Exploratory Multivariate Analysis of Reconstructed Neuropsychological Data in Children with ADHD: A Methodological Study within the Current Neuroimmune Framework

Hicham Lafhal ORCID ,  Ahmed Omar Tohami Ahami ORCID ,  Siham Goutou ,  Atmane Rochdi ORCID
    Received: 6 July 2026; Revised: 28 July 2026; Accepted: 7 August 2026; Published: 17 August 2026

    Abstract

    Attention-deficit/hyperactivity disorder (ADHD) is a heterogeneous neurodevelopmental disorder characterized by executive dysfunction, visuomotor impairment, and visuospatial deficits. Emerging evidence suggests that neuroinflammatory mechanisms may contribute to this cognitive heterogeneity. This exploratory methodological study aimed to illustrate the application of advanced multivariate statistical approaches to a reconstructed neuropsychological dataset and to discuss the resulting analytical patterns within the context of current neuroimmune research. A reconstructed dataset derived from published summary statistics of a previously reported case-control study of 80 children (40 with ADHD and 40 controls) was analyzed focusing on the Bender–Gestalt Test and the Rey–Osterrieth Complex Figure Test. Descriptive statistics, principal component analysis, cluster analysis, multiple regression, mediation analysis, and correlation network visualization were performed. The exploratory multivariate analyses illustrated how advanced statistical methods can identify latent neuropsychological structures within a reconstructed dataset. Principal component analysis summarized the major dimensions of variability, cluster analysis explored alternative cognitive profile solutions, and regression analysis identified variables associated with visuoconstructive performance. Although immune biomarkers were not assessed, the observed cognitive profiles may be interpreted within the current neuroimmune framework as hypothesis-generating. This methodological study illustrates the potential of advanced multivariate analyses for exploring reconstructed neuropsychological datasets. The results should be considered exploratory and hypothesis-generating and require validation using original participant-level data before any clinical or biological interpretation.

    Keywords

    References

      ×