Predicting driver distraction using a single channel ear EEG
DOI: 10.64898/2026.01.24.701469
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 whether cognitive distraction during naturalistic driving can be reliably detected using a single-channel, non-invasive in-ear EEG, addressing the impracticality of conventional multi-channel scalp EEG for real-world deployment. The research was motivated by the need for objective, low-burden monitoring systems to detect cognitive distraction, which lacks overt behavioral markers and poses significant safety risks. The authors hypothesized that in-ear EEG could provide a temporally precise neural marker comparable to full-cap EEG, while also comparing its temporal dynamics against behavioral measures like eye movements and head rotation. Twenty-seven participants (18 women, 9 men; mean age 28.6 years) completed a dual-task paradigm in a highly immersive, full-scale driving simulator (VICTOR). Participants performed continuous vehicle control on a nighttime highway scenario while solving arithmetic problems of varying difficulty (low vs. high working-memory load) presented on an in-vehicle display. Data were recorded concurrently from a single-channel in-ear EEG device, a 24-channel scalp EEG system, Tobii Pro Glasses 3 for eye tracking and head kinematics, and driving performance metrics. The analysis employed time-resolved multivariate pattern analysis (MVPA) using linear discriminant analysis (LDA) to decode working-memory load with millisecond precision. Classifiers were trained on pooled data across participants using 10-fold cross-validation, and performance was quantified via area under the curve (AUC). Temporal generalization matrices were computed to assess the stability of neural and behavioral signatures. Results indicated that cognitive distraction was reliably decoded from the single-channel in-ear EEG, with detection latency and temporal generalization profiles closely matching those of the 24-channel scalp EEG. Although peak decoding performance (AUC) was higher for scalp EEG, the timing and temporal stability of distraction-related neural signatures were largely overlapping between the two neural modalities. Eye velocity provided the earliest and most sensitive behavioral marker of distraction, showing robust off-diagonal generalization indicative of sustained visual engagement. Head rotation contributed complementary but weaker information, with more transient temporal generalization. Scalp EEG topographies suggested that the decoded neural signals were linked to oculomotor and visuomotor processes. Behavioral performance confirmed the manipulation, with high-load trials resulting in slower response times and lower accuracy compared to low-load trials. The findings demonstrate that single-channel in-ear EEG serves as a practical, low-burden neural marker for cognitive distraction during driving. By prioritizing early detection and minimal hardware over maximal classification accuracy, the study identifies a viable operating point for wearable EEG-based driver monitoring systems. This supports the feasibility of fast, off-the-shelf decoding approaches for real-world applications, bridging the gap between laboratory-based neurophysiological research and scalable, real-time neuroadaptive driving technologies.
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 | — | — | 5 | 2026-08-23 |
| clean | success | clean | — | — | 2 | 2026-08-10 |
| chunk | success | chunk | — | — | 2 | 2026-08-10 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 2 | 2026-08-10 |
| promote | success | — | — | — | 1 | 2026-08-09 |
| summarize | success | llm | qwen3.8-27b-gittensor | summ-v5 | 3 | 2026-08-23 |
| tag | success | vector_similarity | — | — | 17 | 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.
- distraction detection algorithms
- drowsiness detection algorithms
- visual
- cognitive
- gaze based attention detection
- neuro workload indices
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
- Methodological Resource: tool software
- Theoretical Contribution: theory or model