Cognitive components of simulated driving performance: Sleep loss effects and predictors
Highlights
► We examined separate cognitive processes that underlie the task of driving. ► Simulated driving and cognitive tasks were assessed after sleep deprivation. ► Psychomotor vigilance was a key component of the variation in driving impairment.
Introduction
Drowsy driving is a contributing factor in a large proportion of motor vehicle accidents and related deaths around the world (Horne and Reyner, 1999, Connor et al., 2001). Sleep-related accidents are particularly prevalent in commercial motor vehicle drivers (Lyznicki et al., 1998, Sabbagh-Ehrlich et al., 2005). These types of accidents do not appear to be solely due to the driver falling asleep at the wheel, as laboratory studies have demonstrated that brief sleep episodes do not fully account for all of the performance decrements evident in sleep-deprived individuals (Welsh et al., 1998, Russo et al., 2000). For example, microsleeps, or bursts of delta or theta activity in the EEG, only preceded 18% of crashes in one driving simulator study; which is too infrequent to explain the incidence of crashes (Welsh et al., 1998). Supporting this view, other studies have reported that drivers are often awake and have their eyes open when they crash (Åkerstedt and Gillberg, 1990). This finding suggests that other aspects of motor, perceptual, and/or cognitive processing may be impaired in sleepy drivers, accounting for the increased sleep-related accident risk.
Driving is a complex task that requires a number of skills. The driver continuously receives information from the road scene, analyses it, and reacts according to knowledge of traffic systems, driving regulations, conditions of the vehicle, applications of the road rules and their previous driving experiences. Driving also involves the processing of complex visual, tactile, and auditory information in order to produce a well-coordinated motor output (Anstey et al., 2005). Simulated driving tasks have been designed to tap into the key processes that are involved in the task of driving. In addition, driving simulations have the ability to examine driving-related performance in a controlled, measurable, and safe environment (Gillberg et al., 1996). Sleep deprivation, sleep restriction, circadian variations and extended periods of time-on-task have been shown to cause a qualitative decrement in driving performance in both on-road and simulated driving tasks (Welsh et al., 1998, Pizza et al., 2004, Åkerstedt et al., 2005, Howard et al., 2007).
As driving involves a number of cognitive processes working in concert, it is difficult to determine from a simulated driving task alone which components are causing the impairment in overall driving performance. A number of cognitive domains have been associated with crash risk in on-road driving studies, including attention and vigilance, visual processes, processing speed and reaction time, working memory and executive function (Anstey et al., 2005). Many of these functions also overlap with neurocognitive impairments observed in sleep deprived individuals (Koslowsky and Babkoff, 1992, Jackson and Van Dongen, 2011).
Driver inattention has been identified as one of the leading causes of motor vehicle accidents (Treat et al., 1977). Low scores on attention and vigilance tasks are associated with higher crash-risk rates and on-road driving performance (Findley et al., 1995, Arnedt et al., 2005). Visual attention performance significantly predicts real-world accident frequency (Owsley et al., 1991). Aspects of motor speed, such as simple visual reaction time, are also important skills in adverse situations (e.g. being able to brake quickly if a pedestrian steps out on the road). Moderate correlations have been observed between simple reaction time tasks and on-road driving performance, with stronger correlations observed for complex reaction time (McKnight and McKnight, 1999). Slowing of reaction times and lapses in attention are also commonly observed after periods of extended wakefulness (Dinges et al., 1997, Van Dongen et al., 2003), and are associated with lane drifting on a simulated driving task (Baulk et al., 2008).
Driving is primarily automatised, although it does involve some shifts to controlled processing when routine reactions are insufficient to deal with novel or complex traffic situations (Lundqvist, 2001). Therefore, information processing speed is an important component of driving. The driver needs to process multiple stimuli simultaneously, select and filter stimuli according to the road situation, and process the information in a short time frame in order to judge the traffic scene and act appropriately (Lundqvist, 2001). The Digit Symbol Substitution Test (DSST) is one measure that assesses information processing and motor speed, and has been shown to be related to simulated driving performance in rested subjects (Szlyk et al., 2002). Impairments in DSST performance have also been observed in some (Williamson and Feyer, 2000, Van Dongen et al., 2003) but not all (Van Steveninck et al., 1999) studies of sleep restricted subjects.
