Fatigue and fatigue research: The Australian experience
Dr Narelle Haworth
Paper presented to 7th Biennial Australasian Traffic Education Conference, Speed, Alcohol, Fatigue, Effects, Brisbane, February 1998.
This paper reviews the extent and nature of fatigue in road crashes in Australia. It then summarises the research that has been undertaken in that area and the issues that have arisen. Countermeasures to reduce fatigue are discussed, along with constraints to their effective implementation.
In recent years there has been much interest in the role of driver impairment in the causation of road crashes. The identification of the role of alcohol in driver impairment and the consequent actions taken to reduce its incidence have played a large part in the reduction in road injuries and fatalities that has occurred in the past decade. An understanding of the role of driver fatigue has potential to lead to further improvements in road safety.
WHAT IS DRIVER FATIGUE?
The problems in specifying an adequate definition of fatigue have been well documented (see Cameron, 1973). However, most definitions include the concept of a deterioration with extended effort in work output, physiological well-being or feelings (Haworth, Triggs and Grey, 1988).
A simple model of driver fatigue was proposed by Hattori, Matsuura, Narumiya, Araki and Ohnaka (1987), who demonstrated three stages of driver performance in extended driving. The driver began in Stage One, characterised by alertness. In Stage Two (drowsy driving) drivers appeared to be sleepy, and "had a tendency to decrease their close attention to safety and to drive gazing vacantly at one unspecified point" (p.249.4). The car speed was kept fairly constant but there was often a delay changing speed in response to change of gradients of the road. This type of driving has been referred to as highway hypnosis (Williams, 1963).
Hattori et. al. (1987) report that in Stage Three (dim driving) "the consciousness level of the driver seemed to become even lower and blinkings were extremely reduced. The steering operation became more dilatory than in stage two and the zigzag driving within the permitted lane became pronounced. During this repeated zigzag driving the car sometimes crossed the centre line or ran off the side of the road" (p.249.4). In Stage Three a high level of fatigue is present and vehicle control is difficult to maintain.
The most dangerous aspect of driver fatigue is falling asleep at the wheel. While drivers affected by fatigue may have slower reactions and impaired visual scanning, they may be able to compensate to some extent for these impairments by, for example, slowing down or being less willing to overtake (Brown, Tickner and Simmonds, 1970). Näätänen and Summala (1978) conclude that "research has not generally been able to show that driver fatigue increases the risk of an accident except by increasing the probability of falling asleep during driving" (pp.27-28).
THE EXTENT AND NATURE OF FATIGUE IN ROAD CRASHES IN AUSTRALIA
An overall estimate of the proportion of accidents in which fatigue is involved is difficult to make and is likely to be misleading. The contribution of fatigue is greater in night-time accidents and in accidents occurring in rural areas. Because rural speeds are generally high and because sleeping drivers do not take evasive action, fatigue-related accidents are often severe.
Fatigue-related accidents are more common on rural highways than on urban roads. One reason for this is that average trip lengths are likely to be longer on these roads and inattention and drowsiness are brought on by the constant speeds and monotony. In addition, on such highways many other causes of accidents - poor access control, presence of unprotected utility poles, sub-standard road geometry, etc. - have been removed.
Identification of crashes as fatigue-related can be difficult. Coronial and police citations of fatigue occur in about 5% of all fatal crashes (Haworth and Rechnitzer, 1993), with percentages up to 14% in less populated states. Analysis of the same database for 1988, 1990 and 1992 showed that fatigue was involved in 5 to 10% of fatal semi-trailer crashes (Hartley, Arnold, Penna, Hochstadt, Corry and Feyer, 1996). However, these figures are based on very strict definitions of fatigue. Including crashes in which loss of concentration may have contributed to the crash increased the prevalence of fatigue in NSW crashes to about 17% (NSW Roads and Traffic Authority, 1993, 1994, cited in Hartley et al., 1996).
