Temporal Graph Learning of Wearable Actigraphy and Sleep Traces for Modelling Adolescent Crystallized Intelligence

2026-09-27

AI Summary

This study addresses the challenge of predicting crystallized intelligence in adolescents from wearable device data, where irregular compliance and complex behavior–environment interactions pose significant difficulties. To this end, we propose SATURN, a network that models participants as temporal graphs, employing dynamic pruning to handle missing data while integrating sociodemographic covariates. The architecture innovatively combines residual GATv2 layers with a masked attention pooling mechanism, effectively avoiding interpolation artifacts and enhancing robustness. Experimental results demonstrate that SATURN achieves an R² of 0.2783, significantly outperforming traditional machine learning and sequential deep learning baselines. Furthermore, the model accurately identifies key behavioral predictors while ensuring fairness across demographic groups.

Abstract

Wearable actigraphy offers a scalable, ecologically valid alternative to episodic clinical assessment. However, predicting continuous adolescent crystallized intelligence ($G_c$) from such traces remains challenging due to irregular device adherence and complex behavioral-environmental interactions. We address this using daily summary data derived from 21-day Fitbit records of 6,091 adolescents in the Adolescent Brain Cognitive Development Study (Release 5.1). We propose SATURN, a Sleep-Activity Temporal Unified Regression Network. It represents participants as 21-node temporal graphs encoding daily behaviors and temporal adjacency. To prevent imputation artifacts, invalid-day edges are dynamically pruned during forward passes. Node embeddings are refined via residual GATv2 layers, aggregated through masked attention pooling, and fused with sociodemographic covariates. Under family-controlled, age-sex-BMI-stratified cross-validation, SATURN achieves $R^2 = 0.2783 \pm 0.0127$, consistently improving upon flattened machine learning (Gradient Boosting, $R^2 = 0.2372$) and sequential deep learning (BiLSTM, $R^2 = 0.2688$) baselines. Explainability analyses identify light activity, metabolic equivalents, and sleep duration as dominant predictors, while Monte Carlo dropout and subgroup analyses confirm equitable performance across sociodemographic strata. Ultimately, SATURN establishes a rigorous computational framework for digital cognitive phenotyping, offering a scalable pathway to complement traditional assessments by highlighting macro-level behavioral anomalies.

Problem

• crystallized intelligence
• wearable actigraphy
• sleep traces
• adolescent
• temporal graph learning

Innovation

• Temporal Graph Learning
• Graph Attention Network
• Wearable Actigraphy
• Crystallized Intelligence
• Digital Phenotyping