Sleepiness and Driving - QUT · Driver sleepiness contributes to a substantial proportion of fatal...

Sleep and wake drives i

The Sleep and Wake Drives: Exploring the Genetic and

Psychophysiological Aspects of Sleepiness, Motivation, and

Performance

Christopher N Watling

BBehavSc(Psych), PostGradDipPsych, MAppSc(Research)

A thesis submitted in fulfilment of the requirements for the Degree of Doctor of

Philosophy

Centre for Accident Research and Road Safety - Queensland

School of Psychology and Counselling

Faculty of Health

Queensland University of Technology

2016

Sleep and wake drives ii

Sleep and wake drives iii

Keywords

Sleepiness, sleep deprivation, performance, driving performance, sleep drive, wake

drive, four-process model, motivation, effort, driving cessation, genetics, single

nucleotide polymorphisms, circadian genes, physiological sleepiness, subjective

sleepiness, hazard perception, psychomotor vigilance.

Sleep and wake drives iv

Sleep and wake drives v

Abstract

Driver sleepiness contributes to a substantial proportion of fatal and severe

road crashes and potentially contributes to a similar, if not a greater, proportion of

less serious crashes. Current sleep-wake models rely on characterising the interaction

between homeostatic and circadian factors for predictions of sleepiness and level of

performance. These models’ predictions have been validated against laboratory and

field data, and are associated with incidents of sleep-related crashes. However, these

monotonic models do not account for the known effects of motivation and individual

differences on sleepiness. A better understanding of how motivation and individual

differences effects the ability to drive safely has important implications for road

safety. Driving performance is particularly susceptible to the effects of sleepiness;

however, motivation to continue driving while sleepy is acknowledged as an

important factor in sleepy driving behaviours. Assessing the factors of motivation

and individual differences for their effects on sleepiness and sleepy driving

behaviours could lead to a better understanding of their role in influencing sleepiness

and performance levels.

This research program examined the factors of motivation and individual

differences, assessed as genetic variations for their effects on sleepiness and

performance through three studies. The first study (N = 131) examined changes in

physiological, subjective, and performance indices of sleepiness while participants

performed a 15-min vigilance task where the last 5 min required them to motivate

themselves to perform better than they had previously. This study also examined 22

genetic variants (i.e., Single Nucleotide Polymorphisms [SNPs]), known to be

involved in circadian functioning, for their associations with the physiological,

subjective, and performance indices from the 15-min Psychomotor Vigilance Task.

Sleep and wake drives vi

Overall, the findings suggest that motivations to improve performance on the

vigilance task had a partial effect on physiological and subjective sleepiness, with a

larger effect on performance indices of sleepiness. Regarding the genotyping results,

overall five SNPs were found to be associated with differences in physiological,

subjective, and performance indices of sleepiness. The different alleles from SNP

rs5751876 (ADORA2A gene) were found to be associated with differences between

physiological and performance measures. Novel findings related to the SNPs,

rs2291738 (TIMELESS gene), rs2735611 (PERIOD1 gene), and rs4630333

(TIMELESS gene), which were limited to the motivated section of the vigilance task,

and the SNP rs1554338 (CRY2 gene) where the genotypes were found to differ in the

physiological and subjective sleepiness measures.

The second study (N = 26) examined the changes in physiological and

subjective sleepiness associated with cessation of a hazard perception task.

Participants were instructed to monitor their sleepiness and stop driving once they

perceived their level of sleepiness was too dangerous to drive safely on the road.

Specifically, this instruction removes the participants’ motivation to continue

performing the task; and thus, understanding the physiological and subjective

sleepiness changes in the absence of motivation to continue to drive is important.

The main finding suggests that after mild sleep restriction, participants’ subjective

sleepiness levels increased, and some of the physiological sleepiness increased over

the duration of the drive. In the absence of motivation to continue to drive while

sleepy, participants in this study were able to identify increasing subjective

sleepiness levels, were able to choose to cease driving before unintended sleep onset

or microsleeps occurred, and this was associated with an increase of blink duration

and greater awareness of the signs of sleepiness.

Sleep and wake drives vii

The third and final study (N = 18), experimentally manipulated the

participants’ motivational levels and manipulated the participants’ level of

sleepiness. Physiological, subjective, and performance indices of sleepiness were

obtained with respect to the effects of the manipulation of motivation and sleepiness

levels. The performance measures included the Hazard Perception Test, which is a

high-order cognitive task, and Psychomotor Vigilance Test, which is a low-order

cognitive task. Specifically, no effect of motivation on data relating to changes in

physiological or subjective sleepiness was observed in either the Hazard Perception

Test data or the Psychomotor Vigilance Test. Physiological and subjective sleepiness

were both greater in the sleepy condition than the alert condition, and over the

duration of both tests, sleepiness levels increased regardless of the motivation or

sleepiness condition. Motivation appeared to have no effect on Hazard Perception

Test performance; however, a three-way interaction was observed in the

Psychomotor Vigilance Test performance data. Reaction-time latencies were shorter

in the alert-motivation condition compared to the other conditions and only a small

increase in reaction-time latencies over the duration of the task was found in the

alert-motivated condition.

Overall, the three studies demonstrate that certain genetic variations can

influence aspects of physiological and subjective sleepiness as well as performance

outcomes. Additionally, motivation can have an effect on sleepiness and

performance, but this effect is certainly limited. Specifically, in the absence of any

form of sleep deprivation, the deterioration of performance on a low-order cognitive

task can be stopped when motivation to improve task performance is implemented.

However, when experiencing moderate sleep deprivation the added resources needed

to improve task performance are not available, and thus motivation has no effect on

Sleep and wake drives viii

improving low-order cognitive performance. Regardless of being alert or sleep

deprived, there is no improvement of performance with motivation when performing

a high-order cognitive task as was the case with the hazard perception task. These

results suggest the sleepiness is resistant to motivation to improve performance,

particularly with the high-order cognitive task of the Hazard Perception Test. As

such, continuing to drive while sleepy whereby the drivers motivate themselves to

apply extra effort to the task of driving are engaging in a risky driving behaviour.

Future research could seek to examine more SNPs with a larger sample, and

there is the potential that including a high-order cognitive task might also further our

understanding of the genetic associations with vulnerability to sleepiness and the

associated impairment of task performance. Future research could also examine the

effects of motivation on other indices of driving performance in high-fidelity driving

simulators, such as lane positioning/lateral positing, driving speed, steering

corrections. Continued research that seeks to extend our understanding of the effects

of motivation and individual differences regarding sleepiness and performance

outcomes is an important component of improving road safety. Such efforts are

important undertakings to reduce road trauma and provide a safer road environment

for all road users.

Sleep and wake drives ix

Table of Contents

Keywords ............................................................................................................... iii Abstract .................................................................................................................... v

Table of Contents ..................................................................................................... ix

List of Tables .......................................................................................................... xv

List of Figures ....................................................................................................... xvii Abbreviations and Terms ....................................................................................... xix

Definition: Sleepiness, Fatigue, and Alertness ........................................................ xxi Statement of Original Authorship .........................................................................xxiii Acknowledgements ............................................................................................... xxv

Publications from this Program of Research ........................................................ xxvii Chapter One: Introduction ......................................................................................... 1

1.1 Overview of Chapter .......................................................................................1 1.2 Why Study Sleepiness? ...................................................................................1

1.2.1 Levels of Societal Sleepiness .................................................................... 3 1.3 Why Study Driver Sleepiness? ........................................................................5 1.4 The Burden of Crashes ....................................................................................6 1.5 The Incidence of Sleep-Related Crashes ..........................................................7

1.5.1 Official Data ............................................................................................. 7 1.5.2 Self-Report Data ....................................................................................... 8 1.5.3 Case Controlled Data ................................................................................ 9

1.6 Level of Crash Risk and Risk Perceptions of Sleepy Driving ......................... 11 1.7 The Cortical Associations of Sleepiness and the Effects on Driving............... 13

1.7.1 The Task of Driving ............................................................................... 15 1.7.1.1 Hazard perception ............................................................................ 17 1.7.1.2 Psychological components of hazard perception .............................. 18 1.7.1.3 Hazard perception and sleepiness ..................................................... 20

1.7.2 Sleepiness and Younger Drivers ............................................................. 20 1.8 Chapter Summary.......................................................................................... 22

Chapter Two: Sleep-Wake Regulation and Performance ......................................... 23 2.1 Overview of Chapter ..................................................................................... 23 2.2 Mechanisms of Sleep-Wake .......................................................................... 23

2.2.1 Genetic Mechanisms............................................................................... 23 2.2.2 Neuroanatomical Mechanisms ................................................................ 25 2.2.3 The Physiological Transition from Wakefulness to Sleep: Sleepiness ..... 26 2.2.4 Regulation of the Sleep-Wake Cycle....................................................... 30

2.3 The Effects of Sleepiness .............................................................................. 33 2.3.1 Cognitive Performance and Task Load ................................................... 34

2.4 Sleepiness and Performance Outcomes .......................................................... 36 2.4.1 Predictions from Biomathematical Sleep-Wake Models .......................... 36

Sleep and wake drives x

2.4.1.1 Two-process model of sleep-wake regulation .................................. 36 2.4.1.2 Three-process model of sleep-wake regulation ................................ 37 2.4.1.3 Utility of biomathematical models ................................................... 39 2.4.1.4 Critical aspects of sleep-wake models .............................................. 40 2.4.1.5 Four-process model of sleep-wake regulation .................................. 42

2.4.1.5.1 Evidence for the wake drive ...................................................... 44 2.4.2 Individual Differences, Sleepiness and Performance............................... 49

2.4.2.1 Genetic aspects ................................................................................ 49 2.5 Aims, Rationale of the Studies, and Significance of the Research Program ... 53

2.5.1 Rationale for Study One ......................................................................... 54 2.5.2 Rationale for Study Two ........................................................................ 56 2.5.3 Rationale for Study Three ...................................................................... 58 2.5.4 Significance of the Research Program .................................................... 59

Chapter Three: Measurement of Sleepiness and Performance ................................. 61 3.1 Overview of Chapter ..................................................................................... 61 3.1 Physiological Measures ................................................................................. 61

3.1.1 Electroencephalography ......................................................................... 61 3.1.2 Electrooculography ................................................................................ 63 3.1.3 Electrocardiography ............................................................................... 64 3.1.4 Physiological Data Acquisition and Processing ...................................... 66

3.2 Subjective Measures ..................................................................................... 69 3.2.1 Pittsburgh Sleep Quality Index ............................................................... 69 3.2.2 Epworth Sleepiness Scale ....................................................................... 71 3.2.3 Sleep Timing Questionnaire ................................................................... 72 3.2.4 Karolinska Sleepiness Scale ................................................................... 73 3.2.5 Motivational Effort Scale ....................................................................... 75 3.2.6 Intrinsic Motivation Inventory ................................................................ 75 3.2.7 Signs of Sleepiness Questionnaire .......................................................... 77

3.3 Performance and Behavioural Measures ........................................................ 78 3.3.1 Psychomotor Vigilance Test ................................................................... 78 3.3.2 Hazard Perception Test .......................................................................... 79 3.3.3 Wrist Actigraphy .................................................................................... 83

3.4 Genotyping Measures ................................................................................... 85 3.5 Room Setup .................................................................................................. 86 3.6 Chapter Summary ......................................................................................... 87

Chapter Four: The Physiological, Subjective, Performance, and Genetic Associations of Sleepiness in Young Adults ................................................................................ 89

4.1 Overview of Chapter ..................................................................................... 89 4.2 Introduction .................................................................................................. 89 4.3 Method ......................................................................................................... 96

4.3.1 Design .................................................................................................... 96 4.3.1.1 Research question one ..................................................................... 96 4.3.1.1 Research question two ..................................................................... 97

Sleep and wake drives xi

4.3.2 Participants ............................................................................................. 98 4.3.3 Measures ................................................................................................ 98

4.3.3.1 Demographic and traffic-related characteristics ................................ 98 4.3.3.2 Physiological measures .................................................................... 98 4.3.3.3 Subjective measures ......................................................................... 99 4.3.3.4 Behavioural measures ...................................................................... 99 4.3.3.5 Genotyping measures ..................................................................... 100

4.3.4 Procedure ............................................................................................. 100 4.3.5 SNP Selection....................................................................................... 101

4.4 Results ........................................................................................................ 102 4.4.1 Sleep Prior to Testing ........................................................................... 102 4.4.2 Data Treatment ..................................................................................... 103 4.4.3 Genotype Distributions ......................................................................... 103 4.4.4 Overall Session Trends ......................................................................... 104 4.4.5 Differences between Genotypes ............................................................ 109

4.5 Discussion ................................................................................................... 114 4.5.1 Time and Motivational Aspects ............................................................ 114 4.5.2 Genotype, Sleepiness, and Performance ................................................ 119 4.5.3 Study Implications ................................................................................ 121 4.5.4 Limitations and Future Research Directions .......................................... 122

4.6 Chapter Summary........................................................................................ 123

Chapter Five: Driving Cessation Motivations Study .............................................. 125 5.1 Overview of Chapter ................................................................................... 125 5.2 Introduction ................................................................................................. 125 5.3 Method ........................................................................................................ 127

5.3.1 Design .................................................................................................. 127 5.3.2 Participants ........................................................................................... 128 5.3.3 Measures .............................................................................................. 129

5.3.3.1 Demographic and traffic-related characteristics .............................. 129 5.3.3.2 Physiological measures .................................................................. 129 5.3.3.3 Subjective measures ....................................................................... 130 5.3.3.4 Performance measures ................................................................... 130

5.3.4 Procedure ............................................................................................. 130 5.4 Results ........................................................................................................ 132

5.4.1 Data Treatment ..................................................................................... 132 5.4.2 Sleep Prior to Testing ........................................................................... 132 5.4.3 Sleepiness and Cessation of the Task .................................................... 133 5.4.4 Physiological and Subjective Data ........................................................ 133 5.4.5 Signs of Sleepiness ............................................................................... 135

5.5 Discussion ................................................................................................... 136 5.5.1 Physiological and Subjective Sleepiness ............................................... 137 5.5.2 Study Implications and Driving Motivation .......................................... 139 5.5.3 Limitations and Future Research Directions .......................................... 141

5.6 Chapter Summary........................................................................................ 143

Sleep and wake drives xii

Chapter Six: Motivation to Continue Driving While Sleepy, the Effects on Sleepiness and Performance ................................................................................................... 145

6.1 Overview of Chapter ................................................................................... 145 6.2 Introduction ................................................................................................ 145 6.3 Method ....................................................................................................... 148

6.3.1 Design .................................................................................................. 148 6.3.1.1 Power analysis and sample size ..................................................... 150

6.3.2 Participants .......................................................................................... 150 6.3.3 Measures .............................................................................................. 151

6.3.3.1 Demographic and traffic-related characteristics ............................. 151 6.3.3.2 Physiological measures.................................................................. 151 6.3.3.3 Subjective measures ...................................................................... 152 6.3.3.4 Performance and behavioural measures ......................................... 153

6.3.4 Procedure ............................................................................................. 154 6.3.5 Motivations Experimental Conditions .................................................. 157

6.4 Results ........................................................................................................ 158 6.4.1 Data Treatment .................................................................................... 158 6.4.2 Manipulation Check and Order Effects ................................................. 159

6.4.2.1 Sleep duration prior to testing ........................................................ 159 6.4.2.2 Subjective sleepiness ..................................................................... 159 6.4.2.3 Hazard perception reliability.......................................................... 159 6.4.2.4 Order Effects ................................................................................. 160

6.4.3 Sleepiness, Motivation, and Performance ............................................. 160 6.4.3.1 Visual examination of the EEG and EOG data ............................... 160 6.4.3.2 Descriptive data ............................................................................. 160 6.4.3.3 Tests of hypotheses ....................................................................... 161

6.4.3.3.1 PVT testing session data ......................................................... 168 6.4.3.3.2 HPT testing session data ......................................................... 170

6.4.3.4 Signs of sleepiness during the HPT sessions .................................. 172 6.4.3.5 Intrinsic subjective experiences ..................................................... 175

6.5 Discussion .................................................................................................. 178 6.5.1 The Effects of Motivation with the HPT ............................................... 178 6.5.2 The Effects of Motivation with the PVT ............................................... 182 6.5.3 Sleepiness and Time-on-Task Effects ................................................... 184 6.5.4 Intrinsic Subjective Experiences ........................................................... 185 6.5.5 Limitations and Future Research Directions ......................................... 186

6.6 Chapter Summary ....................................................................................... 188

Chapter Seven: Overview, General Discussion, and Conclusions .......................... 191 7.1 Overview of Chapter ................................................................................... 191 7.2 Review of the Findings ............................................................................... 191

7.2.1 Study One: The Physiological, Subjective, Performance, and Genetic Associations of Sleepiness in Young Adults.................................................. 191 7.2.2 Study Two: Driving Cessation Motivations Study ................................ 192

Sleep and wake drives xiii

7.2.3 Study Three: Motivation to Continue Driving While Sleepy, the Effects on Sleepiness and Performance .......................................................................... 193

7.3 Theoretical Considerations .......................................................................... 194 7.3.1 Task Performance ................................................................................. 194 7.3.2 Wake Drive Effects .............................................................................. 197

7.4 Practical Considerations .............................................................................. 199 7.4.1 Motivations and On-Road Consideration .............................................. 199 7.4.2 Subjective Perceptions .......................................................................... 201

7.4.2.1 Perceptions of task performance..................................................... 201 7.4.2.2 Perceptions of signs of sleepiness .................................................. 202

7.5 Suggestions for Future Research Directions................................................. 203 7.6 Summary of Chapter ................................................................................... 206

References ............................................................................................................ 211

Appendix A .......................................................................................................... 281

Appendix B........................................................................................................... 283

Sleep and wake drives xiv

Sleep and wake drives xv

List of Tables

Table 2.1. Summary of the Physiological Characteristics of the Progression from Wake to Sleep. ........................................................................................................ 27

Table 2.2. Studies that have Researched the Effect of Motivation on Sleep-Wake Behaviours and Performance Outcomes. ................................................................. 48

Table 4.1. SNPs rs Number, their Associated Gene, and Biological Function. ....... 102

Table 4.2. SNPs and the Genotypes Frequencies with Chi-Square Tests of Hardy-Weinberg Equilibrium........................................................................................... 104

Table 4.3. ANOVA Statistics (3 x 6 Mixed Factorial) of the Physiological, Subjective, and Behavioural Data From the two PVT Sessions. ............................ 110

Table 4.5. Consolidated Genetic Results with the Five SNPs, their Genotype Frequencies, and the Various Phenotype Values of the Physiological, Subjective, and Performance Data. ................................................................................................ 113

Table 5.1. ANOVA Statistics (2 x 2 Mixed Factorial) of the Physiological and Subjective Data During Baselines and Cessation Periods. ..................................... 134

Table 6.1. Study Three Independent and Dependent Variables. ............................. 149

Table 6.2. Mean Amount of Sleep obtained prior to the Alert and Sleepy Testing Sessions for the Sleep Diary and Actigraphic Data. ............................................... 159

Table 6.3. ANOVA Statistics (2 x 2x 2 Repeated Measures) of the Physiological, Subjective, and Performance Data During the PVT Sessions. ................................ 166

Table 6.4. ANOVA Statistics (2 x 2 x 3 Repeated Measures) of the Physiological, Subjective, and Performance Data During the HPT Sessions. ................................ 167

Table 6.5. Post-hoc Comparisons of the Different Levels of the Three-Way Interaction of the PVT Performance Data. ............................................................. 169

Table 6.6. ANOVA Statistics (2 x 12 Repeated Measures) for the Comparisons Between Baseline Signs of Sleepiness Scores and post-HPT Sessions Signs of Sleepiness Scores. ................................................................................................. 174

Table 6.7. ANOVA Statistics (2 x 2 Repeated Measures) for the Intrinsic Subjective Experiences During the HPT Sessions................................................................... 177

Table A.1. Single Nucleotide Polymorphisms to be Examined by Study One. ....... 281

Table B.1. ANOVA Statistics Examining the Order Effects of Completing Two Testing Conditions (i.e., Motivated and Non-motivated) in One Day for the Two Days of Testing. .................................................................................................... 283

Sleep and wake drives xvi

Sleep and wake drives xvii

List of Figures

Figure 1.1. The odds ratios of having a sleep-related crash relative to an “alert” level of subjective sleepiness ........................................................................................... 12

Figure 1.2a-b. Illustrations of cortical areas affected by sleepiness and areas active during driving tasks................................................................................................. 15

Figure 2.1. Neuroanatomical projections that promote wakefulness (left) and the projections that promote sleep (right) ...................................................................... 26

Figure 2.2. Polysomnographic recordings of the transition from alertness to falling asleep ...................................................................................................................... 29

Figure 2.3. Illustration of the homeostatic build-up and circadian rhythm as a function of sleep pressure ........................................................................................ 32

Figure 2.4. The two-process model of sleep-wake regulation .................................. 37

Figure 2.5. The three-process model of sleep regulation .......................................... 38

Figure 2.6. The four-process model of sleep-wake regulation .................................. 43

Figure 3.1. Images of a traffic hazard used in the present study. .............................. 82

Figure 3.2. The laboratory ambient light readings ................................................... 87

Figure 4.1. Data collection points of the study. ........................................................ 97

Figure 4.2. EEG electrode sites utilised in the sleepy gene study. ............................ 99

Figure 4.3. Mean level of EEG sleepiness recorded at the F5, C3, and O1 electrode sites ...................................................................................................................... 106

Figure 4.4. Mean levels of subjective sleepiness (KSS) and reaction-time latency (PVT) ................................................................................................................... 107

Figure 5.1. EEG electrode sites utilised in the driving cessation study. .................. 129

Figure 5.2. Overview of the timeline of Study Two ............................................... 132

Figure 5.3. Mean levels of EEG absolute power, EOG eye blink duration, ECG R-R interval, and subjective sleepiness (KSS) during the baseline (white bars) and cessation (black bars) time periods ........................................................................ 134

Figure 5.4. Importance of the nine signs of sleepiness as an indicator of the level of sleepiness .............................................................................................................. 136

Figure 6.1. EEG electrode sites utilised in the motivation to continue driving while sleepy study. ......................................................................................................... 152

Figure 6.2. Overview of the timeline of Study Three ............................................. 156


Figure 6.4. Mean levels of subjective sleepiness (KSS), motivational effort (MES), and reaction-time latency (PVT) ........................................................................... 163


Sleep and wake drives xviii

Figure 6.6. Mean levels of subjective sleepiness (KSS), motivational effort (MES), and reaction-time latency (HPT) ........................................................................... 165

Figure 6.7. Three-way interaction results between Sleepiness level, Motivation level, and Time Period factors ........................................................................................ 169

Figure 6.8. Importance of the signs of sleepiness as indicators of the participants level of sleepiness ................................................................................................. 173

Figure 6.9. Mean levels of participants’ intrinsic subjective experiences when performing the hazard perception task .................................................................. 177

Sleep and wake drives xix

Abbreviations and Terms

Chronotype Relates to an individual’s diurnal preference, being morning orientated (morning chronotype) or evening orientated (evening chronotype)

DNA Deoxyribonucleic acid is a molecule that contains the majority of genetic instructions required for the development, functioning, and reproduction of all known organisms

ECG Electrocardiography

EEG Electroencephalography

EMG Electromyography

EOG Electrooculography

Epoch A quantified amount of time, also known as a time bin. Typically, 30-sec epochs are utilised with the scoring of sleep studies. Spectral analyses of EEG data utilise epochs of varying duration, but typically 2-4 sec.

ESS Epworth Sleepiness Scale

fRMI functional Magnetic Resonance Imaging

HPT Hazard Perception Test, a reaction-time latency measure to identify a traffic situation that may result in a dangerous situation, requiring a reaction from the driver to avoid an crash

Intraclass correlation coefficients (ICC)

A way to quantify the amount of variance associated with interindividual differences

KSS Karolinska Sleepiness Scale

Lux The Lux is the international system of units measure of illuminance (symbol: lx); quantified as luminous power per area

MSLT Multiple Sleep Latency Test

MWT Maintenance of Wakefulness Test

NREM Non-rapid eye movement

Polysomnography (PSG) A multi-measure test used in the study of sleep and as a diagnostic tool in sleep medicine, otherwise known as a sleep test

PET Positron emission tomography

Power band A grouping or range of frequencies; for example the EEG alpha power band (8-13Hz) which is indicative of drowsiness

PSQI Pittsburgh Sleep Quality Index

PVT Psychomotor Vigilance Test

REM Rapid eye movement

Sleep architecture Represents the structure of sleep as defined by specific electrophysical indicators from EEG, EOG, and EMG recordings

Sleep debt Experiencing sleep deprivation results in the accumulation of a sleep debt; which is the under accumulation of enough sleep to ensure homeostasis

Sleep diary Typically a self-reported account of an individual’s sleep and wake times

SNP Single nucleotide polymorphism

SOL Sleep onset latency

WAVT Wilkinson Auditory Vigilance Test

Sleep and wake drives xx

Sleep and wake drives xxi

Definition: Sleepiness, Fatigue, and Alertness

The term fatigue is used interchangeably with sleepiness in road safety

literature. Sleepiness is likely the greatest contributor to fatigue as well as other

components including task demands and time on task. The term sleepiness is

generally accepted as the inclination for sleep. The term fatigue is generally accepted

as a disinclination to continue performing the task at hand, which is distinctly

different from how sleepiness is conceptualised (Brown, 1994). Moreover, if fatigue

is the disinclination to continue performing a certain task, then why do “fatigued”

drivers choose to continue driving, even when aware of their increasing fatigue?

It has been stated that “Heavy meals, warm rooms, boring lectures, and the

monotony of long-distance automobile driving unmask the presence of physiological

sleepiness but do not cause it” (Roth, Roehrs, Carskadon, & Dement, 2005). Given

the eloquent and zealous arguments for the use of the term sleepiness rather than the

term fatigue by a number of researchers (e.g., Johns, 2001; Roth et al., 2005) as well

as the author’s preference and beliefs, the term sleepiness will be used throughout

this body of work. The term sleepiness is conceptualised as the opposite of alertness.

In essence, sleepiness and alertness are at opposite ends of the same spectrum, and

this is how sleepiness and alertness are conceptualised in the anchors of the

Karolinska Sleepiness Scale (Åkerstedt & Gillberg, 1990).

Sleep and wake drives xxii

Sleep and wake drives xxiii

Statement of Original Authorship

The work contained in this thesis has not been previously submitted to meet

requirements for an award at this or any other higher education institution. To the

best of my knowledge and belief, this thesis contains no material previously

published or written by another person except where due reference is made. The

author of this work was responsible for all of the data collection and the analyses of

the data.

Name: Christopher N Watling

Signature: QUT Verified Signature

Date: August, 2016

Sleep and wake drives xxiv

Sleep and wake drives xxv

Acknowledgements

While this thesis has only one authors name on it, it is an inescapable truth

that so many people have contributed to this thesis in a variety of ways. First of all, I

am grateful for the efforts of my four supervisors, Professor Herbert Biggs, Doctors

Kerry Armstrong and Joanne Voisey, and Associate Professor Simon Smith provided

with shaping this PhD thesis and of course, shaping me into the researcher I am.

Moreover, the autonomy you allowed me to have with the structuring of my PhD

research program was truly welcomed. I am very appreciative of the advice

Associate Professor Göran Kecklund contributed towards aspects of the PhD

program. I am also grateful for the mentorship, but more importantly the friendship

from Associate Professor Igor Radun. In hindsight, Igor you filled an important role

at a critical stage of my PhD “journey” and for that, I am most grateful, thanks mate!

Last, professional editor Adele Fletcher, provided copy-editing and proofreading

services, according to the guidelines laid out in the university-endorsed guidelines

and the Australian Standards for editing research theses.

Financial support for this research program was provided by an Australian

Post Graduate Award as well as an initial top-up scholarship from the Centre for

Accident Research and Road Safety - Queensland for the first year of the PhD

scholarship period, then, during the last two years of the PhD scholarship period, by

a top-up scholarship from the Institute of Health and Biomedical Innovation (IHBI).

The “Sleepy Genes” study (Chapter 4) was financially supported by an IHBI seeding

grant awarded to Aspro Simon Smith and Dr Joanne Voisey. Last, the “Motivation to

Continue Driving” study (Chapter 6) was financially supported from a grant awarded

to Mr Christopher Watling, Aspro Simon Smith, and Dr Kerry Armstrong from the

ARRB Group. General financial support for the PhD research program (e.g.,

Sleep and wake drives xxvi

conference attendance, airline tickets) was provided from two sources, those being a

Faculty of Health Grant-In-Aid and IHBI student funds. The financial support from

all of the above sources was appreciated and welcomed. Last, I would like to

acknowledge the laboratory work performed by Angela O’Keeffe and Dr Joanne

Voisey regarding the processing the saliva samples for the “Sleepy Genes” study.

In many ways, this thesis represents seven years of postgraduate study that

began with undertaking and completing a Master of Applied Science (Research)

degree. However, if one considers the time taken, being eleven years to complete an

undergraduate Bachelor of Behavioural Science (Psychology) course, then a Post

Graduate Diploma in Psychology, and then the Master of Applied Science

(Research) degree, and now the Doctor of Philosophy … well, let us just say I am

ready for some long service leave!

There are just far too many people I would like—no, need to thank for their

support and friendship over the years. While it is tempting to try to mention you all

here, it is highly likely that some people would be omitted, and thus I will refrain

from trying. Instead, I will seek you all out at some time in the future and thank you

personally. While it is nice to be mentioned in an acknowledgements section, it is

more appropriate and real, to thank you in person.

As always, I reserve my greatest thanks for my family. Your love,

unwavering support, and profound tolerance have been so very important with the

completion of not only the present thesis but the previous ones as well. Yes, Mum

and Dad, I will finally be looking for a fulltime job.

Finally and most importantly, to the love of life, my beautiful wife, tack så

mycket och jag älskar dig så, så mycket. Jag tittar kärleksfullt på dig.

Sleep and wake drives xxvii

Publications from this Program of Research

Journal Paper

Watling, C.N., Smith, S.S., & Horswill, M.S. (2016). Psychophysiological changes

associated with self-regulation of sleepiness and cessation from a hazard

perception task. Journal of Psychophysiology, 30(2), 66-75.

Conference Papers

Watling, C. N., & Smith, S. S. (2012). Too sleepy to drive: self-perception and

regulation of driving when sleepy. Paper presented at the Occupational Safety

in Transport Conference. Gold Coast, Queensland.

Watling, C. N., Armstrong, K. A., & Smith, S. S. (2013). Sleepiness : how a

biological drive can influence other risky road user behaviours. Paper

presented at the Proceedings of the 2013 Australasian College of Road Safety

(ACRS) National Conference, Adelaide, South Australia.

Conference Item

Watling, C. N., & Smith, S. S. (2012). Driver sleepiness self-regulation :

physiological and subjective evidence. In the Australian Sleep Association

Conference, Sleep DownUnder2012 : Sleep up Top, Darwin, Northern

Territory.

Sleep and wake drives xxviii

Sleep and wake drives 1

Chapter One: Introduction

If sleep does not serve an absolute vital function, then it is the biggest mistake the

evolutionary process ever made.

–Allan Rechtschaffen

1.1 Overview of Chapter

This chapter presents the background and rationale for the present research

program. Specifically, this chapter will focus on current literature into sleep patterns,

the associations of sleepiness and vehicle crashes, risk perception of sleepy driving,

and the effect sleepiness has on driving, particularly in younger drivers. The sleep-

wake cycle is extremely important cycle that can affect neurobehavioural functioning

of every individual, as no one individual is immune to the effects of sleepiness. The

task of driving is complex as it involves the coordination of a number of

psychological processes. Many key psychological processes involved with driving

are impaired by sleepiness. Consequently, driving while sleepy is a substantial

contributor to crash incidents and is likely to be associated with crashes involving

other risky driving behaviours.

1.2 Why Study Sleepiness?

Human beings have seemingly evolved with a requirement, or a propensity,

such that within each 24-hour period they must sleep for a certain duration of time;

and maintaining complete wakefulness becomes increasingly more difficult after

only a single day (Kleitman, 1963). After only a day without sleep the individual

experiences an increasing drive for sleep, and this is characterised by the occurrence

of microsleeps while the individual is awake and active (Boyle, Tippin, Paul, &


Rizzo, 2008; Lal & Craig, 2005). Consequently, when an individual goes to sleep

after experiencing sleep deprivation, they will, if allowed, sleep for a greater duration

of time than their habitual sleep duration (Borbély, 1982). In effect, the individual

attempts to recover some of their lost sleep time. As such, the apparent drive of sleep

and the recovery mechanism of lost sleep time suggest sleep plays an important role

in human physiology.

The importance of the sleep-wake cycle is also demonstrated by a myriad of

biological, psychological, and physiological changes and impairments that occur

during and after experiencing sleep deprivation. Studies demonstrate that total and

partial sleep deprivation results in altered gene expression (Möller-Levet et al.,

2013), a number of neuroendocrine changes (Wright Jr et al., 2015), impaired

affective control (Anderson & Platten, 2011; Kahn-Greene, Lipizzi, Conrad,

Kamimori, & Killgore, 2006), and impaired cognitive functioning (M. L. Jackson et

al., 2013), including the ability to maintain attention (Dinges et al., 1997). However,

the magnitude of impairment from sleep deprivation can fluctuate over time (Doran,

Van Dongen, & Dinges, 2001) as well as varying between individuals (Van Dongen,

Baynard, Maislin, & Dinges, 2004); suggesting there are certain motivational effects

and trait-like aspects that can influence the magnitude of impairment from sleep

deprivation. The motivational effects and trait-like aspects will be the focus of this

research program. Notwithstanding, the apparent, motivational, and trait-like aspects

associated with sleepiness, no one individual is completely immune to the negative

effects of sleepiness as the quantity and quality of an individual’s sleep is

intrinsically linked with their physiological and daytime functioning.


1.2.1 Levels of Societal Sleepiness

The prevalence and magnitude of sleep disruptions, sleep disorders, poor

sleep habits, and daytime sleepiness is becoming more evident. Sleep disorders (e.g.,

obstructive sleep apnoea, insomnia, disorders of circadian timing) contribute to

excessive daytime sleepiness (Espie, 2002; Kryger, Roth, & Dement, 1994).

However, a substantial proportion of the excessive daytime sleepiness experienced

by modern society’s members might be due to inadequate sleep durations and/or

irregular sleep patterns and habits.

A North American survey conducted by the National Sleep Foundation

(2015) examined various aspects of sleep health and daytime functioning of the

general public. Participants were asked about their typical sleep duration throughout

the week and over the weekend. On average, the participants reported sleeping 6 hr

54 min during the week. However, this average increased to 7 hr 36 min on the

weekends. These increases of sleep duration during the weekend could be indicative

of the individual attempting to recover some of their lost sleep or to “catch up” on

insufficient sleep during the week (Basner et al., 2007). For instance, during the

weekdays 18% of people slept eight to nine hours; however, during the weekend this

percentage increased to 31%. Similarly, 3% of the sample slept for nine or more

hours during the weekdays, but this increased to 13% during the weekends.

Variations in the timing and duration of sleeping time can facilitate circadian phase

shifts (Czeisler, Richardson, Zimmerman, Moore-Ede, & Weitzman, 1981; S. S.

Smith & Trinder, 2005), which can also lead to excessive daytime sleepiness and

impaired cognitive functions (Åkerstedt, 2007; Haimov & Arendt, 1999; Wright,

Lowry, & Lebourgeois, 2012). Overall, 45% of the sample reported that their sleep

duration was not adequate to ensure they could function at their best during the day.


Several research studies have been performed that have examined aspects of

sleep health and daytime functioning of Australians. Two studies found a large

proportion of individuals (29-31%) sleep less than 7 hr per night (Bartlett, Marshall,

Williams, & Grunstein, 2008; Cummins et al., 2012). This might represent

chronically insufficient sleep for many individuals (Carskadon & Dement, 2005;

Roehrs, Carskadon, Dement, & Roth, 2005). That is, neurobehavioural deficits occur

when sleeping time is reduced to 6 hr per night over a two week period (Van

Dongen, Maislin, Mullington, & Dinges, 2003). The magnitude of such

neurobehavioural deficits from two weeks of chronic sleep restriction are equivalent

to neurobehavioural deficits observed after being completely sleep deprived for up to

two nights (Van Dongen, Maislin, et al., 2003). Experiencing excessive daytime

sleepiness is commonly reported in Australians, with approximately two fifths

(38.10%) of individuals in the last two years reporting they have fallen asleep at

work due to excessive daytime sleepiness (Ferguson, Clarkson, & Dawson, 2012). In

the Bartlett et al.’s (2008) study, levels of excessive daytime sleepiness was assessed

using the Epworth Sleepiness Scale (Johns, 1991), with 11.7% of the participants

displaying chronic levels of daytime sleepiness and 9.8% of participants aged 18-24

years reporting chronic daytime sleepiness levels.

Aside from daytime sleepiness and the associated neurobehavioural

impairment, short sleep durations also have an effect psychological functioning. It

was found in the Cummins et al. (2012) study that participants who slept less than 7

hr were significantly below the normative range of personal wellbeing (e.g.,

satisfaction with their lives, quality of life and wellbeing), with steeper decreases in

personal wellbeing associated with fewer hours slept. Other research also

demonstrates that shorter sleep durations are associated with poorer social


relationships and interpersonal communication (Kahn-Greene et al., 2006; Xanidis &

Brignell, 2016). Considered together, these findings suggest that many Australians

have poor sleep health and are potentially accumulating sleep debts.

It is important to note that individual differences exist for sleep requirements.

That is, in the absence of daytime sleepiness there are sleep duration variations

between individuals (Basner et al., 2007; Klerman & Dijk, 2005). However, on

average most survey data suggests that eight hours of sleep is needed for adults to

feel refreshed and well rested (Carskadon & Dement, 2005; Ferguson et al., 2012).

Differences in the duration of sleep across the lifespan are also apparent. Sleep

duration is at its longest during infancy and then reaches its shortest duration around

middle age (Henderson, France, & Blampied, 2011; Ohayon, Carskadon,

Guilleminault, & Vitiello, 2004).

In summary, current trends suggest that sleepiness may be increasing, owing

to societal demands and perceptions of work demands (Ferguson et al., 2012;

Knutson, Van Cauter, Rathouz, DeLeire, & Lauderdale, 2010; National Sleep

Foundation, 2007). In addition, younger individuals also seem to have increasing

levels of excessive daytime sleepiness (Bartlett et al., 2008; Cummins et al., 2012)

with levels of chronic daytime sleepiness in younger demographics being

problematic (Bartlett et al., 2008). These increases in sleepiness are most likely to

lead to impaired daytime functioning of a number of psychological processes, many

of which are important for safety critical tasks.

1.3 Why Study Driver Sleepiness?

The role of sleepiness as a major contributing factor to vehicle crashes is

widely recognised both nationally (ATSB, 2006; Bureau of Infrastructure Transport

and Regional Economics, 2015; Dobbie, 2002) and internationally (Åkerstedt, 2000;


Connor et al., 2002; Gander, Marshall, Harris, & Reid, 2005; Horne & Reyner, 1995;

Nabi et al., 2006). The causes of many vehicle crashes are often multifactorial (Pack

et al., 1995; Shinar, 1978), and sleepiness may also contribute to crashes primarily

attributed to other factors such as alcohol or speeding (Rajaratnam & Jones, 2004;

Watling, Armstrong, & Smith, 2013). Sleepy drivers are at an increased risk of

having a crash when sleepiness affects their attention, cognitive functioning, and

reaction times (Åkerstedt, Connor, Gray, & Kecklund, 2008; Stutts, Wilkins, Scott

Osberg, & Vaughn, 2003). Sleep-related crashes are generally more severe than

crashes associated with other risky driving behaviours. Consequently, sleep-related

crashes have higher fatality rates (Horne & Reyner, 2001; Parsons, 1986; Stutts et al.,

2003). This is due in part to sleep-related crashes more often occurring on roads with

higher posted speed limits (Horne & Reyner, 1995; Radun & Summala, 2004)

resulting in high speed impacts, and is coupled with the inability of sleeping drivers

to make any evasive manoeuvres to mitigate the impact. Thus, the deleterious effect

sleepiness can have on driving safety suggests driver sleepiness research is an

important road safety concern.

1.4 The Burden of Crashes

Injuries resulting from traffic incidents are a significant public health problem

worldwide (World Health Organisation [WHO], 2004). It is predicted that unless

substantial gains are made in the prevention of crashes, traffic-related injuries will

become the third-greatest contributor to the global burden of disease and injury by

2020 (WHO, 2004; WHO, 2013). Approximately 1,100 fatalities occur on Australian

roads each year (Bureau of Infrastructure Transport and Regional Economics, 2015).

These road crashes are an avoidable cause of premature mortality and morbidity (R.

D. Smith, Jan, & Shiell, 1993). Moreover, road crashes exact significant


psychological and physical trauma on the individual, family members, and on the

community (Mitchell, 1997). When the associated property damage, infrastructure

adjustment, and emergency service resources are considered, society bears an

enormous social and economic cost from road crashes (Berry & Harrison, 2007;

Shinar, 2007). In Australia, the Bureau of Infrastructure Transport and Regional

Economics (2009) conservatively estimated that the financial cost of road crashes to

the community was 27 billion AUD a year.

1.5 The Incidence of Sleep-Related Crashes

Accurate and consistent incidence rates for sleep-related crashes are difficult

to ascertain. This is in part due to the absence of an objective test for measuring a

driver’s sleepiness level, such as the blood alcohol content as with alcohol-related

crashes (Pack et al., 1995; Radun, Ohisalo, Radun, & Rajalin, 2012). Generally, there

are three sources of data that are drawn upon to estimate incidence rates; these are

official crash data, self-report data, and case-control studies. The incidence rates of

sleep-related crashes from these three data sources are discussed below.

1.5.1 Official Data

The examination of official crash data generally reveals lower estimates of

sleep-related crash incidences. Attempts to quantify the incidence rates of sleep-

related crashes often rely on subjective reports from crash investigators or police and

on proxy measures based on crash characteristics. Early investigations in the United

States revealed that a low proportion of all crashes (0.5-0.9%) and only 3.6% of fatal

crashes were being classified due to the effect from sleepiness (Knipling & Wang,

1994; Pack et al., 1995). Since these early accounts in the 1990s, the knowledge of

crash investigators regarding specific characteristics of sleep-related crashes has

increased (Horne & Reyner, 2001; Radun, Ohisalo, Radun, Wahde, & Kecklund,


2013) and more crashes are classified as being due to sleepiness. For instance, during

2014 in Queensland, Australia, 13.90% of all fatal road crashes were classified as

being primarily due to driver sleepiness (Department of Transport and Main Roads,

2015).

In contrast, Horne and Reyner (1995) examined the police records relating to

crash characteristics and on-the-spot interviews conducted in England. Their criteria

(proxy definition) for the involvement of sleepiness included: exclusion of alcohol

related and speeding crashes, running off the road or into another vehicle, no signs of

evasive manoeuvres, no mechanical defects, good weather conditions, and whether

the police report recorded the suspicion that sleepiness was the primary cause for the

crash. This analysis found that 20% of motorway crashes and 16% of all crashes on

nonurban roads were directly attributable to sleepiness by these criteria. In Australia,

Dobbie (2002) applied a proxy definition to fatal crashes and found the proxy

definition identified 17% of crashes as being sleep-related crashes; however, police

and coroners reports only identified 7% of crashes as being sleep-related crashes.

The advantages and limitations of proxy definitions for identifying sleep-related

crashes have been explored and commented on for their utility with ascribing sleep-

related crashes (e.g., Armstrong, Filtness, Watling, Barraclough, & Haworth, 2013).

1.5.2 Self-Report Data

An alternative approach has been to gather retrospective data from drivers

regarding their involvement in sleep-related crashes or sleep-related close calls.

However, reporting biases such as social desirability and recall bias can affect the

accuracy of self-reported incidence rates (Neugebauer & Ng, 1990; Wåhlberg, Dorn,

& Kline, 2010). A study undertaken by Maycock (1996) on drivers in the United

Kingdom utilised a self-report postal questionnaire methodology. Overall, the


collected data suggested that 7% of the sample’s crashes were due to “tiredness,”

while higher crash rates occurred on motorways (15%) and in non-built-up areas

(10%). Data from Australia reveal that 17% of drivers reported that they had fallen

asleep while driving, with 8% of these resulting in a crash (Pennay, 2008).

1.5.3 Case Controlled Data

In response to the difficulties in acquiring accurate estimates of sleep-related

crashes, more rigorous methodologies, such as case-control studies, have been used

to assess the incidence of sleep-related crashes. A consensus has emerged among

these more rigorous studies that the proportion of all vehicle crashes related to

sleepiness is in the order of 20% (e.g., Connor et al., 2002; Garbarino, Nobili,

Beelke, De Carli, & Ferrillo, 2001; Kecklund, Anund, Wahlström, Philip, &

Akerstedt, 2012; Nabi et al., 2006).

Connor et al. (2002) conducted a case-control study assessing the magnitude

of fatal and severe sleep-related crashes in New Zealand. The Connor et al. (2002)

study utilised highly structured interviews to elicit specific indicators of driver

sleepiness (i.e., self-report of sleepiness, time of day, prior night and week’s sleep

duration) from case and control participants. The analysis controlled for confounding

variables; such as age, sex, vehicle speed, length of driving time, and use of

recreational drugs; for which many incidence studies have not controlled. Most

importantly, this study measured and controlled for the blood alcohol concentration

of participants. Alcohol use and sleepiness tend to co-occur in many crashes (as

alcohol is often consumed during the evening hours). Connor et al. (2002) estimated

that the population attributable risk for fatal and severe crashes associated with

sleepy driving was 19%. That is, if there were a cessation of all sleep-related crashes,

there would be a 19% decrease in all fatal and severe crashes.


The level of evidence provided by Connor et al.’s (2002) study could be

considered the best estimate of the scale of sleep-related crashes for fatal and severe

crashes. The findings from this study have been replicated with the same rigour and

methodology in a study of a Swedish sample of non-fatally injured drivers.

Preliminary results from a non-peer reviewed Swedish case-control study performed

by Kecklund, Anund, Wahlström, and Åkerstedt (2012) similarly found that 20% of

nonfatal crashes that were admitted to hospital were attributable to acute sleepiness.

The extent of incidence for less severe sleep-related crashes may be even greater

(Pack et al., 1995).

Crashes often have multifactorial causes, and therefore a degree of sleepiness

may be involved in crashes that were primarily attributed to other risky driving

behaviours such as drink driving, distracted driving, or speeding (Watling et al.,

2013). For instance, multiplicative effects are found when sleepiness and alcohol

(even low blood alcohol concentrations) are combined, with impairments of

performance above the levels of impairment from either sleepiness or alcohol alone

(Banks, Catcheside, Lack, Grunstein, & McEvoy, 2004). This interaction has been

observed even when blood alcohol content has fallen to near zero (Barrett, Horne, &

Reyner, 2004). Sleepiness is also associated with deviations from posted speed limits

as well as increases of the variability of these speed deviations (Arnedt, Wilde, Munt,

& Maclean, 2000). The magnitude of the variations of speed deviations also

increases with longer durations of driving (Gillberg, Kecklund, & Åkerstedt, 1996).

Sleepy drivers are also more prone to make short diverted gazes (< 3 sec),

longer diverted gazes (> 3 sec), and to have more distraction-related driving incidents

(e.g., defined as two wheels leaving the driving lane) than when fully rested

(Anderson & Horne, 2013). When a sleep deprived driver has an elevated blood


alcohol content (but below the legal driving level) and uses a hands-free mobile

phone, substantial impairment of driving performance occurs (i.e., speed deviations

and tailgating). As such, the contributions of sleepiness to crash incident rates is

likely to be higher than those estimated from routine crash statistics (Åkerstedt,

2000; Cercarelli & Haworth, 2002; Horne & Reyner, 1995; Watling et al., 2013).

Thus, the evidence indicating sleepiness as a major factor in crashes is compelling.

1.6 Level of Crash Risk and Risk Perceptions of Sleepy Driving

Recent data have described the relationship between sleepiness while driving

and the odds of crashing (Åkerstedt, Connor, et al., 2008). The odds of having a

sleep-related crash rapidly increase in a nonlinear, almost exponential curve as

subjective sleepiness levels increase (see Figure 1.1 over page). Similar rapid

increases in the odds of having a crash are found for alcohol intoxication (e.g.,

Borkenstein, Crowther, Shumate, Zeil, & Zylman, 1964; Evans, 1991), and provide

analogous risks. Moreover, the performance decrement associated with modest sleep

deprivation (20 hours of wakefulness) approaches levels of impairment associated

with blood alcohol content levels of 0.1% (Dawson & Reid, 1997; Williamson &

Feyer, 2000).


Figure 1.1. The odds ratios of having a sleep-related crash relative to an “alert” level of subjective sleepiness, adapted from Åkerstedt, Connor, et al. (2008). Subjective sleepiness was quantified using the Karolinska Sleepiness Scale (range of 1-9 with higher scores indicating greater sleepiness). Odds ratio is the likelihood of an event occurring, for instance a Karolinska Sleepiness Scale score of 7, “sleepy, no effort to stay awake,” relates to an odds ratio of 4, which is a four times greater likelihood of crashing.

While the dangerousness of sleepy driving has been demonstrated in crash

estimates and crash risk estimates, and is acknowledged by police and prosecutors as

a risky driving behaviour (Radun et al., 2013), the risk perception and behaviours of

drivers do not always reflect that dangerousness (Radun, Ohisalo, Radun, &

Kecklund, 2011; Watling, Armstrong, Smith, & Obst, 2015; Watling & Watling,

2015). For instance, some questionnaire studies have shown that sleepy driving is not

viewed by most drivers as a critical issue for road safety (e.g., Pennay, 2008; Radun,

Radun, Wahde, Watling, & Kecklund, 2015; Vanlaar, Simpson, Mayhew, &

Robertson, 2008). Furthermore, a remarkable proportion of individuals (59-77%)

report driving when they are feeling sleepy (Armstrong, Obst, Banks, & Smith, 2010;

Vanlaar et al., 2008; Watling, Armstrong, & Radun, 2015), with a large proportion of

drivers (73%) reporting that they would continue to drive even when aware of


increasing levels of sleepiness (Nordbakke & Sagberg, 2007). Approximately a third

of drivers have admitted to fallen asleep while driving in the past year (National

Sleep Foundation, 2008, 2015). Despite the efforts of road safety agencies to educate

drivers about the dangerousness of sleepy driving, these figures suggest a lack of

awareness or appreciation by drivers of the dangers of driving while sleepy.

1.7 The Cortical Associations of Sleepiness and the Effects on Driving

Neuroimaging studies reveal sleep deprivation results in reductions of

metabolic activity in cortical and subcortical tissue. For instance, Thomas et al.

(2000) found that sleep deprived subjects showed a pronounced global reduction in

glucose metabolic activity quantified by positron emission tomography when

performing the serial addition/subtraction task. Specifically, reductions in metabolic

activity were observed in cortical areas such as the posterior parietal, temporal,

frontal, and prefrontal cortices with reductions in subcortical areas of the thalamus

and the anterior cingulated as well. Less-pronounced reductions were also found in

parts of the cerebellum and anterior parietal lobes. Large reductions in metabolic

activity during sleep deprivation have been independently observed in the

corticothalamic network that mediates attention and high-order cognitive processes

(Jones & Harrison, 2001; Lim et al., 2010; Manly & Robertson, 1997). These data

suggest that sleep deprivation is associated with reductions in metabolic activity in a

number of cortical and subcortical areas subserving a variety of psychological

processes.

Neuroimaging studies also show that many of the cortical and subcortical

areas that display reduced activity during sleep deprivation also correspond to

cortical areas that are active during the task of driving (Graydon et al., 2004;

Horikawa et al., 2005; Uchiyama, Ebe, Kozato, Okada, & Sadato, 2003).


Specifically, studies using functional magnetic resonance imaging (fMRI)

demonstrate that cortical tissue of the occipital lobes are active when perceiving

driving stimuli, cortical tissue in the posterior parietal lobes, premotor cortex,

thalamic and cerebellum are active during driving tasks that require a degree of

attention, and when performing driving navigation tasks and manoeuvres (Spiers &

Maguire, 2007). Additionally, cortical tissue of the right prefrontal cortex are

specifically activated when processing road traffic rules such as retrieving rules from

memory (Spiers & Maguire, 2007). The anterior cingulate is active when performing

a simulated driving tasks of collision avoidance (Spiers & Maguire, 2007) as well as

maintaining safe driving distances (Uchiyama et al., 2003).

As previously described, these cortical areas are also known to be markedly

affected by sleep deprivation; thus, it is possible that sleepiness has specific effects

on the safe operation of a vehicle. The cortical and subcortical brain areas that show

reduced activity during sleep deprivation are shown in Figure 1.2a (over page). The

cortical and subcortical brain areas active during driving tasks are shown in Figure

1.2b. This figure outlines substantial overlap between cortical regions that are

essential with the performance of many driving tasks and cortical areas that have

been found to show significant reduction of activity during sleep deprivation.


Figure 1.2a-b. Illustrations of cortical areas affected by sleepiness and areas active during driving tasks. The light blue areas in Figure 1.2a show cortical areas the display reduced activity during sleep deprivation and yellow areas in Figure 1.2b show the cortical areas that are active during driving tasks. The actual cortical activation is much more diffuse than the illustrations suggest.

1.7.1 The Task of Driving

Driving is a complex task that requires the successful operation and

coordination of a number of psychological processes to ensure the safety of the

driver and other road users. These psychological processes include learning,

memory, perception, motor control, attention, decision-making, and executive

functioning (Groeger, 2002; Horswill & McKenna, 2004; Spiers & Maguire, 2007;

Uchiyama et al., 2003). As such, falling asleep is not the only way sleepiness can

affect driving performance. The driving skill of collision avoidance is also likely to


be impaired by sleepiness. Crash incidence studies suggest that short-duration

sleepers (< 6 hr) are more likely to be involved in a rear-end collisions and single-car

crashes when driving in an urban area during short driving spells (Abe, Komada,

Asaoka, Ozaki, & Inoue, 2011; Abe, Komada, Nishida, Hayashida, & Inoue, 2010).

This evidence is supported from imaging studies that demonstrate the anterior

cingulate has reduced activity during sleepiness (Thomas et al., 2000) and is a

subcortical area that is highly active during simulated driving periods intended to

mimic collision avoidance (Spiers & Maguire, 2007) and driving at safe distances

(Uchiyama et al., 2003).

A number of simulator studies have shown that increases in the levels of

sleepiness can lead to impairment of vehicle control. A measure of vehicle control,

the standard deviation of lane positioning (i.e., lane drifting) has repeatedly been

found to be sensitive to increases in sleepiness levels (e.g., Boyle et al., 2008; Pizza,

Contardi, Ferlisi, Mondini, & Cirignotta, 2008; Watling, Åkerstedt, Kecklund, &

Anund, 2015). Moreover, the magnitude of lane drifting has been shown to increase

with greater levels of sleep deprivation (Åkerstedt, Peters, Anund, & Kecklund,

2005; De Valck & Cluydts, 2001) and to increase as the driving duration increase as

well (Anund, Kecklund, Peters, Forsman, et al., 2008; Watling, Åkerstedt, et al.,

2015). These increases in the magnitude of lane drifting also correlate with increases

of physiological defined indices of sleepiness (Campagne, Pebayle, & Muzet, 2004).

The effect size of sleepiness on lane drifting indices has been determined by the

bootstrap method (i.e., Balkin et al., 2004) to be 0.08 and represents a small effect

size (e.g., Davison & Hinkley, 1997; Efron & Tibshirani, 1993).

One criticism of driving simulators is they may lack the specificity to capture

the range of psychological processes that occur during driving. The major design


concept behind modern driving simulators is that of physical fidelity (Gopher &

Kimchi, 1989) with an emphasis on replicating motor control aspects of driving, such

as vehicle control indices. For the experienced driver, motor control aspects of

driving are regarded by researchers as a highly learned task that requires relatively

few cognitive resources for effective performance (Horswill & McKenna, 2004;

McKenna & Farrand, 1999).

The validity of driving simulators is a critical aspect for their generalisability

to on-road situations. Several studies have reported that the absolute level of

behavioural validity between simulated and on-road driving for motor control indices

(i.e., lane drifting) has been unsatisfactory (Mayhew et al., 2011; Törnros, 1998).

Moreover, motor control indices such as lane deviation or sudden braking are

undemanding “low-order” tasks. Such low-order tasks may be a minor component of

the driving demands made by the complex traffic environments of on-road driving.

Simulator studies that primarily focus on motor control indices have a limited utility

with the dynamic on-road environment (George, 2003). A driving simulator task that

can elicit measurement of a greater number of the psychological processes involved

in driving tasks, especially those impacted by sleepiness, may be necessary to better

understand the impact of sleepiness on driving safety.

1.7.1.1 Hazard perception

Hazard perception, in the context of driving, can be described as the skill

required to determine or to predict that a traffic condition may result in a dangerous

situation, requiring a reaction from the driver to avoid an incident (Horswill &

McKenna, 2004; McKenna & Crick, 1991). Driving hazards that are not quickly

identified and recognised as potentially dangerous situations could, depending on the

reactions of the drivers result in a crash (Underwood, Crundall, & Chapman, 2002).


The principle of the hazard perception paradigm is to examine the cognitive

processes that are in operation during the processing of potential threats presented in

the visual simulation of the road environment (McKenna & Crick, 1991).

When the relationship between the different driving skills (e.g., vehicle

control, skid control, hazard perception) are examined for their association with crash

involvement, only the skill of hazard perception has been consistently found to be

associated crash incidents (Boufous, Ivers, Senserrick, & Stevenson, 2011; Darby,

Murray, & Raeside, 2009; Horswill, Hill, & Wetton, 2015; M. Hull & Christie, 1992;

McKenna & Horswill, 1999; Pelz & Krupat, 1974; Quimby, Maycock, Carter,

Dixon, & Wall, 1986). All other driving skills (e.g., vehicle control, skid control)

have been shown to have inconsistent relationships with crashes (e.g., Katila,

Keskinen, Hatakka, & Laapotti, 2004; A. F. Williams & O'Neill, 1974). Thus, hazard

perception is an important driving task and has a degree of criterion-related validity

with actual on-road crashes.

