Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-7248 (Print)
ISSN : 2234-6473 (Online)
Industrial Engineering & Management Systems Vol.20 No.1 pp.27-34

Effects of Sleep Deprivation on Vigilance, Attention, and Performance During Simulated Train Driving

Clara Theresia, Hardianto Iridiastadi*, Gradiyan Budi Pratama
Department of Industrial Engineering, Faculty of Industrial Technology, Universitas Katolik Parahyangan, Bandung, Indonesia
Departement of Industrial Engineering, Faculty of Industrial Technology, Institut Teknologi Bandung, Bandung, Indonesia
*Corresponding Author, E-mail:
March 10, 2020 July 24, 2020 January 22, 2021


Sleep deprivation has been cited as a major factor that plays an important role in many incidents in the transportation sector. Sleep-deprived train drivers is a fairly common phenomenon in Indonesia, with local reports indicating a good percentage of train drivers who are sleep deprived prior to work. The present study was aimed at quantifying the effects of sleep deprivation on alertness and performance during prolonged simulated train driving. A total of 12 subjects participated in this study and were asked to sleep for approximately 2 h (sleep deprived) and 8 h (normal sleep) the night before the experimental day. The experiment consisted of driving a train simulator for 4 h on a monotonous route. Fatigue and sleepiness were assessed using Psychomotor Vigilance Task (PVT) and Sustained Attention Test (SAT), conducted before and after the driving simulation. Subjective levels of fatigue and sleepiness were determined using questionnaires, while driving performance was measured based on the number of speed-limit violations. Results of this study showed that two hours of sleep was characterized with an initial subjective fatigue and sleepiness measures that were up to two to three times greater than the normal sleep condition. This condition also resulted in poorer driving performance (75% increase in the number of speeding error). While the effects of sleep deprivations on the performance of train driving is probably obvious, the quantitative effects have not been addressed extensively in the literature. It is concluded in this study that the effects of excessive sleep deprivation on fatigue and sleepiness varies, depending on the measures used.



    While there has been only a limited number of major accidents recently, railway safety is still a major issue in Indonesia. From 2010 to 2016, there were a total of 35 major accidents, of which 26% were train collisions and 68% were derailments and/or falling off tracks (KNKT, 2016). At least 55 people died and more than 240 passengers were injured from these accidents. The Indonesian Ministry of Transportation (2012) noted that at least 51% of railway accidents were associated with the humans (particularly the train drivers), while the rests were associated with infrastructures and procedural aspects.

    Previous studies have also indicated the role of human factors in railway accidents. Kim’s study for example, reported as many as 80 train accidents that occurred in the UK, with human failures play role in more than 53% of the accidents (Kim and Yoon, 2013). The remaining 47% were related to technical factors or other factors such as damage of the railway tracks. In Australia, at least 65% of railroad accidents were associated with human error (Baysari et al., 2008). Fatigue, in many transport studies, has been cited as a risk factor that has to be managed very carefully. Fatigued drivers are characterized by substantial reduction in alertness and performance (Dorrian et al., 2006). Fatigue can also result in longer response times and increased errors (Lim and Dinges, 2008).

    Note, however, that fatigue is a phenomenon that can be easily experienced by any individual, but may be somewhat difficult to assess. Several work-related factors can affect fatigue during prolonged driving. These include the monotony that often characterizes train driving tasks, which can decrease the level of alertness and the ability to react to the surrounding conditions (Straussberger, 2004). Another important factor is the fact that train drivers are often given the same routes over and over, that response to sudden events or unusual signals may require a high level of alertness and attention (Dunn and Williamson, 2012). Shift work (particularly night shift) can also affect driver’s fatigue (Anderson et al., 2013).

