Schizotypal traits

We were interested to see whether objectively assessed differences in perceptual inference are associated with self-reported variation in perception and cognition. Therefore, we measured positive, negative and disorganized dimensions of schizotypal traits (that is, personality traits resembling the signs and symptoms of schizophrenia) with the Oxford–Liverpool Inventory of Feelings and Experiences, short version (sO-LIFE³⁹, Hungarian version⁴⁰). An additional purpose of administering this questionnaire was to obtain benchmarks for test–retest stability in the present sample with an instrument optimized for the assessment of individual differences. Here, we used three subscales. Positive schizotypy was measured with the Unusual Experiences subscale (12 items, reliability in the sample Cronbach’s alpha = 0.82/0.82, test–retest reliability Spearman’s rho = 0.92, ICC = 0.96 ), negative schizotypy was measured with the Introvertive Anhedonia subscale (10 items, reliability in the sample Cronbach’s alpha = 0.76/0.74, test–retest reliability Spearman’s rho = 0.86, ICC = 0.94), and disorganized schizotypy was assessed with the Cognitive Disorganisation subscale (11 items, reliability in the sample Cronbach’s alpha = 0.78/0.84, test–retest reliability Spearman’s rho = 0.87, ICC = 0.95). Studies have shown that the sO-LIFE has appropriate psychometric properties to assess schizotypal traits in the non-clinical population (except for the Impulsive Nonconformity dimension, which we therefore did not use in our study, reliability in the sample Cronbach’s alpha = 0.74/0.70, test–retest reliability Spearman’s rho = 0.44, ICC = 0.91)⁴¹. Although the original questionnaire uses a binary response format, we decided on 5-level Likert-format in order to improve reliability (following⁴⁰), which was excellent in the present sample. Additional data (mean, SD, minimum, maximum, skewness, kurtosis) about these subscales can be found in Supplement ¹.