University of Central Florida University of Central Florida

STARS STARS

Electronic Theses and Dissertations, 2004-2019

2010

Effects Of Flight Factors On Pilot Performance, Workload, And Effects Of Flight Factors On Pilot Performance, Workload, And

Stress At Final Approach To Landing Phase Of Flight Stress At Final Approach To Landing Phase Of Flight

Kyongsun Lee University of Central Florida

Part of the Engineering Commons

Find similar works at: https://stars.library.ucf.edu/etd

University of Central Florida Libraries http://library.ucf.edu

This Doctoral Dissertation (Open Access) is brought to you for free and open access by STARS. It has been accepted

for inclusion in Electronic Theses and Dissertations, 2004-2019 by an authorized administrator of STARS. For more

information, please contact STARS@ucf.edu.

STARS Citation STARS Citation Lee, Kyongsun, "Effects Of Flight Factors On Pilot Performance, Workload, And Stress At Final Approach To Landing Phase Of Flight" (2010). Electronic Theses and Dissertations, 2004-2019. 1628. https://stars.library.ucf.edu/etd/1628

EFFECTS OF FLIGHT FACTORS ON PILOT PERFORMANCE,

WORKLOAD, AND STRESS AT FINAL APPROACH TO LANDING

PHASE OF FLIGHT

by

KYONGSUN LEE

B.A., Korea Airforce Academy, 1993

M.S., Chungbuk National University, 2006

A dissertation submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy

in the Department of Industrial Engineering and Management Systems

in the College of Engineering and Computer Science

at the University of Central Florida,

Orlando, Florida

Fall Term

2010

Major Professor: Gene H. Lee

ii

ABSTRACT

Since human errors are one of the major causes of flight accidents, the design and

operation of the modern aircraft system deals with them seriously. Particularly, the pilot

workload on aviation causes human errors. Whenever new procedures are introduced and

operated, the aircraft capabilities have been checked in every aspect. However, there has been

little study on the impact of the new procedures such as LDLP, SCDA, SATS, and Steep

Angle approach on the pilot performance, workload, and stress.

In this study, different methods have been tried to understand the relationship between

new procedures and the pilots in terms of performance, workload, and stress. The flight

factors (e.g. flight experience, gliding angle, and approach area) were examined by the pilot

performance, workload, and stress at the “Final Approach to L/D” phase using the single

engine Cessna 172R type flight simulator. Five students and five instructor pilots from

Embry-Riddle Aeronautical University in Dayton Beach, Florida, participated and they flew

under four different simulation tasks of gliding angle and approach area. Their Heart Rate

Variability (HRV) and NASA-Task Load Index (TLX) were measured to determine their

stress level and subjective workload, respectively. In addition, Landing Performance (LP)

data (e.g. landing distance, landing speed) and Above Glide Path Tracking Performance

(AGPTP) data were also collected to evaluate pilot performance.

iii

As a result, the type of approach area showed a significant effect on pilot performance,

workload, and stress determined by ANOVA (HRV, TLX, LP, AGPTP: all are p < .05).

Flying over “Populated” area (e.g. a large city) resulted in lower pilot performance and higher

pilot workload and stress than that over “Non-Populated” area (e.g. a grass field). Similarly,

the levels of a gliding angle showed the statistical difference on the performance, workload,

and stress (HRV, TLX, and LP: all are p < .05). During the flight with 4.5 degree, the pilots

showed lower performance with higher workload and stress. However, the levels of the flight

experience did not have any influence on the performance, workload, and stress levels

(AGPTP, LP, TLX, HRV: all are p > .05).

In conclusion, flying in Populated area and flying with a 4.5 degree gliding angle

increases the workload and stress level of the pilots. In addition, when the pilots were flying

over Populated area at Final Approach to L/D phase, they showed lower performance on

tracking the glide path. Based on the results, stresses and workload can have a significant

impact on flight performance. Therefore, in order to reduce the workload and stress that can

cause human errors, it is highly recommended to carefully examine the impact of new flight

procedures on pilot workload and stress before they are implemented.

iv

ACKNOWLEDGMENTS

I would like to express my appreciation to my dissertation committee chair, Dr. Gene

Lee, and committee members, Dr. Ahmad Elshennawy, Dr. Pamela McCauley-Bush, and Dr.

Jongwook Kim for their support, guidance, and assistance. I am sincerely grateful for Dr.

Gene Lee‟s encouragement and motivation. His guidance ultimately changed my way of

thinking. I also wish to express my deepest thanks to the staff members of Embry-Riddle

Aeronautical University for their support and high regard for doing an experiment.

I would also like to thank my families, especially my wife, Hyunsuk Park, for her

entire support and prayer on this long endeavor. Without their support, patience, and

understanding, none of this would have been possible.

v

TABLE OF CONTENTS

LIST OF FIGURES ................................................................................................................. vii

LIST OF TABLES ................................................................................................................. viii

LIST OF ACRONYMS ............................................................................................................ ix

CHAPTER ONE: INTRODUCTION ........................................................................................ 1

1.1 Background ...................................................................................................................... 1

1.2 Objectives of the study..................................................................................................... 6

CHAPTER TWO: LITERATURE REVIEW ............................................................................ 8

2.1 Human error and accidents .............................................................................................. 8

2.2 Workload and performance ............................................................................................ 11

2.2.1 Measurements of workload ..................................................................................... 14

2.2.1.1 Psycho-physiological measurements ............................................................... 14

2.2.2.2 Subjective measurements ................................................................................. 15

2.2.2.3 Performance measurements ............................................................................. 17

2.3 Stress and performance .................................................................................................. 18

2.3.1 Measurements of stress ........................................................................................... 20

2.4 Three factors of pilot workload ...................................................................................... 22

2.4.1 Levels of experience ............................................................................................... 23

2.4.2 Levels of gliding angle ............................................................................................ 24

2.4.3 Types of approach area ........................................................................................... 26

2.5 Flight simulation experiment ......................................................................................... 29

CHAPTER THREE: METHODOLOGY ................................................................................ 30

3.1 Experimental design....................................................................................................... 30

3.1.1 Statistical model ...................................................................................................... 30

3.2 Protocol .......................................................................................................................... 31

3.2.1 Experimental set-up ................................................................................................ 31

3.2.2 Subjects ................................................................................................................... 33

3.2.3 Tasks ....................................................................................................................... 35

3.2.4 Procedure ................................................................................................................ 36

3.3 Data collection and processing ...................................................................................... 37

3.3.1 Heart rate variability ............................................................................................... 37

3.3.2 NASA-TLX ............................................................................................................. 39

3.3.3 Flight performances ................................................................................................ 40

3.4 Hypotheses ..................................................................................................................... 41

vi

CHAPTER FOUR: RESULTS ................................................................................................ 43

4.1 Types of Approach Area ................................................................................................ 44

4.1.1 Hypotheses test for the AA effect ........................................................................... 44

4.1.2 Graphical results related in AA ............................................................................... 47

4.2 Levels of Gliding Angle ................................................................................................. 54

4.2.1 Hypotheses test for the GA effect ........................................................................... 54

4.2.2 Graphical results related in GA ............................................................................... 56

4.3 Levels of Flight Experience ........................................................................................... 62

4.3.1 Hypotheses test for the FE effect ............................................................................ 62

4.3.2 Graphical results related in FE ................................................................................ 64

CHAPTER FIVE: DISCUSSION ............................................................................................ 68

5.1 Types of Approach Area ................................................................................................ 69

5.2 Levels of Gliding Angle ................................................................................................. 72

5.3 Levels of Fight Experience ............................................................................................ 75

CHAPTER SIX: COCLUSIONS ............................................................................................. 77

6.1 Conclusions .................................................................................................................... 77

6.2 Findings, Limitations, and Suggested further research .................................................. 79

APPENDIX A: UCF CONSENT FORM ................................................................................ 81

APPENDIX B: APPROVAL OF HUMAN RESEARCH ....................................................... 83

APPENDIX C: SUBJECTIVE QUESTIONNAIRE (NASA-TLX) ........................................ 84

APPENDIX D: COLLECTED DATA .................................................................................... 88

APPENDIX E: STATISTICAL TEST RESULTS .................................................................. 91

REFERENCES ........................................................................................................................ 95

vii

LIST OF FIGURES

Figure 1. Broad Causes of US General Aviation Accident ....................................................... 4

Figure 2. The Accident Rate of Flight Phase in US General aviation ....................................... 5

Figure 3. FRASCA MENTOR ATD Simulator ....................................................................... 32

Figure 4. S801i Heart Rate Monitor ........................................................................................ 33

Figure 5. R-R Interval on ECG Wave ...................................................................................... 38

Figure 6. Mean LP Score of AA within GA ............................................................................ 50

Figure 7. Mean AGPT Ratio of AA within GA ....................................................................... 50

Figure 8. Mean NASA-TLX MD Score of AA within GA ..................................................... 53

Figure 9. Mean HRV-LF Ratio of AA within GA ................................................................... 53

Figure 10. Mean LP Score of GA within AA .......................................................................... 59

Figure 11. Mean AGPT Ratio of GA within AA ..................................................................... 59

Figure 12. Mean NASA-TLX MD Score of GA within AA ................................................... 61

Figure 13. Mean HRV-LF Ratio of GA within AA ................................................................. 61

Figure 14. Mean LP Score between SPG and IPG .................................................................. 66

Figure 15. Mean AGPT Ratio between SPG and IPG ............................................................. 66

Figure 16. Mean NASA-TLX MD Score between SPG and IPG ............................................ 67

Figure 17. Mean HRV-LF Ratio between SPG and IPG ......................................................... 67

viii

LIST OF TABLES

Table 1. Matrix for grouping the subjects ................................................................................ 34

Table 2. Conversion table ........................................................................................................ 41

Table 3. Mean of LP, AGPT, TLX, and HRV between AA .................................................... 47

Table 4. Mean of LP, AGPT, TLX, and HRV between GA .................................................... 57

Table 5. Mean of LP, AGPT, TLX, and HRV between FE ..................................................... 65

ix

LIST OF ACRONYMS

AA Approach Area

AGPTP Above Glide Path Tracking Performance

AF Air Force

ATC Air Traffic Controller

ATSB Australian Transportation Safety Bureau

BHI Black Hole Illusion

Bpm Beat per minute

EDA Electrodermal activity

EEG Electroencephalography

EOG Electrooculograph

FAA Federal Aviation Administration

FE Flight Experience

GPO Glide Path Over estimate

GA Gliding Angle

GP Glide Path

HF High Frequency

HFACS Human Factor Analysis Classification System

HR Heart Rate

HRV Heart Rate Variability

Hz Hertz

ICAO International Civil Aviation Organization

IFR Instrument Flight Rule

IHN Instructor pilot and 4.5 degree GA and Non-Populated area

IHP Instructor pilot and 4.5 degree GA and Populated area

ILN Instructor pilot and 3 degree GA and Non-Populated area

ILP Instructor pilot and 3 degree GA and Populated area

ILS Instrument Landing System

IMC Instrument Meteorological Condition

IP(G) Instructor Pilot (Group)

L/D Landing

LF Low Frequency

Mph Mile per hour

x

NASA-TLX National Aeronautics and Space Administration – Task Load Index

NTSB National Transportation Safety Board

RWY Runway

SLN Student pilot and 3 degree GA and Non-Populated area

SLP Student pilot and 3 degree GA and Populated area

SHN Student pilot and 4.5 degree GA and Non-Populated area

SHP Student pilot and 4.5 degree GA and Populated area

SP(G) Student Pilot (Group)

T/O Take-Off

VFR Visual Flight Rule

VMC Visual Meteorological Condition

VLF Very Low Frequency

1

CHAPTER ONE: INTRODUCTION

1.1 Background

Because of a rapid development of aviation technology in both civil and military

purposes, the operation of an aircraft has also increased dramatically in quantity as well as

quality. In 2009, the total scheduled traffic carried by the airlines of the 190 Member States

of International Civil Aviation Organization (ICAO) amounted for approximately 2,280

million passengers and 38 million tons of freight (ICAO, 2010). ICAO forecasted that world

air traffic volume would increase at an annual rate of 4.4% over the year 2002-2015 (ICAO,

2004). In accordance with it, the airport facilities (e.g. runway, passenger terminal, etc.) have

been expanded toward city area to accommodate this increased air traffic volume. This, in

chain reaction, caused development of the cities near the airport, which was then surrounded

by many metropolitan cities. This occurrence gave negative impact on both surrounding

neighborhoods and pilots for the stress caused by noise and safety concerns, respectively.

This rapid growth of air traffic caused enormous noise problems on the neighborhoods.

To solve this problem, a few noise-reduced approach procedures (e.g. Lower Drag Low

Power (LDLP), Segmented Continuous Descent Approach (SCDA), etc.) were introduced

(Elmenhorst et al., 2009). Moreover, several airports and aircraft manufacture tested Steep

2

Angle Approach (Airbus, 2006; Rob, 2001) to see if it had any impacts on the noise level.

Another problem caused by the intense growth of air traffic was the air traffic congestion. So,

the aviation society introduced the Small Aircraft Transportation System (SATS) to alleviate

this problem (NASA, 2004). These procedures require a steeper gliding angle than normal,

which is usually 2.5 to 3 degrees (°). This steeper gliding angle, however, is known to

increase pilot workload and safety concerns at Final Approach to L/D phase of flight

(Boehm-Davis et al., 2007; Roscoe, 1975).

Stress in a general term refers to some undesirable condition, circumstance, task, or

other factors that impinge upon an individual (McCormick & Sanders, 1993). Some possible

sources of stresses include heavy work, immobilization, extreme temperature conditions,

noise, sleep loss, danger, information overload, boredom, loneliness, and financial insecurity.

These constraints can induce a significant mental (cognitive) workload (Hankins & Wilson,

1998) defined as the ratio between task demands and the capacity of the operator (Veltman &

Gaillard, 1996). In general, the pilots have many tasks to do during a flight in a cockpit. For

example, they should supervise the status of the aircraft system and anticipate future tasks,

including their primary tasks such as flying, navigation, and communications. Thus, aircraft

piloting requires a high level of the cognitive activity associated with various stress factors

such as time constraints, safety threats, and environmental factors. Mental workload can be

3

related to the following consequences: 1) physiological states of stress and effort 2) the

subjective experiences of stress, mental effort, and time pressure 3) the objective measures of

performance levels and 4) breakdowns in performance (Schvaneveldt et al., 1998). Lysaght et

al. (1989) described that an increase in workload would lead to decreases in performance.

Therefore, acceptable performance can be expected at reasonable levels of workload. A

reduction of mental and physical workload on pilots is the top priority in the aviation

community since many accidents happen due to pilot error.

National Transport Safety Board (NTSB) accident database showed that human errors

were cited as a broad cause or factor in 88 to 91 % of all US general aviation accidents

(NTSB, 1998~2005). As it can be seen in Figure 1 below, in 2005, the factors that caused

accidents were human factors (91 %), environmental factors (39 %), and aircraft factors

(25 %). Here, the cumulative percentage exceeds 100 %, which means more than one cause

was involved in a single accident. The 91 % of the human factors are made of 81 % in pilots,

9 % in on and off board personnel, and 1 % in personnel of organization.

4

Figure 1. Broad Causes of US General Aviation Accident

It is well known that the Final Approach to L/D phase of flight is the most work

loaded phase and is a critical phase for the severe accidents. Although it is a very short time

period compared to a total flight time, Final Approach to L/D phase of flight is nearly 50 %

of all accidents, according to the report by Nagel (1988). From the NTSB accident reports in

1998-2005, Figure 2 shows that 55 % of total accidents happened at Final Approach to L/D

phase, 23 % at “T/O & Climb” phase, 18 % at “En-route Fight” phase, and 4 % at “Ground

Operation” phase. Accidents by human errors as well as accidents at Final Approach to L/D

phase of flight are the main factors of previous accidents, as it is shown in Figures 1 and 2.