    Having a good amount of sleep duration is a prerequisite for an optimal performance at work. A decrease in sleep quantity will likely cause drowsiness, a condition that has been shown to be highly related to accidents in the transportation sector (Philip and Akerstedt, 2006). Previous investigations reported that several hours of sleep for a few days in a row can lead to a decrease in cardiovascular and cognitive function, as well as poorer immune system (Banks and Dinges, 2007;Dinges et al., 2005). Van Dongen’s research, through a series of Psychomotor Vigilance Test (PVT) tests, found that the decrease in performance ocurred when the individual experienced a short sleep quantity (6 h/night) for 2 consecutive weeks (Van Dongen et al., 2003). Experimental studies have also shown that short quantity of sleep will adversely affect a number of basic cognitive processes including attention, alertness, and vigilance as well as advanced cognitive processes such as decision making and problem solving (Killgore, 2010). A report by Roach et al. (2003) suggested that a good amount of sleep was closely related to performance and level of alertness. Furthermore, a work conducted by Schleicher and colleagues (2008) also noted a relationship between sleep deprivation and levels of individual alertness. Note, however, that the majority of studies have only addressed partial sleep deprivation (e.g., Åkerstedt et al., 2010;Abe et al., 2011), despite the fact that several hours of sleep prior to train driving is indeed a phenomenon that can be found (though infrequent) in some developing nations.

    In Indonesia, several local reports have suggested sleep deprivation as a common phenomenon among train drivers, but this issue has never been formally investigated. The present study was aimed at comparing and quantifying the effects of sleep deprivation (2 h versus 8 h of sleep) on alertness and performance in simulated train driving. While two hours of sleep could be generally rare, it could occur to some individuals due to irregular (actual) driving roster, religious needs, house-chore responsibilities, and the need for side jobs.

    2. METHODS

    2.1 Experimental Design

    This study employed a full-factorial, within-subject design utilizing train simulator at the Fatigue Laboratory, Institut Teknologi Bandung, Indonesia. Two different sleep durations, 2 hours of sleep (sleep deprived/SD) and 8 hours of sleep (“normal” sleep duration/NS), were investigated in this study. Several baseline data were also collected prior to the actual experiment, in which the participant was required to have a minimum of 15 hours of sleep in the previous two nights.

    2.2 Participants

    Participants in this study were not train drivers, but they were recruited from a pool of potential participants who had very similar physical characteristics and level of knowledge of real Indonesian train drivers. Dunn and Williamson (2012) suggested that there was no difference in the performance of machinists and non-machinists in train driving activities. Participants in this study consisted of 12 healthy men with an average age of 21 years (SE = 0.953), with a minimum education of college degree in engineering. All participants had no history of chronic diseases, and had the ability to operate a PC/laptop. Additionally, smoking and caffeine consumption were not allowed during the experiments.

    2.3 Instruments

    The train simulator employed a software called Railworks Trains Simulator (PI Engineering Software, US), version 2015 by Dovetail Games. The simulation set up consisted of a 42'' LED TV monitor display (LG Electronics, Indonesia) and RailDriver Desktop Cab Controller RD-91-MDT-R (PI Engineering, US) as a train controller (Figure 1). The system was similar to that used by Dunn and Williamson’s research (2012).

    2.3.1 Simulator Scenario

    In this study, each participant was required to drive a fairly straight and monotonous route for four hours continuously. This scenario represented long-distance (between major cities) routes commonly found in Indonesia, with task duration that was similar to the duty duration assigned to Indonesian train drivers. For this particular scenario, the maximum and minimum speed limit were set at 75 km/h and 15 km/h, respectively.

    2.3.2 Subjective Measures

    A 9-point scale Karolinska Sleepiness Scale (KSS) was used to measure the level of sleepiness prior to and immediately after the experiment (Akerstedt and Gillberg, 1990). In addition, KSS score was reported every 20 minutes throughout the duration of the driving task. Another assessment technique employed in this study was the Visual Analogue Scale (VAS), which was used here to represent participant’s overall feeling of fatigue (Dorrian et al., 2006;Petrilli et al., 2005). Similar to the use of KSS, VAS was administered every 20 minutes during the experiment. The Swedish Occupational Fatigue Index (SOFI) was also used to assess fatigue level before and after the experiment. It consisted of 25 measures, which were classified into 5 dimensions: lack of energy, physical exertion, physical discomfort, sleepiness, and lack of motivation (Ahsberg et al., 2000).

