Citation
Chronopsychological learning effects of rapidly-rotating shift work on day-shift attention

Material Information

Title:
Chronopsychological learning effects of rapidly-rotating shift work on day-shift attention
Creator:
McAdaragh, Raymon M., 1951-
Publication Date:
Language:
English
Physical Description:
vii, 125 leaves : ; 29 cm.

Subjects

Subjects / Keywords:
Air traffic control ( jstor )
Alertness ( jstor )
Biological rhythms ( jstor )
Circadian rhythm ( jstor )
Fatigue ( jstor )
Learning ( jstor )
Learning curves ( jstor )
Memory ( jstor )
Shift work ( jstor )
Sleep ( jstor )
Dissertations, Academic -- Instruction and Curriculum -- UF ( lcsh )
Instruction and Curriculum thesis, Ph.D ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1999.
Bibliography:
Includes bibliographical references (leaves 118-124).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Raymon M. McAdaragh.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
030361507 ( ALEPH )
41954000 ( OCLC )

Downloads

This item has the following downloads:


Full Text




CHRONOPSYCHOLOGICAL LEARNING EFFECTS
OF RAPIDLY-ROTATING SHIFT WORK
ON DAY-SHIFT ATTENTION













By

RAYMON M. MCADARAGH


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1999












ACKNOWLEDGMENTS


My most sincere appreciation goes to Dr. Lee Mullally, the chair

of my dissertation committee, for his scholarly support and guidance

in the preparation of this manuscript. Despite his busy schedule, he

spent a great deal of time reading drafts of my dissertation and

providing me with valuable and timely feedback. He was a very

supportive and encouraging advisor.

I would also like to thank Dr. Edward Wolfe for his time and

effort in providing me with the guidance and help that I needed in

the data analysis portion of this study. He contributed many hours

of research and guidance so that the correct analysis procedures

would be used to describe the data in a logical and meaningful way.

My appreciation also goes out to my other committee members

for their input and support. Dr. Sebastian Foti, Dr. Larry Loesch, and

Dr. Jeff Hurt were always there when I needed them.

Finally, I would like to thank my wife, Carol, who was always

helpful and supportive, and my two sons, Jeffrey and Eric, who, with

their mother, were very patient and understanding of my busy

schedule at school and at work. I plan to spend much more time

with them from now on.












TABLE OF CONTENTS

Page
ACKNOW LEDGM ENTS.................................................................................................. iiU

ABSTRACT...................................................................................................................... V

CHAPTERS

I INTRODUCTION .................................................................................................... 1

Statem ent of the Problem ................................................................. 4
Need for the Study........................................................................... 5
Null Hypotheses............................................................................ 10.....
Lim itations................................................................................................ 12
Delim itations........................................................................................... 13
Assum ptions............................................................................................ 16
Sum m ary................................................................................................... 17

2 RELATED LITERATURE ............................................................................. 18

Circadian Rhythm s........................................................................... 18
Circadian Dysrhythm la .................................................................. 27
Shift W ork ........................................................................................... 37
Overview .......................................................................................37
W ork Schedules..................................................................... 38
Sleepiness...................................................................................... 42
Perform ance ................................................................................ 53
Circadian Rhythms and Learning................................................ 57
Sum m ary................................................................................................... 69

3 M ETHODOLOGY .................................................................................................. 70

Participants............................................................................................... 70
Instrum ents.............................................................................................. 70







M aterials/Apparatus ............................................................................ 72
Design. .........................................................................................................73
Procedure ................................................................................................. 74
Null Hypotheses ................................................................................. 75
Variables .................................................................................................... 77
Data Analysis..................................................................................... 77
Sum m ary................................................................................................... 80

4 RESEARCH FIND INGS................................................................................ 81

Results........................................................................................................ 81
Hypothesis 1 ........................................................................... 81
Hypothesis 2........................................................................... 81
Hypothesis 3........................................................................... 82
Hypothesis 4........................................................................... 82
Hypothesis 5........................................................................... 83
Hypothesis 6........................................................................... 83
Hypothesis 7........................................................................... 84
Attention Allocation........................................................................ 84
Learning Curve Comparisons....................................................... 87

