Explainable Deep Learning of Resting-State Functional Connectomes Reveals Network Biomarkers of Adolescent Intelligence
Abstract
Mapping resting-state brain organization to individual differences in cognitive ability remains a major challenge in population neuroinformatics. Although deep learning enables flexible modeling of brain connectivity, limited interpretability restricts its scientific and clinical utility. To address this objective, we developed an explainable deep learning framework based on sparse projected residual networks to predict fluid, crystallized, and total intelligence from resting-state functional magnetic resonance imaging in 5,285 participants from the Adolescent Brain Cognitive Development study. We incorporated three complementary explainability methods (Integrated Gradients, Gradient Shapley Additive Explanations, and Occlusion) to interpret model behavior. The framework outperformed existing approaches, achieving Pearson correlations of 0.44, 0.58, and 0.56 for fluid, crystallized, and total intelligence, respectively, corresponding to predictive improvements of 6 to 9 percent. All three explainability methods produced near-identical feature rankings (pairwise rank correlations greater than 0.99). Consensus maps revealed a dual-layered functional architecture where primary predictive hubs localized within canonical systems, while the strongest global predictive pathways frequently bypassed these hubs through distributed, long-range relay connections. These findings suggest that intelligence emerges from the interaction between localized computational hubs and distributed communication pathways. Ultimately, these normative network architectures provide clinical reference maps to detect individual deviations, supporting earlier diagnosis, cognitive subtype stratification, and treatment monitoring in atypical neurodevelopment.
Related
- NeuroGraphBench: Interacting with Drosophila Connectomes at Scale for Exploring the Functional Logic of Neural Circuits
- Social brain fingerprinting: unique functional connectomes underpin interpersonal behaviours in non-autistic and autistic adults
- A language network in the individualized functional connectomes of 1199 human brains doing arbitrary tasks
- Lesion-Derived Estimated Functional and Structural Connectomes Predict Post-Stroke Recovery Outcomes
- Connectomes: a re-emerging hub dataset for neuroscience
BibTeX
@misc{rahman2026explainable,
title = {{Explainable Deep Learning of Resting-State Functional Connectomes Reveals Network Biomarkers of Adolescent Intelligence}},
author = {Md. Tanvir Rahman and Nabil Anan Orka and Asaduzzaman Khan and Mohammad Ali Moni},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.33422},
eprint = {2609.33422},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.33422},
}
RIS
TY - UNPB
TI - Explainable Deep Learning of Resting-State Functional Connectomes Reveals Network Biomarkers of Adolescent Intelligence
AU - Md. Tanvir Rahman
AU - Nabil Anan Orka
AU - Asaduzzaman Khan
AU - Mohammad Ali Moni
PY - 2026
DA - 2026/09/27
PB - arXiv
DO - 10.48550/arxiv.2609.33422
UR - https://arxiv.org/abs/2609.33422
AB - Mapping resting-state brain organization to individual differences in cognitive ability remains a major challenge in population neuroinformatics. Although deep learning enables flexible modeling of brain connectivity, limited interpretability restricts its scientific and clinical utility. To address this objective, we developed an explainable deep learning framework based on sparse projected residual networks to predict fluid, crystallized, and total intelligence from resting-state functional magnetic resonance imaging in 5,285 participants from the Adolescent Brain Cognitive Development study. We incorporated three complementary explainability methods (Integrated Gradients, Gradient Shapley Additive Explanations, and Occlusion) to interpret model behavior. The framework outperformed existing approaches, achieving Pearson correlations of 0.44, 0.58, and 0.56 for fluid, crystallized, and total intelligence, respectively, corresponding to predictive improvements of 6 to 9 percent. All three explainability methods produced near-identical feature rankings (pairwise rank correlations greater than 0.99). Consensus maps revealed a dual-layered functional architecture where primary predictive hubs localized within canonical systems, while the strongest global predictive pathways frequently bypassed these hubs through distributed, long-range relay connections. These findings suggest that intelligence emerges from the interaction between localized computational hubs and distributed communication pathways. Ultimately, these normative network architectures provide clinical reference maps to detect individual deviations, supporting earlier diagnosis, cognitive subtype stratification, and treatment monitoring in atypical neurodevelopment.
ER -