Ryan, Wright, Hinrichs and McLean (1988) carried out an in-depth study of crashes on rural roads near Adelaide. When all accident-involved drivers were asked, 31.4% responded that they had felt slightly, moderately or very fatigued just prior to the accident. The percentage reporting fatigue was higher for truck drivers (41.7%) and motorcycle riders (50.0%).
Many night-time rural accidents are single-vehicle accidents. Johnson (1980, cited in Chapman, 1985) studied night-time rural accidents in South Australia. In this study 71 per cent of accidents were found to be single-vehicle accidents. Of the single-vehicle accidents, Johnson found that 92 per cent were run-off-road accidents. These accidents mainly occurred either on curves or involved hitting roadside objects (usually trees).
A study of rural single-vehicle accidents in Victoria concluded that it was probable that the driver had fallen asleep in 27% of the crashes investigated (Armour, Carter, Cinquegrana and Griffith, 1988).
Driver fatigue in truck accidents
International evidence has accumulated which suggests that fatigue may be a significant contributor to truck crashes. Transportation Research and Marketing (1985) concluded that fatigue was a primary cause in 41% of heavy truck crashes in the western United States and a probable cause in a further 18%. Jones and Stein (1987) conducted a study in Washington State which found that the crash risk for drivers of articulated vehicles who had driven for more than eight hours was double that of drivers who had driven for less than eight hours. To put this in perspective, this means that someone who has driven for more than eight hours is operating at a similar risk to someone who has a BAC of .05.
The magnitude of the contribution of fatigue to truck crashes in Australia has become clearer in the last few years. In a NSW study, Fell (1987) reported that "articulated trucks have a high involvement in fatigue accidents in comparison with their involvement in other accidents" (p.60). Articulated trucks in NSW were found to have 3.7% of all fatigue-related accidents but only 1.5% of all non-fatigue-related accidents.
Leggett (1988) examined the contribution of fatigue to both car and truck crashes in Tasmania. He found that the contribution of fatigue increased with crash severity, mainly because of the often fatal nature of single-vehicle run-off-road crashes. Analysis of all reported crashes showed a slightly lower involvement of fatigue in rural single-vehicle truck than car crashes. When property damage crashes were removed from this sample, however, fatigue was shown to contribute more to truck than car crashes. Another important finding was that the likelihood of the truck driver being killed was greater in fatigue-related crashes than in other crashes.
A study of driver fatigue in fatal accidents involving a truck in Victoria was undertaken by the author and her colleagues (Haworth, Heffernan and Horne, 1989). Based on Coroners' verdicts, fatigue was a contributing factor in 9.1% of crashes. This figure is likely to be an underestimate because Coroners rarely judged fatigue to be a contributing factor unless there was strong evidence that the driver was likely to have been asleep at the time of the crash.
An alternative estimate of the contribution of fatigue was calculated by classifying as fatigue-related those crashes which involved several of the factors: extended driving hours, evidence of falling asleep at the wheel, comments about tiredness, driving right of centre in the absence of elevated BAC and night-time driving. This resulted in a judgement that fatigue was involved in 19.9% of the sample of fatal crashes involving trucks.
One of the interesting findings from the examination of Coroners' reports was that the car drivers involved in the fatal truck crashes were just as likely (or more likely) to have been fatigued as the truck drivers.
DRIVER FATIGUE RESEARCH
Driver fatigue research in the past can be classified into investigations of:
- the epidemiology of fatigue
- how to measure fatigue
- the factors contributing to the development of driver fatigue
- countermeasures to reduce fatigue
The epidemiology of fatigue
In Australia, several large surveys have been conducted to describe the extent and nature of the experience of fatigue among car (Fell, 1995; Fell & Black, 1996) and truck drivers (Hartley et al., 1996; Haworth, Vulcan, Schulze and Foddy, 1991; Williamson, Feyer, Coumarelos and Jenkins, 1992).
About one-quarter of car drivers reported having had a fatigue-related incident and 4% reported having had a fatigue-related accident in a NSW survey conducted by Fell (1995). Fell and Black (1996) found that 12% of Sydney drivers reported having a fatigue related incident. About a third of these had commenced the trip in a fatigued condition.