5 CONCLUSION ...................................................................................................... 98

Introduction............................................................................................. 98
Discussion................................................................................................ 101
Im plications.......................................................................................... 109
Theoretical. ................................................................................. 109
Practical....................................................................................... 111
Recom m endations............................................................................... 112

REFERENCES.............................................................................................................. 118

BIOGRAPHICA L SKETCH ................................................................................ 125












Abstract of Dissertation Presented to the Graduate School
Of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

CHRONOPSYCHOLOGICAL LEARNING EFFECTS
OF RAPIDLY-ROTATING SHIFT WORK
ON DAY-SHIFT ATTENTION

By

Raymon M. McAdaragh

May 1999

Chairman: Lee Mullally, Ph.D.
Major Department: Instruction and Curriculum

The purpose of this study was to determine if an objectively-

measurable day-shift attention decrement or learning deficiency

exists among rapidly-rotating shift workers. Research using

subjective rating scales indicates that this type of shift work is

associated with a lack of alertness during day-shift hours. An

attention decrement may cause learning deficiencies during day-shift

training sessions, or problems with dynamic situation awareness on

the job.

For this study, 37 air traffic controllers (aged under 46 years)

were recruited from the Jacksonville Air Route Traffic Control Center

as volunteer participants. Group 1 consisted of 18 rapidly-rotating







three-shift workers (15 male/3 female), and Group 2 consisted of 19

day/evening two-shift workers (16 male/3 female).

Each participant completed 25 sessions on the NovaScan

computer-based performance test to determine two baseline

measures of attention allocation (one during a spatial-visualization

task and the other during a tracking task). The mean baseline

measures for the groups were compared using t-tests. The mean

learning curves for the groups were compared on five variables

(attention allocation during the spatial-visualization task, attention

allocation during the tracking task, response flexibility in switching

resources, non-transition reaction to spatial visualization, and

tracking error during tracking). The learning curves' slopes and

residuals were compared using a two-way general linear model,

repeated-measures analysis, which included three statistical tests

(group main effect, time main effect, and group-time interaction).

The t-tests determined that the rapidly-rotating shift workers

have a significantly better attention allocation ability than the

day/evening shift workers on both variables measured. The learning

curve comparisons indicated that the rapidly-rotating shift workers

improved their attention allocation ability during the spatial

visualization task, while the day/evening shift workers did not, and

that the two groups stabilized around their means at different rates

on the learning curve for attention allocation during the tracking







task. No differences were found between the groups' learning curves

for the three other variables. It is suggested that day and evening

shift workers may practice irregular sleep-wake schedules and/or

accumulate a greater sleep debt during the work week than do

rapidly-rotating shift workers.










CHAPTER 1
INTRODUCTION

The occupation of air traffic controller carries with it a great

responsibility in terms of both life and property, in that the

controller's job requires him/her to maintain the safe and

expeditious flow of air traffic in his/her area of responsibility.

McAdaragh (1995) has demonstrated that many controllers work

rotating work schedules that have been shown to induce circadian

dysrhythmia, a misalignment or desynchronization of an individual's

biological and/or psychological daily rhythmic cycles (Hawkins,

1987). Chronopsychological research (the study of behavioral and

psychological rhythms) indicates that these type of work schedules

produce a high degree of subjective fatigue in these individuals

during midnight and day shifts, with the lowest degree being

reported on the evening shifts (Melton, 1985; Monk et al., 1988;

Saldivar et al., 1977; Smith et al., 1971 ). The term "shift-work

insomnia" has been used to describe the condition caused by

circadian dysrhythmia which reduces the total amount of sleep

achieved by affected individuals who work shift work (Akerstedt &

Kecklund, 1991).







2
Babkoffet al. (1991) determined that subjective sleepiness

ratings of individuals during periods of sleep deprivation are

dependent upon the phase of each individual's circadian cycle.

