MovieSTAGE: Scene, Transition, and Global Encoding for Movie-fMRI ADHD Classification
Abstract
Naturalistic movie-fMRI provides a shared, temporally structured probe of brain dynamics, yet predictive models commonly rely on whole-run functional connectivity (FC) or temporally generic representations that are not aligned with narrative events. We introduce MovieSTAGE (Scene, Transition, and Global Encoding), a multiscale framework that combines hypergraph-structured FC-profile organization within scenes, unsigned FC-profile differences across adjacent scenes, and whole-movie FC. We evaluated 260 participants from the CMI-HBN Despicable Me cohort on case-control, ADHD-subtype, and three-class classification using 10 repetitions of stratified five-fold cross-validation, complete out-of-fold (OOF) predictions, and paired subject-cluster bootstrap and permutation tests. MovieSTAGE achieved AUROCs of 0.69, 0.73, and 0.75 and balanced accuracies of 67.6%, 69.8%, and 58.3%, respectively, yielding the highest mean point estimates among the evaluated methods. On the three-class task, the full model outperformed all two-branch variants, the HGNN scene encoder outperformed MLP, GAT, and BNT alternatives under matched settings, and the human-annotated partition outperformed duration-matched random and fixed-count GSBS controls. These controlled results support incremental predictive value from event-aligned scene and transition representations when combined with whole-movie FC in this cohort. Post-hoc model-derived analyses generated network-level hypotheses involving frontoparietal and default-mode systems.
Related
- Eye movement markers of relational encoding predict scene memory precision
- Functional organization and natural scene responses across mouse visual cortical areas revealed with encoding manifolds
- Global tree encoding of atlas-scale single-cell genomics
- Urban Vietnam at the microbiome transition: Co-existing enterotypes reflect the global industrialization patterns
- Divergent nitrogen transition pathways during global agricultural development
BibTeX
@misc{kim2026moviestage,
title = {{MovieSTAGE: Scene, Transition, and Global Encoding for Movie-fMRI ADHD Classification}},
author = {Boseong Kim and Haejun Chung and Ikbeom Jang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.09306},
eprint = {2610.09306},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.09306},
}
RIS
TY - UNPB
TI - MovieSTAGE: Scene, Transition, and Global Encoding for Movie-fMRI ADHD Classification
AU - Boseong Kim
AU - Haejun Chung
AU - Ikbeom Jang
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.09306
UR - https://arxiv.org/abs/2610.09306
AB - Naturalistic movie-fMRI provides a shared, temporally structured probe of brain dynamics, yet predictive models commonly rely on whole-run functional connectivity (FC) or temporally generic representations that are not aligned with narrative events. We introduce MovieSTAGE (Scene, Transition, and Global Encoding), a multiscale framework that combines hypergraph-structured FC-profile organization within scenes, unsigned FC-profile differences across adjacent scenes, and whole-movie FC. We evaluated 260 participants from the CMI-HBN Despicable Me cohort on case-control, ADHD-subtype, and three-class classification using 10 repetitions of stratified five-fold cross-validation, complete out-of-fold (OOF) predictions, and paired subject-cluster bootstrap and permutation tests. MovieSTAGE achieved AUROCs of 0.69, 0.73, and 0.75 and balanced accuracies of 67.6%, 69.8%, and 58.3%, respectively, yielding the highest mean point estimates among the evaluated methods. On the three-class task, the full model outperformed all two-branch variants, the HGNN scene encoder outperformed MLP, GAT, and BNT alternatives under matched settings, and the human-annotated partition outperformed duration-matched random and fixed-count GSBS controls. These controlled results support incremental predictive value from event-aligned scene and transition representations when combined with whole-movie FC in this cohort. Post-hoc model-derived analyses generated network-level hypotheses involving frontoparietal and default-mode systems.
ER -