The surveys of truck drivers found that fatigue is a common problem. Williamson et al. (1992) reported that most truck drivers considered fatigue to be a substantial problem for the road transport industry and one third of drivers reported it as a substantial personal problem. Haworth et al. (1991) found that almost half the drivers reported that they drive on the edge of falling asleep at least sometimes. Hartley et al. (1996) reported that truck drivers in Western Australia perceived fatigue as less of a problem than drivers in other states.
Measures of fatigue
Measures of fatigue generally fall into those which are measured before and after the task and those that are measured during the task.
Before and after measures include:
- reaction time
- critical flicker fusion
- critical tracking task
- simulated driving
- subjective ratings
- analysis of bodily fluids
Measures during driving include
- performance measures (speed, steering wheel movements, lateral position)
- secondary tasks (visual or auditory reaction time)
- physiological measures (EEG, skin conductance, cardiac activity, eye movements, eye closures)
Factors contributing to the development of driver fatigue
The earliest view of driver fatigue was that it was directly related to the number of hours spent driving. Since then a large number of factors have been demonstrated to affect the development of fatigue. Figure 1 shows an early schematic representation of the variety of factors contributing to fatigue and the build up of chronic fatigue if recovery time is insufficient.
Figure 1. Schematic representation of the cumulative effect of daily causes of fatigue (from Grandjean, 1968). Fatigue is compared to the level of liquid in a container, and recovery is shown as the outflow from the container.
Time of day is probably the strongest factor affecting the development of driver fatigue. This was confirmed by the Driver Fatigue and Alertness Study (Wylie, Shultz, Miller, Mitler and Mackie, 1996) which compared driver performance on 10 and 13 hour workdays. The study showed that number of hours or days of driving were not strong or consistent predictors of driver fatigue.
Detailed discussions of the factors contributing to driver fatigue can be found in Haworth (1995) and Haworth, Triggs and Grey (1988).
COUNTERMEASURES TO REDUCE FATIGUE
A large number of measures have been proposed or implemented with the aim of reducing death and injury from fatigue-related crashes. These measures are classified in Table 1, according to their aim and whether they target the driver, vehicle or environment. Fatigue countermeasures can have three aims:
To prevent fatigue by maintaining driver alertness
To prevent fatigued drivers crashing by providing a warning
To reduce the severity of fatigue-related crashes.
Table 1. A classification of fatigue countermeasures (from Haworth, 1990).
|PREVENT FATIGUE||PREVENT CRASHES||REDUCE CRASH SEVERITY|
|DRIVER||Education, Limitation of hours of work|
|VEHICLE||Radio, Ventilation, Reduction of vibration||Fatigue monitors, Antilock brakes||Seat belts, Antilock brakes|
|ENVIRONMENT||Rest breaks||Pavement treatments||Duplication, Better shoulders|
The most widely implemented of these countermeasures have been education, limitation of hours of work, rest breaks, fatigue monitors and pavement treatments.
Educational programs to reduce driver fatigue and the constraints to their success are discussed by Haworth (1996). Some of the specific goals of fatigue educational programs are:
- to educate the public of the dangers of fatigue (information)
- to convince people that fatigue is an important road safety issue (attitude change)
- to get people to plan trips better (behaviour change)
- to get people to stop if feeling tired (behaviour change).
Traditionally, educational programs have been most successful at informing people and least successful in changing behaviour (unless they have been undertaken in support of an enforcement program, Cameron, Haworth, Oxley, Newstead and Le, 1993).
Two methods of educating drivers about the dangers of fatigue and ways of reducing fatigue have been used: media campaigns and incorporation of fatigue education in driver training courses. In general, fatigue advertising has been concentrated in the periods leading up to Christmas and Easter holidays. In New South Wales and Queensland particularly, advertising has focussed on providing support for volunteer-operated rest area refreshment break programs (Driver Reviver, Operation Coffee Break).