Manber et al. (1996) demonstrated that, even when a normal full-

night's sleep is achieved by both groups, individuals who sleep at

irregular schedules report a lesser amount of alertness and a greater

amount of sleepiness during the day than do individuals who sleep at

regular sleep-wake schedules. Billiard et al. (1987), in a study

involving French military draftees, found that sleep difficulties and

irregular sleep-wake schedules were major factors contributing to

excessive daytime somnolence in young men between the ages of 17

and 22 years of age.

Circadian dysrhythmia has also been shown to cause

performance decrements in affected individuals (Higgins, et al., 1975;

Monk et al., 1988), and dysrhythmia and sleep deprivation have

been shown to produce a learning effect (Rankin et al., 1989). Luna

et al. (1997), in a study of air traffic controllers working a rapidly

rotating schedule, found that controllers on the midnight shift of a

forward rapidly-rotating schedule appeared to be falling asleep and

reported increased confusion and fatigue. In a telephone

conversation concerning this study on March 31, 1997, Dr. French

advised the author that, although the performance data collected was








not reported due to a lack of sufficient data, the data that were

collected indicated a significant negative initial learning effect for the

controllers while they were on the midnight shift relative to when

they were on the day or evening shifts. Pilcher and Huffcut (1996)

found, through a meta-analysis of sleep-deprivation studies, that

sleep deprivation has its greatest effect on mood with a lesser effect

on cognitive performance and motor performance in descending

order. They also determined that partial sleep deprivation (sleep

loss which occurs whenever there is a reduction in the usual amounts

of sleep obtained in a 24-hour period), rather than long-term or

short-term sleep deprivation, has a greater negative effect on mood

and cognitive performance. This is important to shift workers,

because partial sleep deprivation results from the condition known

as "shift-work insomnia" which, in turn, results from working

irregular or rotating schedules.

Pilcher and Huffcut also indicate that one clear goal of future

research will be to determine why partial sleep deprivation has such

a pronounced effect on mood and cognitive performance.

For example, partial sleep deprivation may alter certain
circadian rhythm effects on performance and mood. While total
sleep deprivation has been found to interact with circadian
rhythms (Monk et al., 1985; Naitoh et al., 1985), few studies
have investigated the effects of partial sleep deprivation on
circadian rhythms. In addition, partial sleep deprivation may be
similar to fragmented sleep in that subjects in both cases obtain









at least some sleep. Since sleep fragmentation has been shown
to significantly decrease performance and mood (Bonnet, 1986;
Bonnet, 1989), it is possible that the effects of partial sleep
deprivation more closely resemble those of sleep fragmentation
than those of total sleep deprivation. Furthermore, partial sleep
deprivation could have a unique effect on certain psychological
variables. Decreased interest and attention, for example, are
thought to be two prominent variables related to total sleep
deprivation (Meddis, 1982) and could be investigated with
partial sleep deprivation.... In sum, the effects of partial sleep
deprivation need to be more thoroughly investigated,
particularly since partial sleep loss is a relatively common
condition in our society. (p.324)

Statement of the Problem
The Federal Aviation Administration (FAA) defines an

Operational Error as: "An occurrence attributable to an element of

the air traffic system which: (1) Results in less than the applicable

separation minimum between two or more aircraft, or between an

aircraft and terrain or obstacles as required by FAA 7110.65, Air

Traffic Control, and supplemental Instructions. Obstacles include

vehicles, equipment, and personnel on runways; or (2) Aircraft lands

or departs on a runway closed to aircraft operations after receiving

air traffic authorization" (U.S. Department of Transportation, Federal

Aviation Administration, Air Traffic Rules and Procedures Service,

1996, p. 5-1-1). Each year there are a number of operational errors

in the air traffic system, many of which are attributed to controller

error. In 1996, there were 792 operational errors








nation-wide, or about 0.53 operational errors per 100,000 facility

activities in the air traffic system (U.S. Department of Transportation,

Federal Aviation Administration, 1997, p. 6).

