The current study sought to investigate perspective-taking in a group of individuals diagnosed with psychosis. The Social Anxiety Implicit Relational Assessment Procedure (SA-IRAP) contrasted statements and questions referring to ‘anxious’ and ‘non-anxious’ experiences. The SA-IRAP and a Faux-pas (Theory of Mind) test were presented to two groups of participants: a clinical group with a diagnosis of social anxiety and a group of controls. IRAP effects for each group were in the predicted direction and a ROC analysis showed that the IRAP correctly classified 77% of the individuals with social anxiety with a sensitivity level of 97% and a specificity level of 36%. The IRAP was thus successfully used to correctly classify the sample of clinical individuals. However, the study also demonstrated that the two groups were similar with regard to their level of competency on the perspective-taking IRAP. The article also discusses the relationship between the data and recent developments in RFT