Schizophrenia's strong genetic basis is well-established, with familial risk being a significant predictor. However, most individuals with a high genetic predisposition never develop the disorder. This study explores the possibility of milder manifestations of deviant thinking, specifically magical thinking, in individuals at high genetic risk for schizophrenia but without a diagnosis of non-affective psychotic disorders. The study's significance lies in its potential to illuminate the diverse outcomes associated with high genetic risk for psychosis, extending beyond the focus on the development of full-blown schizophrenia. Previous research primarily concentrated on individuals who develop psychosis despite their genetic predisposition. This approach neglects the vast majority of individuals with a schizophrenia-susceptible genotype who do not develop the disorder. While high polygenic risk scores (PRS) predict an increased likelihood of psychosis, their specificity in predicting particular mental disorders remains low, suggesting a broader liability to a range of adverse outcomes. Understanding these alternative outcomes is crucial to a comprehensive understanding of the genetic architecture of schizophrenia and related conditions. Recent research has begun to explore these alternative outcomes in individuals with a schizophrenia-susceptible genotype but without psychosis, focusing on brain activity patterns and cognitive capacities. These studies have produced mixed results. Psychosocial outcomes, however, remain largely uncharted territory. This study focuses on magical thinking, an umbrella term encompassing paranormal beliefs, mystical ideation, and anomalous experiences. The study investigates whether a high polygenic risk for schizophrenia in the absence of psychosis may manifest as increased magical thinking. The use of the longitudinal Young Finns Study, with its extended follow-up period, provides a unique opportunity to examine the developmental trajectory of magical thinking over time.

The literature on the genetics of schizophrenia has evolved from focusing on individual candidate genes to employing polygenic risk scores (PRS) derived from genome-wide association studies (GWAS). These GWAS studies have identified numerous single nucleotide polymorphisms (SNPs) associated with schizophrenia, providing a more comprehensive picture of the genetic architecture of the disorder. However, the percentage of variance in liability explained by these PRS varies across studies. Previous research examining individuals with a schizophrenia-susceptible genotype but without psychosis has yielded mixed findings. While some studies found no association between high PRS and brain structural changes, others reported correlations with increased frontal activity during cognitive tasks and varying effects on cognitive capacity. Studies examining the psychosocial outcomes of a high PRS for schizophrenia in individuals without psychosis are limited and have focused on outcomes such as trait anxiety, creativity, and number of children. The existing research has provided conflicting findings and has not thoroughly explored the possible relationship between genetic risk for schizophrenia and magical thinking. The current DSM-5 recognizes magical thinking as a feature of various mental health conditions, including, but not limited to, schizotypal personality disorder, attenuated psychosis syndrome, and even transiently in other disorders such as obsessive-compulsive disorder and histrionic personality disorder. Therefore, the study of magical thinking within the context of genetic predisposition to schizophrenia can provide valuable insights into the broader phenotype of high genetic risk.

This study utilized data from the population-based Young Finns Study (YFS), a longitudinal study with a 37-year follow-up period. The analysis included 1292 participants who had complete data on polygenic risk score for schizophrenia, psychiatric diagnoses, magical thinking (assessed via the Spiritual Acceptance Scale of the Temperament and Character Inventory at three time points: 1997, 2001, and 2012), childhood family circumstances, and adult socioeconomic factors. Participants with diagnoses of non-affective psychotic disorders were excluded. Polygenic risk scores (PRS) for schizophrenia were calculated using summary statistics from a recent GWAS study. The Spiritual Acceptance Scale measures beliefs in telepathy, miracles, and other paranormal phenomena. Psychiatric diagnoses were obtained from the Finnish Hospital Discharge Register, using a standardized conversion process to align diagnoses across different ICD versions. Adult socioeconomic covariates included income, occupational status, and education level. Childhood psychosocial environment variables included measures of stressful life events, adverse socioeconomic circumstances, and unfavorable emotional family atmosphere, based on parental reports. Data analysis employed growth curve modeling (maximum likelihood estimation) to examine the longitudinal associations between PRS and magical thinking scores, controlling for covariates. The model accounted for individual variations in both the mean level of magical thinking and individual-level variance in magical thinking across the follow-up period. Age was centered to minimize multicollinearity. Additional analyses were conducted to examine the PRS's validity and examine the relationship between the PRS and other subscales of the Self-Transcendence scale. Attrition bias was evaluated using t-tests and chi-square tests.

The study found a significant positive association between high polygenic risk scores (PRS) for schizophrenia and higher levels of magical thinking in adulthood, even among participants who did not develop non-affective psychotic disorders (p=0.001 for weighted PRS; p=0.001 for unweighted PRS). This association remained significant after adjusting for sex, childhood family environment, and adult socioeconomic factors. Furthermore, the study revealed a significant interaction effect between PRS and age (p = 0.034 for weighted PRS; p = 0.011 for unweighted PRS). Specifically, individuals with low PRS showed a steady decrease in magical thinking from age 20 to 50, while those with high PRS exhibited a leveling off or a smaller decrease in magical thinking during middle age, resulting in significantly higher levels of magical thinking than expected for their age. These findings were consistent across both weighted and unweighted PRS analyses. Additional analyses confirmed the validity of the PRS by demonstrating its association with the likelihood of non-affective psychoses. The findings were replicated even after excluding participants with affective disorders. Further investigations revealed that high PRS was also associated with increased self-forgetful experiences but not with transpersonal identification.

The findings of this study demonstrate that a high genetic risk for schizophrenia may manifest as elevated magical thinking in adulthood, particularly in middle age, even in the absence of a diagnosis of non-affective psychosis. This supports the notion that high genetic risk is associated with a broader liability extending beyond the development of the full-blown disorder. The observed age interaction suggests that the impact of genetic risk on magical thinking may become more prominent in middle age. This finding challenges the notion that magical thinking simply acts as a risk factor for psychosis, demonstrating that it may develop independently and perhaps even as a consequence of genetic risk. The association of high PRS with self-forgetful experiences but not with transpersonal identification further supports the idea that high genetic risk may be linked to milder forms of deviant thinking. The absence of an association with transpersonal identification may suggest that this form of spirituality is not maladaptive and may even represent a healthy adaptation. These findings add to our understanding of the diverse phenotypic expressions of high genetic risk for schizophrenia, indicating that seemingly unrelated traits may share common genetic underpinnings.

This study provides robust evidence that high polygenic risk for schizophrenia, even in the absence of non-affective psychosis, predicts higher levels of magical thinking in adulthood, especially during middle age. The distinct developmental trajectories observed highlight the complex relationship between genetic predisposition and phenotypic expression. Future research should investigate the underlying mechanisms linking high PRS to magical thinking, explore potential mediating factors, and replicate these findings in different populations and cultural contexts. Further research could also investigate the specific genetic variants that are contributing to this association.

Several limitations should be considered. The correlates of magical thinking and self-transcendence may be culturally specific. The study sample, although large and population-based, exhibited some attrition bias, with included participants tending to be older, female, and from more advantaged socioeconomic backgrounds. While efforts were made to address potential biases in analyses, the possibility of residual confounding remains. The study's cross-sectional nature limits the ability to infer causality. Finally, although participants were middle-aged when psychiatric diagnoses were collected, the possibility of late-onset psychosis remains, though it is rare.