    2.3.3 Objective Measures

    Changes in both Psychomotor Vigilance Task (PVT) and Sustained Attention Task (SAT) were used as an objective method in assessing fatigue (Dorrian et al 2006, Dorrian et al., 2007;Lamond et al., 2005). The 5-min PVT was used instead of the 10-min PVT, since it has been suggested that there is no differences in performance when both are used (Loh et al., 2004;Roach et al., 2006). The PVT parameters included the mean (pvt-m), number of lapse (pvt-#l), 10% fastest (pvt-f), and 10% slowest (pvt-s), while the SAT parameters included the number of misses (sat-nom) and response time greater than 850 ms (rt-850). Both indicators were administered by using a laptop (before and after a driving task), in which the participants were required to press a key immediately after a visual stimulus had been presented on the screen.

    2.3.4 Sleep Duration

    Sleep quantity was monitored using a smart watch (Fitbit, US). The instrument was used at night before the experiment took place. To obtain 2 hours of sleep, the participants slept between 03:00 until 05:00 (in preceding night before the experiment) in a dedicated sleep room, while those with 8 hours of sleep were asked to be in bed at around 21:00. All participants woke up early at 05:00 due to religious needs and were not allowed to fall asleep at all prior to the beginning of the experiment (started at 08:00). Mean sleep durations for the sleep deprived (SD) and normal sleep (NS) were 2.3 h (SE = 0.30) and 7.9 h (SE = 0.58), respectively. A member of the research team supervised this entire procedure to ensure that the actual duration of sleep was indeed within the established criteria.

    2.4 Statistical Analysis

    Fatigue was determined as changes in all subjective and objective measures as a function of sleep quantity and task duration. Statistical tests consisted of descriptive statistical test (mean and standard deviation), normality test of data with Kolmogorov Smirnov test, Chi-Square test and Analysis of Variance (ANOVA) with p = 05 used as criterion of significance. ANOVA within subject analysis was conducted to test whether the effects of sleep duration on the fatigue measures were significant (Walpole et al., 2012).

    3. RESULTS

    3.1 Subjective Indicator

    KSS scores showed an incremental linear pattern for both SD and the NS conditions (Figure 2). The KSS scores for the SD condition, however, were significantly greater than the scores of the NS condition (a difference of about 94% and 41% for the initial and end scores, respectively). Note that, for both conditions, sleepiness seemed to change in a similar fashion, indicating an increase in sleepiness as a function of driving time.

    Similar to the KSS scores, fatigue as indicated by VAS scores demonstrated an increasing trend for both conditions (Figure 3). Both the initial and the end scores for the SD condition were much greater than the NS condition. Unlike KSS, however, VAS scores for the SD condition increased steeper than the NS condition. Comparisons between the two conditions demonstrated a difference of 108% for the initial VAS score and 52% for the end VAS score.

    Fatigue, as assessed by SOFI scores, was also more pronounced for the SD condition (see Figure 4). Across the four SOFI dimensions, fatigue was associated mostly with lack of energy, followed with lack of motivation and physical exertion. The initial SOFI scores for the SD condition were roughly 1.6 to 2.0 times greater than those of the NS condition. These gaps were narrower when comparing the SOFI scores obtained at the end of the experiment.

    3.2 Objective Alertness

    Table 1 shows PVT and SAT parameters obtained both at the beginning and at the cessation of the simula-tion. Using the Kolmogorov Smirnov tests, all PVT and SAT data were normally distributed (for both the SD and NS conditions).

    Repeated measures ANOVA indicated that the SD condition had a significant effect on PVT on all PVT parameters: pvt-m [F (2, 22) = 9.315, p= 0.001], pvt-f [F (2, 22) = 6.047, p = 0.008], pvt-#l [F (2, 22) = 14.413, p = 0.001] and pvt-s [F (2, 22) = 5.39, p = 0.012]. For the SAT parameters, significant effect was found for the sat-nom parameter [F (2, 22) = 13,065, p = 0.001], but not for the rt-850 parameter [F (2, 22) = 3,365, p = 0.053] which means there was no significant difference between rt-850 parameter test results on the participant's attention level during the experiment.