A Fatigue Management Training Course has been developed for the National Road Transport Commission (National Road Transport Commission, 1996) as part of a Transitional Fatigue Management Scheme. The purpose of the course is
to provide long-distance heavy vehicle drivers with relevant knowledge about the causes and effects of fatigue, and skills to manage fatigue in their driving activities and lifestyles, consistent with the requirements of the national Fatigue Management Model. (p.1)
The two major factors affecting the likelihood of success of educational programs to reduce fatigue are the ability of fatigued drivers to judge the level of risk at which they are operating and the incentives to continue driving.
Wertheim (1978) proposed that drivers be taught to recognise the early signs of fatigue such as misjudgment of velocities, crossing marked lane lines, slow responses and yawning. Once these events occur drivers should rest. However, experimental evidence suggests that subjective estimates of fatigue may not be reliable (e.g., Yabuta, Iizuka, Yanagishima, Kataoka and Seno, 1985). This agrees with the intuitive reasoning that if drivers knew that they were about to fall asleep at the wheel, that they would stop and so avoid crashing.
The effectiveness of attempts to persuade drivers to take rest breaks (including educational attempts) is limited by the real incentives to continue driving. These incentives are particularly strong for truck and bus drivers who have a need to meet schedules. For car drivers, time is also sometimes a factor.
Among young male car drivers, there is something of a machismo element that says that it is a sign of weakness to need to stop. Another widely reported disincentive to stopping for both car and truck drivers is the lack of attractive and practical places to stop. While rest areas exist, they are often not attractive at the time of day when needed most. Very few rest areas have adequate lighting and personal security.
In addition to these factors, Haworth (1996) discusses a number of more general advertising factors which may affect the success of fatigue educational programs.
There have been few evaluations of whether fatigue educational programs actually change behaviour and thus prevent fatigue-related crashes. However, the evaluations which have been conducted suggest that fatigue advertising seems to reach and be recalled by the target audiences. Thus advertising may be successful in informing drivers about the dangers of fatigue. The ultimate benefit of increasing the level of public awareness of fatigue as a road safety issue may be in making more acceptable the possible future introduction of technological or legislative measures which will be effective in reducing fatigue-related crashes (Haworth, 1996).
Limitation of hours of work
Most developed countries have regulations which limit driving hours for drivers of heavy vehicles. The regulations commonly include a limitation on the maximum number of hours that can be driven per day or per week and specifications relating to the length and timing of rest periods. Some jurisdictions limit driving hours (that is, hours behind the wheel) whereas others limit working hours (which may include loading, paperwork, waiting etc.). The underlying assumption is that limiting the hours of driving per day and per session results in drivers who are more alert and are, therefore, involved in fewer crashes.
The current driving hours regulations were framed at a time when little was known about the causes of fatigue and so they do not take into consideration many of the factors that are now known to be important. Feyer and Williamson (1995) note that current driving hours regulations have three critical shortcomings: they place limits on consecutive hours of work and rest irrespective of the time of day, they are not derived from empirical research basis and do not take into account inter- and intra-driver variability. "The omission of important factors affecting alertness levels, such as time of day, activity during rest breaks and prior activity, means that these regulations are incapable of being completely effective, even if problems related to enforcement were solved" (Haworth, 1995, p.46).
While improving safety by reducing fatigue-related crashes has been a major aim of the regulation of driving hours, the ability of these regulations to achieve this aim has been questioned in the last decade (Haworth, 1998). In addition, the development of alternatives to driving hours regulations - driver alertness monitoring or fatigue management programs - have led to an increasing debate about the usefulness in both safety and productivity terms of driving hours regulations.
Advance planning of trips and scheduled rest breaks have been advocated as measures which can be taken by car drivers to minimise the development of driver fatigue and reduce the risk of crashing as a result of falling asleep at the wheel. However, the effectiveness of rest breaks has been shown to depend on a number of factors.
Rest breaks may not be helpful once a considerable level of fatigue has developed (Harris and Mackie, 1977; Lisper and Eriksson, 1978, cited in Lisper, Laurell and van Loon, 1986). Lisper and Eriksson (1977, cited in Lisper and Eriksson, 1980) investigated the effect on truck drivers of a 30 minute break between a 6 and a 5 hour session. They found that breaks were less effective at night than during the day and were more effective for younger than older drivers.
