognitive ability. Although, more controlled manipulations of the VR-PRL are needed before these claims can be supported (e.g., manipulate only motivation). This contrasts with trait negative affect, which did not influence the association between schizotypy and performance in any analysis; suggesting schizotypy and negative affect may influence cognition through separate causal pathways. The results on metacognition presented interesting findings. Across all Chapters, schizotypy was associated with increased self-reflection, thought monitoring, and thought control thinking styles. However, the surveys are designed for clinical samples and define greater scrutiny of thoughts as maladaptive (e.g., rumination). Critically, without cognitive ability assessments, self-report ratings may be misleading, perhaps contributing to current literature inconsistencies. Comparing subjective and behavioural metacognition led to greater insights in the current thesis. Specifically, excessive Need to Control Thoughts (metacognition) partially explained why negative schizotypy was related to poorer social cognition, which is consistent with the above suggestions of avoiding negative internal states. This maladaptive metacognitive belief contrasted with the adaptive metacognitive control technique in the same individuals, wherein those high in negative schizotypy did not act on poorer cognition. Metacognition in the psychosis-spectrum is currently not well understood, but combining psychometric and behavioural measures may help disentangle contrasting findings.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of Correction version of the thesis for internal examiner checking]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "Correction version of the thesis for internal examiner checking")