    3.3 Driving Performance

    Driving performance was assessed by calculating the number of speeding error (or driving speed greater than the speed limit) throughout the duration of the simulation. On average, the sleep deprived participants made about 32 speeding errors, which was four times greater than the number of errors during NS condition. The difference was significant based on Chi-square test (X2= 7.638 > X2α= 3.84 df =1 with p<0.05), indicating the detrimental effect of excessive sleep deprivation.

    3.4 Correlation Analysis

    Correlation analysis was performed across all sub-jective, objective, and performance parameters using Pearson Product Moment correlation test. In the SD condition, the results show that several indicators correlated significantly, including the sat-nom with KSS (r = 0.679; p = 05) and sat-nom with pvt-m (r = 0.643; p = 05). The subjective indicators of SOFI and VAS showed significant correlations with the value of r = 0.705. The number of speeding errors was only marginally correlated with KSS score (r = 0.425).


    This study aimed to quantify the effects of sleep deprivation on individual alertness, attention, and per-formance during 4 hours of simulated train driving. While sleep deprivation certainly affects driving per-formance, the quantitative effects have been seldom addressed. Major finding resulting from this research was that acute lack of sleep, in fact, had severe detri-mental effects on fatigue and performance. The exact effects, however, varied depending on the measures used. Based on the subjective measures, the SD condi-tion was characterized by 50 – 200% greater levels of fatigue. Compared with the NS condition, the SD con-dition also resulted in 5 to 50% poorer response time (as measured by PVT). It should be noted that sleep deprived individuals demonstrated substantially (up to four times) worse driving performance compared to those having sufficient amount of sleep. Such quantita-tive differences have rarely been reported in the litera-ture.

    This study clearly showed that a decrease in cogni-tive functions, slowed response time, and poorer driving performance were found in sleep-deprived individuals. For this condition, other reports also showed similar phe-nomena, including poorer response times and increase errors (Lim and Dinges, 2008). Research has generally showed that awareness and attention are the basic capacities of cognitive aspect and are strongly influenced by lack of sleep (Killgore, 2010). The present study demonstrated that fatigue and sleepiness (due to lack of sleep) were associated with changes in all PVT parameters. This finding is consistent with results from Lamond’s study (2005), but their study further noted the number of lapse as a parameter that was sensitive in detecting cognitive impairment among train drivers in Australia.

    Unlike the airline industry that highly regulates and controls the amount of rests that pilots receive prior to their assigned duties, the railway sector does not generally apply the same rigorous tests for the assessments of fitness for duty. The PVT is used for assessing level of fatigue among Australian train drivers (Lamond et al., 2005), but the method is not used as part of a standard assessment procedure in Indonesia. Train drivers in Indonesia only generally receive a limited number of tests prior to their duties, including the measurements of blood pressure, heart rate, and body temperature. The use of PVT (or other similar methods) is suggested here as a (practical, yet valid) means to assess driver’s ability to maintain continuous attention throughout the driving tasks.

    One of the interests in this study was to determine if other measures (such as the SAT) could be used to indicate fatigue and sleepiness. While PVT is usually administered by asking subjects to respond as quickly as they can to a visual (or auditory) stimulus, SAT parameters are generally obtained based on subject’s short-term memory. In a study conducted by De Valck et al (2015), both SAT parameters were useful in classifying fit vs. unfit train machinists. In our study, however, only one SAT parameter (i.e., the number of miss) that was sensitive to extreme differences in sleep duration. Research has shown that PVT is generally sensitive to sleep deprivation, but how the condition affects memory tests (such as SAT) has received less attention.

    Studies examining performance of train drivers typically indicates the adverse effects of fatigue on driving performance. Reports by Dorrian and colleagues (2007), for example, clearly demonstrated poor train machinist performance (e.g., excessive speed, rough braking or acceleration, and train fuel consumption) as manifestations of fatigue. Such phenomena are fairly consistent with findings from the present study. Note, however, that the results of the present study showed performance decrement that was up to four times poorer due to excessive acute sleep deprivation.