5

Figure 2. The Accident Rate of Flight Phase in US General aviation

During the flight, pilots in the cockpit compete for many important tasks (e.g.

communicating with air traffic controllers, supervising instruments, and looking outside of

cockpit, etc.) to make it a safe flight at all times. They continuously interpret information to

make right decisions. Particularly, the pilots become busier with piloting at Final Approach to

L/D phase (e.g. setting its flap and landing gear, maintaining airspeed, tracking glide path,

etc.). In addition, both internal and external factors (e.g. mechanical failure, severe weather,

etc.) are very crucial in aircraft piloting at Final Approach to L/D phase. This is because the

aircraft is under low altitude and low air speed. Therefore, pilot workload is affected by

6

various factors during the flight and it is frequently blamed for the cause of accidents. If such

potential causes of pilot workload are removed or reduced, the aviation safety will be

improved.

In summary, various landing procedures (e.g. LDLP, SCDA, SATS, Steep Angle

Approach, etc.) at Final Approach to L/D phase have been studied to solve the key issues

regarding noise reduction as well as human health. However, there is still lack of

achievement on either the technical flight safety or the capability of a human operator for the

sophisticated landing procedures (Elmenhorst et al., 2009). As science and technology has

been developed, a machine or system is expected to perform 99 % of its own function

(McCormick & Sanders, 1993), while human beings make errors that could implicate in

various accidents as well as accidents in the aviation field. Therefore, newly introduced

procedures should strongly focus on reducing human errors for improving the safety.

1.2 Objectives of the study

So far, the aviation society has been spending and investing their efforts and money to

fix such problems explained in the introduction section. They also have developed a new

approach procedure such as the Steep Angle approach, which has more than 3º of a gliding

angle, in order to reduce the noise created by aircrafts. Current approach procedures have a

7

common gliding angle of 3º. They have mainly focused on testing the aircraft capability than

the pilot workload and stress. Thus, there are only few researches that report on how the steep

angle over 3º affects pilot workload at Final Approach to L/D phase, which is known as the

most dangerous flight segment. Therefore, the assessment of pilot workload for executing

new procedures at Final Approach to L/D phase is an important aspect to improve the

aviation safety and human health under various environmental factors (e.g. weather, traffic

volume in approach area, type of approach area, etc.).

In this study, how pilot workload is affected by various flight factors while flying at

Final Approach to L/D phase through simulated scenarios by a flight simulator is examined.

This would provide a better understanding of the relationship between pilot workload and an

accident. The objectives of this study are as follows;

1) To compare the level of pilot performance, workload, and stress experienced by

both experienced pilots and less-experienced pilots.

2) To identify and describe pilot performance, workload, and stress with two different

levels of a gliding angle (e.g. 3º and 4.5°).

3) To investigate whether environments of the final approach course (e.g. Non-

Populated & Populated area) has an impact on pilot performance, workload, and

stress or not.

8

CHAPTER TWO: LITERATURE REVIEW

There are various factors that cause unsafe behavior or accidents in a work

environment. Generally, the majority of the worst performance and accidents that still occur

are attributed to human errors. So, it is important to better understand the various roles of the

human operators involved in a system or task to further enhance performance and safety. In

this research, literature that is relevant to the topic of analyzing human error, workload, stress,

and their relationship with performance and accidents is reviewed. In addition, several

methodological approaches are also introduced to measure performance, workload, and stress

of human work.

2.1 Human error and accidents

Numerous definitions have been proposed for human error. From Heinrich‟s (Heinrich

et al., 1931) axioms of the industrial safety to Reason‟s (1990) “Swiss Cheese” model of

human error, a sequential theory of accident causation has been consistently agreed by some

researchers in the field of human error (Wiegmann & Shappell, 2001). According to Reason

(1990), it would be taken as a generic term to encompass all those occasions in which a

planned sequence of mental or physical activities failed to achieve its intended outcome.

9

McCormick and Sanders (1993) also defined that human error was an inappropriate or

undesirable human decision or behavior that reduced, or had the potential for reducing,

effectiveness, safety, or system performance. In the past, human error has been used to

describe operator error. Recently, the broader perspective of human error has considered

other humans (e.g. managers, system designers, maintainers, and coworkers) as a contributor

to an accident. Petersen (1996) stated that human error is the basic cause behind all accidents.

Over the years, numerous error classification schemes have been developed to give

useful insight into the ways for preventing human error. Senders and Moray (1991) suggested

a useful classification scheme of human error. They suggested four types of taxonomy; 1)

phenomenological taxonomy (e.g. omissions, substitutions), 2) internal processes taxonomy

(e.g. capture errors, work overload, and decision errors), 3) neuropsychological mechanisms

taxonomy (e.g. forgetting, stress, attention), and 4) external processes taxonomy (e.g. poor

equipment design). It has been demonstrated that workload is one of the important factors

that cause human errors (Kantowitz & Sorkin, 1983).

Generally, the human being is most reliable under adequate levels of workload that do

not change suddenly and unpredictably. When workload is excessive, errors arise from the

inability of the human operators to cope with high information rates imposed by the

environment (Kantowitz & Casper, 1988). For example, human error has been cited as a

10

cause or contributing factor on disasters and accidents in industries such as nuclear power

(e.g. Three Mile Island accident), aviation (e.g. pilot error), space exploration (e.g., Space

Shuttle Challenger Disaster), and medicine (e.g. medical error) (Reason,1990; Woods, 1990).

Therefore, human error prevention can be the greatest contributor to improve productivity,

quality, and safety.

Many aviation accidents do not happen with a single event or a single mistake. They

rather result from a chain of such events and mistakes culminating with the errors of aircrews,

especially the pilots. Pilot error, called cockpit error in the aviation field, is a term used to

describe the cause of an accident involving an aircraft where the pilot is considered to be

principally or partially responsible. It can be defined as a mistake, oversight, or lapse in

judgment by an aircraft operator during the flight.

A closer examination of the current aviation accident records revealed that between 60 %

and 80 % of all aviation accidents in both military and civil areas are partially due to human

error (Wiegmann & Shappell, 1997; O‟Hare et al., 1994; Yacavone, 1993). It is considered

that multiple perspectives on an aviation accident is required to reduce those accidents. For

quantitative and qualitative improvements in aviation safety, we should primarily focus on

reducing human errors by the pilots and others (mechanics, supervisors, air traffic controllers)

who are involved with the flight safety.

11

2.2 Workload and performance

Workload is defined as the amount of cognitive or attention resources being expended

at any given point in time based on the information processing theory. It is the difference

between the information that needs to be processed to make decisions and the capacity of the

individual to meet those demands (Sweller et al., 1998). Besides, workload is widely used for

various measurement techniques to evaluate the effectiveness of the equipment and work

systems by researchers of human factors.

Many researchers have reported that mental or physical workload is an important factor

in determining human performance in complex systems. Thus, it has been recognized that

optimizing the allocation of workload to individuals can reduce human errors and lead to

increase in productivity (Xie & Salvendy, 2000; Moray, 1988; Gopher & Donchin, 1986).

Some of the factors that can impact workload include: 1) skill of the operator 2) training 3)

operating procedures 4) operating conditions 5) staffing levels and competence 6) task

allocations 7) job task demands 8) organizational expectations 9) task complexity, and 10)

work pace (Cuevas, 2003; Kantowitz, 1987; Bainbridge, 1974).

According to Gawron et al. (1989), workload was commonly used as two different

meanings on the flight task. One meaning depended on work done by pilots in the aircraft.

This indicated that workload varied with the difficulty of the task and the number of tasks

12

being performed. For example, Watson et al. (1996) reported that task difficulty significantly

influenced on pilot workload. Sarno and Wickens (1995) state that the main contributor to

pilot workload is task loading. This is the number of tasks required to be performed at a time.

Moray et al. (1991) showed that there were some limits to performance in the context of

scheduling multiple tasks as well. So, the requirement to perform multiple tasks is a major

contributing factor on workload as well as performance levels (Kantowitz, 1987; Wickens &

Yeh, 1982). The other meaning depended on the environmental conditions or circumstances

under the operation to fly. This suggested that workload was influenced by environmental

factors, under which the tasks were performed

Workload for the pilot varies due to diverse features of flight. Different phases of flight

present different workloads for the pilot. For example, Wilson and Hankins (1994) found that

Instrument Flight Rule (IFR) flight created significantly higher workload than Visual Flight

Rule (VFR) flight. In another study of Hankins and Wilson (1998), they reported that pilots

showed higher workload for Take-Off (T/O), Landing (L/D), and IFR flight phase than

Cruise and VFR flight phase. In addition, the increase on the pilot workload during the most

information-loaded flight phase such as T/O and L/D has been reported by Ylonen et al.

(1997), Sekiguchi et al. (1978), and Opmeer and Krol (1973). Workload is modified further

by factors not under the control of the crew, such as weather, visibility, traffic density, and

13

communication requirements, etc. (Kantowitz & Casper, 1988).

The relationship between workload and performance as described by Lysaght et al.

(1989) showed that high workload results in poor performance. In detail, Searle et al. (1999)

reported that the high work demand condition revealed significantly decreased accuracy in

comparison to the low work demand condition and the partial replication. This has been

supported by many studies in a variety of experimental settings (Cox-Fuenzalida, 2007;

Goldberg & Stewart, 1980; Cumming & Croft, 1973). However, Wickens and Holland (2000)

stated that increases in workload do not necessarily have negative consequences. For example,

Schvaneveldt (1969) demonstrated that performance of relatively simple tasks can be

degraded when coupled with complex, independent tasks. In general, however, over

workload is related to poor performance (Lysaght et al., 1989).

So far, many assessments on workload have been evaluated from performance,

subjective impressions of workload, and physiological indicators of work and stress. Based

on data obtained from these assessments, it turned out that various factors (e.g. ambiguity,

uncertainty, time pressure, level and place of control, and task complexity) can have an

impact on or can be related to work performance in military, medical, and university fields

(Xiao et al., 1996, Kanki, 1996; Urban et al., 1996; Serfaty et al., 1993).

14

2.2.1 Measurements of workload

Levels of cognitive demands can lead to errors that cause catastrophic accidents

(Wilson, 2002a). Therefore, it is necessary to design systems and flight procedures that can

reduce cognitive demands in order to not exceed the capacities of the human operators. The

complexity of flying requires the pilots to use numerous cognitive processes. So, more than

one measurement is required to determine the pilot workload. One measurement cannot be

expected to give a full insight into the multifaceted nature of piloting. The various aspects of

workload have led to distinct means for assessing pilot workload, including psycho-

physiological criteria (e.g. the heart rate, the heart rate variability, the evocation of potentials,

etc.), performance criteria (e.g. quantity and quality of performance), and subjective criteria

(such as rating of level of effort), reported by Schvaneveldt et al. (1998). Details of three

aspects are discussed below.

2.2.1.1 Psycho-physiological measurements

Psycho-physiological measurements used for many years in aviation provide objective

data on the cognitive demands of flying (Wilson, 2002a). As the recording equipment has

become considerably reliable, these measurements have facilitated user acceptance and

increased the potential to collect in-flight data (Wilson, 2001). These measurements include

15

Heart Rate Variability (HRV; Hilburn et al., 1995; Sauvet et al., 2009), Heart Rate (HR;

Roscoe, 1975; Lindholm & Cheatham, 1983; Wilson, 1993; Veltman, 2002), Electrodermal

activity (EDA; Steptoe et al., 1997), Eye Blink (Veltman et al, 1996; Wilson, 1993; Stern et

al., 1984), brain activity (Dussault et al., 2004; Gundel et al., 1997), and EEG (Sterman et

al.,1994).

2.2.2.2 Subjective measurements

A number of different approaches and techniques for measuring subjective workload

have been introduced. For example, Hill et al. (1992) suggested four measurements that were

mostly popular or prospective for application: 1) National Aeronautics and Space

Administration-Task Load Index (NASA-TLX, Hart & Staveland, 1988) 2) Modified

Cooper-Harper Scale (MCH, Cooper & Harper, 1969) 3) the Overall Workload (OW,

Vidulich & Tsang, 1987), and 4) Subjective Workload Assessment Technique (SWAT, Reid

et al., 1981).

Among these measurements, in this study, the paper-and-pencil-based NASA-Task

Load Index (TLX) method was used to measure workload. It was developed through vast

laboratory research by Hart & Staveland in 1988 and its sensitivity and validity has been

reported. According to Hill et al. (1992), the NASA-TLX is consistently superior when

16

considering sensitivity, as measured by factor validity and operator acceptance by comparing

those four measures mentioned above. This method has been extensively tested and

frequently used in human performance studies (Jorgensen et al., 1999) and pilot workload in

real (Shively et al., 1987) and simulated flight tasks (Corwin et al., 1989; Tsang & Johnson,

1989; Battiste & Bortolussi, 1988; Vidulich & Bortolussi, 1988; Nataupsky & Abbott, 1987)

for a long time. In particular, Moroney et al. (1995) stated that it is considered to be a firm

method of measuring subjective workload.

The NASA-TLX is a multidimensional rating technique that requires operators to

assign the ratio on each of six sub-dimensions of subjective workload: Mental Demand (MD),

Physical Demand (PD), Temporal Demand (TD), Performance, Effort, and Frustration. The

NASA-TLX method consists of two parts: 1) ratings and 2) weightings. For ratings,

participants score on six individual 21-point bipolar subscales. For weightings, participants

select the one between each pair which is considered to provide the most significant source of

workload in the tasks. There are 15 pairwise comparisons composed of 6 sub-dimensions,

already mentioned above. The value derived from this paired-comparison procedure is used

to weight the rating associated with each sub-dimension. The weighted values are then

combined to provide an overall workload index.

17

2.2.2.3 Performance measurements

Performance measurements monitor an operator and system performance during a task

and seek to objectively determine the level of successful completion of the task. This

category can be divided into primary and secondary task measurements. The primary task

measurement is used to measure workload under consideration, whereas the secondary task

measurements are typically used to determine the amount of spare cognitive capacity while

the operator is performing the primary task. The primary task measurement provides a direct

indication of performance on the task of interest and the secondary task measurements

provide a useful index of spare cognitive capacity (e.g., the ability to respond to emergencies

or unforeseen events).

The measurements obtained from the primary and secondary tasks are probably an

obvious index of mental workload assessment (Meshkati et al., 1995). Both reaction time and

accuracy are used as the most appropriate ways to measure particular performance by the

nature of the task (Farmer et al., 2000; Fitts, 1966). Flight performance data (e.g. landing

performance, glide path tracking performance, reaction time from the certain events, etc.)

from real flights can help us to understand the usual work demands from the pilots.

18

2.3 Stress and performance

Stress is a dynamic phenomenon caused by the quality and quantity of transactions

between human and environmental demands (Hancock & Warm, 1989; Lazarus & Folkman,

1984). Generally, it has been described as an incompatibility between the individual and his

or her work environment (Humphrey, 1998). For example, evidence from space research and

environments showed that negative responses to stress largely depended on how each

individual evaluated their capacity to manage the output imposed by the stressor (Kanki,

1996; Orasanu & Backer, 1996).

As a part of the workload experienced, stress is typically associated with excessive

high workload levels. Kantowitz (1987) emphasized the complex and multifaceted natures of

workload defined as a subjective experience caused by various factors such as motivation,

ability, expectations, training, timing, stress, fatigue, and circumstances. Brotheridge (2001)

concluded that emotional exhaustion, which is a type of stress, was significantly related to

workload in a study of Canadian government employees. The stress is directly proportional to

the levels of workload. Tyler and Cushway (1995) established a significant and positive link

between workload and stress levels in nurses.

To have a better understanding of the relationship between workload and stress

including various factors, it is important for researchers to determine how to get rid of

19

negative effects of stressors on individuals across organizations. Kirmeyer and Dougherty

(1988) found that the perceived workload of dispatch operators was positively associated

with a measurement of tension-anxiety. Searle et al. (2001) showed that the level of stress for

the high workload condition was significantly higher than that for the low workload condition.

In some cases, Cox-Fuenzalida et al. (2004) asserted that some changes in workload

conditions might serve as a stimulus that induced stress. This implies that the shift of

workload from low to high would raise the level of stress perceived by an individual.