Some operational errors result in aircraft accidents. In 1996,

there were 38 large air carrier, 12 commuter, 88 air taxi, 1,911

general aviation, and 182 rotor craft accidents (U.S. Department of

Transportation, Federal Aviation Administration, 1997, p. 4). Some

of these accidents were the result of controller error. Based on what

is known about chronopsychology and circadian dysrhythmia, it is

very possible that at least some of the so-called controller errors are

the result of a failure on the part of controllers to follow procedures,

either because of degraded performance while working, or because

of a failure to fully acquire information concerning these procedures

during training. These controllers may not be assimilating some

information during training sessions due to a degraded attention

allocation and a negative learning effect which are brought on by the

dysrhythmic condition.

Need for the Study
Air traffic controllers need to be in a high state of alertness

while performing their duties, and while undergoing training on new

procedures or during refresher training of prior learning. Redding's

(1992) analysis of operational errors and workload in air traffic







6
controllers has demonstrated that a loss of situation awareness is the

primary cause of controller error, and Endsley (1995) has described

a model of situation awareness which includes training as a

contributing factor to situation awareness in dynamic situations.

Controllers currently work a variety of shift-work schedules, but

in many facilities they are most likely to receive training during

administrative hours, while they are working day shifts. These

training sessions usually take the form of a group presentation in a

classroom setting and do not address the individual differences

among the learners' chronopsychological rhythms. Based on the

aforementioned research, controllers who work dysrhythmia-

inducing work schedules should be experiencing, to at least some

degree, decrements in performance, somnolence, and most probably

a negative chronopsychological learning effect (a negative learning

effect due to a decreased interest or ability to concentrate and focus

their attention) while attending these training sessions. These

decrements can be expected to be present in these controllers due to

irregular sleep/wakefulness cycles, sleep loss (due to shift work

insomnia), and fatigue, which are all induced by rotating or irregular

work schedules.

Training involves instruction, and instruction is designed with

the goal of accomplishing the objectives of the training. Learning








7
theory and research in the cognitive and behavioral sciences provide

many principles which are used in the design of instruction, and

instructional design addresses many variables in order to accomplish

these instructional objectives. The design process takes into account

learning modalities, styles, and strategies, while giving a great

amount of attention to individual learner characteristics,

instructional settings, instructional media selection and learning

activity development. These variables are all considered during the

design process in order to maximize learning for the target audience.

Research in chronopsychology addresses such variables as short-

term and long-term memory, attention and alertness, subjective

fatigue and mood, and cognitive performance issues such as the

ability to synthesize information, all of which express rhythmic

cycles in humans. Even in instructional settings where learners have

regular diurnal circadian rhythms, these variables have great

implications for the design of instruction. But when it is taken into

consideration that many industrial, military, and government

workers work shift work and are susceptible to its consequences, it

becomes apparent that research in chronopsychology offers a wealth

of information and research opportunity to the field of instructional

design. Even though this is the case, at present there is very little, if








8
any, mention of chronopsychology or its principles in the literature of

instructional design.

Shift work and instructional design are usually found sharing a

common niche in society. Instructional design is most widely used in

industry and the military, where it has its foundations. These are

also areas where a large number of personnel work rotating shift

work and night work. In fact, statistics suggest that approximately

20 percent of the American work force is now engaged in shift work

(Venar et al., 1989).

Shift work and night work have been shown to have a negative

effect on individuals' circadian rhythms (the daily physiological and

psychological rhythms of the body and mind) in much the same

manner as transmeridian flight produces the negative state known as

"jet lag." The condition which results in an individual in either

situation is known as circadian dysrhythmia, a desynchronization of

the body's daily rhythmic cycles.

