Borderline personality disorder (BPD) is a diagnostic label ascribed to a markedly heterogenous psychiatric phenotype, with symptoms ranging from intense emotions to disturbed relatedness, from impulsivity and self-harm to chronic feelings of guilt, shame, and emptiness. While often regarded as an affective disorder, psychotic symptoms such as hallucinations and delusional convictions are also frequently experienced by people with BPD but are habitually dismissed in both research and clinical practice. Their aetiology and the mechanisms underlying them remain poorly understood, largely due to historical and prevailing stigma towards the disorder. By building on recurring findings of psychotic symptom correlates, and by drawing on causal and cognitive models of psychosis in other mental health conditions it is possible to explore potential risk factors and create a preliminary aetiological framework to better inform and direct future research.

This dissertation aims to expand the currently sparse literature on psychotic experiences in BPD, by providing a detailed assessment of the prevalence (Chapter 4), type, and range of hallucinations and delusions (Chapter 5), as well as investigating key factors hypothesised to influence their occurrence or severity (Chapter 6 and 7) with two empirical studies – an online experiment recruiting BPD participants identified by self-report, and a laboratory study of people with a confirmed clinical diagnosis. Both hallucinatory and delusional experiences were significantly more common in people with BPD, presenting on a broad spectrum of frequency and severity – ranging from non-pathological to comparable with psychotic disorders. Dissociation was the strongest predictor for psychotic symptom prevalence and severity for both, perceptual and delusional experiences. Several other factors (including trauma and adversity, disorder severity, psychiatric comorbidities, and common mental distress) were found to generally correlate with (but have limited predictive value for) psychotic experiences. Additionally, current stress levels appeared to influence dissociative and psychotic state symptoms. Results did not support a causal role of aberrant prediction error signalling (as postulated by predictive processing theories) for psychosis proneness in people with borderline personality disorder.

In summary, this thesis furthers our understanding of the prevalence and nature of psychotic symptoms in BPD and provides novel insights into the possible cognitive and aetiological mechanisms underlying these experiences.