The relationship between stress and performance has been verified in many

experiments and has been explained by the Yerkes-Dodson Law: the relationship between

stress and performance has an inverted U-shape (Yerkes & Dodson, 1908; Ahmadi & Alireza,

2007). The law dictates that performance increases with physiological or mental stress, but

only up to a point. When levels of stress become too high, performance decreases. Many

researchers have supported the Yerkes-Dodson Law of the relationship between stress and

performance (Sanders, 1983; Srivastava & Krishna, 1991). However, this law has become the

subject of criticism. Some psychologists have suggested a linear positive relationship

(Meglino, 1977; Hatton et al. 1995) and a negative linear relationship (Jamal, 1985; Vroom,

1964) between stress and performance. Despite the empirical evidence supporting these

20

alternative theories, the inverted-U theory is the most used explanation for the relationship

between stress and performance (Muse et al., 2003).

For military aviator, many surveys were carried out to identify the relationships

between stress and performance. Ahmadi and Alireza (2007) mentioned life stress (both

positive and negative) is an important factor which may have some effects on performance.

Katz (1997) reported that job satisfaction and flight performance were significantly decreased

in relation to personal life events and family problems that cause stress in life. Henn (1996)

found a few significant stressors that decrease the performance ability of aviators such as

irregular work hours, fatigue, training structure, labor-management confrontations, and time

management. Moreover, Carlisle (2001) mentioned that a significant level of stress could

diminish the ability of the experienced pilots to fly safely. It is reported by Edens (1992) that

the psychological stress level has significant relationship with the pilot error that frequently

leads to aviation accidents. Therefore, stress in pilots has a greater tendency to cause

dangerous situation than stress in other people.

2.3.1 Measurements of stress

Stress is measured by physiological measures such as HR, EMG, HRV and eye pupil

response. Cuevas (2003) mentioned that traditional approaches to define stress have followed

21

a physiological stimulus response model that either focuses on the stimuli in the environment

(noise, motion, or workload) or emphasizes the physiological response (such as heart rate or

anxiety) of the individual to the putative stressor. In addition, there are a number of well

validated rating scales for different aspects of stress. Examples include Rated Perceived

Exertion (Borg, 1982), Body Part Discomfort (Corlett & Bishop, 1976), and modified

Cooper-Harper scale (Wierwille & Casali, 1983) as qualitative measurements.

In this study, Heart Rate Variability (HRV) is used to evaluate stress. A HRV signal

contains well-defined rhythms, which have been successfully shown to contain psycho-

physiological information (Akselrod et al., 1981; Sayers, 1973; Penaz et al., 1968). It is used

to measure sympathetic activation (Kamath & Fallen, 1993; Malliani et al., 1991; Pagani et

al., 1989) and especially stress (Paritala, 2009; Seong et al, 2004; Fauvel et al., 2000). Under

the stress, an increase of activity in the sympathetic nervous system, and a decrease of

activity of the parasympathetic nervous system were observed.

There are three different frequency bands (VLF, LF, and HF) in frequency domain

evaluation methods through spectral analysis. Firstly, long period rhythms contained in the

Very Low- Frequency (VLF; 0.00-0.04 Hz) band account for the long-term regulations

mechanisms, probably related to thermoregulation and other humoral factors (Kitney &

Rompelman, 1977). Secondly, in the Low Frequency (LF; 0.03-0.15 Hz) band, there is a

22

rhythm, generally centered on around 0.1 Hz. Kamath & Fallen (1993) reported an increase

in the LF power for the sympathetic activation. Sympathetic activation includes rest-tilt

maneuver, mental stress, hemorrhage, coronary occlusion, and etc. Thirdly, it is possible to

identify a high-frequency component in the HRV signal, generally in a wide range between

0.18 and 0.4 Hz. Such a rhythm, synchronous with the respiration rate, is generally accepted

as an indicator of parasympathetic activation (Cerutti et al., 1995; Pagani et al., 1986). In this

study, LF band ratio was obtained and evaluated to understand the impact of flight factors on

pilot stress.

2.4 Three factors of pilot workload

Workload contains multifaceted and complicated terms controlled by various factors,

which include: 1) the skill of the operator 2) training 3) operating procedures 4) operating

conditions 5) staffing levels and competence 6) task allocations 7) job task demands 8)

organizational expectations 9) task complexity, and 10) work pace (Cuevas, 2003; Kantowitz,

1987). For this research, three factors were considered: 1) levels of experience as a skill of

the operator factor 2) levels of a gliding angle as an operating procedure, and 3) different

approach areas as an operating condition.

23

2.4.1 Levels of experience

In general, workload is affected by levels of an operating skill. A novice is called a

less-skilled or a less-experienced performer and an expert is called a skilled or a much

experienced performer. A novice and an expert can experience different workload even

though they perform the same tasks. Many studies were conducted to evaluate the

relationship between pilot workload and the levels of experience in the aviation field. The

less-experienced pilots showed higher mental workload than the experienced pilots in T/O

and Climb phase (Yao et al. 2008) and also in Final Approach to L/D phase (Dussault et al.,

2004). Stein (1984) found significant differences in the flight workload ratings between

expert pilots and novice pilots when they were at an air transport mission. In some cases,

either novice pilots (Kakimoto et al., 1988) or the pilots who are responsible for the mission

(Hart & Hauser, 1987; Roman, 1965) had higher workloads than those pilots who had lots of

experiences or less responsibility.

In addition, some research verified the effect of the experience level on performance,

which is related to the level of work experience. Mostly, the results reported that obvious

differences existed between novices and experts in the proficiency of performance (Stokes et

al., 1997; Wiggins & O‟Hare, 1995; Fisk et al., 1992; Livingston & Borko, 1989). Because

workload affects pilot performance, such flight workload issues are relevant to aircraft

24

certification, aviation safety, cockpit design, and tactical effectiveness. The workload of

pilots from flying a high-performance aircraft has become a major problem in aviation

(Roscoe, 1993). As a result, the measurement of workload of the pilots has been widely used

for the evaluation of aircraft designs, mission analysis, and the assessment of pilot

performance during the flight operation (Dahlstrom & Nahlinder, 2006; Dussault et al., 2005).

2.4.2 Levels of gliding angle

It is widely accepted that both T/O to Climb and Final Approach to L/D phases are one

of the most difficult phases during the entire flight, while the Cruise Flight phase is known as

the easiest one. At Final Approach to L/D phase, the pilots continuously fly down in

accordance with the 3º glide path. The increase in gliding angle into 4.5º decreases the

distance covered in Final Approach to L/D phase. This decrease in distance would put time

pressure on the pilots. Time pressure can increase or decrease the pilot‟s performance and

efficiency in different conditions. It is generally agreed, however, that the relationship

between time pressure and performance is not linear but curvilinear. For example, Andrew

and Farris (1972) found that the productivity of NASA scientists and engineers increased up

to a certain point as time pressure increased. After that, however, work performance declined

25

as time pressure became even more intense than normal condition at the peak performance

level.

A few studies were conducted for evaluating Steep Approach in terms of workload.

Roscoe (1975) found that increase in heart rate responded to increasing approach angle, such

as 104 (SD; 0.68) beat-per-minute (bpm) for 3º, 108 (SD; 1.52) bpm for 6°, and 115 (SD; 2.2)

bpm for 9°. This result corresponded very closely with the pilot‟s subjective opinion of the

cockpit workload. Boehm-Davis et al. (2007) found that approaches with 5° produced safe

landings with minimal deviations from normal descent (3º) control configuration and were

rated as having a moderate level of workload. Both 6° and 7° gliding angles produced safe

landings but high workload ratings. They concluded that both 6° and 7° angles might be

achievable in the event of an emergency although both angles might not be practical for

routine approaches due to high workload.

Commonly, the descent rate of a steeper approach angle is higher than that of a normal

approach angle. The ideal descent rate varies with different sources; 1) 500 feet-per-minute

(ft/min), which is recommended by the Aircraft Owners and Pilots Association, 2) 500 to 700

ft/min, which is recommended by the FAA, and 3) no more than 1,000 ft/min, which is

recommended by the Flight Safety Foundation (Boehm-Davis et al., 2007). According to

steep approach landing rules, the lowest value of a gliding angle is 4.5°. No upper limits on a

26

gliding angle are given although 7.5° is normally considered as a maximum angle

(Transportation Canada, 2004).

2.4.3 Types of approach area

An aircraft always needs a wide, long, and open area (e.g. runway, grass field, highway,

etc.) to make a safe landing. In the past, almost all airports were built outside of cities.

However, as cities expanded, many airports got surrounded by metropolitan cities and

aircrafts fly over them to land at the airport. However, few research was conducted about the

effect of the simulated ground features (e.g. tree, building, antenna, etc) in Final Approach to

L/D area.

During the VFR flight, the aircrafts are controlled by the perceived orientation of the

ground terrain, which relies on a visual reference such as trees, buildings, etc. As a ground

theory, Calvert (1954) mentioned that the ground features gave some motivations to develop

an approach lighting system, which gives visual reference to the pilot. An absence of the

ground features can create the visual illusion that makes the pilots believe that the aircraft is

at a higher altitude than it actually is. For example, the pilot who does not recognize this

illusion can fly lower while landing over water, darkened areas, and featureless snow fields.

Some simulation studies have provided empirical support showing that pilots tended to

27

overestimate gliding angles toward a schematic (i.e., impoverished) runway (Palmisano &

Gillam, 2005; Lewis & Mertens, 1979; Kraft, 1978; Mertens, 1978). They also found that

increasing true orientation of the ground features improved the subject‟s gliding path

judgment.

However, without any information on the instrument panel, the pilots cannot always

judge the exact height of a ground feature. They can over estimate or under estimate the

height of obstacles. A common factor in aircraft accidents is a failure to maintain an adequate

clearance from the objects or the surface. After examining the performance of helicopter

pilots during the hovering flight task, Johnson and Phatak (1989) found that the pilots

attempted to maintain an altitude by holding a fixed optical location on a ground; a strategy

that led to inappropriate altitude corrections in response to vehicle movements. As a result of

the study on pilots‟ altitude perception under the real situation to operate a helicopter, Ungs

and Sangal (1990) found that the pilots showed wide variation for each test. For example,

some achieved altitudes exceeding the target altitude by 100 % and the others flied below the

target altitude when ascending and descending. Specially, it appeared that more pilots tended

to underestimate their altitudes during descending even though altitude perception of the

pilots is important to fly safely during VFR flight. Since sometimes pilot judgment of height

28

of obstacles is not right, many obstacles at Final Approach to L/D segment can affect pilot

performance negatively.

Therefore, ground features (e.g. buildings, trees, antennas, etc.) can change flight

environment, which can cause psycho-physiological stress on the pilots. The height of a

ground feature is limited by the ICAO, FAA, and local regulations for the pilots to ensure the

safe approach and landing. According to ICAO DOC 9137 and FAA Order 8260 3B (United

States Standard for Terminal Instrument Procedures, TERPS), the ratio is 50:1 (horizontal

distance to the height of ground features) from the end of the runway to 25,000 ft on the

ground horizontally at Final Approach to L/D segment. In other words, the 20 ft high ground

feature can be built 1000ft away from the location at the end of the runway.

In addition, many aircraft accidents were reported because of turbulence and wind

shear at Final Approach to L/D phase. An airflow changed by the ground features can cause

dangerous situation for an aircraft in a low altitude and a low speed at Final Approach to L/D

phase because aircraft movement is easily affected by the wind (speed and direction). Change

of the wind patterns caused by ground obstacles can cause accidents, especially for those

aircraft with low power and low speed. Therefore, accidents can easily happen at Final

Approach to L/D phase.

29

2.5 Flight simulation experiment

Realistic aircraft simulation has become available to many researchers. This makes it

possible to investigate pilot workload under conditions of no actual physical risks. Generally

speaking, simulations have unknown factors: most of the pilots viewed simulations as too far

from the actual aircraft to be of any tangible or real value. However, many people in the

research fields believe in the potential role of simulation in flight training and maintenance of

flight proficiency. Lindholm and Cheatham (1983) stated that extremely dangerous

maneuvers can be investigated without any risks. Also recording instrumentation to get data

is very simple compared to that of the actual flight. Moreover, Magnusson (2002) reported

that simulation and real flight had a close relationship in the field of psycho-physiological

reaction and mental workload. There is enough use simulation to make a realistic assessment

of the relative demands with the different types of flight on simulations. So, flight simulation

for evaluating the pilot workload is plausible.

30

CHAPTER THREE: METHODOLOGY

3.1 Experimental design

3.1.1 Statistical model

Three-factor experiments with repeated measurement were designed in order to

evaluate how flight factors affect pilot performance, workload, and stress level. It was a fixed

model with three replications for each level (Howell, 1987; Winer, 1971). Since all sources of

variability between subjects are excluded from the experimental error, repeated measurement

design provides good precision for comparing treatments or tasks. This experiment was

simulated under the ideal conditions, i.e., without air traffic controller communication,

additional traffic, and adverse weather conditions.

Three factors, which are the independent variables, used in this study are: a level of

Flight Experience (FE), Gliding Angle (GA), and Approach Area (AA). First of all, FE had

two different subjects: a student pilot and an instructor pilot. GA had two different angles of

3º and 4.5º. At last, AA on Final Approach to L/D phase had two different areas: Non-

Populated and Populated. For the dependent variables, three different measurements were

conducted: HRV as a psycho-physiological stress, NASA-TLX score as a subjective

workload, and LP score as a pilot performance. Another dependent variable, which is Above

31

Glide Path Tracking Performance (AGPTP), was also selected for evaluating the pilot

performance.

3.2 Protocol

3.2.1 Experimental set-up

A. A flight simulator (FRASCA MENTOR ATD)

All subjects flew the FRASCA MENTOR ATD as a general aviation training

simulator, shown in Figure 3. It is configured as a single-engine Cessna 172R at Advanced

Flight Safety Center (AFSC) in Embry-Riddle Aeronautical University (ERAU). This model

consists of a flight deck and visual systems. The flight deck includes control yoke (rudder

pedals, toe brakes, a three-lever power quadrant, a flap switch, an electric pitch, and a rudder

trim), the instrument panel, fiberglass shell, and simulator electronics. The visual system

consists of a Visual PC for image generation and a 52 inch diagonal Liquid Crystal Display

for display. The visual PC generates the visual image from a generic database that contains a

runway, hangars, parked aircraft, etc. A display is 1280 by 720 resolutions and is placed in

front of the flight deck to provide a realistic image correctly. In terms of the dynamics and

environments, this flight simulator is as real as an aircraft. It can also be configured for

various flight environments (FRASCA, 2010).

32

Figure 3. FRASCA MENTOR ATD Simulator

B. Heart rate monitoring system

Heart rate data was collected by using a Polar S810i Heart Rate Monitoring system

(Polar Electro, 2010), a portable wireless device shown in Figure 4. It includes a watch, an

elastic strap transmitter, and Polar ProTrainer 5.0 software, which downloads the data via

IrDA (an infra red USB) interface to PC. The elastic strap transmitter comfortably holds at a

right place on the chest. This strap contains micro-electronic circuits, which can sense the

signals from the heart when wrapped around the chest and then transmit the signals to the

watch on the wrist. This heart rate data is directly obtained in the form of beat to beat (R-R)

intervals, the time between the beats. The watch communicates all data from an electrode to a

PC via an infrared connection. A watch and an elastic strap transmitter in a heart monitor

33

system are set on Pulse Mode that shows the current heart rate (beats per minute). A 3V

lithium battery is used for power source of the heart rate monitor set (a watch and elastic

strap transmitter).

Figure 4. S801i Heart Rate Monitor

3.2.2 Subjects

Subjects were divided into two major groups by their FE levels, with five people in

each group (student pilots and instructor pilots). These major groups were divided into 8 sub-

groups according to the levels of GA and AA, as shown in Table 1. All subjects who

participated in this experiment were from ERAU. The student pilots were in the later stage of

34

their “FA 221 Instrument Pilot Single Engine Program” with an average flight hour of 199.7

± 38.7 hr. The flight instructors had “FA 417 (417A and 417I) Flight Instructor Rating” with

795.6 ± 182.5 flight hours. All of the subjects were familiar with a flight simulator since the

flight instructors had experiences with it and the student pilots were trained by the instructors

in ERAU flight school. Each subject read and signed the consent form (Appendix A), which

informed them of their rights before starting the experiment. This study using human subjects

was approved by the UCF Institutional Review Board (Appendix B).