Studies of the minimum length of an effective rest break have been conducted in the laboratory and on the road. In the laboratory, vigilance studies have shown improvements after breaks of only 5 to 10 minutes (Colquhoun, 1959, McCormack, 1958, both cited in Lisper and Eriksson, 1980). In contrast, Lisper and Eriksson (1980) found no difference between 15 and 60 minute rest breaks in a field experiment.
Many rest breaks taken by drivers include consuming food. The issue of the amount of improvement that relates to the rest break alone versus the food is one that has clear practical significance.
Lisper and Eriksson (1980) found that rest breaks with food led to "a comparatively small deterioration in performance after the pause" (p.119). For those subjects who had rest breaks without food, the reaction time increase in the period before the break continued.
We have conducted several experiments on the effects of food and rest breaks on the performance of fatigued drivers at the Monash University Accident Research Centre. In each experiment, drivers performed a reaction time test, drove a simulator during the night and then repeated the reaction time test. Driver performance was measured during simulated driving in terms of the accuracy of steering. More direct measures of alertness comprised the duration and frequency of long eye closures (greater than 0.5 seconds).
The first experiment showed that the improvement resulting from a 15-minute rest break was greater for subjects who were given a snack (muesli bar and orange juice) during the break than for subjects who had a break with no snack.
A second study demonstrated that eating a snack without stopping for a rest break may provide many of the beneficial effects of stopping and eating a snack. It should be noted, however, that eating large meals (rather than snacks) can lead to a reduction in performance - the well-documented post-lunch dip (Christie and McBrearty, 1979; Colquhoun, 1982).
In summary, the studies of rest breaks suggest that they are most beneficial when taken before the driver is very fatigued and should contain food. Food alone (without a rest break) appears to have some beneficial effects. There is some evidence that a rest break does not lead to an improvement in performance, but rather a reduction in the rate of deterioration of performance.
Driver fatigue monitoring
A system for detection of driver fatigue comprises a measure of driver fatigue, a standard against which the value of the measure is compared and a mechanism of communicating a finding that performance is degraded (Haworth, 1992). Systems have taken two forms: performance tests (administered before work or at the roadside) and invehicle systems.
Automotive manufacturers have considered the possibility of invehicle systems for detection of driver fatigue since the late 1960s (see Bishop, Madnick, Walter and Sussman, 1985 for a review of early systems).
The major issues relating to the feasibility of invehicle systems appear to be the need for high detection and low false alarm rates and whether this can be achieved at a realistic cost (Haworth, 1996). The proposal of a two-step detection procedure, unobtrusive measurement of vehicle control then a verbal secondary task (first suggested by Hardee, Dingus and Wierwille, 1985, cited in Haworth, 1996) may be simpler and cheaper than collecting eye closure or other driver variables in addition to steering wheel angle.
In overseas studies pavement treatments have been shown to be effective in alerting the dozing driver.
The best known are rumble strips which are grooves or raised ridges in the asphalt of the shoulder. Tyres running on the rumble strip transmit an audible and vibratory signal to the driver. Rumble strips have been applied to some highways which had a bad record for single-vehicle run-off-road accidents. A treatment on an interstate route running across the Mojave Desert resulted in a 49 percent drop in the number of such accidents (TR News, 1988).
The use of the classic rumble strip is limited to highways with sealed shoulders. A number of adaptations have been developed which can be applied to roads with gravel shoulders.
Ridged edgelining has been extensively installed in Victoria. A series of thermoplastic ridges are applied to the road by a special applicator. The markings are highly reflective and because they are 3 mm thick, the lines are easily visible above road water on rainy days. It is claimed that the thermoplastic edgelining should last six to eight times longer than painted edge lines.
Ridged edgelining is widely used in European countries, where its use is restricted to freeways because it is considered too noisy for use in towns.