1.7.1.2 Psychological components of hazard perception

Hazard perception requires the execution of a number of different

psychological processes. That is, hazard perception is a multicomponent and

multistaged skill with each skill reliant on the one preceding it. First, hazard

perception requires the individual to actively scan the road environment (Chapman,

Underwood, & Roberts, 2002), then register a potentially hazardous event. These two

initial skills likely utilise psychological processes such as perception, attention, and

memory. The driver then has to estimate the likelihood that the hazardous event will

cause a conflict and, decide if a response is necessary (Wetton et al., 2010). A

number of decision-making and memory processes are likely to be used during this

component of hazard perception.


Some hazardous driving situations will contain highly complex stimuli and

interactions (e.g., multiple vehicles, pedestrians, road shape). Such complex traffic

scenarios and the immediacy of a potentially hazardous event will be dynamic;

requiring drivers to attend to a multitude of potential hazards and to make decisions

about their relative likelihood of causing a conflict (Horswill et al., 2008). It is likely

that psychological processes such as selective attention are required for processing

complex traffic scenarios. It has been suggested that the task of hazard perception

requires the active processing and updating of a composite representation of the

complete traffic environment the driver perceives (Horswill & McKenna, 2004;

Wetton, Hill, & Horswill, 2013). The number and interplay of psychological

processes involved in hazard perception has resulted in hazard perception being

considered a high-order cognitive task (Deery, 1999; Horswill & McKenna, 2004; L.

Jackson, Chapman, & Crundall, 2009).

Evidence that hazard perception is a high-order cognitive task is found in

dual-task studies examining hazard perception performance levels. Generally, a high-

order cognitive task shows performance reductions when a concurrent secondary task

is initiated (Baddeley, 1992; Akira Miyake et al., 2000). It has been found that

hazard perception reaction-time latencies are significantly slower (i.e., poorer

performance) and less effective when performing a concurrent secondary task (e.g.,

Horswill & McKenna, 2004; Isler, Starkey, & Williamson, 2009; McKenna & Crick,

1997; McKenna & Farrand, 1999). For instance, performing the two tasks of hazard

perception and a concurrent tracking task, results in slower reaction-time latencies

and poorer hazard perception identification (Isler et al., 2009). Such a performance

decrement would be expected, given that both tasks utilise similar psychological

processes (i.e., perception and motor control). However, hazard perception reaction-


time latencies are significantly slower while performing any concurrent verbal task,

such as random letter generation or verbal responses to stimuli (Horswill &

McKenna, 1999; McKenna & Farrand, 1999). Therefore, this evidence suggests that

hazard perception is a process that requires high-order cognitive resources.

1.7.1.3 Hazard perception and sleepiness

To date there has only been one study that has specifically examined the

effects of sleep deprivation on hazard perception performance. S. S. Smith, Horswill,

Chambers, and Wetton (2009) investigated the effects of extended wakefulness on

hazard perception performance of novice drivers and experienced drivers. It was

found that the hazard perception reaction-time latencies of novice drivers were

longer than those of experienced drivers when tested during the daytime, which is

consistent with previous studies (e.g., L. Jackson et al., 2009; McKenna & Crick,

1991). However, novice drivers displayed a more pronounced reduction in hazard

perception performance than the experienced drivers did during the middle of the

night. The results of this study indicate that younger drivers’ hazard perception

performance is more critically affected by increasing sleepiness levels.

1.7.2 Sleepiness and Younger Drivers

Younger drivers are over-represented in crashes statistics (Australian

Transport Safety Bureau, 2009; Vassallo et al., 2007). Similarly, younger drivers are

also consistently over-represented in the incidence rates of sleep-related crashes

(Connor et al., 2002; Dobbie, 2002; Pack et al., 1995). For instance, Horne and

Reyner (1995) found that approximately one third of their sample that had been

involved in a sleep-related crash were 25 years or younger. Likewise, Connor et al.

(2002) revealed that the largest proportion of drivers (34.2%) involved in fatal or

severe sleep-related crashes were aged between 15 and 24 years of age. Similarly,


Australian data reveals that 35.1% of sleep-related crashes involved young drivers

aged between 17 and 24 years (Dobbie, 2002). Thus, there appears some consensus

among the various studies that younger drivers aged 25 years or less account for

approximately one third of all sleep-related crashes. Yet, drivers aged 25 years or

less account for only 15% of the total driving population of Australia (Department of

Transport and Main Roads, 2009). As such, younger drivers are over-represented in

sleep-related crashes.

The magnitude of the level of impairment of driving performance from

sleepiness appears to be greater for younger drivers than that of older and more

experienced drivers. During periods of sleepiness, younger individuals show greater

impairment on laboratory-based tests of vigilance than older individuals (Adam,

Retey, Khatami, & Landolt, 2006; Philip et al., 2004) and when sleep deprived,

younger drivers often perform worse than older drivers on driving performance

indices (Campagne et al., 2004). At the physiological level, when younger drivers are

compared to middle aged drivers, younger drivers exhibit higher electrophysiological

(i.e., EEG) defined levels of sleepiness during early morning night-time driving

(Lowden, Anund, Kecklund, Peters, & Åkerstedt, 2009). Thus, younger individuals

appear to be critically affected by sleepiness than older individuals.

The risk perceptions that younger drivers attribute to the dangers of having a

sleep-related crash may be erroneous. For instance, S. S. Smith, Carrington, and

Trinder (2005) have shown that younger drivers frequently drive during times of the

day and night when high levels of sleepiness naturally occur, even when they

perceive their sleepiness levels to be elevated. It is likely that limited awareness of

the signs of sleepiness (e.g., Kaplan, Itoi, & Dement, 2007; Watling, Armstrong, &

Radun, 2015) and/or an under-appreciation of the dangers of a sleep-related crash


(e.g., Reyner & Horne, 1998) could contribute to younger drivers erroneous risk

perceptions. The under-appreciation aspects likely contribute to younger drivers’

dangerous practice of accumulating greater sleep debts (less sleep) before embarking

on an long distance driving trips (Philip et al., 1996). Additionally, younger drivers

will drive significantly longer distances than older drivers before stopping for a break

(Philip, Taillard, Quera-Salva, Bioulac, & Åkerstedt, 1999) if they stop for a break

(Watling, 2014). Considered together, younger drivers are at considerable risk for

sleep-related crashes.

1.8 Chapter Summary

A substantial number of fatal and nonfatal crashes are attributable to the

effects of sleepiness. Sleepiness results in a near global reduction of cortical activity

and this can lead to poorer driving performance, particularly as many of the cortical

areas that are active during the performance of many driving tasks are the same

cortical areas that show substantial reductions of activity during sleep deprivation.

Younger drivers appear to be more critically affected by sleepiness. However,

several factors (e.g., genetic, physiological, and behavioural) may contribute to

actual levels of performance. These factors will need closer examination to

understand their relative influences and interactions on sleep-wake regulation and

performance.


Chapter Two: Sleep-Wake Regulation and Performance

Think in the morning. Act in the noon. Eat in the evening. Sleep in the night.

–William Blake


This chapter provides a more detailed review of the mechanisms

underpinning wakefulness and sleep, beginning with a basic overview of the genetics

of the sleep-wake, neuroanatomical mechanisms, and the physiological

manifestations and transitions between sleep and wake states. The review also

examines the effects of sleepiness on performance measures, as well as the

distinction between different cognitive measures that require different levels of

cognitive resources for successful performance. The chapter concludes with a review

of the way natural variations in sleepiness can be modelled using biomathematical

models with an emphasis on the effects of arousal (i.e., motivation, extra effort, or

physical activity) and individual differences.

2.2 Mechanisms of Sleep-Wake

2.2.1 Genetic Mechanisms

The sleep-wake cycle is intrinsically linked to the endogenous circadian

rhythm (Franken & Dijk, 2009). The circadian rhythm is an endogenous cycle of

approximately 24-hours and is predominantly entrained by light-dark transitions

(Reppert & Weaver, 2002; Yamaguchi et al., 2003). Although, some entrainment can

occur from non-photic behaviours (Hastings, Duffield, Smith, Maywood, & Ebling,

1998; Maywood, Smith, Hall, & Hastings, 1997) such as the release of serotonin

(Glass, DiNardo, & Ehlen, 2000), feeding (Castillo, Hochstetler, Tavernier, Greene,


& Bult-Ito, 2004), and activity/exercise (Piercy & Lack, 1988). Nonetheless, at the

neuronal level, the circadian rhythm is controlled by the suprachiasmatic nuclei

(SCN: Albers, Lydic, Gander, & Moore-Ede, 1984; Gachon, Nagoshi, Brown,

Ripperger, & Schibler, 2004; Kafka, Marangos, & Moore, 1985); yet at the

molecular level the SCN are coordinated via the interactions of numerous genes

(Gachon et al., 2004; Reppert & Weaver, 2002).

A number of genes have been identified that contribute to the functioning of

the genetic mechanisms of the sleep-wake cycle. While a detailed description of the

transcription and translation of the numerous genes is far beyond the scope of this

thesis, a description of the main genes that have been identified with circadian

functioning is presented here. The core clock genes identified with the genetic

mechanism include the CLOCK, PERIOD1, PERIOD2, PERIOD3, CSNK1E, ARNTL

(BMAL1), CRY1, CRY2, and NR1D1 (Rev-Erbα) which are all involved in the

transcription and translation feedback loops that cycle through a 24-hour cycle

(Reppert & Weaver, 2002; Yamaguchi et al., 2003). Other genes that have been

documented to influence the timing of the genetic mechanisms via transcription of

the circadian clock genes include the TIMELESS, DEC11, DEC2, and the NPAS2

(Dardente, Dardente, & Cermakian, 2007; Darlington et al., 1998; Honma et al.,

2002). There are also other genes (e.g., NFATC2, SCP2, CACNA1C, POLE, FAM3D)

as well as the previously mentioned core clock genes that are involved in various of

biological functions that have been associated with various disorders of sleep-wake

and circadian functioning (Ebisawa et al., 2001; Iwase et al., 2002; Shimada et al.,

2010).

The genetic control of sleep architecture has been illustrated in twin studies,

with monozygotic twins displaying a large correspondence with transitions through


the various sleep stages (Landolt, 2008a; Zung & Wilson, 1967). Moreover,

comparisons of spectral power of NREM sleep stages determined from

electroencephalography (EEG) data between monozygotic twins produces large

intraclass correlation coefficients (.65-.93), but only moderate intraclass correlation

coefficients (.24-.60) for dizygotic twins (De Gennaro et al., 2008; Kuna et al.,

2012). These same intraclass trends for monozygotic and dizygotic twins also extend

to similarities in spectral components of the waking EEG (Ambrosius et al., 2008;

Linkowski, Kerkhofs, Hauspie, & Mendlewicz, 1991). Considered together, these

studies all highlight a large genetic component of sleep staging and

electrophysiological manifestations of sleep-wake states.

2.2.2 Neuroanatomical Mechanisms

The distinct states of sleep and wake can be characterised by differences in

the activation of cortical and subcortical brain areas and in the neurotransmitters that

subserve these areas (see Figure 2.1 over page). Wakefulness is promoted by the

ascending arousal system (Edlow et al., 2012; Saper, Chou, & Scammell, 2001). One

major input to this arousal system comes from distinct structures in the brainstem

that innervate on to the thalamus. The second input comes from other structures in

the brainstem that project to the cerebral cortex for its activation. This coordinated

ascending arousal system also inhibits cells in the hypothalamus that promote sleep

(Saper, Scammell, & Lu, 2005). The transition from wakefulness to sleep involves

the inhibition of the ascending arousal system; the inhibition activity originates from

an area of the hypothalamus (Szymusiak, Gvilia, & McGinty, 2007). Several specific

neurotransmitters are involved in wakefulness and sleep. These coordinated

inhibitory systems have been called a “flip-flop” switch due to the rapid transition

the two mutually inhibitory systems permit between wakefulness and sleep (Saper et


al., 2001). This inhibition also ensures that spontaneous switching between sleep-

wake states does not occur outside of normal sleep-wake transition times; resulting in

sleep being a predictable, once-a-day event.

Figure 2.1. Neuroanatomical projections that promote wakefulness (left) and the projections that promote sleep (right), from Saper et al. (2005).

2.2.3 The Physiological Transition from Wakefulness to Sleep: Sleepiness

The transition from wakefulness to sleep can be directly observed; however,

standard research paradigms utilise measures included in a standardised

physiological array from full polysomnography (PSG). The measures typically

included in PSG are electroencephalography (EEG), electrooculography (EOG),

electromyography (EMG), and electrocardiography (ECG); other measures can be

included but are typically reliant on the studies focus (e.g., respiration measures for

investigations of sleep-disordered breathing). EEG records the electrical activity of

the cerebral cortex; EOG detects eye movements through the difference of electrical

potential between the back of the eye (negative charge) and the front of the eye

(positive charge); EMG records electrical activity of the muscle under observation;

and ECG records the electrical activity of the heart. The recording techniques of


EEG, EOG, EMG, and ECG provide a direct assessment of the electrophysiological

processes involved during wake-sleep transition. Falling asleep generally follows a

similar pattern for all ostensibly healthy individuals; and involves a progression

through stages of wakefulness, sleepiness, and finally sleep. Table 2.1 is a summary

of this progression and includes the associated physiological activity.

Table 2.1. Summary of the Physiological Characteristics of the Progression from Wake to Sleep.

Arousal level EEG EOG EMG ECG

Awake Beta (13-30 Hz) Typical frequency and duration of blinking

High activity level

Typical heart rate

Sleepiness Eyes open

Beta + alpha (8-13 Hz)

Increased frequency and duration of eye blinks

Reduced activity

Reduced heart rate

Eyes closed Increased occurrence of alpha

Slow-rolling eye movements

Further reduced

Further reduced

Light sleep: NREM1

Theta (4-8 Hz) Approximates theta waves

Further reduced

Further reduced

Standardised sleep staging includes four periods of distinct sleep stages

which can be determined from electrophysical recordings (American Academy of

Sleep Medicine & Iber, 2007). These sleep stages include nonrapid eye movement

(NREM) 1-3 and rapid eye movement (REM), and it takes approximately 90 min to

progress through one complete cycle involving the four stages during normal night

sleep (Carskadon & Dement, 2005; Rechtschaffen & Kales, 1968). Upon falling

asleep, NREM1 sleep can be observed on electrophysical recordings and the EEG is

dominated by theta waves. Behaviourally, NREM1 sleep is reached when the eyes

are closed and the person is no longer attending to the visual environment. NREM1

is the lightest of all the sleep stages from which arousal is the easiest, with arousal


being more difficult from the progressively deeper stages of sleep (i.e., NREM2-3

and REM).

The progression from wake to sleep as described in Table 2.1, can be seen in

the EEG as a slowing of the frequency of cortical electrical activity, with a

corresponding increase in amplitude (shown in Figure 2.2a-d, over page). These

events reflect the increases in cortical synchronisation associated with sleep. It can be

seen that during wakefulness, beta waves (13-30 Hz) are the predominant frequency

of cortical electrical activity. Figure 2.2a shows beta activity and is characterised by

low amplitude of fast cycling electrical activity.

When levels of sleepiness increase, the EEG reflects this change with bursts

of alpha waves (8-13 Hz). Alpha waves are higher in amplitude with a slower

frequency of electrical activity compared to beta waves (Figure 2.2b, over page). The

EOG also reveals changes in arousal levels with more frequent eye blinks that are

slower in duration, and the ECG displays an increasing of the interbeat-interval.

Bursts of theta waves (4-8 Hz) can also occur in the waking EEG when the eyes are

open and are typically indicative of extreme sleepiness (Finelli, Baumann, Borbély,

& Achermann, 2000; Strijkstra, Beersma, Drayer, Halbesma, & Daan, 2003). A

microsleep is defined as 3-15 sec duration of theta activity in the EEG trace, which

typically occurs when the eyes are closed (Boyle et al., 2008; Moller, Kayumov,

Bulmash, Nhan, & Shapiro, 2006). Closing of the eyes when sleepy results in an

increase in alpha waves and slow-rolling eye movements in the EOG trace (Figure

2.2c, over page). Upon falling asleep theta waves (4-8 Hz) appear (Figure 2.2d, over

page). Theta waves are higher in amplitude and slower in their cycling of electrical

activity than alpha waves; scoring rules are used to determine sleep onset and the

sleep stage of NREM1 (American Academy of Sleep Medicine & Iber, 2007).


Figure 2.2. Polysomnographic recordings of the transition from alertness to falling asleep (30 sec epochs).



2.2.4 Regulation of the Sleep-Wake Cycle

It is generally accepted that the regulation of the sleep-wake cycle is

governed by two intrinsic factors: a homeostatic component and a circadian

component (Borbély, 1982; De Valck & Cluydts, 2003). The homeostatic component

describes an increasing need to sleep that gradually develops from the time of

awakening, until sleep onset where the homeostatic pressure is gradually dissipated

during NREM sleep (Borbély & Achermann, 1999; Dijk, Brunner, Beersma, &

Borbély, 1990) as displayed in Figure 2.3 top (over page). The homeostatic

component is usually described mathematically as an exponential relationship. A

reduction in sleep time leads to only a partial discharge of homeostatic pressure and

results in subsequent impairments of neurobehavioural functioning during

wakefulness (Borbély & Achermann, 1999; Dinges et al., 1997).

The exact neuroanatomical location of the homeostatic component is

unknown. Similarly, the actual neuronal mechanisms that govern the build-up of

homeostatic pressure are unknown. It has been suggested that the accumulation of

the neurotransmitter adenosine is linked to the build-up of homeostatic pressure

(Gallopin et al., 2005; Ticho & Radulovacki, 1991). Sleep deprivation leads to an

increased accumulation of adenosine with levels of adenosine decreasing during

sleep (Porkka-Heiskanen & Kalinchuk, 2011). However, the exact role that

adenosine plays is yet to be determined (Saper et al., 2005).

The circadian component represents the endogenous circadian rhythm of an

individual. The suprachiasmatic nuclei (SCN) of the hypothalamus are the circadian

master clocks that regulate the 24-hour rhythms in mammals (Albers et al., 1984;

Kafka et al., 1985). The SCN has few direct connections to sleep-wake regulator

sites. However, there are a vast number of projections to surrounding hypothalamic


cells that control a number of cyclic processes (e.g., thermoregulation, activity,

feeding, corticosteroid secretions; Saper et al., 2005). Generally, the entrainment of

the 24-hour oscillation of the SCN occurs via intricate pathways from the retina and

melatonin secretion from the pineal gland (Johnson, Moore, & Morin, 1988;

Marumoto, Murakami, Katayama, Kuroda, & Murakami, 1996). Lesions of the SCN

eliminate the daily routines of behaviour and physiological processes (i.e., sleep-

wake and feeding patterns) leading to unsynchronised disarray (Abrahamson &

Moore, 2006; Moore & Eichler, 1972).

Mathematically, the circadian component has a sinusoid function during a 24

hour period, which results in low sleep propensity during the day and the highest

sleep propensity typically during the hours of 02:00-06:00 (Richardson, Carskadon,

Orav, & Dement, 1982). Additionally, there is an ultradian fluctuation of the

circadian rhythm, with an increase in sleep propensity corresponding to 14:00-16:00,

when the descending phase begins. However, this so called “post-lunch dip” is lower

in magnitude than the early morning increase in sleep propensity (Carskadon &

Dement, 1992) as the physiological drive for sleep is greatest during the night.

An illustration of the circadian sleep propensity process can be seen in Figure

2.3 bottom (over page). Both body temperature cooling (Gilbert, van den Heuvel,

Ferguson, & Dawson, 2004; Gillberg & Åkerstedt, 1982) and melatonin expression

(a sleep promoting hormone), follow a similar temporal pattern to that depicted in

Figure 2.3 bottom, with the coolest body temperature periods and peak melatonin

expression corresponding with the circadian nadir (Cassone, 1990; Pandi-Perumal et

al., 2007). A number of hormonal, metabolic, and physiological processes are

synchronous with the circadian rhythm. These processes help regulate the sleep-wake

periods as sleep signalling systems during the descending phase of the circadian


rhythm. When the homeostatic and circadian components are combined, sleep-wake

and efficient cognitive functioning are similarly predictable (e.g., Dijk, Duffy, &

Czeisler, 1992).

Figure 2.3. Illustration of the homeostatic build-up and circadian rhythm as a function of sleep pressure. Grey bars represent periods of sleep.

As mentioned previously, the circadian rhythm has alternating periods of an

ascending and descending phase (a sinusoid function) during a 24-hour period

(Broughton, 1975; Moore & Eichler, 1972). During the descending phase increases

in sleepiness are observed as changes in physiological (Gillberg & Åkerstedt, 1982),

subjective (Dijk et al., 1992), and behaviourally indices (Dinges et al., 1997). During


the circadian nadir between 02:00-06:00 the circadian rhythm exerts it greatest

effects on sleep propensity (Richardson et al., 1982). The effects of a partial

discharge of homeostatic pressure are mediated by circadian rhythm factors. That is,

during the ascending phase of the circadian rhythm, performance and subjective

measures of sleepiness typically show only minor increases of impairment levels.

However, during the subsequent descending phase, impairments levels substantially

increase (Schmidt, Collette, Cajochen, & Peigneux, 2007).

2.3 The Effects of Sleepiness

Daytime sleepiness is typically experienced after sleep deprivation, which is

any reduction of the amount of sleep required by an individual to ensure homeostasis

or optimal neurobehavioural functioning during waking hours. A number of factors

including sleep disruption, sleep disorders, and poor sleep habits can result in sleep

deprivation. Sleep deprivation can be acute/partial (i.e., 1-4 hours reduction of sleep),

total (i.e., no sleep), or chronic (i.e., repeated long-term reductions of the optimal

amount of sleep). Additionally, having an untreated sleep disorder (Benbadis et al.,

1999; McArdle & Douglas, 2001) or being a shift worker (Di Milia, 2006; Saksvik,

Bjorvatn, Hetland, Sandal, & Pallesen, 2011; Wright Jr, Bogan, & Wyatt, 2013) can

facilitate chronic sleep deprivation via reductions of the optimal amount of sleep

obtained (i.e., sleep fragmentation) and the reductions of time spent in NREM3 and

REM sleep stages, or from circadian disruption. Experiencing sleep deprivation

results in the accumulation of a sleep debt, which is the under-accumulation of

sufficient sleep to ensure homeostasis (Lo et al., 2012; Van Dongen, Rogers, &

Dinges, 2003). Reductions in the amount of sleep leads to only a partial discharge in

the homeostatic pressure, that then results in an increased homeostatic pressure to

sleep during waking (Aeschbach et al., 2001).


Sleepiness is not only experienced during sleep deprivation, but can be a

function of the endogenous circadian rhythm. Circadian disruptions in the form of

fluctuating sleep-wake times (phase advances and retreats: (Czeisler et al., 1981; S.

S. Smith & Trinder, 2005), shiftwork (Åhsberg, Kecklund, Åkerstedt, & Francesco,

2000; Gillberg, Kecklund, Göransson, & Åkerstedt, 2003), and transmeridian travel

(Caldwell, 2001; Gundel, Drescher, Maa, Samel, & Vejvoda, 1995; Roma, Hursh,

Mead, & Nesthus, 2012) can also lead to increases in sleepiness despite maintaining

an adequate duration of sleep. The increases in sleepiness that are due to circadian

disruptions impair neurobehavioural functioning when wakefulness is enforced at the

time sleep promoting processes are scheduled to occur.

2.3.1 Cognitive Performance and Task Load

Sleepiness has a detrimental effect on a number of cognitive domains.

Specifically, sleepiness impairs alertness (Dinges, 1995; Murphy, Richard, Masaki,

& Segalowitz, 2006; Thomas et al., 2003), impairs a number of cognitive functions

(Choo, Lee, Venkatraman, Sheu, & Chee, 2005; Frey, Badia, & Wright, 2004; Mu et

al., 2005; M. E. Smith, McEvoy, & Gevins, 2002), and reduces psychomotor

performance (Dinges et al., 1997; Philip et al., 2005; S. S. Smith, Kilby, Jorgensen,

& Douglas, 2007). A meta-analysis of 143 studies conducted by Pilcher and Huffcutt

(1996) showed that sleep deprivation had a robust negative effect on general

psychological functioning. Specifically, partial sleep deprivation considerably

impairs cognitive performance (Pilcher & Huffcutt, 1996).

The effects of sleepiness may have differential effects on specific cognitive

tasks. A meta-analysis performed by Lim and Dinges (2010) examined the effects of

sleepiness on different cognitive tasks. Individual effect sizes for different cognitive

tasks were calculated, with the effects from sleepiness on simple attentional tasks


having the greatest combined effect size followed by working memory tasks. The

authors did acknowledge that the pooling of a number of working memory task

might have influenced estimates of effect size. The Lim and Dinges’ (2010) meta-

analysis demonstrated the utility of working memory tasks to detect sleepiness.

However, the sensitivity of a performance task to detect changes in sleepiness could

ultimately be a factor in the cognitive resources the task requires.

High-order cognitive tasks require the co-ordinated use of a number of

psychological processes for their successful execution. The cognitive resources

required for the successful execution of high-order task are greater than those

required for low-order tasks (Galy, Cariou, & Mélan, 2012). When cognitive

resources are limited due to global reductions from sleepiness, the coordination of

several psychological processes is arduous and results in performance impairments

(Thomas et al., 2003). For instance, M. E. Smith et al. (2002) investigated the effects

of sleep deprivation on cognitive function. They found that extended wakefulness

produced deficits on both low- and high-order tasks, but that the high-order task was

more severely affected by the increased time awake. Similarly, Nilsson et al. (2005)

found that participants who were deprived of sleep for over 30 hours had

significantly poorer performance than controls (no sleep deprivation) on a number of

performance tasks that rely on executive functioning.

A low-order attentional task that does have a high sensitivity to sleepiness is

the psychomotor vigilance task (PVT: Dinges & Powell, 1985). The PVT measures

performance levels of the attentional systems, which is quantified by reaction times.

The task requires a speedy response to the presentation of a stimulus (e.g., running

timer, simple dot on a black screen, auditory tone). The appearance of the stimulus

has a random interstimulus interval (ISI) of 2-10 sec. The PVT is a simple perceptual


task that requires very few cognitive resources to complete and consequently can be

considered a low-order task. Importantly, the PVT is sensitive to sleepiness due to

the high attentional load required to respond to the random ISI stimulus in

conjunction with the low contextual stimulation from the task (Kribbs & Dinges,

1994). The PVT has been shown to be valid behavioural measure of sleepiness of

studies of total sleep deprivation (Basner, Mollicone, & Dinges, 2011; Kaida et al.,

2006) and partial sleep deprivation (Dinges et al., 1997; 2007). The indices of the

PVT have been able to demonstrate changes in behavioural performance due to

homeostatic and circadian influences (Van Dongen et al., 2007). Considered

together, certain low- and high-order cognitive tasks appear useful as measures of

cognitive impairment due to sleepiness.

2.4 Sleepiness and Performance Outcomes

2.4.1 Predictions from Biomathematical Sleep-Wake Models

A number of biomathematical sleep-wake models have been proposed to

predict alertness and/or performance levels. The complexity of the required inputs

and the resulting outputs differ substantially from model to model (Mallis, Mejdal,

Nguyen, & Dinges, 2004). Nonetheless, nearly all models integrate the two primary

components, the intrinsic homeostatic component and the circadian component.

2.4.1.1 Two-process model of sleep-wake regulation

The two-process model of sleep-wake regulation (see Figure 2.4, over page)

was modelled as a summation of the homeostatic (process S) and the circadian

(process C) components (Borbély, 1982; Borbély & Achermann, 1992, 1999). The

time course of process S was originally derived from electrophysiological data

(EEG) with process C modelled on laboratory results (Borbély & Achermann, 1992;

Daan, Beersma, & Borbély, 1984). Values of alertness are the output metric


produced by the two-process model (Mallis et al., 2004). One issue with this model’s

prediction was that it did not fit data obtained immediately after the sleep period. The

two-process model suggests that alertness is maximal immediately after waking, but

actual data suggests that alertness is impaired during this period, a state known as

sleep inertia (Jewett et al., 1999; Lavie & Weler, 1989).

Figure 2.4. The two-process model of sleep-wake regulation, adapted from Borbély (1982).

2.4.1.2 Three-process model of sleep-wake regulation

The three-process model (Folkard & Åkerstedt, 1991) includes a homeostatic

component (process S) and a circadian component (process C) similar to the two-

process model. However, it is the inclusion of the component of sleep inertia

(process W) in its biomathematical formula that differentiates the three-process

model from the two-process model. Sleep inertia is defined as a transient state of

extreme sleepiness which is also associated with a transient period of impaired

cognitive functioning (Muzet, Nicolas, Tassi, Dewasmes, & Bonneau, 1995; Tassi &

Muzet). The effects of sleep inertia can last from several minutes for shorter nap

durations (e.g., Sallinen, Härmä, Åkerstedt, Rosa, & Lillqvist, 1998) to half an hour

for longer nap durations (e.g., Caldwell & Caldwell, 1998; Dinges, Orne,


Whitehouse, & Orne, 1987). Laboratory observations have confirmed the impairing

effect sleep inertia has on cognitive functioning (Balkin et al., 2002; Jewett et al.,

1999; Wertz, Ronda, Czeisler, & Wright, 2006). The effects of sleep inertia,

specifically the extreme sleepiness upon awakening, is in direct conflict with

predictions from the two-process model that predicts that upon awakening sleepiness

will be at its lowest (Borbély, 1982). The inclusion of process W (sleep inertia) in the

three-process model provides a more realistic prediction of the variations in

sleepiness and thus, the three-process model was designed to make specific

biomathematical predictions regarding an individual’s alertness and performance

levels. The three-process model is shown in Figure 2.5. A number of other

biomathematical models exist, with each one incorporating process S and C

(Achermann, 2004; Mallis et al., 2004).

Figure 2.5. The three-process model of sleep regulation, adapted from Åkerstedt, Connor et al.(2008). S’ is the reversal of process S beginning at sleep onset. S+C is the arithmetic sum of process S and process C. Process W is not shown.


2.4.1.3 Utility of biomathematical models

The utility of any model rests with the accuracy of its predictions when

compared to actual data. This accuracy can be assessed using statistics such as the

amount of variance accounted for by the model. Statistics for current

biomathematical models are examined below.

The accuracy of the models for predicting physiological and subjective

sleepiness is quite accurate. Åkerstedt and Folkard (1995) compared the predictions

from the three-process model to pre-existing grouped data from ambulatory EEG

recordings of highway night-time truck drivers (i.e., Kecklund & Åkerstedt, 1993)

and night-time train drivers (i.e., Torsvall & Åkerstedt, 1987). The reported results of

variance (R2) accounted for the three-process model for EEG alpha power was .8 for

the truck drivers and .94 for night-time train drivers. Predictions from the model

were also applied to data from a laboratory based study (i.e., Åkerstedt & Gillberg,

1990). The model accounted for .95 for EEG alpha power and .7 for subjective

reports of sleepiness of the variance in the data. However, model predictions could

only account for R2 = .3 of the variance of subjective sleepiness data collected from

paper mill shift workers (i.e., Åkerstedt, Axelsson, & Kecklund, 2007). These studies

show that model predictions parallel actual physiological and subjective sleepiness

data.

The relationship between model predictions and actual levels of performance

is equivocal. Correspondence between the three-process model predictions and

performance on a 30-min auditory vigilance task were R2 = .79 during a laboratory

session of sleep deprivation (Åkerstedt & Folkard, 1997). However, other

assessments of biomathematical modelling and actual performance differ,

particularly outside of the laboratory. For instance, assessment of performance levels


on the PVT (reaction times, reciprocal reaction times, and lapses) by flight attendants

during working periods were compared to predicted performance levels with

associations in the order of R2 = .75, .05, and .49 respectively (Roma et al., 2012).

The comparisons of pre- and post-work periods on PVT indices were varied as well

(R2 = .03-.89). It was noted that the predicted physiological and subjective sleepiness

from biomathematical models aligned with the obtained data well in the Roma et al.

(2012) study. However, the correspondence between predicted performance levels

and actual data are less accurate, particularly data collected outside the laboratory.

2.4.1.4 Critical aspects of sleep-wake models

The studies presented above suggest that in a laboratory environment or

during the high sleepiness periods while performing an uneventful procedure (night-

time driving) the predications from the models are consistent with physiological data

and slightly less so with subjective data. When model predictions are compared to

data that is collected from more dynamic work situations the utility of the predictions

decreases. The effects of dynamic versus non-dynamic situations and the resulting

effects on arousal levels have already been noted in section 2.4.1.3. Briefly,

contextual situations or actions that invoke higher arousal levels (e.g., social

interactions, physical movements) lead to reductions of sleepiness (Åkerstedt,

Kecklund, & Axelsson, 2008; Dijkman et al., 1997; Horne & Foster, 1996; Johns,

2002). That is, situations where sleep drive is high and arousal levels are similarly

high are likely to result in conflicting predictions from the biomathematical models.

Biomathematical models have limited capacity to account for variations in arousal

levels that can occur, because they only focus on endogenous arousal levels (process

S and C) of the individual, not the interaction between the individual and the context

in which they are engaged.


The heterogeneity of cognitive processes used for differing performance tasks

is potentially another aspect for which biomathematical models cannot account. That

is, the predictions of a model predicated on one type of performance measure may

not cross-translate to accurate predictions of another performance measure; given the

varied cognitive processes utilised between differing measures (Dinges &

Achermann, 1999; M. L. Jackson et al., 2013). Notwithstanding that many cognitive

process have a certain level of correspondence (Collette, Hogge, Salmon, & Van der

Linden, 2006; A. Miyake et al., 2000); evidence suggests that different cognitive

processes are differentially affected by sleepiness (Tucker, Whitney, Belenky,

Hinson, & Van Dongen, 2010; Williamson, Feyer, Mattick, Friswell, & Finlay-

Brown, 2001).

Another issue for biomathematical models such as the two- and three-process

models and others [the sleep, activity, fatigue, and task effectiveness model (Hursh et

al., 2004), circadian alertness simulator (Moore-Ede et al., 2004)] is their modelling

using grouped data and accordingly predict a “grouped response” (Dinges, 2004;

Mallis et al., 2004). However, individual differences in response to sleep and wake

variations can be pronounced; approximately 70% of the total variance of PVT

performance between individuals can be accounted for by interindividual differences

(Van Dongen et al., 2004). This suggests that the mechanism of individual variation

in task performance is stable. Currently, individual differences in performance are

accounted for with post-hoc analysis using mixed modelling by including a random

effect that aligns with the inter-individual differences (Olofsen, Dinges, & Van

Dongen, 2004; Van Dongen et al., 2007).

In Summary, the two primary constructs of process S and process C have

proved to be a rich platform for the examination of sleep-wake and performance.


These factors have been shown to be associated with physiological, subjective, and

performance outcomes. However, the strength of the relationships has varied

depending on the contextual situations and individual level factors. It is likely that

other factors can be accounted for by sleep-wake models and may be reflected in

performance levels.

2.4.1.5 Four-process model of sleep-wake regulation

Despite the predictive utility of process S and C, the mechanisms behind

sleep-wake and neurobehavioural performance are not fully understood, and there are

a number of potential mechanisms that could alter sleep-wake and performance

levels. Included among these is the notion of an opponent process to sleep, an arousal

component or a wake drive (Bonnet & Arand, 2001; Edgar, Dement, & Fuller, 1993).

One major limitation of models that are solely based on process S and C is that they

fail to explicitly account for contextual or transient arousal effects (e.g., motivation,

movement, and mental stimulation) that can increase wakefulness (Bonnet & Arand,

1999; Bonnet & Arand, 2000, 2001; Johns, 1998).

An expanded model of sleep and wakefulness, termed the four-process

model, was conceptualised by Johns (1993). Unlike the two-process model that is

entirely reliant upon factors that compel the individual towards sleep (i.e., sleep

drive); the four-process model includes an arousal or wakefulness component (i.e.,

wake drive). It is suggested that whether an individual is asleep or awake is

dependent upon the comparative strengths of the sleep and wake drive systems. That

is, the two drives are mutually inhibitory systems with the drive with the greatest

overall strength ultimately deciding the state of consciousness (Johns, 1998). The

support for this model has increased with developments of the flip-flop

neuroanatomical model of sleep-wake switching (Saper et al., 2001; Saper et al.,


2005). It is also suggested that the sleep and wake drive systems each have primary

and secondary components (Johns, 1998). The primary components originate within

the central nervous system in different neuronal clusters and conceivable could be

influenced by genetic components. In contrast, the secondary components are linked

to homeostatic processes and are influenced by behavioural factors. Figure 2.6 shows

the four-process model.

Figure 2.6. The four-process model of sleep-wake regulation, adapted from Johns (1998). The state of sleep or wake is determined by the drive with the greatest strength. When the strength of a drive is greater than the other drive, the systems “tips” towards that state of consciousness; similar to the neuroanatomical flip-flop switch of Saper et al., (2005).

The primary sleep drive system is thought to be governed by intrinsic activity

from various neuronal centres located within the central nervous system. The relative

strength of the primary sleep drive varies with the endogenous circadian rhythm. The

secondary sleep drive system is the equivalent of process S from the two-process

model (Johns, 1998). Similarly, the relative strength of the secondary sleep drive

system gradually increases with time awake and is discharged progressively after

non-REM sleep onset. These aspects of the four-process model are modelled from

the earlier two-process model.


The primary wake drive is proposed to have the inverse timing of the

endogenous circadian rhythm, and it is thought of as the equivalent of process C

from the two-process model. The secondary wake drive is proposed to be the

outcome from neuronal impulses from the afferent nerve system (i.e., postural

muscles and movement/activity), inputs from the visual system, and mental

stimulation (Johns, 1993; Johns, 1998). As such, the secondary wake drive is

believed to be partially under voluntary control by the individual. This is potentially

a key aspect of the sleep-wake model proposed by Johns (1998) as it addresses some

of the specific deficiencies identified in well-established biomathematical models

and obtained research findings. However, there has been no direct empirical

validation or testing of the model. Examinations of this model could bridge the

discrepancy between endogenous arousal factors (process S and C) and contextual or

individually based arousal factors.

2.4.1.5.1 Evidence for the wake drive

The notion of a wake drive informs the re-evaluations of existing data. One

aspect of the wake drive, being increased activity or physical movements has been

examined for their effects on sleep-wake durations. The Multiple Sleep Latency Test

(MSLT; Carskadon & Dement, 1982; Carskadon et al., 1986) that measures the

duration taken to fall asleep and the Maintenance of Wakefulness Test (MWT;

Mitler, Gujavarty, & Browman, 1982) that measures the ability to stay awake are

both influenced by physical activity. Various levels of physical activity have shown

to increase sleep onset latencies for both the MSLT and MWT (Bonnet & Arand,

1998; 1999) suggesting a promotion of wakefulness by activity. The level of activity

seems to mediate the duration of wake promoting effects (Bonnet & Arand, 2005b;

Horne & Foster, 1996; Sallinen et al., 2008).


Activity also has an effect on physiologically and subjectively derived

sleepiness levels and performance measures. The effects of standing when compared

to sitting, attenuates EEG signs of sleepiness and reduces performance impairment

on the PVT during a night of sleep deprivation (Caldwell, Prazinko, & Hall, 2000).

Activity also increases subjective sleepiness levels, leading to incongruent

associations with EEG recordings (Gillberg et al., 2003; Watling, 2012). These

incongruences are minimised by eliciting subjective ratings after a “calm-down”

period entailing being seated with closed eyes for 1-min (Yang, Lin, & Spielman,

2004). In effect, the ‘calm-down’ period could be considered a period where the

activation of the wake drive reduces as seen in the Caldwell et al. (2000) study.

These findings lend partial support for the aspect of the secondary wake drive from

the four-process model.

Mental stimulation may also enhance wakefulness (Åkerstedt, Kecklund, et

al., 2008; Johns, 1998). Mental stimulation may encompass a number of

psychological facets; however, the dimension of motivation is a potentially salient

factor. When motivation of task engagement levels are recorded during a night of

sleep deprivation, motivation levels decrease along with sleepiness and performance

indices (J. T. Hull, Wright, & Czeisler, 2003; Ikegami et al., 2009). Moreover,

intrinsic motivation can be modified, receiving feedback associated with

performance levels can augment performance level (Steyvers & Gaillard, 1993;

Wilkinson, 1961). An increase in the strength of the wake drive could also occur

when a certain level of performance is requested or required from following either

task instructions (Bonnet & Arand, 2005a) or from momentary incentives (Hsieh, Li,

& Tsai, 2010).


The four-process model can guide predictions regarding sleep-wake and

performance levels. The effects of heightened sleep drive on performance are

consistent with predictions of current sleep-wake models. However, the magnitude of

effect of the wake drive is largely unknown. The following section will examine

aspects of the wake drive, specifically the effects of motivation on sleep-wake and

performance levels. Table 2.2 (over page) lists the studies that have incorporated

motivation into their study methodology.

These data show that manipulation of motivation levels with task

performance can have an effect on task performance levels and the associated

physiological indices. Nonetheless, several aspects of the effects of motivation are

unanswered— particularly the duration of effects. That s, the length of the effect that

motivation can have on the type of cognitive performance (i.e., low- vs. high-order

tasks) is unresolved. Horne and Pettitt (1985) demonstrated that motivation

facilitated a complete restoration of performance during a 30-min low-order task

when sleep deprived. However, as shown in Hsieh, Li, and Tsai’s (2010) study, the

effects of motivation on sleep-deprived individuals were not enough to attenuate the

performance deficits for a high-order task. The equivocal performance effects of

motivation on cognitive tasks may potentially be an issue of cognitive resources, as

high-order tasks seemingly require more cognitive resources to improve levels of

performance (Galy et al., 2012; Hsieh et al., 2010). Thus, a within-subjects

examination of both low- and high-order cognitive tasks when sleep deprived and

when fully rested during the same testing situations is seemingly required.

While the effect of motivation on performance levels is equivocal,

physiological indices suggest that an orientation of attention towards improving

performance occurs. The studies of Boksem, Meijman, and Lorist (2006) and Hsieh,


et al. (2010) both clearly show effects of motivation on an increase in the amplitude

of the error related negativity (ERN) which is an error-related event potential

component. Increases in the ERN component are indicative of heightened error

detection processes (Gehring, Goss, Coles, & Meyer, 1993; Scheffers, Coles,

Bernstein, Gehring, & Donchin, 1996). These increases in the amplitude of the ERN

are also associated with a slight improvement of task performance accuracy (Boksem

et al., 2006; Hsieh et al., 2010). Thus, the effect of motivation on one type

physiological index is evident. However, it is unknown if the effect from motivation

extends to other forms of physiological indices of sleepiness, such as EEG defined

sleepiness from spectral analyses.


Table 2.2. Studies that have Researched the Effect of Motivation on Sleep-Wake Behaviours and Performance Outcomes.

Study Participants Measures Design & intervention Outcome Effect size (Cohen’s d)

Hartse, Roth, and Zorick (1982)

Ostensibly healthy individuals

MSLT: SOL • Within-subjects design • Condition 1: “try to fall asleep” • Condition 2: “try to stay awake”

• SOL was significantly longer for condition 2 • Condition 1 progressed more rapidly to deeper

stages of sleep

• 0.37 • 0.53

Bonnet and Arand (2005a)


MSLT and MWT: SOL

• Within-subjects design • Baseline: standard instructions • Condition 1: “fall asleep rapidly” • Condition 2: “try to stay awake”

• Condition 2 had longer SOL for MSLT than baseline measures

• Condition 1 had shorter SOL for MWT than at baseline

• 0.63

• 0.44

Shreter, Peled, and Pillar (2006)

Excessively sleepy individuals

MWT: SOL • Cross-sectional design • Motivated to stay awake during

MWT

• 91% of participants were able to remain awake • SOL results were well above the normal cut-offs • The authors argued that participant’s motivations to

keep their licenses could account for the results

NA

Horne and Pettitt (1985)


WAVT (low-order) over two nights of complete sleep deprivation

• Between-subjects design • Group 1: “sleep deprivation -

incentive” • Group 2: “sleep deprivation - no

incentive” • Controls: “no sleep deprivation, no

incentive”

• Group 2 performed significantly worse over the two nights of complete sleep deprivation

• Group 1 actually performed better than baseline results and not significantly different to Controls

• Group 1 performance eventually faded during the second night of testing but was still superior than Group 2 performance

• 0.65

Boksem, Meijman, and Lorist (2006)


Two hour high-order task (speed and accuracy) responses with concurrent ERP recordings

• Cross-sectional design • Congruent responses (stimuli and

response matched) • Incongruent responses (stimuli and

response did not match) • Motivation: best performer of last 20

min received extra payment

• Motivation period produced either a increase in speed or accuracy of responses, but could not completely attenuate time-on-task effects

• Speed responders: had higher physiological anticipatory and preparatory processes

• Accuracy responders: had a higher physiological error detection process

• 0.73: speed • 0.52: accuracy

• 0.81

• 0.69

Hsieh, Li, and Tsai (2010)


Flanker task (high-order) after a regular nights’ sleep and no sleep

• Within-subjects, counterbalanced design

• Factor 1: Incentive (incentive/no-incentive)

• Factor 2: Sleep (sleep/no-sleep)

• Main effect for sleep: impaired performance for reaction time and response %

• Main effect for incentive: higher response % only • No interactive effects

Higher physiological error detection process for incentives

• 0.47

• 0.47

Note: SOL = sleep onset latency; WAVT = Wilkinson Auditory Vigilance Test - a discrimination test between a target tone and other tones with various pitches.



2.4.2 Individual Differences, Sleepiness and Performance

The absolute amount of sleep deprivation required to produce equivalent

neurobehavioural deficits can vary between individuals (Frey et al., 2004; Van

Dongen, Rogers, et al., 2003). This individual variation is likely to be stable. For

instance, Rupp, Wesensten, and Balkin (2012) have shown that the intraclass

correlation coefficients for performance outcomes on low- and high-order cognitive

tasks were all greater than .86 between a night of total sleep deprivation and a week

of partial sleep deprivation. Similarly, two sessions of sleep deprivation separated by

a number of weeks, produced performance deficits that were highly replicable for a

given individual (Leproult et al., 2003). These findings suggest that the

neurobehavioural response to sleep deprivation is trait-like.

2.4.2.1 Genetic aspects

Individual differences have a number of sources that can potentially

contribute to sleep-wake and performance. These can include endogenous factors

such as task engagement, personality, stress and coping styles, and IQ. While

research has demonstrated that these endogenous factors can affects sleep-wake and

performance, these endogenous factors have not been found to have an entirely stable

or replicable effect on sleep-wake and task performance outcomes (Blagrove &

Akehurst, 2001; Sadeh, Keinan, & Daon, 2004; Soehner, Kennedy, & Monk, 2007;

Waters, Adams, Binks, & Varnado, 1993; Yamamoto, Tanaka, Motoko, Yamazaki,

& Sirakawa, 2000). However, genetic variation is emerging as one source of

individual variability regarding vulnerability to sleepiness, and, as previously

discussed in Section 2.2.1, circadian functioning, is largely coordinated via genetic

mechanisms.


Single nucleotide polymorphisms (SNPs) are the most common type of

genetic variation among individuals; SNPs determine our unique physical features

and contribute to our behavioural characteristics (Brookes, 1999; Butler, 2012).

Several genes and their polymorphisms have associations with specific sleep

disorders (Archer et al., 2003; Dauvilliers & Tafti, 2006; Mignot, Young, Lin, &

Finn, 1999) and with aspects of sleep-wake and performance outcomes (Allebrandt

et al., 2010; Jones et al., 2007). A number of specific genes (e.g., PERIOD1/2/3,

CLOCK, CKI∑, BMAL1) that are expressed in the SCN have been found to

coordinate circadian functioning (Dunlap, 1999; Lowrey & Takahashi, 2011), and

thus these circadian genes offer a number of candidate genes to examine their

association with sleep-wake and performance outcomes.

A number of studies have examined the PERIOD gene family and

particularly the PERIOD3 gene. The PERIOD3 gene has a number of different facets

associated with sleep-wake behaviours and performance outcomes. PERIOD3 is

located on chromosome 1p36.23 (Zylka, Shearman, Weaver, & Reppert, 1998) and

contains a variable-number tandem-repeat (VNTR) polymorphism coding for 18

amino acids on exon 18. This VNTR polymorphism repeats either four (PERIOD34/4)

or five (PERIOD34/5, PERIOD35/5) times (Ebisawa et al., 2001). The longer VNTR

polymorphism of PERIOD35/5 has been associated with diurnal preference of

morningness, with the shorter allele PERIOD34/4 having been associated with

eveningness and delayed sleep phase syndrome (Archer et al., 2003; Ebisawa et al.,

2001).

A number of differences between the VNTR polymorphisms of the PERIOD3

genotypes have also been found regarding physiological, subjective, and

performance measures. A consistent finding is the effect of genotype on individual


response to total sleep deprivation. Specifically, most differences in performance

measures between individuals with long and short PERIOD3 genotypes seem to

present after the circadian nadir when the peak melatonin expression occurs (Groeger

et al., 2008; Lo et al., 2012; Viola et al., 2007). However, it is noted that the effect of

the different genotypes on individual response to lower levels of sleep pressure (i.e.,

acute and chronic partial sleep deprivation) is equivocal.

While research focus has predominantly focused on the PERIOD3 gene, a

number of studies have identified a role the ADORA2A gene (SNP rs5751876)

performs with electrophysiological expression of sleepiness as well as levels of

performance. Previous research that has shown that individuals with the TT genotype

display higher levels of EEG defined sleepiness in the range of 7-10 Hz during sleep

as well as wakefulness than the other genotypes of the ADORA2A gene (Rétey et al.,

2005; Rétey et al., 2007). Moreover, in the Rétey et al. (2007) study, TT genotypes

were also more sensitive to the effects of caffeine, which suggests that the different

genotypes could have differing strength of arousal mechanisms (i.e., wake drive).

Performance differences between the genotypes of the ADORA2A gene have also

been noted in the extant literature. Individuals with the TT genotype display longer

reaction-time latencies and more PVT lapses than participants with the CT genotype

(Bodenmann et al., 2012; Rupp, Wesensten, Newman, & Balkin, 2013).

Other genes with an identified circadian function such as CLOCK,

TIMELESS, ARNTL, and NPAS2 genes and have also been associated with specific

sleep disorders (e.g., narcolepsy, insomnia) and affective disorders that can also

present with significant issues relating to sleep duration and timing such as seasonal

affective disorder, depression, and bipolar disorder (Kim et al., 2015; J. J. Liu et al.,

2015; Utge et al., 2010). Moreover, these genes of circadian functioning are also


associated with aspects of diurnal preference and sleep structure (Archer, Viola,

Kyriakopoulou, von Schantz, & Dijk, 2008; Garaulet et al., 2011; Viola et al., 2007).

Moreover, several studies have demonstrated that diurnal preference is associated

with differential levels of sleepiness and performance across the day for morning

versus evening chronotypes (Correa, Molina, & Sanabria, 2014; Goldstein, Hahn,

Hasher, Wiprzycka, & Zelazo, 2007; Lara, Madrid, & Correa, 2014) and this

differential performance could in part, be due to genetic factors. It is acknowledged

that while research that focuses on the genetic mechanisms related to sleep-wake

behaviours and their disorders is ongoing; little research has been conducted that

examines known circadian genes and their associations with physiological and

subjective sleepiness levels as well as their association with performance measures.

While the extant research has predominantly focused on genes with known

circadian functioning such as the PERIOD, CLOCK, and TIMELESS genes, several

other genes that do not have specific circadian function are also related to individual

variations of physiological, subjective, and performance outcomes. For instance, the

catechol-O-methyltransferase (COMT) is a gene that has no identified circadian

function; however, it has been found to modulate arousal levels via modulation of

dopamine levels in frontal and prefrontal cortices (Tunbridge, Bannerman, Sharp, &

Harrison, 2004). Individuals with the Val genotype have been found to display lower

levels of EEG defined sleepiness (i.e., alpha waves 8-13 Hz) when sleep deprived

(Bodenmann et al., 2009) and correspondingly, individuals with the alternate Met

allele display poorer performance on vigilance tasks such as the PVT (Lim et al.,

2012). Another non-circadian gene, the human leukocyte antigen DQB1*0602 has

also demonstrated individual variations regarding wake and sleep electrophysiology

(Goel, Banks, Mignot, & Dinges, 2010; Mignot et al., 1999); however, these


differences purportedly do not extend to differences in cognitive performance (Goel

et al., 2010).

An increasing number of studies are demonstrating the considerable influence

individual differences can have on the variations of sleep-wake and performance

outcomes. Moreover, these interindividual differences have been documented to

account for a substantial proportion of the variance on sleep-wake and performance

outcomes (Friedman et al., 2008; Leproult et al., 2003; Van Dongen et al., 2004; Van

Dongen, Rogers, et al., 2003). The cited literature demonstrates that genetic

variations can account for differences regarding sleep-wake indices and performance

impairment from sleepiness. Additionally, taking into consideration that circadian

functioning is largely influenced by genetic components (Gachon et al., 2004;

Reppert & Weaver, 2002), exploring the effects genetic variations and specifically,

circadian gene variations can represent has the potential to further our understanding

of individual differences in relation to sleep-wake and performance outcomes.

2.5 Aims, Rationale of the Studies, and Significance of the Research Program

The aim of the current research program is to extend knowledge in this field

by examining the influence of motivation and individual differences on sleepiness

and performance. Current sleep-wake models rely on homeostatic and circadian

components for their predictions of sleepiness and performance levels. These models

have been empirically validated and are predictive of performance. However, these

monotonic models cannot account for the effects of motivation and individual

differences; such effects have historically been ignored or treated as error variance.

The reviewed literature suggests increases in the strength of the wake drive (via

motivations and/or individual variations) can affect levels of sleepiness and

performance outcomes. However, the magnitude, the duration, and, in particular, the


effects on low- and high-order cognitive tasks have not been comprehensively

assessed. Additionally, individual differences relating to the effects of motivation

also need to be assessed. The research program will examine the effects of these two

factors in a systematic fashion. The following research questions (RQ) are proposed:

RQ1. What are the magnitude and duration of the effects of motivation on

sleepiness and performance levels?

RQ2. What effects do individual differences, assessed as genetic variations

have on sleepiness and performance?

The major focus of this research program is the effects of motivation on

sleepiness and performance levels with the examination of genetic aspects the

secondary focus. As such, each of the three studies has included some form of

manipulation of motivation. Regarding the genetic component of the research

program, only Study One was appropriate to run a genetic analysis. This is due to

considerations of the Hardy-Weinberg principle, which is discussed below.

2.5.1 Rationale for Study One

As noted previously, sleep-wake functioning has a number of genetic

mechanisms, with several sleep disorders having genetic underpinnings. Moreover, a

number of consistent individual differences, in terms of performance impairments

from sleepiness are well documented in the extant literature. Some of these

performance impairments have been associated with specific genes (e.g., PERIOD3,

COMT). Therefore, examining a set of genes that have known associations with

circadian function and effects from sleepiness can improve our understanding of

sleepiness and performance.

Study One examined the physiological, subjective, performance, and genetic

associations of sleepiness in young adults and will mainly explore research question


two. This study examined a set of genetic markers (see Appendix A) with known

associations with circadian function and trait sleepiness to determine their

association with physiological, subjective, and performance indices of sleepiness.

The extant literature has predominantly focused on the long and short VNTR

polymorphisms of the PERIOD3 gene, as well as the ADORA2A, COMT, and the

human leukocyte antigen DQB1*0602 and their associations with sleepiness and

cognitive performance during periods of extended wakefulness or sleep deprivation

(e.g., Groeger et al., 2008; Vandewalle et al., 2009; Viola et al., 2007). However,

there are numerous genes involved in circadian timing and aspects of sleep-wake

functions that have yet to be examined.

Several studies in the sleep field have not adhered to an important principle of

genetic association studies, being the Hardy-Weinberg principle or Hardy-Weinberg

equilibrium (HWE). The HWE states the genotype frequencies should remain

constant from generation to generation in a population as long as no reduction,

mutation, or migration occurs (Guo & Thompson, 1992; Salanti, Amountza, Ntzani,

& Ioannidis, 2005). Practically, it has been documented that genetic studies that do

not maintain HWE find erroneous associations (Xu, Turner, Little, Bleecker, &

Meyers, 2002), and these associations have failed to remain once statistical

corrections to the data were made (Trikalinos, Salanti, Khoury, & Ioannidis, 2006).

Thus, maintaining HWE is an important principle to adhere to with genetic

association studies, as it is a basis for genetic inference (Guo & Thompson, 1992;

Trikalinos et al., 2006). Consequently, large sample sizes are needed for genetic

association studies. Therefore, in the present research program, only Study One will

include a genetic component in its study design.


Study One includes a manipulation of the wake drive and therefore, also

address research question one. A motivational component was included in this study

where participants were instructed to “better your previous performance by

motivating yourself to try even harder”. Goel et al., (2009), performed the only study

that has examined the associations of the PERIOD3 gene, motivations and

performance on the PVT. However, in this study, there was no manipulation of

motivation levels and only the VNTR polymorphisms of the PERIOD3 genome was

examined. As such, Study One can extend our understanding of the effects of

motivation and specific genes associated with circadian functioning and trait

sleepiness for their association with sleepiness and performance. Identification of

SNPs associated with physiological aspects of sleep and motivational differences

may be informative as to why some individuals are at an increased risk for sleep-

related crashes.

2.5.2 Rationale for Study Two

Study Two was a pilot study based on the reanalysis of an existing dataset.

The data was originally collected to determine whether partially sleep-deprived

drivers could choose to cease driving while completing a hazard perception driving

task (see S. S. Smith, Watling, Horswill, & Douglas, 2011). Participants were

instructed to “Stop when you think you would be too sleepy to drive safely on the

road” (see Section 5.3.4 for greater detail of the procedure). This study informed the

first research question regarding the effects motivation has on sleepiness and

performance levels.

