FEEL: Quantifying Heterogeneity in Physiological Signals for Generalizable Emotion Recognition

TL;DR

Pragya Singh, Ankush Gupta, Somay Jalan, Mohan Kumar, Pushpendra Singh (IIIT-Delhi / RIT). arXiv 2604.05926 (posted 2026-04-07 America/Chicago). NeurIPS 2025 Datasets & Benchmarks.

Abstract

Emotion recognition from physiological signals has substantial potential for applications in mental health and emotion-aware systems. However, the lack of standardized, large-scale evaluations across heterogeneous datasets limits progress and model generalization. We introduce FEEL, the first large-scale benchmarking study of emotion recognition using electrodermal activity (EDA) and photoplethysmography (PPG) signals across 19 publicly available datasets. We evaluate 16 architectures spanning traditional machine learning, deep learning, and self-supervised pretraining approaches, structured into four representative modeling paradigms. Our study includes both within-dataset and cross-dataset evaluations, analyzing generalization across variations in experimental settings, device types, and labeling strategies. Our results showed that fine-tuned contrastive signal-language pretraining (CLSP) models (71/114) achieve the highest F1 across arousal and valence classification tasks, while simpler models like Random Forests, LDA, and MLP remain competitive (36/114). Models leveraging handcrafted features (107/114) consistently outperform those trained on raw signal segments, underscoring the value of domain knowledge in low-resource, noisy settings. Further cross-dataset analyses reveal that models trained on real-life setting data generalize well to lab (F1 = 0.79) and constraint-based settings (F1 = 0.78). Similarly, models trained on expert-annotated data transfer effectively to stimulus-labeled (F1 = 0.72) and self-reported datasets (F1 = 0.76). Moreover, models trained on lab-based devices also demonstrated high transferability to both custom wearable devices (F1 = 0.81) and the Empatica E4 (F1 = 0.73), underscoring the influence of heterogeneity. More information about FEEL can be found on our website https://alchemy18.github.io/FEEL_Benchmark/.

Verbatim from arXiv (opens in a new tab) · arXiv metadata (abstracts are released under CC0)

Results on ChatPPG leaderboards

All leaderboards

No results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.

Tasks & signals

Tasks

Stress

Signals

PPG EDA

Devices

Empatica E4 Muse S EEG headband PixArt PAG-2016BA wrist watch RespiBAN chest band ScientISST Shimmer ECG Unit Shimmer3 GSR+

Keywords

emotionvalencearousalbenchmarkCLSPEmpatica-E4NeurIPS-DBheterogeneitycross-dataset

Datasets

Show 6 more datasets

How it works

Cite this paper

APA

FEEL: Quantifying Heterogeneity in Physiological Signals for Generalizable Emotion Recognition. (2026). NeurIPS 2025 Datasets and Benchmarks. https://arxiv.org/abs/2604.05926

BibTeX

@inproceedings{feelheterogeneit2026feel,
  title         = {{FEEL: Quantifying Heterogeneity in Physiological Signals for Generalizable Emotion Recognition}},
  year          = {2026},
  booktitle     = {NeurIPS 2025 Datasets and Benchmarks},
  eprint        = {2604.05926},
  archivePrefix = {arXiv}
}

Built from the hub record. Check it against the publisher before submitting.