Executive functions alterations across neurodevelopmental disorders – chasing a transdiagnostic approach

Background: The term "excutive functions" refers to a group of cognitive processes involved in goal-directed behaviours, allowing the interaction with complex and constantly changing environments [1]. Deficits of these domains are common across various psychiatric disorders, particularly those with neurodevelopmental etiology[2][3]. Such impairments may play a key role, offering useful insights into shared and distinct cognitive mechanisms. Aim: The aim of this study is to investigate weather specific EF deficit can characterize specific diagnostic categories, acting as transdiagnostic markers and contributing to a more dimensional understanding of psychopathology. Methods: For subtests from the Brief Assessment of Cognition in Schizophrenia Scale[4] (BACS) were used. The sample was composed of 159 adult patients: 36 with autism spectrum disorder (ASD), 57 with attention deficit/hyperactivity disorder (ADHD) and 66 with schizophrenia. Data analysis was performed using R. Group differences were assessed via ANOVA or Kruskal-Wallis test, followed by post hoc analysis as appropriate. Multiple linear regressions were conducted to identify predictors of cognitive performance, including diagnosis, education level, medication use and Mini Mental State Examination (MMSE, available for ASD and schizophrenia subgroups). Results: Significant group differences emerged in all domains except verbal fluency. In Working Memory, ASD patients outperformed both schizophrenia (p=0.001) and ADHD (p=0.009), while no difference was found between these two. Conversely, in Symbol Coding and Tower of London tasks, schizophrenia patients performed significantly worse than both ASD (p=0.00002; p=0.005) and ADHD (p= 0.07, p=0.04). No significant difference were observed between ASD and ADHD in these tasks. Regression analysis identified diagnosis and years of education as the strongest predictors of performance, with schizophrenia consistently associated with lower scores (from -2.9 to -11 across task). Valproate use was linked to poorer performance in verbal fluency and symbol coding, while other medication did not show significant effects in any test. In the ASD and schizophrenia subgroup, MMSE emerged as the most relevant predictor, especially for Verbal Fluency (model R2 = 0.50). Mediation analysis showed that MMSE explained up to 70% of education effect on verbal fluency, suggesting a mediating role in EF performance in these patients. Conclusions: This study reveals distinct executive dysfunction patterns across neurodevelopmental disorders. ADHD and schizophrenia patients performed similarly in working memory, both scoring lower than ASD. In contrast, schizophrenia was associated with worse impairment in symbol coding and problem solving while ADHD showed intermediate deficits. In ASD and schizophrenia subgroups, MMSE emerged as a key predictor of performance, particularly in verbal fluency, where it mediated much of the education effect. This emphasizes the important of considering general cognitive status when interpreting executive functions outcomes. These findings support the hypothesis of disorder-specific cognitive profiles, suggesting the value of transdiagnostic cognitive phenotyping. Further research should investigate MMSE mediating role and explore EF outcomes across diagnostic subtypes (Treatment resistant vs Treatment respondent schizophrenia or Inattentive vs Hyperactive prevalent ADHD, among others), to obtain more personalized and precise clinical approaches.