The arousal paradox in critical task performance in automated driving during sleep inertia using a quasi experimental approach

Tomzig, Markus; Wörle, Johanna; Neukum, Alexandra; Baumann, Martin · 2025 · Crossref

DOI: 10.1038/s41598-025-08726-4

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 "arousal paradox" observed during sleep inertia in automated driving scenarios, where drivers must resume control immediately after forced awakenings. While traditional theories view arousal as a holistic state where cortical, physiological, and subjective measures align, recent evidence suggests discrepancies in high-stress, immediate-action contexts. The research aims to determine if cortical, physiological, and self-reported arousal diverge when drivers are awakened from different sleep stages to take over vehicle control, challenging the assumption that these dimensions are mediated by the same mechanisms. The researchers employed a quasi-experimental design using a high-fidelity driving simulator with 24 participants. Participants completed four drives: one baseline condition where they remained awake, and three conditions where they were instructed to sleep for 20, 40, or 60 minutes during automated driving before receiving a takeover request (TOR). The sleep stage immediately preceding the TOR served as the independent variable, categorized as wakefulness (W), Stage 1 (N1), or deeper sleep (N2/N3 combined). Data collection included electroencephalogram (EEG) recordings to measure cortical arousal (delta, theta, alpha, and beta power), heart rate monitoring for physiological arousal, and the Karolinska Sleepiness Scale for subjective sleepiness. Multilevel regression analyses and repeated measures correlations were used to assess the impact of the last sleep stage on these arousal metrics during the takeover period and the subsequent manual drive. The results revealed a significant divergence in arousal markers, termed the "arousal paradox." Following awakenings from N2 or N3, cortical arousal was low, indicated by increased delta, theta, and alpha activity. However, physiological arousal was elevated, evidenced by increased beta activity and higher heart rates compared to baseline. Significant positive correlations were found between delta activity, heart rate, and self-reported sleepiness. This pattern contradicts the holistic arousal model, as participants exhibited signs of both low cortical activation (sleep inertia) and high physiological activation (stress response) simultaneously. The authors hypothesize that the heightened physiological response is driven by the stress of performing demanding tasks under sleep inertia rather than general alertness. The study concludes that forced awakenings from N2 or N3 should be avoided in automated driving systems to mitigate performance impairments. If such awakenings are unavoidable, sufficient time must be allowed between awakening and duty resumption to permit arousal normalization. These findings imply that arousal is not a unitary construct in critical, time-pressured scenarios, necessitating a reevaluation of safety guidelines for pilots, emergency responders, and automated vehicle users who may face immediate takeover demands after napping.

Key finding

Awakenings from N2 or N3 sleep stages during automated driving result in an 'arousal paradox' characterized by low cortical arousal but elevated physiological arousal and stress.

Methodology

simulator

Sample size: 24

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 2026-06-10; verification: verified.

Topics

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

• drowsiness
• sleep deprivation
• drowsiness detection algorithms
• time on task
• circadian factors
• microsleep

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, behavioral performance data
• Theoretical Contribution: theory or model