Typically driving is a task-oriented activity of proceeding from point A to

point B. The instruction that was used in Study Two was “Stop when you think you

would be too sleepy to drive safely on the road” which can be considered a


manipulation the wake drive. That is, the instruction can be viewed as removing the

need for participants to maintain a certain demand on their wake drive to remain

awake to perform the driving task. For instance, when participants are instructed to

either remain awake or fall asleep during assessments with the Multiple Sleep

Latency Test or Maintenance of Wakefulness Test, their subsequent sleep latencies

are significantly altered (e.g., Bonnet & Arand, 2005a). Such that, that instruction of

“remain awake” can be considered an instruction that requires an increase of the

wake drive; whereas, the instruction “fall asleep” can be considered an instruction

that necessitates the reduction of the wake drive. This logic used by Bonnet and

Arand (2005a) was incorporated into the instruction given to participants in Study

Two.

When sleep deprived participants are instructed to complete a driving task, it

is often found that they will apply extra effort to remain awake during periods when

they are “fighting” sleep onset due to their sleep deprivation and the lowered arousal

levels from tedious driving scenarios (e.g., Reyner & Horne, 1998; Thiffault &

Bergeron, 2003). In essence, when a participant’s motivation or task instruction is to

complete a driving task, this can lead to increased motivation to remain awake by

fighting sleep onset until the task is completed. A number of studies have used this

type of paradigm to examine driving performance during periods of extreme

sleepiness (e.g., complete sleep deprivation or extended wakefulness), which is

associated with increases in physiological and subjective sleepiness, as well as

impaired driving performance (e.g., Anund, Kecklund, Peters, & Åkerstedt, 2008;

Arnedt, Geddes, & MacLean, 2005; Lowden et al., 2009). However, little is known

about drivers’ ability to monitor their sleepiness levels and then choose to cease

driving before they reach such extreme levels of sleepiness. Furthermore, the


changes in physiological and subjective sleepiness when a participant does not have

to fight sleepiness for an extended period have not been examined. As such, Study

Two can extend our understanding of the changes in physiological and subjective

sleepiness associated with cessation of a hazard perception task when the motivation

is to stop driving when too sleepy. This study has important implications as to an

individual’s driving preconceptions and motivations.

2.5.3 Rationale for Study Three

The rational for Study Three is the direct experimental manipulation of the

wake drive via manipulations of motivational levels. This study informed the first

research question regarding the magnitude and duration of the effects motivation has

on sleepiness and performance levels.

In the context of road safety, an individual’s motivation for continuing to

drive while sleepy is potentially an important factor. Indeed, previous research

suggests drivers cite factors associated with destination arrival (i.e., time pressures,

close to destination) as reasons for continuing to drive while sleepy (Armstrong et

al., 2010; Nordbakke & Sagberg, 2007). Moreover, motivation to reach a destination

has been associated with truck drivers’ self-reported instances of falling asleep at the

wheel (McCartt, Rohrbaugh, Hammer, & Fuller, 2000). Consequently, a remarkable

proportion of drivers (69-73%) report that they would continue to drive even when

aware of increasing levels of sleepiness (Nordbakke & Sagberg, 2007; Watling,

Armstrong, & Radun, 2015); even though driving while sleepy is acknowledged as a

risky driving behaviour, motivation to continue driving while sleepy is a pertinent

factor for performing sleepy driving behaviours (Armstrong et al., 2010; Fletcher,

McCulloch, Baulk, & Dawson, 2005; Nordbakke & Sagberg, 2007; Watling,

Armstrong, Obst, & Smith, 2014).


While several survey studies have quantified self-reported aspects of drivers’

motivations to drive while sleepy, there appears to be lack of studies that examine the

actual psychophysiological and performance consequences of motivating oneself to

apply extra effort to the task of driving when sleepy. The four-process model of

sleep-wake suggests that visual stimulation, movement, and mental

stimulation/motivation can increase the strength of the wake drive (Johns, 1993).

Indeed, postural movements while driving are associated with decreases in sleepiness

(Rogé, Pebayle, & Muzet, 2001) and increased visual stimulation in the form to more

visually dynamic or non-monotonous road scenery is associated with increased levels

of arousal (Larue, Rakotonirainy, & Pettitt, 2011; Thiffault & Bergeron, 2003); these

findings are consistent with predictions from the four-process model. Yet, no data

exists that has examined the effect of motivation on sleepiness and performance in

the context of driving. Therefore, Study Three can extend our understanding of the

effects of motivation in the context of driving including the effects of motivation on

levels of sleepiness and performance.

2.5.4 Significance of the Research Program

The significance of the research program is based on a number of aspects.

First, the three studies are very much exploratory in nature - that is, some literature

exists to inform the research questions but it is limited in its scope to provide specific

directions for the research program. The exploratory nature of Study One can extend

our understanding of the effects of motivation and specific genes associated with

circadian functioning and trait sleepiness for their association with sleepiness and

performance. In particular, many of the SNPs that were chosen for examination have

only been examined for their associations with sleep disorders and sleep-wake

timing. Thus, examining these SNPs in relation to their associations with sleepiness


and performance levels is an important undertaking and furthers our understanding of

the genetic associations with sleepiness and performance outcomes.

Studies Two and Three utilised an experimental methodology with sleepiness

and driving performance outcomes. The significance of Study Two rests with the

examination of several physiological indices and there relative change during a

simulated driving task when participants have to cease driving before they get too

sleepy to drive safely. This study has important implications not only for an

individual’s driving awareness of sleepiness but also for their motivations. Study

Three is perhaps the most significant study pertaining to the effect of motivation, as

this study experimentally manipulates the wake and sleep drive. Additionally, the

third study extends on the findings from Study One by including a low- and high-

order cognitive task in its design. Thus, this study systematically explores the effects

of motivation on sleepiness and performance when alert and when sleepy on both

low- and high-order cognitive tasks.


Chapter Three: Measurement of Sleepiness and Performance

I always know how sleepy I am… it usually depends on how boring the social

situation is I fall asleep in.

–Unknown author


This chapter presents the measures used in the three studies. Several

measures have been used in previous research to assess levels of sleepiness and

performance; these can be broadly grouped into three general categories:

physiological, subjective, and performance as well as behavioural measures. Each of

these measures is described, and their use in the research program is justified.

3.1 Physiological Measures

The physiological measures provide a direct measurement of a particular

aspect of physiology. The research program used three types of physiological

measures. They are electroencephalography, electrooculography, and

electrocardiography.

3.1.1 Electroencephalography

Electroencephalography (EEG) is the recording of electrical activity from the

scalp; the electrical activity is believed to be the summation of post synaptic

potentials (Sanei & Chambers, 2008). The analysis of electroencephalographic

recordings can be done by visual scoring of the wave traces or by the application of

spectral analyses. Visual scoring is time consuming and the human eye can fail to

notice the subtle changes that can occur in the wavelets. Alternatively, spectral

analysis can accurately categorise an entire EEG recording. Spectral analyses utilise


signal processing (typically fast-Fourier transform) where the signal is transformed

from the time domain to the frequency domain for a specified duration of time (or

epoch). The absolute power of the frequency bands is calculated as the areas under

the curve for each specified frequency band. The fundamental frequency bands for

EEG spectral analyses are delta (0.5-4.00 Hz), theta (4.01-8.00 Hz), alpha (8.01-

13.00 Hz), and beta (13.00-30.00 Hz). The power of the specified frequency bands,

measured in microvolts squared (μV2) is then determined for each epoch, which can

typically range from 2-30 sec in duration.

Studies that have examined the test-retest stability of EEG recordings have

demonstrated a high level of reliability. Gasser, Bächer, and Steinberg (1985)

assessed test-retest scores of EEG recordings of 3-min duration that were subjected

to spectral analysis. Significant and large correlations were found for C3 and C4

recording sites for the theta power band (3.5-7.5 Hz) r = .79 and .71, for the alpha1

power band (7.5-9.5 Hz) r = .81 and .74, alpha 2 power band (9.5-12.5 Hz) r = .77

and .74, for the beta 1 power band (12.5-17.5 Hz) r = .63 and .64, and for the beta 2

power band (17.5-25.0 Hz) r = .73 and .70 over the two testing sessions. Lal and

Craig (2005) assessed the reproducibility of the magnitude of EEG activity during a

tedious simulated driving task. In this study significant correlations were found

between the two testing sessions in the order of r = .76 for beta activity, r = .81 for

alpha activity, r = .99 for theta activity, and r = .97 for delta activity.

The validity of EEG recordings to detect increases in sleepiness is adequate.

Changes in the absolute power in the EEG theta and alpha bands reflect changes in

cortical arousal levels. That is, increases in sleepiness are observed as increases in

the power of the EEG theta and alpha bands across time (Akerstedt, Kecklund, &

Knutsson, 1991; Gorgoni et al., 2014; Kecklund & Åkerstedt, 1993). In contrast,


decreases in sleepiness that occur after awakening from a nap, correspond to

reductions in the theta and alpha frequency bands (Hayashi, Ito, & Hori, 1999; Horne

& Reyner, 1996; Watling, Smith, & Horswill, 2014). Several studies have shown that

EEG recordings reflect variations in sleepiness as concurrently assessed by

subjective measures and performance tasks (Åkerstedt & Gillberg, 1990; Kaida et al.,

2006), including simulated driving tasks (Campagne et al., 2004; Gillberg et al.,

1996; Lal & Craig, 2005) and on-road driving studies (Macchi, Boulos, Ranney,

Simmons, & Campbell, 2002; Simon et al., 2011). Considered together, spectral

analysis of EEG recorded data can detect changes in sleepiness levels.

3.1.2 Electrooculography

Electrooculography (EOG) detects eye movements via the potential

difference between the back of the eye (negative charge) and the front of the eye

(positive charge). A number of ocular movements have been found to be sensitive to

sleepiness. For instance, increases in sleepiness have been related to increases in

blink frequency (Barbato et al., 2007; Schleicher, Galley, Briest, & Galley, 2008;

Stern, Boyer, & Schroeder, 1994), decreases in the closing velocity of the eyelid

during blinking (Sandberg et al., 2011), and increases in the duration of eye blinks

(Ingre, Åkerstedt, Peters, Anund, & Kecklund, 2006; Jammes, Sharabty, & Esteve,

2008; Numata, Kitajima, Goi, & Yamamoto, 1998). Increases in sleepiness as

indexed by ocular measures have also been significantly correlated with EEG indices

of sleepiness (Cajochen, Khalsa, Wyatt, Czeisler, & Dijk, 1999; De Padova, Barbato,

Conte, & Ficca, 2009), subjective sleepiness (Caffier, Erdmann, & Ullsperger, 2003;

Ingre et al., 2006), and behavioural measures of sleepiness (Johns, 1991).

The reliability of electrooculography has been found to be satisfactory. For

instance, Wang, Toor, Gautam, and Henson (2011) has found significant test-retest


correlations of blink frequency and blink duration to be r = .71 and r = .93

respectively. Similarly, Franks (1963) found a test-retest correlation of r = .72 for

blinks frequency. Ocular indices of the amplitude-velocity ratios have been found to

have a test-retest correlation of r = .8 when repeated within a two hour period (Johns,

Crowley, Chapman, Tucker, & Hocking, 2008).

The validity of blink measures is also satisfactory as measures of sleepiness.

Increases in sleepiness as indexed by ocular measures have also been significantly

correlated with EEG indices of sleepiness (Cajochen et al., 1999; De Padova et al.,

2009), subjective sleepiness measures (Caffier et al., 2003; Ingre et al., 2006), and

behavioural measures of sleepiness (Johns, 1991). However, blink duration may

prove to be a more sensitive index for sleepiness than other ocular indices (Hallvig,

Anund, Fors, Kecklund, & Akerstedt, 2014; Ingre et al., 2006; Schleicher et al.,

2008).

3.1.3 Electrocardiography

Electrocardiography is the recording of electrical activity of the heart. It is

well established that increases in sympathetic activity result in elevated heart rate,

including sitting upright from a supine position, physical activity, mental stress, and

cognitive performance (Allen, Obrist, Sherwood, & Crowell, 1987; Cole, 1989;

Melillo, Bracale, & Pecchia, 2011; Toivonen, Helenius, & Viitasalo, 1997; Treffner,

Barrett, & Petersen, 2002). Parasympathetic tone (or vagal tone) increases during

resting conditions (Brownley, Hurwitz, & Schneiderman, 2007; Goldberger, 1999;

Rajendra Acharya, Kannathal, Mei Hua, & Mei Yi, 2005). Analysis of the changes in

the interbeat-intervals in terms of the time and frequency domains can provide

assessments of sympathetic and parasympathetic tone. These interbeat-intervals are

sensitive to dynamic changes in underlying arousal levels (Cacioppo, Tassinary, &


Berntson, 2007; Chua et al., 2012; Tran, Wijesuriya, Tarvainen, Karjalainen, &

Craig, 2009) and can provide a fine temporal resolution of dynamic changes in

arousal levels (Task Force of the European Society of Cardiology and the North

American Society of Pacing and Electrophysiology [TFESCNASPE], 1996).

A number of time and frequency domain indices can be computed for ECG

recording. Commonly examined time domain indices include: mean heart rate, mean

R-R, standard deviation of R-R intervals (SDNN), root mean square SD (RMSSD),

and percentage of the number of times consecutive normal sinus intervals exceeded

50 ms (pNN50) (TFESCNASPE, 1996; Tran et al., 2009). Indications of sleepiness

for time domain indices typically result in increased HRV indices (Tran et al., 2009;

Verwey & Zaidel, 1999). Frequency domain indices commonly examined include the

absolute and relative power densities in the low frequency (LF: 0.04-0.15 Hz), high

frequency (HF: 0.15-0.4 Hz) bands and the LF/HF ratio. Typically, increases in LF

spectra, decreases in HF spectra, and increases in the LF/HF ratio are indicative of

increases in sleepiness (Apparies, Riniolo, & Porges, 1998; Tran et al., 2009)

The reliability of heart rate indices has shown good stability. The time

domain indices SDNN and RMSSD when collected two weeks apart have test-retest

correlations of .74-.85 and .75-.98 respectively (Guijt, Sluiter, & Frings-Dresen,

2007). Similarly, correlations between .71-.98 for mean R-R, SDNN, and RNSSD

indices have been observed (Nussinovitch, Cohen, Kaminer, Ilani, & Nussinovitch,

2012). Test-retest correlations for frequency domain indices of absolute LF and HF

have been found to be .83 and .85 respectively and for normalised LF and HF spectra

were .76 for both (Schroeder et al., 2004). These findings indicate adequate

reliability for time and frequency domain indices.

The validity of heart rate variability indices has similarly been acceptable.


Examinations of the heart rate time domain indices have associations with aerobic

output during cycling exercise (Guijt et al., 2007). During an overnight sleep

deprivation study frequency domain indices of LF have increased and HF have

decreased with increasing time awake and were validated against increases in EEG

theta and alpha absolute power (Kaida, Åkerstedt, Kecklund, Nilsson, & Axelsson,

2007). Several other studies have found HRV indices to be associated with increases

in sleepiness (Tran et al., 2009; Verwey & Zaidel, 1999). Changes in HRV have been

found for increased cognitive functioning (Luft, Takase, & Darby, 2009; Mukherjee,

Yadav, Yung, Zajdel, & Oken, 2011) including increases in cognitive task demand

(Backs & Seljos, 1994; Fairclough & Houston, 2004; Fairclough & Roberts, 2011).

3.1.4 Physiological Data Acquisition and Processing

The following section will present the technical aspects of the physiological

data acquisition as well as the processing of the data for subsequent use in statistical

analyses. The EEG data recorded in the three studies of the present research program

used slightly different electrode montage (i.e., recording sites). The EEG data

recorded in Study One and Three was recorded from four bipolar derivations

positioned at F5-A2, C3-A2, O1-A2 and F6-A1 using Ag-Al electrodes. Whereas,

the EEG data recorded in Study Two, which was collected prior to the present

research program but was reanalysed for Study Two, was recorded from four bipolar

derivations positioned at C3-A2, O1-A2, C4-A1, and O2-A1. All EEG data was

recorded using Ag-Al electrodes filled with a commercial conductive paste (Ten20;

Weaver and Company, Colorado, United States of America).

The EOG data was recorded with a vertical pairing of electrodes. Electrodes

were placed centrally to the left pupil, and were located 3 cm above and 2 cm below

the centre of the pupil. This vertical-EOG channel was used with the calculation of


the blink duration index. The ECG data was recorded using a modified two-lead

pairing of electrodes. One electrode was placed approximately 3-5 cm below the

right clavicle (collarbone) with the second electrode placed on the left lower ribcage

at the V6 location. A ground electrode was placed approximately 3-5 cm below the

left clavicle. Self-adhesive electrodes were used for the EOG, ECG, and ground

recording sites. The skin beneath all electrode sites were abraded with a commercial

skin abrasion product (Nuprep; Weaver and Company, Colorado, United States of

America) until an impedance of 5 kΩ was achieved, which is consistent with

guidelines for electrophysical studies (Leary, 2007). In addition, the signal quality

was visually confirmed onscreen before the data collection began.

The physiological data was recorded with ProFusion PSG 3 v3.4 Build 345

software (Compumedics, Melbourne, Australia). The EEG, ECG, and EOG data was

sampled online at 256 Hz using a 102 Hz low-pass and a 0.16 Hz high-pass filter. An

offline low-pass filter of 0.50 Hz and a 46 Hz filter was then applied to the EEG data

prior to FFT for the spectral analyses. The EEG data was subject to a Fourier Fast

transformation utilising a 4-sec Hanning window with a 50% overlap. The absolute

power levels (μV2) were determined for two frequency bands, theta (4-8 Hz) and

alpha (8-13 Hz). The EOG data had a 10 Hz low-pass filter was applied to the data.

The EOG signals were scored as blinks if they were greater than 100 µV in

amplitude and were visually confirmed as blinks on the EOG electrophysical trace.

The index of blink duration, measured in milliseconds (ms) was extracted from the

data, measured as the duration at half blink amplitude, which is a commonly used

index of blinks amplitude (Caffier et al., 2003; Hallvig et al., 2014; Ingre et al., 2006;

Schleicher et al., 2008). The left eye EOG trace was used for all eye blink analyses.


As described in section 3.1.3, numerous indices which can be extracted and

computed from ECG data. The cardiac interbeat-interval or R-R interval, was chosen

as the cardiac index of arousal as interbeat-intervals are sensitive to dynamic changes

in underlying arousal levels (Cacioppo et al., 2007; Chua et al., 2012; Tran et al.,

2009), they provide a fine temporal resolution of dynamic changes in arousal levels

(TFESCNASPE, 1996), and are argued as a stable and reliable measure of cardiac

arousal (Porges & Bohrer, 1990). Thus, the interbeat-interval or R-R interval,

measured in milliseconds (ms), was extracted from artefact-free ECG data using a

peak detection algorithm (Tarvainen et al., 2014) that produced the R-R interval data.

A common problem for the processing of physiological data into useful

indices for analysis is the contamination of the electrophysical recording by

superfluous electrophysical signals or artefact. Movement artefact produced when

the participant moves can cause issues for several electrophysical recordings.

Specifically, movement artefact can cause increases in the power across all

frequency domains in EEG recordings, but predominantly in the alpha and beta

frequency ranges (Campos Viola et al., 2009; Gevins, Yeager, Zeitlin, Ancoli, &

Dedon, 1977). Regarding ECG and EOG recordings, movement artefact can distort

the characteristic shape of the normal sinus rhythm or the eye blink (TFESCNASPE,

Schlögl et al., 2007; 1996), and thus such recordings with movement artefact should

not be processed further (Gratton, 2007). To counteract these effects, all recordings

were visually scored for movement artefacts and any segment of movement artefact

was rejected from further processing.

Eye blink artefact is a particular concern for the analysis of EEG data.

Various methods have been proposed to correct eye blink artefacts (see Hatskevich,

Itkis, & Maloletnev, 1992; Jung et al., 2000; Ktonas, Osorio, & Everett, 1979;


Teixeira, Tomé, Lang, Gruber, & Martins da Silva, 2006; Woestenburg, Verbaten, &

Slangen, 1983). The frequency bands that are of most interest to the current study are

the theta and alpha frequency bands, due to their association with sleepiness

(Campagne et al., 2004; Kecklund & Åkerstedt, 1993). Artefact from blinking can

result in increases in the delta and theta frequency bands, but eye blink artefact can to

a lesser degree also contaminate the alpha frequency bank (Gasser, Sroka, & Möcks,

1985; Hagemann & Naumann, 2001). Thus, to mitigate the effects of eye blink

artefact in the EEG, independent component analysis was used to remove eye blink

artefacts prior to spectral analysis.

3.2 Subjective Measures

Several subjective measures were used in the present program of research.

The first two measures described in this section, the Pittsburgh Sleep Quality Index

and the Epworth Sleepiness Scale were used as assessment tools to screen potential

participants reporting sleep disturbances and excessive daytime sleepiness

respectively. The Sleep Timing Questionnaire was used to measure participants

habitual sleep timing. The Karolinska Sleepiness Scale was used to assess how

participants perceived their current level of sleepiness. Subjective ratings of

sleepiness provide important information regarding participants’ self-perception of

sleepiness and on sleepiness-related driving performance levels. Thus, subjective

measures can be informative regarding the state of the driver and they complement

physiological measures by providing a multi-measure or multi-assessment of the

behaviour or phenomena of interest (Hunsley, 2003; Hunsley & Meyer, 2003).

3.2.1 Pittsburgh Sleep Quality Index

The Pittsburgh Sleep Quality Index (PSQI; Buysse, Reynolds, Monk,

Berman, & Kupfer, 1989) is a self-report questionnaire that assesses subjective sleep


quality and sleep disturbances during the preceding month. The items of the PSQI

represent standard themes that sleep clinicians routinely assess (Buysse et al., 1989).

The questionnaire utilises 19-items to generate seven component scores ranging from

0-3. The seven components are: subjective sleep quality, sleep latency, sleep

duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication,

and daytime dysfunction. The seven component scores are summated to produce a

global PSQI score that has a range of 0-21, with higher scores indicative of poorer

sleep quality. In accordance with Buysse et al. (1989), a score of five or less was

utilised as a cut-off point between “good” and “bad” sleepers.

The PSQI has repeatedly demonstrated high levels of reliability. Test-retest

reliability was assessed by Backhaus, Junghanns, Broocks, Riemann, and Hohagen

(2002) over an approximate six week period, with a resultant correlation of r = .86, p

< .001 between the two testing times. Cronbach’s alpha coefficients between .70

and.85 have been reported for the PSQI, indicating that the internal consistency is

also high (e.g., Backhaus et al., 2002; Beck, Schwartz, Towsley, Dudley, &

Barsevick, 2004; Buysse et al., 1989; Carpenter & Andrykowski, 1998; Doi et al.,

2000; Skouteris, Wertheim, Germano, Paxton, & Milgrom, 2009).

The validity of the PSQI has been found to be satisfactory. Carpenter and

Andrykowski (1998) assessed the construct validity of the PSQI by determining the

convergent and discriminant validity of the index. It was found that similar

constructs were moderately or highly correlated with the PSQI (i.e., there was

convergent validity), whereas dissimilar constructs were poorly correlated (i.e., there

was discriminant validity). Additionally, Backhaus et al. (2002) found that the PSQI

was significantly and highly correlated with sleep log data (r = .81, p < .001).

Finally, a number of studies have found that the cut-off score of five or less can


discriminate between good and poor sleepers (e.g., Aloba, Adewuya, Ola, & Mapayi,

2007; Backhaus et al., 2002; Buysse et al., 1989; Buysse et al., 1991; Stein, Chartier,

& Walker, 1993; Stein, Kroft, & Walker, 1993).

3.2.2 Epworth Sleepiness Scale

The Epworth Sleepiness Scale (ESS; Johns, 1991) is a measure of general

level of excessive daytime sleepiness in adults. The ESS was constructed based on

observations of the occurrence and nature of daytime sleepiness (Johns, 1991).

Participants respond to eight items relating to how likely they are to doze off or fall

asleep in various situations (e.g., “sitting and reading”, “sitting and talking to

someone”, and “in a car, while stopped for a few minutes in the traffic”). Potential

responses range are: 0 = would never doze, 1 = slight chance of dozing, 2 = moderate

chance of dozing, and 3 = high chance of dozing. The range of possible composite

scores is 0-24, with increasing scores being indicative of greater daytime sleepiness.

A score of 10 or less is considered to be within the normal range of daytime

sleepiness, such that there is little likelihood or potential sleep disorder (Johns, 1993;

Johns, 2000; Johns & Hocking, 1997).

The ESS has demonstrated good reliability with test-retest evaluations and

with measures of internal consistency. The test-retest properties of the EES were

assessed over a five month period with a large correlation found (r = .82, p < .001)

(Johns, 1992). Measures of internal consistency of the ESS (i.e., Cronbach’s alpha)

are similarly satisfactory. The Cronbach’s alpha measure of the ESS when completed

by 150 individuals with various sleep disorders and 104 ostensibly healthy medical

students was .88 and .73 respectively (Johns, 1992). Similarly, a Chinese version of

the ESS recorded a Cronbach’s alpha of .81 when completed by 30 individuals with

sleep disorders (N. H. Chen et al., 2002).


While the ESS has been found to have good reliability, concerns around the

validity of the measure have been raised. For instance, inconsistent results regarding

criterion validity was found when compared to mean sleep latencies of the Multiple

Sleep Latency Test (MSLT; Carskadon & Dement, 1982; Carskadon et al., 1986).

Comparisons with the MSLT, which is considered the gold standard physiological

measure of sleep propensity (Carskadon et al., 1986; Harrison & Horne, 1996a) has

yielded inconsistent results. Correlations ranging from moderate (e.g., Johns, 1991;

Johns, 1994; Vignatelli et al., 2003), small (e.g., Blaivas, Patel, Hom, Antigua, &

Ashtyani, 2007; Chervin, Aldrich, Pickett, & Christian, 1997), and non-significant

(e.g., Benbadis et al., 1999; Fong, Ho, & Wing, 2005) have been found. It has been

suggested that these contradictory findings can be explained by incompatibilities

between the MSLT, a physiological measures and the ESS, a retrospective subjective

measure of sleepiness (Johns, 2000). Nonetheless, the ESS has been shown to

distinguish between healthy, good sleepers and individuals with sleep disorders that

have been confirmed by alternate physiological measures (e.g., N. H. Chen et al.,

2002; Johns, 1991).

3.2.3 Sleep Timing Questionnaire

The Sleep Timing Questionnaire (STQ; Monk et al., 2003) is a self-report,

single-administration questionnaire that assesses the habitual sleep-wake timing of an

individual. Measures of an individual’s habitual bedtime (goodnight time, GNT) and

habitual wake-time (good morning time, GMT) are produced from the questionnaire

for the weeknights and weekend nights respectively. Additionally, the STQ produces

a measure of an individual’s stability of habitual bedtime and wake-time. This

measure of stability is measured on an arbitrary scale of 1-11, with higher scores

indicative of greater instability.


While the STQ has not had its psychometric properties extensively assessed,

it has been found to have good reliability. Monk et al. (2003) assessed the test-retest

reliability of the STQ of less than a year’s duration (mean 105 days, range 6-349

days). The test-retest correlations of GNT and GMT were both significant and large,

with correlations of r = .71 and r = .83 respectively.

The validity of the STQ has also been found to be acceptable. The convergent

validity of the STQ was assessed against sleep diary records of two weeks.

Significant correlations were found for GNT (r = .79) and GMT (r = .79) and the

sleep diary data (Monk et al., 2003). Significant and large negative correlations were

found by Monk et al. (2003) between morningness values assessed by the Composite

Scale of Morningness (C. S. Smith, Reilly, & Midkiff, 1989) and GNT (r = -.71) and

GMT (r = -.74) indicating that morning types fall asleep and wake-up earlier. Last,

the convergent validity of the STQ was assessed against data chronicled from

actigraphs from 23 healthy control individuals. Significant correlations of r = .59 and

r = .77 were found for GNT and GMT and the data obtained from the actigraphs

(Monk et al., 2003).

3.2.4 Karolinska Sleepiness Scale

The Karolinska Sleepiness Scale (KSS; Åkerstedt & Gillberg, 1990) is a self-

report measure of the level of subjective sleepiness an individual is experiencing.

Individuals are required to indicate on a 9- point Likert scale how sleepy they are

currently feeling. The modified version of the KSS (Reyner & Horne, 1998) includes

verbal anchors for every step (1 = extremely alert, 2 = very alert, 3 = alert, 4 = rather

alert, 5 = neither alert nor sleepy, 6 = some signs of sleepiness, 7 = sleepy, no effort

to stay awake, 8 = sleepy, some effort to stay awake, and 9 = very sleepy, great effort


to keep awake, fighting sleep. The question posed to the participants is “Right now

how sleepy are you feeling?”

The KSS has been found to be a reliable measure of sleepiness and in

particular with sleep-deprived individuals. A sleep deprivation study that utilised the

KSS over two nights of testing that were one week apart found no significant

differences between the KSS scores over the two testing sessions (Gillberg,

Kecklund, & Åkerstedt, 1994). Gillberg et al. (1994) found the KSS to be sensitive to

the amount of time spent awake during a night of sleep deprivation, such that KSS

scores gradually increased over a night of sleep deprivation. Similarly, Kaida et al.

(2006) found that KSS scores increased with time spent awake in a strong linear

trend.

The validity of the KSS has been repeatedly shown to be quite high when

compared to concurrently recorded EEG. For example, Åkerstedt and Gillberg(1990)

found correlations between KSS scores and EEG theta and alpha power levels that

were all in excess of r = .5. Moreover, van den Berg, Neely, Nilsson, Knutsson, and

Landström (2005)found correlations between the KSS scores and EEG

measurements of sleepiness of r = .76-.91. The concurrent validity of the KSS was

assessed against response times of the PVT which has been shown to be sensitive to

arousal decrements and sleep deprivation (Dinges et al., 1997). The KSS scores

significantly correlated (r = .57) with mean response times of the PVT (Kaida et al.,

2006). Last, KSS scores have been found to be a significant predictor of crash

involvement, with higher scores related to increases in crash likelihood (Åkerstedt,

Connor, et al., 2008).


3.2.5 Motivational Effort Scale

The Motivational Effort Scale (MES) was custom designed for use in the

present research program. The MES is a self-report measure of the level of effort the

individual is currently applying to maintain their level of task performance.

Participants indicate on a 9-point Likert scale how much effort they are currently

applying to the task. Higher scores on the MES are indicative of greater effort (1 =

no effort and 9 = great effort). Currently, no psychometric data is available for this

measure.

The MES was developed as the existing scales that assess applied effort, the

Subjective Mental Effort Questionnaire (SMEQ; Zijlstra, 1993) and the NASA Task

Load Index (NASA-TLX; Hart & Staveland, 1988) were not entirely suitable. That

is, both the SMEQ and the NASA-TLX are after scenario scales that are

administered to the participant/s once the task has been completed. Moreover, the

present studies required a scale that could be administered quickly during the

performance of task. The SMEQ has nine verbal anchors that would require a

participant to read and assess which verbal anchor is appropriate for their response at

the time. The NASA-TLX is comprised by six Likert scales that are used for

assessing the level of workload the participant experienced or exerted during the

task. Thus, completing either of these two scales would take a short amount of time;

however, the present study required a rapidly administered scale and the

Motivational Effort Scale, a 1-item Likert scale, with only two verbal anchors was

specifically developed for this purpose.

3.2.6 Intrinsic Motivation Inventory

The Intrinsic Motivation Inventory (IMI; Ryan, 1982) is a self-report

measures of an individual’s experiences while completing a given task. The IMI


measures these experiences via seven subscales, those subscales being:

interest/enjoyment, perceived competence, effort/importance, value/usefulness, felt

pressure/tension, perceived choice, and relatedness. The items of the IMI are

generically worded, such that the task of interest can be substituted in the wording of

the items (McAuley, Duncan, & Tammen, 1989; Ryan, 1982). Participants respond

to 45 items on a 7-point Likert scale with higher scores indicative of greater

agreement (1 = not at all true and 7 = very true). The perceived choice (i.e., of

completing the task) and relatedness (i.e., to another person) subscales were deemed

not applicable and thus were not used in the current study. Removal of subscales that

are not applicable or the focus of the research study does not change the

psychometric properties of the IMI (Ryan, 1982) and has been performed in other

studies (e.g., Choi, Mogami, & Medalia, 2010; Ferrer-Caja & Weiss, 2000).

The IMI has been found to demonstrate adequate reliability. McAuley et al.

(1989) have reported Cronbach’s alpha values to range between .68 to .84 for the

subscales of the IMI. In addition, McAuley et al.’s confirmatory factor analysis

substantiated the factor structure of the IMI subscales. Subsequent, use of the IMI

have reported Cronbach’s alpha values between .72 to .91 for the subscales of the

IMI (Markland & Hardy, 1997). Test-retest reliability of the IMI subscales when

completed a week apart have demonstrated adequate reliability, with test-retest

correlations between r = .63 and .92 (Tsigilis & Theodosiou, 2003). Similarly, Choi

et al. (2010) have found test-retest correlations for the IMI subscales between r = .71

and .85 when completed 4 weeks part.

The validity of the IMI has also been found to be acceptable. For instance,

Ferrer-Caja and Weiss (2000) have found the IMI to predict adolescent students

motivated behaviours (i.e., effort and persistence) associated with physical education.


The convergent validity of the IMI has been demonstrated by Choi et al. (2010) with

significant and positive correlations with germane constructs such as the Perceived

Competency Scale (G. Williams, Freedman, & Deci, 1998) level of autonomy task

engagement using the Treatment of Self-Regulation Scale (Ryan, Plant, & O'Malley,

1995).

3.2.7 Signs of Sleepiness Questionnaire

After examination of the relevant peer reviewed literature regarding signs of

sleepiness, nine signs of sleepiness were selected for the Signs of Sleepiness

Questionnaire (SoSQ). The nine signs of sleepiness included physical and

psychological signs (i.e., difficulty keeping eyes open, slow eye blinks, difficulty

concentrating, yawning, slower reactions, changing position frequently, frequent eye

blinks, mind wandering, easily distracted). Participants rated the importance of each

of the sign of sleepiness as an indicator of their level of sleepiness on a 7-point Likert

scales from 1 = not important to 7 = very important. Currently, no psychometric data

is available for this measure.

The SoSQ was custom designed for use in the present research program, as

the existing questionnaires were not completely relevant and/or applicable. The

Physical Symptoms Questionnaire (T. Nilsson, Nelson, & Carlson, 1997) only

included physical sensations and no psychological signs of sleepiness are included in

this measure. Thus, the Physical Symptoms Questionnaire was limited in its utility

given some signs of sleepiness are psychological in nature. Alternatively, the

Sleepiness Symptoms Questionnaire (Howard et al., 2014) predominately listed signs

(or symptoms) of sleepiness that would be experienced during periods of high to

extreme levels of sleepiness (e.g., vision becoming blurred, nodding off to sleep,

head dropping down). Thus, experiencing early signs of sleepiness (e.g., yawning,


slow eye blinks, changing position frequently) were not assessed by the Sleepiness

Symptoms Questionnaire and some items were not applicable with use of the Hazard

Perception Test (e.g., difficulty maintaining the correct speed).

3.3 Performance and Behavioural Measures

3.3.1 Psychomotor Vigilance Test

The Psychomotor Vigilance Test (PVT; Dinges & Powell, 1985; Wilkinson

& Houghton, 1982) is a reaction-time task that measures tonic alertness, and is a

low-order cognitive task. The PVT requires participants to respond as quickly as

possible to a stimulus. Participants fixate on a cross on the computer screen which

then disappears and after a varying interstimulus intervals (1-10 sec) and in the

version of the PVT used in the present study, a red dot appears where the fixation

cross was previously. Participants have to respond as quickly as they can to the

presentation of the red dot with a press of a spacebar. The reaction time (measured in

ms) of each trial appears on screen after the spacebar is pressed (i.e., performance

feedback is provided). The standard duration of the PVT is 10-min in length;

however, 5-min version has also demonstrated sensitivity in detecting sleepiness

(Loh, Lamond, Dorrian, Roach, & Dawson, 2004). The PVT has virtually no

learning curve before proficiency is reached (Graw, Kräuchi, Knoblauch, Wirz-

Justice, & Cajochen, 2004).

The several indices can be extracted from the obtained PVT data. These

include simple mean reaction time, standard deviation of reaction time, mean of the

fastest and slowest 10% of reaction times, false starts, and lapses. False starts are a

response before the stimulus has appeared on screen. Lapses are a reaction time that

is greater than 500 ms in duration. The appearance of lapses during repeated testing

sessions have been shown to be indicative of greater levels of sleepiness from


extended wakefulness, partial, and complete sleep deprivation (Banks, Van Dongen,

Maislin, & Dinges, 2010; Basner & Dinges, 2011; Loh et al., 2004; Rupp et al.,

2012).

The reliability and the validity of the PVT has been found to quite adequate.

Regarding test-retest correlations of the effects of total sleep deprivation that were

one week apart found correlations of .89 and .86 for lapses and mean reaction times

respectively. Similarly, the validity of the PVT has been established with its

sensitivity to arousal decrements due to increases in homeostatic sleep pressure and

to circadian effects (Caldwell, Prazinko, & Caldwell, 2003; Dinges et al., 1997;

Kaida et al., 2006). Last, significant and positive correlations have been found

between PVT indices (lapses, mean RT, fastest RT, slowest RT, and median RT) and

theta and alpha absolute power during a night of total sleep deprivation (Caldwell et

al., 2003; Kaida et al., 2006) which demonstrate the validity of the PVT as a measure

of sleepiness.

3.3.2 Hazard Perception Test

Hazard perception has been described as the skill to identify a traffic scenario

that may result in a dangerous situation, requiring a reaction from the driver to avoid

an incident (Horswill & McKenna, 2004; McKenna & Crick, 1991). Driving is a

complex task that requires the successful operation and coordination of a number of

psychological processes. These processes used when driving include learning,

memory, perception, motor control, attention, decision-making, and executive

functioning (Horswill & McKenna, 2004; Uchiyama et al., 2003). The principle of

the hazard perception paradigm is to examine the cognitive processes that are in use

during the processing of potential threats presented in the visual simulation of the

road environment (McKenna & Crick, 1991).


The major design concept behind modern driving simulators is that of

physical fidelity and replicating vehicle control tasks such as lane keeping and speed

regulation (i.e., tracking tasks). However, sophisticated driving simulators may lack

the specificity to capture the range of psychological processes that occur during

driving and the graphics displayed to the participant can be limiting in terms of

clarity, perspective, and levels of task immersion. In contrast, the Hazard Perception

Test, utilises a number of psychological processes (e.g., memory, perception, motor

control, attention, decision-making) and is considered a high-order cognitive task

(Horswill & McKenna, 2004; McKenna & Crick, 1991). The visual stimuli of the

Hazard Perception Test are actual recordings of real footage recorded from the

driver’s perspective of actual hazards and potential hazards that have occurred on the

road, which allows for a higher level of genuineness and task immersion. Notably,

when vehicle control skills and hazards perception skills are examined for their

association with crash involvement, only the driving skill of hazard perception has

consistently been found to be associated crash incidents (Boufous et al., 2011; Darby

et al., 2009; Horswill et al., 2015; M. Hull & Christie, 1992; McKenna & Horswill,

1999; Pelz & Krupat, 1974; Quimby et al., 1986). Thus, hazard perception is an

important driving task and has a degree of criterion-related validity with actual on-

road crashes.

The Hazard Perception Test (HPT) is a reaction-time latency measure of the

ability to identify potential traffic conflicts. Traditionally, the HPT has been utilised

as a measure of the experience level of younger drivers with graduated driver

licencing schemes (McKenna & Crick, 1991). However, a growing number of

studies have used the HPT with various road safety objectives, including examining

the perceptual abilities of younger and older driving (Underwood, Phelps, Wright,


van Loon, & Galpin, 2005), the effects of a traumatic brain injury on hazard

perception performance (Preece, Horswill, & Geffen, 2011), the hazard perception

abilities of individuals with autistic spectrum disorder (Sheppard, Ropar,

Underwood, & Loon, 2009), and the degradation of cognitive functioning associated

with aging on hazard perception ability (Horswill, Anstey, Hatherly, & Wood, 2010).

The HPT is completed by watching video footage of on-road traffic

situations. The footage utilised for the test was that of real footage recorded from the

driver’s perspective (during daylight hours). Each video segment requires the

individual to determine if a potentially hazardous situation could eventually lead to

traffic incident if no defensive action was initiated by the on-road driver. If a

potentially hazardous situation is identified by the individual, they then have to click

on the potential hazard (e.g., car suddenly braking in front of them) with a mouse

pointer. The reaction-time latency is determined from when the potential hazard first

appears in the video to when it is clicked on with the mouse. A previously developed

and validated methodology developed by McGowan and Banbury (2004) was used

which requires participants to click on the identified hazard with a mouse pointer.

This scoring method avoids ambiguous responses, because participants have to

identify the location as well as the timing of each hazard identified. The HPT was

run from desktop computer, with the video footage displayed to participants on a

monitor (60 Hz refresh rate), situated directly in front of them. Figure 3.1 (over page)

shows an example of a potential hazard sequence that was utilised in the PhD

research program.


Figure 3.1. Images of a traffic hazard used in the present study. The blue car brakes, which means that the car containing the camera (effectively the participant’s vehicle) would have to slow in order to avoid a collision. Importantly, it is possible for those participants with good hazard perception ability to anticipate this traffic conflict long before the blue car starts braking and start to slow down. As long as they are actively scanning the road beyond the blue car where it is possible to see a taxi manoeuvring and block the road ahead as seen in the second and third images).

The reliability of any hazard perception test is a critical issue and ultimately

determines its sensitivity. The majority of hazard perception studies have found

differences between novice and experienced drivers. Horswill et al. (2008), S. S.

Smith et al. (2009), and Wetton et al. (2010) studies have utilised the identical

footage as the current HPT with reported reliability coefficients of α = .85, .90, and

.95. Last. the study performed by S. S. Smith et al. (2009) demonstrated the utility of

the HPT with its use in repeated measures design, with a large correlation of r = .83

between two version of the HPT used in that study.

Evidence of the validity of the HPT has been repeatedly demonstrated in

assessments of crash risk, driver experience levels, and ratings of driver


performance. For instance, the HPT has been shown to differentiate inexperienced

from experienced drivers (Borowsky, Oron-Gilad, & Parmet, 2009; L. Jackson et al.,

2009; McKenna & Crick, 1991; S. S. Smith et al., 2009; Underwood et al., 2005).

Further, slower hazard perception performance has been related to increased crash

risk (A. E. Drummond, 2000; Horswill et al., 2015; M. Hull & Christie, 1992;

McKenna & Horswill, 1999; Pelz & Krupat, 1974; Pollatsek, Narayanaan, Pradhan,

& Fisher, 2006). Most importantly, the skill of hazard perception is sensitive to

increases in sleepiness levels (S. S. Smith et al., 2009; Watling, 2012).

As mentioned above, hazard perception is one of many driving skills, but is

an important driving skill that has validity data with on road crashes (Boufous et al.,

2011; Darby et al., 2009; Horswill et al., 2015; M. Hull & Christie, 1992; McKenna

& Horswill, 1999; Pelz & Krupat, 1974; Quimby et al., 1986). Whereas other driving

skills such as lane keeping and speed maintenance (i.e., tracking tasks) that are

assessed in driving simulators have uncertain value with on road incidents (George,

2003). Which driving skill is more important to determine if a particular task (Hazard

perception test, simulated driving or others) should be called “driving” the author is

currently unaware of any widely accepted conventions. Thus, the use of the word

“driving” refers to any skill/s that is involved in the task of driving.

3.3.3 Wrist Actigraphy

Actigraphy is not a direct measure of sleepiness, but is a way of quantifying

the sleep-wake timing. Actigraphy is a non-invasive method of inferring the sleep-

wake cycles of an individual. Often worn on the wrist, an actimetry sensor measures

the relative movement of the individual. Rest/activity periods are calculated using

custom computer software and sleep-wake periods are subsequently inferred based

on these calculations. Increased movement is regarded as an indicator of wake


periods and reduced movements is thought to signify sleep. Actigraphs provide a

non-invasive and objective measure of the rest/activity periods.

The actigraph model used was the Actiwatch®-L (Mini Mitter Co., Inc.). The

accelerometer of this device has a sampling frequency of 32 Hz, a bandwidth of 3-11

Hz, and a sensitivity of .05 g-force. The sampling epochs for this study were set to 1

min. Each recorded epoch is assigned as sleep or wake periods using the algorithm

that was developed by Oakley (1997) and the utility of this algorithm has been

authenticated against polysomnographic recorded sleep staging.

The actigraph has been found to be reliable and valid instrument for assessing

sleep-wake periods. For instance, Benson et al. (2004) found no significant

differences between actigraph recordings over two nights from a sample of good

sleepers. Additionally, Sadeh, Sharkey, and Carskadon (1994) found that actigraphs

worn on the dominant or non-dominant wrist were both equally reliable measures of

sleep-wake durations. A correspondence of 92.54% for the dominant and 92.58% for

the non-dominant wrist was obtained between actigraphic and polysomnographic

measures (Sadeh et al., 1994).

A number of studies have repeatedly found actigraphs to demonstrate high

levels of validity. Cole, Kripke, Gruen, Mullaney, and Gillin (1992) recorded

accuracy rates of 87.97% for determining sleep-wake periods for actigraphy

measures when compared against polysomnography measures in a validation sample

of healthy individuals. Jean-Louis et al. (1996) found a significant and large

correlation (r = .93) between total sleep time measured by actigraphs and

polysomnography. Additionally, a large correlation of r = .78 for sleep onset latency

was found between the two measures. Considered together, actigraphy is a valid

measure of sleep for normal/good sleepers (Cole et al., 1992; Jean-Louis et al., 1996;


Sadeh et al., 1994; Tryon, 2004)

3.4 Genotyping Measures

The assessment of DNA used the 2 ml self-collection kits (DNA Genotek Inc,

Ottawa, Ontario, Canada). The Oragene kits are non-invasive (as opposed to blood

sample collections), are easy to use, and are a reliable alternative to DNA extraction

from blood (Rylander-Rudqvist, Håkansson, Tybring, & Wolk, 2006; Viltrop,

Krjutškov, Palta, & Metspalu, 2010). Moreover, several studies have shown that the

Oragene•DNA (OG-500) kits produce significantly higher DNA yields than other

oral collection methods (Birnboim, 2011; Whittle et al., 2008). Last, the Oragene kits

have been successfully used in a number of previous studies (Bruenig, White,

Young, & Voisey, 2014; Lim et al., 2012).

Standard protocols were followed with the collection of the saliva samples

(DNA Genotek Inc., 2012). In short, the process involves the participant holding the

tubed container up to their mouth and they expel 2 ml of saliva into the container.

After providing the neccesary amount of saliva the participant then closes the lid and

releases a stabilisation solution.

Saliva samples were genotyped by the Australian Genome Research Facility

using a homogeneous MassEXTEND Sequenom assay and were consistent with a

previously published genotyping study from Voisey et al. (2010). The homogeneous

MassEXTEND Sequenom assay is based on the annealing of an oligonucleotide

primer (homogeneous MassEXTEND Sequenom primer) adjacent to the SNP that is

of interest to the research study. The addition of a DNA polymerase along with a

mixture of terminator nucleotides allows extension of the homogeneous

MassEXTEND Sequenom primer through the polymorphic location and generates

allele-specific extension products, each having a unique molecular mass. These


allele-specific extension products are then analysed by matrix-assisted laser

desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF MS), and a

genotype is assigned in real time.

3.5 Room Setup

All studies were performed in the same laboratory. All participants were

seated in a standard office high-backed chair. Participants were seated with their

head approximately 60 centimetres from the monitor, however, this distance varied

slightly depending on the participants’ preferred seating position. The monitor was

elevated slightly to ensure it was positioned in line with the participant’s eyes to

minimise the potential of neck or shoulder strain. The laboratory’s temperature was

controlled at 23 °C. No time cues, including wristwatches or mobile phones were

available to the participants while they were undergoing a testing session.

A bank of fluorescent lights was located on the ceiling of the laboratory. The

ambient light level of the laboratory were recorded using a Gossen Mavolux light

meter (5032B USB, certified to DIN 5032-7 and CIE 69 standards, initial sensitivity

of 0.01 lx). Light readings were taken prior to the first participant completing each

study; the range of light readings with the Gossen Mavolux light meter can be seen in

Figure 3.2 (over page).


Figure 3.2. The laboratory ambient light readings for the 19 in, 4 by 3 ratio aspect monitor used for displaying the Hazard Perception Test (left) and 27 in, 16 by 9 ratio aspect monitor used for displaying the Psychomotor Vigilance Test to participants for the “sleepy gene” study (right). Not to scale.

3.6 Chapter Summary

As has been documented in previous sections, that measuring sleepiness as

well as sleep-wake variables can be achieved in several distinct ways. Grouped

broadly, sleepiness can be measured via physiological recordings (i.e., EEG, EOG,

ECG), subjective measures (i.e., KSS), and behavioural measures (i.e., PVT, HPT).

While each measure has inherent pros and cons, it is generally accepted that multi-

measure studies are more comprehensive (Hunsley, 2003; Hunsley & Meyer, 2003)

and in particular, studies concerned with sleepiness and performance are enhanced

by the use of a number of convergent methodologies, such as physiological,

subjective, and behavioural measures (Åkerstedt et al., 2010; Moller et al., 2006).

Therefore, the measures proposed for the studies of the current research program

should be sufficient coverage to explore the psychophysiological and performance

outcomes associated with motivation and genetic differences.

19 in LCD Monitor

60 cm

30 cm 30 cm

390 lx 354 lx 382 lx

27 in LCD Monitor

60 cm

30 cm 30 cm

381 lx 364 lx 374 lx


Chapter Four: The Physiological, Subjective, Performance, and

Genetic Associations of Sleepiness in Young Adults

I'm not asleep... but that doesn't mean I'm awake.

–Unknown Author


This chapter presents the findings of the first study conducted as part of the

research program. Study one aimed to address both of the research questions; those

being research question one, what are the effects of motivation in terms of magnitude

and duration on sleepiness and performance?, and research question two, what

effects do individual differences, assessed as genetic variations, have on sleepiness

and performance?

4.2 Introduction

The absolute amount of sleep deprivation required to produce equivalent

levels of neurobehavioural deficits can vary between individuals. Several studies

have demonstrated that differential vulnerability to sleep loss can be found between

individuals during both partial or complete sleep deprivation (Frey et al., 2004;

Leproult et al., 2003; Rupp et al., 2012; Van Dongen et al., 2004). Indeed, intraclass

correlation coefficients, suggest that 48-68% of the variance in PVT lapses (i.e.,

reaction-time latency greater than 500 ms) can be accounted for by individual

differences (Lim, Choo, & Chee, 2007; Van Dongen, Maislin, et al., 2003).

Moreover, Rupp et al. (2012) have shown that the intraclass correlation coefficients

for performance outcomes on low- and high-order cognitive tasks were all greater

than .86 between a night of total sleep deprivation and a week of partial sleep


deprivation. Considered together, these findings suggest that the neurobehavioural

response to total and partial sleep deprivation is robustly trait-like and possibly

suggestive of a role of genetic variation.

As previously noted in Chapter 2, sleep-wake and performance outcomes are

intrinsically linked to sleep homeostatic as well as circadian rhythm processes. The

functioning of the circadian rhythm is coordinated by the transcription and

translation of numerous genes (Gachon et al., 2004; Reppert & Weaver, 2002) and

similarly a number of aspects of sleep and wake regulation are seemingly influenced

by genetic components. For instance, the genetic influence on sleep architecture has

been demonstrated from results of twin studies. Larger intraclass correlation

coefficients are observed with spectral power of NREM sleep periods between

monozygotic twins than dizygotic twins regarding (De Gennaro et al., 2008; Kuna et

al., 2012; Landolt, 2008a; Zung & Wilson, 1967). The same trends between

monozygotic and dizygotic twins also extend to similarities in spectral power

components of the waking EEG (Ambrosius et al., 2008; Linkowski et al., 1991) as

well as levels of cognitive performance when sleep deprived (Kuna et al., 2012).

Considered together, these studies all highlight a genetic influence on sleep-wake

functioning and cognitive performance.

Notwithstanding the trait-like responses to sleep deprivation and the

underlying influence of a multitude of genes involved not only with known circadian

functioning but also with sleep and wake regulation (e.g., Porkka-Heiskanen &

Kalinchuk, 2011), relatively little is known about the genetic impacts on sleepiness

and more specifically cognitive performance (Goel & Dinges, 2011; Landolt, 2008a;

Van Dongen et al., 2004). The majority of the genetic association studies focusing on

sleepiness and performance have been performed during acute sleep loss and have


predominantly focused on the variable-number tandem-repeats (VNTR) of the

PERIOD3 gene and the catechol-O-methyltransferase (COMT) Val158Met

polymorphism. In short, the studies examining the relationships between the

PERIOD3 gene and cognitive performances suggest that the longer PERIOD3 allele

seems to confer vulnerability to the effects of sleep deprivation, but only for high-

order cognitive tasks. These effects on high-order tasks appear to be the only

consistent finding (e.g., Groeger et al., 2008; Lo et al., 2012). Whereas, inconsistent

findings have emerged with different genotypes of the PERIOD3 gene, with

performance on low-order cognitive tasks, such as the PVT (e.g., Goel, Banks,

Mignot, & Dinges, 2009; Groeger et al., 2008; Viola et al., 2007). Additionally, the

main effects from the longer PERIOD3 allele only seem to fully emerge in response

to extended wakefulness, with most of the salient differences between genotypes

emerging after the circadian nadir (i.e., peak melatonin expression).

The protein coding gene ADORA2A (an adenosine A2 receptor gene) appears

to be another central gene involved in sleep-wake regulation. The accumulation of

adenosine is thought to be associated or involved in the accumulation of homeostatic

pressure (Gallopin et al., 2005; Ticho & Radulovacki, 1991) with a reduction of

adenosine levels occurring during sleep periods (Porkka-Heiskanen & Kalinchuk,

2011). Thus, adenosine may act as a signalling system for the need to sleep (Landolt,

2008b). While the exact role that adenosine plays in sleep-wake regulation is still to

be conclusively established (Basheer, Strecker, Thakkar, & McCarley, 2004; Saper et

al., 2005), genetic association studies are demonstrating a role the ADORA2A gene

has with differential effects on sleepiness and performance.

A number of studies have identified the role that the ADORA2A gene (SNP

rs5751876) performs with electrophysiological expression of sleepiness as well as


performance levels. Previous research has shown that individuals with the TT

genotype display higher levels of EEG defined sleepiness in the range of 7-10 Hz

during sleep and wakefulness, than individuals with the other genotypes of the

ADORA2A gene (Rétey et al., 2005; Rétey et al., 2007). Moreover, in the Rétey et al.

(2007) study, TT genotypes were also more sensitive to the effects of caffeine, which

suggests that the different genotypes could have distinctive arousal mechanisms.

Other differences between the genotypes of the ADORA2A gene have also been

noted in the extant literature in relation to performance outcomes. Individuals with

the TT genotype when compared to individuals with the CT genotype display longer

reaction-time latencies and more PVT lapses (Bodenmann et al., 2012; Rupp et al.,

2013).

Another SNP that is of interest regarding sleep-wake and performance

outcomes is the COMT gene Val158Met polymorphism (rs4680), which has its

chromosomal location in proximity to the ADORA2A. The COMT gene is not

associated with circadian functioning per se; however, the importance of the COMT

gene in terms of cognitive performance stems from the involvement of COMT in

modulating dopamine levels in frontal and prefrontal cortices (Huotari, García-

Horsman, Karayiorgou, Gogos, & Männistö, 2004; Tunbridge et al., 2004).

Dopamine has a critical role in goal-directed activity such as performance on

cognitive tasks, including executive tasks (Bertolino et al., 2006; Malhotra et al.,

2014), but more importantly, vigilance tasks (Lim et al., 2012; Tzschentke, 2001).

The COMT Met allele has a lower dopamine enzymatic activity than the Val allele,

and this results in a reduced modulation of dopamine availability in cortical tissue for

individuals with the Met allele (J. Chen et al., 2004; Lachman et al., 1996).

Neuroimaging studies have shown that individuals with the Met allele typically


displays larger areas of active cortical tissue during performance of a working

memory task than individuals with the Val allele, even with similar performance

levels on a working memory task (Bertolino et al., 2006; Egan et al., 2001). This

larger area of active cortical tissue in Met allele individuals is interpreted as a

compensatory mechanism due to the limited efficiency of their dopamine modulation

(Bertolino et al., 2006; Egan et al., 2001; Winterer et al., 2006).

The extant literature has shown that COMT genotypes are associated with

increased sleepiness and poorer performance. For instance, individuals with the Val

allele have been found to display lower levels of the upper frequency range (fast

alpha: 11-13 Hz) of the alpha rhythm when sleep deprived and when fully alert.

Additionally, the peak frequency (i.e., maximum average frequency) recorded during

wakefulness was 1.4 Hz slower in Val allele than in Met allele individuals

(Bodenmann et al., 2009). Enoch, Xu, Ferro, Harris, and Goldman (2003) have also

demonstrated functional difference between Val158Met genotypes, individuals with

the Val allele display higher alpha power from EEG recordings during wakefulness.

These findings all suggest functional differences of cortical arousal levels between

the different COMT genotypes. Regarding performance effects associated with the

Val158Met genotypes, alert (i.e., not sleep deprived) Val/Met genotype participants

have been shown to have poorer performance on a 20-min version of the PVT, with

faster decrements of performance levels (Lim et al., 2012). As previously stated, the

results from Lim et al. (2012) are contrary to previous findings (Bertolino et al.,

2006; Malhotra et al., 2014) and thus suggests more research is needed to understand

the functional difference between Val158Met genotypes.

