InertiaMPDD: Multimodal Personalized Depression Detection via Emotional Inertia-Inspired Dynamics Modeling
Multimodal depression screening estimates risk from speech, visual behavior and language. Current methods mainly predict labels, with limited evidence of departures from a non-depressed reference. Recording-level supervision also leaves fine-grained emotional dynamics unannotated. Motivated by emotional inertia, the persistence of affect, InertiaMPDD learns healthy next-state patterns using a world-model-inspired latent predictor. Using pretrained multimodal representations, it predicts from observed history and context. Reference deviation, residual decay rate and a Jacobian spectral descriptor form reference-derived emotional dynamics descriptors, evaluated as candidate proxy biomarkers for standalone screening and personalized decision correction. Across five cohorts from four datasets, fusion reaches 98.41% F1 on D-Vlog and 84.98% on LMVD. Standalone descriptors outperform fixed additive fusion on MPDD but discriminate weakly on AVEC2014. These findings distinguish reference informativeness from fusion benefit and reveal context-dependent utility.
Authors
- Lang He (ORCID: https://orcid.org/0000-0003-2515-8579)
- Changzeng Fu (ORCID: https://orcid.org/0000-0003-1083-9486)
- Chang Gao (ORCID: https://orcid.org/0009-0006-5387-3026)
- Hongwei Ding (ORCID: https://orcid.org/0000-0002-0851-1994)
- Jiarui Wu
- Xinyi Zhang
- Yong Li
- Shiqi Zhao
- Mo Meng
- Songyang Wang
- Mingze Li
- Peilin Li
Publication Details
- Journal
- arXiv (Cornell University)
- Published
- 2026-11-12
- DOI
- https://doi.org/10.48550/arxiv.2611.16647
- Primary Topic
- Emotion and Mood Recognition
- Type
- preprint