Effect Of Driving Duration On Eeg Fluctuations

Puspasari, Maya Arlini; Iridiastadi, Hardianto; Sutalaksana, Iftikar Zahedi; Sjafruddin, Ade · 2017 · Crossref

DOI: 10.14716/ijtech.v8i6.716

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 the impact of prolonged driving duration on electroencephalogram (EEG) fluctuations to identify the most effective EEG parameters for measuring driver fatigue. Motivated by the high incidence of road accidents in Indonesia, largely attributed to mental fatigue, the research addresses discrepancies in prior literature regarding which specific EEG indicators best correlate with fatigue onset. While EEG is considered the gold standard for fatigue measurement, previous studies have yielded conflicting results concerning the optimal frequency bands and ratios for detection. Consequently, this study aims to evaluate how a three-hour driving duration affects EEG activity and to determine the most accurate EEG parameter for distinguishing between alert and fatigued states. The experimental design involved seven commercial drivers aged 25 to 35, who completed a three-hour driving session using a medium-fidelity simulator. Participants were screened for sleep quality and prohibited from consuming caffeine or smoking prior to the experiment. EEG data were collected using an Emotiv headset, focusing on six frontal channels associated with cognitive control. Measurements were taken for five minutes before and after the driving task. The data underwent pre-processing, including band-pass filtering and Fast Fourier Transform decomposition into delta (0–4 Hz), theta (4–7 Hz), alpha (7–13 Hz), and beta (13–20 Hz) bands. Relative Power Ratios (RPR) and specific ratios (θ/β, θ/(α+β), (θ+α)/β, and (θ+α)/(α+β)) were calculated. Subjective fatigue was assessed using the Karolinska Sleepiness Scale (KSS). Statistical analysis included one-way ANOVA, Pearson correlation, and Receiver Operating Characteristics (ROC) curve analysis to evaluate the sensitivity and specificity of each EEG parameter against subjective fatigue scores. The results demonstrated significant changes in EEG activity after three hours of driving. Specifically, there was a decrement in alpha and beta band activities and an increment in theta and delta activities. Correlation analysis revealed strong positive correlations between alpha-beta and theta-delta bands, and strong negative correlations between the other band combinations. ROC curve analysis identified three parameters as the most effective indicators of fatigue: the RPR of theta, the RPR of alpha, and the ratio of θ/(α+β). These parameters achieved high accuracy rates above 85%, with RPR theta showing the highest Area Under the Curve (AUC) at 91.2%. The study established specific cutoff values for these parameters to classify drivers as fatigued, with RPR theta exceeding 0.223, RPR alpha exceeding 0.238, and θ/(α+β) exceeding 0.426 indicating a fatigued state. The significance of this research lies in its identification of robust EEG markers for driver fatigue, specifically highlighting RPR theta as a superior indicator compared to previously suggested ratios like (θ+α)/β. The findings suggest that EEG-based technology utilizing these specific parameters could be developed for real-time fatigue detection systems in vehicles. Such systems could trigger visual, auditory, or haptic countermeasures to mitigate the risks associated with prolonged driving. The study concludes that a three-hour driving duration significantly induces fatigue, necessitating interventions such as mandatory rest periods. While the use of a medium-fidelity simulator is a limitation, the results provide a validated foundation for future research using high-fidelity simulators or real-world driving conditions to enhance transport safety.

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_with_issues.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• drowsiness detection algorithms
• time on task
• neuro workload indices
• drowsiness
• 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
• Methodological Resource: validation psychometrics