EEG Information Transfer Changes in Different Daily Fatigue Levels During Drowsy Driving

Huang, Kuan-Chih; Tseng, Chun-Ying; Lin, Chin-Teng · 2024 · Crossref

DOI: 10.1109/ojemb.2024.3367496

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 investigates how daily fatigue levels influence brain connectivity and driving performance, addressing a gap in understanding the neurophysiological markers of drowsy driving. While previous research has focused on spectral power changes or functional specialization, this work examines effective connectivity—specifically the directional flow of information between brain regions—across three distinct daily fatigue states: low (normal), medium (reduced), and high (high-risk). The authors hypothesize that variations in daily fatigue alter the brain connectivity-behavior relationship during simulated driving tasks. The experimental design involved sixteen healthy subjects who performed a virtual reality lane-keeping task while wearing a ReadiBand actigraph to objectively measure daily fatigue levels using the Sleep, Activity, Fatigue, and Task Effectiveness (SAFTE) model. Subjects were categorized into three groups based on their effectiveness scores over a month. Electroencephalography (EEG) data were recorded from six key channels (Fz, Cz, C3, C4, Pz, Oz) during the task. The researchers utilized Transfer Entropy (TE), a model-free information-theoretic measure, to calculate effective connectivity between channel pairs. TE values were normalized against baseline performance and analyzed in relation to reaction times (RT) to observe connectivity dynamics as driving performance declined. The results revealed distinct patterns of brain connectivity changes depending on the fatigue level. In the low- and medium-fatigue groups, connectivity exhibited an inverted U-shaped change as performance deteriorated, indicating a complex relationship between connectivity and behavioral lapse. However, this pattern was absent in the high-fatigue group. Specifically, as fatigue increased from low to high levels, connectivity magnitude decreased in frontal regions and increased in occipital regions. Statistical comparisons showed significant differences in TE values between the high-risk state and both the reduced and normal states, particularly in frontal and motor-related channel pairs during short reaction times. In contrast, differences between the reduced and normal states were less pronounced. The high-fatigue state was characterized by a sharp decline in connectivity magnitude at earlier stages of reaction time degradation compared to the other groups. These findings suggest that daily fatigue levels significantly modulate the brain connectivity-behavior relationship during driving. The shift from an inverted U-shaped connectivity pattern in lower fatigue states to a monotonic decline in high-fatigue states indicates a fundamental change in neural processing under severe fatigue. The reduction in frontal connectivity and increase in occipital connectivity highlight specific neurophysiological signatures of drowsy driving. This study underscores the importance of considering daily fatigue levels when analyzing brain connectivity for driver monitoring systems, providing evidence that effective connectivity metrics can distinguish between varying degrees of fatigue-induced performance decline.

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.

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.

• drowsiness detection algorithms
• drowsiness
• time on task
• sleep deprivation
• vigilance

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