Driving behaviour
For the driver the main effect of fatigue is a progressive withdrawal of attention from the road and traffic demands, leading to impaired performance behind the wheel [15]. Research has found that a person who drives after being awake for 17 hours has impaired driving skills comparable to a driver with a 0.05 blood alcohol level. A driver who has gone without sleep for 24 hours has impaired driving skills comparable to a driver with an illegal high BAC of 0, 1 g/l [110]. In the case of sleepy drivers, the ultimate impairment is falling asleep at the wheel.
Several studies [27][91] have shown that fatigue influences driving behaviour in specific ways:
- Slower reaction times: fatigue increases the time taken to react in an emergency
- Reduced vigilance: subjects perform worse on attention-based tasks when sleep-deprived (e.g. a fatigued driver will be slower to notice oncoming hazards, such as roadworks or a railway crossing)
- Reduced information processing: fatigue reduces both the ability to process information and the accuracy of short-term memory (e.g. a fatigued driver may not remember the previous few minutes of driving).
Consequences on specific operational driving aspects
Below findings are reported about specific aspects of driving task control:
Steering
On-the-road studies have indicated that steering performance gradually deteriorates and that performance decrements are correlated with subjective ratings of fatigue [84][94].
In a simulator study, [52] found that even in a limited 2.5-h time-span subjective fatigue and sleepiness increased as a function of time on task. The increase of fatigue and sleepiness was accompanied by an increased aversion to continue driving and a deterioration of steering performance. Interestingly, larger increases of fatigue and sleepiness were associated with larger increases of aversion and a greater deterioration of steering performance. These results suggest that fatigue is accompanied by a decreased motivation to continue with the task.
In a study by Van Winsum [112], participants had to drive for 3 hours in a driving simulator on a two-lane road outside urban areas. As the drive progressed, steering performance became less flexible and the amplitude of steering corrections increased. Van der Hulst et al. [52] also found that, after a 2, 5 hour drive in a simulator, steering performance deteriorated. Desmond [26] found that effect of fatigue on steering performance and on lateral position were greater on straight road sections than in road curves. Desmond concludes that tired drivers have more difficulty regulating attention and performance in situations with low task demand (straight road sections) than in situations with high task demands (curves).
Ǻkerstedt et al. [10] compared performance in a driving simulator of shift-workers after a normal night sleep with performance after a normal night shift. Driving after the night shift was associated with an increased number of occurrences of two wheels outside lane marking and with an increased lateral deviation of the vehicle. Also, performance after nightshift led to increased eye closure and increased subjective sleepiness.
Speed choice
A German simulator study showed that participants drive faster the longer they performed the driving task [42]. However, driving faster did not diminish general driving performance. The researchers considered this as evidence for the hypothesis that drivers attempt to adapt their attention-level (by changing speed). In other words, by changing speed drivers may change sensory input which may spur the body and mind to put in extra effort to notice and respond to signals from the environment.
In a survey study by Oron-Gilad & Shinar [83], 12% of military truck drivers said that they drove slower when they were fatigued; 14% admitted to having difficulties with estimating own speed correctly. Riemersma et al [94] report that falling asleep goes together with lessened muscle force, manifesting itself also in less force on the accelerator pedal.
Following behaviour
In a study of time-on-task effects on car-following performance, Brookhuis et al [14] found that differences between the speed of the following car and the speed of the lead car became larger after 2.5 h of continuous driving. This indicates that accuracy in following the lead car's speed changes was reduced.
In a simulator study [52], it was found that participants who became more fatigued during the prolonged simulator drive increased their headway to a greater extent than participants who reported only a slight increases of fatigue. This applied to a lesser extent for participants who had to perform the task under time pressure. Time pressure may make the task more challenging and less monotonous thereby sustaining motivation to perform well.
| 
