Neural correlates of hallucination proneness in the general population: Behavioural and multimodal MRI-based approaches

Abstract

Hallucinations in healthy people are more common than previously assumed and may occur without the distress or impairment typically associated with clinical psychosis. Studying hallucination proneness (HP) in non-clinical populations provides opportunities to explore underlying mechanisms of hallucinatory experiences while minimising confounding factors such as medication, chronic illness, and comorbidity. This thesis explores the cognitive and neural correlates of HP in healthy individuals. The psychosis continuum model provides a theoretical framework, proposing that healthy hallucinators occupy an intermediate position between clinical hallucinators and healthy controls in terms of anomalous experiences, while remaining below the threshold for clinical intervention. Therefore, a central focus of this research is to evaluate whether the findings support this continuum model of psychosis. The thesis adopts multimodal, multi-cohort cross-sectional design studies, integrating findings from a systematic review, a large-scale population survey, and two empirical studies combining behavioural and magnetic resonance imaging (MRI) methods. First, the systematic review (Chapter 2) synthesised results from 22 MRI studies examining auditory and visual hallucinations in non-clinical populations. Although there was considerable methodological heterogeneity, involvement of the superior temporal gyrus (STG), insula, and anterior cingulate cortex (ACC) was observed. The neural correlates of hallucinations in healthy hallucinators showed both overlapping and distinct features compared to clinical groups, suggesting that similar brain networks may be engaged, but through different mechanisms. Current evidence remains inconclusive as to whether clinical and non-clinical hallucinations share the same underlying pathophysiology or arise from distinct processes. Chapter 3 reports the first psychometric validation of the Arabic version of the Launay-Slade Hallucination Scale-Extended (LSHS-E) in a general population sample from Saudi Arabia (n = 428). The Arabic version demonstrated strong reliability and validity, with a four-factor structure consistent with previous cross-cultural studies. This chapter also introduced the Arabic version of the positive dimension - Community Assessment of Psychic Experiences (PCAPE), supporting future research on HP in Arabic-speaking populations. Chapters 4 and 5 report two empirical MRI studies conducted on a university-based cohort initially comprising 491 participants. From this larger sample, individuals in the highest (n =31) and lowest (n = 35) tertiles of LSHS-E scores were selected for MRI and associated behavioural experiments. Chapter 4 combined two attention-based signal detection tasks with functional MRI (fMRI). Behaviourally, the high HP group showed significantly lower perceptual sensitivity and faster reaction times under conditions of cross-modal distraction, but no consistent group differences in response bias, suggesting altered sensory discrimination and inhibitory control. fMRI revealed robust task-related activations across auditory, visual and attentional networks with only subtle, non-significant group differences. A trend-level increase in activation in the right Rolandic operculum in the high HP group was observed but did not provide support for the continuum hypothesis. Chapter 5 employed diffusion tensor imaging (DTI) to investigate white matter microstructure. Contrary to previous findings in schizophrenia, the high HP group showed significantly lower mean diffusivity (MD) in the right superior longitudinal fasciculus (SLF) and corticospinal tract (CST). Hallucination scores were negatively correlated with diffusivity measures in these tracts. These results challenge the assumption that HP reflects a milder form of clinical pathology and instead raise the possibility that lower diffusivity may reflect more efficient or protective white matter organisation in non-clinical individuals prone to hallucinations. The findings do not support a simple dimensional model of psychosis. Rather, they suggest that HP in healthy individuals is characterised by distinct cognitive and neural features. The thesis offers novel contributions to the understanding of HP and provides important cross-cultural insights by including the first neuroimaging investigation of HP in a healthy, Arabic-speaking population.