Objectives Shame is increasingly implicated in the development and maintenance of several psychological problems including psychosis. The aim of the current paper was to review the research literature concerning the relationship between shame and the psychosis continuum, examining the nature and direction of this relationship. Method Systematic searches of databases Psyc INFO , Medline, Scopus, and Web of Science (from the earliest available database date until November 2016) were undertaken to identify papers that examined the relationship between shame and psychosis or psychotic experiences. Results A total of 20 eligible papers were identified. Risk of bias assessment identified methodological shortcomings across the research in relation to small, unrepresentative samples and failure to control for confounding variables. Narrative synthesis suggested positive associations between shame and paranoia ( n  =   10, r  =   .29–.62), shame and psychosis ( n  =   1, r  =   .40), and shame and affiliation with voices ( n  =   1, β = .26), and suggested that shame was greater in those with psychosis compared to controls ( n  =   4, d  =   0.76–1.16). Conclusions Overall, several studies provide partial support for the theory that shame is an important factor in relation to psychotic experiences in both clinical and non‐clinical populations, particularly paranoia. However, the predominance of cross‐sectional designs prevents any conclusions being drawn concerning the temporal nature of associations. Additional research is necessary to further delineate the role of shame in relation to specific psychotic experiences such as voice‐hearing. Longitudinal research is particularly needed to help establish the directionality and temporal aspects of effects. Practitioner points Research indicates moderate‐to‐strong positive associations between shame and psychotic experiences in the existing literature. The results provide preliminary evidence that shame may play a role in relation to psychosis and, more specifically, paranoia. Findings should be interpreted with caution due to many disparities across the studies reviewed and methodological shortcomings (e.g., small sample sizes). It is not currently possible to determine causality or direction of effect due to the cross‐sectional design of all existing studies.