This thesis explored the role of metacognitive beliefs in paranoia. The thesis is divided into three sections. Paper 1 presents a review of the current evidence relevant to the metacognitive model of paranoia. Paper 2 extends the evidence base by empirically manipulating metacognitive beliefs in non-clinical paranoia. Papers 1 and 2 have been prepared for submission to Acta Psychiatrica Scandinavica and Psychosis respectively. Within the thesis, references, tables and figures are presented in a consistent format to facilitate readability. Paper 3 represents a critical evaluation of the methods carried out in papers 1 and 2. More specifically, paper 1 reports a systematic review that examines the applicability of the metacognitive model to explain the development and maintenance of paranoia. Studies investigating testable predictions of the metacognitive model of paranoia have focused on assessing the relationship between metacognitive beliefs and paranoia. Ten studies meeting the inclusion criteria exploring the relationship between metacognitive beliefs and paranoia were identified through database searching and were included in the review. Results showed a lack of evidence with regards to the causal role of metacognitive beliefs in the development of paranoia, thus providing limited support for the metacognitive model of paranoia. The strengths and weakness of the studies and of the review were discussed along with theoretical and clinical applications and recommendations for future research. In the absence of evidence supporting a causal relationship between metacognitive beliefs and paranoia, Paper 2 aimed at investigating the direct impact of manipulating metacognitive beliefs on paranoia frequency and associated distress. Paper 2 reports an experimental analogue study in which participants (n = 110) were randomized to either a positive or negative manipulation group intended to alter beliefs about paranoia before entering a paranoia induction task. The metacognitive beliefs manipulation was partially successful. In line with predictions, the positive group reported an increase in paranoia frequency after the paranoia induction, whereas, contrary to predictions, participants in the negative beliefs group reported a decrease in paranoia related distress. Clinical and research implications of the experimental findings are considered. Paper 3 provides a critical appraisal of the research process as a whole. Strengths and limitations of the research are presented along with clinical implications, and suggestions for future research.