Table 1. Matrix for grouping the subjects

Student pilots

3º

Non-Populated area1)

SLN

Repeated

3 times.

Populated area2)

SLP

4.5°

Non-Populated area1)

SHN

Populated area2)

SHP

Instructor pilots

3º

Non-Populated area1)

ILN

Populated area2)

ILP

4.5°

Non-Populated area1)

IHL

Populated area2)

IHP

Note. 1) Non-Populated area represents the country side including small mountains, rivers, and the sea.

2) Populated area represents the big cities including buildings and taller towers.

35

3.2.3 Tasks

The pilots‟ flight task was to fly a series of approaches at a single-runway airport with

an Instrument Landing System (ILS). The pilots executed a “Runway 7L ILS Procedure” at

the Daytona Beach International airport. Among the full procedures, the partial segment of

the procedure, fixed from 5 mile to a runway, was selected for the experiment. ILS is a

ground-based instrument approach system that provides precision guidance to an aircraft

approaching and landing on a runway. For example, it provides the pilots with their relative

vertical and lateral positions, in terms of the established glide path of the ILS procedures.

In this study, the pilots were instructed to follow Precision Approach Path Indicator

(PAPI) on the side of a landing runway, which provides vertical guidance information for

helping the pilots acquire and maintain the established glide path to the runway. During this

experiment, the pilots used the Localizer Indicators in the cockpit to track and control the

aircraft for their lateral positions.

There were four different scenarios for this experiment as shown in Table 1, section

3.2.2. For the scenario 1, the pilots flew with GA of 3º in Non-Populated area at Final

Approach to L/D phase of flight. Flying with the 3º in Populated area was the scenario 2. For

example, to describe Populated area, various buildings whose heights are limited from the

36

regulation of FAA Order 8260 3B were placed. The pilots flew with the 4.5º for the scenario

3 in Non-Populated area and for the scenario 4 in Populated area, respectively.

All scenarios were the Straight-In Approach and L/D starting 5 nautical miles at the

revised ILS procedure mentioned above. At the starting point, an aircraft was configured at

100 mile-per-hour (mph) in a straight and level position with no flap extensions. The initial

altitude was set to 1600 feet (ft) for the scenarios 1 and 2, and 2400 ft for the scenarios 3 and

4, respectively. An aircraft was intercepted on the glide path. The weather conditions were set

with 7 miles in visibility, moderate turbulence in wind, and a scattered cloudy sky. Prior to

the experiment, the pilots were provided with scheduled weather conditions, GA, and AA.

All of the simulation flights were terminated when an aircraft landed on or off the runway,

crashed, or did Go-around.

3.2.4 Procedure

All scenarios contained 4 steps. STEP 1 was an orientation for this study: the

experiment procedures of the flight simulation were individually provided to all pilots.

STEPs 2 to 4 were composed of 5-minute of flying for recording (STEP 2), 5-minute of

measuring for NASA-TLX (STEP 3), and 10-minute of resting (STEP 4). Here is more

detailed explanation of the STEPs. In STEP 1, each subject was informed with instructions

37

for the study (e.g., total experiment procedure, goal of experiment, notice, etc). Before sitting

on the cockpit seat, the subjects were equipped with HR monitors. The pilots were seated

comfortably on the cockpit seat in a noise-attenuated, shielded room. STEP 2 contained the

work for 5-minute flying, which means the task was finished when an aircraft landed on the

runway. After completing STEP 2, the pilots left the cockpit and rated the workload

associated with the task using the NASA-TLX shown in Appendix C for STEP 3. Then, in

STEP 4, the pilots took a rest for 10 minutes to recover HR. In all cases, HR was recorded by

a heart rate monitor during the flight. The pilots who finished scenario 1 continued these

steps for complying with scenarios 2, 3, and 4.

3.3 Data collection and processing

3.3.1 Heart rate variability

R-R interval of HR was recorded throughout the scenarios with a Polar S810i Heart

Rate Monitoring system of a sampling rate of 1000 Hz. The heart rate signal of

Electrocardiogram (ECG) consists of P, Q, R, S and T waves. The R-R interval means the

time interval from a peak point of R wave to the next peak as shown in Figure 5. The R-R

interval time differs every time since there are lots of phenomena that influence people‟s

38

body and mental condition (Murai et al., 2004).

Figure 5. R-R Interval on ECG Wave

Based on the raw R-R interval, a Polar S810i Heart Rate Monitoring system provides

HRV analysis in time domain or in frequency domain. The values of the time domain

analysis contain the mean R-R interval, the standard deviation of all R-R intervals (SDNN),

the root mean square of differences (RMSSD) of successive R-R intervals, the difference

between adjacent R-R intervals of more than 50 ms (NN50), and the proportion of difference

between the adjacent R-R intervals of more than 50 ms (pNN50). The frequency domain

analysis shows the variability of the R-R signal as a function of frequency by looking at the

proportion of the frequencies related to the original R-R signal. The frequency bands used for

frequency domain analysis consist of VLF of 0.00-0.04 Hz, LF of 0.04- 0.15 Hz and HF of

0.15-0.4 Hz (Cerutti et al., 1995). The frequency domain analysis includes the values of LF

39

and HF, and the ratio of LF to HF with a normalized unit (nu). For the present study, the LF

value was used for our statistical analysis (Seong et al., 2004; Murai et al., 2004; Sato et al.,

1995).

3.3.2 NASA-TLX

Most of the literature on NASA-TLX, which was created by Hart and Staveland in

1988, have reported workload data using the composite indices with both rating and

weighting as explained in Chapter 2. However, Moroney et al. (1995) mentioned many

applications of a single sub-dimension of workload in various fields. They suggested that

information for 6 sub-dimensions provided by the individual rating may be obscured and

even ignored. Indeed, because sub-dimension rating is sensitive to different task demands, it

is important to preserve a sub-dimension‟s property. In 1988, Vidulich and Bortolussi

reported that six sub-dimensions of NASA-TLX are sensitive to different task demands in

helicopter combat simulations: while weighted TLX failed to demonstrate the differences

between tasks, the physical demand and effort of sub-dimensions positively exhibited the

difference during high workload segments of the missions. The six sub-dimensions of

NASA-TLX have been supported in a series of simulation and field tests by Christ et al.

40

(1993). They reported that the sub-dimensions identified the type of workload associated with

different mission segments and crew positions in the helicopter-simulation study.

In addition, several studies also showed a significant correlation between weighted and

sub-dimensional ratings. Byers et al. (1989) compared the weighted and sub-dimensional

methods in the test with the evaluation of 5 Army systems. They reported a very high

correlation (r = 0.977) between the two ratings by the 5 independent studies. These empirical

tests were supported by Moroney et al. (1992) and Nygren (1991). Moroney et al. (1995)

argued that more attention should be paid to the individual sub-dimensional measurement

since it is valid for measuring workload.

In this study, the NASA-TLX method was used for the assessment on subjective

workload. The pilots evaluated their tasks after finishing them. They scored them from 1 to

21 points on six sub-dimensions of NASA-TLX. One of these sub-dimensions, Mental

Demand (MD) rating, was evaluated and was compared with workload.

3.3.3 Flight performances

Flight performances were evaluated in terms of LP and AGPTP. The pilots were

instructed to land at the point of 1,000 ft from the end of the runway, which is the general

landing zone for the pilots. To determine where the pilots land, reference lines were drawn on

41

the runway with 300 ft intervals from the landing point designated. The pilots were also

supposed to land at 65 mph. In terms of the distance and the speed, the flight parameters

collected by an experimenter were figured and converted into scores in accordance with

Table 2. For evaluating the glide path tracking performance, the flight path flown by the

pilots was printed out from the simulation database and was manually calculated to get the

percent deviation from the established glide path.

Table 2. Conversion table

L/D distance

Deviations

(± feet) 300 600 900 >1200

Scores 0 -1 -2 -3

L/D speed

Deviations

(± mph) 5 10 15 >20

Scores 0 -1 -2 -3

3.4 Hypotheses

The objectives of this study were to identify and describe the pilot performance,

workload, and stress level under the various flight factors at Final Approach to L/D phase. It

was hypothesized that the pilot stress, workload, and performance would be affected by the

levels of flight factors (FE, GA, and AA). Hypotheses in this study were as follows:

42

1. The relationship between the level of FE and the pilot performance, workload, and

stress level at Final Approach to L/D phase of flight.

- H0: FE does not affect the pilot performance, workload, and stress level.

- H1: FE does affect the pilot performance, workload, and stress level.

2. Changes in the pilot performance, workload and stress level as a function of AA.

- H0: Type of AA does not affect the pilot performance, workload, and stress level.

- H1: Type of AA does affect the pilot performance, workload, and stress level.

3. Pilot performance, workload and stress levels vary between the different GA values.

- H0: GA does not affect the pilot performance, workload, and stress level.

- H1: GA does affect the pilot performance, workload, and stress level.

43

CHAPTER FOUR: RESULTS

The questions of interest in this study were: 1) how pilot performance varies with the

different flight factors 2) what is the relationship between pilot performance and their

workload and stress at Final Approach to L/D phase of flight. To address these questions,

flight performance was analyzed first and then workload ratings and stress level were

evaluated.

The simulation output data and the subjective rating data, which are shown in

Appendix D, were analyzed with descriptive statistics and analysis of variance (ANOVA)

with MINITAB 15 statistical software to see the overall effect of the independent variables

and their interactions. Both Landing Performance (LP) and Above Glide Path Tracking

(AGPT) data were used for the flight performance to verify how it is affected by different

fight factors, i.e. Approach Area (AA), Gliding Angle (GA), and Flight Experience (FE).

Additionally, in terms of pilot workload and mental stress, NASA-TLX MD rating and HRV-

LF were obtained and processed with each flight factor.

The null hypothesis for this analysis is that the mean value of task groups is not

significantly different. The alpha level, α, used for acceptance and rejection was set to .05. If

the ANOVA shows significance and if the null hypothesis is rejected, it is assumed that at

least one mean value is different from the others. FE was the between-subjects variable tested

44

with two levels (SPG and IPG). As the within-subjects variables, two levels of GA (3º and

4.5º) and AA (Non-Populated and Populated) were tested. In this chapter, the effect of each

flight factor on the pilot performance, workload, and stress level was explained and discussed

by both statistical and graphical results.

4.1 Types of Approach Area

4.1.1 Hypotheses test for the AA effect

Hypotheses were tested to verify the relationship between the flight factors (AA) and

flight performances, which are shown by LP and AGPT scores. In addition, the NASA-TLX

score for the subjective workload and HRV-LF ratio for the stress were hypothesized and

tested for their relationships to different flight factors. The results of the ANOVA analysis of

these measurements are shown in detail in Appendix E.

Hypothesis 1a (H1a): The relationship between the level of AA and LP score at

Final Approach to L/D phase of flight.

- H0: AA does not affect pilot‟s LP score.

- H1: AA does affect pilot‟s LP score.

45

On the within-subjects variable of the levels of AA (Non-Populated and Populated

area), when a critical α of .05 was set, the LP on the levels of AA was statistically significant,

F(1, 108) = 53.26, p < .05. Thus, the pilot performance was different with respect to the

levels of AA.

Hypothesis 1b (H1b): The relationship between the levels of AA and the glide path

tracking performance.

- H0: AA does not affect the glide path tracking performance.

- H1: AA does affect the glide path tracking performance.

The p-value for the AGPTP was < .05 for the levels of AA. Based on the p-value

obtained, the null hypothesis was rejected. This indicates that the levels of AA had a

statistically significant effect on AGPTP.

Hypothesis 1c (H1c): The relationship between the level of AA and TLX score at

Final Approach to L/D phase of flight.

- H0: AA does not affect pilot‟s TLX score.

- H1: AA does affect pilot‟s TLX score.

46

At the specified α = .05 level, the TLX rating score was F(1, 108) = 29.88, p < .05. The

null hypothesis was rejected; thus, the level of AA influenced pilot workload.

Hypothesis 1d (H1d): The relationship between the level of AA and HRV ratio at

Final Approach to L/D phase of flight.

- H0: AA does not affect pilot HRV ratio.

- H1: AA does affect pilot HRV ratio.

Under the same condition above, the HRV on the levels of AA was F(1, 108) = 14.85,

p < .05, thus, pilot stress level depends on types of approach area.

To sum, the results from the statistical test indicated that pilot performance was

significantly influenced by the levels of AA at Final Approach to L/D phase of flight. The

subjective workload rating (TLX) and mental stress level (HRV) were also statistically

different depending on the levels of AA.

47

4.1.2 Graphical results related in AA

Types of the approach area were one of the independent variables of this study. Table 3

shows the mean of each dependent variable between Non-Populated and Populated area

groups. Populated group recorded a lower score for the LP and a higher score for the AGPT.

This means Populated group got a bad performance compared to Non-Populated group. The

mean LP score between Non-Populated and Populated group was -1.15 points and -1.68

points, respectively. The difference in AGPT scores between Non-Populated group (mean:

12%) and Populated group (mean: 25.71%) was noticeable with a maximum difference of

13.71%. The TLX mean score between Non-Populated and Populated group was 5.48 points

and 7.2 points, respectively. On HRV, the mean value was 22.78 % for Non-Populated group

and 25.82 % for Populated group. Both TLX and HRV data showed that the Populated area

group had more workload and stress, which leads to bad performance.

Table 3. Mean of LP, AGPT, TLX, and HRV between AA

LP(points) AGPT (%) TLX(points) HRV (%)

Non-Populated -1.15 12.00 5.48 22.78

Populated -1.68 25.71 7.20 25.82

48

Figures 6 to 9 show the effect of AA within the levels of GA (3º and 4.5º) for the

dependent variables. The LP score was compared in 3º group. As shown in Figure 6, SN

(student pilots who flew over Non-Populated area) group recorded the best score (mean: -

1.07) and IP (instructor pilots who flew over Populated area) group showed the worst

performance (mean: -1.21). Both SN and IN (instructor pilots who flew over Non-Populated

area) groups recorded better performance than SP (student pilots who flew over Populated

area) and IP. It could be interpreted that the pilot performance was affected by the different

AA within 3º GA (L). In addition, while flying with 4.5º GA (H), IN recorded the best score

(mean: -1.27) and the worst performance (mean: -2.27) was shown by SP. In accordance with

the results of 3º, both SN and IN showed better score than both SP and IP. Thus, the pilot

landing performance was highly affected by the types of area, where they flew over.

Moreover, LP score for the 4.5º (SP: -2.27, IP: -2.07) decreased more than that for the 3º (SP:

-1.20, IP: -1.21). According to the graphical results (Fig. 6) and the statistical test (H1a), it

could be interpreted that pilot performance was affected considerably by flying over the

Populated area as well as flying with 4.5º.

Figure 7 depicts AA effect within the levels of GA for the AGPT score. When flying

with 3º, SN recorded the lowest ratio (mean: 15.99) while SP showed the highest ratio (mean:

31.49). During the flight with 4.5º, IN recorded the lowest score (mean: 10.59) while SP

49

showed the highest score (mean: 26.36). The pilots showed better performance when they

flew over the Non-Populated than over the Populated area, which showed that the pilot glide

path tracking performance was affected by the types of AA, where they flew over. In addition,

the pilots showed higher AGPT ratio for the Populated area with 3º than with 4.5º. Thus,

based on graphical (Fig. 7) and statistical result (H1 b), it can be concluded that glide path

tracking performance was highly influenced by flying over the Populated area as well as

flying with 3º.

50

Figure 6. Mean LP Score of AA within GA

Note. S, I, L, H, N, P stand for Student, Instructor, 3º, 4.5º, Non-Populated area, Populated area,

respectively, as explained in Chapter 3.

