Reliability of a driving simulation task for evaluation of sleepiness

Driving Simulators reproduce situations that require tracking and visual searching, the main features of real driving. This study measured the reliability of a monotonous driving scenario to detect the circadian variations of alertness in healthy subjects. Five men and five women underwent a monotonous 30 min driving simulation task every 2 h. Before each driving task subjects completed the Stanford Sleepiness Scale (SSS) and the Visual Analogue Scale (VAS) to correlate the subjective measurements of sleepiness to the objective data of the simulator. Driving performances deteriorated or improved according to the circadian variation of alertness. The scenario is suitable to detect the consequences of sleepiness related to the circadian variations of alertness. The standard deviation of lane position, comparing the differences among the 10 min blocks in each task is the parameter most significant for the evaluation of sleepiness.