Meta-beliefs about the senses: Cognitive and neural mechanisms
Olawole-Scott, Helen. 2025. Meta-beliefs about the senses: Cognitive and neural mechanisms. Doctoral thesis, Goldsmiths, University of London [Thesis]
|
Text (Meta-beliefs about the senses: Cognitive and neural mechanisms)
PSY_thesis_Olawole-ScottH_2025.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB) | Preview |
Abstract or Description
To effectively interact with the environment, we must estimate the reliability or ‘precision’ of what we perceive. Assessing our confidence in our senses depends on metacognitive monitoring mechanisms in the brain, which many think are influenced by contextual information and prior beliefs. However, it remains unclear whether and how our brains generate these ‘beliefs about precision’, and how they influence confidence, awareness and behaviour. This thesis provides an insight into these questions.
The thesis is arranged as follows. After reviewing the relevant literature in Chapter 1, Chapter 2 describes empirical and computational work testing the influential theory that the mind is ‘Bayesian’ – the idea that agents form expectations about precision and use these to guide confidence and awareness. In each experiment, participants acquired probabilistic expectations about the likely strength of upcoming signals, while making confidence (Exps 1-2) or subjective visibility ratings (Exp 3). Computational modelling (Exp 4) revealed that the effects of these expectations on awareness could be well-explained by a predictive learning model that infers the precision (strength) of current signals as a weighted combination of incoming evidence and top-down expectation. These results suggest that agents do not only ‘read out’ the reliability of information arriving at their senses, but also take into account prior knowledge about how reliable or ‘precise’ different sources of information are likely to be.
Chapter 3 investigated the neural mechanisms underpinning the formation and use of expectations about precision. After adapting and piloting the paradigm used in Chapter 2 (Exp 5) we conducted a 3T functional magnetic resonance imaging (fMRI) experiment to investigate the how sensory uncertainty is represented in the brain (Exp 6). Using multivariate pattern decoding, we found representations of sensory uncertainty in the insula, which critically were also modulated by expectations about precision. One possibility is that the insula plays a role in encoding ‘precision prediction errors’ needed to form the expectations about precision seen in Chapter 2.
In Chapter 4 we investigated whether the ‘expected precision’ mechanisms identified in previous chapters generalise from the visual domain to our perception of speech – and how these mechanisms may be connected to unusual hallucination-like experiences. Across two studies (Exps 7 & 8) we find that expectations about precision can similarly bias subjective impressions of spoken voices. Moreover, we find some evidence that those prone to hallucinations relied less on these contextual cues when estimating the reliability of the sensory world. This result suggests that the fundamental cognitive and neural mechanisms identified in previous chapters could be disrupted in psychotic illness.
In Chapter 5 we investigated how expectations about precision influence one aspect of metacognitive control: the control of evidence accumulation (Exp 9). And in Chapter 6 we investigated how representations of sensory uncertainty more generally control another aspect of metacognitive control: seeking information (Exps 10 & 11). Chapter 7 provides a Discussion.
Item Type: |
Thesis (Doctoral) |
Identification Number (DOI): |
|
Keywords: |
Precision, visual perception, auditory perception, metacognition, meta-beliefs, fMRI |
Departments, Centres and Research Units: |
|
Date: |
30 June 2025 |
Item ID: |
39168 |
Date Deposited: |
11 Jul 2025 13:07 |
Last Modified: |
11 Jul 2025 13:14 |
URI: |
View statistics for this item...
![]() |
Edit Record (login required) |