Research indicates that jet lag or shift work affects many

individuals with decrements to health (Vener et al., 1989), and, as

mentioned earlier, subjective sleepiness and fatigue during day and

night shifts, as well as overall decrements in performance. It has

also been demonstrated that irregular sleep-wake schedules are a

cause of alertness disorders (Billiard et aL, 1987), and that the timing








of sleep termination may be closely related to the

sleepiness/alertness rhythm (Akerstedt & Gillberg, 1980). This line

of research implies that shift workers, such as air traffic controllers,

may suffer from an alertness disorder that could cause an attention

decrement. 'f such a decrement were to be identified in controllers

who are engaged in rotating shift work during day shift hours (when

training sessions are scheduled at many facilities), it could imply a

need to amend future instructional strategies to address this variable

when considering instructional activities and media selection.

Research concerning diurnal psychological rhythms also

demonstrates implications for instructional design for normal day-

shift workers. According to Folkard and Monk (1980), short-term

and long-term memory rhythms demonstrate implications for the

timing of instruction depending on the type of instructional objective.

Englund (1979) has demonstrated a diurnal function of reading rate,

comprehension, and efficiency which implies a relationship with

memory, body temperature, and general activity cycles of circadian

rhythms. Reading rate and comprehension peak at different times of

the day, while reading efficiency is maintained.

According to Adan (1993), analysis of the psychological or

behavioral variables of circadian rhythms (chronopsychology) is

recent and still developing. She indicates that most psychological







10
research ignores the time-of-day factor as both a procedural variable

and as a means of control in experiments. This oversight can cause

erroneous results in that behavioral parameters are not trivial and

can be equivalent in magnitude to the effects experienced from

limiting sleep to three hours or from ingesting the legal driving limit

of alcohol.

The fact that chronopsychological parameters are significant in

diurnal circadian conditions is ground enough to consider addressing

these psychological rhythms when designing instruction. By so

doing, designers may take advantage of learners' abilities to

accomplish different objectives at different times of the day.

Furthermore, the complex relationship of these same psychological

variables and their resulting learner characteristics, when influenced

by a variety of work schedule rotations, can and should be the focus

of further research. Only then can the typical psychological

principles concerning the chronopsychological effects of each type of

schedule become understood. Only then can instruction be designed

to take advantage of these same variables for shift workers.

Null Hypotheses
1. Air traffic controllers who work rapidly-rotating work schedules,

will show no significant difference from controllers who work day-

shift and/or evening-shift schedules on their NovaScan mean








11
baseline rating of attention allocation for dial monitoring during the

spatial visualization task (Task #1).
2. Air traffic controllers who work rapidly-rotating work schedules

will show no significant difference from controllers who work day-

shift and/or evening-shift schedules on their NovaScan mean

baseline rating of attention allocation for dial monitoring during the

tracking task (Task #2).

3. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of

controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of reaction time on dial

monitoring during the spatial visualization task (Task #1).

4. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of

controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of reaction time on dial

monitoring during the tracking task (Task #2).

5. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of








controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of reaction time for the transition

from Task #2 to Task #1 (which indicates response flexibility in

switching resources).

6. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of

controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of reaction time on Task #1 non-

transition object orientation (which indicates spatial visualization).

7. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of

controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of tracking error on Task #2

(which indicates mind/motor coordination).

Limitations
Work schedule, the predictor variable, was the only variable

used to differentiate the two groups of participants. Because of this,

work schedule was considered as the suspected cause for any

detriments that were found in this study. Circadian rhythmicity can








only be considered as a probable cause because no other social or

psychological differences among the participants were controlled.

The first group of participants only included controllers who

work rapidly-rotating schedules which are usually composed of day

shifts, evening shifts, and night shifts. To be included in this study,

schedules for participants in this group included at least one

midnight shift and one day shift each week so that behavioral and

psychological circadian rhythms could be expected to be dysrhythmic

in the individuals working them. According to the literature, this

group should be characterized by diurnal biological rhythms and

desynchronous behavioral and psychological rhythms. Therefore,

only desynchronous behavioral and psychological circadian rhythms

were suspected as a probable cause for any decrement to day-shift

attention that was determined. No determinations were made

concerning biological circadian rhythms.

The second group only included controllers who work day-shift

and evening-shift rotating work schedules. According to the

literature, this group is expected to be characterized by diurnal

circadian rhythms, with rotating work and leisure time.

