Multisubject “Learning” for Mental Workload Classification Using Concurrent EEG, fNIRS, and Physiological Measures
archive: archived pipeline: cataloged
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study addresses the challenge of accurately classifying mental workload levels in real-time neuroergonomic applications, specifically focusing on reducing the lengthy calibration time typically required for individualized brain-computer interface (BCI) decoders. The primary motivation is to develop a "multisubject learning" approach that leverages data from multiple participants to improve workload classification accuracy when only limited data is available from a target subject. The research integrates electroencephalogram (EEG), functional near-infrared spectroscopy (fNIRS), and physiological measures (heart rate variability, respiration) to classify three levels of cognitive load induced by an n-back working memory task. The experimental design involved 21 healthy right-handed participants who performed 0-back, 2-back, and 3-back tasks while simultaneously recording EEG (26 channels), prefrontal fNIRS (16 optodes), and physiological signals via a chest band. Data were processed to extract specific features: EEG band powers (delta, theta, alpha, beta) at the single-stimulus level; fNIRS oxy-hemoglobin and deoxy-hemoglobin amplitudes at the block level; and heart rate variability (HRV) spectral bands along with average heart and breath rates. Classification was performed using Linear Discriminant Analysis (LDA) with Naïve-Bayes fusion for multimodal combinations. To test the multisubject learning hypothesis, the study compared traditional calibration (using only target subject data) against a proposed method that combined target subject data with data from other subjects, using a weighted average of mean and covariance matrices (λ = 0.5). Performance was evaluated across varying calibration durations (13, 26, and 39 minutes) using a repeated learning-testing protocol. Results confirmed that integrating EEG and fNIRS significantly improved workload classification accuracy compared to using either modality alone. However, the inclusion of physiological measures did not significantly enhance classification performance beyond what was achieved with EEG or fNIRS alone. Crucially, the multisubject learning approach demonstrated that incorporating data from other subjects improved classification accuracy, particularly when the calibration data from the target subject was small. This suggests that inter-subject signal patterns are sufficiently similar to allow for effective transfer learning, thereby reducing the time required to train robust workload classifiers. The significance of this work lies in its potential to streamline the deployment of BCI systems for mental workload monitoring. By demonstrating that decoders can be trained using pooled data from multiple users, the study offers a practical solution to the "calibration bottleneck" that hinders the widespread adoption of neuroergonomic tools. This approach could facilitate faster setup times for operators in high-stakes environments, such as aviation or industrial settings, where continuous workload monitoring is essential for maintaining safety and performance. The findings support the viability of shared or semi-shared calibration protocols in multimodal neuroimaging applications.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-08-09 |
| archive | success | canonical_url | — | — | 1 | 2026-08-09 |
| extract | success | cached | — | — | 4 | 2026-08-23 |
| clean | success | clean | — | — | 1 | 2026-08-09 |
| chunk | success | chunk | — | — | 1 | 2026-08-09 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-08-09 |
| enrich | success | semantic_scholar | — | — | 1 | 2026-08-09 |
| promote | success | — | — | — | 1 | 2026-08-09 |
| summarize | success | llm | qwen3.8-27b-gittensor | summ-v5 | 3 | 2026-08-23 |
| tag | success | vector_similarity | — | — | 11 | 2026-08-11 |
| verify | success | — | — | — | 2 | 2026-08-09 |
Summary generated by qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Empirical Findings: physiological data