A systematic review of studies investigating the impact of sleep deprivation on drivers’ physiology and driving performance

Abstract

Driving is a multifaceted and risky activity that requires activation and utilisation of both cognitive and physiological capabilities. Sleep deprivation tends to impair cognitive function, which compromises drivers’ capabilities and increases the likelihood of crashes. Researchers have developed driver monitoring systems that can detect driving impairment by utilising driving performance measures and drivers’ physiological measures. However, sleep deprivation could induce specific physiological variations accompanied by changes in driving performance, thereby rendering detection of driving impairment challenging. This study aims to categorise drivers’ physiological and driving performance indicators associated with sleep deprivation and to evaluate existing evidence using a systematic review. Additionally, by examining the combined measures of behavioural and physiological state of vigilance, this review identifies correlations between various driving performance metrics and drivers’ physiological responses that can help in detecting the state transitions of drivers. The twenty-five studies that met the review criteria were chosen in accordance with the PRISMA framework from four research databases: Scopus, Web of Science, Transportation Research International Documentation (TRID), and IEEE Xplore digital library. Findings from this systematic review provide consistent evidence that sleep deprived driving induces physiological variations and leads to driving performance deficits. Sleep deprived driving resulted in increased electroencephalographic slow activity (alpha and theta power) of the brain and correlated with driving performance deficits. Ocular markers, including saccadic velocity, mean blink duration, variations in gaze behaviour, and PERCLOS, were able to detect physiological impairments while driving in sleep-deprived conditions. Combining physiological measures, such as slow eye movements and increased power in the alpha and theta bands of the EEG, also served as a robust measure of impaired driving performance. Notably, this review acknowledges limitations due to the diversity of methodologies across the studies, which complicates direct comparisons of findings. Nonetheless, these research findings will give directions for future research in developing strategies for robust real-time warning systems incorporating hybrid measures to mitigate the consequences of sleep deprived driving.