Bifactor Structure of the Schizotypal Personality Questionnaire Across the Schizotypy Spectrum

Schizotypal traits in a large sample of high-school and university students from Tunisia: correlates and measurement invariance of the arabic schizotypal personality questionnaire across age and sex

BACKGROUND

The main goal of the present study was to examine the characteristics of schizotypal traits and their correlations with genetic (i.e., family history of mental illness), demographic (i.e., age, sex), environmental (e.g., income, urbanicity, tobacco/alcohol/cannabis use), and psychological (i.e., personal history of mental illness other than psychosis) factors in Tunisian high-school and university students. Our secondary goal was to contribute the literature by examining the factor structure and factorial invariance of the Arabic Schizotypal Personality Questionnaire (SPQ) across sex and age (adolescents [12-18 years] vs. young adults [18-35 years]) groups.

METHOD

This was a cross-sectional study involving 3166 students: 1160 (36.6%) high-school students (53.0% females, aged 14.9 ± 1.8); and 2006 (63.4%) university students (63.9% females, aged 21.8 ± 2.3). All students were asked to complete a paper-and-pencil self-administered questionnaire containing sociodemographic characteristics as well as the Arabic version of the SPQ.

RESULTS

The total sample yielded total SPQ scores of 24.1 ± 16.6 out of 74. The SPQ yielded good composite reliability as attested by McDonald's omega values ranging from .68 to .80 for all nine subscales. Confirmatory Factor Analysis indicated that fit of the 9-factor model of SPQ scores was acceptable. This model is invariant (at the configural, metric and structural levels) across sex and age. Except for "Odd or eccentric behavior", all schizotypy features were significantly higher among female students compared to males. Multivariable analyses showed that female sex, being a university student, lowest family incomes, tobacco use, and having a personal history of psychiatric illness were significantly associated with higher positive, negative and disorganized schizotypy subscales scores.

CONCLUSION

Future research still needs to confirm our findings and investigate the contribution of the identified factors in the development of clinical psychosis. We can also conclude that the Arabic SPQ is appropriate for measuring and comparing schizotypy across age and sex in clinical and research settings. These findings are highly relevant and essential for ensuring the clinical utility and applicability of the SPQ in cross-cultural research.

Trait schizotypy and the psychosis prodrome: Current standard assessment of extended psychosis spectrum phenotypes

Schizotypy has become an increasingly important construct for elaborating psychotic disorders that vary along the schizophrenic spectrum. However, different schizotypy inventories vary in conceptual approach and measurement. In addition, commonly used schizotypy scales have been seen as qualitatively different from screening instruments for prodromal schizophrenia like the Prodromal Questionnaire-16 (PQ-16). Our study investigated the psychometric properties of three schizotypy questionnaires (the Schizotypal Personality Questionnaire-Brief, Oxford-Liverpool Inventory of Feelings and Experiences, and the Multidimensional Schizotypy Scale) as well as the PQ-16 in a cohort of 383 non-clinical subjects. We initially evaluated their factor structure using Principal Component Analysis (PCA) and used Confirmatory Factor Analysis (CFA) to test a newly proposed composition of factors. PCA results support a three-factor structure of schizotypy that accounts for 71 % of the total variance, but also shows cross-loadings of some schizotypy subscales. CFA of the newly composed schizotypy factors (together with an added neuroticism factor) shows good fit. Analyses including the PQ-16 indicate considerable overlap with measures of trait schizotypy, suggesting that the PQ-16 might not be quantitatively or qualitatively different from schizotypy measurements. Taken together, results indicate that there is good support for a three-factor structure of schizotypy but also that different schizotypy measurements grasp facets of schizotypy differently. This points towards the need for an integrative approach for assessing the construct of schizotypy.

People with high schizotypy experience more illusions in the Pattern Glare Test: Consistent with the hyperexcitability hypothesis

Individuals diagnosed with schizophrenia spectrum disorders (SSD) exhibit a constellation of sensory and perceptual impairments, including hyporeactivity to external input. However, individuals with SSD also report subjective experiences of sensory flooding, suggesting sensory hyperexcitability. To identify the extent to which behavioural indices of hyperexcitability are related to non-psychotic symptoms of schizophrenia, we tested a non-clinical population measured for schizophrenia-like traits (schizotypy), and a behavioural measure of sensory hyperexcitability, specifically the number of illusions seen in the Pattern Glare Test. Two samples totaling 913 individuals completed an online version of the Schizotypal Personality Questionnaire - Brief Revised (SPQ-BR) and the Pattern Glare Test. Individuals with higher schizotypy traits reported more illusions in the Pattern Glare Test. Additionally, one of the three SPQ-BR factors, the disorganized factor, significantly predicted the number of illusions reported. These data illustrate the potential for research in non-clinical samples to inform clinically relevant research.

Association of Multidimensional Schizotypy with PID-5 Domains and Facets

The underlying vulnerability for schizophrenia-spectrum disorders is expressed across a continuum of clinical and subclinical symptoms referred to as schizotypy. Schizotypy is a multidimensional construct with positive, negative, and disorganized dimensions. The present study examined associations of positive, negative, and disorganized schizotypy with pathological personality traits and facets assessed by the Personality Inventory for DSM-5 (PID-5) in 1,342 young adults. As hypothesized, positive schizotypy was associated with the PID-5 psychoticism domain and facets, negative schizotypy was associated with the detachment domain and facets and the restricted affectivity facet, and disorganized schizotypy's strongest associations were with the distractibility and eccentricity facets and the negative affect domain. The PID-5 facets accounted for upwards of two thirds of the variance in each schizotypy dimension. The authors conclude by providing regression-based algorithms for computing positive, negative, and disorganized schizotypy scores based on the PID-5 facets.

The Sensory Gating Inventory-Brief

The Sensory Gating Inventory (SGI) is a 36-item measure used to assess an individual's subjective ability to modulate, filter, over-include, discriminate, attend to, and tolerate sensory stimuli. Due to its theoretical and empirical link with sensory processing deficits, this measure has been used extensively in studies of psychosis and other psychopathology. The current work fills a need within the field for a briefer measure of sensory gating aberrations that maintains the original measure's utility. For this purpose, large samples (total n = 1552) were recruited from 2 independent sites for item reduction/selection and brief measure validation, respectively. These samples reflected subgroups of individuals with a psychosis-spectrum disorder, at high risk for a psychosis-spectrum disorder, nonpsychiatric controls, and nonpsychosis psychiatric controls. Factor analyses and item-response models were used to create the SGI-Brief (SGI-B; 10 Likert-rated items), a unidimensional self-report measure that retains the original SGI's transdiagnostic (ie, present across disorders) utility and content breadth. Findings show that the SGI-B has excellent psychometric properties (alpha = 0.92) and demonstrates external validity through strong associations with measures of psychotic symptomatology, theoretically linked measures of personality (eg, perceptual dysregulation), and modest associations with laboratory-based sensory processing tasks in the auditory and visual domains on par with the original version. Accordingly, the SGI-B will be a valuable tool for dimensional and transdiagnostic examination of sensory gating abnormalities within clinical science research, while reducing administrator and participant burden.