The pathophysiological mechanisms underlying metamemory disruptions

Abstract

Aberrations to metacognition—the ability to monitor and control one’s own cognitive processes—are a feature of poor mental health. While empirical research has largely focused on establishing an association between disruptions to metacognition and psychopathology, the origins of inter-individual differences in metacognition are yet to be determined. The Metacognitive Model of PTSD proposes that maladaptive metacognitive beliefs are activated following exposure to a traumatic event, driving and maintaining symptoms of poor mental health. In this thesis, I investigate the role of stress as one pathway through which alterations to metacognition may arise. Emerging evidence that metacognitive deficits are common across multiple psychiatric disorder categories suggests that a transdiagnostic approach may better delineate the clinical manifestations of metacognitive disruptions. To further investigate the relationship between metacognition and psychopathology, measures of three transdiagnostic symptom clusters (‘anxious-depression’, ‘compulsivity and intrusive thought’, and ‘social withdrawal’) were incorporated into Chapters 2 and 3. Throughout the thesis, metacognitive metrics were quantified using a metamemory task that involved the encoding of stimuli followed by a test requiring participants to make a first-order response (‘Have you seen this image before?’), followed by a second-order introspective confidence rating assigned to that response (‘How confident are you in your choice?’). In Chapter 2, I began by investigating whether metamemory disruptions are present following trauma, differing systematically between individuals with versus without elevated post-traumatic stress symptoms. It was hypothesised that individuals in the clinical sample would exhibit poorer metacognitive performance relative to those in the control group. To further decompose the relationship between metamemory and psychopathology, metacognitive efficiency was predicted from scores on three transdiagnostic factors. Contrary to our hypothesis, metamemory performance did not differ between individuals with versus without elevated symptoms following trauma. However, higher scores on the compulsivity and intrusive thought symptom dimension were associated with poorer metacognitive efficiency, in the absence of any changes to objective memory performance. In Chapter 3, an experimental stress manipulation was used to investigate whether acute stress responses are associated with state-level disruptions to metamemory in a healthy population. It was hypothesised that metamemory performance would be poorer following a stress induction relative to performance at baseline. While no overall changes in metacognitive bias or efficiency were observed from pre- to post-stress, heightened psychological (but not physiological) responses to the stress induction were associated with poorer metacognitive efficiency, in the absence of any changes to first-order memory. Interestingly, compulsivity and intrusive thought symptoms were unrelated to metacognitive performance pre-stress, but a relationship between these symptoms and metacognitive efficiency emerged following the stress manipulation, suggesting that the relationship between the psychiatric dimension and metamemory only emerged under stressful conditions. In Chapter 4, the final experimental chapter of this thesis, I further examined the proposed relationship between autonomic function and metamemory abilities. In this study, I investigated whether blocking noradrenergic action using propranolol would remediate arousal-related disruptions to metamemory performance. It was hypothesised that heightened autonomic arousal would predict disruptions to metacognitive efficiency. I additionally hypothesised that blocking noradrenergic activity via propranolol would successfully remediate this effect and that metacognitive efficiency would be greater in the propranolol group relative to the placebo group. While a higher pulse rate was associated with poorer metacognitive efficiency, systolic blood pressure, diastolic blood pressure, state anxiety, and pupil dilation measurements were unrelated to metacognitive abilities. Propranolol did not remediate the relationship between pulse rate and metamemory performance, and neither metacognitive bias nor metacognitive efficiency differed between the placebo and propranolol groups, indicating that propranolol was unsuccessful in improving metamemory performance. Together, the results reported in this thesis provide weak evidence for a relationship between psychological and physiological responses to stress and aberrations to metamemory performance. Nevertheless, our findings show that metamemory performance relates closely to compulsivity and intrusive thought symptoms and that this relationship may be triggered following exposure to acute stress. Chapter 5 provides a broader discussion of these findings within the context of clinical psychology and cognitive neuroscience frameworks of metacognition, in relation to existing research, and in light of the studies’ methodological limitations.