Factors predicting psychosis proneness and aberrant salience among florence’s students from the academy of fine arts: an observational study

Clinical intuition and anecdotal evidence have long suggested that people choosing to undergo a creative career in the arts might display an increased risk for schizotypy and psychosis spectrum disorders; these claims have been sup- ported by old and recent data [1]. In order to further test this hypothesis, we administered a battery of psycho- metric tests and questionnaires to the students of Florence’s Academy of Fine Arts. In particular, we measured the proneness to psychosis through the Aberrant Salience Inventory (ASI) [2]; moreover, we administered tests for personality assessment (Big Five Inventory) [3], obsessive psychopathology (Obsessive Be- liefs Questionnaire, OBSQ) [4] and convergent/divergent thinking (Remote Associate Test, RAT; Anagrams Test, ANAG) [5]. We also collected basic de- mographic data (age, sex). Including RAT and ANAG allowed us to test the following secondary hypothesis: high divergent thinking skills (ANAG) could predict increased ASI scores, while high convergent thinking skills (RAT) could be a protective factor toward psychosis proneness. Descriptive statistics for psychometric and demographic variables were computed. Linear regression models for overall ASI scores and ASI subscales scores (“Heightened Cognition”, “Sharpening of Senses”, “Enhanced Interpreta- tion and Emotionality”) were then computed using the following variables: Big Five Inventory subscales (Extraversion, Neuroticism, Openness, Conscientious- ness, Agreeableness), age, sex, OBSQ score, RAT and ANAG. P-value threshold for model significance was set to p < 0.05. Descriptive statistics’s results are shown in Table 1. The sample was composed mostly of females < 25 years old; ASI scores were quite consistently higher than in healthy controls as shown in previous literature; as a reference, Cicero origi- nally proposed 14 as a cut-off score in order to identify high risk subjects [2]. Regression models on ASI overall score and ASI’s “Enhanced Interpretation and Emotionality” subscale (ASIei) reached statistical significance. Within the first model (overall ASI score, p 1⁄4 0.039, R2 1⁄4 0.382), two coefficients reached sta- tistical significance: BFI Openness (p 1⁄4 0.04, standardized coefficient β 1⁄4 0.476) and RAT score (p 1⁄4 0.021, standardized coefficient β 1⁄4 -0.379). Two coefficients significantly predicted ASIei: OBSQ (p 1⁄4 0.016, standardized coefficient β 1⁄4 0.391) and RAT score (p 1⁄4 0.001, standardized coefficient β 1⁄4 -0.575). The relation between the Openness score on BFI and the ASI score is noteworthy. Openness has in fact often been accounted as a factor closely related to schizo- typy and even psychotic psychopathology. At the same time, such personality dimensions have also been linked to creativity, novelty-seeking, intelligence and curiosity. The capability of Openness to predict psychosis proneness, as attested by our results, gives new support to these hypotheses. On the other hand, convergent thinking skills (as measured by the RAT) were found to predict ASI negatively. This reflected our hypothesis, despite ANAG not reaching statistical significance (p 1⁄4 0.169, t 1⁄4 1.4). Our findings shed new light on risk factors in a population potentially prone to psychosis; thus, our study has both clinical and psychopathological implications.