Figure 7. Mean AGPT Ratio of AA within GA

S I S I

L H

N -1.07 -0.73 -1.53 -1.27

P -1.20 -1.21 -2.27 -2.07

-2.50 -2.00 -1.50 -1.00 -0.50 0.00

Pe

na

lty S

co

re

Landing Performance Score

S I S I

L H

N 15.99 9.34 12.10 10.59

P 31.49 27.35 26.36 17.64

0.00

10.00

20.00

30.00

40.00

Perc

enta

ge

Above Glide Path Tracking Score

51

The pilots‟ subjective workload rate by NASA-TLX is shown in Figure 8. During the

flight with 3º GA, SP (student pilots who flew over Populated area) rated 6.73 points, while

IN (instructor pilots who flew over Non-Populated area) rated 4.07 points. In accordance with

it, SP rated 9.2 points and IN rated 6.0 points during the flight with 4.5º GA. From the pilot

groups who showed high subjective workload level for the Populated area, the student pilot

group (SPG) recorded a higher score than that of the instructor pilot group (IPG). This meant

that pilot workload was affected by flying over the Populated area. In addition, the workload

level that the pilots showed during the flight with 4.5º was increased compared to the

workload level that the pilots showed during the flight with 3º. Based on the graphical (Fig. 8)

and statistical test (H1C) results, it could be concluded that pilot workload was influenced by

the glide path angle and the landscape of approach area.

As shown in Figure 9, when pilots flew with3º GA, SP recorded the highest score

(mean: 26.65), while IN recorded the lowest (mean: 18.65) on HRV ratio. During the flight

with 4.5º GA, SP recorded the highest ratio (mean: 27.56), while IN recorded the lowest ratio

(mean: 24.24). Moreover, in comparison to the pilots who flew over Non-populated area, the

pilots who flew over Populated area showed higher stress level. This means that pilot stress

level was highly influenced by Populated area that the pilots flew over. In addition, the pilots

showed higher stress level during the flight with 4.5º than during the flight with 3º. From

52

these results, it could be interpreted pilot stress level was influenced considerably by flying

over Populated area as well as with 4.5º (Fig. 9 and H1d).

Overall, the pilot performance, workload, and stress level are statistically different

depending on the types of approach area. Pilots showed lower performance, higher workload,

and higher stress level when they flew over Populated area than over Non-Populated area.

53

Figure 8. Mean NASA-TLX MD Score of AA within GA

Figure 9. Mean HRV-LF Ratio of AA within GA

S I S I

L H

N 5.00 4.07 6.87 6.00

P 6.73 5.47 9.20 7.40

0.00 2.00 4.00 6.00 8.00

10.00

Sco

re

NASA-TLX Mental Demand Rate

S I S I

L H

N 21.66 18.65 26.55 24.24

P 25.65 24.65 27.56 25.43

0.00

10.00

20.00

30.00

Perc

enta

ge

Heart Rate Variability

54

4.2 Levels of Gliding Angle

4.2.1 Hypotheses test for the GA effect

Hypotheses were tested to verify the relationship between the flight factors, GA, and

flight performances, shown by LP and AGPT score. In addition, the NASA-TLX score for

the subjective workload and HRV-LF ratio for the mental stress were hypothesized and tested

for their relationships to different flight factors.

Hypothesis 2a (H2a): The relationship between the level of GA and LP score at

Final Approach to L/D phase of flight.

- H0: GA does not affect pilot‟s LP score.

- H1: GA does affect pilot‟s LP score.

On the within-subjects variable of the levels of GA (3 degree and 4.5 degree), when a

critical value of α (0.05) was set, the LP on the levels of GA was statistically significant F(1,

108) = 27.61, p < .05. This indicated that the pilot performance was different with respect to

the levels of GA.

55

Hypothesis 2b (H2b): The relationship between the levels of GA and the glide path

tracking performance.

- H0: GA does not affect the glide path tracking performance.

- H1: GA does affect the glide path tracking performance.

The p-value for the AGPTP was greater than .05 for the levels of GA. Based on the p-

value obtained, the null hypothesis was not rejected. This shows that the levels of GA did not

have any significant effect on AGPTP.

Hypothesis 2c (H2c): The relationship between the level of GA and TLX score at

Final Approach to L/D phase of flight.

- H0: GA does not affect pilot‟s TLX score.

- H1: GA does affect pilot‟s TLX score.

When a critical value of α, .05, was set, TLX rating score was F(1, 108) = 30.2, p < .05.

The null hypothesis was rejected. Therefore, the level of GA did influence pilot workload.

56

Hypothesis 2d (H2d): The relationship between the level of GA and HRV ratio at

Final Approach to L/D phase of flight.

- H0: GA does not affect pilot HRV ratio.

- H1: GA does affect pilot HRV ratio.

When a critical value of α, .05, was set, the HRV on the levels of GA was F(1, 108) =

7.12, p< .05. Therefore, the pilot stress level was significantly different depending on levels

of gliding angle.

To sum, the results of statistical test indicated that pilot landing performance was

significantly different with respect to the levels of GA at Final Approach to L/D phase of

flight. However, it revealed that AGTP was not statistically different by the levels of GA. The

subjective workload rating (TLX) and stress level (HRV) were also significantly different

depending on the levels of GA.

4.2.2 Graphical results related in GA

Table 4 illustrated the mean value of each dependent variable between the levels of GA

(3º and 4.5º). The mean LP score of both 3º GA group and 4.5º GA group was -1.05 point and

57

-1.78 point, respectively. This showed that the pilots scored low for 4.5º. The difference in

AGPT between 3º (mean: 21.04%) and 4.5º (mean: 16.88%) was 4.16% and this was not

significant. Overall, the pilot showed lower performance for 4.5º than 3º. The TLX score

revealed that the pilots rated higher subjective workload during the flight with 4.5º than with

3º. In addition, the average volume of HRV-LF on each group was 22.64 % for 3º group and

25.95 % for 4.5º group, respectively. The pilots who flew with 4.5º showed more stress level.

Therefore, it could be concluded that the level of GA did significantly influence pilot

performance, subjective workload, and stress level.

Table 4. Mean of LP, AGPT, TLX, and HRV between GA

LP(points) AGPT (%) TLX(points) HRV (%)

3 degree -1.05 21.04 5.32 22.65

4.5 degree -1.78 16.68 7.37 25.95

Figure 10 to 13 showed GA effect within the levels of AA for the dependent variables.

In Non-Populated area group (N), SL (student pilot who flew with 3º GA) recorded the best

score (mean: -1.07) and SH (student pilot who flew with 4.5º GA) showed the worst

performance (mean: -1.53) as shown in Figure 10. Both SLN and ILN recorded better

58

performance than SHN and IHN. It could be interpreted that the pilot performance was

affected more by the 4.5º within Non-Populated area than 3º within Non-Populated area.

While flying over Populated area (P), SL recorded the best score (mean: -1.2) and the worst

performance (mean: -2.27) was shown by SH. In accordance with the results of Non-

Populated group, both SLP and ILP showed better scores than both SHP and IHP. Thus, the

pilot performance was highly influenced by flying with 4.5º. Moreover, LP score for Populate

group was higher than that for Non-Populated group. From these results, it could be

interpreted that pilot performance was highly affected by flying with 4.5º as well as over

Populated area. This is supported by the statistical test (H2a and H3a) as shown in this

chapter.

Figure 11 depicted the GA effect within the levels of AA for the AGPT score. When

flying over Non-Populated area (N), SL recorded the lowest score (mean: 15.99), while IL

showed the highest score (mean: 9.34). During the flight over the Populated area (P), IH

recorded the lowest score (mean: 17.64), while SL showed the highest score (mean: 31.49).

IPG showed the better performance for flying with 4.5º for the both Non-Populated and

Populated, while SPG didn‟t show a particular pattern. In addition, the difference of mean

value between 3º and 4.5º groups was small. Therefore, it could be concluded that the levels

of GA didn‟t influence on AGPT performance.

59

Figure 10. Mean LP Score of GA within AA

Figure 11. Mean AGPT Ratio of GA within AA

S I S I

N P

L -1.07 -0.73 -1.2 -1.21

H -1.53 -1.27 -2.27 -2.07

-2.5-2

-1.5-1

-0.50

Pen

alty S

core

Landing Performance Score

S I S I

N P

L 15.99 9.34 31.49 27.35

H 12.10 10.59 26.36 17.64

0.00

10.00

20.00

30.00

40.00

Perc

enta

ge

Above Glide Path Tracking Score

60

The pilots‟ subjective workload rating by NASA-TLX was shown in Figure 12. In

Non-Populated group (N), SH (student pilots who flew with 4.5º GA) rated 6.87 and IL

(instructor pilots who flew with 3º GA) rated 4.07, while SHP rated 9.2 and ILP rated 5.47

during the flight over Populated area, respectively. The pilots showed high subjective

workload level for the flight with 4.5º. This indicated that pilot workload was more

influenced by the flight with 4.5º than with 3º GA. In addition, the workload level that the

pilots showed during the flight over Populated area was higher than the results shown during

the flight over Non-Populated area. From these results, it can be interpreted that pilot

workload was indeed influenced by the flight with 4.5º as well as over Populated area.

As shown in Figure 13, SHN recorded the highest score (mean: 26.65), and ILN

recorded the lowest (mean: 18.65) on HRV ratio. During flying over Populated area (P), SH

recorded the highest score (mean: 27.56), and IL recorded the lowest (mean: 24.65).

Moreover, the pilots showed high stress levels for flying with 4.5º. This indicated that pilot

stress level was highly affected by flying with 4.5º. In addition, the pilots showed higher

stress level during flying over Populated area than during the flight over Non-Populated area.

From these results, it could be interpreted that the pilot stress level was affected considerably

by the flight with 4.5º as well as over Populated area

61

Figure 12. Mean NASA-TLX MD Score of GA within AA

Figure 13. Mean HRV-LF Ratio of GA within AA

S I S I

N P

L 5.00 4.07 6.73 5.47

H 6.87 6.00 9.2 7.40

0.00 2.00 4.00 6.00 8.00

10.00

Score

NASA-TLX Mental Demand Rate

S I S I

N P

L 21.66 18.65 25.65 24.65

H 26.55 24.24 27.56 25.43

0.00

10.00

20.00

30.00

Perc

enta

ge

Heart Rate Variability

62

4.3 Levels of Flight Experience

4.3.1 Hypotheses test for the FE effect

Hypotheses were tested to verify the relationship between the levels of FE and flight

performances, shown by LP and AGPT score. In addition, the NASA-TLX score for the

subjective workload and the HRV-LF ratio for the mental stress were hypothesized and tested

for their relationship to different flight factors.

Hypothesis 3a (H3a): The relationship between the level of FE and LP score at

Final Approach to L/D phase of flight.

- H0: FE does not affect pilot‟s LP score.

- H1: FE does affect pilot‟s LP score.

On the between-subjects variable of the levels of FE (student pilot and instructor pilot),

when a critical value α, .05, was set, the LP on the levels of FE was not statistically

significant, F(1, 108) = 0.00, p > .05. Thus, the pilot performance is not significantly

different with respect to the levels of FE.

63

Hypothesis 3b (H3b): The relationship between the levels of FE and the glide path

tracking performance.

- H0: FE does not affect the glide path tracking performance.

- H1: FE does affect the glide path tracking performance.

The p-value for the AGPTP was greater than .05 for the levels of FE. Based on the p-

value obtained, the null hypothesis was not rejected. This indicated that the levels of FE did

not have a statistically different effect on AGPTP.

Hypothesis 3c (H3c): The relationship between the level of FE and TLX score at

Final Approach to L/D phase of flight.

- H0: FE does not affect pilot‟s TLX score.

- H1: FE does affect pilot‟s TLX score.

When a critical value α (.05) was set, the TLX rating score was F(1, 108) = 1.26, p

> .05. The null hypothesis was not rejected; thus, the level of FE did not significantly

influence pilot workload.

64

Hypothesis 3d (H3d): The relationship between the level of FE and HRV ratio at

Final Approach to L/D phase of flight.

- H0: FE does not affect pilot HRV ratio.

- H1: FE does affect pilot HRV ratio.

When a critical value α (.05) was set, the HRV on the levels of FE was F(1, 108) =

0.61, p > .05, thus, the pilot stress levels were not different for levels of flight experience.

To sum, the results of statistical test indicated that pilot performance was not

significantly different with respect to the levels of FE at Final Approach to L/D phase of

flight. The subjective workload rating (TLX) and stress level (HRV) were not statistically

different depending on levels of FE.

4.3.2 Graphical results related in FE

On the pilot performance, two levels of flight experience (student pilots and instructor

pilots) were evaluated for the LP score and the Above Glide Path Tracking (AGPT) score.

From the Table 5 and Figure 14, the mean values of LP for SPG (student pilot group) and

IPG (instructor pilot group) were -1.52 points and -1.32 points, respectively. The difference

65

of LP scores between the groups was small. Although SPG showed higher performance for

the tasks, there was no significant difference on levels of performance between SPG and IPG.

Statistical results, H3a, confirmed this relationship. Figure 15 showed that SPG recorded

higher score (mean: 21.46 %) than IPG (mean: 16.23 %) for the AGPT score. The results of

ANOVA indicated that FE did not have a significant impact on AGPT (H3b, p > .05).

Therefore, the level of FE didn‟t affect either pilot workload or AGPT for this study.

This was coherent to NASA-TLX and HRV-LF results. The average scores of TLX-

MD rating of both groups was represented in Figure 16 for the subjective workload. The

mean scores of SPG and IPG were 6.95 points and 5.73 points, respectively. On the mental

stress, the average score of HRV-LF ratio of both groups were illustrated in Figure 17. It was

25.36 % for SPG and 23.24 % for IPG, respectively. Overall, FE, one of the flight factors

tested, showed no impact on the performance, workload, and stress based on the statistical

results (H3c and H3d).

Table 5. Mean of LP, AGPT, TLX, and HRV between FE

LP(points) AGPT (%) TLX(points) HRV (%)

SPG -1.52 21.49 6.95 25.36

IPG -1.32 16.23 5.73 23.24

66

Figure 14. Mean LP Score between SPG and IPG

Figure 15. Mean AGPT Ratio between SPG and IPG

Mean

SPG -1.52

IPG -1.32

-2.00

-1.50

-1.00

-0.50

0.00

Pen

alty S

core

Landing Performance

Mean

SPG 21.49

IPG 16.23

0

10

20

30

Perc

enta

ge

Above Glide Path Tracking Performance

67

Figure 16. Mean NASA-TLX MD Score between SPG and IPG

Figure 17. Mean HRV-LF Ratio between SPG and IPG

Mean

SPG 6.95

IPG 5.73

0.0

2.0

4.0

6.0

8.0

Score

NASA-TLX Mental Demand Rating

Mean

SPG 25.36

IPG 23.24

0.0

10.0

20.0

30.0

Perc

enta

ge

Herat Rate Variability-Low Frequency

68

CHAPTER FIVE: DISCUSSION

The purpose of this study was to determine the effect of the flight factors on the pilot

performance, workload, and stress level using a flight simulator. The independent variables

included approach areas (AA), gliding angles (GA), and flight experience (FE). The

dependent variables were LP & AGPTP, NASA-TLX, and HRV, with respect to pilot

performance, workload, and stress level. Ten subjects participated in this repetitive

experiment. They were divided into two groups, with respect to their FE, and executed four

scenarios, which consisted of the treatment combinations of GA and AA. Upon completion of

the experiment, collected data were graphically and statistically analyzed. As a result of

repeated-measurements-univariate ANOVA, the following results were revealed.

Firstly, one of the flight factors examined was the level of AA, and its hypothesis was

that the type of AA does affect pilot performance, workload, and stress level. Since the

hypothesis was true, it can be concluded that the levels of AA have a significant impact on

pilot performance, workload, and stress level. This effect was more recognizable in the flight

that flew over Populated area. Secondly, the levels of GA also had significant impact on pilot

performance, workload, and stress level, as shown in the results of hypothesis test. The pilots

showed lower performance, higher workload, and higher stress at 4.5º than that of 3º. Finally,

for the level of FE, a group of student pilots, who had less flight experiences, scored higher

69

HRV, NASA-TLX, but lower performance than a group of instructor pilots, who had more

flight experiences. However, the difference of mean value on the measurements was not

statistically significant. The hypothesis test results showed that the level of flight experience

did not have significant impact on the pilot performance, workload, and stress levels. These

results are discussed in detail in the following sections.

