Neurolinguistic Communication Intervention

Articles

Characterizing Language Network Connectivity Changes in Autism Spectrum Disorder Using Graph Theory and Machine Learning: A Multisite Functional Magnetic Resonance Imaging Study

Downloads

Ma, H., Wu, J., Chen, L., Song, J., Zhang, H., Hao, Z., Zhan, L., Cheng, L., & Jia, X. (2026). Characterizing Language Network Connectivity Changes in Autism Spectrum Disorder Using Graph Theory and Machine Learning: A Multisite Functional Magnetic Resonance Imaging Study. Neurolinguistic Communication Intervention, 1(1), 42–60. Retrieved from https://ojs.ukscip.com/index.php/nci/article/view/2271
Volume 1 Issue 1 (2026)

Authors

  • Huibin Ma

    School of Information and Electronics Technology, Jiamusi University, Jiamusi, 154007, China
  • Jinying Wu

    School of Information and Electronics Technology, Jiamusi University, Jiamusi, 154007, China
  • Lanfen Chen

    School of Medical Imaging, Shandong Second Medical University, Weifang, 261053, China
  • Jiaying Song

    School of Information and Electronics Technology, Jiamusi University, Jiamusi, 154007, China
  • Hang Zhang

    School of Medical Imaging, Shandong Second Medical University, Weifang, 261053, China
  • Zeqi Hao

    School of Psychology, Zhejiang Normal University, Jinhua, 321004, China
  • Linlin Zhan

    Faculty of Western Languages, Heilongjiang University, Harbin, 150080, China
  • Lulu Cheng

    School of Foreign Studies, China University of Petroleum (East China), Qingdao, 266580, China
  • Xize Jia

    School of Psychology, Zhejiang Normal University, Jinhua, 321004, China；Intelligent Laboratory of Zhejiang Province in Mental Health and Crisis Intervention for Children and Adolescents, Zhejiang Normal University, Jinhua, 321004, China

Autism spectrum disorder (ASD) is associated with atypical language-related functional connectivity (FC), though network-level substrates of verbal ability across large multisite cohorts require further characterization. In this study, resting-state fMRI data from 219 individuals with ASD and 322 healthy controls with valid verbal IQ (VIQ) scores were obtained from 12 sites in the ABIDE database. ROI-based FC and whole-brain network metrics were computed and correlated with VIQ. Three feature sets—raw connectivity and topology, group-difference metrics, and metrics combined with VIQ—were evaluated using SVM and manifold learning. ASD participants showed widespread FC alterations with overall increases and selective decreases in specific networks. At the global level, ASD patients exhibited a significantly lower AUC specifically for the normalized clustering coefficient γ (p < 0.05), while other metrics showed no significant group differences. Nodal analyses revealed increased degree centrality and efficiency in subcortical and frontal regions and reductions in limbic and temporal areas (all p < 0.05). Anterior cingulate centrality correlated negatively with VIQ, while occipital metrics correlated positively. The classification reached a peak AUC of 0.7514, which is competitive for large-scale multisite analysis and demonstrates the relevance of VIQ-related features in a proof-of-principle context. These results provide a proof-of-principle for the relevance of VIQ-related network features. These findings suggest that aberrant FC linked to verbal ability reflects key aspects of ASD pathophysiology, emphasizing the importance of integrating behavioral scales into neuroimaging research.