This study investigated whether individual differences in obsessive-compulsive symptoms and in thought–action fusion are related to theory of mind abilities. One hundred and ninety-two adult participants completed self-reports of obsessive–compulsive symptoms (OCI-R), thought–action fusion (TAF), private self-consciousness (PSC) and self-reflectiveness (SR) as mentalising abilities, and anxiety and depression. A nonintrospective method examining participants' implicit structure of their lexicon for ‘knowing’ was used to assess theory of mind. Private self-conciousness and SR added to the prediction of OCD symptoms independently of TAF and depression but did not mediate the relationship between TAF and OCD symptoms. Participants high in thought–action fusion gave a greater emphasis to the certainty dimension of the mental lexicon and placed lesser importance on the source of information dimension than those low in TAF. Our results provide preliminary evidence of a relationship between theory of mind and thought–action fusion. People disposed to thought–action fusion are more likely to make a significance judgment about ‘knowing’ based on the degree of certainty than on reference to the source of knowledge. Identifying disruptions to theory of mind abilities in OCD provides links to solid theory and evidence about metacognitive development and may help integrate cognitive processing and cognitive appraisal models of OCD.

A growing literature suggests that personality traits may be endophenotype markers for psychiatric illnesses. Although the phenotypic relationships between obsessive–compulsive disorder (OCD) and high neuroticism and low extraversion have been well documented, underlying genetic and environmental contributions to these associations have not been explored previously. Five hundred and twenty-four monozygoitc (MZ) and 228 dizygotic (DZ) pairs of adolescent and young adult twins (aged 13–24 years) drawn from the South Korean Twin Registry completed the Maudsley Obsessive Compulsive Inventory (MOCI) and the Neuroticism and Extraversion scale of the Eysenck Personality Scale by mail. The total score of MOCI (MOCIT) was significantly and positively correlated with Neuroticism (r = .44), but only weakly and negatively related to Extraversion (r = –.10). A trivariate Cholesky model was applied to the data. The additive genetic correlations in the best-fitting model were .51 between Neuroticism and MOCIT and –.17 between Extraversion and MOCIT, suggesting that additive genetic factors that lead to high neuroticism and low extraversion overlap with those genetic factors influencing high OC symptoms. These findings add to the cumulative evidence of the shared genetic etiology for the associations between a personality profile of high neuroticism and low extraversion and mental illnesses.