5.1 Types of Approach Area

The statistical test and graphical results show that the types of Approach Area (AA,

Non-Populated and Populated area) have an effect on the pilot performance, workload, and

stress level. It seems like the programmed Populated area influenced pilot landing

performance score, NASA-TLX score, and HRV-LF ratio. As an environmental factor on

workload and stress, many objects programmed for the generating Populated area influence

pilots by increasing the density of the pilot‟s visual information. And pilots considered the

flight over Populated area as a new experience and felt stress from the task itself.

For the pilot performance, the pilots recorded the worst score for Populated area on

both LP and AGPT as show in Table 3 and Figures 6 and 7. In general, experience is a very

important variable moderating human performance (Salthouse, 1987) and undoubtedly a

major factor leading to greater effectiveness in performance (Shriver, 1953). So, the flight

70

over Populated area became inexperienced work which caused lower performance, higher

workload, and higher stress level than the flight over Non-Populated area, which the pilots

are used to.

Many objects at Final Approach to L/D phase may make the pilots fly higher than the

established glide path at the airport. Thus, flight performance got worse during the flight over

the Populated area than Non-Populated area. The density of visual clues can affect pilots‟

performance at Final Approach to L/D phase. Gibb and Gray (2006) said that the global and

local terrain features allow a pilot to accurately judge the glide path. According to Foyle et al.

(1992), visible texture density and its changes have been proposed as a clue for altitude

awareness, with increasing density associated with higher perceived altitudes. This was

supported by Oshima and Lintern (1992), who found that the pilots flew progressively higher

approaches as the density of the visual information (objects) increased. Similar results were

reported by Lintern and Walker (1991), Lirdan and Koonce (1991), and Reardon (1988).This

simulation resulted in the highest AGPT score for the Populated area.

NASA-TLX results showed that the flight over Populated area increased in pilots‟

subjective workload level. In accordance with TLX results, HRV showed that the pilots who

flew over Populated area were exposed to more stress than those who flew over Non-

Populated area. Tyler and Cushway (1995) reported that there is a significant and direct

71

proportional link between workload and stress levels. For example, Sheridan (1979)

explained workload as the stress that people experience from the various stressors such as

ambient (environmental) and performance. In 1983, Hockey stated that stress is usually

considered to be a property of the environment, all other individual factors being equal. The

traditional approach has investigated both noise and temperature as environmental sources of

stress (Hancock & Warm, 1989). In this study, many buildings were programmed in

Populated area. Therefore, it may play a role as an environmental stressor. From this, pilots

can experience stress from the possibility of unexpected accidents while flying over

Populated area.

Moreover, many research reported that the pilot workload varies on the flight task type

and environments. For example, the pilots who fly under IFR showed significantly higher

workload (Wilson, 2002b; Hankins & Wilson, 1998; Hasbrook & Rasmussen, 1970). It was

claimed that IFR flight is a difficult cognitive task resulting in the increased workload. Based

on the study result of Bennett and Schwirzke (1992), the aviation accident rate for IFR flight

during the nighttime was almost eight times greater than that for VFR flight during the

daytime. It can be said that pilot workload, caused by environmental stressor, decreases the

flight performance.

72

Moreover, pilots may feel stress from the flight over Populated area. According to

Hancock and Warm (1989), the task itself is considered as a primary influence in the

generation of stress from the integrated view of stress and performance.

Overall, it can be suggested that both the new work conditions (environmental factors)

and the tasks themselves were strongly related to increase in pilot workload and stress.

Therefore, increased workload and stress may cause the different flight performance

depending on the levels of AA.

5.2 Levels of Gliding Angle

The levels of Gliding Angle (GA, 3 degree and 4.5 degree) significantly affected the

pilot performance, workload, and stress level. Pilots showed lower performance, higher

workload, and higher stress levels for the flight with 4.5 degree than with 3 degree GA. It

seems like 4.5 degree ILS task was new experience task, difficult task, and the task that

caused time pressure to the pilots.

Usually, the pilots have to follow the ILS procedure that is established with 3º GA to

land on the runway. Unlike this 3º ILS, the task programmed in this study was to do 4.5° ILS,

which was a new experience for both participant groups. Accordingly, the pilots who flew

with 4.5º ILS became novices in the tasks. However, the pilots who flew with 3º ILS could be

73

an expert for these tasks. In 1988, Chi et al. stated that experts were faster and error free than

novices at performing skills related to the tasks. While novices will use effort and time

consuming knowledge-based-behaviors on the tasks, experts will be able to aggregate these

behaviors into schemas which can then be enacted at a skill based level (Rasmussen, 1983;

Stanton et al., 2009). As a result, there is a difference on the cognitive capacity that affects

decision making. This is supported by Boehm-David et al. (2007) who reported that the pilot

performance obtained from simulated IFR flight decreased gradually as the degree of GA

increased. From Table 4 and Figure 6, the lower pilot performance on AGPT is recorded by

3º ILS. AGPT performance of the pilots was highly affected by density of visual clues

(buildings) at Final Approach to L/D phase as explained in Section 5.1. When the pilots fly

with 4.5º ILS, the effect of visual clues may be less on the flight.

According to pilot performance, 4.5º ILS is considered a difficult task for the pilots in

this study. It is verified by the NASA-TLX results: the pilots rated higher subjective

workload level for the 4.5º ILS. As reported by Boehm-Davis et al. (2007), the participants‟

ratings of subjective workload obtained from NASA-TLX data increased gradually as the

degree of GA increased. Various studies also showed that the NASA-TLX is sensitive to

changes in mental workload (Collet et al., 2009; DiDomenico & Nussbaum, 2008). Figure 8

shows that 4.5º ILS task generated more stress, which means that the pilots consider 4.5º ILS

74

as the difficult task. According to Cnossen et al. (2004) and Vicente et al. (1987), an increase

in task demands (difficulty) is accompanied by an increase in mental workload. Again, Tyler

and Cushway (1995) reported that there is a significant and direct proportional link between

workload and stress levels. In other words, it can be said that mental stress level increases as

degree of task difficulty increases. Many researchers reported the increase of a HRV-LF

volume as an increase in the difficulty of task. For example, Paritala (2009) also reported a

HRV-LF increase for the mental task compared to that for physical task. Seong et al. (2004)

reported that a HRV-LF increased for the mental task. Malliani et al. (1991) reported that

mental stress, induced by arithmetic calculation, increased a HRV-LF.

Another possible reason for increase of stress level is time pressure. Hendy et al. (2001)

stated that time pressure affected operator performance, error production, and judgment of

workload from the information-processing model. Hence, time pressure is claimed as a

principal stressor in the human information-processing context. For example, Andrew and

Parris (1972) reported that the productivity of NASA scientists and engineers is highly

affected by time pressure. In this study, the distance from the touchdown point on the runway

to decision height (DH, the altitude that pilots decide whether they continue to land) was

getting shorter for the 4.5º ILS. In the cockpit, the pilots pay attention to many important

tasks (e.g. to communicate with air traffic control, to supervise instruments, to look outside of

75

cockpit, etc.) to make safe flight at any given time. The pilots who are at the most loaded

phase of flight may be busier because the total time that the pilots can spend to land on the

runway is reduced for 4.5º ILS task. This study‟s results support those reports mentioned

above.

5.3 Levels of Fight Experience

The levels of Flight Experience (FE, less-experienced (student) and experienced

(instructor) pilots) did not influence pilot performance, workload, and stress level. It seems

that the differences of performance, workload, and stress level between groups were reduced

by the effect of both the new work conditions (Populated approach area and 4.5 degree ILS)

and the flight done at the most loaded phase of flight in this study.

As shown in Figures 14 and 15, IPG recorded a higher score than SPG for the LP and

AGPT scores. However, the difference between both groups‟ score was not statistically

significant. The new work conditions of GA and AA could be applied for both groups. In

general, experience is a very important variable that moderates human performance

(Salthouse, 1987) and undoubtedly it is a major factor leading to greater effectiveness in

performance (Shriver, 1953). Mostly, the experts show the better performance than the

76

novices (Stokes et al., 1997; Wiggins & O‟Hare, 1995; Fisk et al., 1992; Livingston & Borko,

1989).

However, this study showed that the levels of experience didn‟t affect the pilot

performance. In 1988, Chi et al. stated that experts are faster and error free than novices at

performing skills related to the tasks. While novices will use effort and time consuming

knowledge-based-behaviors on the tasks, experts will be able to aggregate these behaviors

into schemas which can then be enacted at a skill based level (Rasmussen, 1983; Stanton et

al., 2009). As a result, there is a difference in the cognitive capacity affecting decision

making.

Additionally, the flight at Final Approach to L/D phase, which is the most loaded

phase of flight, increases that pilot workload (Hankins and Wilson, 1998; Sekiguchi et al.,

1978; Opmeer & Krol, 1973). In this study, new work conditions may increase workload on

both groups at Final Approach to L/D phase. Moreover, this simulation flight was supposed

to be done under the most loaded phase of flight. Tyler and Cushway (1995) reported that

there is a significant and positive link between workload and stress levels. Therefore, FE

made small difference in statistics on performance, workload, and stress level, although SPG

recorded lower performance and represented the higher subjective workload rating, as well as

mental stress level.

77

CHAPTER SIX: COCLUSIONS

6.1 Conclusions

The study investigated the overall effect of flight factors on the pilot performance,

workload, and stress level. The results showed that the pilots recorded lower performance,

higher workload, and higher stress level when they were flying both at 4.5º and over

Populated area (e.g. large city) than both at 3º and over the Non-Populated area (e.g. grass

field, river, etc.). However, the pilot performance, workload, and stress level between

experienced pilots and novice pilots were not significantly different for this experiment.

Based on these results, the following issues can be considered.

The design process of developing new procedures (LDLP, SCDA, Steep Approach, etc.)

for the landing of aircrafts should be considered from the aspects of pilot workload, safety,

passenger comfort, and economy. However, the economy factor is likely to be considered

first, even over safety and pilot workload. In particular, the workload of the crew in the

cockpit is very high since they need to change configuration, decelerate speed, and flare to

land on the runway. Generally, these new procedures require increased height, decreased

thrust (power), and delayed configuration to help reduce fuel consumption and noise

problems. However, this would increase pilot workload at Final Approach to L/D phase

78

because pilots would be busier by attempting steeper approach that makes the shorter Final

Approach to L/D segment.

There are obvious safety issues related to the allowance of structures in the

navigational airspace, particularly at Final Approach to L/D area. However, economic

development desires of land owners and developers to expand and provide new housing

facilities are unlimited. Structures encroaching upon the area need to remain clear for safe

operation of aircraft because development of airport areas can cause problems by

constructing many skyscrapers that cause steeper approach for the aircraft. Therefore,

development at Final Approach to L/D segment needs to be limited since it increases pilot

workload and causes safety concern based on the results of this study.

Conclusively, in order to reduce the workload and stress that can cause human errors, it

is highly recommended to carefully examine the impact of new flight procedures on pilot

workload and stress before they are implemented. In addition, safety issues related to

development in the navigational airspace should be considered in terms of pilot performance,

workload, and stress, although, in this study, all pilots made a safe landing under simulated

flight factors such as Populated area and 4.5 degree ILS.

The study provides useful information for researchers interested in the performance,

workload, and stress in aviation, but the results themselves are important. In particular, for

79

the risk and safety management to improve aviation safety, these results may be useful to

policy makers in developing policy on new approach procedures such as Steep Approach by

providing reference data. In addition, the results help to manage navigational airspace of

airports in big cities by showing the effect of obstacles on pilots who fly at Final Approach to

L/D segment.

6.2 Findings, Limitations, and Suggested further research

Findings from this study are:

(1) Pilot‟s workload was affected by the levels of gliding angle and approach area.

(2) Pilots kept higher glide path when they were flying over Populated area (e.g. large

city) than over Non-Populated area (e.g. country town).

(3) Low Frequency band in HRV is sensitive to the consequence of sympathetic

activity (rest-tilt maneuver, mental stress, hemorrhage, coronary occlusion, etc.).

(4) The individual rating scale, Mental Demand (MD) of NASA-TLX, corresponds

well to pilot‟s workload indication.

It is important to note that these results may be restricted to the conditions examined in

this study for the following reasons. First, methodological limitations (e.g., the artificial and

controlled airspace situation) have to be considered. In the presence of such stressful real-

80

world factors (e.g. other traffic, communications with air traffic controller, etc.), pilots may

be more likely to experience severe stress from heightened risk, causing increased workload.

Second, participants began each simulation when the aircraft was already on the descending

approach angle. Allowing the pilots to capture the glide path from an “en-route” phase of

flight may produce different results. Finally, these results are limited to the similar category

of aircraft, Cessna 172 R. Medium and heavy category aircrafts may show differences in

results since those have different capabilities and characters in flight. All of these conditions

should be examined in future research before a solid conclusion can be made.

For enhancing the validity of the results, further studies can be suggested by

performing this research and surveying the literature. The suggested further studies are:

(1) To use many subjects, including female pilots, to enhance the validity of the

research. It is worldwide trend that aviation field allow female pilots.

(2) To include other factors to enhance the reliability of the result. For example, these

factors might include a night time flight, various weather condition, etc.

(3) To extend the experiment from the beginning of the flight to the completion of the

flight in order to strengthen the validity of the study.

(4) To do the experiment with different aircrafts such as carrier aircrafts, fighter jets,

and helicopters.

81

APPENDIX A: UCF CONSENT FORM

82

83

APPENDIX B: APPROVAL OF HUMAN RESEARCH

84

APPENDIX C: SUBJECTIVE QUESTIONNAIRE (NASA-TLX)

85

86

87

88

APPENDIX D: COLLECTED DATA

Data Set #1

89

Data Set #2

90

Data Set #3

91

APPENDIX E: STATISTICAL TEST RESULTS

92

93

94

95

REFERENCES

Ahmadi, K. & Alireza, K. (2007). Stress and job satisfaction among Air Force military pilots.

Journal of Social Sciences, 3(3), 159-163.

Airbus. (2006). A318 – Steep approach operation. Retrieved from

http://www.smartcockpit.com/.../A318_Steep_Approach_Operations.pdf.

Akselrod, S., Golden, D. S., Ubel, F. A., Shannon, D. C. & Cohen, R. J. (1981). Power

spectrum analysis of heart rate fluctuation: A quantitative probe of beat to beat

cardiovascular control. Science, 213(4504), 220-222.

Andrew, F. M., & Farris, G. F. (1972). Time pressure and performance of scientists and

engineers: A five year panel study. Organizational Behavior and Human Performance,

8(2), 185-200.

Bainbridge, L. (1974). The adaptation of cognitive processes to task demands and mental

capacity: Some reasons for the lack of correlation between objective and subjective mental

workload. Le Travail Humain, 37 (2), 279-302.

Battiste, V., & Bortolussi, M. (1988). Transport pilot workload: A comparison of two

subjective techniques. Proceedings of the Human Factors and Ergonomics Society 32nd

Annual Meeting (pp. 150-154). Santa Monica, CA: Human Factors Society and Ergonomic

Society.

Bennett, C. T., & Schwirzke, M. (1992). Analysis of accidents during instrument approaches.

Aviation, Space, and Environmental Medicine, 63(4), 253-261.

Boehm-Davis, D. A., Casali, J. G., Kleiner, B. M., Lancaster, J. A., Saleem, J. J., &

Wochinger, K. (2007). Pilot performance, strategy, and workload while executing

approaches at steep angles and with lower landing minima. Human Factors, 49(5), 759-

772.

Borg, G. A. (1982). Psychophysical bases of perceived exertion. Medicine and Science in

Sports and Exercise, 14(5), 377-381.

96

Brotheridge, C. M. (2001). A comparison of alternative models of coping: Identifying

relationships among coworker support, workload, and emotional exhaustion in the

workplace. International Journal of Stress Management, 8(1), 1-14.

Byers, J. C., Bittner, A. C., & Hill, S. G. (1989). Traditional and raw Task Load Index (TLX)

correlations: Are paired comparisons necessary? In A. Mital (Ed.), Advances in industrial

ergonomics and safety: 1 (pp.481-485). London, UK: Taylor & Francis.

Calvert, E. S. (1954). Visual judgments in motion. The Journal of Navigation, 7(3), 233-251.

Carlisle, D. (2001). Pilots stress and what to do about it. Business and commercial aviation,

89(5), 92-97.

