BACKGROUND: Social anxiety and paranoia are connected by a shared suspicion framework. Based on cognitive-behavioural approaches, there is evidence for treating social anxiety and psychosis. However, mechanisms underlying the relationship between social anxiety and paranoia remain unclear. AIMS: To investigate mediators between social anxiety and paranoia in schizophrenia such as negative social appraisals (i.e. stigma or shame; Hypothesis 1), and safety behaviours (i.e. anxious avoidance or in situ safety behaviours; Hypothesis 2). METHOD: A cross-sectional study was conducted among Asian out-patients with schizophrenia (January-April 2020). Data on social anxiety, paranoia, depression, shame, stigma, anxious avoidance, and in situ behaviours were collected. Associations between social anxiety and paranoia were investigated using linear regressions. Mediation analysis via 10,000 bias-corrected bootstrap samples with 95% confidence intervals (CI) was used to test the indirect effects (ab) of mediators. RESULTS: Participants (n=113, 59.3% male) with a mean age of 44.2 years were recruited. A linear relationship between social anxiety and paranoia was found. In multiple mediation analyses (co-varying for depression), stigma and shame (Hypothesis 1) did not show any significant indirect effects with ab=.004 (95%CI=-.013, .031) and -.003 (-.023, .017), respectively, whereas in situ behaviours (Hypothesis 2) showed a significant effect with ab=.110 (.038, .201) through the social anxiety-paranoia relationship. CONCLUSIONS: Social anxiety and paranoia are positively correlated. In situ safety behaviours fully mediated the social anxiety and paranoia relationship. Targeted interventions focusing on safety behaviours could help reduce paranoia in psychosis. Symptom severity should be measured to help characterise the participants' characteristics.