Stationary gaze entropy predicts lane departure events in sleep-deprived drivers

Shiferaw, Brook; Downey, Luke A.; Westlake, Justine; Stevens, Bronwyn; Rajaratnam, Shantha M. W.; Berlowitz, David J.; Swann, Phillip; Howard, Mark E. · 2018 · OpenAlex-citations

DOI: 10.1038/s41598-018-20588-7

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 sleep deprivation alters visual scanning behavior and whether these changes predict driving performance decrements, specifically lane departure events. While blink parameters are established indicators of drowsiness, the impact of sleep deprivation on gaze entropy—a measure of visual scanning uncertainty and structure—had not previously been examined. The research aimed to determine if sleep deprivation and task-induced fatigue disrupt gaze distribution and scanning patterns, thereby impairing visual awareness and increasing crash risk. The researchers conducted a randomized, repeated-measures study with nine healthy participants who drove a fully instrumented vehicle on a closed track for two hours under two conditions: after a night of normal sleep and after 24 hours of total sleep deprivation. Eye movement activity, including fixations, blinks, and saccades, was recorded alongside lane departure events. Gaze entropy was quantified using stationary gaze entropy ($H_s$), which measures the spatial dispersion of fixations, and transition gaze entropy ($H_t$), which assesses the predictability of scanning patterns. Statistical models were employed to analyze the effects of condition and driving duration on ocular metrics and to predict the likelihood of lane departures. Results indicated that sleep deprivation significantly increased blink rate, blink duration, saccade amplitude, and both stationary and transition gaze entropy, while reducing fixation rate. These impairments worsened with driving duration. Sleep-deprived drivers exhibited a 3.79-fold higher rate of lane departure events compared to rested drivers, and six of nine participants terminated the drive early due to fatigue. Crucially, a prediction model incorporating ocular measures revealed that stationary gaze entropy and fixation rate were significant predictors of lane departure events. Specifically, the odds of a lane departure increased by 7% for every 1% increase in stationary entropy, whereas higher fixation rates reduced the odds. Transition entropy did not significantly contribute to the prediction model. The findings demonstrate that sleep deprivation impairs driving not only by increasing eyelid closure but also by disrupting top-down control of visual attention, leading to more dispersed and random gaze patterns. This deterioration in visual sampling reduces the driver’s ability to monitor their lane position effectively. The study highlights stationary gaze entropy as a promising physiological indicator for detecting drowsiness-related impairment. Integrating gaze entropy analysis into driver alerting systems could provide more precise monitoring of driver state relative to environmental demands, potentially reducing accidents caused by fatigued driving.

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-prismaquant on 2026-06-26; verification: verified.

Topics

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

• drowsiness
• drowsiness detection algorithms
• sleep deprivation
• vigilance
• microsleep
• eye movements scanning

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: behavioral performance data, physiological data