Cerutti, S., Bianchi, A. M., & Mainardi, L. T. (1995). Spectral analysis of the heart rate

variability signal. In M. Malik, A. J. Camm (Eds), Heart Rate Variability (pp.63-74)

Armonk, NY: Futura Publishing company, Inc.

Chi, M. T. H., Glaser, R., & Farr, M. J. (1988). The Nature of Expertise. Hillsdale, New

Jersey: Lawrence Erlbaum Associate Publishers.

Christ, R. E., Hill, S. G., Byers, J. C., Lavecchia, H. M., Zaklad, A. L., & Bittner A. C.

(1993). Applications and validation of workload assessment techniques (Tech. Rep. No.

974). Alexandria, VA: U.S. Army Research Institute for the Behavioral and Social

Sciences.

Cnossen, F., Meijman, T., & Rothengatter, T. (2004). Adaptive strategy changes as a function

of task demands: A study of car drivers. Ergonomics, 47(2), 218-236.

Collet, C., Averty, P., & Dittmar, A. (2009). Autonomic nervous system and subjective

ratings of strain in air-traffic control. Applied Ergonomics, 40(1), 23-32.

Cooper, G. E., & Harper, R. P. (1969). The use of pilot rating in the evaluation of aircraft

handling qualities (NASA TN-D-5153). Moffett Field, CA: NASA Ames Research Center.

Corlett, E. N. & Bishop, R. P. (1976). A technique for assessing postural discomfort.

Ergonomics, 19(2), 175-182.

97

Corwin, W. H., Sandry-Garza, D. L., Biferno, M. H., Boucek, G. P., Logan, A. L., Jonsson, J.

E., & Metalis, S. A. (1989). Assessment of crew workload measurement methods,

techniques, and procedures: Process, methods and results (Report WRDC-TR-89-7006).

Wright-Patterson Air Force Base, OH: Wright Research and Development Centre, Air

Force Systems Command.

Cox-Fuenzalida, L. E., Swickert, R., & Hittner, J. B. (2004). Effects of neuroticism and

workload history on performance. Personality and Individual Differences, 36(2), 444-456.

Cox-Fuenzalida, L. E. (2007). Effect of workload history on task performance. Human

Factors, 49(2), 277-291.

Cuevas, H. (2003). The pilot personality and individual difference in the stress response.

Proceedings of the Human Factors and Ergonomics Society 47th Annual Meetings

(pp.1092-1096). Santa Monica, CA: Human Factors and Ergonomics Society.

Cumming, R. W., & Croft, P. G. (1973). Human information processing under varying task

demand. Ergonomics, 16(5), 581-586.

Dahlstrom, N., & Nahlinder, S. (2006). A comparison of two recorders for obtaining in-flight

heart rate data. Applied Psychophysiology and Biofeedback, 31(3), 273-279.

DiDomenico, A., & Nussbaum, M. A. (2008). Interactive effects of physical and mental

workload on subjective workload assessment. International Journal of Industrial

Ergonomics, 38(11-12), 977-983.

Dussault, C., Jouanin, J. C., Philippe, M., & Guezennec, C. Y. (2004). EEG and ECG

changes during selected flight sequences. Aviation, Space, and Environmental Medicine,

75(10), 889-897.

Dussault, C., Jouanin, J. C., Philippe, M., & Guezennec, C. Y. (2005). EEG and ECG

changes during simulator operation reflect mental workload and vigilance. Aviation, Space,

and Environmental Medicine, 76(4), 344-351.

98

Edens, E. S. (1992). Individual differences underlying pilot cockpit error (Doctorial

dissertation, George Mason University). Retrieved from

http://oai.dtic.mil/oai/oai?verb=getRecord&metadataPrefix=html&identifier=ADA236107.

Elmenhorst, E. M., Vejvoda, M., Maass, H., Wenzel, J., Plath, G., Schubert, E., & Basner, M.

(2009). Pilot workload during approaches: Comparison of simulated standard and noise-

abatement profiles. Aviation, Space, and Environmental Medicine, 80(4), 364-370.

Farmer, E. W., van Rooij, J., Riemersma, J., Jorna, P., & Moraal, J. (2000). Handbook of

Simulator-Based Training. Brookfield, VT: Ashgate Publishing Limited.

Fauvel, J. P., Cerutti, C., Quelin, P., & Laville, M. (2000). Mental stress-induced increase in

blood pressure is not related to baroreflex sensitivity in middle-aged healthy men. Journal

of Hypertension, 35(4), 887-891.

Federal Aviation Administration. United States standard for terminal instrument procedures

(TERPS). Retrieved from

http://www.faa.gov/regulations_policies/orders_notices/index.cfm/go/document.informatio

n/documentID/11698.

Fitts, P. M. (1966). Cognitive aspects of information processing: III. Set for speed versus

accuracy. Journal of Experimental Psychology, 71(6), 849-857.

Fisk, A. D., & Rogers, W. A. (1992). The application of consistency principles for the

assessment of skill development. In J. W., Regian, & V. J., Shute (Eds), Cognitive

approaches to automated instruction (pp 171-194). Hillsdale, NJ: Lawrence Erlbaum

Associates.

Foyle, D. C., Kaiser, M. K., & Johnson, W. W. (1992). Visual cues in low-level flight:

Implication for pilotage, Training, Simulation, and Enhanced/Synthetic vision system.

From American Helicopter Society 48th Annual Forum (pp. 253-260).

Frasca international, Inc, 2010, Qualification approval guide document (No. CERT 144253

Rev.B).

99

Gawron, V. J., Schiflett, S. G., & Miller, J. C. (1989). Measures of in-flight workload. In R.

S., Jensen (Ed.), Aviation psychology (pp. 240-287). Aldershot, Brookfield: Gower

Technical.

Gibb, R., & Gray, R. (2006). Terrain orientation theory for a visual approach to landing.

Proceedings of the Human Factors and Ergonomics Society 50th Annual Meeting. Santa

Monica, CA: Human Factors and Ergonomics Society.

Goldberg, R. A., & Stewart, M. R. (1980). Memory overload or expectancy effect?

„Hysteresis‟ revisited. Ergonomics, 23(12), 1173-1178.

Gopher, D. & Donchin, E. (1986). Workload-An examination of the concept. In K. R., Boff,

L., Kaufman, & J. P., Thomas (Eds.), Handbook of Perception and Human

performance,II(pp.41-1 to 41-49). New York, NY: Wiley.

Gundel, A., Polyakov, V. V., & Zulley, J. (1997). The alteration of human sleep and

circadian rhythms during space flight. Journal of Sleep Research, 6(1), 1-8.

Hancock, P. A. & Warm, J. S. (1989). A dynamic model of stress and sustained attention.

Human Factors, 31(5), 519-537.

Hankins, T. C., & Wilson, G. F. (1998). A comparison of heart rate, eye activity, EEG and

subjective measures of pilot mental workload during flight. Aviation, Space, and

Environmental Medicine, 69(4), 360-367.

Hart, S. G., & Hauser, J. R. (1987). In-flight application of three pilot workload measurement

techniques. Aviation, Space, and Environmental Medicine, 58(5), 402-410.

Hart, S. G., & Staveland, L. E. (1988), Development of NASA-TLX (Task Load Index):

Results of Empirical and Theoretical Research. In P. A., Hancock, N., Meshkati (Eds.),

Human mental workload (pp. 139-183). New York, NY: Elsevier.

Hasbrook, A. H., & Rassmussen, P. G. (1970). Pilot heart rate during in-flight simulated

instrument approaches in a general aviation aircraft. Aerospace Medicine, 41(10), 1148-

1152.

100

Hatton, C., Brown, R., Caine, A., & Emerson, E. (1995). Stressors, coping strategies, and

stress related outcomes among direct care staff in staffed houses for people with learning

disabilities. Mental Handicap Research, 8(4), 252-271.

Heinrich, H. W., Petersen, D., & Roos, N. (1931). Industrial accident prevention: A safety

management approach (1st ed.). New York, NY: McGraw-Hill.

Hendy, K. C., Farrell, P. S. E., & East, K. P. (2001). An information-processing model of

operator stress and performance. In P. A., Hancock, & P. A., Desmond (Eds.), Stress,

Workload, and Fatigue (pp.34-80). Mahwah, NJ: Lawrence Erlbaum Associate.

Henn, S. M. (1996), MAI 34/01 p.4317, Pacific Lutheran University.

Hilburn, B., Jorna, P. G. A. M., & Parasuraman, R. (1995). The effect of advanced ATC

automation on mental workload and monitoring performance: An empirical investigation

in Dutch airspace. Proceedings of the 8th International Symposium on Aviation Psychology,

1, 387-391. Columbus, OH: The Ohio State University.

Hill, S. G., Lavecchia, H. P., Bittner, A. C., Byers, J. C., Zaklad, A. L., & Christ R. E., (1992).

Comparison of four subjective workload rating scales. Human Factors, 34(4), 429-439.

Hockey, R. (1983). Stress and fatigue in human performance. Chichester, New York: Wiley.

Howell, D. C. (1987). Statistical Methods for Psychology (2nd ed.). Boston, MA: Duxbury

Press.

Humphrey, J. H. (1998). Job Stress. Needman Heights, MA: Allyn & Bacon.

International Civil Aviation Organization. (2004). Outlook for air transport to the year 2015.

Retrieved from http://www.icao.int/icao/en/atb/fep/C304_Outlook_En.pdf.

International Civil Aviation Organization. (2010). Annual report of the council(2009).

Retrieved from http://www.icao.int/annualreports/.

Jamal, M. (1985). Relationship of job stress to job performance: A study of managers and

Blue-Collar workers. Human Relations, 38(5), 409-424.

101

Johnson, W. W., & Phatak, A. V. (1989). Optical variables and control strategy used in a

visual hover task. Proceeding of the IEEE International Conference on Systems, Man and

Cybernetics (pp. 719-724). Cambridge, MA: Institute of Electrical and Electronics

Engineers.

Jorgensen, A. H., Garde, A. H., Laursen, B., & Jensen, B. R. (1999). Applying the concept of

workload to IT work. From the Virtual Proceedings of CybErg: The Second International

Virtual Conference on Ergonomics. Perth, Australia: International Ergonomics

Association/Curtin University of Technology.

Kakimoto, Y., Nakamura, A., Tarui, H., Nagasawa, Y., & Yagura, S. (1988). Crew workload

in JASDF C-1 transport flights: 1. Change in heart rate and salivary cortisol. Aviation,

Space, and Environmental Medicine, 59(6), 511-516.

Kamath, M. V., & Fallen, E. L. (1993). Power spectral analysis of heart rate variability: A

non-invasive signature of cardiac autonomic functions. Critical Reviews in Biomedical

Engineering, 21(3), 245-311.

Kanki, B. G. (1996). Stress and aircrew performance: A team-level perspective. In J. E.,

Driskell, & E., Salas (Eds.), Stress and human performance (pp. 127-162). Mahwah, NJ:

Lawrence Erlbaum Associates.

Kantowitz, B. H., & Sorkin, R. D. (1983), Human factors: understanding people-system

relationships. NewYork, NY: Wiley.

Kantowitz, B. H. (1987). Mental workload. In P. A. Hancock (Ed.), Human Factors

Pshychology (pp. 81-121). New York, NY: Elsevier Science Pub. Co.

Kantowitz, B. H., & Casper, P. A. (1988). Human workload in aviation. In E. L., Wiener, &

D. C., Nagel (Eds.), Human factors in aviation (pp. 157-162). San Diego, CA: Academic

Press.

Katz, L. C. (1997). Assessment of aircrew stress (Report No. 97-37). Fort Rucker, Alabama:

U.S. Army Aeromedical Research Laboratory.

102

Kirmeyer, S. L., & Dougherty, T. W. (1988). Workload, tension, and coping: Moderating

effects of supervisor support. Personnel Psychology, 41(1), 125-139.

Kitney, R. & I., Rompelman, O. (1977). Analysis of the interaction of the human blood

pressure and thermal systems. In J., Perkins (Ed.), Biomedical Computing (pp. 49-54).

London, UK: Pitman Medical.

Kraft, C. L. (1978). A psychophysical contribution to air safety: Simulator studies of visual

illusions in night visual approaches. In: H. L., Pick, H. W., Leibowitz, J. E., Singer, A.,

Steinschneider., & H. W., Stevenson (Eds.). Psychology: From Research to Practice

(pp.363-385). New York, NY: Plenum Press.

Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, NY:

Springer.

Lewis, M. F., & Mertens, H. W. (1979). Pilot performance during simulated approaches and

landings made with various computer-generated visual glide path indicators. Aviation,

Space, and Environmental Medicine, 50(10), 991-1002.

Lindholm, E., & Cheatham, C. M. (1983). Autonomic activity and workload during learning

of a simulated aircraft carrier landing task. Aviation, Space, and Environmental Medicine,

54(5), 435-439.

Lintern, G., & Walker, M. B. (1991). Scene content and runway breadth effects on simulated

landing approaches. The International Journal of Aviation Psychology, 1(2), 117-132.

Lirdan, G., & Koonce, J. M. (1991). Display magnifications for simulated landing approaches.

The International Journal of Aviation Psychology, 1(1), 59-72.

Livingston, C., & Borko, H. (1989). Expert-novice differences in teaching: A cognitive

analysis and implications for teacher education. Journal of Teacher Education, 40(4), 36-

42.

103

Lysaght, R. J., Hill, S. G., Dick, A. O., Plamondon, B. D., Wherry, R. J., Zaklad, A. L., &

Bittner, A. C. (1989). Operator workload: Comprehensive review and evaluation of

operator workload methodology (ARI Technical Report 851). Fort Bliss, TX: US Army

Research Institute, Field unit.

Magnusson, S. (2002). Similarities and differences in psychophysiological reactions between

simulated and real air-to-ground missions. The International Journal of Aviation

Psychology, 12(1), 49-61.

Malliani, A., Pagani, M., Lombardi, F., & Cerutti, S. (1991). Cardiovascular neural regulation

explored in the frequency domain. Circulation, 84, 482-492.

McCormick, E. J., & Sanders, M. S. (1993). Human factors in Engineering and Design (7th

ed.). New York: McGraw-Hill, Inc.

Meglino, B. M. (1977). Stress and performance: Are they always incompatible? Supervisory

Management, 22(3), 2-12.

Mertens, H. W. (1978). Comparison of the visual perception of a runway model in pilots and

non-pilots during simulated night landing approach. Aviation, Space, and Environmental

Medicine, 49(9), 1043-1055.

Meshkati, N., Hancock, P. A., Rahimi, M., & Dawes, S. M. (1995). Techniques in mental

workload assessment. In J. R. Wilson, & E. N., Corlett (Eds.), Evaluation of Human Work:

a practical ergonomics methodology (2nd ed.)(pp.749-782). Bristol, PA: Taylor & Francis

Ltd.

Moray, N. (1988). Mental workload since 1979. International Review of Ergonomics, 2, 123-

150.

Moray, N., Dessouky, M. I., Kijowski, B. A., & Adapathya, R. (1991). Strategic behavior,

workload, and performance in task scheduling. Human Factors, 33, 607-629.

104

Moroney, W. F., Biers, D. W., Eggemeier, F. T., & Mitchell, J. A. (1992). Comparison of two

scoring procedures with the NASA Task Load Index in a simulated flight task.

Proceedings of the IEEE National Aerospace Electronics Conference (pp. 734-740).

Dayton, OH: Institute of Electrical and Electronics Engineers.

Moroney, W. F., Biers, D.W., & Eggemeier, F. T. (1995). Some measurement and

methodological considerations in the application of subjective workload and measurement

techniques. The International Journal of Aviation Psychology, 5(1), 87-106.

Murai, K., Hayashi, Y., Nagata, N., & Inokuchi, S. (2004). The mental workload of a ship's

navigator using heart rate variability. Interactive Technology and Smart Education, 1(2),

127 - 133.

Muse, L., Harris, S. & Field, H. (2003). Has the inverted-U theory of stress and job

performance had a fair test? Human Performance, 16(4), 349-364.

Nagel, D. C. (1988). Human error in aviation operations, In E. L., Wiener, & D. C., Nagel

(Eds.), Human Factors in Aviation (pp.263-303). San Diego, CA: Academic Press.