The reviewed studies suggest that a number of genes could be associated with

variations of sleepiness and performance. Two aspects that have seemingly been


overlooked in previous research include the effects of the various genes at different

levels of sleep pressure. That is, most genetic association studies that have assessed

task performance have been conducted under sleep deprivation conditions (Rupp et

al., 2013; Viola et al., 2007). While some studies have demonstrated that their

findings are invariant under different sleep pressures (Bodenmann et al., 2009; Goel

et al., 2010; Groeger et al., 2008), there are still uncertainties regarding the exact

influence of genes on performance outcomes when sleep pressure is not at an

elevated level. Certainly, this is also the case with studies that examine performance

outcomes on the PVT. Importantly, most of the extant literature concerning the PVT

is based on acute sleep deprivation studies (Anderson & Horne, 2008; Dinges et al.,

1997; Loh et al., 2004; Rupp et al., 2012). Thus, examining normal variations in

sleepiness without the addition of experimentally manipulated sleep deprivation as

well as examining the relationship between normal variations in sleepiness with

different genotypes is important for furthering knowledge regarding normal day-to-

day functioning of the sleep and wake drives.

The effects of motivation on sleep-wake and performance outcomes have

been scarcely been examined, especially in genetic association studies. To date only

Goel et al.’s (2009) study has unsystematically and, perhaps unintentionally,

included any form of motivation with their study paradigm. Specifically, in Goel et

al.’s (2009) study, participants were instructed to “perform to the best of their ability

and to use compensatory effort to maintain performance.” Thus, without a

manipulation of motivation or a baseline measurement, it is difficult to quantify the

magnitude of the effect from motivation or apply extra effort.

The findings from Goel et al.’s (2009) study can be tentatively compared to

the findings from the Lo et al. (2012) as both studies used similar performance


measures of both low- and high-order cognitive tasks. Lo et al. (2012) found that

participants with the PERIOD35/5 allele displayed the typical vulnerability to

sleepiness as reported in other studies. In contrast, Goel et al. (2009) found no

difference between the PERIOD3 genotypes, which potentially could have been an

effect from the instruction “to use compensatory effort to maintain performance”.

Neuroimaging studies also supports the importance of increased effort, with the

study by Vandewalle et al., (2009) demonstrating participants with the PERIOD34/4

allele recruited supplemental cortical tissue during a working memory task, whereas

the shorter alleles did not show this tendency. Consequently, task performance

between the two genotypes did not differ significantly in Vandewalle et al.’s (2009)

study. Given the apparently intertwined relationship between sleepiness and

motivation, examining the genetic associations with increased motivation in a

systematic way would appear warranted.

Thus, the purpose of Study One was to address research question one, which

was concerned with the effect motivation can have on sleepiness and performance,

and research question two, being what are the genetic association with sleepiness and

performance? Younger individuals are more vulnerable to the impairing effects of

sleepiness (e.g., Adam et al., 2006; Lowden et al., 2009), as well as increased sleep-

related crash risk for 18-25 year olds (Connor et al., 2002). Additionally, circadian

related gene expressions may be more influential with younger individuals (Jones et

al., 2007). Therefore, examining genetic differences in relation to physiological,

subjective, and performance outcomes with younger individuals is an important

endeavour.


4.3 Method

4.3.1 Design

4.3.1.1 Research question one

A cross-sectional design was used to address the study’s research questions.

To address research question one, a series of 3 x 6 mixed factorial design ANOVAs

were used to examine changes in the physiological, subjective, and performance data

during the testing session. The first factor, a within-subjects factor, was Time Period

(0-5 min, 5-10 min, 10-15 min). The first two time periods (0-5 min, 5-10 min)

encompass the PVT10 testing session which was undertaken with the standard PVT

instructions, and the last time period (10-15 min) encompassing the PVT5 testing

session was undertaken with a motivated instruction set. The KSS data was collected

prior to the PVT10, at the end of the PVT10 and at the PVT5. Figure 4.1 (over page)

displays the data collection points of the study.

The second factor, a between-subjects factor, was the Time of Day of Testing

factor, during which the participant undertook the study. The Time of Day of Testing

factor included six levels (09:00, 10:30, 12:00, 13:30, 15:00, and 16:30). The Time

of Day of Testing factor was included to control for the associated increases in

sleepiness that occur throughout the day due to homeostatic and circadian processes.

The dependent variables were physiological (i.e., EEG: theta and alpha absolute

power for the electrode sites of F5, C3, and O1), subjective (i.e., KSS), and

performance variables (i.e., PVT: mean reaction-time latency). Significant main

effects were further examined with a set of Bonferroni adjusted planned comparisons

that examined the change from the 0-5 min to the 5-10 min time period and the 5-10

min to the 10-15 min time period. If the sphericity assumption was breached the

Greenhouse-Geisser correction was reported.


4.3.1.1 Research question two

To examine research question two, the genotype associations with sleepiness

and performance, a series of Univariate ANOVAs were conducted to examine the

levels of physiological, subjective, and performance data during the PVT10 as well

as the PVT5 testing sessions with different genotypes of the candidate SNPs. Given

the difficulties with ensuring ANOVA designs are adequately powered to detect

significant interaction effects (Cohen, 1988), and the small main effect sizes

commonly found with genetic association studies (Ioannidis, Ntzani, Trikalinos, &

Contopoulos-Ioannidis, 2001; Park et al., 2010), separate Univariate ANOVAs were

conducted on the PVT10 data as well as the PVT5 data for each of the 22 candidate

SNPs. As such, the independent variables were the different genotypes of the

candidate SNPs. The dependent variables were physiological (i.e., EEG: theta and

alpha absolute power for the electrode sites of F5, C3, and O1), subjective (i.e.,

KSS), and performance variables (i.e., PVT: mean reaction-time latency, and total

lapses). Significant main effects were followed up with post-hoc tests with Games-

Howell adjustments, as the genotypes were likely to have different sample (i.e., cell)

sizes.

PVT10 PVT5

0-5 min 5-10 min 10-15 min

Standard PVT Instructions Motivated Instructions

KSS 1 KSS 2 KSS 3 Figure 4.1. Data collection points of the study.


4.3.2 Participants

Participants were recruited via the university’s email distribution lists,

individual approach on campus and by advertisement on the first-year psychology

participant pool webpage. All participants were compensated for their involvement in

the study; the first-year psychology students received course credit while the

remaining participants received 20 AUD. In total, 131 participants took part in the

study. The mean age was 19.95 years (SD = 2.35; range = 18-25) with 68 participants

(51.91%) being female. Participants were excluded from participating if they had any

significant health problems or a sleep disorder, were a shift worker, had travelled

overseas in the past month, and took prescription medications that altered arousal

levels or illicit drugs. No other exclusion criteria were used.

On average the participants usual weekday bedtime was at 23:00 (SD = 73.00

min, range: 20:30-03:00) and usual weekday wake-time was at 07:25 (SD = 91.00

min, 03:00-12:30) and thus the average normal weekday sleep duration for the

participants was M = 507.802 min (SD = 84.50; range = 300-750) as assessed by the

Sleep Timing Questionnaire (Monk et al., 2003).

4.3.3 Measures

4.3.3.1 Demographic and traffic-related characteristics

Demographic and traffic-related information was collected, including age,

sex, duration of having a licence (including duration of their learner licence), and

number of hours driven per week.

4.3.3.2 Physiological measures

The physiological measure used in the current study were EEG theta and

alpha absolute power levels; spectral analyses were performed on the F5-A2, C3-A2,

and O1-A2 derivations. Please refer to section 3.1 for specific details of this EEG


measure, including issues of reliability and validity and section 3.1.4 for the details

of data acquisition and processing. Figure 4.2 shows the electrode sites recorded in

the current study.

Figure 4.2. EEG electrode sites utilised in the sleepy gene study.

4.3.3.3 Subjective measures

The subjective measures used in the current study were the KSS and STQ.

The KSS was used to quantify the participants’ levels of subjective sleepiness, while

the STQ was used to quantify habitual sleep timing. Please see sections 3.2 for

specific details about these measures including aspects of reliability and validity.

4.3.3.4 Behavioural measures

The behavioural measures used in the current study were the metrics derived

from the PVT. These metrics were mean reaction-time latency and lapses. Please see

sections 3.3 for specific details of the PVT as well as issues regarding the reliability

and validity of the PVT.


4.3.3.5 Genotyping measures

The Oragene self-collection kits were used for collection of human saliva

samples in the current study. Please see section 3.4 for specific details regarding the

Oragene self-collection kits.

4.3.4 Procedure

Ethical clearance to conduct research with human participants was obtained

from the University Human Research Ethics Committee (clearance number

1200000552) prior to beginning the study as well as the Health and Safety approval.

All participants were tested individually. Participants were instructed to maintain

their habitual sleep and wake times the night before the testing session to ensure they

were fully rested. Upon arrival at the laboratory, the study was explained to

participants in greater detail before participants signed a consent form. Participants

then had the EEG, EOG, and ECG electrodes attached. While the electrodes were

being attached, participants completed the study questionnaires (e.g., demographic,

Sleep Timing Questionnaire). Once the electrodes were attached and the participants

had completed the study questionnaires they received standardised instructions about

completing the testing session. The standardised instruction included a brief

description of how to complete the PVT and the stimulus the participant would

perceive. Although, it is noted that the PVT has virtually no learning curve before

proficiency is reached (Graw et al., 2004). The experimenter left the laboratory so

that each participant completed the entire testing session alone.

Before participant started the PVT10 their subjective sleepiness was assessed

with the KSS. Participants then completed a 10-min version of the PVT (PVT10) and

received standard PVT instructions and at the conclusion of the PVT10 the

participants’ subjective sleepiness was assessed again. The participants then


undertook a 5-min version of the PVT (PVT5), however, they received some

additional onscreen instructions before starting the PVT5. These instructions were

“We now want you to try to BETTER your previous performance. This means

motivating yourself to try even harder than before.” The participants’ reaction time

for each response was displayed onscreen, and thus the participants were aware of

their level of task performance. After completing the PVT5 the participants’

subjective sleepiness level was assessed for the last time. Previous research has

demonstrated that when experimenters interact with participants to verbally assess

the participants level of subjective sleepiness or to deliver task instructions, moderate

increases of arousal occur (Kaida et al., 2007). Thus, to avoid this effect the

assessment of the KSS and the delivery of the PVT10 and PVT5, instructions were

displayed onscreen.

At the end of the PVT5 testing, the experimenter re-entered the laboratory

and explained to the participant the correct method for providing the saliva sample.

While the participant was providing the saliva sample the electrodes were removed.

Once the saliva sample was provided, and all the electrodes were removed, the first-

year psychology students received course credit (via an online webpage), with the

other participants provided with 20 AUD compensation. Lastly, all participants were

thanked for their participation in the study before leaving the laboratory.

4.3.5 SNP Selection

In total, 22 SNPs from 16 genes were selected for examination in the current

study. Table 4.1 (over page) displays the SNPs “rs” number (Reference SNP cluster

ID), its associated gene, and known biological function.


4.4 Results

4.4.1 Sleep Prior to Testing

In total, the amount of sleep obtained the night before the testing session was

M = 502.23 min (SD = 81.91; range = 340-705) as determined by a simple sleep

diary. No significant difference was found between the participants average normal

weekday sleep duration (M = 507.802 min, SD = 84.50) as assessed by the STQ and

the amount of sleep obtained the night before the testing session (M = 502.23 min,

SD = 81.91), t(130) = 1.34, p = .18. Additionally, the correlation between the

participants average normal weekday sleep duration and the amount of sleep

obtained the night before the testing session was r = .73, p < .001.

Table 4.1. SNPs rs Number, their Associated Gene, and Biological Function.

SNP Gene Biological function

rs10462020 PERIOD3 Circadian rhythm regulation rs10774044 CACNA1C Calcium ion channel gating rs1082214 TIMELESS Circadian rhythm regulation rs11541353 NPAS2 Circadian rhythm regulation rs1154155 HLA-DQB1*0602 Immune system rs1554338 CRY2 Circadian rhythm regulation rs17107484 SCP2 Protein encoder rs1799990 PRNP Protein production rs1801260 CLOCK Circadian rhythm regulation rs1868049 ARNTL Circadian rhythm regulation rs2291738 TIMELESS Circadian rhythm regulation rs2298383 ADORA2A Adenosine receptor rs2304672 PERIOD2 Circadian rhythm regulation rs2735611 PERIOD1 Circadian rhythm regulation rs2797687 PERIOD3 Circadian rhythm regulation rs3736544 CLOCK Circadian rhythm regulation rs3749474 CLOCK Circadian rhythm regulation rs4242909 POLE DNA maintenance and regulation rs4630333 TIMELESS Circadian rhythm regulation

rs4680 COMT Dopaminergic role rs5751876 ADORA2A Adenosine receptor, protein encoding rs8119787 NFATC2 Immune system


4.4.2 Data Treatment

All data were examined for their distribution properties with histograms,

skewness, and kurtosis statistics. The histograms, skewness, and kurtosis statistics all

confirmed that the data displayed normal distributions; however, there was some

minor departures from normality (positive skewing) occurring with the KSS data.

When transformations were applied to the data, the skew was removed, but no

practical change to the statistical results occurred when they were compared to the

statistical results using the untransformed data, thus the untransformed data used.

The distributional properties of the genotype data will be examined below.

4.4.3 Genotype Distributions

Table 4.2 (over page) displays the results of the genotyping of the 22

candidate SNPs and the distribution of genotypes. Overall, 10 of the 22 genes

assayed had a pass rate of 100% with the lowest assay pass rate being 95.42% for the

NPAS2 gene SNP (rs11541353). After examining genotype frequencies all SNPs

were found to be in Hardy-Weinberg Equilibrium except SNP rs17107484 (p =

0.001). Thus, SNP rs17107484 was removed from the subsequent analyses as the

distribution of genotypes breached Hardy-Weinberg Equilibrium.


Table 4.2. SNPs and the Genotypes Frequencies with Chi-Square Tests of Hardy-Weinberg Equilibrium.

Nucleotide sequences and genotype frequencies Hardy-

Weinberg

Equilibriuma SNP

Common

homozygote Heterozytote

Rare homozygote

Assay

pass rate

rs10462020 TT: 98 (74.81%) GT: 31 (23.66%) GG: 2 (1.53%) 100.00% .80 rs10774044 GG: 120 (92.31%) GA: 9 (6.92%) AA: 1 (0.77%) 99.23% .10 rs1082214 CC: 115 (88.46%) CT: 15 (11.54%) 0 (0.00%) 99.24% .49 rs11541353 CC: 88 (70.40%) CT: 33 (26.40%) TT: 4 (3.20%) 95.42% .68 rs1154155 TT: 99 (75.57%) GT: 28 (21.37%) GG: 4 (3.05%) 100.00% .26 rs1554338 AA: 113 (86.26%) GA: 18 (13.74%) 0 (0.00%) 100.00% .40 rs17107484 GG: 121 (92.37%) GA: 8 (6.11%) AA: 2 (1.53%) 100.00% .00 rs1799990 AA: 59 (45.74%) GA: 57 (44.19%) GG: 13 (10.08%) 98.47% .89 rs1801260 TT: 68 (52.71%) TC: 49 (37.98%) CC: 12 (9.30%) 98.47% .47 rs1868049 CC: 87 (67.44%) CT: 39 (30.23%) TT: 3 (2.33%) 98.47% .57 rs2291738 TT: 37 (28.24%) TC: 60 (45.80%) CC: 34 (25.95%) 100.00% .34 rs2298383 TT: 48 (36.92%) TC: 57 (43.85%) CC: 25 (19.23%) 99.24% .28 rs2304672 GG: 111 (88.10%) GC: 14 (11.11%) CC: 1 (0.79%) 96.18% .46 rs2735611 AA: 88 (68.75%) GA: 35 (27.34%) GG: 5 (3.91%) 97.71% .52 rs2797687 GG: 129 (98.47%) GT: 2 (1.53%) 0 (0.00%) 100.00% .93 rs3736544 GG: 61 (46.56%) GA: 57 (43.51%) AA: 13 (9.92%) 100.00% .95 rs3749474 CC: 48 (37.50%) CT: 62 (48.44%) TT: 18 (14.06%) 97.71% .78 rs4242909 TT: 49 (38.28%) CT: 58 (45.31%) CC: 21 (16.41%) 97.71% .59 rs4630333 CC: 50 (38.17%) CT: 55 (41.98%) TT: 26 (19.85%) 100.00% .13

rs4680 AA: 33 (25.19%) GA: 66 (50.38%) GG: 32 (24.43%) 100.00% .93 rs5751876 CC: 48 (37.50%) CT: 57 (44.53%) TT: 23 (17.97%) 97.71% .40 rs8119787 AA: 31 (23.66%) AG: 69 (52.67%) GG: 31 (23.66%) 100.00% .54 a denotes the obtained significance level (p value) from the Chi-square test.

4.4.4 Overall Session Trends

Figure 4.3 and Figure 4.4 (over page) displays the mean and standard error of

the mean values of the physiological, subjective, and performance data. Overall, the

trends of the physiological data suggest that mean theta absolute power values did

not change over the course of the testing session. This observed trend in the mean

theta absolute power values did not appear to occur with the mean alpha absolute

power values. Specifically, mean alpha absolute power values appeared to increase

over the duration of the PVT10 for all electrode sites. Yet, it can be seen in Figure


4.3 that the mean alpha absolute power values appear to increase only for the O1

electrode site from the 5-10 min time period to levels found during the PVT5, with

no increase or little increase with the F5 and C3 electrode sites. Subjective sleepiness

appeared to increase from levels prior to the testing session to the end of the PVT10

session, but they appeared not to increase from levels at the end of the PVT10 to

levels at the end of the PVT5 session. The mean reaction-time latencies of the PVT

performance data increased from the 0-5 min section to the 5-10 min section of the

PVT10 session. Mean reaction times latencies of the PVT5 (the motivated test

session) reduced to similar levels found during the 0-5 min section of the PVT10.

To examine research questions one, the effects of motivation, a series of 3 x 6

mixed factorial design ANOVAs were performed, with the results tabulated in Table

4.3 (over page). If the sphericity assumption was breached the Greenhouse-Geisser

correction was reported. Significant main effects for the Time Period factor were

observed for F5 alpha, C3 alpha, and O1 alpha data. The planned comparisons show

that both the F5 and C3 alpha absolute power increased from time period one (0-5

min time period) to time period two (5-10 min time period), but did not increase

from time period two to time period three. Regarding the O1 alpha absolute power

data, it was found they increased from time period one to two and again from time

period two to three. The levels of subjective sleepiness assessed via the KSS,

increased prior to the undertaking the PVT10 to levels found at the end of the

PVT10. However, mean KSS levels did significantly change from the end of the

PVT10 to the end of the PVT5. Last, the performance data, being the mean reaction-

time latency from the PVT increased significantly from time period one to time

period two during the PVT10 session. Yet, the mean reaction-time latency

significantly decreased from time period two to time period three.


Figure 4.3. Mean level of EEG sleepiness recorded at the F5, C3, and O1 electrode sites during the PVT10 (black bars) and the PVT5 (grey bar). Error bars represent standard error of the mean.

0.00

5.00

10.00

15.00

20.00

25.00

30.00

0-5 min 5-10 min 10-15 min

EEG

abs

olut

e po

wer

, µV2

Time period

F5 Mean Theta Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

0-5 min 5-10 min 10-15 min

EEG

abs

olut

e po

wer

, µV2

Time period

F5 Mean Alpha Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

0-5 min 5-10 min 10-15 min

EEG

abs

olut

e po

wer

, µV2

Time period

C3 Mean Theta Power

0.00

5.00

10.00

15.00

20.00

25.00

0-5 min 5-10 min 10-15 min

EEG

abs

olut

e po

wer

, µV2

Time period

C3 Mean Alpha Power

0.002.004.006.008.00

10.0012.0014.0016.0018.0020.00

0-5 min 5-10 min 10-15 min

EEG

abs

olut

e po

wer

, µV2

Time period

O1 Mean Theta Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

0-5 min 5-10 min 10-15 min

EEG

abs

olut

e po

wer

, µV2

Time period

O1 Mean Alpha Power


Figure 4.4. Mean levels of subjective sleepiness (KSS) and reaction-time latency (PVT) during the PVT10 (black bars) and the PVT5 (grey bar). Error bars represent standard error of the mean.

1.00

2.00

3.00

4.00

5.00

6.00

7.00

Prior toPVT10

End ofPVT10

End ofPVT5

KSS

val

ue

Time period

Mean KSS

300.00310.00320.00330.00340.00350.00360.00370.00380.00390.00

0-5 min 5-10 min 10-15 min

Rea

ctio

n-tim

e la

tenc

y, m

s

Time period

Mean PVT


Table 4.3. ANOVA Statistics (3 x 6 Mixed Factorial) of the Physiological, Subjective, and Behavioural Data From the two PVT Sessions.

Time period ToD Time period*ToD Planned comparisonsa

Data source F df ηp2 F df ηp

2 F df ηp2 0-5 to 5-10 5-10 to 10-15

EEG

F5 theta 1.98 1.47,179.16 .02 1.18 5,122 .05 0.61 7.34,179.16 .03 - -

F5 alpha 3.99* 1.46,177.55 .03 0.96 5,122 .04 0.79 7.28,177.55 .03 -2.35* -0.68

C3 theta 2.03 2,244 .02 0.30 5,122 .02 0.50 10,244 .02 - -

C3 alpha 4.62* 1.59,194.52 .04 0.27 5,122 .01 1.64 7.97,194.52 .06 -2.88** -1.02

O1 theta 2.52 1.80,218.93 .02 0.38 2,122 .02 1.26 8.97,218.93 .05 - -

O1 alpha 11.06** 1.15,140.80 .08 0.21 2,122 .01 0.35 5.77,140.80 .02 -4.58** -2.99**

KSS 59.15** 1.77,221.41 .32 0.77 5,125 .03 0.72 8.86,221.41 .03 -9.89** 0.64

PVT 38.16 1.85,231.76 .23 1.54 5,125 .06 1.32 9.27,231.76 .05 -8.72** 7.50**

Note. ToD = Time of Day of Testing factor. a Planned comparisons refer to Bonferroni adjusted t-tests. * < .05, ** < .01.



4.4.5 Differences between Genotypes

To examine research question two, the genotype associations with sleepiness

and performance, a series of Univariate ANOVAs were performed to examine

differences between genotypes and the levels of physiological, subjective, and

performance data during the PVT10 as well as the PVT5 testing sessions. Given the

previous results, only the O1 electrode site was examined to limit the number of

analyses performed with the genotype data.

As it can be seen in Table 4.4 (over page), five SNPs demonstrated

significant differences between the different genotypes. Participants with the AA

genotype of rs1554338 (CRY2) displayed a significantly higher mean level of theta

absolute power levels during the PVT5 (M = 18.06, SD = 7.39, n = 110) than

participants with the GA genotype (M = 13.78, SD = 5.66, n = 18), p = .04. In

relation to subjective sleepiness values, participants with the AA genotype similarly

displayed a significantly higher mean level of KSS (M = 6.02, SD = 1.52, n = 113)

compared to participants with the GA genotype (M = 5.00, SD = 1.72, n = 18) at the

end of PVT10 session, p = .01. Last, this trend between the AA and GA genotypes

remained the same at the end of the PVT5 session, the AA genotype having

significantly higher mean level of KSS (M = 5.93, SD = 1.55, n = 113) compared to

participants with the GA genotype (M = 5.11, SD = 1.71, n = 18) at the end of PVT5

session, p = .04.


Table 4.4. ANOVA Statistics (Univariate) for Physiological, Subjective, and Performance Data as a Function of the 21 Candidate SNPs.

EEG O1 electrode site KSS PVT

PVT10 PVT5 PVT10 PVT5 PVT10 PVT5

Theta AP Alpha AP Theta AP Alpha AP KSS KSS MRTL Lapses MRTL Lapses

SNP Gene F df ηp2 F df ηp

2 F df ηp2 F df ηp




2

rs10462020 PERIOD3 1.05 2 .02 0.49 2 .01 1.40 2 .02 0.3 2 .01 1.12 2 .02 2.87 2 .04 2.20 2 .03 1.06 2 .02 1.18 2 .02 1.19 2 .02

rs10774044 CACNA1C 1.08 2 .02 0.11 2 .00 0.35 2 .01 0.14 2 .00 0.25 2 .00 0.01 2 .00 1.04 2 .02 0.03 2 .03 0.49 2 .01 0.55 2 .01

rs1082214 TIMELESS 0.49 1 .00 0.20 1 .00 0.42 1 .00 0.01 1 .00 0.02 1 .00 1.74 1 .01 0.96 1 .01 3.68 1 .02 1.21 1 .01 0.01 1 .00

rs11541353 NPAS2 0.84 2 .01 0.16 2 .00 0.63 2 .01 0.03 2 .00 3.05 2 .05 1.68 2 .03 1.47 2 .02 1.43 2 .01 2.40 2 .04 1.63 2 .03

rs1154155 DQB1*0602 0.89 2 .01 0.96 2 .02 0.50 2 .01 0.71 2 .01 0.53 2 .01 0.15 2 .00 1.28 2 .02 0.44 2 .02 0.77 2 .01 0.59 2 .01

rs1554338 CRY2 3.13 1 .02 1.52 1 .01 4.35* 1 .03 1.85 1 .01 6.69* 1 .05 4.20* 1 .03 0.29 1 .00 2.05 1 .02 0.59 1 .01 2.97 1 .02

rs1799990 PRNP 0.88 2 .01 2.14 2 .03 1.45 2 .02 1.69 2 .03 0.03 2 .00 0.33 2 .01 0.58 2 .01 1.13 2 .02 1.29 2 .02 0.41 2 .01

rs1801260 CLOCK 1.82 2 .03 2.18 2 .03 1.45 2 .02 1.14 2 .02 1.76 2 .03 2.52 2 .04 0.45 2 .01 0.96 2 .02 0.98 2 .02 1.12 2 .02

rs1868049 ARNTL 2.10 2 .03 1.73 2 .03 1.00 2 .02 1.01 2 .02 2.12 2 .03 2.34 2 .04 1.15 2 .02 1.23 2 .02 0.80 2 .01 1.14 2 .02

rs2291738 TIMELESS 0.70 2 .01 0.28 2 .00 0.53 2 .01 0.13 2 .00 0.22 2 .00 1.22 2 .02 0.66 2 .01 1.01 2 .02 0.17 2 .00 4.12* 2 .06

rs2298383 ADORA2A 0.79 2 .01 2.16 2 .03 1.37 2 .02 2.30 2 .04 0.05 2 .00 0.52 2 .01 0.49 2 .01 2.24 2 .04 0.12 2 .00 0.15 2 .00

rs2304672 PERIOD2 0.77 2 .01 0.85 2 .01 1.41 2 .02 0.60 2 .01 0.02 2 .00 0.03 2 .00 0.03 2 .00 0.08 2 .00 0.71 2 .01 0.21 2 .00

rs2735611 PERIOD1 0.75 2 .01 0.62 2 .01 0.79 2 .01 0.45 2 .01 0.03 2 .00 0.39 2 .01 2.51 2 .04 2.75 2 .04 4.00* 2 .06 3.33* 2 .05

rs2797687 PERIOD3 0.01 1 .00 0.05 1 .00 0.03 1 .00 0.03 1 .00 0.01 1 .00 0.03 1 .00 1.00 1 .01 0.76 1 .01 0.77 1 .01 0.50 1 .00

rs3736544 CLOCK 1.67 2 .03 0.36 2 .01 1.14 2 .02 0.33 2 .01 0.11 2 .00 0.27 2 .00 0.63 2 .01 0.55 2 .01 0.34 2 .01 0.59 2 .01

rs3749474 CLOCK 0.77 2 .01 0.31 2 .01 0.24 2 .00 0.36 2 .01 2.59 2 .04 0.71 2 .01 0.02 2 .00 0.02 2 .00 0.70 2 .01 0.50 2 .01

rs4242909 POLE 2.15 2 .03 0.73 2 .01 2.46 2 .04 0.52 2 .01 0.64 2 .01 1.73 2 .03 0.60 2 .01 1.03 2 .02 0.22 2 .00 0.65 2 .01

rs4630333 TIMELESS 0.57 2 .01 1.50 2 .02 1.07 2 .02 1.38 2 .02 0.36 2 .01 0.31 2 .01 0.11 2 .00 0.22 2 .00 0.33 2 .01 4.75* 2 .07

rs4680 COMT 0.07 2 .01 0.38 2 .01 0.08 2 .00 0.31 2 .01 0.16 2 .00 0.63 2 .01 2.00 2 .03 1.93 2 .03 2.51 2 .04 1.38 2 .02

rs5751876 ADORA2A 1.31 2 .02 3.19* 2 .05 1.85 2 .03 3.13* 2 .05 0.12 2 .00 0.71 2 .01 0.92 2 .02 3.37* 2 .05 0.47 2 .01 0.27 2 .00

rs8119787 NFATC2 0.16 2 .00 0.48 2 .01 0.17 2 .00 1.25 2 .02 0.02 2 .00 0.16 2 .00 2.26 2 .03 0.71 2 .01 2.46 2 .04 1.47 2 .02

Note. AP = absolute power; MRLT = mean reaction-time latency; DQB1*0602 = HLA-DQB1*0602. * < .05, ** < .01.



Regarding the PVT data and the SNP rs2291738 (TIMELESS gene),

participants with the TT genotype displayed a significantly higher mean level of PVT

lapses during PVT5 session (M = 3.38, SD = 3.81, n = 37) than participants with the

TC genotype (M = 1.88, SD = 1.92, n = 60), p = .02; but not compared to participants

with the CC genotype (M = 2.00, SD = 2.02, n = 34), p = .45. No other comparisons

between the genotypes for the SNP rs2291738 were significant. Another TIMELESS

gene, SNP rs4630333 was found to have significant differences between its

genotypes. Participants with the TT genotype displayed a significantly higher mean

level of PVT lapses during PVT5 session (M = 3.69, SD = 3.80, n = 26) than

participants with the CC genotype (M = 1.78, SD = 1.82, n = 50), p = .01, but not

compared to participants with the CT genotype (M = 2.20, SD = 2.52, n = 55), p =

.05. No other comparisons between the genotypes for the SNP rs2291738 were

significant.

Significant differences were found for the SNP rs2735611 (PERIOD1 gene),

such that participants with the GA genotype had significantly longer PVT5 mean

reaction-time latencies (M = 373.15, SD = 31.09, n = 35) than participants with the

AA genotype (M = 358.48, SD = 25.65, n = 88), p = .02, but not compared to

participants with the GG genotype (M = 372.80, SD = 21.39, n = 5), p = 1.00.

Moreover, the same participants with the GA genotype had significant more PVT

lapses during the PVT5 session (M = 3.26, SD = 3.46, n = 35) than participants with

the AA genotype (M = 1.98, SD = 2.30, n = 88), p = .04, but not compared to

participants with the GG genotype (M = 3.40, SD = 0.89, n = 5), p = .91.

The last gene to have significant differences between it genotypes was the

SNP rs5751876 (ADORA2A gene), regarding mean values of PVT10 and PVT5

alpha absolute power as well as the PVT10 session lapses. During the PVT10


session, participants with the TT genotype (M = 31.69, SD = 26.78, n = 23) had

significantly higher O1 alpha absolute power levels than participants with the CT

genotype (M = 21.13, SD = 15.76, n = 55), p = .16, but not compared to participants

with the CC genotype (M = 22.98, SD = 13.01, n = 47), p = .56. The comparisons

between the participants with the CT genotype and the CC genotype were not

significantly different, p =.60. For the PVT5 alpha absolute power, participants with

the TT genotype (M = 40.30, SD = 46.36, n = 23) had significantly higher O1 alpha

absolute power levels during the PVT5 than participants with the CT genotype (M =

23.68, SD = 23.52, n = 55), p = .04, but not compared to participants with the CC

genotype (M = 25.83, SD = 17.80, n = 47), p = .08. No other comparisons with the

alpha absolute power levels were significant. Similarly, participants with the TT

genotype (M = 8.26, SD = 8.88, n = 23) had significantly more PVT lapses during the

PVT10 than participants with the CT genotype (M = 4.49, SD = 4.90, n = 55), p = .03

but not compared to participants with the CC genotype (M = 5.77, SD = 5.79, n =

47), p = .81. The comparisons between the participants with the CT genotype and the

CC genotype were not significantly different, p =.69. Table 4.5 displays a

consolidated list of results for the genetic analysis; such that SNPs that were found to

have statistically significant differences between the genotypes are displayed,

including the genotypes frequencies obtained in Study One.


Table 4.5. Consolidated Genetic Results with the Five SNPs, their Genotype Frequencies, and the Various Phenotype Values of the Physiological, Subjective, and Performance Data.

Phenotype Genotype and frequency

SNP Gene Study measure Common homozygote Heterozytote Rare homozygote

rs1554338 CRY2 AA: 113 (86.26%) GA: 18 (13.74%)

PVT5 theta AP (µV2) M = 18.06, SD = 7.39 M = 13.78, SD = 5.66

PVT10 KSS M = 6.02, SD = 1.52 M = 5.00, SD = 1.72

PVT5 KSS M = 5.93, SD = 1.55 M = 5.11, SD = 1.71

rs2291738 TIMELESS TT: 37 (28.24%) TC: 60 (45.80%) CC: 34 (25.95%)

PVT5 lapses M = 3.38, SD = 3.81 M = 1.88, SD = 1.92 M = 2.00, SD = 2.02

rs2735611 PERIOD1 AA: 88 (68.75%) GA: 35 (27.34%) GG: 5 (3.91%)

PVT5 MRTL (ms) M = 358.48, SD = 25.65 M = 373.15, SD = 31.09 M = 372.80, SD = 21.39


rs4630333 TIMELESS CC: 50 (38.17%) CT: 55 (41.98%) TT: 26 (19.85%)


rs5751876 ADORA2A CC: 48 (37.50%) CT: 57 (44.53%) TT: 23 (17.97%)

PVT10 alpha AP (µV2) M = 22.98, SD = 13.01 M = 21.13, SD = 15.76 M = 31.69, SD = 26.78

PVT5 alpha AP (µV2) M = 25.83, SD = 17.80 M = 23.68, SD = 23.52 M = 40.30, SD = 46.36


Note. AP = Absolute power; MRLT = Mean reaction-time latency; bolded values were significantly different.



4.5 Discussion

This first study of the research program examined the effects of motivation as

well as differences on physiological, subjective, and performance data between

different genotypes during a vigilance task. This study examined a relatively large

sample of young ostensibly healthy individuals using a cross-sectional design with

physiological, subjective, and performance measures of sleepiness. The following

discussion examines the outcomes of the two research questions sequentially.

4.5.1 Time and Motivational Aspects

The obtained data clearly revealed an increase in physiological, subjective,

and performance indices of sleepiness over the duration of the PVT10 session.

Specifically, the EEG data revealed an increase of alpha absolute power levels from

period one to period two for the electrode sites of F5, C3, and O1. Similarly,

subjective sleepiness increased from KSS levels obtained at the start of the PVT10 to

levels found at the end of the PVT10. Additionally, performance deteriorated during

the PVT10 with mean reaction-time latencies increasing. These results from the

various measures all substantiate an increase in sleepiness over the duration of the

PVT10.

It must be noted that no statistically significant changes occurred with the

other index of physiological sleepiness, being EEG theta absolute power. However,

as this study was performed during the daytime and participants in this study were

not experiencing any form of experimentally manipulated sleep deprivation, it is not

entirely unexpected that EEG theta absolute power did not increase. Although, it

must be noted the 09:00 start might have resulted in some participants having to

wake up earlier than normal to make this testing time and is potentially a source of

sleep restriction. Nonetheless, as EEG indices of theta and alpha absolute power are


commonly used measures of physiological sleepiness, theta waves are typically seen

in the waking EEG during extreme bouts of sleepiness (Ogilvie & Wilkinson, 1984)

and are an indicator of light sleep or microsleeps (Harrison & Horne, 1996b; Moller

et al., 2006). Thus, the lack of statistically significant change with the EEG theta

absolute power is consistent with previous findings regarding physiological

sleepiness.

The results clearly demonstrate that sleepiness increased over the duration of

the PVT10 despite the exclusion of any protocol-necessitated sleep-deprivation in the

current study design. Typically, the PVT is utilised in studies where the protocol

requires participants to be experiencing sleepiness through either sleep deprivation

(complete, partial, or chronic) or extended wakefulness (Basner & Dinges, 2011;

Dinges et al., 1997; Howard et al., 2014). Moreover, participants typically complete

repeated sessions of the PVT over the duration of the testing sessions. Such studies

often find time-of-day effects, whereby increases in sleep propensity due to circadian

rhythm functioning is associated with poorer performance on the PVT as well as

increases in physiological and subjective sleepiness (Basner & Dinges, 2011;

Caldwell et al., 2003; Kaida et al., 2006). However, surprisingly no time-of-day

effects were found with the obtained data in Study One. Yet, this lack of time-of-day

effects in the current study is not an isolated finding; previous research has also

shown a lack of time-of-day effects on performance of the PVT by sleep-deprived

participants (Corsi-Cabrera, Arce, Ramos, & Lorenzo, 1996; Gorgoni et al., 2014)

and when participants are fully alert (Gorgoni et al., 2014).

While the PVT10 data suggested an increase in sleepiness over the duration

of the PVT10 task, differential effects on the various measures were obtained after

the motivation instruction of the PVT5 session. Participants were asked to “try to


better your previous performance. This means motivating yourself to try even

harder” The EEG data suggested that physiological sleepiness, measured as alpha

absolute power, continued to increase during the PVT5 session, but only at the O1

electrode site was a statistically significant increase observed. Increases in EEG

alpha absolute power were attenuated at the F5 and C3 electrode sites during the

PVT5 session. The potential reasons for this differential effect of the motivation

instruction on the F5 and C3 electrode sites are not immediately obvious. However,

studies that have concurrently employed fMRI scanning and EEG recording have

shown a negative relationship between the EEG recorded alpha rhythm and blood-

oxygen-level dependent (BOLD) activity, such that decreases in the alpha rhythm are

associated with increases in the BOLD activity (Goldman, Stern, Engel Jr, & Cohen,

2002; Laufs et al., 2003). Furthermore, other studies using fMRI scanning, have

demonstrated that additional cortical areas of the dorsolateral prefrontal cortex,

parietal lobes, and subcortical structures are initiated when additional effort is needed

for task performance while experiencing some sleepiness (S. P. A. Drummond,

Brown, Salamat, & Gillin, 2004; Vandewalle et al., 2009). It is possible that under

the motivational PVT5 session participants recruited additional cortical areas to

facilitate better task performance. This recruitment of additional cortical areas could

have resulted in greater activation of frontal and parietal areas which might have

explained the lack of increase in EEG defined sleepiness in the form of alpha

absolute power at the F5 and O3 electrode sites.

Differential effects regarding the motivational instruction were also observed

with the subjective and performance measures. Subjective sleepiness did not increase

from levels found at the end of the PVT10 to levels found at the end of the PVT5,

instead they were sustained, as evidenced by there being no significant change in


KSS levels. This is somewhat inconsistent with previous research which typically

demonstrates that subjective sleepiness increases over the duration of the PVT

(Corsi-Cabrera et al., 1996; Dorrian, Rogers, & Dinges, 2005; Lim et al., 2012). It is

possible that lack of change in the subjective sleepiness levels relates to changes in

the performance data.

A substantial decrease in reaction-time latencies occurred following the

motivational instruction. This finding in the current data is consistent with previous

research which has demonstrated that financial reward (i.e., financial motivation) for

improvement on a low-order cognitive task indeed resulted in cognitive performance

improvement after 24 hours of continual wakefulness (Horne & Pettitt, 1985).

Regarding the relationship between task performance and subjective sleepiness, it

has been suggested that subjective sleepiness mediates self-perceptions of task

performance (Biggs et al., 2007; Dorrian, Lamond, & Dawson, 2000). However,

Dorrian et al. (2000) acknowledged that the causality of this relationship had not

been firmly established.

Considering the results of the present study and the findings discussed thus

far, it is also likely that the direction of the relationship between task performance

and perceptions of subjective sleepiness could be bi-directional. Indeed, in the

current study, performance on the PVT5 significantly improved over the course of

the session and this fact would be known by participants as the reaction-time

latencies are displayed after each response. Certainly, knowledge of task

performance has been known to attenuate the impairing effects of sleepiness. Thus, it

is also possible that knowledge of improved task performance found in the current

study could influence subjective sleepiness ratings.


An alternate explanation with the results obtained following the motivational

instruction can be offered. The time between the PVT10 and PVT5 session included

a KSS rating and the delivery of the motivational instructions. Previous research

demonstrated that interruptions while performing a vigilance task can affect

sleepiness levels. That is, a momentary yet transient increase in arousal occurs when

re-orientating ones attention (Anderson & Horne, 2006; Gillberg et al., 2003; Kaida

et al., 2007; Oron-Gilad, Ronen, & Shinar, 2008). Similarly, simply asking for a

verbal rating of sleepiness has a small effect on reducing physiological sleepiness

levels (Kaida et al., 2007). However, the increase of arousal documented in Kaida et

al.’s (2007) study was transient and only occurred once from the five data collection

points. Nonetheless, the KSS rating and the delivery of the PVT5 instruction could

have influenced the change in sleepiness.

The improved performance levels observed in the current study are more

likely to be due to participants applying extra effort than being due to arousal from

the KSS rating and delivery of the PVT5 instructions. Kaida et al. (2007), in their

study, did not find any improvement of performance levels in relation to any of the

KSS ratings. Similarly, Horne and Burley (2010) did not observe any performance

improvement when their study participants completed a 10-min PVT with KSS

ratings occurring every 2 min. Moreover, Horne and Burley (2010) also reported that

the rating of KSS every 2 min did not reduce (or maintain) the participants subjective

sleepiness levels, which might have been expected with such a constant routine of

rating subjective sleepiness levels. Considered together, the findings from the extant

literature lend support to the conclusion that improvement in performance during the

PVT5 when participants were asked to improve their performance was due to the

instruction rather than the KSS rating and the delivery of the PVT5 instruction.


4.5.2 Genotype, Sleepiness, and Performance

The majority of the genotype associations studies focusing on sleepiness and

performance have been performed during acute sleep loss and these studies have

predominantly focused on the VNTRs of the PERIOD3 gene, or the HLA-

DQB1*0602, and the COMT polymorphisms. In the present research, a number of

novel results were found in the genotype comparisons with the physiological,

subjective, and performance indices of sleepiness.

Differences on physiological as well as performance indices of sleepiness

were observed with the SNP rs5751876 (ADORA2A gene). Specifically, the current

results demonstrate that the TT genotype had higher levels of alpha absolute power

during the PVT10 and the PVT5 as well as more PVT lapses during the PVT10. This

finding of higher EEG defined levels of sleepiness with the TT genotype is consistent

with previous research that has shown the TT genotype to display higher levels of

EEG defined sleepiness in the range of 7-10 Hz (i.e., slow alpha) during sleep and

wakefulness (Rétey et al., 2005). Performance differences have also been noted in

the extant literature between the genotypes of the ADORA2A gene (SNP rs5751876).

That is, several studies have shown participants with the TT genotype displays longer

reaction-time latencies and PVT lapses than participants with the CT genotype when

sleep deprived (Bodenmann et al., 2012; Rupp et al., 2013). However, the functional

effects of the ADORA2A genotypes potentially extend to periods of full alertness

(i.e., non-sleep deprived) as demonstrated in the current results and as with previous

research (e.g., Rétey et al., 2005). Further examinations of the relationships of the

different genotypes with sleepiness and performance are seemingly warranted.

During the PVT5 session, several significant differences were found. The two

SNPs of the TIMELESS gene (rs2291738 and rs4630333) were both found to display


significant differences on the PVT5 lapse measure. No previous research has

specifically examined the TIMELESS gene for it associations with sleepiness and

performance indices and thus, the finding in the current study are novel. Nonetheless,

the TIMELESS gene has been associated with subclinical levels of sleep maintenance

insomnia as well as self-reports of increased levels of excessive daytime sleepiness

(Utge et al., 2010). These results regarding the TIMELESS gene need to be

interpreted with a degree of caution. The TIMELESS gene does not exert a direct

effect on circadian timing but has a subordinate role with circadian functioning (e.g.,

Dardente et al., 2007; Darlington et al., 1998; Honma et al., 2002); this subordinate

role is potentially reflected by previous findings that demonstrate TIMELESS was not

associated with self-reported morningness-eveningness in a population-based

epidemiology study (Pedrazzoli et al., 1999).

The other SNP that was associated with performance differences during the

PVT5 session was the PERIOD1 gene (rs2735611). The PERIOD3 gene has been

extensively studied regarding associations with sleep-wake disorders as well as

cognitive performance during sleep deprivation. Several of these studies have

demonstrated differences of levels of cognitive performance, sleep-wake timing, and

sleep disorders with the different VNTRs of the PERIOD3 gene (Archer et al., 2003;

Rupp et al., 2013; Vandewalle et al., 2009). The present results contribute to the

extant literature by demonstrating a functional difference on the PVT5 performance

measures of mean reaction-time latency and PVT lapse occurrence between different

genotypes of the SNP rs2735611 of the PERIOD1 gene.

Regarding previous findings concerned with the PERIOD1 gene, only one

other study that examined diurnal preference (i.e., chronotype) and its association

with specific circadian genes has found any difference between the PERIOD1


genotypes. Carpen, Schantz, Smits, Skene, and Archer (2006) reported that the major

allele of the PERIOD1 gene (rs2735611) was found more often in individuals with

extreme morning preference than in subjects with extreme evening preference

(Carpen et al., 2006). Research has also demonstrated that morning chronotypes tend

to have a more consistent level of simulated driving performance throughout

repeated daytime driving sessions (Correa et al., 2014) as well as during the

performance of working memory tasks (Lara et al., 2014). Nonetheless, while the

current findings are highly preliminary, considering the extant literature concerned

with the PERIOD1 gene, further research is needed regarding the effects of

motivation on task performance.

4.5.3 Study Implications

The implications from the present study mainly relate to the distinct effect of

the motivation instruction. Following the motivation instruction, there was a clear

improvement of performance on the PVT. Thus, it appears from the obtained results

that the manipulation of one aspect of the wake drive, being motivation, was

successful in negating increases in subjective sleepiness as well as physiological

sleepiness (except for the occipital site). However, several aspects of how effective

the wake drive can be with negating sleepiness remain. It is still unknown how

effective augmentation of the wake drive (excluding the use of psychopharmacology

stimulants) can be in negating sleepiness when the sleep drive is increased (i.e.,

during sleep deprivation). This is the next aspect of the four-process model that

needs to be examined and potentially is the most important aspect in need of

empirical examination. That is, there are numerous situations where sleepiness can

naturally increase (e.g., post-lunch dip, working night shift, long-distance driving)


and reducing sleep drive with the use of naps to is not easily implemented nor widely

accepted (e.g., napping at work).

The effect of motivation on high-order cognitive task performance while alert

and while sleepy also needs to be systematically examined. As stated earlier,

performance of a high-order cognitive task requires the coordination of several

psychological processes and accordingly additional cognitive resources needed for

the successful performance of a high-order cognitive task (Galy et al., 2012; Thomas

et al., 2003). Thus, it is currently unclear if motivation can overcome sleepiness-

related impairments on a high-order cognitive.

The genetic analyses suggested that a number of genes were associated with

differences of physiological and subjective sleepiness as well as performance levels.

Due to the small sample size (discussed below in more detail), the implications of

this study are quite tentative. The ADORA2A gene (rs2735611) results were

promising in terms of identifying a gene that could confer vulnerability to sleepiness.

Several studies have demonstrated that ADORA2A genotypes have functioning

differences in terms of physiologically sleepiness and performance levels on the PVT

(Bodenmann et al., 2012; Rétey et al., 2005; Rétey et al., 2007; Rupp et al., 2013)

with these functional differences holding during periods of sleep deprivation as well

as when fully alert. However, the unique findings observed with the CRY2,

PERIOD1, and TIMELESS genes need to be interpreted with some caution given the

paucity of other studies corroboration the present study’s findings.

4.5.4 Limitations and Future Research Directions

The main limitation of Study One was the limited sample size. Certainty, the

131 participants were enough for the purposes of the psychophysiological and

performance analyses. However, genetic association studies require considerably


more participants than 131 participants and thus, the current study is considered a

small sample. The limited sample size also meant the interaction terms were not

tested with the physiological, subjective, and performance data against the genetic

data. Given the extant literature, it is likely that certain genotypes of the COMT or

HLA-DQB1*0602 genes might displays differential effects of motivation on

sleepiness and performance indices. Exploring these possibilities might enhance

current understandings of the effects of genes with sleepiness and performance.

Another limitation was the shorter duration of the PVT5 used in Study One. The

decision to use the PVT5 with the motivation instruction was due to time constraints

given the arduous testing schedule outlined in section 4.3.1. However, the decision to

utilise the PVT5 was also a considered one, as previous studies have shown the

PVT5 is a sensitive test of changes in levels of sleepiness (Lamond, Dawson, &

Roach, 2005; Loh et al., 2004). Nonetheless, extending the duration of the motivation

session to 10 min or more will provide new information regarding how long

participants can upregulate their wake drive to increase their level of performance.

Last, the lack of a measure for motivation or effort the participants were applying to

the task was a limitation of the study. Having a measure of the participants’ effort

level could have furthered the understanding of the relationships between motivation,

sleepiness, and performance levels.

4.6 Chapter Summary

This chapter has described the first study of the research program. This study

examined changes in physiological, subjective, and performance indices of

sleepiness while participants performed, in total, a 15-min vigilance task. The study

addressed both research questions, being what are the effects of motivation on

sleepiness and performance levels?, and what genetic differences are associated with


sleepiness and performance levels? Both motivations and genetics have been

previously shown to be integral components of sleep-wake behaviours and

performance levels.

The overall finding from the first study suggests that motivations to improve

performance on a low-order cognitive task had a partial effect on physiological and

subjective sleepiness, with a larger effect occurring on performance indices of

sleepiness. Specifically, motivation to improve performance levels on the PVT5

seemingly attenuated physiological sleepiness at the frontal and central electrode

sites, with physiological sleepiness continuing to increases at the occipital electrode

site. Moreover, the motivation instruction was associated with attenuation of

subjective sleepiness and a substantial improvement of performance on the PVT5 as

demonstrated with significant reduction of reaction-time latencies.

Regarding the genetic results, overall five SNPs were found to be associated

with differences of physiological, subjective, and performance indices of sleepiness.

The SNP rs5751876 (ADORA2A) was found to have differences on physiological and

performance measures which were consistent with extant research findings. The first

novel finding in the current study relates to the associations between SNPs

rs2291738 (TIMELESS), rs2735611 (PERIOD1), rs4630333 (TIMELESS) genotypes

with performance measures during the PVT5 session. A second novel finding relates

to the SNP rs1554338 (CRY2 gene) where the genotypes were found to exhibit

differences with the physiological and subjective sleepiness measures.


Chapter Five: Driving Cessation Motivations Study

What hath night to do with sleep?

–John Milton


This chapter presents the findings of the driving cessation motivations study.

This study was based on the reanalysis of an existing dataset 1. The data was

originally collected to examine if drivers could choose to cease driving while

undertaking a Hazard Perception Test after partial sleep restriction. The driving

cessation motivations study addresses the first research question, which is concerned

with the effect that motivation can have on sleepiness and performance. In essence,

the driving cessation motivation study removes the participants’ motivations to

continue driving when sleepy. Thus, understanding the psychophysiological changes

in sleepiness when external motivations to continue driving are removed is important

in the context of sleepy driving behaviours.

5.2 Introduction

As noted earlier, the effects of sleepiness have been shown to have a number

of detrimental neurobehavioural outcomes. Specifically, sleep deprivation decreases

alertness (Åkerstedt & Gillberg, 1990; Kribbs & Dinges, 1994), impairs cognitive

functions (M. L. Jackson et al., 2013), and reduces psychomotor performance

(Dinges et al., 1997). As a consequence of these reductions in alertness, cognitive

1 The data collection period for Study One was prior to the author’s PhD candidature; see S. S. Smith et al. (2011) for the original study. The author of the present dissertation was a principal investigator and collected the data for the original study. He subsequently reanalysed this data during his PhD candidature because of the motivational aspect of the instruction used.


functioning, and psychomotor performance from sleepiness, driving performance is

also impaired.

Driving is a task-oriented activity, such that, when a driver begins driving,

they task themselves (i.e., task set) to drive from point A to point B. Consequently, a

driver’s motivations to reach their destination has been shown to be important facet

for choosing to continue to drive while sleepy (McCartt et al., 2000; Watling,

Armstrong, et al., 2014). Indeed, Watling, Armstrong, et al. (2014) demonstrated that

motivations to drive while sleepy have the strongest relationship with continuing to

drive while sleepy, over and above their crash risk perceptions for driving while

sleepy.

Similarly, the task instruction given to a participant could influence the

participant’s behaviour when performing a simulated driving task. Due to the

monotonous nature of sleepy driving research scenarios, it is often found that sleep

deprived participants’ levels of subjective sleepiness increase quickly during the

driving task. Specifically, extreme levels of subjective sleepiness (measured on the

KSS) such as level 8 (sleepy, some effort to stay awake) or 9 (very sleepy, great

effort to keep awake, fighting sleep) are often reported at the end of driving in many

studies (e.g., Howard et al., 2014; Ingre et al., 2006; Watling, Åkerstedt, et al., 2015).

When the participants’ task set is to “complete a driving task”, it is likely that once

the individual reaches a high level of sleepiness they would apply extra effort to

remain awake during periods when they are ‘fighting’ sleep onset due to their sleep

deprivation (e.g., Anund, Kecklund, Peters, Forsman, et al., 2008; Arnedt et al.,

2005; Lowden et al., 2009). Applying extra effort to stay awake and fight sleep onset

can be considered to correspond to increased activity of the wake drive.


As discussed earlier, the four-process model (Johns, 1993) suggests that the

conscious state of sleep or wakefulness is determined by the outcome of two

opponent processes, the sleep and wake drives. Regarding the primary and secondary

wake drives of the four-process model, it is suggested that the secondary wake drive

is partially under voluntary control by the individual. Fighting sleep onset and

applying extra effort to the performance of a task is likely to involve augmenting the

secondary wake drive (i.e., increased mental stimulation) to counteract the increased

drive to sleep.

An aspect of sleepy driving that has not thoroughly been examined is when

drivers choose to cease driving before reaching an extreme level of sleepiness.

Specifically, the progression of sleepiness once an individual does not have to fight

sleepiness by augmenting their wake drive is unknown. As such, the aim of Study

Two was to examine the changes in physiological and subjective sleepiness

associated with cessation from a hazard perception task.

5.3 Method

5.3.1 Design

A 2 x 2 mixed factorial design was used that included a within-subjects factor

of Time Period (baseline and cessation) and a between-subjects factor of Time of

Testing (a.m. or p.m.). The between-subjects factor of Time of Testing (a.m. or p.m.)

was included to assess for the associated increases in sleepiness that occur with the

descending phase of the circadian rhythm and increased homeostatic drive as the data

was collected during the morning and afternoon. An equal number of participants

were randomly assigned to undertake the testing during the morning (n = 13, starting

at 09:00) and in the afternoon (n = 13, starting at 14:00). The dependent variables

were physiological (i.e., EEG: theta and alpha absolute power for the electrode site


of O1; EOG: blink duration; and ECG: interbeat-interval or R-R) and subjective (i.e.,

KSS) measures of sleepiness. Unfortunately, no performance data could be used in

this study (see section 5.3.3.4 for further details). If the sphericity assumption was

breached the Greenhouse-Geisser correction was reported.

5.3.2 Participants

Participants were recruited into the study by an email sent via the University

student intranet. Participants were excluded from participating if they had a habitual

bedtime later than 12 midnight, any significant health problems or a sleep disorder,

had excessive daytime sleepiness (assessed as a Epworth Sleepiness Scale of > 10:

Johns, 1991) or had sleeping difficulties (assessed as a Pittsburgh Sleep Quality

Index score of > 5: Buysse et al., 1989), were a shift worker, had travelled overseas

in the past month, took prescription medications that altered arousal levels or illicit

drugs, or drank more than three cups of coffee per day and/or more than two standard

drinks of alcohol per day.

Overall, 26 participants (19 females and 7 males) took part in the study. The

mean age of participants was M = 23.77 years (SD = 2.32; range = 20-28). On

average the participants usual weekday bedtime was 22:29 (SD = 39.00 min) and

usual weekday wake-time was 06:52 (SD = 51.00 min), and thus the average normal

weekday sleep duration for the participants was M = 508.27 min (SD = 41.81; range

= 420-585) as assessed by the Sleep Timing Questionnaire (Monk et al., 2003). All

participants had a valid driver licence with a mean duration of licensure being M =

5.65 (SD = 2.46; range = 2-10), and all participants were current drivers. Participants

were compensated 100 AUD for taking part the study.


5.3.3 Measures



sex, duration of having a licence (including duration of their learner licence), number

of hours driven per week.


The physiological measures used in the current study were EEG, ECG, and

EOG. The EEG data was used to quantify the absolute power of the theta and alpha

frequency bands and a spectral analysis was performed on the O1-A2 derivations.

The EOG data was used to quantify blink duration and the ECG data was used to

quantify the interbeat-interval or R-R duration. Please see sections 3.1 for specific

details as well as issues of reliability and validity regarding these measures and

section 3.1.4 for the details of data acquisition and processing. Figure 5.1 shows the

electrode sites recorded in Study Two.

Figure 5.1. EEG electrode sites utilised in the driving cessation study.



The subjective measures used in the current study were the KSS, ESS, PSQI,

STQ, and the SoSQ. The KSS was used to quantify the participants’ levels of

subjective sleepiness. The ESS and PSQI were used as screening measures for

excessive daytime sleepiness, while the STQ was used to quantify habitual sleep

behaviours. The SoSQ quantified the importance that participants believed certain

signs of sleepiness were as indicators of their level of sleepiness. Please see section

3.2 for specific details of each measure as well as issues of reliability and validity.

