Estimating cognitive workload using a commercial in-ear EEG headset
archive: archived pipeline: cataloged
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study investigates the feasibility of estimating cognitive workload using a commercial in-ear electroencephalography (EEG) headset, the IDUN ‘Guardian’, to address the practical limitations of traditional scalp EEG in brain-computer interface (BCI) applications. While passive BCIs that monitor mental states like workload offer more natural interaction than active control paradigms, they are often hindered by the cumbersome setup of conventional EEG caps. In-ear EEG provides a more comfortable, user-friendly alternative suitable for everyday use, but its signal quality and classification accuracy have not been thoroughly benchmarked against standard systems. The research specifically examines whether a commercial in-ear device can distinguish between different levels of mental workload and evaluates the contribution of high-frequency gamma band activity to this estimation. The experimental design involved 16 participants performing two classical workload tasks: an n-back task with four difficulty levels (0-back to 3-back) and a mental arithmetic task with eight difficulty levels defined by q-value. Data was simultaneously recorded using a 26-electrode conventional EEG system and the 2-electrode IDUN Guardian in-ear headset. To facilitate a comprehensive comparison, the study analyzed signals in both low-frequency (1–35 Hz) and high-frequency (1–100 Hz) ranges. Additionally, surrogate in-ear signals were derived from the conventional EEG data to simulate multi-channel in-ear configurations (1, 2, and 4 channels) and to test advanced denoising algorithms. For the n-back task, a regularized linear discriminant analysis classifier was used, while ridge regression was employed for the mental arithmetic task to estimate q-values. Results indicated that the in-ear EEG system achieved statistically significant workload estimation performance, surpassing chance levels with 44.1% accuracy for four classes and 68.4% for two classes in the n-back task. However, conventional EEG demonstrated significantly superior performance, achieving 80.3% and 92.9% accuracy for the respective tasks. The surrogate measures derived from conventional EEG showed improved results over the actual in-ear device, with the 4-channel high-frequency surrogate reaching 57.5% and 85.5% accuracy. A key finding was that high-frequency signals (including the gamma band) consistently outperformed low-frequency counterparts in terms of classification accuracy, validating the importance of gamma band features in workload estimation. Statistical tests confirmed that all approaches performed significantly better than chance, though conventional EEG remained the most robust method. The significance of this work lies in providing a realistic assessment of commercial in-ear EEG technology for passive BCI applications. The study demonstrates that while current commercial in-ear headsets can detect workload changes, their accuracy is substantially lower than that of conventional scalp EEG. The findings suggest that enhancing in-ear systems with additional channels and leveraging high-frequency gamma band activity can improve performance. These insights offer guidelines for future development of more effective, comfortable, and practical EEG-based interfaces for everyday cognitive monitoring.
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 | openalex | — | — | 5 | 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 |
| 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