    Findings from research examining performance of car drivers also indicate similar phenomena. Individuals with lack of sleep will exhibit poor driving performance. A previous study, with 19 professional drivers and 30-min of driving, has shown an association between lack of sleep and 20% and 30% increase in lane lateral position and speed violation, respectively (Jackson et al., 2013). Another study (Killgore, 2010) found positive correlation between response times (assessed by PVT parameters) and lane lateral position and meeting Australian’s speed limit of 80 km/h.

    The present study also sought to evaluate how in-sufficient amount of sleep affected subjective feelings of sleepiness. This study found two interesting phenomena – first, sleepiness tended to increase linearly as a function of task duration, with overall patterns that were somewhat similar (for both conditions) and, second, sleepiness was rated twice as much in the SD condition throughout the driving task, particularly in the first hour of the experiment. This is in line with a report by Schleicher and colleagues (2008) that indicated a decrease in alert level that was proportional to the in-creased sleepiness in the 2 hours of driving activity. Based on Tempesta et al. (2010), the lack of (even one night’s) sleep will decrease the subjective rating score of one’s emotions and mood when compared to normal 8 hours of sleep. It should be noted that 4 hours of driving was considered acceptable (for the NS condition), while individuals from the SD condition reported moderate levels of sleepiness at the end of the experiment. The latter phenomenon was rather surprising, considering that it was expected that 2 hours of sleep should result in fairly severe level of sleepiness.

    Subjective indicators for assessing fatigue in the present investigation were the visual analogue scale (VAS) and the Swedish Occupational Fatigue Inventory (SOFI). Similar to the sleepiness data, fatigue tended to increase linearly throughout the duration of the task. At the beginning of the driving task, excessive lack of sleep resulted in fatigue levels that were two times greater than those reported by subjects having adequate amount of sleep. This discrepancy tended to decrease toward the end of the driving task (a difference of roughly 50%). With respect to SOFI data, lack of sleep was associated with feelings of fatigue that were up to two times as much as those reported from the normal sleep condition. For train driving activity, ‘lack of energy’ was the feeling that ranked first, followed by ‘lack of motivation’ and ‘lack of physical exertion’. This finding was similar to that reported by Åhsberg et al. (2000) despite differences in task context (working at night).

    Another key issue typically discussed in the litera-ture is correlations among various fatigue measures. This is due to the fact that fatigue is not a simple construct, and an effort to understand manifestations of fatigue will require the use of several approaches simultaneously. In this study, very strong correlations were generally found between subjective measures of fatigue. Strong positive correlation was also found between SAT parameter (e.g., number of miss) and PVT parameter (mean response times). Fatigue as assessed by the SAT parameter was also highly correlated with the subjective measures. Driving performance, in contrast, only moderately correlated with sleepiness. This is somewhat in line with the work by Kosmadopoulos and colleagues (2017) that showed correlation between subjective and performance indicator of fatigue during car driving activities. Such finding clearly shows that fatigue and sleepiness cannot be evaluated solely by employing a limited number of indicators. The use of one particular subjective indicator over the others, however, could be sufficient.

    Several limitations of the study are worth noting. First, the participants of the study were not actual train drivers. In reality, these train drivers might have been very familiar with (and were not that susceptible to) the monotony of the task. The number of experimental subjects was also relatively small, resulting in moderate level of statistical power. Generalization of the finding should, therefore, be done cautiously. While a couple of hours of sleep is probably a rare phenomenon, lack of sleep is indeed very common among Indonesian train drivers. Such condition will likely result in substantial degraded performance, potentially becoming a safety risk. Controls to mitigate the risk should be comprehensively sought, by simultaneously considering company’s operational requirements and the needs to have sufficient amount of rests among the train drivers. The second limitation was that this study utilized a train simulator that lacked several environmental stimuli that, if present, could (to some extent) improve cognitive functioning of drowsy drivers. Lastly, real train drivers are actually allowed to smoke and drink coffee prior to driving. These are factors that are known to improve alertness, which could modify the detrimental effects of acute sleep deprivation.