5.3.3.4 Performance measures

The HPT was used as the performance measure of sleepiness and was the

driving stimulus as well; however, the obtained data could not be analysed due to

varying durations of participants driving sessions. This resulted in some participants

not having enough data points required to ensure a minimum level of confidence in

the internal consistency of the HPT scores and therefore the HPT data could not be

analysed (M. S. Horswill, personal communication, July 14, 2014). Please refer to

section 3.3 for specific details of this measure as well as issues of reliability and

validity of this measure.

5.3.4 Procedure



0800000823) prior to beginning the study as well as the Health and Safety approval.

On the day of testing, the participant awoke at 05:00 and abstained from

ingesting any caffeine or alcohol until completion of the testing session. Upon arrival

at the laboratory, the participant had the EEG, EOG, and ECG electrodes attached.

Before starting the HPT, the participant was instructed to “stop when you think you


would be too sleepy to drive safely on the road”. This instruction can be considered a

manipulation of the wake drive, such that the participant does not have to increase

their wake drive to continue the simulated drive when they are highly sleepy. The

participant’s subjective sleepiness was assessed with the KSS before starting the

hazard perception task. At the beginning of the driving task, the participant watched

a 5-min instructional video showing how to complete the HPT, which included two

examples of “typical” hazards. This was intended to reduce any ambiguities

regarding what constituted a hazard that the participant was required to identify. The

standardised 5-min instructional video has been validated in numerous research

studies (e.g., Horswill et al., 2008; S. S. Smith et al., 2009; Wetton et al., 2010).

Each participant completed the testing alone. When the participant chose to

cease driving, they spoke into a microphone they were too sleepy and were ready to

stop, at which point the experimenter noted the duration of driving, then re-entered

the laboratory and administered the KSS. Lastly, the participant was thanked for their

participation in the study before leaving the laboratory. Each participant was seated

in a comfortable high-backed chair in an environment with controlled sound,

temperature, and light (see section 3.5).

Figure 5.2 (over page) displays an overview of the timeline of Study Two and

the points where subjective sleepiness ratings were obtained and the sections of

physiological recordings used in the subsequent analyses. A duration of 5-min was

chosen for the baseline and cessation episodes as all participants duration of driving

were different, which prevented comparisons of equal time durations if for instance

the first half of driving was to be compared to the last half of driving. Moreover, as

the participant was instructed to stop once they thought they were too sleepy it was

reasoned that the last five minutes prior to cessation would encompass the period of


greatest sleepiness and likewise the baseline period would encompass the time when

the participant was most alert during the study. Lastly, previous research has

similarly used a 5-min baseline and driving cessation periods for statistical

comparisons (e.g., Åkerstedt et al., 2013).

Figure 5.2. Overview of the timeline of Study Two. Not to scale.

5.4 Results


The EEG, EOG, ECG, and KSS data were examined for their distribution

properties with all but the EEG alpha absolute power data displaying a satisfactory

normal distribution. The EEG alpha absolute power data had a slight positive skew,

which was corrected with a square root transformation. Separate ANOVAs were run,

with the transformed and untransformed EEG alpha absolute power data. However,

there was no change the overall significance of the ANOVA results with either data.

Thus, the untransformed EEG alpha absolute power data was used in the analysis.

5.4.2 Sleep Prior to Testing

In total, the amount of sleep obtained prior to testing was approximately 6.50

hours (M = 390.96; SD = 39.12 min) and the amount of sleep debt the participants

were experiencing was approximately two hours (M = 117.31 min; SD = 36.75) as

determined by a simple sleep diary. No significant difference was found between the

duration of sleep prior to the morning (M = 387.69, SD = 38.17) and afternoon

testing session (M = 394.23, SD = 41.32), t(24) = -0.42, p = .68. Similarly, no

Hazard Perception test

KSS 1 KSS 2

Baseline 5 min

Cessation 5 min

Setup


significant difference was found between the morning (M = 125.38, SD = 34.31) and

afternoon testing session participants (M = 109.23, SD = 38.67), t(24) = 1.13, p = .27

for the amount of sleep debt. As such, the estimated need for sleep between the

morning and afternoon participants was considered equivalent.

5.4.3 Sleepiness and Cessation of the Task

On average the participants chose to cease the hazard perception task after

approximately 40 min (M = 36.13; SD = 17.70; range = 12.50-73.00). An

independent samples t-test found no time-of-day effects with the duration of driving

between the morning (M = 34.58, SD = 14.47) and afternoon durations of driving (M

= 37.69, SD = 20.92), t(24) = -0.44, p = .66. Visual inspection of the EEG and EOG

data revealed that none of the participants could be judged to have fallen asleep by

current sleep staging criteria (i.e., American Academy of Sleep Medicine & Iber,

2007) or having experienced any microsleeps (i.e., 3-15 sec duration of theta

activity) before the point of driving cessation. This suggests that participants did not

fight sleepiness during the testing session.

5.4.4 Physiological and Subjective Data

A series of 2 x 2 mixed factorial ANOVAs were performed with a within-

subjects factor of Time Period (baseline or cessation) and a between-subjects factor

of Time of Testing (a.m. or p.m.). Figure 5.3 (over page) displays the mean and

standard error of the mean values of the physiological and subjective data and Table

5.1 displays the mixed factorial ANOVAs results. The level of subjective sleepiness

increased significantly from the beginning of the simulated drive to cessation of the

task, this was a large effect. A significant increase in blink duration was found from

baseline to cessation, but the effect size was smaller in magnitude than the subjective

sleepiness effect size. No other significant effects were observed in the data.


Figure 5.3. Mean levels of EEG absolute power, EOG eye blink duration, ECG R-R interval, and subjective sleepiness (KSS) during the baseline (white bars) and cessation (black bars) time periods. Error bars represent standard error of the mean.

Table 5.1. ANOVA Statistics (2 x 2 Mixed Factorial) of the Physiological and Subjective Data During Baselines and Cessation Periods.

Baseline-Cessation (within factor)

Time of Testing (between factor)

Baseline-Cessation* Time of testing


2 F df ηp2

Physiological sleepiness EEG theta 4.13 1,24 .11 2.61 1,24 .10 2.54 1,24 .10 EEG alpha 1.48 1,24 .06 3.19 1,24 .09 2.39 1,24 .10 EOG blink duration 9.84** 1,24 .29 0.99 1,24 .04 1.65 1,24 .06 ECG R-R interval 0.08 1,24 .01 0.30 1,24 .01 0.58 1,24 .02

Subjective sleepiness KSS 134.21** 1,24 .85 2.04 1,24 .08 .09 1,24 .01

* < .05, ** < .01.

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

Theta Alpha

EEG

abs

olut

e po

wer

μV2

Baseline-cessation data

EEG O1 Mean Power

0.00

20.00

40.00

60.00

80.00

100.00

120.00

140.00

EOG baseline EOG cessation

ms


EOG Eye Blink Duration

0.00100.00200.00300.00400.00500.00600.00700.00800.00900.00

ECG baseline ECG cessation

ms


ECG R-R Interval

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

KSS baseline KSS cessation

KSS

vla

ue


Mean KSS


5.4.5 Signs of Sleepiness

Participants rated the importance of nine signs of sleepiness as an indicator of

their level of sleepiness prior to and after ceasing the driving task. Figure 5.4

displays mean scores for the importance of the nine signs of sleepiness. To determine

if the importance of the signs of sleepiness changed after performing the driving task,

a 2 x 9 repeated measures ANOVA was performed; the factors being, Baseline-

Cessation Ratings and Signs of Sleepiness (the 9 signs of sleepiness, see Figure 5.4,

over page). The Greenhouse-Geisser correction was reported if the Sphericity

assumption was breached and post-hoc tests used the Bonferroni adjustment for

multiple comparisons.

A significant main effect was found for the Baseline-Cessation Ratings factor

F(1, 25) = 15.99, p < .001, ηp2 = .39; indicating the level of importance of the signs

of sleepiness increased from levels at the beginning of the driving task (baseline) to

levels found at the completion of the driving task (cessation). A significant main

effect was found for the Signs of Sleepiness factor F(5.30,132.41) = 31.66, p < .001,

ηp2 = .56; suggestive that the individual signs of sleepiness were rated differently for

their level of importance. The main effect for the Signs of Sleepiness factor was not

examined further, as the importance of each of the individual signs of sleepiness

were not a focus of the analyses, only the change in the ratings of importance. The

interaction between the Pre-Post Task and Signs of Sleepiness factors was significant

F(4.71,117.70) = 20.86, p < .001, ηp2 = .46. Thus, Bonferroni adjusted post-hoc

comparisons were performed on the pre- and post-driving importance values of each

individual signs of sleepiness. Overall, the importance ratings of the signs of

sleepiness ratings of slow eye blinks (t(25) = -4.87, p < .01), difficulty concentrating

(t(25) = -3.78, p < .05), slower reactions (t(25) = -7.63, p < .01), changing position


(t(25) = -7.68, p < .01), and frequent eye blinks (t(25) = -10.09, p < .01) increased

from levels found at the beginning of the driving task to levels reported after ceasing

the driving task.

Figure 5.4. Importance of the nine signs of sleepiness as an indicator of the level of sleepiness prior to (white bars) and after ceasing the driving task (black bars). Error bars represent standard error of the mean.

5.5 Discussion

The main finding from this study was that all participants chose to cease the

hazard perception task before unintended sleep onset occurred. After mild sleep

restriction, the participants reported moderate subjective sleepiness at the start of the

task. The participants’ subjective sleepiness increased during the driving sessions to

a high level of subjective sleepiness at driving cessation (corresponding KSS anchor

of sleepy, some effort to stay awake). Participants appeared able to identify increases

in subjective sleepiness, and were all able to cease the hazard perception task before

unintentional sleep onset occurred. None of the participants were judged to have

fallen asleep by standard sleep staging criteria (i.e., American Academy of Sleep

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

Impo

rtan

ce

Sign of sleepiness


Medicine & Iber, 2007) during the hazard perception task prior to cessation; that is,

there were no sudden sleep attacks or microsleeps, which suggests that participants

were not fighting sleepiness for an extended period of time.

5.5.1 Physiological and Subjective Sleepiness

The aim of Study Two was to examine the changes in physiological and

subjective sleepiness associated with cessation from a hazard perception task. In this

study, it was found that subjective sleepiness increased for each participant over the

duration of the driving session. It is noted that subjective sleepiness levels were

slightly elevated at the beginning of the task (6.65 on the KSS) which corresponds to

scores (and verbal anchors) on the KSS of 6 (some signs of sleepiness) and 7 (sleepy,

no effort to stay awake). These higher subjective sleepiness levels are likely due the

sleep restriction protocol. Subjective sleepiness levels were found to be significantly

higher (8.15 on the KSS) at the point of cessation from the hazard perception task.

Regarding the physiological changes associated with cessation of the hazard

perception task, only the blink duration measure increased significantly. This

increase in blink duration over the duration of the hazard perception task is consistent

with previous sleepy driving research that has found blink durations increase with

increases of sleepiness (Ingre et al., 2006; Sandberg et al., 2011). In contrast, no

significant changes in EEG or ECG indices of sleepiness were found, which was

contrary to previous work (e.g., Kecklund & Åkerstedt, 1993; Tran et al., 2009). One

potential explanation for the lack of effect with the EEG measure could be the level

of sleepiness the participants were experiencing at the beginning of the test session.

That is, the physiological defined sleepiness could have already been relatively high

as the participants were moderately sleep deprived and therefore, would not have

changed considerably.


The discrepancy between the physiological and subjective measures deserves

examination. While a detailed discussion of the potential reasons why physiological

and subjective measures of sleepiness do not always correspond is outside the scope

of this thesis, it is necessary to discuss some key reasons for the lack of

correspondence. One potential reason for inconsistencies between physiological and

subjective sleepiness could be due to poor self-awareness of level of sleepiness.

Evidence that supports this notion comes from findings that demonstrate some

individuals have limited awareness of their own sleepiness and typically report

experiencing fewer signs of sleepiness when performing a monotonous computer

task after only two hours of sleep (Kaplan et al., 2007). Additionally, sleep-deprived

individuals’ perceptions of sleepiness does not always relate to objectively defined

EEG theta bursts of greater than 3 sec (Herrmann et al., 2010). However, earlier

research has shown that perceptions of the transition between the states of sleepiness

and sleep are rather imprecise (Bonnet & Moore, 1982). What these studies suggest

is that at extreme levels of sleepiness when the individual is likely fighting sleep

onset, perceptions of subjective sleepiness could be erroneous.

Another potential explanation for the inconsistencies between physiological

and subjective findings could be due to the effect from arousal or rather the wake

drive. As discussed earlier the four-process model (Johns, 1993) suggests that the

conscious state of sleep or wakefulness is determined by the outcome of two

opponent processes, the sleep and wake drives. The four-process model allows for

the situation where an individual could be sleepy, and thus their EEG recordings

would display a certain level of EEG defined sleepiness (i.e., alpha or theta activity),

although, the individual could experience some kind of arousal that could manifest

more prominently in subjective sleepiness assessments. Indeed, Yang et al. (2004)


have demonstrated that a “calm-down” procedure (participants seated quietly with

their eyes closed), resulted in several moderate correlations between the subjective

sleepiness and objective measures of sleepiness, such as sleep onset latency of a nap

and reaction times from the continuous performance test. Whereas, no correlations

were found between subjective sleepiness and objective measures with participants

that did not perform the calm-down procedure in Yang et al.’s (2004) study.

Considered together, arousal could mask underlying sleepiness that the individual is

experiencing.

5.5.2 Study Implications and Driving Motivation

It is likely that the instructions and the task set given to participants had an

effect on the participant’s awareness of their sleepiness and decision-making. The

instruction that participants received prior to driving was to “stop when you think

you would be too sleepy to drive safely on the road” The participants’ only task in

this study was to monitor their sleepiness levels while completing the simulated

driving task. As previously discussed, no participant could have been judged to have

fallen asleep by standard sleep scoring criteria nor were any mircosleeps observed.

Additionally, it was found that the ratings of importance of certain signs of

sleepiness as indicators of subjective sleepiness levels increased from prior to after

the simulated driving task. This suggests that participant might have been more

conscious of these signs of sleepiness during the simulated drive, and thus the greater

awareness might have assisted the participants with deciding when to cease the

driving task.

Previous work suggests that motivation to reach a destination is an important

reason drivers cite for continuing to drive when they are aware of their elevated

sleepiness levels (Nordbakke & Sagberg, 2007; Watling, Armstrong, et al., 2014). It


is likely that increased motivation could have a two-fold effect, being a) a focusing

of awareness away from perceiving the signs of sleepiness; and b) obscuring or

biasing a driver’s decision-making processes. Regarding the first point, the present

data revealed that the importance of several signs of sleepiness as indicators of the

level of sleepiness increased from the beginning of the drive to the point of cessation.

However, the data set from Study Two only had an “non-motivated” condition and

thus a comparison of motivated and non-motived condition could not be made.

Nonetheless, it has previously been suggested that sleepy drivers consider the signs

of sleepiness as inconsequential (Dinges, 1995) and partially supports the obtained

results.

The second point that motivation obscures or biases a driver’s decision-

making processes is predicated on previous research findings. It has been noted in

the literature that drivers often fail to recognise the seriousness of driving while

sleepy (Horne & Reyner, 2001). It has previously been shown that sleepy driving is

typically rated as a low crash-risk factor, below speeding and drink driving which are

subject to strictly enforced legal limits (Pennay, 2008; Radun et al., 2015; Watling,

Armstrong, Smith, et al., 2015). Additionally, sleepiness leads to increases in risk-

taking propensity as well as to poorer decision-making (Harrison & Horne, 2000;

Rossa, Smith, Allan, & Sullivan, 2014), and both could effect a sleepy driver’s

decision-making processes. In this study, reaching a “destination” was not an explicit

requirement of the task, only stopping when the participants felt too sleepy to drive

safely. Subsequently, it was observed that all participants reported increases in

subjective sleepiness, perceived a number of signs of sleepiness, and were able to

choose to cease the hazard perception task before unintended sleep onset or a

microsleep occurred.


The duration of driving before cessation was relatively short, with most

participants stopping within one hour. Similar results have been reported with an on-

road night-time driving study with participants stopping driving prematurely after 43

min due to high levels of sleepiness (Åkerstedt et al., 2013). Survey data suggests

one third of drivers report having briefly fallen asleep within one hour of beginning

driving (Pennay, 2008). Current road safety campaigns recommend drivers take a

break after every two hours of driving. The current data suggests that drivers can

experience significant sleepiness levels well within this limit, and drivers may need

to be better informed about such rapid onset of sleepiness - particularly when driving

in rural locations. Although, laboratory conditions invoke lower arousal levels than

on-road driving situations (Hallvig et al., 2013; Philip et al., 2005) and this effect

needs to be considered with interpreting the current results from Study Two.

Another factor that could have influenced the participant’s decisions to cease

driving was the cash payment for partaking in the study. Participants could choose

when to stop the task, and it is possible that some participants ceased driving when

they thought they had completed enough of the study. That is, participants may have

only completed a certain amount of testing to ensure their payment for participation.

The motivational effects of participant payments have been noted previously (e.g.,

Ripley, 2006); however, the exact impact of participant payments for engagement in

task outcomes is not well understood (McKeganey, 2001). Future work may need to

consider this potential effect.


Study Two had three key limitations. Firstly, it lacked a performance measure

for the simulated driving task and could not address the performance aspect of

research question one. It is possible that an individual’s decision to cease driving due


to sleepiness could be based on their perception of their driving performance, or

levels of subjective sleepiness, or both. It has been suggested that perceptions of

subjective sleepiness mediates the perceptions of task performance (Dorrian et al.,

2000). A second limitation of the current study was the lack of a comparison

condition. A second condition that, for example instructed participants to “complete

the driving task” would have provided additional information regarding the

progression of sleepiness, in particular physiological sleepiness. Thirdly, the study

design only collected subjective sleepiness data at the beginning and end of the

hazard perception task; therefore, it is unknown for how long participants were at the

highest level of nine on the KSS.

The design and procedure of Study Three took into consideration many of the

issues raised by Studies One and Two. Addressing the issues raised by Study One,

the final study included a manipulation of the wake drive, but also a manipulation of

the sleep drive via partial sleep deprivation. These manipulations ensured that the

effects of motivation on sleepiness and performance levels were examined both when

the participants were alert and when they were sleepy. In addition, in Study Three a

low-order cognitive task (the PVT) was employed as well as a high-order cognitive

task (the HPT), which allowed the effect of motivation to be examined on both low-

and high-order cognitive tasks when the participants were alert and when they were

sleepy. Thus, Study Three extends on Study One with a more thorough examination

of the effect of motivation. Regarding the issues from Study Two, in Study Three it

was possible to analyse the driving performance data from the Hazard Perception

Test because all participants completed the task in the same length of time. In

addition, Study Three included several measures of participant motivation or effort

level which was lacking in the previous two studies.


5.6 Chapter Summary

This chapter has described the second study of the research program. The

study reanalysed an existing data set to examine the changes in physiological and

subjective sleepiness associated with cessation from a hazard perception task. The

driving cessation motivations study addresses the first research question, which is

concerned with the effect that motivation can have on sleepiness and performance.

The instruction given to participants in this study was to monitor their sleepiness and

stop driving once they perceived their level of sleepy was too dangerous to drive

safely on the road. Specifically, this instruction removes the participants’ motivations

to continue performing the task or to continually fight sleep onset to continue

performing the task; and thus, understanding the psychophysiological changes in

sleepiness when external motivations to continue driving are removed are important.

The main finding from this study was that all participants chose to cease the

hazard perception task before unintended sleep onset or microsleeps occurred. The

results suggest that after mild sleep restriction participants’ subjective (KSS) and

physiological (blink duration) sleepiness increased over the duration of the drive, but

no significant change in EEG or ECG data was found. It was also found that the

importance of several of the signs of sleepiness as indicators of the individual level

of sleepiness similarly increased from the start of the hazard perception task to the

point of cessation. In the absence of motivation to continue to drive while sleepy,

participants in this study were able to identify increasing subjective sleepiness levels,

and were able to choose to cease driving before unintended sleep onset or

microsleeps occurred, and this was associated with an increase of blink duration and

greater awareness of the signs of sleepiness. Although, due to the study’s limitation

of not having a second alternative condition that instructed participants to complete


the driving task the study results should be interpreted with a degree of caution.

Nonetheless, the current study suggests that motivation to cease driving is an

important aspect of sleepy driving behaviours.


Chapter Six: Motivation to Continue Driving While Sleepy, the

Effects on Sleepiness and Performance

It does not matter how slowly you go as long as you do not stop.

–Confucius


This chapter presents the results of the final study of the PhD research

program. This study’s contribution to the research program is the direct experimental

manipulation of the wake drive via the participants’ motivational levels as well as a

manipulation of the sleep drive via partial sleep deprivation. This final study

addressed the first research question regarding the magnitude and duration of the

effects motivation has on sleepiness and performance levels. Several hypotheses

were proposed to examine the effect of motivation on sleepiness and performance,

and they are outlined in the following sections.

6.2 Introduction

A number of studies have demonstrated that some drivers choose to continue

to drive while sleepy despite being aware of an increasing level of sleepiness.

Specifically, a remarkable proportion of individuals (59-77%) report driving when

feeling sleepy (Armstrong et al., 2010; Vanlaar et al., 2008; Watling, Armstrong, &

Radun, 2015). Moreover, a large proportion of drivers (69-73%) report that they

would continue to drive even when aware of increasing levels of sleepiness

(Nordbakke & Sagberg, 2007; Watling, Armstrong, & Radun, 2015).

A number of survey studies suggest that motivation to continue driving while

sleepy is an important factor for performing sleepy driving behaviours (Armstrong et


al., 2010; Fletcher et al., 2005; Nordbakke & Sagberg, 2007; Watling, Armstrong, et

al., 2014). The driver’s motivation to reach their destination is a pertinent factor in

their decision to continue driving while sleepy, overshadowing their perception of

crash risk associated with sleepy driving (Watling, Armstrong, et al., 2014). Indeed,

the study by Watling et al. (2014) demonstrated that younger drivers had stronger

motivations to continue driving while sleepy as well as lower crash risk perceptions

for sleepy driving. Moreover, in a study that examined younger drivers’ strategies to

deal with driver sleepiness, the third most popular strategy was increasing the effort

and attention directed at driving safely (Lucidi et al., 2006). Considered together,

drivers’ motivations to continue driving when sleepy appear to be a highly relevant

factor that can influence sleepy driving behaviours, particularly for younger drivers.

While several survey studies have quantified self-reported aspects of this

behaviour, there appears to be a lack of studies that examine the actual

psychophysiological and performance outcomes of motivating oneself to apply extra

effort to the task of driving when sleepy. Regarding the effects of motivation on

physiological indices of sleepiness, the literature review in Section 2.4.1.5.1

describes that motivation in the form of instructions to stay awake, results in

increases in sleep onset latency during Multiple Sleep Latency tests (Bonnet &

Arand, 2005a; Hartse et al., 1982; Shreter et al., 2006). Additionally, motivation in

the form of incentives for correct responses results in electrophysiology changes of

the attentional component of event related potentials (Boksem et al., 2006; Hsieh et

al., 2010).

The studies cited in Section 2.4.1.5.1 suggest an effect of motivation on some

physiological indices. However, it is not known how motivation affects more general

measures of cortical arousal, such as measures of absolute power of the theta and


alpha bands derived from spectral analyses. Additionally, the effect of motivation on

subjective sleepiness measures is also somewhat unknown. It was found in Study

One that motivation to improve performance on the PVT resulted in maintaining

levels of subjective sleepiness. This finding is in contrast to previous research that

typically demonstrates that subjective sleepiness increases while performing the PVT

(Dinges et al., 1997; Horne & Burley, 2010; Rupp et al., 2012).

The effect of motivation on performance indices appears to have certain

trends regarding low- and high-order cognitive tasks. For instance, Horne and Pettitt

(1985) demonstrated that motivation facilitated a complete restoration of

performance during a 30-min low-order task when participants were sleep deprived.

Similar to the results from Horne and Pettitt, clear effects of motivation were

observed in Study One, whereby motivation resulted in a complete restoration of

performance on PVT reaction-time latencies. However, as shown in Hsieh et al.’s

(2010) study, the effects of motivation on sleep-deprived individuals were not

enough to attenuate the performance deficits on a high-order task.

Examining the effect of motivation on sleepiness and performance is an

important road safety undertaking, especially when considering the results of survey

data cited earlier. In particular, the magnitude and the duration of motivational

effects are largely unknown. Additionally, the effect of motivation on low- versus

high-order cognitive tasks also appears to be unresolved. In order to examine these

unresolved issues, five hypotheses were proposed. The magnitude of the effect from

motivation was tested with two hypotheses: (H1) higher levels of motivation will

lead to an overall decrease in physiological, subjective, and behavioural indices of

sleepiness; and (H2) motivation will be greater for the non-sleep deprived condition

than the sleep-deprived condition. The duration of the motivational effect was tested


against two hypotheses: (H3) the effect from motivation will be transient in duration

with physiological, subjective, and performance indices of sleepiness eventually

increasing after the motivation effect has stopped; and (H4) the transient effects of

motivation will be less for the sleepy conditions than the alert conditions. Finally, the

potential for differential effects of motivation on low- and high-order tasks was

examined with a single hypothesis: (H5) the effects of motivation will be greater in

magnitude for low-order than high-order behavioural tasks. Secondary analyses that

focused on the signs of sleepiness and intrinsic experiences were performed to assess

the perceptions the participants experienced during the task. Understanding drivers’

self-perception of the signs of sleepiness and intrinsic experiences during driving

tasks may be a critical consideration in attempts to change driver behaviour.

6.3 Method

6.3.1 Design

This study used an experimental repeated measures design with three factors

(independent variables). The first factor, Sleepiness Level (alert or sleepy) was the

manipulation of level of sleepiness. The second factor, Motivation Level (motivated

or non-motivated) was the direct manipulation of motivational levels. The third

factor, Time Period is the segment of time during the study. The data obtained during

the PVT testing session was segmented into two time periods (time periods one and

two), whereas, the data obtained during the HPT testing session was segmented into

three time periods (time periods one, two, and three). The duration of the PVT was

10 min, whereas the duration of the HPT was 60 min and thus only two and three

time periods were used for the PVT and HPT respectively. The dependent variables

were physiological (i.e., EEG: theta and alpha absolute power for the electrode sites

of F5, C3, and O1), subjective (i.e., KSS, MES scores), and behavioural indices (i.e.,


HPT and PVT reaction-time latencies) of sleepiness. Table 6.1 displays the levels of

the factors and the outcome variables.

As a repeated measures design was used, practice and order effects had the

potential to influence the results. As such, the undertaking of the motivation and

sleep factors were counterbalanced across participants. The order of undertaking the

four different versions of the HPT was conducted according to a balanced Latin

square design. The counterbalancing and balanced Latin square design should

mitigate any order or practice effects; nonetheless, potential order effects were

investigated prior to running the main analyses. If the Sphericity assumption was

breached the Greenhouse-Geisser correction was reported and given the exploratory

nature of Study Three, significant ANOVA main effects and interactions were

further examined with post-hoc, Bonferroni corrected comparisons.

Table 6.1. Study Three Independent and Dependent Variables.

Independent variables for the PVT testing sessions

Factor one: Sleepiness Level

Factor two: Motivation Level

Factor three: Time Period

Dependent variables

1) Alert 2) Sleepy

1) Motivated 2) Non-motivated

1) 0-5 min 2) 5-10 min

Physiological: • EEG (theta and alpha power) Subjective: • KSS • MES Performance: • PVT (reaction-time latency)

Independent variables for the HPT testing sessions

Factor one: Sleepiness Level

Factor two: Motivation Level

Factor three: Time Period

Dependent variables

1) Alert 2) Sleepy

1) Motivated 2) Non-motivated

1) 0-20 min 2) 20-40 min 3) 40-60 min

Physiological: • EEG (theta and alpha power) Subjective: • KSS • MES Performance: • HPT (reaction-time latency)


6.3.1.1 Power analysis and sample size

A power analysis was performed to estimate the minimum sample size

needed. Table 2.2 displays the effect sizes of relevant previous studies and it can be

seen that mainly medium with some medium-larger effect sizes have been found

with motivation during task performance (e.g., Boksem et al., 2006; Hsieh et al.,

2010). Taking a conservative approach, the required minimum sample size was

calculated with the parameters of a medium effect size of f = 0.25, an alpha level of

.05, and a power level of .95 for a within-subjects design. It was estimated from these

parameters that a minimum sample size of 23 participants was required.

6.3.2 Participants

Participants were recruited into the study via the university’s email

distribution lists. The inclusion criterion was that participants had at least two years

on-road driving experience. The hazard perception literature consistently reveals an

association between faster hazard perception identification and more on-road

experience (see Horswill & McKenna, 2004). Therefore, this inclusion criterion was

required to eliminate individuals that may have slower hazard perception skill due to

a lack of on-road experience. Participants were excluded from participating if they

had a habitual bedtime later than 12 midnight, any significant health problems or a

sleep disorder, had excessive daytime sleepiness (assessed as a Epworth Sleepiness

Scale of > 10: Johns, 1991) or had sleeping difficulties (assessed as a Pittsburgh

Sleep Quality Index score of > 5: Buysse et al., 1989), were a shift worker, had

travelled overseas in the past month, took prescription medications that altered

arousal levels or illicit drugs, or drank more than three cups of coffee per day and/or

more than two standard drinks of alcohol per day.


Overall, 18 participants (9 females and 9 males) took part in the study. The

mean age of participants was M = 22.29 years (SD = 1.76; range = 20-25). On

average the participants usual weekday bedtime was at 22:52 (SD = 52 min) and

usual weekday wake-time was at 07:05 (SD = 53 min), and thus the average normal

weekday sleep duration for the participants was M = 492.35 min (SD = 48.39; range

= 420-540) as assessed by the Sleep Timing Questionnaire (Monk et al., 2003).

All participants had a valid driver licence with a mean duration of licensure

being M = 4.73 years (SD = 0.94; range = 3-6) and all participants were current

drivers. Additionally, the sample reported having driven on average 7.06 hr per week

(SD = 4.41; range = 2-21). Altogether, six participants reported having a sleep-

related close call (i.e., a near-crash while driving or driving outside of the designated

lane); however, no participant reported ever having had a sleep-related crash (i.e.,

where they were the driver and there was damage to property or persons). All

participants were paid 100 AUD for partaking in the study.

6.3.3 Measures



sex, duration of having a licence (including the duration of their learner licence), and

number of hours driven per week.


The physiological measure used in the current study were EEG theta and

alpha absolute power levels; spectral analyses were performed on the F5-A2, C3-A2,

and O1-A2 derivations. Please see sections 3.1 for specific details as well as issues

regarding reliability and validity of these measures and section 3.1.4 for the details of


data acquisition and processing. Figure 6.1 shows the electrode sites recorded in

Study Three.

Figure 6.1. EEG electrode sites utilised in the motivation to continue driving while sleepy study.


The subjective measures used in the current study were the KSS, MES, IMI,

ESS, PSQI, STQ, SoSQ. The KSS was used to quantify the participants’ levels of

subjective sleepiness, while the MES and IMI quantified the amount of effort the

participants were using to complete the task. The ESS and PSQI were used as

screening measures for excessive daytime sleepiness, while the STQ was used to

quantify habitual sleep behaviours. The SoSQ quantified the importance that

participants believed certain signs of sleepiness were as indicators of their level of

sleepiness. The SoSQ was slightly modified for use in Study Three, whereby the sign

of sleepiness “being easily distracted” was removed and replaced with four different

signs of sleepiness including, feeling jittery/jumpy, experiencing a dreamlike state,


becoming irritable, and feeling cold. These new inclusions were so a wider range of

physical and psychological signs of sleepiness was assessed by the SoSQ. Please

refer to section 3.2 for specific details of these measures as well as issues of

reliability and validity of these measures.

6.3.3.4 Performance and behavioural measures

The performance and behavioural measures used in the current study were the

PVT, HPT, and actigraphy. The PVT and HPT were used to quantify the

participants’ performance levels with the manipulations of Sleepiness Level and

Motivation Level factors. The actigraphs were used to ensure the participants

compliance with the study’s sleep duration protocols. Please refer to section 3.3 for

specific details of these measures including the reliability and validity of these

measures.

As a repeated measures design was used, four equivalent versions of the HPT

were developed to mitigate practice effects from viewing the same footage on several

occasions using video clips validated in previous research (i.e., Horswill et al., 2008;

S. S. Smith et al., 2009; Wetton et al., 2010). The four versions of the HPT were

designed to be as equivalent as possible, such that each version contained 31 hazards

(or traffic conflicts) with 10 hazards appearing in the first 20 mins, 11 hazards

appearing in the second 20 mins, and 10 hazards appearing in the last 20 mins of the

hour long tests. Due to the extended length of time required for this study (four hours

in total), several video clips did not contain any hazards. Clips that did contain

hazards could contain one or several hazards. The distribution of video clips that

contained hazards and video clips that did not contain any hazards was

approximately the same across the duration of the four different versions of the HPT.

The PVT has virtually no learning curve before proficiency is reached (Graw et al.,


2004) and similarly, practice effects are not evident with improvement of

performance levels of repeated attempts on the PVT (e.g., Dinges et al., 1997; Loh et

al., 2004). Thus, repeated performances of the PVT are not problematic.

6.3.4 Procedure



1400000437) as well as Health and Safety approval prior to beginning the study.

The current study took place over three distinct sessions. The first session

was a baseline session and the subsequent two sessions were testing sessions. The

baseline session required the participant to arrive at the laboratory between 10:00 and

13:00. This baseline session entailed the participant completing baseline

questionnaires (i.e., PSQI and ESS) to determine their eligibility to partake in the

current study. Once the PSQI and ESS scores confirmed their eligibility to partake in

the study, the study procedure was explained to the participant in greater detail. If the

participant wanted to continue with the study, they then signed a consent form, and

they were given an actigraph to wear and a simple sleep diary to complete for a

period of two weeks. The participant then completed a 10-min practice HPT session

and a 5-min PVT that included practice with using the KSS and MES to ensure they

were familiar with the rating scales. Lastly, participants also completed the baseline

assessment of the SoSQ.

Exactly one week later, each participant returned to complete the first of two

4-hr testing sessions. One of these two testing sessions required the participant to be

partially sleep deprived. The participant was instructed to wake up 2 hours before

their normal waking time, which was confirmed by the actigraph and sleep diary

data. For the alert condition, the participant maintained their habitual sleep-wake


times. On the day of testing, the participant was also instructed not to ingest any form

of caffeine or alcohol until after the testing session had been completed. The order of

undertaking the test under sleep-deprived or non-sleep-deprived conditions as well as

the motivated or non-motivated conditions was counterbalanced across all

participants. Each of the two laboratory testing sessions included completing both of

the motivation conditions (i.e., motivated and non-motivated), so that both

motivation conditions were undertaken when alert and when sleepy.

Upon arrival at the laboratory for the testing session, the participant had the

EEG, EOG, and ECG electrodes attached. Once all of the electrodes were correctly

recording the electrophysical data, the participant then completed a 10-min PVT,

before undertaking a 60-min HPT simulated drive. Before each HPT driving tasks,

all participants watched a 5-min instructional video showing how to complete the

HPT, which included two examples of “typical” hazards. This was intended to

reduce any ambiguities regarding what constituted a hazard that the participant was

required to identify. The standardised 5-min instructional video has been validated in

numerous research studies (e.g., Horswill et al., 2008; S. S. Smith et al., 2009;

Wetton et al., 2010). After approximately five minutes, the participant entered their

KSS as well as their MES score, a procedure that continued at 5-min intervals

throughout the HPT sessions. At the completion of the HPT, each participant then

had a 40-min respite in the laboratory while the experimenter present.

The activities of the participants during the respite were monitored and were,

within reason, regulated to ensure a degree of similarity between all participants. For

instance, all participants were offered some drinking water and were given the

opportunity to visit the bathroom. All participants were required to stand up for a

maximum of 10 min during the respite interval during which they were free to move


around the laboratory in a non-energetic manner. Previous research has shown that

the greater amount of energy that is expended during physical activities, the greater

the increases in alertness (e.g., Bonnet & Arand, 2005b; Horne & Foster, 1996;

Sallinen et al., 2008). Thus, the level of energetic activity was controlled. Aside from

these activities, participants were free to use the desktop computer (to check emails,

browse the internet) and light conversation was made between the experimenter and

all participants.

Following the 40-min respite the participant undertook another 10-min PVT

and another 60-min HPT drive under the remaining motivation condition (i.e.,

motivated or non-motivated). At the end of each PVT and HPT session, the

participant completed post-experimental questionnaires that included the Signs of

Sleepiness Questionnaire, as well as perceptions and subjective experiences during

the task. Figure 6.2 displays the order of completing the various tasks during one

testing session.

Figure 6.2. Overview of the timeline of Study Three. Not to scale.

Exactly one week later, the participant returned for their final testing session,

where they completed either an alert or a sleepy condition, depending on which

condition they undertook during the first testing session. Previous research suggests

that a week between laboratory testing sessions should be sufficient to ensure

cognitive recovery for those participants that undertook the sleepy (i.e., partial sleep

08:30 start time 40 min

3 hr 40 min

60 min 60 min

HPT Setup HPT PVT10 PVT10 Interval


deprivation) condition during the first testing session (e.g., Ikegami et al., 2009). For

all testing sessions, each participant was tested alone, while seated in a comfortable

high-backed chair in an environment with controlled sound, temperature, and light

(see section 3.5). All participants received standardised instructions relating to the

experimental protocol.

6.3.5 Motivations Experimental Conditions

The experimental condition includes a motivated and non-motivated

condition. It was found in the literature review that motivation has previously been

manipulated using monetary incentives and/or social comparisons (i.e., competition).

These two types of motivations have little contextual utility with non-professional

drivers. The context or reasons cited by drivers when they continue to drive despite

being aware of increasing sleepiness are related to being close to their destination

and/or time demands (Armstrong et al., 2010; Fletcher et al., 2005; Nordbakke &

Sagberg, 2007; Watling, Armstrong, et al., 2014). Importantly, it was also shown in

Study One that an instructional motivator had an effect on physiological, subjective

and performance levels during the PVT5. Therefore, monetary incentives and/or

social comparisons were not used in the current study.

The type of motivation chosen for the current study was an instructional

motivator. The instructional motivator was delivered to the participant prior to

undertaking the HPT or the PVT. The participant was told that “You are about to

undertake a simulated driving task called the Hazard Perception Test/reaction-time

task. During this task, it is very important2 that you do not fall asleep at any time

as this could affect the study results. It is also very important for the study that you

perform to the best of your ability on the hazard perception task/reaction-time task. 2 Bolded text was emphasised by a change in the intonation of the experimenter’s voice (the PhD candidate) and thus a high level of consistency with the motivational instructions and the change in intonation was achieved.


This means motivating yourself to apply as much extra effort as you possibly can

during the testing session. Do you understand these instructions or have any

questions?” In contrast, the non-motivated condition received the instruction “You

are about to undertake a simulated driving task called the Hazard Perception

Test/reaction-time task. Do you have any questions?”

6.4 Results

The following section describes the results of the tests of hypotheses 1 to 5.

The first section examines the amount of sleep participants obtained prior to the two

testing days. Then manipulation checks regarding participants’ subjective sleepiness

from the manipulation of Sleepiness Level factor will be examined, followed by

examination of the reliability statistics of the four Hazard Perceptions Tests, and

testing for the potential of order effects with the study data. Last, the study’s

hypotheses was first be performed on the PVT data and then on the HPT data.


All data were examined for their distribution properties, with some minor

departures from normality (positive skewing) occurring with the EEG and KSS data.

When transformations were applied to the data, the skew was removed, but no

practical change to the statistical results occurred when they were compared to the

statistical results using the untransformed data. Moreover, as ANOVA is relatively

robust to moderate breaches of normality (Field, 2009; Tabachnick & Fidell, 2007),

and physiological data does not always conform to normal distributions, the

untransformed data was consequently used for all the analyses.


6.4.2 Manipulation Check and Order Effects

6.4.2.1 Sleep duration prior to testing

The participants’ duration of sleep the night prior to testing was examined to

ensure their compliance with the study protocol. Table 6.2 displays the participants’

mean amounts of sleep obtained the night prior to testing (presented in minutes) via

self-report from the sleep diaries and objective data from the actigraphs.

Table 6.2. Mean Amount of Sleep obtained prior to the Alert and Sleepy Testing Sessions for the Sleep Diary and Actigraphic Data.

Sleep diary data Actigraph data

Testing session M SD M SD

Alert testing session 480.05 38.47 473.80 35.89

Sleepy testing session 370.23 35.63 363.38 41.16

6.4.2.2 Subjective sleepiness

To assess whether the reduced sleep time on the sleep-deprived testing day

induced any subjective daytime sleepiness in the participants, the mean values of the

KSS prior to testing were inspected. The participants’ level of subjective sleepiness

were examined with a 2 x 2 repeated measures ANOVA, the two factors being

Sleepiness Level (alert or sleepy) and Motivation Level (motivated or non-

motivated). A significant difference was found for the Sleep Duration factor F(1, 17)

= 12.24, p = .003, with mean KSS levels being higher when sleep deprived (M =

6.00, SE = 0.40) than after a normal duration of sleep (M = 4.88, SE = 0.28). No

significant difference was found for the Motivation Level factor, F(1, 17) = 0.04, p =

.85, nor for the interaction term, F(1, 17) = 0.11, p = .75.

6.4.2.3 Hazard perception reliability

The internal consistency of the four versions of the Hazard Perception Test

was evaluated with the Cronbach’s alpha statistic. The values obtained for tests 1-4


were .86, .88, .87, and .90 respectively. These values suggest that the reliability of

the four versions of the Hazard Perception Test was adequate.

6.4.2.4 Order Effects

The physiological, subjective, and behavioural data were examined for

potential order effects. As participants completed two testing sessions on the same

day with a 40-min interval between the two testing sessions (see Figure 6.2), there is

a potential for carry-over effects, even with the counterbalancing. Potential carry-

over effects were examined by adding two between-subjects factors into the repeated

measures ANOVAs. These two between-subjects factors represented the order the

participant undertook the motivated or non-motivated conditions for the first testing

session (i.e., week one: alert or sleepy) and the second testing session (i.e., week two:

alert or sleepy). The results of the ANOVAs examining the potential order effects are

tabulated in Appendix B, Table B.1 and no significant main effects or interactions

were observed.

6.4.3 Sleepiness, Motivation, and Performance

6.4.3.1 Visual examination of the EEG and EOG data

Visual inspection of the EEG and EOG data revealed that none of the

participants could be judged as having fallen asleep by current sleep staging criteria

(i.e., American Academy of Sleep Medicine & Iber, 2007) or as having experienced

any microsleeps (i.e., 3-15 sec duration of theta activity) during the testing sessions.

6.4.3.2 Descriptive data

Figures 6.3 and 6.4 (over page) display the means and standard error of the

mean of the physiological, subjective, and behavioural data for the PVT testing

sessions. It is clear from the data displayed in Figures 6.3 and 6.4 that levels of

physiological and subjective sleepiness were greater for the two sleepy conditions


than the two alert conditions and that levels of sleepiness increased over the duration

of the PVT task. Similarly, on average the PVT data appears to have increased over

the duration of the task. However, the alert-motivated condition appears to have

yielded shorter reaction-time latencies than the other three conditions during the PVT

testing.

Figures 6.5 and 6.6 (over page) display the means and standard error of the

mean of the physiological, subjective, and behavioural data for the HPT testing

sessions. It is clear from the data displayed in Figures 6.5 and 6.6 that, similar to the

PVT data trends, levels of physiological and subjective sleepiness were greater for

the two sleepy conditions than the two alert conditions and that the level of

sleepiness generally increased over the duration of the task. Similarly, the HPT

reaction-time latency data appears to have increased over the duration of the task and

there were no clear trends for any of the Sleepiness Level or Motivation Level

factors.

6.4.3.3 Tests of hypotheses

The tests of the study hypotheses are presented in two sections. The first

section examines the study hypotheses on the PVT data. A series of 2 x 2 x 2

repeated measures ANOVAs were performed on the physiological, subjective, and

performance data obtained during the PVT session, except for the MES data, as this

was only collected once, at the end of the testing session (see Table 6.3, over page).

The MES data was examined with a 2 x 2 repeated measures ANOVA. The second

section examines the testing of the study hypotheses on the HPT data. A series of 2 x

2 x 3 repeated measures ANOVAs was performed on the physiological, subjective,

and performance data obtained during the HPT sessions (see Table 6.4, over page).


Figure 6.3. Mean level of EEG sleepiness recorded at the F5, C3, and O1 electrode sites during the alert-motivated condition (white bars), the alert-non-motivated condition (black bars), the sleepy-motivated condition (light grey bars), and the sleepy-non-motivated condition (dark grey bars) during the PVT testing sessions. Error bars represent standard error of the mean.

0.00

10.00

20.00

30.00

40.00

50.00

60.00

0-5 min 5-10 min

EEG

abs

olut

e po

wer

, µV2

Time Period

F5 Mean Theta Power

0.005.00

10.0015.0020.0025.0030.0035.0040.0045.0050.00

0-5 min 5-10 min

EEG

abs

olut

e po

wer

, µV2

Time Period

F5 Mean Alpha Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

0-5 min 5-10 min

EEG

abs

olut

e po

wer

, µV2

Time Period

C3 Mean Theta Power

0.00

10.00

20.00

30.00

40.00

50.00

60.00

0-5 min 5-10 min

EEG

abs

olut

e po

wer

, µV2

Time Period

C3 Mean Alpha Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

0-5 min 5-10 min

EEG

abs

olut

e po

wer

, µV2

Time Period

O1 Mean Theta Power

0.0010.0020.0030.0040.0050.0060.0070.0080.0090.00

100.00

0-5 min 5-10 min

EEG

abs

olut

e po

wer

, µV2

Time Period

O1 Mean Alpha Power


Figure 6.4. Mean levels of subjective sleepiness (KSS), motivational effort (MES), and reaction-time latency (PVT) during the alert-motivated condition (white bars), the alert-non-motivated condition (black bars), the sleepy-motivated condition (light grey bars), and the sleepy-non-motivated condition (dark grey bars) during the PVT testing sessions. Error bars represent standard error of the mean.

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

Prior to PVT End of PVT

KSS

val

ue

Time Period

Mean KSS

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

End of PVT

MES

val

ue

Time Period

Mean MES

300.00320.00340.00360.00380.00400.00420.00440.00460.00480.00500.00

0-5 min 5-10 min

Rea

ctio

n-tim

e la

tenc

y, m

s

Time Period

Mean PVT


Figure 6.5. Mean level of EEG sleepiness recorded at the F5, C3, and O1 electrode sites during the alert-motivated condition (white bars), the alert-non-motivated condition (black bars), the sleepy-motivated condition (light grey bars), and the sleepy-non-motivated condition (dark grey bars) during the HPT testing sessions. Error bars represent standard error of the mean.

0.00

10.00

20.00

30.00

40.00

50.00

60.00

0-20 min 20-40 min 40-60 min

EEG

abs

olut

e po

wer

, µV2

Time Period

F5 Mean Theta Power

0.00

10.00

20.00

30.00

40.00

50.00

60.00

0-20 min 20-40 min 40-60 min

EEG

abs

olut

e po

wer

, µV2

Time Period

F5 Mean Alpha Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

0-20 min 20-40 min 40-60 min

EEG

abs

olut

e po

wer

, µV2

Time Period

C3 Mean Theta Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

0-20 min 20-40 min 40-60 min

EEG

abs

olut

e po

wer

, µV2

Time Period

C3 Mean Alpha Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

0-20 min 20-40 min 40-60 min

EEG

abs

olut

e po

wer

, µV2

Time Period

O1 Mean Theta Power

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

0-20 min 20-40 min 40-60 min

EEG

abs

olut

e po

wer

, µV2

Time Period

O1 Mean Alpha Power


Figure 6.6. Mean levels of subjective sleepiness (KSS), motivational effort (MES), and reaction-time latency (HPT) during the alert-motivated condition (white bars), the alert-non-motivated condition (black bars), the sleepy-motivated condition (light grey bars), and the sleepy-non-motivated condition (dark grey bars) during the HPT testing sessions. Error bars represent standard error of the mean.

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

0-20 min 20-40 min 40-60 min

KSS

val

ue

Time Period

Mean KSS

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

0-20 min 20-40 min 40-60 min

MES

val

ue

Time Period

Mean MES

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

0-20 min 20-40 min 40-60 min

Rea

ctio

n-tim

e la

tenc

y, s

ec

Time Period

Mean HPT


Table 6.3. ANOVA Statistics (2 x 2x 2 Repeated Measures) of the Physiological, Subjective, and Performance Data During the PVT Sessions.

SLP MOT TIME SLP*MOT SLP*TIME MOT*TIME SLP*MOT*TIME

Data

source

F df ηp2 F df ηp



2 F df ηp2

EEG

F5 theta 5.44* 1,16 .25 1.24 1,16 .07 19.14** 1,16 .55 0.33 1,16 .02 2.24 1,16 .12 0.01 1,16 .01 1.95 1,16 .11

F5 alpha 7.42* 1,16 .32 0.59 1,16 .04 9.30** 1,16 .37 0.12 1,16 .01 0.52 1,16 .03 2.63 1,16 .14 0.04 1,16 .01

C3 theta 15.57** 1,16 .49 0.19 1,16 .01 7.94* 1,16 .33 0.02 1,16 .01 0.09 1,16 .01 1.77 1,16 .10 0.52 1,16 .03

C3 alpha 5.08* 1,16 .24 0.02 1,16 .01 8.75** 1,16 .35 1.79 1,16 .10 0.34 1,16 .02 0.03 1,16 .01 0.89 1,16 .05

O1 theta 13.39** 1,16 .46 1.27 1,16 .07 12.13** 1,16 .43 0.08 1,16 .01 0.50 1,16 .03 0.22 1,16 .01 1.04 1,16 .06

O1 alpha 5.60* 1,16 .26 0.17 1,16 .01 5.51* 1,16 .26 0.08 1,16 .01 0.20 1,16 .01 0.19 1,16 .01 0.01 1,16 .01

KSS 16.98** 1,16 .52 0.01 1,16 .01 22.02** 1,16 .58 0.03 1,16 .01 2.49 1,16 .13 1.58 1,16 .09 0.81 1,16 .05

MESa 8.59** 1,16 .35 7.14* 1,16 .31 - - - 0.37 1,16 .02 - - - - - - - - -

PVT 12.50** 1,16 .44 9.90** 1,16 .38 81.87** 1,16 .84 16.76** 1,16 .51 7.08* 1,16 .31 5.30* 1,16 .25 4.72* 1,16 .23

Note. SLP = Sleepiness Level factor; MOT = Motivational Level factor; TIME = Time Period factor. a Note that the MES data was only collected at the end of the PVT testing, thus a 2 x 2 ANOVA was calculated. * < .05, ** < .01.



Table 6.4. ANOVA Statistics (2 x 2 x 3 Repeated Measures) of the Physiological, Subjective, and Performance Data During the HPT Sessions.

SLP MOT TIME SLP*MOT SLP*TIME MOT*TIME SLP*MOT*TIME




2 F df ηp2

EEG

F5 theta 7.12* 1,16 .31 0.03 1,17 .01 23.90** 2,32 .60 0.06 1,17 .01 4.76** 2,32 .23 0.49 1.50,23.96 .03 2.24 1.27,20.29 .12

F5 alpha 5.22* 1,16 .25 0.01 1,17 .01 15.49** 1.48,23.64 .49 0.56 1,17 .03 2.20 2.32 .12 1.01 2,32 .06 1.01 2,32 .06

C3 theta 30.12** 1,16 .65 0.03 1,17 .01 40.54** 2.32 .72 0.30 1,17 .02 20.80** 1.36,21.77 .56 0.96 2,32 .06 2.36 2,32 .13

C3 alpha 8.52* 1,16 .35 0.02 1,17 .01 19.33** 1.37,21.85 .55 0.01 1,17 .01 0.54 1.45,23.16 .03 0.96 2,32 .06 2.97 1.50,23.95 .15

O1 theta 12.68** 1,16 .44 1.79 1,17 .10 17.06** 1.43,22.84 .52 0.01 1,17 .01 1.86 1.47,23.52 .10 0.75 2,32 .04 0.32 1.25,20.06 .02

O1 alpha 7.17* 1,16 .31 1.73 1,17 .10 29.28** 1.49,23.88 .65 0.15 1,17 .01 4.43* 1.31,21.01 .15 4.73* 2,32 .23 1.28 2,32 .07

KSS 63.66** 1,16 .80 1.89 1,17 .11 65.70** 1.35,21.65 .80 1.19 1,17 .07 0.26 1.31,20.89 .02 0.27 1.30,20.79 .02 0.94 2,32 .06

MES 5.64* 1,16 .26 9.69** 1,17 .38 9.87** 2.32 .38 1.01 1,17 .06 1.79 2,32 .10 0.27 1.18,18.80 .02 0.09 1.14,18.25 .01

HPT 4.76* 1,16 .23 3.48 1,17 .18 149.11** 1.45,23.20 .90 0.27 1,17 .02 0.26 1.47,23.53 .01 1.23 2,32 .06 1.61 2,32 .09

Note. SLP = Sleepiness Level factor; MOT = Motivational Level factor; TIME = Time Period factor. * < .05, ** < .01.



6.4.3.3.1 PVT testing session data

Table 6.3 displays the ANOVAs of the physiological, subjective, and

performance data obtained during the PVT testing sessions. The first hypothesis (H1)

is concerned with the magnitude of the effect from motivation, such that motivation

would lead to an overall decrease in physiological, subjective, and performance

indices of sleepiness. This hypothesis was partly supported by the PVT session data.

No significant effects were found with Motivation Level factor for any of the

physiological or subjective sleepiness data. However, the main effect of the

Motivation Level factor evident in the MES data was that the level of motivational

effort participants were applying was greater in the motivational conditions than

during the non-motivational conditions, as can be seen in Figure 6.4. The Motivation

Level factor’s main effect evident in the PVT reaction-time latency data is that the

reaction-time latencies were shorter for the motivated conditions than the non-

motivated conditions.

The results of the ANOVAs for the PVT reaction-time latency data revealed a

number of main effects and interactions that partly support H1. The Motivation Level

factor main effect shows that motivated conditions had shorter reaction-time

latencies than the non-motivated conditions. A number of significant interactions

were found with the PVT reaction-time latency data; however, rather than exploring

all of the simple main effects with all of these interactions, only the three-way

interaction was explored. The three-way interaction can be seen in Figure 6.7 (over

page) with the post-hoc Bonferroni corrected comparisons listed in Table 6.5. When

examining the change in reaction-time latency from the first 5 min to the second 5

min time period for the four conditions, it was found that the reaction-time latencies

for all but the alert-motivated condition significantly increased over the duration of

the PVT testing. Additionally, the alert-motivated PVT reaction-time latencies were

shorter at the 0-5 min and 5-10 min time periods.


Figure 6.7. Three-way interaction results between Sleepiness level, Motivation level, and Time Period factors. Error bars represent standard error of the mean.

Table 6.5. Post-hoc Comparisons of the Different Levels of the Three-Way Interaction of the PVT Performance Data.

Level of comparison Comparison at 0-5 min Comparison at 5-10 min

Condition 0-5 to 5-10

with AM

with AN

with SM

with SN

with AM

with AN

with SM

with SN

AM -3.37 - - AN -4.93** -3.78** -4.47** - SM -6.46** -3.90** 0.33 - -9.21** -0.12 - SN -4.10** -4.14** 0.35 -0.20 - -8.78** 0.15 0.40 -

Note. AM = alert-motivated; AN = alert-non-motivated; SM = sleepy-motivated; SN = sleepy-non-motivated. * < .05, ** < .01.

The second hypothesis (H2), that the effect of increased motivation would be

greater for the non-sleep-deprived condition than the sleep-deprived condition was

only partly supported. Specifically, no significant Sleepiness Level factor by

Motivation Level factor interaction was found for the physiological or subjective

sleepiness data. A significant Sleepiness Level by Motivation Level interaction was

found for the PVT reaction-time latency data. However, as a three-way interaction

was observed in the PVT data interpreting the Sleepiness level by Motivation level

interaction might be erroneous without consideration of the Time period factor. As

such, examining the post-hoc comparisons in Table 6.5 is appropriate. Table 6.5

300.00320.00340.00360.00380.00400.00420.00440.00460.00480.00500.00

0-5 min 5-10 min

Rea

ctio

n-tim

e la

tenc

y

Time period

Alert-motivated

Alert-non-motivated

Sleepy-motivated

Sleepy-non-motivated


shows that only the alert-motivated condition reaction time latencies were

maintained during the PVT, with reaction-time latencies increasing for the remaining

conditions. Thus, only partial support was provided for the second hypothesis.

The third hypothesis (H3), concerning the duration of the effect from

motivation, whereby the effects from motivation will be transient in duration with

physiological, subjective, and performance indices of sleepiness eventually

increasing shortly after the motivational instruction, was supported in the current

data. Significant main effects for the Time Period factor were found for all of the

data sources and as can be seen in Figures 6.3 and 6.4 all data sources (baring the

MES data) increased.

The fourth hypothesis (H4), that the transient effect of motivation on

physiological, subjective, and performance indices of sleepiness will reduce more

quickly for the sleepy conditions than the alert conditions, was only partly supported

from the obtained data. None of the physiological or subjective sleepiness data

supported this hypothesis; however, a significant three-way interaction was observed

for the PVT reaction-time latency data. As previously shown (see Figure 6.7 and

Table 6.5), the alert-motivated condition resulted in shorter reaction-time latencies

than any of the other conditions and this trend occurred during the first 5 min of the

PVT session and continued during the second 5 min of the PVT session.

6.4.3.3.2 HPT testing session data

The outcomes of the repeated measures ANOVAs on the physiological,

subjective, and performance data obtained during the HPT sessions can be seen in

Table 6.4. The first hypothesis (H1), which was concerned with the magnitude of the

effect from motivation, whereby motivation would lead to an overall decrease in

physiological, subjective, and performance indices of sleepiness, was not supported.