Executive, higher order function is required for integrating new introspective, sensory and situational information, whilst suppressing distracting information by focusing attention on relevant stimuli and planning a response. A number of tasks that tap into executive functions have been found to correlate with driving skills (Lundqvist, 2001, Daigneault et al., 2002, Szlyk et al., 2002, Ramaekers et al., 2006a, Ramaekers et al., 2006b). The frontal cortex is largely thought to control attention and executive function and is vulnerable to even a single night of sleep deprivation, as demonstrated in neuroimaging studies (Drummond et al., 1999, Thomas et al., 2000, Jackson et al., 2011). Conditions that impair frontal lobe functioning, such as sleep deprivation and aging, may negatively impact on driving performance. This can potentially lead to a driver taking inappropriate risks, having poor insight into performance deficits, perseverating on maladaptive thoughts and actions, and having problems making behavioural modifications based on new information from the road scene.
Precisely what aspects of driving performance are affected in sleep deprived individuals remains unclear. Neurocognitive tasks may detect more subtle underlying impairments in an individuals’ driving performance not detected by real-world driving or driving simulators (Szlyk et al., 2002). In particular, the relationship and predictive value of neurocognitive tasks for simulated driving under conditions of sleep deprivation has not been examined. To determine which cognitive functions are associated with sleep-related driving impairment, this study employed a range of neurocognitive tasks that assess different cognitive components of driving and that have previously been shown to relate to crash risk. The aim of this study was to examine simulated driving and neurocognitive performance after a single night of sleep deprivation, and also examined the association between neurocognitive outcomes and driving performance measures.
Section snippets
Subjects
Nineteen professional drivers (1 female), aged between 23 and 62 years (mean age (sd) = 45.3 (9.1) years) participated. A medical practitioner interviewed subjects obtained other physiological measures (e.g. weight, height, blood pressure). Subjects were excluded if they had a medical condition which could be exacerbated by sleep deprivation such as cardiovascular disease, hypertension, epilepsy, diabetes, or psychiatric illness; a sleep disorder (Multivariate Apnea Prediction score > 0.5 (Maislin
Prior sleep of the subjects
The sleep diary results for the averaged hours of sleep each night for one week prior to each session did not differ between the normal sleep session (mean (sd) = 6.76 (0.99) h) and the sleep deprivation session (6.79 (1.14) h; p = 0.55). Similarly, there was no difference in the number of hours of sleep recorded on the night prior to the normal sleep session (6.68 (0.88) h) and the sleep deprivation session (6.90 (0.92) h; p = 0.98).
Effect of total sleep deprivation on performance and subjective sleepiness
Means and standard deviations of the simulated driving task
Discussion
Sleep deprivation is associated with increased accident risk and driving impairment, although the cognitive dysfunction underlying this impairment is not well understood. The current study sought to examine separate cognitive processes that are used in the task of driving, to determine which of these were most affected by sleep loss, and further, to examine which cognitive functions were associated with driving impairment. The main finding of this study was that measures of psychomotor
Conclusion
The findings from the present experiment suggest that a period of acute sleep deprivation can lead to deterioration in a number of driving-related processes, which in turn may affect individuals’ ability to drive safely. We identified decrements in driving and psychomotor function and subjective sleepiness following 27 h of acute sleep deprivation during a circadian peak period. While executive function was not significantly affected in this study, combined sleep deprivation and circadian
Acknowledgment
The authors would like to thank Professor Hans Van Dongen for his statistical advice.
References (60)
- et al.
Cognitive, sensory and physical factors enabling driving safety in older adults
Clinical Psychology Review
(2005) - et al.
Comparative sensitivity of a simulated driving task to self-report, physiological, and other performance measures during prolonged wakefulness
Journal of Psychosomatic Research
(2005) - et al.
How do prolonged wakefulness and alcohol compare in the decrements they produce on a simulated driving task?
Accident Analysis and Prevention
(2001) - et al.
Chasing the silver bullet: measuring driver fatigue using simple and complex tasks
Accident Analysis and Prevention
(2008) - et al.
The role of driver sleepiness in car crashes: a systematic review of epidemiological studies
Accident Analysis and Prevention
(2001) - et al.
Vigilance and automobile accidents in patients with sleep apnea or narcolepsy
Chest
(1995) - et al.
Using a driving simulator to identify older drivers at inflated risk of motor vehicle crashes
Journal of Safety Research
(2003) - et al.
Multivariate analysis of age-related driver ability and performance deficits
Accident Analysis and Prevention
(1999) - et al.
Fatigue, sleep restriction and driving performance
Accident Analysis and Prevention
(2005) - et al.
A driving simulation task: correlations with multiple sleep latency test
Brain Research Bulletin
(2004)