Commonly experienced acute and cumulative 8-hour sleep loss differentially affect simulated driving during a night-time to early-morning commute: a randomized controlled crossover trial.
Stefan Lakämper, Michael Scholz, Thomas Kraemer et al.
✦ AI-curated · Sources linked
In 30 seconds
This randomized within-subject crossover trial compared simulated driving after normal sleep, four nights with 2 h less sleep per night, and one night without sleep in 20 healthy men. Each condition was followed by a nighttime-to-early-morning commute simulation. Acute total sleep deprivation markedly impaired vehicle control: lateral control deteriorated within 4-6 min, with impairment comparable to that reported at high blood alcohol concentrations (1.2-1.4 g/L). Cumulative sleep restriction produced smaller and less consistent driving changes, despite increased subjective sleepiness. The findings suggest that acute and cumulative sleep loss do not affect simulated driving performance in the same way.
Key findings
- After acute total sleep deprivation, lateral control deteriorated within 4-6 min, with impairment comparable to that reported at blood alcohol concentrations of 1.2-1.4 g/L.
- Acute deprivation also worsened steering stability and integrated driving performance.
- Four nights of sleep restriction, each with 2 h less sleep, produced smaller and less consistent driving changes.
- Vigilance task performance and speed variability were comparatively preserved across conditions.
Why it matters
The results indicate that the pattern of sleep loss may matter when assessing driving impairment: one sleepless night was associated with rapid deterioration in operational vehicle control, whereas several nights of restricted sleep had less consistent effects in this simulation. Distinguishing these patterns may help frame evaluations of sleep-loss-related driving risk, while recognizing that simulated performance does not establish real-world crash risk.
What to keep in mind
The study included 20 healthy male participants and assessed driving in a simulator, so findings may not generalize to other populations or directly quantify real-world crash risk.