No significant effects were found with the Motivation Level factor for any of the

physiological, subjective sleepiness, or performance data. However, there were small

effects present in the KSS and HPT data that failed to reach significance levels. Yet

again, the Motivation Level factor main effect for the MES data reveals that the

motivational effort participants were applying was greater in the motivated

conditions than the non-motivated conditions, as can be seen in Figure 6.6.

Regarding the second hypothesis (H2), that the effect of increased motivation

would be greater for the non-sleep-deprived condition than the sleep-deprived

condition, was not supported by the data obtained during the HPT testing sessions.

No significant interactions were found for the physiological or subjective sleepiness

as well as the performance data from the HPT.

The third hypothesis (H3), concerning the duration of the effect from

motivation, and suggesting that the effects from motivation will be transient in

duration with increases of physiological, subjective, and performance indices of

sleepiness eventually occurring was not supported from the current data. No

significant interactions between the Motivation Level factor and the Time Period

factor were observed from the obtained data. Significant main effects for the Time

Period factor were found for all of the data sources as shown in Table 6.4.

Specifically, all of the physiological, subjective, and performance data increased for

all conditions over the duration of testing.

The fourth hypothesis (H4) that the transient effect of motivation on

physiological, subjective, and performance indices will be lower for the sleepy

conditions than the alert conditions was not supported from the data obtained during

the HPT sessions. None of the physiological or subjective sleepiness data or the

performance data supported this hypothesis.


6.4.3.4 Signs of sleepiness during the HPT sessions

At the completion of each HPT session, the participants rated the importance

of signs of sleepiness as indicators of their level of sleepiness during the four

different conditions. These post-HPT signs of sleepiness scores were compared to the

baseline scores of the importance of the signs of sleepiness that were collected at the

baseline session (see section 6.3.4). Figure 6.8 (over page) displays mean scores for

the importance of the signs of sleepiness for the baseline session and the four

experimental conditions.


Figure 6.8. Importance of the signs of sleepiness as indicators of the participants level of sleepiness during the baseline session (bars with the diagonal line fill), alert-motivated condition (white bars), the alert-non-motivated condition (black bars), the sleepy-motivated condition (light grey bars), and the sleepy-non-motivated condition (dark grey bars). Error bars represent standard error of the mean.

1.00

2.00

3.00

4.00

5.00

6.00

7.00

Impo

rtan

ce

Sign of Sleepiness



A series of four, 2 x 12 repeated measures ANOVAs were performed (see

Table 6.6), with the two factors being Baseline-Post Session factor (baseline scores

or post-HPT scores) and Signs of Sleepiness factor (the 12 signs of sleepiness, see

Figure 6.8). The Greenhouse-Geisser correction was reported if the Sphericity

assumption was breached and post-hoc tests used the Bonferroni adjustment for

multiple comparisons. Significant main effects were found for all of the Baseline-

Post Session factors; indicating that the level of importance of the signs of sleepiness

increased from levels recorded at the baselines session to levels recorded at the

completion of the HPT sessions. Similarly, significant main effects were found for

the Signs of Sleepiness factors and suggest that the individual signs of sleepiness

were rated differently for their level of importance. The main effect for the Signs of

Sleepiness factors were not further examined, as the importance of each of the

individual signs of sleepiness were not a focus of the analyses, only the change in the

ratings of importance.

Significant interaction terms were found for both of the comparisons of the

baseline importance scores with the alert-non-motivated condition, as well as the

comparison of the baseline importance scores to the sleepy-non-motivated condition.

Table 6.6. ANOVA Statistics (2 x 12 Repeated Measures) for the Comparisons Between Baseline Signs of Sleepiness Scores and post-HPT Sessions Signs of Sleepiness Scores.

Baseline-Post Session Signs of Sleepiness Interaction

Session comparison F df ηp2 F df ηp

2 F df ηp2

Baseline with AM 7.37* 1,17 .32 8.00** 6.35,86.66 .33 1.51 11,177 .09

Baseline with AN 5.16* 1,17 .24 7.59** 11,177 .32 2.59** 1,17 .14

Baseline with SM 4.52* 1,17 .22 7.38** 11,177 .32 1.23 1,177 .07

Baseline with SN 15.22** 1,17 .49 7.34** 11,177 .32 1.87* 11,17 .10

Note. AM = alert-motivated; AN = alert-non-motivated; SM = sleepy-motivated; SN = sleepy-non-motivated. * < .05, ** < .01.


Bonferroni adjusted post-hoc comparisons were performed on the baseline and post-

HPT importance values of each individual signs of sleepiness. After controlling for

multiple comparisons, it was found that only the slower reactions sign of sleepiness

was significantly rated more important after completing the alert-non-motivated

condition (t(17) = -3.734, p < .05) and the sleepy-non-motivated condition slower

reactions (t(17) = -3.344, p < .05) when compared to the baseline ratings of

importance.

It is worth noting that the trend of the interaction effect sizes. That is, bigger

effect sizes were found with the two non-motivated conditions when compared to

their same Sleepiness Level factor motivated conditions. This potential motivation

trend was further explored with a 2 x 2 x 12 repeated measures ANOVA with the

three factors being Sleepiness Level (alert or sleepy), Motivation Level (motivated or

non-motivated), and Signs of Sleepiness factor (the 12 signs of sleepiness). Only the

post-HPT importance scores were used in the 2 x 2 x 12 repeated measures ANOVA.

The only significant finding was for the main effect of Sign of Sleepiness factor,

F(11,177) = 8.88, p = .071, ηp2 = .36. However, the two next greatest effect sizes

were found for the Sleepiness Level*Motivation Level interaction, F(1,17) = 6.89, p

= .071, ηp2 = .19 and for the main effect of the Motivation Level factor, F(11,17) =

2.85, p = .091, ηp2 = .15; which were both approaching significance levels.

6.4.3.5 Intrinsic subjective experiences

An examination of the participants’ intrinsic subjective experiences in

relation to HPT testing sessions was performed. These intrinsic subjective

experiences included interest/enjoyment, perceived competence, effort/importance,

and pressure/tension subscales from the Intrinsic Motivation Inventory. Figure 6.9

(over page) displays the mean and standard error of the mean values for the four


subjective experiences for each condition. The internal consistency of the four

subscales from the Intrinsic Motivation Inventory assessed during the four conditions

(i.e., alert-motivated, alert-non-motivated, sleepy-motivated condition, and the

sleepy-non-motivated conditions) were evaluated with the Cronbach’s alpha statistic

and all ranged between .72-.88.Thus, these values indicate that the reliability of the

Intrinsic Motivation Inventory subscales were adequate. Table 6.7 (over page)

displays the outcomes of the repeated measures ANOVAs for each intrinsic

subjective experience.

It can be seen in Table 6.7 that main effects were found for both the

Sleepiness Level factor and the Motivation Level factor for the perceived

competence and effort/importance subjective experiences. Regarding perceived

competence subjective experience, the alert conditions as well as the motivated

conditions had higher mean values of perceived competence. Similarly, for the

effort/importance subjective experience, the alert conditions as well as the motivated

conditions had higher mean values of effort/competence. Last, significant main

effects were found for the Sleepiness Level and the Motivation Level factors.

However, a significant interaction was observed with the pressure/tension subjective

experience. Post-hoc Bonferroni corrected comparisons revealed that the

pressure/tension subjective experience was significantly greater during the sleepy-

motivated condition than during the sleepy-non-motivated condition, t(16) = 3.24, p

< .05; during the alert-motivated condition, t(16) = 16.96, p < .01; and the alert-non-

motivated condition, t(16) = 13.46, p < .01.


Figure 6.9. Mean levels of participants’ intrinsic subjective experiences when performing the hazard perception task during the alert-motivated condition (white bars), the alert-non-motivated condition (black bars), the sleepy-motivated condition (light grey bars), and the sleepy-non-motivated condition (dark grey bars). Error bars represent standard error of the mean.

Table 6.7. ANOVA Statistics (2 x 2 Repeated Measures) for the Intrinsic Subjective Experiences During the HPT Sessions.

SLP MOT SLP*MOT

Data source F df ηp

2

F df ηp2 F df ηp

2

Interest/enjoyment 0.01 1,17 .01 1.38 1,17 .08 0.12 1,17 .01

Perceived competence 5.66* 1,17 .26 6.01* 1,17 .27 0.18 1,17 .01

Effort/importance 19.27** 1,17 .55 20.02** 1,17 .60 2.52 1,17 .14

Pressure/tension 43.12** 1,17 .61 14.11** 1,17 .47 5.33* 1,17 .25

Note. SLP = Sleepiness Level factor; MOT = Motivational Level factor. * < .05, ** < .01.

1.00

2.00

3.00

4.00

5.00

6.00

7.00

Inte

rest

/enj

oym

ent

Condition

1.00

2.00

3.00

4.00

5.00

6.00

7.00

Perc

eive

d co

mpe

tenc

e

Condition

1.00

2.00

3.00

4.00

5.00

6.00

7.00

Effo

rt/im

port

ance

Condition

1.00

2.00

3.00

4.00

5.00

6.00

7.00

Pres

sure

/tens

ion

Condition


6.5 Discussion

This final study of the research program extended upon the findings of the

previous two studies with a more thorough examination of the effects of motivation

on sleepiness and performance. The final study utilised an experimental design with

a Sleepiness Level factor (sleepy and alert) and a Motivation Level factor (motivated

and non-motivated) to examine, in isolation, the effects of the two factors on

sleepiness and performance. The study also used the 10-min PVT as well as the HPT

to examine potential differential effects of motivation on low- and high-order

cognitive tasks respectively. Rather than discuss each of the study hypotheses

individually, the following discussion will examine the results of the analyses in a

broader sense to understand the effects of motivation obtained in the current study

and how these findings relate to the extant literature.

6.5.1 The Effects of Motivation with the HPT

Differential effects for motivation on the low- and high-order tasks were

found in the Study Three. Beginning with the HPT results (i.e., the high-order

cognitive task), no effects were found for motivation on physiological, subjective, or

performance indices of sleepiness. This was contrary to the expected hypotheses.

While no effect on sleepiness was detected, differences were found for the MES,

which quantified the amount of effort participants were applying to maintain their

level of task performance. The MES results demonstrate that participants were on

average, applying more effort during the two motivation conditions when compared

to the non-motivation conditions. Further, no significant main effect of Motivation

level was observed with the HPT reaction-time latencies; however, a small effect was

found. This suggests that performance on the HPT could have been improved in the


motivational conditions; however, a lack of power likely restricted the statistical

analyses.

The trends of the MES data deserve some further discussion. The results

obtained demonstrate that the amount of effort participants exerted as assessed via

the MES was significantly different at different levels of sleepiness; with the level of

effort higher during the sleepy conditions. While this finding is consistent with

previous research (Odle-Dusseau, Bradley, & Pilcher, 2010; Pilcher & Walters,

1997), it is contrary to other findings (Sallinen et al., 2013). A point of difference

with the Sallinen et al. (2013) study could be one of different designs. The Sallinen et

al. (2013) study incorporated a partial and cumulative sleep restriction paradigm over

a period of five nights. It has been suggested that some unknown adaptation process

occurs when an individual is subjected to several nights of partial sleep restriction

(Drake et al., 2001; Horne, 2010). Indeed, some studies have even shown that

subjective sleepiness ratings and, in some instances, performance levels begin to

stabilise after two to three nights of repeated partial sleep restriction (Dinges et al.,

1997; Purnell, Feyer, & Herbison, 2002); thus, it is possible that this adaptation

process occurred in the Sallinen et al. (2013) study, potentially explains why

participants in that study did not perceive an increase in effort during the testing

sessions.

The results from Study Three seem to suggest that the effects of motivation

are non-existent with reducing physiological, subjective, and performance measures

used during the HPT sessions. However, some of the obtained data do support an

effect of motivation. Specifically, participants were able to maintain wakefulness

during the sleepy conditions. Visual inspection of the EEG and EOG data revealed

that no single participant could be judged to have fallen asleep by current sleep


staging criteria (i.e., Iber & American Academy of Sleep Medicine, 2007) or having

experienced any microsleeps (i.e., 3-15 sec duration of theta activity). Previous

research has shown that microsleeps during simulated driving study occur even with

fully alert participants that were not sleep-deprived (Moller et al., 2006). As no

participant in the current study displayed any overt visual signs of sleep onset or

having experienced any microsleeps using standard criteria, even though EEG theta

and alpha absolute power levels increased over the duration of the HPT session,

some type of effect of motivation is likely.

The lack of effect of motivation on the HPT performance deserves some

discussion. The extant literature suggests that high-order cognitive processes require

the coordinated use of a number of psychological processes for their successful

execution. High-order psychological processes are subserved by the corticothalamic

networks as well as frontal and prefrontal cortices (Jones & Harrison, 2001; Lim et

al., 2010; Manly & Robertson, 1997). The metabolic activity of these cortical and

subcortical areas is substantially reduced when experiencing sleepiness (Lim et al.,

2010; Thomas et al., 2000). Moreover, the cognitive resources required for the

successful execution of high-order task are greater than those required for low-order

tasks (S. P. A. Drummond et al., 2005; S. P. A. Drummond et al., 2004; Galy et al.,

2012; M. E. Smith et al., 2002).When cognitive resources are limited due to the

effect from sleepiness, the co-ordination of several psychological processes is

arduous and results in performance deficits (M. E. Smith et al., 2002; Thomas et al.,

2003). Thus, the lack of effect of motivation on HPT performance when sleepy is

somewhat understandable with the data from Study Three.

Given the discussion above, it was expected that a motivation effect would be

found in the HPT results, particularly with the alert-motivated condition. Certainly,


hazard perception performance can be improved by remedial interventions. For

instance, remedial interventions such as ‘what happens next’ protocols or prediction

exercises (Poulsen, Horswill, Wetton, Hill, & Lim, 2010; Wetton et al., 2013),

running commentary (i.e., think-aloud) of visual search strategies (Isler et al., 2009;

Poulsen et al., 2010), and expert driver commentary training (Crundall, Andrews,

van Loon, & Chapman, 2010; Wetton et al., 2013) have all demonstrated that hazard

perception performance can be improved. While the majority of these studies have

examined if inexperienced drivers’ hazard perception performance can be improved

with training (Crundall et al., 2010; Isler et al., 2009; Poulsen et al., 2010), it is

important to note that hazard perception performance can be improved by more

experienced drivers (> 10 years driving) as well (Horswill, Taylor, Newnam, Wetton,

& Hill, 2013). Thus, the potential did exist for improvement of hazard perception

performance in Study Three, particularly with the alert-motivated condition.

The performance outcomes in Study Three revealed that no such

improvement of hazard perception performance emerged in the HPT data. Although,

this lack of an improvement in hazard perception ability could be due to the lack of

training intervention as described above (e.g., Crundall et al., 2010; Horswill et al.,

2013; Isler et al., 2009). Nonetheless, the present findings are consistent with a

previous study by Hsieh et al. (2010), which had a similar design as Study Three. In

the study by Hsieh et al. (2010) Motivation and Sleepiness Level factors were

similarly experimentally manipulated, however, the motivation effect could not

attenuate the effect that sleep deprivation had on performance levels on a high-order

cognitive task. It must be noted that participants in the Hsieh et al. (2010) study were

completely sleep deprived and this is a considerable point of difference between the

two studies.


Another potential explanation can be offered. The effect of motivation has, in

some studies, resulted in differential effects on performance measures. That is, some

individuals when asked to improve their performance choose to increase either the

speed of responses or accuracy of their responses (Boksem et al., 2006; Hsieh et al.,

2010). It is possible that participants in this study increased their effort on “accuracy”

of their performance. Successful hazard perception is not only predicated on fast

response times but also on comprehensive visual scanning of the road environment

(Chapman et al., 2002; Underwood et al., 2005). Sleepiness also reduces the size of

the useful visual field when performing a driving simulator task (Rogé, Pébayle,

Hannachi, & Muzet, 2003), and the effect from sleepiness also reduces the

magnitude of visual scanning paths (Lavine, Sibert, Gokturk, & Dickens, 2002).

Therefore, participants focusing on accuracy might have increased the amount of

visual scanning or increased another perceptual task/behaviour involved with

attending to the hazard perception stimuli. Considered together, and with respect to

the dynamic and visually engaging nature of the HPT, it could be that greater visual

scanning was a logical way for participants to improve their performance on the

HPT, rather than applying effort to respond faster to the hazards on screen. Future

research could seek to explore this possibility.

6.5.2 The Effects of Motivation with the PVT

The outcomes of the PVT sessions in relation to the physiological and

subjective data were similar to the trends found for HPT physiological and subjective

data. That is, the significant main effects that were found for the Sleepiness Level

factor as well as the Time Period factor both indicated that physiological and

subjective sleepiness levels were greater during the sleepy conditions and that

sleepiness levels increased over the duration of the PVT session. Previous literature


has demonstrated similar findings in relation to the Sleepiness Level factor (Adam et

al., 2006; Dinges et al., 1997; Graw et al., 2004) as well as the Time Period factor

(Corsi-Cabrera et al., 1996; Dorrian et al., 2005; Lim et al., 2012).

Yet, differential effects relating to the PVT performance measures were

observed in the data. Specifically, a significant three-way interaction was observed in

the PVT reaction-time latency data (see Table 6.5). That is, across the duration of the

task, significant increases in reaction-time latencies were observed for all conditions

except for the alert-motivated condition. The alert-motivated condition reaction-time

latency did not increase over the duration of the PVT task, after controlling for

multiple comparisons. Moreover, the mean reaction-time latencies for the alert-

motivated condition were significantly shorter than for the other conditions during

the first and last 5 min of the task.

To date, the PVT results concerning the effect of motivation and sleepiness

on the PVT performance obtained in the current research are unique. What is unique

about this finding is that alert and motivated participants can perform at a higher

level, but no motivation effect was found in the other three conditions. Potentially,

having a benchmark with which to gauge task improvement for applying extra effort

is a key consideration with the obtained PVT performance results. Previous research

has demonstrated that motivation can in some ways attenuate performance

impairment due to sleepiness (Boksem et al., 2006; Steyvers, 1987; Steyvers &

Gaillard, 1993) or even improve task performance back to baseline levels (Horne &

Pettitt, 1985). Certainly, some increases of performance would have been expected

with the sleepy-motivated condition, particularly with the PVT being a low-order

cognitive task (e.g., Horne & Pettitt, 1985). However, studies of motivation effect on

low-order task have typically introduced the motivational component towards the


end of the testing sessions, such that participants have their previous performance

levels to use as a gauge from which to base their comparisons. It has previously been

shown that knowledge of prior results or levels of performance are associated with

shorter reaction-time latencies than when no performances feedback is received

(Steyvers & Gaillard, 1993; Wilkinson, 1961). This seemingly was the same effect

that occurred with results from Study One. Knowledge of previous performance in

effect, creates a competitive situation which is also know to drive an improvement in

performance levels (McAuley & Tammen, 1989). Thus, lack of knowledge of

previous performance among participants in this study might have explained the lack

of motivation effect in the PVT data with the sleepy-motivated condition.

6.5.3 Sleepiness and Time-on-Task Effects

While no significant main effects were found for the motivation conditions,

several significant main effects were found for the Sleepiness Level factor and for

the Time Period factor for both the HPT and PVT data. Beginning with the

Sleepiness Level main effects, all of the study variables were found to be sensitive to

the two levels of the sleepiness. Specifically, EEG absolute power (for all electrode

sites) and subjective sleepiness values were higher during the sleepy conditions than

the alert conditions. Additionally, the HPT and PVT reaction-time latencies were

significantly longer in the sleepy conditions than the alert conditions. A significant

main effect for Time Period was found in the obtained data. During the HPT sessions

and regardless of the experimental condition, physiological and subjective sleepiness

increased, with HPT reaction-time latencies similarly increasing. Both of these

findings of increased sleepiness and poorer performance when sleepy versus alert

and increasing sleepiness and poorer performance over the duration of the HPT


session are consistent with numerous studies (Åkerstedt et al., 2010; Gillberg et al.,

1996; Otmani, Pebayle, Roge, & Muzet, 2005; Sandberg et al., 2011).

6.5.4 Intrinsic Subjective Experiences

The obtained results of the intrinsic subjective experiences that participants

had during the HPT testing session (assessed at the conclusion of the session) also

revealed differences associated with the Sleepiness Level and Motivation Level

factors. Regardless of the Sleepiness Level or Motivation Level manipulations, low

levels of task interest/enjoyment were found and these values did not change across

the experimental conditions. The lack of change in interest/enjoyment is likely due to

the laboratory environment and the tedious tasks the participants were completing,

both of which are known to be unstimulating for participants (e.g., Hallvig et al.,

2013; Thiffault & Bergeron, 2003). However, significant differences were found for

the perceived competence, effort/importance, and pressure/tension experiences.

Regarding the perceived competence experiences, participants rated their

perceived competence during the HPT session higher when alert than when sleepy.

This finding is consistent with previous work that demonstrates task performance is

self-rated as poor when sleep deprived, and as performance continues to deteriorate

due to increasing sleepiness, self-rating of task performance similarly reduces (Biggs

et al., 2007; Dorrian et al., 2000). However, the perceived competence scale is

somewhat different to self-ratings of task performance, as this scale simply assesses

the participants’ perception of performance. The perceived competence scale gauges

the participants perceptions of how successfully they feel they performed the task

(McAuley, Wraith, & Duncan, 1991; Miller & McAuley, 1987).

The intrinsic experiences of pressure/tension also merit some consideration.

The results from the current study demonstrate that participants’ perceptions of


pressure/tension were greater during the sleepy-motivated condition. Sleep

deprivation has been known to effect mood states; specifically, increases in tension-

anxiety have been documented during sleep deprivation (Liu, Zhou, Liu, & Zhao,

2015; Short & Louca, 2015). However, the greater levels of pressure/tension

experienced during the sleepy-motivated condition found in this study are unique.

Emotional regulation has been documented to be impaired when sleep deprived

(Anderson & Platten, 2011; Killgore et al., 2008) and increases in stress are known to

take place during arduous (high workload) performance testing (Robert & Hockey,

1997) including simulated driving (Brookhuis & de Waard, 2001; Matthews &

Desmond, 2002). Thus, it is likely that multiplicative effects from being sleepy and

the increased strain from the motivation instruction to perform at a higher level could

have contributed to the increased subjective experiences of pressure/tension during

the sleepy-motivated condition. The practical implications of such situations, when

sleepy drivers motivate themselves to continue to drive when sleepy could have

important road safety implications.


Limitations of the third and final study will be examined. The sample size

used for Study Three was likely too small to detect any small effect sizes. This

limitation particularly relates to the KSS and HPT analyses from the driving task

sessions, but also to the signs of sleepiness analyses and intrinsic subjective

experiences where a number of small effects failed to reach significance levels.

Future research will likely need to have a greater number of participants to fully

evaluate these small effects that were observed in Study Three.

Another limitation of Study Three concerns the laboratory environment the

study was performed in; and certain caveats are always linked with laboratory-based


research. Several research groups have now shown that driving on the actual road

versus driving in a simulator is associated with higher levels of arousal (Hallvig et

al., 2013; Philip et al., 2005; Törnros, 1998). Thus, it is possible that motivation

could possibly have a larger effect when driving on actual roads, particularly with the

possibility of actual injury due to crashing. However, the ethical as well as practical

issues make such a study extremely difficult to conduct. After all, driving in a real

car on actual roads with EEG and EOG recording equipment connected and having a

researcher sitting in the seat next to you, would likely still feel like an experimental

simulation when compared to normal driving situations. Moreover, ethical and

Health and Safety requirement necessitate that such studies need to be conducted in a

dual-control car and this circumstance could potentially reduce crash risk perceptions

and any subsequent behaviour, if consideration is given to risk homeostasis theory

(e.g., Wilde, 1982).

Overt performance feedback is seemingly an important facilitator of

improving performance levels. The PVT delivers immediate performance feedback

by displaying the reaction time for each trial on screen; however, the HPT does not

give such overt performance feedback. This potentially limited the efficacy of the

motivation instruction that intended to improve participants’ performance levels.

Future research could address this limitation of Study Three, by providing immediate

performance feedback regarding hazard perception reaction-time latencies, similar to

the feedback provided by the PVT. However, providing such HPT performance

feedback drastically changes the characteristics and test engagement with completing

a HPT, as performance feedback is typically not provided to the participant.

A couple of limitations of Study Three relate to the testing protocol. First, it

must be noted the 08:30 start might have resulted in some participants having to


wake up slightly earlier than normal to make this testing time and is potentially a

source of sleep restriction. Second, as described in section 6.3.4, on each of the two

testing days participants completed a motivated and a non-motivated condition on the

same day, which took in total 3 hr 40 min to complete. The 3 hr 40 min duration of

the testing could be considered demanding, although a 40 min interval was included

in the study’s design to ensure that any time-on-task effects accumulated in the first

condition did not carry through to the second condition that was to be completed for

that days testing. While the potential for order effects was controlled for with the

counterbalancing and was also statistically assessed, with no significant order effects

found, it would be restrictive to not consider that other more subtle effects or changes

(e.g., blood glucose change: Fairclough & Houston, 2004) did not occur with

participants completing the 3 hr 40 min testing schedule. While it has been shown

that more demanding cognitive performance tests leads to, on average larger

decreases in blood glucose, a direct correlation between greater reductions of blood

glucose levels and actual task performance is more difficult to substantiate (Backs &

Seljos, 1994; Fairclough & Houston, 2004) and the precise relationship between

mental effort, blood glucose levels, and task performance is complex (Scholey,

Laing, & Kennedy, 2006). Nonetheless, future studies might address this potential

limitation by testing each of the conditions on separate days.

6.6 Chapter Summary

This chapter has described the third and final study of the research program.

This final study addressed the first research question regarding the magnitude and

duration of the effects motivation has on sleepiness and performance levels. This

study experimentally manipulated an aspect of the wake drive, via manipulation of

the participants’ motivational levels and also manipulated the participants’ level of


sleepiness. Physiological, subjective, and performance indices of sleepiness were

obtained to measure the effects of the manipulation of motivation and sleepiness

levels. The performance measures included the HPT, which is a high-order cognitive

task, and the PVT, which is a low-order cognitive task.

The overall finding from the final study was that there was a differential

effect of motivation on the HPT and PVT. Specifically, no effect was observed in the

HPT data in terms of changes in sleepiness or performance levels. However,

physiological and subjective sleepiness were higher in the sleepy conditions and HPT

performance was poorer. Main effects for time were also observed, where sleepiness

increased over the duration of the HPT sessions regardless of the sleepiness or

motivation conditions. This finding in particular highlights the dangerousness of

sleepy driving.

Similar to the HPT findings, no motivation effect was observed in the

physiological or subjective sleepiness data. Physiological and subjective sleepiness

was higher in the sleepy conditions, and physiological and subjective sleepiness

increased over the duration of the PVT task. Regarding the PVT performance data, a

three-way interaction was observed. Reaction-time latencies were faster in the alert-

motivation condition compared to the other conditions and reaction-time latencies

over the duration of the task were maintained in the alert-motivated condition.

Considered together, these findings suggest that sleepiness is very resistant to

motivations to stay alert and improve performance levels. The results from Study

Three suggest that drivers who continue to drive while sleepy by motivating

themselves to apply extra effort to the task of driving are engaging in a risky driving

behaviour.


Chapter Seven: Overview, General Discussion, and Conclusions

You don't start out writing good stuff. You start out writing crap and thinking it's

good stuff, and then gradually you get better at it. That's why I say one of the most

valuable traits is persistence.

–Octavia Butler


The aim of the present research program was to further our understanding of

the effect of motivation on sleepiness and performance as well as the genetic

associations with sleepiness and performance. The obtained findings suggest that

motivation does have an effect on sleepiness and performance, but that this effect is

certainly limited. Yet, motivations to cease driving had a substantial effect on

sleepiness and self-regulation of participants’ behaviour in the HPT. In total, five

different SNPs were identified as being associated with sleepiness and performance.

This final chapter will give an overview of research findings from the three studies,

as well as discussing the theoretical and practical considerations from the findings of

this research program. Last, the limitations of the research program will be discussed,

some suggestions for future research will be proposed, and concluding comments

will be made.

7.2 Review of the Findings

7.2.1 Study One: The Physiological, Subjective, Performance, and Genetic

Associations of Sleepiness in Young Adults

The findings from the first study of this research program suggest that

motivations to improve performance on a low-order cognitive task (the PVT5)


partially influenced physiological and subjective sleepiness. Specifically, when

participants were motivated to improve their performance on the PVT5, the

physiological sleepiness measured at the frontal and central electrode sites did not

increase further. However, physiological sleepiness continued to increase at the

occipital electrode site. Moreover, the motivation instruction was associated with an

attenuation of subjective sleepiness, as measured by the KSS. In addition to the

influence on physiological and subjective sleepiness, a large effect of motivation on

performance indices of sleepiness was observed. That is, a substantial improvement

on PVT5 performance, as demonstrated with a significant reduction of reaction-time

latencies, was found after the application of motivation.

7.2.2 Study Two: Driving Cessation Motivations Study

Study Two of this research program addressed the first research question,

which was concerned with the effect that motivation can have on sleepiness and

performance. To this end, existing data from a simulated driving study examining the

point at which sleepy driving becomes dangerous was reanalysed. In this study,

sleep-deprived participants were instructed to monitor their sleepiness and stop

driving once they perceived their level of sleepiness was too dangerous to drive

safely on the road. The nature of the instructions utilised in this study effectively

removed the participants’ motivations to continue performing the task once they

reached a particular threshold of sleepiness and thus, participants would not have to

fight sleep onset to continue performing the task. As such, this study presented an

opportunity to examine psychophysiological changes in sleepiness when external

motivations to continue driving are removed, which could provide unique

information regarding the increases of physiological and subjective sleepiness when

drivers feel they are too sleepy.


As expected, results of this study showed that after mild sleep restriction

participants’ subjective (as measured by the KSS) and physiological (as measured by

EOG-based blink duration) sleepiness increased over the duration of the drive. Yet,

no significant change in EEG or ECG data was found. In addition to these subjective

and physiological measures, it was also found that the importance of several of the

signs of sleepiness as indicators of the individual level of sleepiness increased from

the start of the hazard perception task to the point of cessation. Importantly, in terms

of driving cessation, this study demonstrated that all participants chose to cease the

hazard perception task before unintended sleep onset or microsleeps had occurred.

These results suggest that, in the absence of motivation to continue to drive while

sleepy, all participants were able to identify increasing subjective sleepiness levels,

and could choose to cease driving.

7.2.3 Study Three: Motivation to Continue Driving While Sleepy, the Effects on

Sleepiness and Performance

This section describes the third and final study of the research program. This

final study addressed the first research question regarding the magnitude and

duration of the effects motivation has on sleepiness and performance levels. This

study experimentally manipulated participants’ motivational levels and level of

sleepiness and measured the effect of these manipulations on physiological,

subjective, and performance indices of sleepiness. Performance indices included both

a high-order cognitive task (the HPT) and a low-order cognitive task (the PVT).

The main results of this study suggested that motivation differentially affects

higher and lower order cognitive tasks. Specifically, no effect was observed in the

HPT data in terms of changes to either sleepiness or performance levels due to the

motivation manipulation. As expected, physiological and subjective sleepiness were


higher in the sleepy conditions and HPT performance was poorer in the sleepy

conditions as well. A main effect for the Time Period factor also demonstrated that

physiological and subjective sleepiness increased over the duration of the HPT

session and HPT performance became progressively worse. This finding in particular

highlights the dangerous of continuing to drive while sleepy.

Similar to the HPT findings, no effect for motivation was observed in the

PVT physiological or subjective sleepiness data. Physiological and subjective

sleepiness was higher in the sleepy conditions and physiological and subjective

sleepiness increased over the duration of the PVT task. Regarding the PVT

performance data, a three-way interaction was observed. Reaction-time latencies

were faster in the alert-motivation condition compared to the other conditions and

only a small increase in reaction-time latencies over the duration of the task were

found in the alert-motivated condition. Considered together, these findings suggest

that sleepiness is very resistant to motivations to stay as alert and improve

performance levels.

7.3 Theoretical Considerations

7.3.1 Task Performance

This section integrates the findings from the three studies of the research

program. Beginning with the overall effects of motivation, the different protocols

used with the implementation of the motivation instruction have provided interesting

insights into motivation and subsequent performance outcomes. For instance, in

Study One, when task performance was deteriorating, as it was at the end of the

PVT10 session, and the motivational instruction was given to participants, there was

a substantial improvement in the PVT reaction-time latencies during the PVT5

session. At the end of the PVT10 session, participants’ reaction-time latencies were


quite slow and subsequently there was room for improvement with the effect of

motivation. Indeed, this sentiment is consistent with the state instability hypothesis

(Doran et al., 2001). The state instability hypothesis attempts to account for

performance impairment and the associated increase in the variability of performance

levels as sleepiness increases. Doran et al. (2001) suggest that the instability of

performance is due to variations in levels of sleepiness and effort; such that,

increases in sleepiness lead to poor performance, which are then countered by an

increase in effort to resist sleepiness and continue performance of the task, and

subsequently, performance in improves (Doran et al., 2001). The state instability

hypothesis has been supported by data from several studies (Balkin et al., 2004;

Doran et al., 2001; Dorrian et al., 2005; Durmer, 2005; Kleitman, 1963; Zhou et al.,

2011); however, several important aspect of this hypothesis have not been previous

examined.

It was previously suggested that knowledge of the results from increased

effort is an important reinforcer (Steyvers & Gaillard, 1993). Indeed, the PVT

provides immediate feedback regarding the level of task performance, and it is likely

that knowledge of the results is potentially an important aspect contributing to

improved performance as documented in Doran et al.’s (2001) data using the PVT as

a performance metric where the state instability hypothesis was elucidated. It was

observed in Study Three that the effect of effort was only limited to the alert-

motivated condition and no improvement was observed in the sleepy-motivated

condition. This is likely due to no benchmark of previous performance results being

available to participants. It is acknowledged that the study protocol did have

participants perform a motivated and a non-motivated PVT session on the same day.

However, it is likely that participants might not have been able to recall the results of


the first PVT testing session, as there was a large interval of time between the two

PVT testing sessions (i.e., 1 hr 40 min), which were performed on the same day for

the motivated and non-motivated conditions. An inability to recall the previous PVT

results between testing sessions of each condition would mean no participant had a

benchmark for which to gauge their subsequent PVT performance on and possibly

improve upon with the motivated conditions.

Regarding the protocol differences between Study One and Three, the

motivation session of Study One took place straight after performing the PVT10.

Whereas, Study Three’s protocol separated the motivated and non-motivated

conditions into separate conditions. Given the above discussion regarding

benchmarking, it would be useful to perform the HPT sessions in a similar manner to

the protocol of Study One. It is possible that including a motivation session towards

the end of the HPT could provide some ecological validity with drivers’ on-road

practices. However, the study by Boksem et al. (2006) used this type of protocol with

a monetary motivator for improved performance on a high-order cognitive task and

found only limited effectiveness of motivation. Participants in the Boksem et al.

(2006) study increased either accuracy or the speed of their responses, but the levels

of improved task performance did not completely attenuate time-on-task effects.

A large proportion of drivers (69-73%) report that they would continue to

drive even when aware of their increasing sleepiness (Nordbakke & Sagberg, 2007;

Watling, Armstrong, & Radun, 2015). Additionally, increasing motivation and effort

to the task of driving is a commonly used tactic by younger drivers, once they

become sleepy when driving on the road (Lucidi et al., 2006). Thus, for road safety it

is important to determine what effects increased motivation effort can have on

driving performance and sleepiness levels when applied towards the end of a drive


once the driver is quite sleepy. It must be noted that participants in Boksem et al.

(2006) received performance feedback, and errors in task performance were

associated with physiological differences in the increase in the amplitude of the

error-related negativity (ERN), which is an error-related event related potential

component. This finding suggests that participants in the Boksem et al. (2006) were

engaging in the task and attempting to improve their performance levels. While the

current study utilised a different measure of physiological arousal (i.e., spectral

analysis) a measure of participants’ effort throughout the task and post-hoc measures

of effort were included in the study protocol. It was previously highlighted in section

6.2, that more effort is needed to maintain task performance when sleepy than when

alert and this finding was supported in the present study.

7.3.2 Wake Drive Effects

It is noted in the four-process model that there are a number of neuronal and

cognitive inputs into the secondary wake drive. Specifically, the secondary wake

drive is proposed to be the outcome from neuronal impulses from the afferent nerve

system (i.e., postural muscles and movement/activity), inputs from the visual system,

and mental stimulation (Johns, 1993; Johns, 1998). Only one aspect of the secondary

wake drive was manipulated in the present research program and no significant effect

of this manipulation was observed in any of the physiological data during the HPT

testing sessions. While no effects were found in the physiological data, it is possible

that other measures of brain activity could have detected changes in brain activity

during the motivated condition. Certainly, there are sufficient neuroimaging studies

that have demonstrated that additional cortical areas are recruited during task

performance when sleep deprived (S. P. A. Drummond et al., 2004; Vandewalle et


al., 2009). However, subcortical structures could also be involved with increases of

the wake drive.

Certainly, the flip-flop switch described by Saper et al. (2005) suggests that

wakefulness is promoted by the ascending arousal system. One major input to this

arousal system comes from distinct structures in the brainstem that innervate the

thalamus. The second input comes from other structures in the brainstem that project

to the cerebral cortex for its activation (Saper, Fuller, Pedersen, Lu, & Scammell,

2010; Saper et al., 2005). Other neuroimaging studies demonstrate that many of these

subcortical structures, such as the corticothalamic networks as well as frontal and

prefrontal cortices are involved when performing vigilance tasks (Jones & Harrison,

2001; Lim et al., 2010). The anterior cingulate cortex is also another candidate

cortical area that could demonstrate increased activity with augmentation of the wake

drive. That is, neuroimaging studies show the anterior cingulate cortex is highly

active during attentional tasks such as, error monitoring, selection-for-action tasks,

and monitoring of performance (Botvinick, Nystrom, Fissell, Carter, & Cohen, 1999;

Devinsky, Morrell, & Vogt, 1995; MacDonald, Cohen, Stenger, & Carter, 2000).

Only future neuroimaging studies can determine the functions particular cortical and

subcortical areas have with augmentation of the wake drive.

Another aspect of the wake drive that the present research program

potentially highlights that could be the association of particular genes. Study One

demonstrated that the SNP rs1554338 (CRY gene) was associated with EEG Theta

absolute power levels during the PVT10 session and the SNP rs5751876 (ADORA2A

gene) was associated with EEG alpha absolute power levels during the PVT10

session and during the motivated PVT5 testing session. In particular, the ADORA2A

gene is worthy of continued examination as significant reductions in alpha absolute


power levels were observed in the current study. The ADORA2A gene is a protein-

encoding gene of the adenosine A2 receptor and appears to be involved in

homeostatic regulation. That is, the accumulation of adenosine in cortical tissue

reflects the duration of time elapsed since awakening and thus, is believed to reflect

increases of homeostatic pressure (Gallopin et al., 2005; Ticho & Radulovacki,

1991). As the TT genotype of the SNP rs5751876 (ADORA2A gene) was found to

have higher levels of alpha absolute power levels during the PVT5 motivated session

this might suggest a physiological vulnerability of the TT genotype to applying extra

effort when motivated.

7.4 Practical Considerations

This section will examine and explore the practical considerations from the

three studies of the research program. The second and third study of the research

program had a specific focus on driving and thus some consideration of the results is

seemingly warranted. This section will first examine the overall effects of motivation

from Study Two and Study Three and what this might mean for actual driving on the

road when sleepy. Then a discussion of the relationship between motivations and

self-perceptions of sleepiness, the signs of sleepiness, and performance will be

undertaken.

7.4.1 Motivations and On-Road Consideration

While few significant effects of motivation were found on sleepiness and

performance indices, if consideration is given to the combined results of Study Two

and Study Three then motivation potentially has a major effect on road safety.

Beginning with Study Two, it was found that when participants were motivated to

cease driving they were able to recognise increasing sleepiness (via the KSS and

blink duration measures), and they were able to cease driving before unintentional


sleep onset or microsleeps occurred. In contrast with Study Three, motivations were

not shown to lead to a reduction of physiological or subjective sleepiness, and task

performance on the HPT deteriorated significantly over the duration of the task. As

such, sleepy drivers who notice an elevated level of sleepiness and choose not to

cease driving and instead choose to prolong their driving by increasing their

motivation are likely to be at considerable risk of crashing. Indeed, it has been found

that as the level of subjective sleepiness increases, there is a corresponding near-

exponential increase in the likelihood of departing the driving lane (Reyner & Horne,

2000; Watling, Åkerstedt, et al., 2015) and a having a sleep-related crash (Åkerstedt,

Connor, et al., 2008).

Rewards are important with any goal-directed activity, and aspects of

intrinsic and extrinsic motivation certainly facilitate the effort allocated to any task

(Deci, Koestner, & Ryan, 1999). Sleepy drivers who choose to motivate themselves

to fight sleepiness and continue driving, and remain awake are potentially reinforcing

their own behaviour. Particularly, if they do not have a sleep-related close call or a

crash, and they arrive at their destination without any negative incidents. It has been

shown that when goals are set by individuals, regardless of the actual outcome of

their performance, they rate the level of performance higher than those who do not

set goals (Miller & McAuley, 1987). Considering this finding in the context of sleepy

driving, if a sleepy driver’s goal is to remain awake by fighting sleepiness by

increasing the effort applied to the driving task, and they arrive at their destination,

they could perceive their driving performance as acceptable while they were fighting

sleepiness to remain awake. Moreover, electrophysical studies of attention and task

performance suggest that error evaluation is impaired when experiencing high levels

of sleepiness (Murphy et al., 2006), and behavioural studies have also shown that


risk-taking is more pronounced when sleepy (Rossa et al., 2014). Considered

together, the decision not to cease driving but to continue to drive while sleepy and

apply extra effort is a highly concerning issue for road safety.

7.4.2 Subjective Perceptions

7.4.2.1 Perceptions of task performance

In the absence of a reliable and validated technology that can monitor the

level of sleepiness of a driver, the individual driver is reliant on their own

perceptions of their subjective sleepiness and driving performance to determine when

they are too sleepy to drive safely. While the findings of Study Three did not show a

significant effect of motivation on subjective sleepiness, the motivated PVT session

of Study One did suggest that motivation had an effect on subjective sleepiness.

Certainly, the overall findings from the present research program and the extant

findings suggest that motivation has more of an effect for improving performance on

low-order cognitive tasks than high-order cognitive tasks (Boksem et al., 2006;

Bonnet & Arand, 2005a; Horne & Pettitt, 1985; Steyvers, 1987).

Maintaining lane positioning is an important driving skill (van Loon,

Brouwer, & Martens, 2015). Increases in lateral positioning have been demonstrated

to occur when driving while sleepy and the variations (amount of sideways

movement) increases as the level of sleepiness increases (Arnedt et al., 2005;

Campagne et al., 2004). Driving outside of the road lane and crashing is a common

consequence of sleepy driving (Horne & Reyner, 1995; Radun & Summala, 2004).

Lane positioning could be considered a low-order cognitive task (i.e., a tracking task)

and thus could potentially be improved with extra effort. For instance, it has been

shown that alertness and driving performance assessed by lane positioning, improve

when concurrently performing an interactive cognitive task aimed at increasing


arousal levels, such as motivation might (e.g., Gershon, Ronen, Oron-Gilad, &

Shinar, 2009; Verwey & Zaidel, 1999). Should a motivated sleepy driver improve

their lane positioning while driving, this might be taken by the driver to indicate that

their extra effort is effective at improving the driving performance. However, it has

been demonstrated in Study Three that motivation to perform better when sleepy

does nothing to improve hazard perception performance, nor does it lead to any

attenuation or reduction of physiological or subjective sleepiness.

Improvements in lane positioning might also be taken by the driver to

indicate that their subjective sleepiness is reducing. Study One has indicated that

improvements in task performance might influence perceptions of subjective

sleepiness. It has previously been suggested that subjective sleepiness mediates

perceptions of task performance (Biggs et al., 2007; Dorrian et al., 2000); however,

the results from Study One and consideration of the extant literature (Biggs et al.,

2007; Dorrian et al., 2000) suggests that this relationship might be bi-directional.

7.4.2.2 Perceptions of signs of sleepiness

Some discussion is warranted with the results from the signs of sleepiness

ratings. It was found in Study Two that for certain signs of sleepiness (slow eye

blinks, difficulty concentrating, slower reactions, changing position, and frequent eye

blinks) their importance ratings as indicator of level of sleepiness increased from

baseline levels to levels reported after cessation of the hazard perception task. It is

possible that the importance ratings of these signs of sleepiness increased as

participants were instructed to monitor their sleepiness levels and to stop driving

when too sleepy. In essence, this instruction could have caused participants to

reorientate their attention from that of “passive attention” to that of an “active” or

“focused attention” (i.e., Posner, Nissen, & Ogden, 1978; Treisman & Gelade, 1980)


with perceiving the signs of sleepiness they were experiencing. Such an effect has

been noted with the level of attention focused on the awareness of experiencing of

emotions (Bush, Luu, & Posner, 2000; Lambie & Marcel, 2002; Lane et al., 1998) as

well as the perception of visual stimuli (Beanland & Chan, 2016; Simons &

Ambinder, 2005).

Indeed, such an interpretation is also useful for accounting for the outcomes

of the signs of sleepiness analyses in Study Three. That is, significant main effects

for the Pre-Post Session factors were found and indicate that on average, the ratings

of importance of the signs of sleepiness increased after completing the HPT sessions.

This effect could be due to the procedure used in Study Three whereby KSS data was

collected every 5 min during the HPT sessions. It is possible that the periodic ratings

of subjective sleepiness “focused” the participants’ attention or awareness on the

signs of sleepiness they were experiencing. It was incidentally shown from the

results of Study Three that participants’ physiological and subjective sleepiness

increased and this too could have contributed to the increased ratings of importance

of the signs of sleepiness at the end of the hazard perception driving tasks.

Alternatively, the increase could simply be due to the baseline ratings of the signs of

sleepiness being erroneously low than what the participants’ actual ratings of

importance of the signs of sleepiness for indicating how sleepy they were.

7.5 Suggestions for Future Research Directions

Three key suggestions for future research based on findings from the present

program of research are presented here. First, an increased sample size and exploring

the relationships of other SNPs with sleepiness and performance outcomes is

definitely warranted. The effects sizes from the ANOVAs statistics suggested that

small effect sizes were present, but the power to detect these effects was potentially


limited by the small sample size. Additionally, a larger sample size will also allow

for the testing of interactions between the various genotypes and variations in time of

sleepiness and performance levels. Certain genotypes that confer specific

vulnerability to sleepiness (e.g., SNPs from PERIOD3, ADORA2A) might possibly

be associated with greater increases in levels of sleepiness and greater performance

decrements.

Numerous genes are involved in the regulation of the human circadian

rhythm and only a small number of circadian SNPs were examined in Study One. As

such, analysing more SNPs in a larger sample is also a suggestion for future research.

If future research was to focus on a larger number of circadian SNPs, then it is also

worthy to consider the influence chronotype of the individuals can have on

performance measures. Several studies have demonstrated many circadian genes are

associated with diurnal preference (Allebrandt et al., 2010; Archer et al., 2003;

Ebisawa et al., 2001; Jones et al., 2007). Moreover, previous research has shown

morning chronotypes tend to have a more stable cognitive and simulated driving

performance during daytime testing, with performance worsening during the

afternoon and evening. In contrast, evening chronotypes cognitive and simulated

driving performance is typically poorest during the mornings and then stabilises or

improves during the afternoon and evening (Correa et al., 2014; Hahn et al., 2012;

Lara et al., 2014). Considered together, examining the association of specific

circadian SNPs and chronotype on sleepiness and performance outcomes could be

beneficial to understanding why some individuals in the absence of any sleep

disorder or severe sleep debt have sleep-related crashes.

Secondly, the present research program used the HPT as the driving stimuli

and the measure of driving performance; however, sophisticated high-fidelity driving


simulators can capture a number of other metrics of driving performance (e.g., lane

positioning/lateral positing, driving speed, steering corrections). It has been

suggested in section 7.4.2 Motivations and On-Road Considerations, that increased

effort might facilitate a reduction in the amount of variation of lane positioning a

driver might exhibit while sleepy. Exploring potential situations relating to

motivation and vehicle control indices as well as drivers’ perceptions of changes in

their own performance and subjective sleepiness levels would increase our

understanding of the effect of motivation and driving performance levels. It was

discussed in section 4.5.1, that knowledge of improvement in PVT task performance

might be associated with a maintaining of subjective sleepiness level. The occurrence

of improved driving performance possibly influencing subjective sleepiness levels

could be an important road safety consideration and needs to be explored with future

research.

The third suggestion relates to when the motivation to improve performance

is initiated. Conducting a driving simulator study that instructs the participant to

apply extra effort after they have been driving for a certain amount of time would

have a higher level of ecological validity with particular sleepy driving behaviours.

As shown in Study Two as well as an on road study performed by Åkerstedt et al.

(2013), high levels of sleepiness can be experienced on average after approximately

40 minutes. It is possible that a sleepy driver will after only 40 minutes of driving,

choose to push on and motivate them self to continue driving and to reach their

destination. Previous research has documented that a large proportion of drivers (69-

73%) report that they would continue to drive even when aware of increasing levels

of sleepiness (Nordbakke & Sagberg, 2007; Watling, Armstrong, & Radun, 2015).

Thus, as continuing to drive while sleepy remains a prevalent behaviour on the roads,


research examining the effects of motivations on driver’s sleepiness and performance

levels is still warranted.

7.6 Summary of Chapter

In conclusion, this program of research sought to examine the effects of

motivations and genetic associations on levels of sleepiness and performance.

Overall, the current results suggest that motivation has a limited effect on

physiological and subjective sleepiness. Differential effects of motivation on

performance levels were found. Specifically, the effect of motivation on improving

task performance was observed on the PVT, a low-order cognitive task. In contrast,

motivation seemingly had no effect on improving performance levels on the HPT, a

high-order cognitive task. Regarding the genetic association with sleepiness and

performance, five SNPs were identified that were associated with different aspects of

sleepiness and performance when completing the PVT. These findings suggest

individual genetic variations can influences certain aspect of physiological and

subjective sleepiness as well as performance outcomes.

This research program has made a number of important contributions to the

area of sleepiness and performance. Regarding the theoretical contribution, this

research program has improved our understanding of the sleep and wake drives of

the four-process model of sleep-wake (Johns, 1993). Studies conducted in the

program have demonstrated that one aspect of the wake drive, being motivation,

could have a limited effect on sleepiness with a greater effect on performance.

Additionally, findings relating to specific genes were also shown to be related to the

motivational aspect of the wake drive. Regarding the practical contributions, several

survey studies have quantified self-reported aspects of driving while sleepy and

motivations to continue driving while sleepy, but there appears to be dearth of


studies that have examined the actual psychophysiological and performance sequel

of motivating oneself to apply extra effort to the task of driving when sleepy. The

program of research presented in this dissertation has demonstrated some important

aspects of the effect of motivations on sleepiness and performance. The importance

of progressing educational campaigns regarding the dangerousness of sleepy driving

and in particular continuing to drive while sleepy is seemingly warranted and this

research program can potentially be of some value. Such efforts are important

undertakings to reduce road trauma and provide a safer road environment for all road

users.


Writing is the flip side of sex - it's good only when it's over.

–Hunter S. Thompson


References

Abe, T., Komada, Y., Asaoka, S., Ozaki, A., & Inoue, Y. (2011). Questionnaire‐

based evidence of association between sleepiness while driving and motor

vehicle crashes that are subjectively not caused by falling asleep. Sleep and

Biological Rhythms, 9(3), 134-143.

Abe, T., Komada, Y., Nishida, Y., Hayashida, K., & Inoue, Y. (2010). Short sleep

duration and long spells of driving are associated with the occurrence of

Japanese drivers’ rear‐end collisions and single‐car accidents. Journal of

sleep research, 19(2), 310-316.

Abrahamson, E. E., & Moore, R. Y. (2006). Lesions of suprachiasmatic nucleus

efferents selectively affect rest-activity rhythm. Molecular and Cellular

Endocrinology, 252(1-2), 46-56.

Achermann, P. (2004). The Two-Process Model of Sleep Regulation Revisited.

Aviation, Space, and Environmental Medicine, 75(3,Sect2), A37-A43.

Adam, M., Retey, J. V., Khatami, R., & Landolt, H.-P. (2006). Age-related changes

in the time course of vigilant attention during 40 hours without sleep in men.

Sleep, 29(1), 55-57.

Aeschbach, D., Postolache, T. T., Sher, L., Matthews, J. R., Jackson, M. A., & Wehr,

T. A. (2001). Evidence from the waking electroencephalogram that short

sleepers live under higher homeostatic sleep pressure than long sleepers.

Neuroscience, 102(3), 493-502.

Åhsberg, E., Kecklund, G., Åkerstedt, T., & Francesco, G. (2000). Shiftwork and

different dimensions of fatigue. International Journal of Industrial

Ergonomics, 26(4), 457-465.


Åkerstedt, T. (2000). Consensus Statement: Fatigue and accidents in transport

operations. Journal of Sleep Research, 9(4), 395-395.

Åkerstedt, T. (2007). Altered sleep/wake patterns and mental performance.

Physiology & Behavior, 90(2-3), 209-218.

Åkerstedt, T., Axelsson, J., & Kecklund, G. (2007). Individual validation of model

predictions of sleepiness and sleep hours. Somnologie - Schlafforschung und

Schlafmedizin, 11(3), 169-174.

Åkerstedt, T., Connor, J., Gray, A., & Kecklund, G. (2008). Predicting road crashes

from a mathematical model of alertness regulation--The Sleep/Wake

Predictor. Accident Analysis & Prevention, 40(4), 1480-1485.

Åkerstedt, T., & Folkard, S. (1995). Validation of the S and C components of the

three-process model of alertness regulation. Sleep, 18(1), 1-6.

Åkerstedt, T., & Folkard, S. (1997). The three-process model of alertness and its

extension to performance, sleep latency, and sleep length. Chronobiology

International, 14(2), 115-123.

Åkerstedt, T., & Gillberg, M. (1990). Subjective and objective sleepiness in the

active individual. International Journal of Neuroscience, 52(1-2), 29-37.

Åkerstedt, T., Hallvig, D., Anund, A., Fors, C., Schwarz, J., & Kecklund, G. (2013).

Having to stop driving at night because of dangerous sleepiness - awareness,

physiology and behaviour. Journal of Sleep Research, 22(4), 380-388.

Åkerstedt, T., Ingre, M., Kecklund, G., Anund, A., Sandberg, D., Wahde, M., . . .

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.


Åkerstedt, T., Kecklund, G., & Axelsson, J. (2008). Effects of context on sleepiness

self-ratings during repeated partial sleep deprivation. Chronobiology


Akerstedt, T., Kecklund, G., & Knutsson, A. (1991). Spectral analysis of sleep

electroencephalography in rotating three-shift work. Scandinavian Journal of

Work, Environment & Health, 17(5), 330-336.

Åkerstedt, T., Peters, B., Anund, A., & Kecklund, G. (2005). Impaired alertness and

performance driving home from the night shift: A driving simulator study.

Journal of Sleep Research, 14(1), 17-20.

Albers, H. E., Lydic, R., Gander, P. H., & Moore-Ede, M. C. (1984). Role of the

suprachiasmatic nuclei in the circadian timing system of the squirrel monkey.

I. The generation of rhythmicity. Brain Research, 300(2), 275-284.

Allebrandt, K. V., Teder-Laving, M., Akyol, M., Pichler, I., Müller-Myhsok, B.,

Pramstaller, P., . . . Roenneberg, T. (2010). CLOCK Gene Variants Associate

with Sleep Duration in Two Independent Populations. Biological Psychiatry,

67(11), 1040-1047.

Allen, M. T., Obrist, P. A., Sherwood, A., & Crowell, M. D. (1987). Evaluation of

myocardial and peripheral vascular responses during reaction time, mental

arithmetic, and cold pressor tasks. Psychophysiology, 24(6), 648-656.

Aloba, O. O., Adewuya, A. O., Ola, B. A., & Mapayi, B. M. (2007). Validity of the

Pittsburgh Sleep Quality Index (PSQI) among Nigerian university students.

Sleep Medicine, 8(3), 266-270.

Ambrosius, U., Lietzenmaier, S., Wehrle, R., Wichniak, A., Kalus, S., Winkelmann,

J., . . . Friess, E. (2008). Heritability of sleep electroencephalogram.

Biological Psychiatry, 64(4), 344-348.


American Academy of Sleep Medicine, & Iber, C. (2007). The AASM manual for the

scoring of sleep and associated events: rules, terminology and technical

specifications. Illinois, United States of America: American Academy of

Sleep Medicine.

Anderson, C., & Horne, J. A. (2006). Sleepiness Enhances Distraction During a

Monotonous Task. Sleep, 29(4), 573-576.

Anderson, C., & Horne, J. A. (2008). Placebo response to caffeine improves reaction

time performance in sleepy people. Human Psychopharmacology: Clinical

and Experimental, 23(4), 333-336.

Anderson, C., & Horne, J. A. (2013). Driving drowsy also worsens driver distraction.


Anderson, C., & Platten, C. R. (2011). Sleep deprivation lowers inhibition and

enhances impulsivity to negative stimuli. Behavioural Brain Research,

217(2), 463-466.

Anund, A., Kecklund, G., Peters, B., & Åkerstedt, T. (2008). Driver sleepiness and

individual differences in preferences for countermeasures. Journal of Sleep

Research, 17(1), 16-22.

Anund, A., Kecklund, G., Peters, B., Forsman, Å., Lowden, A., & Åkerstedt, T.

(2008). Driver impairment at night and its relation to physiological

sleepiness. Scandinavian Journal of Work, Environment & Health, 34(2),

142-150.

Apparies, R. J., Riniolo, T. C., & Porges, S. W. (1998). A psychophysiological

investigation of the effects of driving longer-combination vehicles.

Ergonomics, 41(5), 581-592.


Archer, S. N., Robilliard, D. L., Skene, D. J., Smits, M., Williams, A., Arendt, J., &

von Schantz, M. (2003). A length polymorphism in the circadian clock gene

per3 is linked to delayed sleep phase syndrome and extreme diurnal

preference. Sleep, 26(4), 413-415.