    This study was aimed at quantifying the effects of severe sleep deprivation on fatigue during simulated train driving. It was concluded that acute, excessive lack of sleep clearly affected fatigue, sleepiness, and performance. Compared with those having sufficient amount of sleep the night before, individuals with lack of sleep exhibited much poorer attention and response, greater level of fatigue and sleepiness, and extremely degraded driving performance. Two major implications of this study can be argued here. First, findings of this study could roughly be used to estimate the amount of degradation in fatigue, attention, and performance as a function of sleep duration the night before a driving task. Second, study findings could be used to convince railway companies that even marginal levels of fatigue and sleepiness can actually imply much poorer driving performance, a phenomenon that should be carefully considered during fitness for duty assessments. It should be noted, however, that this study employed a train driving simulator, and generalization of the findings in the real driving conditions should be done with great care.

    Clara Theresia is a lecturer in the Department of Industrial Engineering, Faculty of Industrial Technology at Universitas Katolik Parahyangan, Indonesia. Her research interests include fatigue in the transportation sector, ergonomics, and mental workload analysis.

    Hardianto Iridiastadi is currently an Associate Professor within the Faculty of Industrial Technology at Institute Teknologi Bandung, Indonesia. His research interests include occupational ergonomics, design of patient handling assistive device, and fatigue/workload in the transportation sector.

    Gradiyan Budi Pratama is a lecturer in the Industrial Engineering Department at ITB. His research interests include occupational ergonomics, fatigue in the transportation sector, and human factors engineering.


    The authors wish to express their gratitude to Direc-torat of Higher Education (DIKTI) for supporting funding in this research.



    Railway simulator used in the study.


    Karolinska sleepiness scale score.


    Visual analog scale score.


    Swedish occupational fatigue index (SOFI) score.


    Number of speeding error.