Nataupsky, M., & Abbott, T. S. (1987). Comparison of workload measures on computer-

generated primary flight displays. Proceedings of the Human Factors and Ergonomic

Society 31st Annual Meeting (pp. 548-552). Santa Monica, CA: Human Factors and

Ergonomic Society.

National Aeronautic and Space Administration. (2004). Small aircraft transportation system.

Retrieved from http://www.nasa.gov/centers/langley/news/factsheets/SATS.html.

National Transportation Safety Board, (1998 to 2005), Annual review of aircraft accident

data: US air carrier operations. Retrieved from http:// www.ntsb.gov/Publictn/A_stat.htm.

Nygren, T. E. (1991). Psychometric properties of subjective workload measurement

techniques: Implications for their use in the assessment of perceived mental workload,

Human Factors, 33(1), 17-33.

O'Hare, D., Wiggins, M., Batt, R., & Morrison, D. (1994). Cognitive failure analysis for

aircraft accident investigation. Ergonomics, 37(11), 1855-1869.

105

Opmeer, C. H., Krol, J. P. (1973). Towards an objective assessment of cockpit workload. I.

Physiological variables during different flight phases. Aerospace Medicine, 44(5), 527-532.

Orasanu, J. M., & Backer, P. (1996). Stress and Military Performance. In J. E., Driskell, & E.,

Salas (Eds.), Stress and Human Performance (pp. 89-125). Mahwah, NJ: Lawernce

Erlbaum Associates.

Oshima, C. R., & Lintern, G. (1992). Effects of scene content and horizon on glide slope

control. Proceeding of the IEEE International Conference on Systems, Man, and

Cybernetics. Chicago, IL: Institute of Electrical and Electronics Engineers.

Pagani, M., Lombardi, F., Guzzetti, S., (1986). Power spectral analysis of heart rate and

arterial pressure variabilities as a marker of sympatho-vagal interaction in man and

conscious dog. Circulation Research, 59(2), 178-193.

Pagani, M., Furlan, R., Pizzinelli, P., Crivellaro, W., Cerutti, S., & Malliani, A. (1989).

Spectral analysis of R-R and arterial pressure variabilities to assess sympatho-vagal

interaction during mental stress in humans. Journal of Hypertension, Supplement, 7(6), 14-

15.

Palmisano, S., & Gillam, B. (2005). Visual perception of touchdown point during simulated

landing. Journal of Experimental Psychology: Applied, 11(1), 19-32.

Paritala, S. A. (2009). Effects of physical and mental tasks on heart rate variability (Master

Thesis, Louisiana State University). Retrieved from http://etd.lsu.edu/docs/available/etd-

08222009-204359/unrestricted/thesis.pdf.

Penaz, J., Roukenz, J., & Van, D. W. H. J. (1968). Spectral analysis of some spontaneous

rhythms in the circulation. In H., Drischel, & N., Tiedt (Eds.), Biokybernetik (pp.233).

Leipzig, Germany: Karl Marx University.

Petersen, D. (1996). Human-error reduction and safety management (3rd ed.). New York,

NY: Van Nostrand Reinhold.

Polar Electro. (2010). S810i manual. Retrieved from

http://www.polar.fi/en/support/product_support?product=3126.

106

Rasmussen, J. (1983). Skills, rules, and knowledge; signals, signs, and symbols, and other

distinctions in human performance models. IEEE Transactions on Systems, Man, and

Cybernetics, 13, 257-266.

Reardon, K. A. (1988). The effects of nested texture on a landing judgment task. Proceedings

of the Human Factors and Ergonomic Society 32nd Annual Meeting (pp. 10-14). Santa

Monica, CA: Human Factors Society.

Reid, G. B., Shingledecker, C. A., & Eggemeier, F. T. (1981). Application of conjoint

measurement to workload scale development. Proceedings of the Human Factors Society

25th Annual Meeting (pp.522-525). Santa Monica, CA: Human Factors Society.

Reason, J. (1990). Human Error. Cambridge, UK: Cambridge University Press.

Rob, E. (2001). Steep approach at London city airport. Retrieve from

http://www.storksmt.com/downloads/Aerospace/FS/.../6-2104142/PAG_22.PDF.

Roman, J. A. (1965). Risk and responsibility as factors affecting heart rate in test pilots:

Flight research program 2. Aerospace Medicine, 36, 518-523.

Roscoe, A. H. (1975). Heart rate monitoring of pilots during steep-gradient approaches.

Aviation, Space, and Environmental Medicine, 46(11), 1410-1413.

Roscoe, A. H. (1993). Heart rate as a psychophysiological measure for in-flight workload

assessment. Ergonomics, 36, 1055-1062.

Salthouse, T. A. (1987). The role of experience in cognitive aging. Annual review of

gerontology and geriatrics, 7, 135-158.

Sanders, A. F. (1983). Towards a model for stress and human performance. Acta

Psychologica, 53, 61-97.

Sarno, K., Wickens, C. D. (1995). The role of multiple resources in predicting time-sharing

efficiency: An evaluation of three workload models in a multiple task setting. International

Journal of Aviation Psychology, 5(1), 107-130.

107

Sato, N., Miyake, S., Akatsu, J., & Kumashiro, M. (1995). Power spectral analysis of heart

rate variability in healthy young women during the normal menstrual cycle. Psychosomatic

Medicine, 57(4), 331-335.

Sauvet, F., Jouanin, J. C., Langrume, C., Van Beers, P., Papelier, Y., & Dussault, C. (2009).

Heart Rate Variability in novice pilots during and after a multi-leg cross-country flight.

Aviation, Space, and Environmental Medicine, 80(10), 862-869.

Sayers, B. M. (1973). Analysis of heart rate variability. Ergonomics, 16, 17-32.

Schvaneveldt, R. W. (1969). Effects of complexity in simultaneous reaction-time tasks.

Journal of Experimental Psychology, 81, 289-296.

Schvaneveldt, R. W., Reid, G. B., Gomez, R. L., & Rice, S. (1998). Modeling mental

workload (AFRL-HE-WP-SR-2000-1000). Wright Patterson AFB, OH: US Airforce

Research Laboratory Human Effectiveness Directorate Crew System Interface Division.

Sekiguchi, C., Handa, Y., Gotoh, M., Kurihara, Y., Nagasawa, A., & Kuroda, I. (1978).

Evaluation method of mental workload under flight condition. Aviation, Space, and

Environmental Medicine, 49(7), 920-925.

Seong, J. S., Lee, T. M., Shin, W. S., Kim, Y. R., Yoon, Y. R., & Yoon, H. M. (2004). The

Analysis of Mental Stress using Time-Frequency Distribution of Heart Rate Variability

Signal. Proceedings of the 26th Annual International Conference of the IEEE EMBS. San

Francisco, CA: Institute of Electrical and Electronics Engineers.

Serfaty, D., Entin, E. E., & Volpe, C. (1993). Adaptation to stress in team decision making

and coordination. Proceedings of the Human Factors and Ergonomics Society 37th annual

meeting (pp. 1228-1232). Santa Monica, CA: Human Factors and Ergonomics Society.

Searle, B., Bright, J. E. H., & Bochner, S. (1999). Testing the 3-factor model of occupational

stress: The impact of demands, control and social support on a mail sorting task. Work &

Stress, 13(3), 268-279.

Searle, B., Bright, J. E. H., & Bochner, S. (2001). Helping people to sort it out: The role of

social support in the job strain model. Work & Stress, 15(4), 328-346.

108

Senders, J. W., & Moray, N. P. (1991). Human Error: Cause, Prediction and Reduction.

Hillsdale, NJ: Lawrence Erlbaum Associates.

Sheridan, T. B. (1979). Definitions, models and measures of human workload. In Moray, N.

(Ed.). Mental Workload: Its theory and measurement (pp.219-233). New York, NY:

Plenum Press.

Shively, R., Battiste, V., Matsumoto, J., Pepiton, D., Bortolussi, M., & Hart, S. G. (1987). In

flight evaluation of pilot workload measures for rotorcraft research. Proceedings of the

Fourth Symposium on Aviation Psychology (pp. 637-643). Columbus, OH: Department of

Aviation, Ohio State University.

Shriver, B. M. (1953). Age and behavior (Research project No.21-0202-005). Randolph Field,

TX: USAF School of Aviation Medicine.

Srivastava, A. K. & Krishna, A. (1991). A test of inverted U hypothesis of stress performance

relationship in the industrial context. Psychological Studies, 36(1), 34-38.

Stanton, N. A., Salmon, P. M., Walker, G. H., & Jenkins, D. P. (2009). Genotype and

phenotype schemata as models of situation awareness in dynamic command and control

teams. International Journal of Industrial Ergonomics, 39(3), 480-489.

Stein, E. S. (1984). The measurement of pilot performance: A master-journeyman approach

(DOT/FAA/CT-83/15). Atlantic, NJ: FAA Technical Center.

Stern, J. A., Walrath, L. C., & Goldstein, R. (1984). The endogenous eyeblink.

Psychophysiology, 21(1), 22-33.

Sterman, M. B., Mann, C. A., Kaiser, D. A., & Suyenobu, B. Y. (1994). Multiband

topographic EEG analysis of a simulated visuo-motor aviation task. International Journal

of Psychophysiology, 16(1), 49-56.

Steptoe, A., Evans, O., & Fieldman, G. (1997). Perceptions of control over work: Psycho-

physiological response to self-paced and externally-paced tasks in an adult population

sample. International Journal of Psychophysiology, 25(3), 211-220.

109

Stokes, A. F., Kemper, K., & Kite, K. (1997). Aeronautical decision making, cue recognition,

and expertise under time pressure. In C. Zsambok & G. Klein (Eds.), Naturalistic decision

making (pp. 183-196). Mahwah, NJ: Lawrence Erlbaum Associates.

Sweller, J., Van Merriënboer, J., & Paas, F. (1998). Cognitive architecture and instructional

design. Educational Psychology Review, 10(3), 251-296.

Transportation Canada. (2004). Approval of Steep Approach Landing Capability of Transport

Category Aeroplanes, AC 525-011. Retrieved from

http://www.tc.gc.ca/CivilAviation/certification/guidance/525/525-011.htm.

Tsang, P. S., & Johnson, W. W. (1989). Cognitive demand in automation. Aviation, Space,

and Environmental Medicine, 60, 130-135.

Tyler, P., & Cushway, D. (1995). Stress in nurses: The effects of coping and social support.

Stress Medicine, 11(4), 243-251.

Ungs, T. J., & Sangal, S. P. (1990). Perception of near earth altitudes by pilots: ascending vs.

descending over both a land and water surface. Aviation Space and Environmental

Medicine, 61(12), 1098-1101.

Urban, J. M., Weaver, J. L., Bowers, C. A., & Rhodenizer, L. (1996). Affects of workload

and structure on team processes and performance: Implications for complex team decision

making. Human Factors, 38(2), 300-310.

Veltman, J. A., & Gaillard, A. W. (1996). Physiological indices of workload in a simulated

flight task. Biological Psychology, 42(3), 323-342.

Veltman, J. A. (2002). A comparative study of psychophysiological reactions during

simulator and real flight. The International Journal of Aviation Psychology, 12(1), 33-48.

Vidulich, M. A., & Tsang, P. S. (1987). Absolute magnitude estimation and relative judgment

approaches to subjective workload assessment. Proceedings of the Human Factors and

Ergonomics Society 31st Annual Meeting (pp.1057-1061). Santa Monica, CA: Human

Factors Society.

110

Vidulich, M. A., & Bortolussi, M. R. (1988). Speech recognition in advanced rotorcraft:

Using speech controls to reduce manual control overload. Proceedings of the American

Helicopter Society National Specialists’ Meeting-Automation Applications of Rotorcraft.

Atlanta, GA: Southeast Region of the American Helicopter Society.

Vicente, K. J., Thornton, D. C., & Moray, N. (1987). Spectral analysis of sinus arrhythmia: A

measure of mental effort. Human Factors, 29(2), 171-182.

Vroom, V. H. (1964). Work and Motivation. New York, NY: Wiley Press.

Watson, A. R., Ntuen, C., & Park, E. (1996). Effects of task difficulty on pilot workload.

Computers and Industrial Engineering, 31(1), 487-490.

Wickens, C. D., & Yeh, Y. Y. (1982). The dissociation of subjective ratings and performance.

Proceedings of the IEEE International Conference on Cybernetics and Society (pp. 584-

587). Seattle, WA: Institute of Electrical and Electronics Engineers.

Wickens, C. D., & Hollands, J. G. (2000). Engineering Psychology and Human Performance

(3rd ed.). Upper Saddle River, NJ: Prentice Hall.

Wiegmann, D. A., & Shappell, S. C. (1997). Human Factors Analysis of post-accident data:

Applying theoretical taxonomies of human error. The International Journal of Aviation

Psychology, 7(1), 67-81.

Wiegmann, D. A, & Shappell, S. C. (2001). Human error perspectives in aviation.

International Journal of Aviation Psychology, 11(4), 341-357.

Wierwille, W. S., & Casali. (1983). A validated rating scale for global mental workload

measurement applications. Proceedings of the Human Factors and Ergonomic Society 27th

Annual Meeting (pp. 129-133). Santa Monica, CA: Human Factors Society.

Wiggins, M. W., & O‟ Hare, D. (1995). Expertise in aeronautical weather-related decision

making: A cross-sectional analysis of general aviation pilots. Journal of Experimental

Psychology: Applied, 1(4), 304-319.

111

Wilson, G. F. (1993). Air-to-ground training missions: A psycho-physiological workload

analysis. Ergonomics, 36(9), 1071-1087.

Wilson, G. F., & Hankins, T. (1994). EEG and subjective measures of private pilot workload.

Proceedings of the Human Factors and Ergonomic Society 38th Annual Meeting (pp.1322-

1330). Santa Monica, CA: Human Factors Society.

Wilson, G. F. (2001). Psychophysiological in-flight monitoring, In J., Fahrenberg, M.,

Myrtek (Eds.), Progress in ambulatory assessment: Computer-assisted psychological and

psychophysiological methods in monitoring and field studies (pp.435-454). Seattle, WA:

Hogrefe & Huber.

Wilson, G. F., (2002a). Psychophysiological test methods and procedures, In S. G., Charlton,

T. G., O‟Brien (Eds.), Handbook of human factors testing and evaluation (2nd ed.)

(pp.127-156). Mahwah, NJ: Lawrence Erlbaum Associates.

Wilson, G. F. (2002b). An analysis of mental workload in pilots during flight using multiple

psychophysiological measures. The International Journal of Aviation Psychology, 12(1), 3-

18.

Winer, B. J. (1971). Statistical Principles in Experimental design (2nd ed.). New York, NY:

McGraw-Hill.

Woods, D. D. (1990). Modeling and predicting human error. In J., Elkind, S., Card, J.,

Hochberg, & B., Huey (Eds.), Human performance models for computer-aided

engineering (pp.248-274). Boston, MA: Academic Press.

Xie, D., & Salvendy, G. (2000). Prediction of mental workload in single and multiple tasks

environments. International Journal of Cognitive Ergonomics, 4(3), 213-242.

Xiao, Y, Mackenzie, C. F., Harper, B. D. (1996). Simulator limitations and their affects on

decision making. Proceedings of the Human Factors and Ergonomics 40th Annual

Meeting (pp.747-751). Santa Monica, CA: Human Factors and Ergonomics Society.

112

Yacavone, D. W. (1993). Mishap trends and cause factors in naval aviation: A review of

naval safety center data, 1986-90. Aviation, Space, and Environmental Medicine, 64(5),

392-395.

Yao, Y. J., Chang, Y. M., Xie, X. P., Cao, X. S., Sun, X. O., & Wu., Y. H. (2008). Heart rate

and respiration responses to real traffic pattern flight. Applied Psychophysiology and

Biofeedback, 33(4), 203-209.

Yerkes, R. M, Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-

formation. Journal of Comparative Neurology and Psychology, 18(5), 459-482.

Ylonen, H., Lyytinen, H., Leino, T., Leppaluoto, J., & Kuronen, P. (1997). Heart rate

responses to real and simulated BA Hawk 51 Flight. Aviation, Space, and Environmental

Medicine, 68(7), 601-605.