SU26. Measuring Implicit Theory of Mind in Schizophrenia Research: Cross-Cultural Validation

摘要

>Background: Implicit theory of mind (iToM) is part of early social development (Schneider et al, 2012). It is crucial for building sensitivity to the other person or event (Apperly and Butterfill, 2009) and for tracking the other’s mental states. This process occurs without conscious awareness (Low and Perner, 2012). Measuring iToM may be important for understanding and rehabilitating social cognitive impairments in people with schizophrenia. The Social Attribution Task–Multiple Choice (SAT-MC), which was first developed for studying people with autism, has been successfully employed to distinguishing individuals with schizophrenia (SZ) from healthy controls (HC) in iToM in western countries (Bell et al, 2010). Because it uses geometric shapes enacting a social drama, the stimuli are nonverbal and does not include culturally loaded materials compared to other social cognition measures. We therefore hypothesized that the SAT-MC would be valid for use in schizophrenia research across cultures. >Methods: A total of 120 (30 Korean SZ; 30 Korean Healthy controls; 30 US SZ; 30 US Healthy controls) were randomly selected from existing databases to examine reliability and validity of the SAT-MC in participants in both nations. First, factor structures were investigated using a confirmatory factor analysis. Then, internal consistency coefficients were calculated to examine reliability. Moreover, a series of ANOVAs with between-group variables (ie, SZ vs HC and Korean vs American) were conducted to examine mean differences in potential extraneous variables including demographic information. Lastly, a series of correlations analyses were conducted to examine relationships of the SAT-MC with other social cognition tasks (eg, The Eye Test, The Hinting Task) and psychiatric symptoms. >Results: SAT-MC has good internal consistency independent of group and culture (Cronbach’s α = 0.85 for total samples, 0.82 for SZ, 0.81 for HC, 0.80 for Korean SZ, 0.84 for Korean HC, 0.84 for US SZ, 0.78 for US HC). Confirmatory factor analysis showed that the 1-factor model had a good model fit (χ² = 260.391, RMSEA = 0.077 [90% CI: 0.061, 0.093], SRMR = 0.077). There were no significant differences in demographic variables (eg, age, education and gender), premorbid IQ, age at onset, and psychiatric symptoms between groups and cultures. SZ were significantly lower on SAT-MC than HC, regardless of cultures (F(3,116) = 11.00, P < .001). Moreover, the SAT-MC was significantly correlated with other social cognition tasks in SZ in Korea and the United States. >Conclusion: The results suggest that the SAT-MC is a reliable and valid measure for measuring iToM in both Korean and the United States. It is a measure that may be less vulnerable to cultural and language differences than other measures of ToM.