    Mean and standard error of PVT and SAT


    1. Abe, T., Nonomura, T., Komada, Y. Asaoka, S., Sasai, T., Ueno, A., and Inoue, Y. (2011), Detecting deteriorated vigilance using percentage of eyelid closure time during behavioural maintenance of wakefulness tests, International Journal of Psychophysiology, 82(3), 269-274,
    2. Ahsberg, E, Kecklund, G., Akerstedt, T., AND Gamberale, F. (2000), Shiftwork and different dimensions of fatigue, International Journal of Industrial Ergonomics, 26(4), 457-465,
    3. Akerstedt, T. and Gillberg, M. (1990), Subjective and objective sleepiness in the active individual, International Journal of Neuroscience, 52(1-2), 29-37,
    4. Akerstedt, T., Ingre, M., Kecklund, G., Anund, A., Sandberg, D., Wahde, M., Phillip, P., and Kronberg, P. (2010), Reaction of sleepiness indicators to partial sleep deprivation, time of day and time on task in a driving simulator – the DROWSI project, Journal of Sleep Research, 19(2), 298-309,
    5. Anderson, C., Grunstein, R. R., and Rajatnam, S. (2013), Hours of work and rest in the rail industry, International Medicine, 43(6), 717-721,
    6. Banks, S. and Dinges, D. F. (2007), Behavioral and physiological consequences of sleep restriction, Journal of Clinical Sleep Medicine,3(5), 519-528,
    7. Baysari, M. T., McIntosh, A. S., and Wilson, J. R. (2008), Understanding the human factors contribution to railways accidents and incidents in Australia, Accident Analysis and Prevention, 40(5), 1750-1757,
    8. De Valck, E., Smeekens, L., and Vantrappen, L. (2015), Periodic psychological examination of train drivers' fitness in Belgium: Deficits observed and efficacy of the screening procedure, Journal of Occupational and Environmental Medicine, 57(4), 445-452,
    9. Dinges, D. F., Roggers, N. L., and Baynard, M. D. (2005), Chronic Sleep Deprivation. Principles and Practice of Sleep Medicine (4th ed), WB Saunders.
    10. Dorrian, J., Hussey, F., and Dawson, D. (2007), Train driving efficiency and safety: Examining the cost of fatigue, Journal of Sleep Research, 16(1), 1-11,
    11. Dorrian, J., Roach, G. D., Fletcher, A., and Dawson, D. (2006), The effect of fatigue on train handling during speed restrictions, Transportation Research Part F: Traffic Psychology and Behaviour, 9(4), 243-257,
    12. Dunn, N. and Williamson, A. (2012), Driving monotonous routes in train simulator: The effect of task demand on driving performance and subjective experience, Ergonomics, 55(9), 997-1008,
    13. Jackson, M. L., Croft, R. J., Kennedy, G. A., Owens, K., and Howard, M. E. (2013), Cognitive components of simulated driving performance: Sleep loss effects and predictors, Accident Analysis and Prevention, 50, 483-444,
    14. Killgore, W. D. S. (2010), Effect of sleep deprivation on cognition, Progress in Brain Research Brain Research, 185, 105-129,
    15. Kim, D. S. and Yoon, W. C. (2013), An accident causation model for railway industry: Application of the model to 80 rail accident investigation reports from the UK, Safety Science, 60, 57-68,
    16. KNKT-Database about train accident by National Transportation Safety Committee (2016), cited 2016 October 31,
    17. Kosmadopoulos, A., Sargent, C., Zhou, X., Darwent, D., Matthews, R. W., Dawson, D., and Roach, G. D. (2017), The efficacy of objective and subjective predictors of driving performance during sleep restriction and circadian misalignment, Accident Analysis and Prevention, 99, 445-451,
    18. Lamond, N., Darwent, D., and Dawson, D. (2005), Train drivers’ sleep and alertness during short relay operations, Applied Ergonomics, 36(3), 313-318,
    19. Lim, J. and Dinges, D. F. (2008), Sleep deprivation and vigilant attention, Analysis of the New York Academy of Sciences, 1129(1), 305-322,
    20. Loh, S., Lamond, N., Dorrian, J., Roach, G., and Dawson, D. (2004), The validity of psychomotor vigilance task of less than 10 minutes duration. Behaviour, Research Methods, Instrument & Computer, 36(2) 339-346,
    21. Petrilli, R. M., Jay, S. M., Dawson, D., and Lamond, N. (2005), The impact of sustained wakefulness and time of day on OSPAT performance. The centre for sleep research the queen elizabeth hospital: Woodville Australia, Industrial Health, 43(1), 186-192,
    22. Philip, P. and Akerstedt, T. (2006), Transport and industrial safety, how are they affected by sleepiness and sleep restriction?, Sleep Medicine Review, 10(5), 347-356,
    23. Roach, G. D., Dawson, D., and Lamond, N. (2006), Can a shorter PVT be used as a reasonable substitute for the 10-minute PVT? Centre for sleep research University of South Australia: Adelaide, Chronobiology International, 23(6), 1379-1387,
    24. Roach, G. D., Reid, K. J., and Dawson, D. (2003), The amount of sleep obtained by locomotive engineers: Effects of break duration and time of break onset, Occupational Environment Med, 60(12), 17-22,
    25. Schleicher, R., Galley, N., Briest, S., and Galley, L. (2008), Blink and saccades as indicator of fatigue in sleepiness warning: Looking tired?, Ergonomics, 51(7), 982-1010,
    26. Straussberger, S. (2004), A psychological investigation of the concept of monotony in ATC: Effect of traffic repetitiveness and traffic density, In: International conference on research in air transportation (ICRAT), 199-208, Available from:
    27. Tempesta, D., Couyoumdijian, A., Curcio, G., Moroni, F., Marzano, C., De Gennaro, L., and Ferrara, M. (2010), Lack of sleep affect the evaluation of emotional stimuli, Brain Research Bulletin, 82(1-2), 104-108,
    28. Van Dongen, H. P., Maislin, G., Mullington, J. M., and Dinges, D. F. (2003), The cumulative cost of additional wakefulness: Dose response effect on Neurobehavioral functions from chronic sleep restriction and total sleep deprivation, Sleep, 26(2), 117-126,
    29. Walpole, R. E., Myres, R. H., Myres, S. L., and Ye, K. (2012), Probability and Statistics for Engineers and Scientists: Correlation testing (9th ed), Prentice Hall, Boston, MA.
    Do not open for a day Close