Adaptive Filtering for Improved EEG-Based Mental Workload Assessment of Ambulant Users
DOI: 10.3389/fnins.2021.611962
archive: archived pipeline: cataloged verified
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 assessing mental workload (MW) using electroencephalography (EEG) in ambulant users, where movement artifacts severely degrade signal quality. While mobile EEG devices enable MW monitoring in ecological settings, conventional enhancement algorithms designed for ocular and muscle artifacts are suboptimal for removing motion-induced noise. The authors propose an adaptive filtering technique that utilizes accelerometer data as a reference signal to isolate and remove movement artifacts, thereby improving the reliability of EEG-based MW assessment during physical activity. The experimental design involved 48 participants who performed the Revised Multi-Attribute Task Battery-II (MATB-II) under low and high mental workload conditions. Participants engaged in three levels of physical activity (none, medium, high) while either walking/jogging on a treadmill or using a stationary bicycle. EEG data was collected via an 8-channel portable headset, and movement data was recorded using a chest-worn accelerometer. The proposed adaptive filter, based on the normalized least mean squares algorithm, was applied to the EEG signals. These enhanced signals were then processed using various benchmark algorithms, including Artifact Subspace Reconstruction (ASR), ADJUST, and the Harvard Automated Processing Pipeline for Electroencephalography (HAPPE). Features extracted included power spectral density, phase and magnitude spectral coherence, and amplitude modulation rate-of-change. Mental workload classification was performed using random forest and support vector machine classifiers. The results demonstrate that the adaptive filter significantly improved classification accuracy, particularly under high physical activity conditions where movement artifacts are most pronounced. When combined with the HAPPE enhancement pipeline and a fusion of all feature sets, the random forest classifier achieved a peak accuracy of 97.90% for distinguishing between low and high mental workload states. This performance was consistent across different physical activity levels and types, indicating that the proposed method effectively mitigates the negative impact of motion artifacts. Additionally, the study found that amplitude modulation features and phase/magnitude coherence provided complementary information to traditional power spectral features, contributing to the high classification performance. The significance of this work lies in its demonstration that accurate, real-time mental workload monitoring is feasible for ambulant users, such as first responders, who operate in dynamic environments. By integrating adaptive filtering with existing enhancement algorithms, the study overcomes a major limitation of mobile EEG applications. Furthermore, the analysis of top-ranked features revealed increased gamma activity in the parietal cortex during high workload, suggesting a link between sensorimotor integration, attention, and mental workload in moving subjects. These findings support the development of closed-loop systems that can monitor cognitive states in real-world, high-mobility scenarios.
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 | — | — | 124 | 2026-08-10 |
| 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 |
| promote | success | — | — | — | 1 | 2026-08-09 |
| summarize | success | llm | qwen3.6-27b-nvidia | summ-v5 | 123 | 2026-08-10 |
| tag | success | vector_similarity | — | — | 11 | 2026-08-11 |
| verify | success | — | — | — | 2 | 2026-08-10 |
Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.
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