The vast distances to be covered in Australia have made this country a fertile ground for fatigue and fatigue research. The dangers of long hours of driving, often under monotonous and uncomfortable conditions, have long been recognised. The more recent knowledge about other factors leading to fatigue - even in short distance driving - mean that the view of fatigue as a consequence of driving too long is no longer adequate and we cannot afford to limit our thinking to strategies to limit hours of driving by private or professional drivers.
Armour, M., Carter, R., Cinquegrana, C. and Griffith, J. (1988). Study of single vehicle rural accidents. Interim report (Report No. RN/88/1). Melbourne: Road Traffic Authority.
Bishop, H., Madnick, B., Walter, R. and Sussman, E.D. (1985). Potential for driver attention monitoring system development (DOT-HS-806-744, DOT-TSC-NHTSA-85-1). Washington, D.C.: National Highway Traffic Safety Administration.
Brown, I. D., Tickner, A. H. and Simmonds, D. C. V. (1970). Effect of prolonged driving on overtaking criteria. Ergonomics, 13, 239-242.
Cameron, C. (1973). A theory of fatigue. Ergonomics, 16, 638-648.
Cameron, M. H., Haworth, N., Oxley, J., Newstead, S. and Le, T. (1993). Evaluation of Transport Accident Commission road safety television advertising (Report No. 52). Melbourne: Monash University Accident Research Centre.
Chapman, P. (1985). Rural road accident study: Literature review (Working paper 2/85 Part 3). Adelaide: Road Safety Division.
Christie, M. J. and McBrearty, E. M. T. (1979). Psychophysiological investigations of post lunch state in male and female subjects. Ergonomics, 22, 307-323.
Colquhoun, P. (1982). Biological rhythms and performance. In W. B. Webb (Ed.), Biological rhythms, sleep and performance (pp.59-86). Chichester: Wiley.
Fell, D. (1987). A new view of driver fatigue. Draft version. Rosebery, NSW: Traffic Authority of NSW.
Fell, D. (1995). The road to fatigue: circumstances leading to fatigue accidents. In L. Hartley (Ed.), Fatigue and driving: Driver impairment, driver fatigue and driving simulation. London: Taylor & Francis.
Fell, D. and Black, B. (1996). Driver fatigue in the city. In L. Hartley (Ed.), Proceedings of The Second International Conference on Fatigue and Transportation: engineering, enforcement and education solutions. Fremantle February 11-16 1996. Canning Bridge: Promaco Conventions. (pp.165-187).
Feyer, A.-M. and Williamson, A. M. (1995). Managing driver fatigue in the long-distance road transport industry: interim report of a national research programme. In L. Hartley (Ed.), Fatigue and driving. Driver impairment, driver fatigue and driving simulation. London: Taylor and Francis. (pp.25-32)
Grandjean, E. (1968). The Societys Lecture, 1968. Ergonomics, 11, 427-436.
Hartley, L.R., Arnold, P.K., Penna, F., Hochstadt, Corry, A. and Feyer, A.M. (undated). Fatigue in the Western Australian transport industry. Part One. The principle and comparative findings (Report No.117). Institute for Research in Safety & Transport, Murdoch University, Western Australia.
Hattori, H., Matsuura, Y., Narumiya, K., Araki, K. and Ohnaka, H. (1987). Effect of vibration stimulus in lowering alertness levels of drivers. Proceedings of the Fourth International Pacific Conference of Automotive Engineers (pp.249.1-249.10). Society of Automotive Engineers.
Haworth, N. L. (1990). Current countermeasures and areas for countermeasure development. In M. Henderson (Ed.) Workshop on driver fatigue: Report of proceedings (CR2/90). Road Safety Bureau, Roads and Traffic Authority of NSW.
Haworth, N.L. (1992). Systems for detection of driver fatigue. In Proceedings 16th ARRB Conference, Part 4. Melbourne: Australian Road Research Board. (pp.189-201).