Archer, S. N., Viola, A. U., Kyriakopoulou, V., von Schantz, M., & Dijk, D.-J.

(2008). Inter-individual differences in habitual sleep timing and entrained

phase of endogenous circadian rhythms of BMAL1, PER2 and PER3 mRNA

in human leukocytes. Sleep, 31(5), 608-617.

Armstrong, K. A., Filtness, A. J., Watling, C. N., Barraclough, P., & Haworth, N.

(2013). Efficacy of proxy definitions for identification of fatigue/sleep-related

crashes: An Australian evaluation. Transportation Research Part F: Traffic

Psychology and Behaviour, 21(0), 242-252.

Armstrong, K. A., Obst, P., Banks, T., & Smith, S. S. (2010). Managing driver

fatigue: Education or motivation? Road & Transport Research, 19(3), 14-20.

Arnedt, J. T., Geddes, M. A. C., & MacLean, A. W. (2005). Comparative sensitivity

of a simulated driving task to self-report, physiological, and other

performance measures during prolonged wakefulness. Journal of

Psychosomatic Research, 58, 61-71.

Arnedt, J. T., Wilde, G., Munt, P., & Maclean, A. (2000). Simulated driving

performance following prolonged wakefulness and alcohol consumption:

separate and combined contributions to impairment. Journal of Sleep

Research, 9(3), 233-241.

Australian Transport Safety Bureau. (2006). Characteristics of fatal road crashes

during national holiday periods. Canberra: Commonwealth of Australia


Australian Transport Safety Bureau. (2009). Road deaths Australia 2008: Statistical

summary (DOTARS 50249). Canberra: Commonwealth of Australia

Backhaus, J., Junghanns, K., Broocks, A., Riemann, D., & Hohagen, F. (2002). Test-

retest reliability and validity of the Pittsburgh Sleep Quality Index in primary

insomnia. Journal of Psychosomatic Research, 53(3), 737-740.

Backs, R. W., & Seljos, K. A. (1994). Metabolic and cardiorespiratory measures of

mental effort: the effects of level of difficulty in a working memory task.

International Journal of Psychophysiology, 16(1), 57-68.

Baddeley, A. (1992). Working memory. Science, 255(5044), 556-559.

Balkin, T. J., Bliese, P. D., Belenky, G., Sing, H. C., Thorne, D. R., Thomas, M. L., .

. . Wesensten, N. J. (2004). Comparative utility of instruments for monitoring

sleepiness-related performance decrements in the operational environment.


Balkin, T. J., Braun, A. R., Wesensten, N. J., Jeffries, K., Varga, M., Baldwin, P., . . .

Herscovitch, P. (2002). The process of awakening: A PET study of regional

brain activity patterns mediating the reestablishment of alertness and

consciousness. Brain: A Journal of Neurology, 125(10), 2308-2319.

Banks, S., Catcheside, P., Lack, L., Grunstein, R. R., & McEvoy, R. D. (2004). Low

levels of alcohol impair driving simulator performance and reduce perception

of crash risk in partially sleep deprived subjects. Sleep, 27(6), 1063-1067.

Banks, S., Van Dongen, H. P. A., Maislin, G., & Dinges, D. F. (2010).

Neurobehavioral dynamics following chronic sleep restriction: Dose-response

effects of one night for recovery. Sleep, 33(8), 1013-1026.


Barbato, G., De Padova, V., Paolillo, A. R., Arpaia, L., Russo, E., & Ficca, G.

(2007). Increased spontaneous eye blink rate following prolonged

wakefulness. Physiology & Behavior, 90(1), 151-154.

Barrett, P. R., Horne, J. A., & Reyner, L. A. (2004). Alcohol continues to affect

sleepiness related driving impairment, when breath alcohol levels have fallen

to near-zero. Human Psychopharmacology: Clinical and Experimental,

19(6), 421-423.

Bartlett, D. J., Marshall, N. S., Williams, A., & Grunstein, R. R. (2008). Sleep health

New South Wales: chronic sleep restriction and daytime sleepiness. Internal

Medicine Journal, 38(1), 24-31.

Basheer, R., Strecker, R. E., Thakkar, M. M., & McCarley, R. W. (2004). Adenosine

and sleep–wake regulation. Progress in Neurobiology, 73(6), 379-396.

Basner, M., & Dinges, D. F. (2011). Maximizing sensitivity of the psychomotor

vigilance test (PVT) to sleep loss. Sleep, 34(5), 581-591.

Basner, M., Fomberstein, K. M., Razavi, F. M., Banks, S., William, J. H., Rosa, R.

R., & Dinges, D. F. (2007). American time use survey: Sleep time and its

relationship to waking activities. Sleep, 30(9), 1085-1095.

Basner, M., Mollicone, D., & Dinges, D. F. (2011). Validity and sensitivity of a brief

psychomotor vigilance test (PVT-B) to total and partial sleep deprivation.

Acta Astronautica, 69, 949-959.

Beanland, V., & Chan, E. H. C. (2016). The relationship between sustained

inattentional blindness and working memory capacity. Attention, Perception,

& Psychophysics, 78(3), 808-817.


Beck, S. L., Schwartz, A. L., Towsley, G., Dudley, W., & Barsevick, A. (2004).

Psychometric evaluation of the Pittsburgh sleep quality index in cancer

patients. Journal of Pain and Symptom Management, 27(2), 140-148.

Benbadis, S. R., Mascha, E., Perry, M. C., Wolgamuth, B. R., Smolley, L. A., &

Dinner, D. S. (1999). Association between the Epworth sleepiness scale and

the multiple sleep latency test in a clinical population. Annals of Internal

Medicine, 130(4 Pt 1), 289-292.

Benson, K., Friedman, L., Noda, A., Wicks, D., Wakabayashi, E., & Yesavage, J.

(2004). The measurement of sleep by actigraphy: Direct comparison of 2

commercially available actigraphs in a nonclinical population. Sleep, 27(5),

986-989.

Berry, J. G., & Harrison, J. E. (2007). Serious injury due to land transport accidents,

Australia, 2003–04 (INJCAT 107). Canberra: Australian Institute of Health

and Welfare

Bertolino, A., Blasi, G., Latorre, V., Rubino, V., Rampino, A., Sinibaldi, L., . . .

Scarabino, T. (2006). Additive effects of genetic variation in dopamine

regulating genes on working memory cortical activity in human brain. The

Journal of Neuroscience, 26(15), 3918-3922.

Biggs, S. N., Smith, A., Dorrian, J., Reid, K., Dawson, D., van den Heuvel, C. J., &

Baulk, S. (2007). Perception of simulated driving performance after sleep

restriction and caffeine. Journal of Psychosomatic Research, 63(6), 573-577.

Birnboim, H. C. (2011). DNA yield with an Oragene self-collection kit (PD-WP-

001). Ontario, Canada: DNA Genotek Inc.


Blagrove, M., & Akehurst, L. (2001). Personality and the modulation of effects of

sleep loss on mood and cognition. Personality and Individual Differences,

30(5), 819-828.

Blaivas, A. J., Patel, R., Hom, D., Antigua, K., & Ashtyani, H. (2007). Quantifying

microsleep to help assess subjective sleepiness. Sleep Medicine, 8(2), 156-

159.

Bodenmann, S., Hohoff, C., Freitag, C., Deckert, J., Retey, J. V., Bachmann, V., &

Landolt, H.-P. (2012). Polymorphisms of ADORA2A modulate psychomotor

vigilance and the effects of caffeine on neurobehavioural performance and

sleep EEG after sleep deprivation. British Journal of Pharmacology, 165(6),

1904-1913.

Bodenmann, S., Rusterholz, T., Durr, R., Stoll, C., Bachmann, V., Geissler, E., . . .

Landolt, H.-P. (2009). The functional Val158Met polymorphism of COMT

predicts interindividual differences in brain alpha oscillations in young men.

The Journal of Neuroscience, 29(35), 10855-10862.

Boksem, M. A. S., Meijman, T. F., & Lorist, M. M. (2006). Mental fatigue,

motivation and action monitoring. Biological Psychology, 72(2), 123-132.

Bonnet, M. H., & Arand, D. L. (1998). Sleepiness as measured by modified multiple

sleep latency testing varies as a function of preceding activity. Sleep, 21(5),

477-483.

Bonnet, M. H., & Arand, D. L. (1999). Level of arousal and the ability to maintain

wakefulness. Journal of Sleep Research, 8(4), 247-254.

Bonnet, M. H., & Arand, D. L. (2000). The impact of music upon sleep tendency as

measured by the multiple sleep latency test and maintenance of wakefulness

test. Physiology & Behavior, 71(5), 485-492.


Bonnet, M. H., & Arand, D. L. (2001). Arousal components which differentiate the

MWT from the MSLT. Sleep, 24(4), 441-447.

Bonnet, M. H., & Arand, D. L. (2005a). Impact of motivation on multiple sleep

latency test and maintenance of wakefulness test measurements. Journal of

Clinical Sleep Medicine, 1(4), 386-390.

Bonnet, M. H., & Arand, D. L. (2005b). Sleep latency testing as a time course

measure of state arousal. Journal of Sleep Research, 14(4), 387-392.

Bonnet, M. H., & Moore, S. E. (1982). The threshold of sleep: Perception of sleep as

a function of time asleep and auditory threshold. Sleep, 5(3), 267-276.

Borbély, A. A. (1982). A two process model of sleep regulation. Human

Neurobiology, 1(3), 195-204.

Borbély, A. A., & Achermann, P. (1992). Concepts and models of sleep regulation:

An overview. Journal of Sleep Research, 1(2), 63-79.

Borbély, A. A., & Achermann, P. (1999). Sleep homeostasis and models of sleep

regulation. Journal of Biological Rhythms, 14(6), 557-568.

Borkenstein, R. F., Crowther, R. F., Shumate, R. P., Zeil, W. W., & Zylman, R.

(1964). The role of the drinking driver in traffic accidents. Bloomington,

Indiana: Indiana University

Borowsky, A., Oron-Gilad, T., & Parmet, Y. (2009). Age and skill differences in

classifying hazardous traffic scenes. Transportation Research Part F: Traffic


Botvinick, M., Nystrom, L. E., Fissell, K., Carter, C. S., & Cohen, J. D. (1999).

Conflict monitoring versus selection-for-action in anterior cingulate cortex.

Nature, 402(6758), 179-181.


Boufous, S., Ivers, R., Senserrick, T., & Stevenson, M. (2011). Attempts at the

Practical On-Road Driving Test and the Hazard Perception Test and the Risk

of Traffic Crashes in Young Drivers. Traffic Injury Prevention, 12(5), 475-

482.

Boyle, L. N., Tippin, J., Paul, A., & Rizzo, M. (2008). Driver performance in the

moments surrounding a microsleep. Transportation Research Part F: Traffic


Brookes, A. J. (1999). The essence of SNPs. Gene, 234(2), 177-186.

Brookhuis, K. A., & de Waard, D. (2001). Assessment of Drivers' Workload:

Performance and Subjective and Physiological Indexes. Stress, workload and

fatigue.

Broughton, R. (1975). Biorhythmic variations in consciousness and psychological

functions. Canadian Psychological Review/Psychologie canadienne, 16(4),

217-239.

Brown, I. D. (1994). Driver fatigue. Human Factors, 36(2), 298-314.

Brownley, K. A., Hurwitz, B. E., & Schneiderman, N. (2007). Cardiovascular

Psychophysiology. In J. T. Cacioppo (Ed.), Handbook of psychophysiology

(3rd ed.). New York, NY US: Cambridge University Press.

Bruenig, D., White, M. J., Young, R. M., & Voisey, J. (2014). Subclinical Psychotic

Experiences in Healthy Young Adults: Associations with Stress and Genetic

Predisposition. Genetic Testing and Molecular Biomarkers, 18(10), 683-689.

Bureau of Infrastructure Transport and Regional Economics. (2009). Cost of road

crashes in Australia 2006. Canberra, Australia: Department of Infrastructure,

Transport, Regional Development and Local Government


Bureau of Infrastructure Transport and Regional Economics. (2015). Road trauma

Australia 2014 statistical summary. Canberra, Australia: Department of

Infrastructure and Transport

Bush, G., Luu, P., & Posner, M. I. (2000). Cognitive and emotional influences in

anterior cingulate cortex. Trends in cognitive sciences, 4(6), 215-222.

Butler, J. M. (2012). Chapter 12 - Single Nucleotide Polymorphisms and

Applications Advanced Topics in Forensic DNA Typing (pp. 347-369). San

Diego: Academic Press.

Buysse, D. J., Reynolds, C. F., 3rd, Monk, T. H., Berman, S. R., & Kupfer, D. J.

(1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric

practice and research. Psychiatry Research, 28(2), 193-213.

Buysse, D. J., Reynolds, C. F., 3rd, Monk, T. H., Hoch, C. C., Yeager, A. L., &

Kupfer, D. J. (1991). Quantification of subjective sleep quality in healthy

elderly men and women using the Pittsburgh Sleep Quality Index (PSQI).

Sleep, 14(4), 331-338.

Cacioppo, J. T., Tassinary, L. G., & Berntson, G. G. (2007). Handbook of

psychophysiology (3rd ed.). New York, NY US: Cambridge University Press.

Caffier, P. P., Erdmann, U., & Ullsperger, P. (2003). Experimental evaluation of eye-

blink parameters as a drowsiness measure. European Journal of Applied

Physiology, 89(3-4), 319-325.

Cajochen, C., Khalsa, S. B., Wyatt, J. K., Czeisler, C. A., & Dijk, D. J. (1999). EEG

and ocular correlates of circadian melatonin phase and human performance

decrements during sleep loss. The American Journal of Physiology, 277(3 Pt

2), R640-R649.


Caldwell, J. A. (2001). The impact of fatigue in air medical and other types of

operations: A review of fatigue facts and potential countermeasures. Air

Medical Journal, 20(1), 25-32.

Caldwell, J. A., & Caldwell, J. L. (1998). Comparison of the effects of zolpidem-

induced prophylactic naps to placebo naps and forced rest periods in

prolonged work schedules. Sleep, 21(1), 79-90.

Caldwell, J. A., Prazinko, B., & Caldwell, J. L. (2003). Body posture affects

electroencephalographic activity and psychomotor vigilance task

performance in sleep-deprived subjects. Clinical Neurophysiology, 114(1),

23-31.

Caldwell, J. A., Prazinko, B. F., & Hall, K. K. (2000). The effects of body posture on

resting electroencephalographic activity in sleep-deprived subjects. Clinical

Neurophysiology, 111(3), 464-470.

Campagne, A., Pebayle, T., & Muzet, A. (2004). Correlation between driving errors

and vigilance level: Influence of the driver's age. Physiology & Behavior,

80(4), 515-524.

Campos Viola, F., Thorne, J., Edmonds, B., Schneider, T., Eichele, T., & Debener, S.

(2009). Semi-automatic identification of independent components

representing EEG artifact. Clinical Neurophysiology, 120(5), 868-877.

Carpen, J. D., Schantz, M., Smits, M., Skene, D. J., & Archer, S. N. (2006). A silent

polymorphism in the PER1 gene associates with extreme diurnal preference

in humans. Journal of Human Genetics, 51(12), 1122-1125.

Carpenter, J. S., & Andrykowski, M. A. (1998). Psychometric evaluation of the

Pittsburgh Sleep Quality Index. Journal of Psychosomatic Research, 45(1), 5-

13.


Carskadon, M. A., & Dement, W. C. (1982). The multiple sleep latency test: What

does it measure? Sleep, suppl2, S128-146.

Carskadon, M. A., & Dement, W. C. (1992). Multiple sleep latency tests during the

constant routine. Sleep, 15(5), 396-399.

Carskadon, M. A., & Dement, W. C. (2005). Normal Human Sleep: An Overview. In

W. H. Kryger, T. Roth, & W. C. Dement (Eds.), Principles and practice of

sleep medicine (pp. 16-26). Philadelphia: Elsevier/Saunders.

Carskadon, M. A., Dement, W. C., Mitler, M. M., Roth, T., Westbrook, P. R., &

Keenan, S. (1986). Guidelines for the multiple sleep latency test (MSLT): A

standard measure of sleepiness. Sleep, 9(4), 519-524.

Cassone, V. M. (1990). Melatonin: time in a bottle. Oxf Rev Reprod Biol, 12, 319-

367.

Castillo, M. R., Hochstetler, K. J., Tavernier, R. J., Greene, D. M., & Bult-Ito, A.

(2004). Entrainment of the master circadian clock by scheduled feeding.

American Journal of Physiology-Regulatory, Integrative and Comparative

Physiology, 287(3), R551-R555.

Cercarelli, L. R., & Haworth, N. L. (2002). The determination of fatigue-related

crashes using routinely collected road crash data (RR120). Crawley, WA:

Injury Research Centre, The University of Western Australia

Chapman, P., Underwood, G., & Roberts, K. (2002). Visual search patterns in trained

and untrained novice drivers. Transportation Research Part F: Traffic


Chen, J., Lipska, B. K., Halim, N., Ma, Q. D., Matsumoto, M., Melhem, S., . . .

Apud, J. (2004). Functional analysis of genetic variation in catechol-O-

methyltransferase (COMT): effects on mRNA, protein, and enzyme activity


in postmortem human brain. The American Journal of Human Genetics,

75(5), 807-821.

Chen, N. H., Johns, M. W., Li, H. Y., Chu, C. C., Liang, S. C., Shu, Y. H., . . . Wang,

P. C. (2002). Validation of a Chinese version of the Epworth sleepiness scale.

Quality of Life Research, 11(8), 817-821.

Chervin, R. D., Aldrich, M. S., Pickett, R., & Christian, G. (1997). Comparison of

the results of the Epworth Sleepiness Scale and the Multiple Sleep Latency

Test. Journal of Psychosomatic Research, 42(2), 145-155.

Choi, J., Mogami, T., & Medalia, A. (2010). Intrinsic Motivation Inventory: An

Adapted Measure for Schizophrenia Research. Schizophrenia Bulletin, 36(5),

966-976.

Choo, W.-C., Lee, W.-W., Venkatraman, V., Sheu, F.-S., & Chee, M. W. L. (2005).

Dissociation of cortical regions modulated by both working memory load and

sleep deprivation and by sleep deprivation alone. NeuroImage, 25(2), 579-

587.

Chua, E. C.-P., Tan, W.-Q., Yeo, S.-C., Lau, P., Lee, I., Mien, I. H., . . . Gooley, J. J.

(2012). Heart rate variability can be used to estimate sleepiness-related

decrements in psychomotor vigilance during total sleep deprivation. Sleep,

35(3), 325-334.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. New Jersey:

Lawrenced Erlbaum Associates, Inc. Publishers.

Cole, R. J. (1989). Postural baroreflex stimuli may affect EEG arousal and sleep in

humans. Journal of Applied Physiology, 67(6), 2369-2375.


Cole, R. J., Kripke, D. F., Gruen, W., Mullaney, D. J., & Gillin, J. C. (1992).

Automatic sleep/wake identification from wrist activity. Sleep, 15(5), 461-

469.

Collette, F., Hogge, M., Salmon, E., & Van der Linden, M. (2006). Exploration of

the neural substrates of executive functioning by functional neuroimaging.


Connor, J., Norton, R., Ameratunga, S., Robinson, E., Civil, I., Dunn, R., . . .

Jackson, R. (2002). Driver sleepiness and risk of serious injury to car

occupants: population based case control study. British Medical Journal,

324(7346), 1125-1130.

Correa, A., Molina, E., & Sanabria, D. (2014). Effects of chronotype and time of day

on the vigilance decrement during simulated driving. Accident Analysis &

Prevention, 67, 113-118.

Corsi-Cabrera, M., Arce, C., Ramos, J., & Lorenzo, I. (1996). Time course of

reaction time and EEG while performing a vigilance task during total sleep

deprivation. Sleep.

Crundall, D., Andrews, B., van Loon, E., & Chapman, P. (2010). Commentary

training improves responsiveness to hazards in a driving simulator. Accident

Analysis & Prevention, 42(6), 2117-2124.

Cummins, R. A., Woerner, J., Weinberg, M., Collard, J., Hartley-Clark, L., Perera,

C., & Horfiniak, K. (2012). Australian unity wellbeing index survey 27.0 part

A : the report The wellbeing of Australians - quantity and quality of sleep.

Geelong, Victoria: Deakin University.

Czeisler, C. A., Richardson, G. S., Zimmerman, J. C., Moore-Ede, M. C., &

Weitzman, E. D. (1981). Entrainment of human circadian rhythms by light-


dark cycles: a reassessment. Photochemistry And Photobiology, 34(2), 239-

247.

Daan, S., Beersma, D. G., & Borbély, A. A. (1984). Timing of human sleep:

recovery process gated by a circadian pacemaker. The American Journal of

Physiology, 246(2 Pt 2), R161-R183.

Darby, P., Murray, W., & Raeside, R. (2009). Applying online fleet driver

assessment to help identify, target and reduce occupational road safety risks.

Safety Science, 47(3), 436-442.

Dardente, H., Dardente, H., & Cermakian, N. (2007). Molecular circadian rhythms in

central and peripheral clocks in mammals. Chronobiology International,

24(2), 195-213.

Darlington, T. K., Wager-Smith, K., Ceriani, M. F., Staknis, D., Gekakis, N.,

Steeves, T. D., . . . Kay, S. A. (1998). Closing the circadian loop: CLOCK-

induced transcription of its own inhibitors per and tim. Science, 280(5369),

1599-1603.

Dauvilliers, Y., & Tafti, M. (2006). Molecular genetics and treatment of narcolepsy.

Annals of Medicine, 38(4), 252-262.

Davison, A. C., & Hinkley, D. V. (1997). Bootstrap Methods and Their Application:

Cambridge University Press.

Dawson, D., & Reid, K. (1997). Fatigue, alcohol and performance impairment.

Nature, 388(6639), 235-235.

De Gennaro, L., Marzano, C., Fratello, F., Moroni, F., Pellicciari, M. C., Ferlazzo,

F., . . . Sforza, E. (2008). The electroencephalographic fingerprint of sleep is

genetically determined: a twin study. Annals of neurology, 64(4), 455-460.


De Padova, V., Barbato, G., Conte, F., & Ficca, G. (2009). Diurnal variation of

spontaneous eye blink rate in the elderly and its relationships with sleepiness

and arousal. Neuroscience Letters, 463(1), 40-43.

De Valck, E., & Cluydts, R. (2001). Slow-release caffeine as a countermeasure to

driver sleepiness induced by partial sleep deprivation. Journal of Sleep

Research, 10(3), 203-209.

De Valck, E., & Cluydts, R. (2003). Sleepiness as a state-trait phenomenon,

comprising both a sleep drive and a wake drive. Medical Hypotheses, 60(4),

509-512.

Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of

experiments examining the effects of extrinsic rewards on intrinsic

motivation. Psychological Bulletin, 125(6), 627-668.

Deery, H. A. (1999). Hazard and risk perception among young novice drivers.

Journal of Safety Research, 30(4), 225-236.

Department of Transport and Main Roads. (2009). Queensland Road Toll 2008.

Brisbane: Department of Transport and Main Roads

Department of Transport and Main Roads. (2015). 2014 Summary Road Crash

Report: Queensland Road Fatalities. Brisbane: Department of Transport and

Main Roads

Devinsky, O., Morrell, M. J., & Vogt, B. A. (1995). Contributions of anterior

cingulate cortex to behaviour. Brain, 118(1), 279-306.

Di Milia, L. (2006). Shift work, sleepiness and long distance driving. Transportation

Research Part F: Traffic Psychology and Behaviour, 9(4), 278-285.


Dijk, D. J., Brunner, D. P., Beersma, D. G., & Borbély, A. A. (1990).

Electroencephalogram power density and slow wave sleep as a function of

prior waking and circadian phase. Sleep, 13(5), 430-440.

Dijk, D. J., Duffy, J. F., & Czeisler, C. A. (1992). Circadian and sleep/wake

dependent aspects of subjective alertness and cognitive performance. Journal

of Sleep Research, 1(2), 112-117.

Dijkman, M., Sachs, N., Levine, E., Mallis, M., Carlin, M. M., Gillen, K. A., . . .

Dinges, D. F. (1997). Effects of reduced stimulation on neurobehavioral

alertness depend on circadian phase during human sleep deprivation. Sleep

Research, 26, 265.

Dinges, D. F. (1995). An overview of sleepiness and accidents. Journal of Sleep

Research, 4(s2), 4-14.

Dinges, D. F. (2004). Critical research issues in development of biomathematical

models of fatigue and performance. Aviation, Space, and Environmental

Medicine, 75(3 Suppl), A181-A191.

Dinges, D. F., & Achermann, P. (1999). Commentary: Future considerations for

models of human neurobehavioral function. Journal of Biological Rhythms,

14(6), 598-601.

Dinges, D. F., Orne, M. T., Whitehouse, W. G., & Orne, E. C. (1987). Temporal

placement of a nap for alertness: contributions of circadian phase and prior

wakefulness. Sleep, 10(4), 313-329.

Dinges, D. F., Pack, F., Williams, K., Gillen, K. A., Powell, J. W., Ott, G. E., . . .

Pack, A. I. (1997). Cumulative sleepiness, mood disturbance, and

psychomotor vigilance performance decrements during a week of sleep

restricted to 4-5 hours per night. Sleep, 20(4), 267-277.


Dinges, D. F., & Powell, J. W. (1985). Microcomputer Analyses of Performance on a

Portable, Simple Visual RT Task during Sustained Operations. Behavior

Research Methods, 17(6), 652-655.

DNA Genotek Inc. (2012). Laboratory protocol for manual purification of DNA from

whole sample (PD-PR-015). Ontario, Canada: DNA Genotek Inc.

Dobbie, K. (2002). Fatigue-related crashes: An analysis of fatigue-related crashes

on Australian roads using an operational definition of fatigue (OR 23).

Canberra: Australian Transport Safety Bureau

Doi, Y., Minowa, M., Uchiyama, M., Okawa, M., Kim, K., Shibui, K., & Kamei, Y.

(2000). Psychometric assessment of subjective sleep quality using the

Japanese version of the Pittsburgh Sleep Quality Index (PSQI-J) in

psychiatric disordered and control subjects. Psychiatry Research, 97(2-3),

165-172.

Doran, S. M., Van Dongen, H. P., & Dinges, D. F. (2001). Sustained attention

performance during sleep deprivation: evidence of state instability. Archives

Italiennes De Biologie, 139(3), 253-267.

Dorrian, J., Lamond, N., & Dawson, D. (2000). The ability to self-monitor

performance when fatigued. Journal of Sleep Research, 9(2), 137-144.

Dorrian, J., Rogers, N. L., & Dinges, D. F. (2005). Psychomotor vigilance

performance: Neurocognitive assay sensitive to sleep loss. In K. C. (Ed.),

Sleep Deprivation. New York: Marcel Dekker New York.

Drake, C. L., Roehrs, T. A., Burduvali, E., Bonahoom, A., Rosekind, M., & Roth, T.

(2001). Effects of rapid versus slow accumulation of eight hours of sleep loss.

Psychophysiology, 38(6), 979-987.


Drummond, A. E. (2000). Paradigm lost! Paradise gained? An Australian’s

perspective on the novice driver problem. Paper presented at the Novice

Driver Conference (1st–2nd June), Bristol, United Kingdom.

Drummond, S. P. A., Bischoff-Grethe, A., Dinges, D. F., Ayalon, L., Mednick, S. C.,

& Meloy, M. J. (2005). The neural basis of the psychomotor vigilance task.

Sleep, 28(9), 1059-1068.

Drummond, S. P. A., Brown, G. G., Salamat, J. S., & Gillin, J. C. (2004). Increasing

Task Difficulty Facilitates the Cerebral Compensatory Response to Total

Sleep Deprivation. Sleep, 27(3), 445-451.

Dunlap, J. C. (1999). Molecular Bases for Circadian Clocks. Cell, 96(2), 271-290.

Durmer, J. (2005). Neurocognitive Consequences of Sleep Deprivation. Seminars in

Neurology, 25(1), 117-129.

Ebisawa, T., Uchiyama, M., Kajimura, N., Mishima, K., Kamei, Y., Katoh, M., . . .

Yamauchi, T. (2001). Association of structural polymorphisms in the human

period3 gene with delayed sleep phase syndrome. EMBO Reports, 2(4), 342-

346.

Edgar, D. M., Dement, W. C., & Fuller, C. A. (1993). Effect of SCN lesions on sleep

in squirrel monkeys: Evidence for opponent processes in sleep-wake

regulation. The Journal of Neuroscience, 13(3), 1065-1079.

Edlow, B. L., Takahashi, E., Wu, O., Benner, T., Dai, G., Bu, L., . . . Kinney, H. C.

(2012). Neuroanatomic connectivity of the human ascending arousal system

critical to consciousness and its disorders. Journal of Neuropathology &

Experimental Neurology, 71(6), 531-546.

Efron, B., & Tibshirani, R. (1993). An Introduction to the Bootstrap: Chapman &

Hall.


Egan, M. F., Goldberg, T. E., Kolachana, B. S., Callicott, J. H., Mazzanti, C. M.,

Straub, R. E., . . . Weinberger, D. R. (2001). Effect of COMT Val108/158

Met genotype on frontal lobe function and risk for schizophrenia.

Proceedings of the National Academy of Sciences, 98(12), 6917-6922.

Enoch, M.-A., Xu, K., Ferro, E., Harris, C. R., & Goldman, D. (2003). Genetic

origins of anxiety in women: a role for a functional catechol-O-

methyltransferase polymorphism. Psychiatric genetics, 13(1), 33-41.

Espie, C. A. (2002). INSOMNIA: Conceptual Issues in the Development,

Persistence, and Treatment of Sleep Disorder in Adults. Annual Review of

Psychology, 53(1), 215-243.

Evans, L. (1991). Traffic safety and the driver. New York: Van Nostrand Reinhold.

Fairclough, S. H., & Houston, K. (2004). A metabolic measure of mental effort.

Biological Psychology, 66(2), 177-190.

Fairclough, S. H., & Roberts, J. S. (2011). Effects of performance feedback on

cardiovascular reactivity and frontal EEG asymmetry. International Journal

of Psychophysiology, 81(3), 291-298.

Ferguson, S. A., Clarkson, L., & Dawson, D. (2012). Sealy Sleep Census: Australia's

Biggest Sleep Survey. Adelaide: Central Queensland University.

Ferrer-Caja, E., & Weiss, M. R. (2000). Predictors of Intrinsic Motivation among

Adolescent Students in Physical Education. Research Quarterly for Exercise

and Sport, 71(3), 267-279.

Field, A. P. (2009). Discovering statistics using SPSS: (and sex, drugs and

rock'n'roll) (3rd ed.). London: SAGE.

Finelli, L. A., Baumann, H., Borbély, A. A., & Achermann, P. (2000). Dual

electroencephalogram markers of human sleep homeostasis: Correlation


between theta activity in waking and slow-wave activity in sleep.


Fletcher, A., McCulloch, K., Baulk, S. D., & Dawson, D. (2005). Countermeasures

to driver fatigue: A review of public awareness campaigns and legal

approaches. Australian and New Zealand Journal of Public Health, 29, 471-

476.

Folkard, S., & Åkerstedt, T. (1991). A three process model of the regulation of

alerness and sleepiness. In R. Ogilvie & R. Broughton (Eds.), Sleep, arousal

and performance: Problems and promises (pp. 11-26). Boston: Birkhauser.

Fong, S. Y. Y., Ho, C. K. W., & Wing, Y. K. (2005). Comparing MSLT and ESS in

the measurement of excessive daytime sleepiness in obstructive sleep apnoea

syndrome. Journal of Psychosomatic Research, 58(1), 55-60.

Franken, P., & Dijk, D. J. (2009). Circadian clock genes and sleep homeostasis.

European Journal of Neuroscience, 29(9), 1820-1829.

Franks, C. M. (1963). Ocular movements and spontaneous blink rate as functions of

personality. Perceptual And Motor Skills, 16(1).

Frey, D. J., Badia, P., & Wright, J. K. P. (2004). Inter- and intra-individual

variability in performance near the circadian nadir during sleep deprivation.


Friedman, N. P., Miyake, A., Young, S. E., DeFries, J. C., Corley, R. P., & Hewitt, J.

K. (2008). Individual differences in executive functions are almost entirely

genetic in origin. Journal of Experimental Psychology: General, 137(2), 201-

225.


Gachon, F., Nagoshi, E., Brown, S. A., Ripperger, J., & Schibler, U. (2004). The

mammalian circadian timing system: from gene expression to physiology.

Chromosoma, 113(3), 103-112.

Gallopin, T., Luppi, P. H., Cauli, B., Urade, Y., Rossier, J., Hayaishi, O., . . . Fort, P.

(2005). The endogenous somnogen adenosine excites a subset of sleep-

promoting neurons via A2A receptors in the ventrolateral preoptic nucleus.

Neuroscience, 134(4), 1377-1390.

Galy, E., Cariou, M., & Mélan, C. (2012). What is the relationship between mental

workload factors and cognitive load types? International Journal of


Gander, P. H., Marshall, N. S., Harris, R., B. , & Reid, P. (2005). Sleep, sleepiness

and motor vehicle accidents: A national survey. Australian and New Zealand

Journal of Public Health, 29(1), 16-21.

Garaulet, M., Sánchez-Moreno, C., Smith, C. E., Lee, Y.-C., Nicolás, F., & Ordovás,

J. M. (2011). Ghrelin, sleep reduction and evening preference: relationships

to CLOCK 3111 T/C SNP and weight loss. PLoS One, 6(2), e17435.

Garbarino, S., Nobili, L., Beelke, M., De Carli, F., & Ferrillo, F. (2001). The

contributing role of sleepiness in highway vehicle accidents. Sleep, 24(2),

203-206.

Gasser, T., Bächer, P., & Steinberg, H. (1985). Test-retest reliability of spectral

parameters of the EEG. Electroencephalography and Clinical


Gasser, T., Sroka, L., & Möcks, J. (1985). The transfer of EOG activity into the EEG

for eyes open and closed. Electroencephalography and Clinical



Gehring, W. J., Goss, B., Coles, M. G., & Meyer, D. E. (1993). A neural system for

error detection and compensation. Psychological Science, 4(6), 385-390.

George, C. F. P. (2003). Driving simulators in clinical practice. Sleep Medicine

Reviews, 7(4), 311-320.

Gershon, P., Ronen, A., Oron-Gilad, T., & Shinar, D. (2009). The effects of an

interactive cognitive task (ICT) in suppressing fatigue symptoms in driving.

Transportation Research Part F: Traffic Psychology and Behaviour, 12(1),

21-28.

Gevins, A. S., Yeager, C. L., Zeitlin, G. M., Ancoli, S., & Dedon, M. F. (1977). On-

line computer rejection of EEG artifact. Electroencephalography and Clinical


Gilbert, S. S., van den Heuvel, C. J., Ferguson, S. A., & Dawson, D. (2004).

Thermoregulation as a sleep signalling system. Sleep Medicine Reviews, 8(2),

81-93.

Gillberg, M., & Åkerstedt, T. (1982). Body temperature and sleep at different times

of day. Sleep, 5(4), 378-388.

Gillberg, M., Kecklund, G., & Åkerstedt, T. (1994). Relations between performance

and subjective ratings of sleepiness during a night awake. Sleep, 17, 236-241.

Gillberg, M., Kecklund, G., & Åkerstedt, T. (1996). Sleepiness and performance of

professional drivers in a truck simulator--comparisons between day and night

driving. Journal of Sleep Research, 5(1), 12-15.

Gillberg, M., Kecklund, G., Göransson, B., & Åkerstedt, T. (2003). Operator

performance and signs of sleepiness during day and night work in a simulated

thermal power plant. International Journal of Industrial Ergonomics, 31(2),

101-109.


Glass, J. D., DiNardo, L. A., & Ehlen, J. C. (2000). Dorsal raphe nuclear stimulation

of SCN serotonin release and circadian phase-resetting. Brain research,

859(2), 224-232.

Goel, N., Banks, S., Mignot, E., & Dinges, D. F. (2009). PER3 Polymorphism

Predicts Cumulative Sleep Homeostatic but Not Neurobehavioral Changes to

Chronic Partial Sleep Deprivation. PloS One, 4(6), 1-13.

Goel, N., Banks, S., Mignot, E., & Dinges, D. F. (2010). DQB1* 0602 predicts

interindividual differences in physiologic sleep, sleepiness, and fatigue.

Neurology, 75(17), 1509-1519.

Goel, N., & Dinges, D. F. (2011). Behavioral and genetic markers of sleepiness.

Journal of Clinical Sleep Medicine: JCSM: Official Publication of the

American Academy of Sleep Medicine, 7(5 suppl), S19.

Goldberger, J. J. (1999). Sympathovagal balance: how should we measure it?

American Journal of Physiology - Heart and Circulatory Physiology, 276(4),

H1273-H1280.

Goldman, R. I., Stern, J. M., Engel Jr, J., & Cohen, M. S. (2002). Simultaneous EEG

and fMRI of the alpha rhythm. Neuroreport, 13(18), 2487.

Goldstein, D., Hahn, C. S., Hasher, L., Wiprzycka, U. J., & Zelazo, P. D. (2007).

Time of day, intellectual performance, and behavioral problems in Morning

versus Evening type adolescents: Is there a synchrony effect? Personality and

Individual Differences, 42(3), 431-440.

Gopher, D., & Kimchi, R. (1989). Engineering psychology. Annual Review of

Psychology, 40, 431-455.

Gorgoni, M., Ferlazzo, F., Ferrara, M., Moroni, F., D'Atri, A., Fanelli, S., . . . De

Gennaro, L. (2014). Topographic electroencephalogram changes associated


with psychomotor vigilance task performance after sleep deprivation. Sleep

Medicine, 15(9), 1132-1139.

Gratton, G. (2007). Biosignal processing. In J. T. Cacioppo, L. G. Tassinary, G. G.

Berntson, J. T. Cacioppo, L. G. Tassinary, & G. G. Berntson (Eds.),

Handbook of psychophysiology (3rd ed.). (pp. 834-858). New York, NY, US:


Graw, P., Kräuchi, K., Knoblauch, V., Wirz-Justice, A., & Cajochen, C. (2004).

Circadian and wake-dependent modulation of fastest and slowest reaction

times during the psychomotor vigilance task. Physiology & Behavior, 80(5),

695-701.

Graydon, F. X., Young, R., Benton, M. D., Genik Ii, R. J., Posse, S., Hsieh, L., &

Green, C. (2004). Visual event detection during simulated driving:

Identifying the neural correlates with functional neuroimaging.

Transportation Research Part F: Traffic Psychology and Behaviour, 7(4-5),

271-286.

Groeger, J. A. (2002). Trafficking in cognition: Applying cognitive psychology to

driving. Transportation Research Part F: Traffic Psychology and Behaviour,

5(4), 235-248.

Groeger, J. A., Viola, A. U., Lo, J. C. Y., von Schantz, M., Archer, S. N., & Dijk, D.-

J. (2008). Early morning executive functioning during sleep deprivation is

compromised by a PERIOD3 polymorphism. Sleep, 31(8), 1159-1167.

Guijt, A. M., Sluiter, J. K., & Frings-Dresen, M. H. W. (2007). Test-Retest

Reliability of Heart Rate Variability and Respiration Rate at Rest and during

Light Physical Activity in Normal Subjects. Archives of Medical Research,

38(1), 113-120.


Gundel, A., Drescher, J., Maa, H., Samel, A., & Vejvoda, M. (1995). Sleepiness of

civil airline pilots during two consecutive night flights of extended duration.

Biological Psychology, 40(1-2), 131-141.

Guo, S. W., & Thompson, E. A. (1992). Performing the exact test of Hardy-

Weinberg proportion for multiple alleles. Biometrics, 361-372.

Hagemann, D., & Naumann, E. (2001). The effects of ocular artifacts on (lateralized)

broadband power in the EEG. Clinical Neurophysiology, 112(2), 215-231.

Hahn, C., Cowell, J. M., Wiprzycka, U. J., Goldstein, D., Ralph, M., Hasher, L., &

Zelazo, P. D. (2012). Circadian Rhythms in Executive Function during the

Transition to Adolescence: The Effect of Synchrony between Chronotype and

Time of Day. Developmental Science, 15(3), 408-416.

Haimov, I., & Arendt, J. (1999). The prevention and treatment of jet lag. Sleep

Medicine Reviews, 3(3), 229-240.

Hallvig, D., Anund, A., Fors, C., Kecklund, G., & Akerstedt, T. (2014). Real driving

at night--predicting lane departures from physiological and subjective

sleepiness. Biological psychology, 101, 18-23.

Hallvig, D., Anund, A., Fors, C., Kecklund, G., Karlsson, J. G., Wahde, M., &

Akerstedt, T. (2013). Sleepy driving on the real road and in the simulator--A

comparison. Accident Analysis & Prevention, 50, 44-50.

Harrison, Y., & Horne, J. A. (1996a). "High sleepability without sleepiness". The

ability to fall asleep rapidly without other signs of sleepiness. Clinical


Harrison, Y., & Horne, J. A. (1996b). Occurrence of 'microsleeps' during daytime

sleep onset in normal subjects. Electroencephalography and Clinical



Harrison, Y., & Horne, J. A. (2000). The impact of sleep deprivation on decision

making: A review. Journal of Experimental Psychology: Applied, 6(3), 236-

249.

Hart, S. G., & Staveland, L. E. (1988). Development of NASA-TLX (Task Load

Index): Results of empirical and theoretical research. Advances in

psychology, 52, 139-183.

Hartse, K. M., Roth, T., & Zorick, F. J. (1982). Daytime sleepiness and daytime

wakefulness: The effect of instruction. Sleep, 5(2), 107-118.

Hastings, M. H., Duffield, G. E., Smith, E. J. D., Maywood, E. S., & Ebling, F. J. P.

(1998). Entrainment of the Circadian System of Mammals by Nonphotic

Cues. Chronobiology International, 15(5), 425-445.

Hatskevich, C. W., Itkis, M. L., & Maloletnev, V. I. (1992). Off-line methods for

detection and correction of EEG artefacts of various origin. International

Journal of Psychophysiology, 12(2), 179-185.

Hayashi, M., Ito, S., & Hori, T. (1999). The effects of a 20-min nap at noon on

sleepiness, performance and EEG activity. International Journal of


Henderson, J. M. T., France, K. G., & Blampied, N. M. (2011). The consolidation of

infants' nocturnal sleep across the first year of life. Sleep Medicine Reviews,

15(4), 211-220.

Herrmann, U. S., Hess, C. W., Guggisberg, A. G., Roth, C., Gugger, M., & Mathis, J.

(2010). Sleepiness is not always perceived before falling asleep in healthy,

sleep-deprived subjects. Sleep Medicine, 11(8), 747-751.


Honma, S., Kawamoto, T., Takagi, Y., Fujimoto, K., Sato, F., Noshiro, M., . . .

Honma, K.-i. (2002). Dec1 and Dec2 are regulators of the mammalian

molecular clock. Nature, 419(6909), 841-844.

Horikawa, E., Okamura, N., Tashiro, M., Sakurada, Y., Maruyama, M., Arai, H., . . .

Itoh, M. (2005). The neural correlates of driving performance identified using

positron emission tomography. Brain and Cognition, 58(2), 166-171.

Horne, J. A. (2010). Sleepiness as a need for sleep: When is enough, enough?

Neuroscience and Biobehavioral Reviews, 34(1), 108-118.

Horne, J. A., & Burley, C. V. (2010). We know when we are sleepy: Subjective

versus objective measurements of moderate sleepiness in healthy adults.

Biological Psychology, 83(3), 266-268.

Horne, J. A., & Foster, S. C. (1996). Can exercise overcome sleepiness? [Abstract].

Sleep Research, 24A, 437.

Horne, J. A., & Pettitt, A. N. (1985). High incentive effects on vigilance performance

during 72 hours of total sleep deprivation. Acta Psychologica, 58(2), 123-

139.

Horne, J. A., & Reyner, L. A. (1995). Sleep related vehicle accidents. British

Medical Journal, 310(6979), 565-567.

Horne, J. A., & Reyner, L. A. (1996). Counteracting driver sleepiness: effects of

napping, caffeine, and placebo. Psychophysiology, 33(3), 306-309.

Horne, J. A., & Reyner, L. A. (2001). Sleep-related vehicle accidents: Some guides

for road safety policies. Transportation Research Part F: Traffic Psychology

and Behaviour, 4(1), 63-74.

Horswill, M. S., Anstey, K. J., Hatherly, C. G., & Wood, J. M. (2010). The crash

involvement of older drivers is associated with their hazard perception


latencies. Journal of the International Neuropsychological Society, 16(05),

939-944.

Horswill, M. S., Hill, A., & Wetton, M. (2015). Can a video-based hazard perception

test used for driver licensing predict crash involvement? Accident Analysis &


Horswill, M. S., Marrington, S. A., McCullough, C. M., Wood, J., Pachana, N. A.,

McWilliam, J., & Raikos, M. K. (2008). The harard perception ability of

older drivers. The Journals of Gerontology, 63B(4), 212-218.

Horswill, M. S., & McKenna, F. P. (1999). The development, validation, and

application of a video-based technique for measuring an everyday risk-taking

behavior: Drivers' speed choice. Journal of Applied Psychology, 84(6), 977-

985.

Horswill, M. S., & McKenna, F. P. (2004). Driver’s hazard perception ability:

Situation awareness on the road. In S. Banbury & S. Tremblay (Eds.), A

Cognitive Approach to Situation Awareness (pp. 155-174). Hampshire, UK:

Ashgate Publishing.

Horswill, M. S., Taylor, K., Newnam, S., Wetton, M., & Hill, A. (2013). Even highly

experienced drivers benefit from a brief hazard perception training

intervention. Accident Analysis & Prevention, 52(0), 100-110.

Howard, M. E., Jackson, M. L., Berlowitz, D., O'Donoghue, F., Swann, P., Westlake,

J., . . . Pierce, R. J. (2014). Specific sleepiness symptoms are indicators of

performance impairment during sleep deprivation. Accident Analysis &



Hsieh, S., Li, T.-H., & Tsai, L.-L. (2010). Impact of monetary incentives on

cognitive performance and error monitoring following sleep deprivation.

Sleep, 33(4), 499-507.

Hull, J. T., Wright, K. P., Jr., & Czeisler, C. A. (2003). The Influence of Subjective

Alertness and Motivation on Human Performance Independent of Circadian

and Homeostatic Regulation. Journal of Biological Rhythms, 18(4), 329-338.

Hull, M., & Christie, R. (1992). Hazard perception test: The Geelong trial and future

development. Paper presented at the Proceedings of the National Road Safety

Seminar, Wellington, New Zealand, Wellington, New Zealand.

Hunsley, J. (2003). Introduction to the special section on incremental validity and

utility in clinical assessment. Psychological Assessment, 15(4), 443-445.

Hunsley, J., & Meyer, G. J. (2003). The incremental validity of psychological testing

and assessment: conceptual, methodological, and statistical issues.

Psychological Assessment, 15(4), 446-455.

Huotari, M., García-Horsman, J. A., Karayiorgou, M., Gogos, J. A., & Männistö, P.

T. (2004). D-amphetamine responses in catechol-O-methyltransferase

(COMT) disrupted mice. Psychopharmacology, 172(1), 1-10.

Hursh, S. R., Redmond, D. P., Johnson, M. L., Thorne, D. R., Belenky, G., Balkin, T.

J., . . . Eddy, D. R. (2004). Fatigue models for applied research in

warfighting. Aviation, Space, and Environmental Medicine, 75(3 Suppl),

A44.

Ikegami, K., Ogyu, S., Arakomo, Y., Suzuki, K., Mafune, K., Hiro, H., & Nagata, S.

(2009). Recovery of cognitive performance and fatigue after one night of

sleep deprivation. Journal of Occupational Health, 51(5), 412-422.


Ingre, M., Åkerstedt, T., Peters, B., Anund, A., & Kecklund, G. (2006). Subjective

sleepiness, simulated driving performance and blink duration: examining

individual differences. Journal of Sleep Research, 15(1), 47-53.

Ioannidis, J. P. A., Ntzani, E. E., Trikalinos, T. A., & Contopoulos-Ioannidis, D. G.

(2001). Replication validity of genetic association studies. Nature genetics,

29(3), 306-309.

Isler, R. B., Starkey, N. J., & Williamson, A. R. (2009). Video-based road

commentary training improves hazard perception of young drivers in a dual

task. Accident Analysis & Prevention, 41(3), 445-452.

Iwase, T., Kajimura, N., Uchiyama, M., Ebisawa, T., Yoshimura, K., Kamei, Y., . . .

Katoh, M. (2002). Mutation screening of the human Clock gene in circadian

rhythm sleep disorders. Psychiatry research, 109(2), 121-128.

Jackson, L., Chapman, P., & Crundall, D. (2009). What happens next? Predicting

other road users' behaviour as a function of driving experience and processing

time. Ergonomics, 52(2), 154-164.

Jackson, M. L., Gunzelmann, G., Whitney, P., Hinson, J. M., Belenky, G., Rabat, A.,

& Van Dongen, H. P. A. (2013). Deconstructing and reconstructing cognitive

performance in sleep deprivation. Sleep Medicine Reviews, 17(3), 215-225.

Jammes, B., Sharabty, H., & Esteve, D. (2008). Automatic EOG analysis: A first step

toward automatic drowsiness scoring during wake-sleep transitions.

Somnologie - Schlafforschung und Schlafmedizin, 12(3), 227-232.

Jean-Louis, G., von Gizycki, H., Zizi, F., Fookson, J., Spielman, A., Nunes, J., . . .

Taub, H. (1996). Determination of sleep and wakefulness with the actigraph

data analysis software (ADAS). Sleep, 19(9), 739-743.


Jewett, M. E., Wyatt, J. K., Ritz-De Cecco, A., Khalsa, S. B., Dijk, D. J., & Czeisler,

C. A. (1999). Time course of sleep inertia dissipation in human performance

and alertness. Journal of Sleep Research, 8(1), 1-8.

Johns, M. W. (1991). A new method for measuring daytime sleepiness: the Epworth

sleepiness scale. Sleep, 14(6), 540-545.

Johns, M. W. (1992). Reliability and factor analysis of the Epworth Sleepiness Scale.

Sleep, 15(4), 376-381.

Johns, M. W. (1993). Daytime sleepiness, snoring, and obstructive sleep apnea: The

Epworth Sleepiness Scale. Chest, 103(1), 30-37.

Johns, M. W. (1994). Sleepiness in different situations measured by the Epworth

Sleepiness Scale. Sleep, 17(8), 703-710.

Johns, M. W. (1998). Rethinking the assessment of sleepiness. Sleep Medicine

Reviews, 2(1), 3-15.

Johns, M. W. (2000). Sensitivity and specificity of the multiple sleep latency test

(MSLT), the maintenance of wakefulness test and the epworth sleepiness

scale: Failure of the MSLT as a gold standard. Journal of Sleep Research,

9(1), 5-11.

Johns, M. W. (2001). Assessing the drowsiness of drivers. Melbourne, Australia:

VicRoads

Johns, M. W. (2002). Sleep propensity varies with behaviour and the situation in

which it is measured: the concept of somnificity. Journal of Sleep Research,

11(1), 61-67.

Johns, M. W., Crowley, K., Chapman, R., Tucker, A., & Hocking, C. (2008). The

Test-Retest Reliability of an Ocular Measure of Drowsiness. Sleep, 31, A118.


Johns, M. W., & Hocking, B. (1997). Daytime sleepiness and sleep habits of

Australian workers. Sleep, 20(10), 844-849.

Johnson, R. F., Moore, R. Y., & Morin, L. P. (1988). Loss of entrainment and

anatomical plasticity after lesions of the hamster retinohypothalamic tract.

Brain Research, 460(2), 297-313.

Jones, K., Ellis, J., Von Schantz, M., Skene, D. J., Dijk, D.-J., & Archer, S. N.

(2007). Age-related change in the association between a polymorphism in the

PER3 gene and preferred timing of sleep and waking activities. Journal of

Sleep Research, 16(1), 12-16.

Jones, K., & Harrison, Y. (2001). Frontal lobe function, sleep loss and fragmented

sleep. Sleep Medicine Reviews, 5(6), 463-475.

Jung, T. P., Makeig, S., Humphries, C., Lee, T. W., McKeown, M. J., Iragui, V., &

Sejnowski, T. J. (2000). Removing electroencephalographic artifacts by blind

source separation. Psychophysiology, 37(2), 163-178.

Kafka, M. S., Marangos, P. J., & Moore, R. Y. (1985). Suprachiasmatic nucleus

ablation abolishes circadian rhythms in rat brain neurotransmitter receptors.

Brain Research, 327(1-2), 344-347.

Kahn-Greene, E. T., Lipizzi, E. L., Conrad, A. K., Kamimori, G. H., & Killgore, W.

D. S. (2006). Sleep deprivation adversely affects interpersonal responses to

frustration. Personality and Individual Differences, 41(8), 1433-1443.

Kaida, K., Åkerstedt, T., Kecklund, G., Nilsson, J. P., & Axelsson, J. (2007). The

effects of asking for verbal ratings of sleepiness on sleepiness and its masking

effects on performance. Clinical Neurophysiology, 118(6), 1324-1331.

Kaida, K., Takahashi, M., Åkerstedt, T., Nakata, A., Otsuka, Y., Haratani, T., &

Fukasawa, K. (2006). Validation of the Karolinska sleepiness scale against


performance and EEG variables. Clinical Neurophysiology, 117(7), 1574-

1581.

Kaplan, K. A., Itoi, A., & Dement, W. C. (2007). Awareness of sleepiness and ability

to predict sleep onset: Can drivers avoid falling asleep at the wheel? Sleep

Medicine, 9(1), 71-79.

Katila, A., Keskinen, E., Hatakka, M., & Laapotti, S. (2004). Does increased

confidence among novice drivers imply a decrease in safety?: The effects of

skid training on slippery road accidents. Accident Analysis & Prevention,

36(4), 543-550.

Kecklund, G., & Åkerstedt, T. (1993). Sleepiness in long distance truck driving: An

ambulatory EEG study of night driving. Ergonomics, 36(9), 1007-1017.

Kecklund, G., Anund, A., Wahlström, M. R., Philip, P., & Akerstedt, T. (2012).

Sleepiness and the risk of car crash : a case-control study. Journal of Sleep

Research, 21(S1), 307-307.

Killgore, W. D. S., Kahn-Greene, E. T., Lipizzi, E. L., Newman, R. A., Kamimori,

G. H., & Balkin, T. J. (2008). Sleep deprivation reduces perceived emotional

intelligence and constructive thinking skills. Sleep Medicine, 9(5), 517-526.

Kim, H.-I., Lee, H.-J., Cho, C.-H., Kang, S.-G., Yoon, H.-K., Park, Y.-M., . . . Kim,

L. (2015). Association of CLOCK, ARNTL, and NPAS2 gene

polymorphisms and seasonal variations in mood and behavior.

Chronobiology International, 32(6), 785-791.

Kleitman, N. (1963). Sleep and wakefulness: University of Chicago Press.

Klerman, E. B., & Dijk, D.-J. (2005). Interindividual variation in sleep duration and

its association with sleep debt in young adults. Sleep, 28(10).


Knipling, R. R., & Wang, J.-S. (1994). Crashes and fatalities related to driver

drowsiness/fatigue. Washington, DC: Office of Crash Avoidance Research,

United States Department of Transportation

Knutson, K. L., Van Cauter, E., Rathouz, P. J., DeLeire, T., & Lauderdale, D. S.

(2010). Trends in the prevalence of short sleepers in the USA: 1975-2006.

Sleep, 33(1), 37-45.

Kribbs, N. B., & Dinges, D. (1994). Vigilance decrement and sleepiness. In R. D.

Ogilvie & J. R. Harsh (Eds.), Sleep onset: Normal and abnormal processes.

(pp. 113-125). Washington, United States of America: American

Psychological Association.

Kryger, M. H., Roth, T., & Dement, W. C. (1994). Principles and practice of sleep

medicine. Philadelphia: W. B. Saunders Co.

Ktonas, P. Y., Osorio, P. L., & Everett, R. L. (1979). Automated detection of EEG

artifacts during sleep: Preprocessing for all-night spectral analysis.

Electroencephalography and Clinical Neurophysiology, 46(4), 382-388.

Kuna, S. T., Maislin, G., Pack, F. M., Staley, B., Hachadoorian, R., Coccaro, E. F., &

Pack, A. I. (2012). Heritability of performance deficit accumulation during

acute sleep deprivation in twins. Sleep, 35(9), 1223-1233.

Lachman, H. M., Papolos, D. F., Saito, T., Yu, Y.-M., Szumlanski, C. L., &

Weinshilboum, R. M. (1996). Human catechol-O-methyltransferase

pharmacogenetics: description of a functional polymorphism and its potential

application to neuropsychiatric disorders. Pharmacogenetics and Genomics,

6(3), 243-250.


Lal, S. K. L., & Craig, A. (2005). Reproducibility of the spectral components of the

electroencephalogram during driver fatigue. International Journal of


Lambie, J. A., & Marcel, A. J. (2002). Consciousness and the varieties of emotion

experience: a theoretical framework. Psychological review, 109(2), 219.

Lamond, N., Dawson, D., & Roach, G. D. (2005). Fatigue assessment in the field:

validation of a hand-held electronic psychomotor vigilance task. Aviation,

space, and environmental medicine, 76(5), 486-489.

Landolt, H.-P. (2008a). Genotype-dependent differences in sleep, vigilance, and

response to stimulants. Current Pharmaceutical Design, 14(32), 3396-3407.

Landolt, H.-P. (2008b). Sleep homeostasis: a role for adenosine in humans?

Biochemical Pharmacology, 75(11), 2070-2079.

Lane, R. D., Reiman, E. M., Axelrod, B., Yun, L.-S., Holmes, A., & Schwartz, G. E.

(1998). Neural correlates of levels of emotional awareness: Evidence of an

interaction between emotion and attention in the anterior cingulate cortex.

Journal of Cognitive Neuroscience, 10(4), 525-535.

Lara, T., Madrid, J. A., & Correa, Á. (2014). The Vigilance Decrement in Executive

Function Is Attenuated When Individual Chronotypes Perform at Their

Optimal Time of Day. PLoS One, 9(2), e88820.

Larue, G. S., Rakotonirainy, A., & Pettitt, A. N. (2011). Driving performance

impairments due to hypovigilance on monotonous roads. Accident Analysis &

Prevention, 43(6), 2037-2046.

Laufs, H., Kleinschmidt, A., Beyerle, A., Eger, E., Salek-Haddadi, A., Preibisch, C.,

& Krakow, K. (2003). EEG-correlated fMRI of human alpha activity.

Neuroimage, 19(4), 1463-1476.


Lavie, P., & Weler, B. (1989). Timing of naps: effects on post-nap sleepiness levels.


Lavine, R. A., Sibert, J. L., Gokturk, M., & Dickens, B. (2002). Eye-tracking

measures and human performance in a vigilance task. Aviation, space, and

environmental medicine, 73(4), 367-372.

Leary, E. (2007). Patient Preparation. In N. Butkov & T. Lee-Chiong (Eds.),

Fundamentals of Sleep Technology (pp. 241-252). Philadelphia: Lippincott

Williams and Wilkins.

Leproult, R., Colecchia, E. F., Berardi, A. M., Stickgold, R., Kosslyn, S. M., & Van

Cauter, E. (2003). Individual differences in subjective and objective alertness

during sleep deprivation are stable and unrelated. American Journal of

Physiology - Regulatory, Integrative and Comparative Physiology, 284(2),

280-290.

Lim, J., Choo, W.-C., & Chee, M. W. L. (2007). Reproducibility of changes in

behaviour and fMRI activation associated with sleep deprivation in a working

memory task. Sleep, 30(1), 61-70.

Lim, J., Ebstein, R., Tse, C. Y., Monakhov, M., Lai, P. S., Dinges, D. F., & Kwok,

K. (2012). Dopaminergic polymorphisms associated with time-on-task

declines and fatigue in the Psychomotor Vigilance Test. PLoS One, 7(3),

e33767.

Lim, J., Wu, W.-c., Wang, J., Detre, J. A., Dinges, D. F., & Rao, H. (2010). Imaging

brain fatigue from sustained mental workload: An ASL perfusion study of the

time-on-task effect. NeuroImage, 49(4), 3426-3435.


Linkowski, P., Kerkhofs, M., Hauspie, R., & Mendlewicz, J. (1991). Genetic

determinants of EEG sleep: a study in twins living apart.


Liu, J. J., Sudic Hukic, D., Forsell, Y., Schalling, M., Ösby, U., & Lavebratt, C.

(2015). Depression-associated ARNTL and PER2 genetic variants in

psychotic disorders. Chronobiology International, 32(4), 579-584.

Liu, Q., Zhou, R., Liu, L., & Zhao, X. (2015). Effects of 72 hours total sleep

deprivation on male astronauts' executive functions and emotion.

Comprehensive Psychiatry, 61, 28-35.

Lo, J. C., Groeger, J. A., Santhi, N., Arbon, E. L., Lazar, A. S., Hasan, S., . . . Dijk,

D. J. (2012). Effects of partial and acute total sleep deprivation on

performance across cognitive domains, individuals and circadian phase. PLoS

One, 7(9), e45987.

Loh, S., Lamond, N., Dorrian, J., Roach, G. D., & Dawson, D. (2004). The validity

of psychomotor vigilance tasks of less than 10-minute duration. Behavior

Research Methods, 36(2), 339-346.

Lowden, A., Anund, A., Kecklund, G., Peters, B., & Åkerstedt, T. (2009).

Wakefulness in young and elderly subjects driving at night in a car simulator.

Accident Analysis & Prevention, 41(5), 1001-1007.

Lowrey, P. L., & Takahashi, J. S. (2011). Genetics of circadian rhythms in

Mammalian model organisms. Advances In Genetics, 74, 175-230.

Lucidi, F., Russo, P. M., Mallia, L., Devoto, A., Lauriola, M., & Violani, C. (2006).

Sleep-related car crashes: Risk perception and decision-making processes in

young drivers. Accident Analysis & Prevention, 38(2), 302-309.


Luft, C. D. B., Takase, E., & Darby, D. (2009). Heart rate variability and cognitive

function: Effects of physical effort. Biological Psychology, 82(2), 196-201.

Macchi, M. M., Boulos, Z., Ranney, T., Simmons, L., & Campbell, S. S. (2002).

Effects of an afternoon nap on nighttime alertness and performance in long-

haul drivers. Accident Analysis & Prevention, 34(6), 825-834.

MacDonald, A. W., Cohen, J. D., Stenger, V. A., & Carter, C. S. (2000).

Dissociating the role of the dorsolateral prefrontal and anterior cingulate

cortex in cognitive control. Science, 288(5472), 1835-1838.

Malhotra, A. K., Kestler, L. J., Mazzanti, C., Bates, J. A., Goldberg, T., & Goldman,

D. (2014). A functional polymorphism in the COMT gene and performance

on a test of prefrontal cognition. American Journal of Psychiatry.

Mallis, M. M., Mejdal, S., Nguyen, T. T., & Dinges, D. F. (2004). Summary of the

key features of seven biomathematical models of human fatigue and

performance. Aviation, Space, and Environmental Medicine, 75(3 Suppl),

A4–A14.

Manly, T., & Robertson, I. H. (1997). Sustained attention and the frontal lobes. In P.

Rabbitt (Ed.), Methodology of frontal and executive function (pp. 135-153).

Hove: Psychology Press.

Markland, D., & Hardy, L. (1997). On the Factorial and Construct Validity of the

Intrinsic Motivation Inventory: Conceptual and Operational Concerns.

Research Quarterly for Exercise and Sport, 68(1), 20-32.

Marumoto, N., Murakami, N., Katayama, T., Kuroda, H., & Murakami, T. (1996).

Effects of daily injections of melatonin on locomotor activity rhythms in rats

maintained under constant bright or dim light. Physiology & Behavior, 60(3),

767-773.


Matthews, G., & Desmond, P. A. (2002). Task-induced fatigue states and simulated

driving performance. The Quarterly Journal of Experimental Psychology:

Section A, 55(2), 659-686.

Maycock, G. (1996). Sleepiness and driving: The experience of UK car drivers.


Mayhew, D. R., Simpson, H. M., Wood, K. M., Lonero, L., Clinton, K. M., &

Johnson, A. G. (2011). On-road and simulated driving: Concurrent and

discriminant validation. Journal of Safety Research, 42(4), 267-275.

Maywood, E. S., Smith, E., Hall, S. J., & Hastings, M. H. (1997). A Thalamic

Contribution to Arousal‐induced, Non‐photic Entrainment of the Circadian

Clock of the Syrian Hamster. European Journal of Neuroscience, 9(8), 1739-

1747.

McArdle, N., & Douglas, N. J. (2001). Effect of continuous positive airway pressure

on sleep architecture in the sleep apnea–hypopnea syndrome: a randomized

controlled trial. American Journal of Respiratory and Critical Care Medicine,

164(8), 1459-1463.

McAuley, E., Duncan, T., & Tammen, V. V. (1989). Psychometric properties of the

Intrinsic Motivation Inventory in a competitive sport setting: A confirmatory

factor analysis. Research Quarterly for Exercise and Dport, 60(1), 48-58.

McAuley, E., & Tammen, V. V. (1989). The effects of subjective and objective

competitive outcomes on intrinsic motivation. Journal of Sport and Exercise

Psychology, 11(1), 84-93.

McAuley, E., Wraith, S., & Duncan, T. E. (1991). Self‐Efficacy, Perceptions of

Success, and Intrinsic Motivation for Exercise1. Journal of Applied Social

Psychology, 21(2), 139-155.


McCartt, A. T., Rohrbaugh, J. W., Hammer, M. C., & Fuller, S. Z. (2000). Factors

associated with falling asleep at the wheel among long-distance truck drivers.


McGowan, A., & Banbury, S. (2004). Evaluating interruption-based techniques using

embedded measures of driver anticipation. In S. P. Banbury & S. Tremblay

(Eds.), A Cognitive Approach to Situation Awareness (pp. 176-192).

Hampshire, UK: Ashgate Publishing.

McKeganey, N. (2001). To pay or not to pay: respondents' motivation for

participating in research. Addiction, 96(9), 1237-1238.

McKenna, F. P., & Crick, J. L. (1991). Hazard perception in drivers: A methodology

for testing and training (Final Report). Crowthorne, UK: Transport Research

Laboratory

McKenna, F. P., & Crick, J. L. (1997). Developments in hazard perception (297).

UK: Transport Research Laboratory

McKenna, F. P., & Farrand, P. (1999). The role of automaticity in driving. In G. B.

Grayson (Ed.), Behavioural Research in Road Safety IX. Crowthorne, UK:

Transport Research Laboratory.

McKenna, F. P., & Horswill, M. S. (1999). Hazard perception and its relevance for

driver licensing. Journal of the International Association of Traffic and Safety

Sciences, 23(1), 26-41.

Melillo, P., Bracale, M., & Pecchia, L. (2011). Nonlinear Heart Rate Variability

features for real-life stress detection. Case study: students under stress due to

university examination. Biomedical Engineering Online, 10, 96-96.

Mignot, E., Young, T., Lin, L., & Finn, L. (1999). Nocturnal sleep and daytime

sleepiness in normal subjects with HLA-DQB1*0602. Sleep, 22(3), 347-352.


Miller, J. T., & McAuley, E. (1987). Effects of a goal-setting training program on

basketball free-throw self-efficacy and performance. The Sport Psychologist,

1(2), 103-113.

Mitchell, M. (1997). Death and injury on the road. In M. Mitchell (Ed.), The

aftermath of road accidents: Psychological, social and legal consequences of

an everyday trauma (pp. 3-14). London: Routledge.

Mitler, M. M., Gujavarty, K. S., & Browman, C. P. (1982). Maintenance of

Wakefulness test: A polysomnographic technique for evaluating treatment

efficacy in patients with excessive somnolence. Electroencephalography and

Clinical Neurophysiology, 53(6), 658-661.

Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager,

T. D. (2000). The unity and diversity of executive functions and their

contributions to complex "Frontal Lobe" tasks: a latent variable analysis.

Cognitive Psychology, 41(1), 49-100.

Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager,

T. D. (2000). The unity and diversity of executive functions and their

contributions to complex “frontal lobe” tasks: A latent variable analysis.

Cognitive psychology, 41(1), 49-100.

Möller-Levet, C. S., Archer, S. N., Bucca, G., Laing, E. E., Slak, A., Kabiljo, R., . . .

Dijk, D.-J. (2013). Effects of insufficient sleep on circadian rhythmicity and

expression amplitude of the human blood transcriptome. Proceedings of the

National Academy of Sciences, 110(12), E1132-E1141.

Moller, H. J., Kayumov, L., Bulmash, E. L., Nhan, J., & Shapiro, C. M. (2006).

Simulator performance, microsleep episodes, and subjective sleepiness:


Normative data using convergent methodologies to assess driver drowsiness.

Journal of Psychosomatic Research, 61(3), 335-342.

Monk, T. H., Buysse, D. J., Kennedy, K. S., Pods, J. M., DeGrazia, J. M., &

Miewald, J. M. (2003). Measuring sleep habits without using a diary: The

sleep timing questionnaire. Sleep, 26(2), 208-212.

Moore-Ede, M., Heitmann, A., Guttkuhn, R., Trutschel, U., Aguirre, A., & Croke, D.

(2004). Circadian alertness simulator for fatigue risk assessment in

transportation: application to reduce frequency and severity of truck

accidents. Aviation, Space, and Environmental Medicine, 75(3 Suppl), A107-

A118.

Moore, R. Y., & Eichler, V. B. (1972). Loss of a circadian adrenal corticosterone

rhythm following suprachiasmatic lesions in the rat. Brain Research, 42(1),

201-206.

Mu, Q., Mishory, A., Johnson, K. A., Nahas, Z., Kozel, F. A., Yamanaka, K., . . .

George, M. S. (2005). Decreased brain activation during a working memory

task at rested baseline is associated with vulnerability to sleep deprivation.

Sleep, 28(4), 433-446.

Mukherjee, S., Yadav, R., Yung, I., Zajdel, D. P., & Oken, B. S. (2011). Sensitivity

to mental effort and test–retest reliability of heart rate variability measures in

healthy seniors. Clinical Neurophysiology, 122(10), 2059-2066.

Murphy, T. I., Richard, M., Masaki, H., & Segalowitz, S. J. (2006). The effect of

sleepiness on performance monitoring: I know what I am doing, but do I

care? Journal of Sleep Research, 15(1), 15-21.


Muzet, A., Nicolas, A., Tassi, P., Dewasmes, G., & Bonneau, A. (1995).

Implementation of napping in industry and the problem of sleep inertia.

Journal of Sleep Research, 4(s2), 67-69.

Nabi, H., Guéguen, A., Chiron, M., Lafont, S., Zins, M., & Lagarde, E. (2006).

Awareness of driving while sleepy and road traffic accidents: prospective

study in GAZEL cohort. British Medical Journal, 333(7558), 75-75.

National Sleep Foundation. (2007). 2007 Sleep in America Poll: Women and Sleep -

Summary of Findings. Washington: National Sleep Foundation.

National Sleep Foundation. (2008). 2008 Sleep in America Poll: Sleep, Performance,

and the Workplace - Summary of Findings. Washington: National Sleep

Foundation.

National Sleep Foundation. (2015). 2015 Sleep in America Poll: Sleep and Pain.

Sleep Health, 1(2), e14-e375.

Neugebauer, R., & Ng, S. (1990). Differential recall as a source of bias in

epidemiologic research. Journal of Clinical Epidemiology, 43(12), 1337-

1341.

Nilsson, J. P., Söderström, M., Karlsson, A. U., Lekander, M., Åkerstedt, T.,

Lindroth, N. E., & Axelsson, J. (2005). Less effective executive functioning

after one night's sleep deprivation. Journal of Sleep Research, 14(1), 1-6.

Nilsson, T., Nelson, T. M., & Carlson, D. (1997). Development of fatigue symptoms

during simulated driving. Accident Analysis & Prevention, 29, 479-488.

Nordbakke, S., & Sagberg, F. (2007). Sleepy at the wheel: Knowledge, symptoms

and behaviour among car drivers. Transportation Research Part F: Traffic



Numata, N., Kitajima, H., Goi, Y., & Yamamoto, K. (1998). Analysis of drivers'

behavior before and after crashes in simulated expressway driving to predict

sleepiness levels for doze alarm activation. JSAE Review, 19(3), 269-275.

Nussinovitch, U., Cohen, O., Kaminer, K., Ilani, J., & Nussinovitch, N. (2012).

Evaluating reliability of ultra-short ECG indices of heart rate variability in

diabetes mellitus patients. Journal of Diabetes and its Complications, 26(5),

450-453.

Oakley, N. R. (1997). Validation with polysomnography of the Sleepwatch

sleep/wake scoring algorithm used by the Actiwatch activity monitoring

system. Oregon, USA: Mini-Mitter Co., Inc.

Odle-Dusseau, H. N., Bradley, J. L., & Pilcher, J. J. (2010). Subjective perceptions of

the effects of sustained performance under sleep-deprivation conditions.

Chronobiology International, 27(2), 318-333.

Ogilvie, R. D., & Wilkinson, R. T. (1984). The detection of sleep onset: Behavioral

and physiological convergence. Psychophysiology, 21(5), 510-520.

Ohayon, M. M., Carskadon, M. A., Guilleminault, C., & Vitiello, M. V. (2004).

Meta-Analysis of Quantitative Sleep Parameters From Childhood to Old Age

in Healthy Individuals: Developing Normative Sleep Values Across the

Human Lifespan. Sleep, 27(7), 1255-1273.

Olofsen, E., Dinges, D. F., & Van Dongen, H. P. A. (2004). Nonlinear mixed-effects

modeling: Individualization and prediction. Aviation, Space, and

Environmental Medicine, 75, A134-A140.

Oron-Gilad, T., Ronen, A., & Shinar, D. (2008). Alertness maintaining tasks (AMTs)

while driving. Accident Analysis & Prevention, 40(3), 851-860.


Otmani, S., Pebayle, T., Roge, J., & Muzet, A. (2005). Effect of driving duration and

partial sleep deprivation on subsequent alertness and performance of car

drivers. Physiology & Behavior, 84, 715-724.

Pack, A. I., Pack, A. M., Rodgman, E., Cucchiara, A., Dinges, D. F., & Schwab, C.

W. (1995). Characteristics of crashes attributed to the driver having fallen

asleep. Accident Analysis & Prevention, 27(6), 769-775.

Pandi-Perumal, S. R., Smits, M., Spence, W., Srinivasan, V., Cardinali, D. P., Lowe,

A. D., & Kayumov, L. (2007). Dim light melatonin onset (DLMO): A tool for

the analysis of circadian phase in human sleep and chronobiological

disorders. Progress in Neuro-Psychopharmacology and Biological

Psychiatry, 31(1), 1-11.

Park, J.-H., Wacholder, S., Gail, M. H., Peters, U., Jacobs, K. B., Chanock, S. J., &

Chatterjee, N. (2010). Estimation of effect size distribution from genome-

wide association studies and implications for future discoveries. Nature

genetics, 42(7), 570-575.

Parsons, M. (1986). Fits and other causes of loss of consciousness while driving. The

Quarterly Journal Of Medicine, 58(227), 295-303.

Pedrazzoli, M., Ling, L., Finn, L., Kubin, L., Young, T., Katzenberg, D., & Mignot,

E. (1999). A polymorphism in the human timeless gene is not associated with

diurnal preferences in normal adults. Sleep research online: SRO, 3(2), 73-

76.

Pelz, D. C., & Krupat, E. (1974). Caution profile and driving record of undergraduate

males. Accident Analysis & Prevention, 6(1), 45-58.


Pennay, D. (2008). Community Attitudes to Road Safety: 2008 survey report

(INFRA-08366). Canberra, Australia: Australian Government Department of

Infrastructure, Transport, Regional Development and Local Government

Philip, P., Ghorayeb, I., Stoohs, R., Menny, J. C., Dabadie, P., Bioulac, B., &

Guilleminault, C. (1996). Determinants of sleepiness in automobile drivers.

Journal of Psychosomatic Research, 41(3), 279-288.

Philip, P., Sagaspe, P., Taillard, J., Valtat, C., Moore, N., Åkerstedt, T., . . . Bioulac,

B. (2005). Fatigue, sleepiness, and performance in simulated versus real

driving conditions. Sleep, 28(12), 1511-1516.

Philip, P., Taillard, J., Quera-Salva, M. A., Bioulac, B., & Åkerstedt, T. (1999).

Simple reaction time, duration of driving and sleep deprivation in young

versus old automobile drivers. Journal of Sleep Research, 8(1), 9-14.

Philip, P., Taillard, J., Sagaspe, P., Valtat, C., Sanchez-Ortuno, M., Moore, N., . . .

Bioulac, B. (2004). Age, performance and sleep deprivation. Journal of Sleep

Research, 13(2), 105-110.

Piercy, J., & Lack, L. (1988). Daily exercise can shift the endogenous circadian

phase. Sleep Res, 17(393), b42.

Pilcher, J. J., & Huffcutt, A. I. (1996). Effects of sleep deprivation on performance:

A meta analysis. Sleep, 19, 318-326.

Pilcher, J. J., & Walters, A. S. (1997). How Sleep Deprivation Affects Psychological

Variables Related to College Students' Cognitive Performance. Journal of

American College Health, 46(3), 121-126.

Pizza, F., Contardi, S., Ferlisi, M., Mondini, S., & Cirignotta, F. (2008). Daytime

driving simulation performance and sleepiness in obstructive sleep apnoea

patients. Accident Analysis & Prevention, 40(2), 602-609.


Pollatsek, A., Narayanaan, V., Pradhan, A., & Fisher, D. L. (2006). Using eye

movements to evaluate a PC-based risk awareness and perception training

program on a driving simulator. Human Factors, 48(3), 447-464.

Porges, S. W., & Bohrer, R. E. (1990). The analysis of periodic processes in

psychophysiological research. In J. T. Cacioppo, L. G. Tassinary, J. T.

Cacioppo, & L. G. Tassinary (Eds.), Principles of psychophysiology:

Physical, social, and inferential elements. (pp. 708-753). New York, NY, US:


Porkka-Heiskanen, T., & Kalinchuk, A. V. (2011). Adenosine, energy metabolism

and sleep homeostasis. Sleep Medicine Reviews, 15(2), 123-135.

Posner, M. I., Nissen, M. J., & Ogden, W. C. (1978). Attended and unattended

processing modes: The role of set for spatial location. Modes of perceiving

and processing information, 137, 158.

Poulsen, A. A., Horswill, M. S., Wetton, M. A., Hill, A., & Lim, S. M. (2010). A

brief office-based hazard perception intervention for drivers with ADHD

symptoms. Australian and New Zealand Journal of Psychiatry, 44(6), 528-

534.

Preece, M. H. W., Horswill, M. S., & Geffen, G. M. (2011). Assessment of drivers'

ability to anticipate traffic hazards after traumatic brain injury. Journal of

Neurology, Neurosurgery & Psychiatry, 82(4), 447-451.

Purnell, M. T., Feyer, A. M., & Herbison, G. P. (2002). The impact of a nap

opportunity during the night shift on the performance and alertness of 12-h

shift workers. Journal of Sleep Research, 11(3), 219-227.


Quimby, A. R., Maycock, G., Carter, I. D., Dixon, R., & Wall, J. G. (1986).

Perceptual abilities of accident involved drivers (27). Berkshire, England:

Transport and Road Research Laboratory

Radun, I., Ohisalo, J., Radun, J., & Kecklund, G. (2011). Night Work, Fatigued

Driving and Traffic Law: The Case of Police Officers. Industrial Health,

49(3), 389-392.

Radun, I., Ohisalo, J., Radun, J., & Rajalin, S. (2012). Law defining the critical level

of driver fatigue in terms of hours without sleep: Criminal justice

professionals' opinions and fatal accident data. International Journal of Law,

Crime and Justice, 40(3), 172-178.

Radun, I., Ohisalo, J., Radun, J., Wahde, M., & Kecklund, G. (2013). Driver fatigue

and the law from the perspective of police officers and prosecutors.

Transportation Research Part F: Traffic Psychology and Behaviour, 18, 159-

167.

Radun, I., Radun, J. E., Wahde, M., Watling, C. N., & Kecklund, G. (2015). Self-

reported circumstances and consequences of driving while sleepy.

Transportation Research Part F: Traffic Psychology and Behaviour, 32, 91-

100.

Radun, I., & Summala, H. (2004). Sleep-related fatal vehicle accidents:

characteristics of decisions made by multidisciplinary investigation teams.

Sleep, 27(2), 224-227.

Rajaratnam, S. M. W., & Jones, C. B. (2004). Lessons about sleepiness and driving

from the Selby rail disaster case: R v Gary Neil Hart. Chronobiology



Rajendra Acharya, U., Kannathal, N., Mei Hua, L., & Mei Yi, L. (2005). Study of

heart rate variability signals at sitting and lying postures. Journal of

Bodywork and Movement Therapies, 9(2), 134-141.

Rechtschaffen, A., & Kales, A. (1968). A manual of standardized terminology,

technique and scoring system for sleep stages of human subjects. Washington

DC: Public Health Service, US Government Printing Office

Reppert, S. M., & Weaver, D. R. (2002). Coordination of circadian timing in

mammals. Nature, 418(6901), 935-941.

Rétey, J., Adam, M., Honegger, E., Khatami, R., Luhmann, U., Jung, H., . . . Landolt,

H.-P. (2005). A functional genetic variation of adenosine deaminase affects

the duration and intensity of deep sleep in humans. Proceedings of the

National Academy of Sciences, 102(43), 15676-15681.

Rétey, J., Adam, M., Khatami, R., Luhmann, U., Jung, H., Berger, W., & Landolt,

H.-P. (2007). A genetic variation in the adenosine A2A receptor gene

(ADORA2A) contributes to individual sensitivity to caffeine effects on sleep.

Clinical Pharmacology & Therapeutics, 81(5), 692-698.

Reyner, L. A., & Horne, J. A. (1998). Falling asleep whilst driving: Are drivers

aware of prior sleepiness? International Journal of Legal Medicine, 111(3),

120-123.

Reyner, L. A., & Horne, J. A. (2000). Early morning driver sleepiness: effectiveness

of 200 mg caffeine. Psychophysiology, 37(02), 251-256.

Richardson, G. S., Carskadon, M. A., Orav, E. J., & Dement, W. C. (1982).

Circadian variation of sleep tendency in elderly and young adult subjects.

Sleep, 5(S), S82-94.


Ripley, E. B. D. (2006). A Review of Paying Research Participants: It's Time to

Move Beyond the ethical Debate. Journal of Empirical Research on Human

Research Ethics, 1(4), 9-20.

Robert, G., & Hockey, J. (1997). Compensatory control in the regulation of human

performance under stress and high workload: A cognitive-energetical

framework. Biological Psychology, 45(1–3), 73-93.

Roehrs, T., Carskadon, M. A., Dement, W. C., & Roth, T. (2005). Daytime

Sleepiness and Alertness. In W. H. Kryger, T. Roth, & W. C. Dement (Eds.),

Principles and practice of sleep medicine (pp. 42-53). Philadelphia:

Elsevier/Saunders.

Rogé, J., Pébayle, T., Hannachi, S. E., & Muzet, A. (2003). Effect of sleep

deprivation and driving duration on the useful visual field in younger and

older subjects during simulator driving. Vision Research, 43(13), 1465-1472.

Rogé, J., Pebayle, T., & Muzet, A. (2001). Variations of the level of vigilance and of

behavioural activities during simulated automobile driving. Accident Analysis

& Prevention, 33, 181-186.

Roma, P. G., Hursh, S. R., Mead, A. M., & Nesthus, T. E. (2012). Flight Attendant

Work/Rest Patterns, Alertness, and Performance Assessment: Field

Validation of Biomathematical Fatigue Modeling (DOT/FAA/AM-12/12).

Washington, DC: Office of Aerospace Medicine

Rossa, K. R., Smith, S. S., Allan, A. C., & Sullivan, K. A. (2014). The effects of

sleep restriction on executive inhibitory control and affect in young adults.

Journal of Adolescent Health, 55(2), 287-292.

Roth, T., Roehrs, T. A., Carskadon, M. A., & Dement, W. C. (2005). Daytime

Sleepiness and Alertness. In W. H. Kryger, T. Roth, & W. C. Dement (Eds.),


Principles and practice of sleep medicine (pp. 40-49). Philadelphia:

Elsevier/Saunders.

Rupp, T. L., Wesensten, N. J., & Balkin, T. J. (2012). Trait-like vulnerability to total

and partial sleep loss. Sleep, 35(8), 1163-1172.

Rupp, T. L., Wesensten, N. J., Newman, R., & Balkin, T. J. (2013). PER3 and

ADORA2A polymorphisms impact neurobehavioral performance during

sleep restriction. Journal of Sleep Research, 22(2), 160-165.

Ryan, R. M. (1982). Control and information in the intrapersonal sphere: An

extension of cognitive evaluation theory. Journal of Personality and Social

Psychology, 43(3), 450-461.

Ryan, R. M., Plant, R. W., & O'Malley, S. (1995). Initial motivations for alcohol

treatment: Relations with patient characteristics, treatment involvement, and

dropout. Addictive behaviors, 20(3), 279-297.

Rylander-Rudqvist, T., Håkansson, N., Tybring, G., & Wolk, A. (2006). Quality and

Quantity of Saliva DNA Obtained from the Self-administrated Oragene

Method—A Pilot Study on the Cohort of Swedish Men. Cancer

Epidemiology Biomarkers & Prevention, 15(9), 1742-1745.

Sadeh, A., Keinan, G., & Daon, K. (2004). Effects of stress on sleep: the moderating

role of coping style. Health Psychology, 23(5), 542-545.

Sadeh, A., Sharkey, K. M., & Carskadon, M. A. (1994). Activity-based sleep-wake

identification: an empirical test of methodological issues. Sleep, 17(3), 201-

207.

Saksvik, I. B., Bjorvatn, B., Hetland, H., Sandal, G. M., & Pallesen, S. (2011).

Individual differences in tolerance to shift work – A systematic review. Sleep

Medicine Reviews, 15(4), 221-235.


Salanti, G., Amountza, G., Ntzani, E. E., & Ioannidis, J. P. A. (2005). Hardy-

Weinberg equilibrium in genetic association studies: an empirical evaluation

of reporting, deviations, and power. European Journal of Human Genetics,

13(7), 840-848.

Sallinen, M., Härmä, M., Åkerstedt, T., Rosa, R., & Lillqvist, O. (1998). Promoting

alertness with a short nap during a night shift. Journal of Sleep Research,

7(4), 240-247.

Sallinen, M., Holm, A., Hiltunen, J., Hirvonen, K., Härmä, M., Koskelo, J., . . .

Müller, K. (2008). Recovery of cognitive performance from sleep debt: Do a

short rest pause and a single recovery night help? Chronobiology


Sallinen, M., Onninen, J., Tirkkonen, K., Haavisto, M.-L., Härmä, M., Kubo, T., . . .

Porkka-Heiskanen, T. (2013). Effects of cumulative sleep restriction on self-

perceptions while multitasking. Journal of Sleep Research, 22(3), 273-281.

Sandberg, D., Anund, A., Fors, C., Kecklund, G., Karlsson, J. G., Wahde, M., &

Akerstedt, T. (2011). The Characteristics of Sleepiness During Real Driving

at Night-A Study of Driving Performance, Physiology and Subjective

Experience. Sleep, 34(10), 1317-1325.

Sanei, S., & Chambers, J. A. (2008). EEG Signal Processing. Chichester: John Wiley

& Sons, Ltd.

Saper, C. B., Chou, T. C., & Scammell, T. E. (2001). The sleep switch: hypothalamic

control of sleep and wakefulness. Trends in Neurosciences, 24(12), 726-731.

Saper, C. B., Fuller, P. M., Pedersen, N. P., Lu, J., & Scammell, T. E. (2010). Sleep

State Switching. Neuron, 68(6), 1023-1042.


Saper, C. B., Scammell, T. E., & Lu, J. (2005). Hypothalamic regulation of sleep and

circadian rhythms. Nature, 437(7063), 1257.

Scheffers, M. K., Coles, M. G., Bernstein, P., Gehring, W. J., & Donchin, E. (1996).

Event-related brain potentials and error-related processing: an analysis of

incorrect responses to go and no-go stimuli. Psychophysiology, 33(1), 42-53.

Schleicher, R., Galley, N., Briest, S., & Galley, L. (2008). Blinks and saccades as

indicators of fatigue in sleepiness warnings: looking tired? Ergonomics,

51(7), 982-1010.

Schlögl, A., Keinrath, C., Zimmermann, D., Scherer, R., Leeb, R., & Pfurtscheller,

G. (2007). A fully automated correction method of EOG artifacts in EEG

recordings. Clinical neurophysiology, 118(1), 98-104.

Schmidt, C., Collette, F., Cajochen, C., & Peigneux, P. (2007). A time to think:

Circadian rhythms in human cognition. Cognitive Neuropsychology, 24(7),

755-789.

Scholey, A. B., Laing, S., & Kennedy, D. O. (2006). Blood glucose changes and

memory: Effects of manipulating emotionality and mental effort. Biological

Psychology, 71(1), 12-19.

Schroeder, E. B., Whitsel, E. A., Evans, G. W., Prineas, R. J., Chambless, L. E., &

Heiss, G. (2004). Repeatability of heart rate variability measures. Journal of

Electrocardiology, 37(3), 163-172.

Sheppard, E., Ropar, D., Underwood, G., & Loon, E. (2009). Brief Report: Driving

Hazard Perception in Autism. Journal of Autism and Developmental

Disorders, 40(4), 504-508.

Shimada, M., Miyagawa, T., Kawashima, M., Tanaka, S., Honda, Y., Honda, M., &

Tokunaga, K. (2010). An approach based on a genome-wide association


study reveals candidate loci for narcolepsy. Human Genetics, 128(4), 433-

441.

Shinar, D. (1978). Psychology on the road: The human factor in traffic safety. New

York: John Wiley & Sons.

Shinar, D. (2007). Traffic safety and human behavior. London: Elsevier Science.

Short, M. A., & Louca, M. (2015). Sleep deprivation leads to mood deficits in

healthy adolescents. Sleep Medicine, 16(8), 987-993.

Shreter, R., Peled, R., & Pillar, G. (2006). The 20-min trial of the Maintenance of

Wakefulness Test is profoundly affected by motivation. Sleep and Breathing,

10(4), 173-179.

Simon, M., Schmidt, E. A., Kincses, W. E., Fritzsche, M., Bruns, A., Aufmuth, C., . .

. Schrauf, M. (2011). EEG alpha spindle measures as indicators of driver

fatigue under real traffic conditions. Clinical Neurophysiology, 122(6), 1168-

1178.

Simons, D. J., & Ambinder, M. S. (2005). Change blindness theory and

consequences. Current Directions in Psychological Science, 14(1), 44-48.

Skouteris, H., Wertheim, E. H., Germano, C., Paxton, S. J., & Milgrom, J. (2009).

Assessing sleep during pregnancy: A study across two time points examining

the Pittsburgh sleep quality index and associations with depressive

symptoms. Women's Health Issues, 19(1), 45-51.

Smith, C. S., Reilly, C., & Midkiff, K. (1989). Evaluations of three circadian rhythm

questionnaires with suggestions for an improved measure of morningess.

Journal of Applied Psychology, 74, 728-738.


Smith, M. E., McEvoy, L. K., & Gevins, A. (2002). The impact of moderate sleep

loss on neurophysiologic signals during working-memory task performance.

Sleep, 25(7), 784-794.

Smith, R. D., Jan, S., & Shiell, A. (1993). Can we assess what Australians are

willing to pay for road safety?: A report to the federal office of road safety,

Department of Transport and Communications. Canberra: Australian

Transport Safety Bureau

Smith, S. S., Carrington, M., & Trinder, J. (2005). Subjective and predicted

sleepiness while driving in young adults. Accident Analysis & Prevention,

37(6), 1066-1073.

Smith, S. S., Horswill, M. S., Chambers, B., & Wetton, M. (2009). Hazard

perception in novice and experienced drivers: the effects of sleepiness.


Smith, S. S., Kilby, S., Jorgensen, G., & Douglas, J. A. (2007). Napping and

nightshift work: Effects of a short nap on psychomotor vigilance and

subjective sleepiness in health workers. Sleep and Biological Rhythms, 5(2),

117-125.

Smith, S. S., & Trinder, J. (2005). Morning sunlight can phase advance the circadian

rhythm of young adults. Sleep and Biological Rhythms, 3(1), 39-41.

Smith, S. S., Watling, C. N., Horswill, M. S., & Douglas, J. A. (2011). The effects of

rest breaks on driver fatigue. New South Wales: NRMA-ACT Road Safety

Trust and the Roads and Traffic Authority

Soehner, A., Kennedy, K., & Monk, T. (2007). Personality Correlates with Sleep-

Wake Variables. Chronobiology International, 24(5), 889-903.


Spiers, H. J., & Maguire, E. A. (2007). Neural substrates of driving behaviour.

NeuroImage, 36(1), 245-255.

Stein, M. B., Chartier, M., & Walker, J. R. (1993). Sleep in nondepressed patients

with panic disorder: I. Systematic assessment of subjective sleep quality and

sleep disturbance. Sleep, 16(8), 724-726.

Stein, M. B., Kroft, C. D. L., & Walker, J. R. (1993). Sleep impairment in patients

with social phobia. Psychiatry Research, 49(3), 251-256.

Stern, J. A., Boyer, D., & Schroeder, D. (1994). Blink rate: A possible measure of

fatigue. Human Factors, 36(2), 285-297.

Steyvers, F. J. (1987). The influence of sleep deprivation and knowledge of results

on perceptual encoding. Acta Psychologica, 66(2), 173-187.

Steyvers, F. J., & Gaillard, A. W. (1993). The effects of sleep deprivation and

incentives on human performance. Psychological Research, 55(1), 64-70.

Strijkstra, A. M., Beersma, D. G. M., Drayer, B., Halbesma, N., & Daan, S. (2003).

Subjective sleepiness correlates negatively with global alpha (8-12 Hz) and

positively with central frontal theta (4-8 Hz) frequencies in the human resting

awake electroencephalogram. Neuroscience Letters, 340(1), 17-20.

Stutts, J. C., Wilkins, J. W., Scott Osberg, J., & Vaughn, B. V. (2003). Driver risk

factors for sleep-related crashes. Accident Analysis & Prevention, 35(3), 321-

331.

Szymusiak, R., Gvilia, I., & McGinty, D. (2007). Hypothalamic control of sleep.


Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.).

Boston, MA: Allyn & Bacon.


Task Force of the European Society of Cardiology and the North American Society

of Pacing and Electrophysiology. (1996). Heart rate variability. Standards of

measurement, physiological interpretation, and clinical use. European Heart

Journal, 17(3), 354-381.

Tassi, P., & Muzet, A. Sleep inertia. Sleep Medicine Reviews, 4(4), 341-353.

Teixeira, A. R., Tomé, A. M., Lang, E. W., Gruber, P., & Martins da Silva, A.

(2006). Automatic removal of high-amplitude artefacts from single-channel

electroencephalograms. Computer Methods and Programs in Biomedicine,

83(2), 125-138.

Thiffault, P., & Bergeron, J. (2003). Monotony of road environment and driver

fatigue: A simulator study. Accident Analysis & Prevention, 35(3), 381-391.

Thomas, M. L., Sing, H. C., Belenky, G., Holcomb, H. H., Mayberg, H. S., Dannals,

R. F., . . . Redmond, D. P. (2003). Neural basis of alertness and cognitive

performance impairments during sleepiness: II. Effects of 48 and 72 h of

sleep deprivation on waking human regional brain activity. Thalamus &

Related Systems, 2(3), 199-229.

Thomas, M. L., Sing, H. C., Belenky, G., Holcomb, H. H., Mayberg, H. S., Dannals,

R. F., . . . Redmond, D. (2000). Neural basis of alertness and cognitive

performance impairments during sleepiness. I. Effects of 24 h of sleep

deprivation on waking human regional brain activity. Journal of Sleep

Research, 9(4), 335-352.

Ticho, S. R., & Radulovacki, M. (1991). Role of adenosine in sleep and temperature

regulation in the preoptic area of rats. Pharmacology Biochemistry and

Behavior, 40(1), 33-40.


Toivonen, L., Helenius, K., & Viitasalo, M. (1997). Electrocardiographic

repolarization during stress from awakening on alarm call. Journal of the

American College of Cardiology, 30(3), 774-779.

Törnros, J. (1998). Driving behaviour in a real and a simulated road tunnel--a

validation study. Accident Analysis & Prevention, 30(4), 497-503.

Torsvall, L., & Åkerstedt, T. (1987). Sleepiness on the job: Continuously measured

EEG changes in train drivers. Electroencephalography and Clinical


Tran, Y., Wijesuriya, N., Tarvainen, M., Karjalainen, P., & Craig, A. (2009). The

Relationship Between Spectral Changes in Heart Rate Variability and

Fatigue. Journal of Psychophysiology, 23(3), 143-151.

Treffner, P., Barrett, R., & Petersen, A. (2002). Stability and skill in driving. Human

Movement Science, 21(5-6), 749-784.

Treisman, A. M., & Gelade, G. (1980). A feature-integration theory of attention.

Cognitive Psychology, 12(1), 97-136.

Trikalinos, T. A., Salanti, G., Khoury, M. J., & Ioannidis, J. P. A. (2006). Impact of

Violations and Deviations in Hardy-Weinberg Equilibrium on Postulated

Gene-Disease Associations. American Journal of Epidemiology, 163(4), 300-

309.

Tryon, W. W. (2004). Issues of validity in actigraphic sleep assessment. Sleep, 27(1),

158-165.

Tsigilis, N., & Theodosiou, A. (2003). Temporal stability of the intrinsic motivation

inventory. Perceptual and Motor Skills, 97(1), 271-280.


Tucker, A. M., Whitney, P., Belenky, G., Hinson, J. M., & Van Dongen, H. P. A.

(2010). Effects of sleep deprivation on dissociated components of executive

functioning. Sleep, 33(1), 47-57.

Tunbridge, E., Bannerman, D., Sharp, T., & Harrison, P. (2004). Catechol-o-

methyltransferase inhibition improves set-shifting performance and elevates

stimulated dopamine release in the rat prefrontal cortex. The Journal of

Neuroscience, 24(23), 5331-5335.

Tzschentke, T. M. (2001). Pharmacology and behavioral pharmacology of the

mesocortical dopamine system. Progress in Neurobiology, 63(3), 241-320.

Uchiyama, Y., Ebe, K., Kozato, A., Okada, T., & Sadato, N. (2003). The neural

substrates of driving at a safe distance: A functional MRI study.

Neuroscience Letters, 352(3), 199-202.

Underwood, G., Crundall, D., & Chapman, P. (2002). Selective searching while

driving: The role of experience in hazard detection and general surveillance.

Ergonomics, 45(1), 1-12.

Underwood, G., Phelps, N., Wright, C., van Loon, E., & Galpin, A. (2005). Eye

fixation scanpaths of younger and older drivers in a hazard perception task.

Ophthalmic and Physiological Optics, 25(4), 346-356.

Utge, S. J., Soronen, P., Loukola, A., Kronholm, E., Ollila, H. M., Pirkola, S., . . .

Paunio, T. (2010). Systematic analysis of circadian genes in a population-

based sample reveals association of TIMELESS with depression and sleep

disturbance. PLoS One, 5(2), e9259.

van den Berg, J., Neely, G., Nilsson, L., Knutsson, A., & Landström, U. (2005).

Electroencephalography and subjective ratings of sleep deprivation. Sleep

Medicine, 6(3), 231-240.


Van Dongen, H. P., Baynard, M. D., Maislin, G., & Dinges, D. F. (2004). Systematic

interindividual differences in neurobehavioral impairment from sleep loss:

Evidence of trait-like differential vulnerability. Sleep, 27(3), 423-433.

Van Dongen, H. P., Maislin, G., Mullington, J. M., & Dinges, D. F. (2003). The

cumulative cost of additional wakefulness: dose-response effects on

neurobehavioral functions and sleep physiology from chronic sleep restriction

and total sleep deprivation. Sleep, 26(2), 117-126.

Van Dongen, H. P., Mott, C. G., Huang, J. K., Mollicone, D. J., McKenzie, F. D., &

Dinges, D. F. (2007). Optimization of biomathematical model predictions for

cognitive performance impairment in individuals: Accounting for unknown

traits and uncertain states in homeostatic and circadian processes. Sleep,

30(9), 1129-1143.

Van Dongen, H. P., Rogers, N. L., & Dinges, D. F. (2003). Sleep debt: Theoretical

and empirical issues. Sleep and Biological Rhythms, 1(1), 5-13.

van Loon, R. J., Brouwer, R. F. T., & Martens, M. H. (2015). Drowsy drivers’ under-

performance in lateral control: How much is too much? Using an integrated

measure of lateral control to quantify safe lateral driving. Accident Analysis

& Prevention, 84, 134-143.

Vandewalle, G., Archer, S. N., Wuillaume, C., Balteau, E., Degueldre, C., Luxen, A.,

. . . Dijk, D.-J. (2009). Functional magnetic resonance imaging-assessed brain

responses during an executive task depend on interaction of sleep

homeostasis, circadian phase, and PER3 genotype. The Journal of

Neuroscience, 29(25), 7948-7956.


Vanlaar, W., Simpson, H., Mayhew, D., & Robertson, R. (2008). Fatigued and

drowsy driving: A survey of attitudes, opinions and behaviors. Journal of

Safety Research, 39, 303-309.

Vassallo, S., Smart, D., Sanson, A., Harrison, W., Harris, A., Cockfield, S., &

McIntyre, A. (2007). Risky driving among young Australian drivers: Trends,

precursors and correlates. Accident Analysis & Prevention, 39(3), 444-458.

Verwey, W. B., & Zaidel, D. M. (1999). Preventing drowsiness accidents by an

alertness maintenance device. Accident Analysis & Prevention, 31(3), 199-

211.

Vignatelli, L., Plazzi, G., Barbato, A., Ferini-Strambi, L., Manni, R., Pompei, F., &

D'Alessandro, R. (2003). Italian version of the Epworth sleepiness scale:

external validity. Neurological Sciences, 23(6), 295-300.

Viltrop, T., Krjutškov, K., Palta, P., & Metspalu, A. (2010). Comparison of DNA

extraction methods for multiplex polymerase chain reaction. Analytical

Biochemistry, 398(2), 260-262.

Viola, A. U., Archer, S. N., James, Lynette M., Groeger, J. A., Lo, J. C. Y., Skene,

D. J., . . . Dijk, D.-J. (2007). PER3 Polymorphism Predicts Sleep Structure

and Waking Performance. Current Biology, 17(7), 613-618.

Voisey, J., Swagell, C. D., Hughes, I. P., Connor, J. P., Lawford, B. R., Young, R.

M., & Morris, C. P. (2010). A polymorphism in the dysbindin gene

(DTNBP1) associated with multiple psychiatric disorders including

schizophrenia. Behavioral and Brain Functions, 6(1), 1-7.

Wåhlberg, A. E. a., Dorn, L., & Kline, T. (2010). The effect of social desirability on

self reported and recorded road traffic accidents. Transportation Research

Part F: Traffic Psychology and Behaviour, 13(2), 106-114.


Wang, Y., Toor, S. S., Gautam, R., & Henson, D. B. (2011). Blink frequency and

duration during perimetry and their relationship to test-retest threshold

variability. Investigative Ophthalmology & Visual Science, 52(7), 4546-4550.

Waters, W. F., Adams, S. G., Binks, P., & Varnado, P. (1993). Attention, stress and

negative emotion in persistent sleep-onset and sleep maintenance insomnia.

Sleep, 16(2), 128-136.

Watling, C. N. (2012). Stop and revive? : the effectiveness of nap and active rest

breaks for reducing driver sleepiness. (Masters by Research), Queensland

University of Technology, Brisbane, Australia.

Watling, C. N. (2014). Sleepy driving and pulling over for a rest: Investigating

individual factors that contribute to these driving behaviours. Personality and

Individual Differences, 56, 105-110.

Watling, C. N., Åkerstedt, T., Kecklund, G., & Anund, A. (2015). Do repeated

rumble strip hits improve driver alertness? Journal of Sleep Research, (in

press).

Watling, C. N., Armstrong, K. A., Obst, P. L., & Smith, S. S. (2014). Continuing to

drive while sleepy: The influence of sleepiness countermeasures, motivation

for driving sleepy, and risk perception. Accident Analysis & Prevention, 73,

262-268.

Watling, C. N., Armstrong, K. A., & Radun, I. (2015). Examining signs of driver

sleepiness, usage of sleepiness countermeasures and the associations with

sleepy driving behaviours and individual factors. Accident Analysis &


Watling, C. N., Armstrong, K. A., & Smith, S. S. (2013). Sleepiness : how a

biological drive can influence other risky road user behaviours. Paper


presented at the 2013 Australasian College of Road Safety (ACRS) National

Conference, Adelaide, Australia.

Watling, C. N., Armstrong, K. A., Smith, S. S., & Obst, P. L. (2015). Crash risk

perception of sleepy driving and its comparisons with drink driving and

speeding: Which behavior is perceived as the riskiest? Traffic Injury

Prevention, 17(4), 400-405.

Watling, C. N., Smith, S. S., & Horswill, M. S. (2014). Stop and revive? The

effectiveness of nap and active rest breaks for reducing driver sleepiness.


Watling, C. N., & Watling, H. (2015). Sleepy driving and drink driving: attitudes,

behaviours, and perceived legitimacy of enforcement of younger and older

drivers. Paper presented at the Australasian Road Safety Conference 2015,

Gold Coast Convention and Exhibition Centre, Queensland,

Australia. http://eprints.qut.edu.au/89249/

Wertz, A. T., Ronda, J. M., Czeisler, C. A., & Wright, K. P. (2006). EFfects of sleep

inertia on cognition. JAMA, 295(2), 159-164.

Wetton, M. A., Hill, A., & Horswill, M. S. (2013). Are what happens next exercises

and self-generated commentaries useful additions to hazard perception

training for novice drivers? Accident Analysis & Prevention, 54, 57-66.

Wetton, M. A., Horswill, M. S., Hatherly, C., Wood, J. M., Pachana, N. A., &

Anstey, K. J. (2010). The development and validation of two complementary

measures of drivers' hazard perception ability. Accident Analysis &

Prevention, 42(4), 1232-1239.

Whittle, D., Whidborne, R., Witt, C., Buffery, S., Smith, L., Castley, A., &

Christiansen, F. (2008). Utilising the Oragene DNA Self-Collection kit for

http://eprints.qut.edu.au/89249/


the isolation of DNA from saliva for HLA typing by sequence based typing

(SBT) and Luminex beads. Human Immunology, 69, (1 Suppl), S71.

Wilde, G. J. (1982). The theory of risk homeostasis: implications for safety and

health. Risk Analysis, 2(4), 209-225.

Wilkinson, R. T. (1961). Interaction of lack of sleep with knowledge of results,

repeated testing, and individual differences. Journal of Experimental

Psychology, 62(3), 263-271.

Wilkinson, R. T., & Houghton, D. (1982). Field test of arousal: A portable reaction

timer with data storage. Human Factors, 24, 487-493.

Williams, A. F., & O'Neill, B. (1974). On-the-road driving records of licensed race

drivers. Accident Analysis & Prevention, 6(3-4), 263-270.

Williams, G., Freedman, Z., & Deci, E. (1998). Supporting autonomy to motivate

glucose control in patients with diabetes. Diabetes Care, 21(10), 1644-1651.

Williamson, A. M., & Feyer, A.-M. (2000). Moderate sleep deprivation produces

impairments in cognitive and motor performance equivalent to legally

prescribed levels of alcohol intoxication. Occupational and Environmental

Medicine, 57(10), 649-655.

Williamson, A. M., Feyer, A.-M., Mattick, R. P., Friswell, R., & Finlay-Brown, S.

(2001). Developing measures of fatigue using an alcohol comparison to

validate the effects of fatigue on performance. Accident Analysis &

Prevention, 33(3), 313-326.

Winterer, G., Musso, F., Vucurevic, G., Stoeter, P., Konrad, A., Seker, B., . . .

Weinberger, D. R. (2006). COMT genotype predicts BOLD signal and noise

characteristics in prefrontal circuits. Neuroimage, 32(4), 1722-1732.


Woestenburg, J., Verbaten, M., & Slangen, J. (1983). The removal of the eye-

movement artifact from the EEG by regression analysis in the frequency

domain. Biological Psychology, 16(1), 127-147.

World Health Organisation. (2004). The world health report. Geneva: World Health

Organisation.

World Health Organization. (2013). WHO global status report on road safety 2013:

supporting a decade of action: World Health Organization.

Wright Jr, K. P., Bogan, R. K., & Wyatt, J. K. (2013). Shift work and the assessment

and management of shift work disorder (SWD). Sleep Medicine Reviews,

17(1), 41-54.

Wright Jr, K. P., Drake, A. L., Frey, D. J., Fleshner, M., Desouza, C. A., Gronfier,

C., & Czeisler, C. A. (2015). Influence of sleep deprivation and circadian

misalignment on cortisol, inflammatory markers, and cytokine balance.

Brain, Behavior, and Immunity, 47, 24-34.

Wright, K. P., Lowry, C. A., & Lebourgeois, M. K. (2012). Circadian and

wakefulness-sleep modulation of cognition in humans. Frontiers in

Molecular Neuroscience, 5, 50-50.

Xanidis, N., & Brignell, C. M. (2016). The association between the use of social

network sites, sleep quality and cognitive function during the day. Computers

in Human Behavior, 55, 121-126.

Xu, J., Turner, A., Little, J., Bleecker, E. R., & Meyers, D. A. (2002). Positive results

in association studies are associated with departure from Hardy-Weinberg

equilibrium: hint for genotyping error? Human Genetics, 111(6), 573-574.


Yamaguchi, S., Isejima, H., Matsuo, T., Okura, R., Yagita, K., Kobayashi, M., &

Okamura, H. (2003). Synchronization of cellular clocks in the

suprachiasmatic nucleus. Science, 302(5649), 1408-1412.

Yamamoto, Y., Tanaka, H., Motoko, M., Yamazaki, K., & Sirakawa, S. (2000).

Characteristics of personality on sleep quality: Study for neuroticism and

extroversion–introversion. Japanese Journal of Health Psychology, 13(1),

13-22.

Yang, C.-M., Lin, F.-W., & Spielman, A. J. (2004). A Standard Procedure Enhances

the Correlation Between Subjective and Objective Measures of Sleepiness.

Sleep, 27(2), 329-332.

Zhou, X., Ferguson, S. A., Matthews, R. W., Sargent, C., Darwent, D., Kennaway,

D. J., & Roach, G. D. (2011). Dynamics of neurobehavioral performance

variability under forced desynchrony: Evidence of state instability. Sleep,

34(1), 57-63.

Zijlstra, F. R. H. (1993). Efficiency in work behaviour: A design approach for

modern tools: TU Delft, Delft University of Technology.

Zung, W. W., & Wilson, W. P. (1967). Sleep and dream patterns in twins: Markov

analysis of a genetic trait. In J. Wortis (Ed.), Recent advances in biological

psychiatry (pp. 119-130). New York: Springer.

Zylka, M. J., Shearman, L. P., Weaver, D. R., & Reppert, S. M. (1998). Three period

Homologs in Mammals: Differential Light Responses in the Suprachiasmatic

Circadian Clock and Oscillating Transcripts Outside of Brain. Neuron, 20(6),

1103-1110.


Appendix A

Table A.1. Single Nucleotide Polymorphisms to be Examined by Study One.

No. Gene SNP MAF 1 rs10462020 PERIOD3 0.22 2 rs10774044 CACNA1C 0.08 3 rs1082214 TIMRLESS 0.39 4 rs11541353 NPAS2 0.48 5 rs1154155 HLA-DQB1*0602 0.48 6 rs1554338 CRY2 0.24 7 rs17107484 SCP2 0.26 8 rs1799990 PRNP 0.29 9 rs1801260 CLOCK 0.38 10 rs1868049 ARNTL 0.40 11 rs2291738 TIMELESS 0.09 12 rs2298383 ADORA2A 0.30 13 rs2304672 PERIOD2 0.10 14 rs2735611 PERIOD1 0.12 15 rs2797687 PERIOD3 0.10 16 rs3736544 CLOCK 0.40 17 rs3749474 CLOCK 0.34 18 rs4242909 POLE 0.22 19 rs4630333 TIMELESS 0.20 20 rs4680 COMT 0.10 21 rs5751876 ADORA2A 0.29 22 rs8119787 NFATC2 0.20 Note. SNP = Single Nucleotide Polymorphisms; MAF = Minor allele frequency.


Appendix B

Table B.1. ANOVA Statistics Examining the Order Effects of Completing Two Testing Conditions (i.e., Motivated and Non-motivated) in One Day for the Two Days of Testing.

MOT condition 1st on 1st day session

MOT condition 1st on 2nd day session

Interaction


2 F df ηp2

PVT sessionsa EEG F5 theta 0.06 1,14 .00 2.05 1,14 .12 1.02 1,14 .07 F5 alpha 0.51 1,14 .04 2.29 1,14 .17 1.99 1,14 .01 C3 theta 0.14 1,14 .01 1.26 1,14 .15 0.77 1,14 .06 C3 alpha 0.05 1,14 .00 1.43 1,14 0.7 0.48 1,14 .01 O1 theta 1.11 1,14 .07 0.77 1,14 .06 0.90 1,14 .06 O1 alpha 0.84 1,14 .06 0.47 1,14 .04 2.02 1,14 .13 KSS 0.94 1,14 .07 2.04 1,14 .11 1.91 1,14 .10 MES 1.16 1,14 .08 0.41 1,14 .03 0.98 1,14 .07 PVT 1.14 1,14 .08 0.37 1,14 .03 0.34 1,14 .03 HPT sessionsb EEG F5 theta 0.06 1,14 .01 0.53 1,14 .02 1.36 1,14 .06 F5 alpha 0.25 1,14 .02 0.96 1,14 .03 1.96 1,14 .05 C3 theta 0.48 1,14 .04 0.02 1,14 .02 1.11 1,14 .07 C3 alpha 0.44 1,14 .03 0.15 1,14 .01 0.98 1,14 .01 O1 theta 0.48 1,14 .04 0.01 1,14 .00 0.67 1,14 .05 O1 alpha 0.55 1,14 .04 0.08 1,14 .00 1.89 1,14 .08 KSS 1.81 1,14 .12 1.10 1,14 .08 0.70 1,14 .01 MES 2.01 1,14 .13 0.94 1,14 .07 1.31 1,14 .09 HPT 0.01 1,14 .01 1.14 1,14 .11 0.59 1,14 .04 a PVT sessions were calculated with a 2 x 2 x 2 x 2 x 2 mixed factorial ANOVA was used with two between-factors (Motivation Condition First-First Day [two levels] and Motivation Condition First-Second Day [two levels] factors) and three within-factors (Motivation Level [two levels], Sleepiness Level [two levels], and Time Period [three levels] factors). Except for the MES data which used a 2 x 2 x 2 x 2 mixed factorial ANOVA, which had exactly the same factors previously described except there was no Time Period factor. b HPT sessions were calculated with a 2 x 2 x 2 x 2 x 3 mixed factorial ANOVA was used with two between-factors (Motivation Condition First-First Day [two levels] and Motivation Condition First-Second Day [two levels] factors) and three within-factors (Motivation Level [two levels], Sleepiness Level [two levels], and Time Period [three levels] factors). * < .05, ** < .01.

Sleepiness and Driving - QUT · Driver sleepiness contributes to a substantial proportion of fatal...

Documents

Transcript of Sleepiness and Driving - QUT · Driver sleepiness contributes to a substantial proportion of fatal...