Haworth, N. (1995). The role of fatigue research in setting driving hours regulations. In L. Hartley (Ed.), Fatigue and driving. Driver impairment, driver fatigue and driving simulation. London: Taylor and Francis. (pp.41-47)
Haworth, N. (1996). Factors affecting the success of educational programs to reduce driver fatigue. In L. Hartley (Ed.), Proceedings of The Second International Conference on Fatigue and Transportation: engineering, enforcement and education solutions. Fremantle February 11-16 1996. Canning Bridge: Promaco Conventions. (pp.561-572).
Haworth, N. (1998). Does regulating driving hours improve safety? Paper presented to The Third International Conference on Fatigue and Transportation: Coping with the 24 Hour Society. Fremantle 9-13 1998.
Haworth, N. L., Heffernan, C. J. and Horne, E. J. (1989). Fatigue in truck accidents (Report No. 3). Melbourne: Monash University Accident Research Centre.
Haworth, N. L. and Rechnitzer, G. (1993). Description of fatal crashes involving various causal variables (CR119). Canberra: Federal Office of Road Safety.
Haworth, N. L., Triggs, T. J. and Grey, E. M. Driver fatigue: Concepts, measurement and crash countermeasures (CR72). Canberra: Federal Office of Road Safety.
Haworth, N. L., Vulcan, P., Schulze, M. T. and Foddy, B. (1991). Truck driver behaviour and perceptions study (Report No.18). Melbourne: Monash University Accident Research Centre.
Jones, I. S. and Stein, H. S. (1987). Effect of driver hours of service on tractor-trailer crash involvement. Washington, D.C.: Insurance Institute for Highway Safety.
Leggett, L. M. W. (1988). Truck accidents, fatigue and driving hours in Tasmania (Research Report 63). Hobart: Transport Tasmania.
Lisper, H-O. and Eriksson, B. (1980). Effects of length of a rest break and food intake on subsidiary reaction-time performance in an 8-hour driving task. Journal of Applied Psychology, 65, 117-122.
Lisper, H.-O., Laurell, H. and van Loon, J. (1986). Relation between time to falling asleep behind the wheel on a closed track and changes in subsidiary reaction time during prolonged driving on a closed track. Ergonomics, 29, 445-453.
Näätänen, R. and Summala, H. (1978). Fatigue in driving and accidents. In Driver fatigue in road traffic accidents (pp.26-32) (EUR 6065 EN). Luxembourg: Commission of the European Communities.
National Road Transport Commission. (1996). Fatigue Management Training Course. Draft prepared by SkillsLink.
Ryan, G. A., Wright, J. N., Hinrichs, R. W. and McLean, A. J. (1988). An in-depth study of rural road crashes in South Australia (Report Nos. FORS CR78, RSD 13/88). Adelaide: NH&MRC Road Accident Research Unit, University of Adelaide.
Transportation Research and Marketing. (1985). A report on the determination and evaluation of the role of fatigue in heavy truck accidents. Prepared for the AAA Foundation for Traffic Safety in cooperation with the Commercial Vehicle Safety Alliance.
Wertheim, A. H. (1978). Explaining highway hypnosis: Experimental evidence for the role of eye movements. Accident Analysis and Prevention, 10, 111-129.
Williams, G. W. (1963). Highway hypnosis: an hypothesis. International Journal of Clinical and Experimental Hypnosis, 103, 143-151.
Williamson, A.M., Feyer, A.M., Coumarelos, C. and Jenkins, T.. (1992). Strategies to combat fatigue in the long distance road transport industry - The industry perspectives (Report No.CR108). Canberra: Federal Office of Road Safety.
Wylie, C.D., Shultz, T., Miller, J.C., Mitler, M.M. and Mackie, R.R. (1996). Commercial Motor Vehicle Driver Fatigue and Alertness Study: Project Report (FHWA-MC-97-002). Washington, D.C.: Federal Highway Administration.
Yabuta, K., Iizuka, H., Yanagishima, T., Kataoka, Y. and Seno, T. (1985). The development of drowsiness warning devices. Paper presented to the 10th International Technical Conference on Experimental Safety Vehicles, Oxford, England.