Delimitations
The instrument used in this study measured variables which

were used to determine attention allocation baseline ratings and







14
learning curves for the participants. The NovaScan instrument was

designed and validated to measure an individual's performance

against his/her own baseline of attention allocation, once it has been

established, in order to determine readiness for duty. It has also

been validated for establishing these ratings of attention allocation,

and for establishing both individual and group learning curves for

the variables which it measures (O'Donnell, 1992, p. 19). This was

the first time that this instrument was used to compare the attention

allocation baselines of two groups of individuals who are expected to

differ on this variable according to literature. Evaluation of these

ratings determine if one group displayed a higher mean attention

allocation baseline rating, as determined by NovaScan, than the

other group.

Because NovaScan has been validated for measuring attention

allocation, it has logical (or face) validity for use in this study. It also

has construct validity because it has been validated through research

for its extreme sensitivity to decrements caused by drugs or alcohol

(O'Donnell, 1992, p. 26).

The learning curve data were used to make determinations

concerning any possible learning effects between the groups

compared in this study. The criterian variables measured by

NovaScan in this study determined each participant's learning curve







15
and baseline rating for the following: (a) Two measures of attention

allocation (Variables 1 and 2), (b) response flexibility in switching

resources during multiple tasks performance (Variable 3), (c) spatial

visualization ability (Variable 4), and (d) straight mind/motor

coordination (Variable 5).

All of these variables have been shown by prior research to be

negatively affected by acute shifts in routine. For example, Monk et

al. (1988), demonstrated that a phase advancement of six hours in

the wake period (induced jet lag in the laboratory) with middle-age

male subjects resulted in several negative effects. These negative

effects included the following: (a) Decreased subjectively-rated

alertness (which corresponds to NovaScan Variables #1 and #2), (b)

decreased performance at search tasks (which corresponds to

NovaScan Variable #3), and (c) decreased performance at manual

dexterity tasks (which corresponds to NovaScan Variable #5).

Higgins et al. (1975) demonstrated that a 12-hour shift in the wake-

sleep cycle of 15 male participants (ages 20 to 28) produced

deficiencies in individuals' multiple tasks performance (which

corresponds to NovaScan Variable #3). Luna et al. (1997) found that

Air Force controllers reported greater amounts of confusion (which

corresponds to NovaScan Variable #4) and less vigor on night shifts

than when on evening or day shifts of a rapidly-rotating schedule.








The difference between this study and the previously cited

research is that this study determined objective measures by

NovaScan in a natural setting for all the variables tested, whereas,

the previous research has only reported subjective ratings in natural

or laboratory settings, or objective ratings determined in contrived

laboratory environments.

Assumptions
It was assumed that the participants recruited for this study

had no significant cognitive differences and that they were free of

any attention or learning decrements not related to dysrhythmia.

Air traffic controllers possess unique cognitive skills and abilities.

These cognitive attributes become unique to controllers through a

weeding-out process and through cognitive training. Controllers

undergo pre-employment cognitive and psychological testing which

has been developed to eliminate individuals who do not possess the

necessary mental attributes required of air traffic control work in

general. The training that controllers obtain continues to eliminate

individuals who do not possess the particular skills and abilities that

are required, and to develop those skills and abilities in the

individuals who do possess them. The cognitive variables that were

investigated in this study are some of the same cognitive variables

that are developed through controller training.








Summary
In the previous pages, the problem statement and need for the

study were identified. The problem is that, each year many

controller errors lead to aircraft accidents. This study is necessary

because many air traffic controllers work rotating shift work and are

susceptible to the adverse effects of circadian dysrhythmia, a

condition which causes decrements in alertness and performance

ability. These decrements could be a factor in controller learning

deficiencies during training sessions, and/or deficiencies in situation

awareness during dynamic air traffic control situations. In order to

address these factors, the hypotheses to be answered by this study

were proposed, and the underlying limitations, delimitations and

assumptions of the investigation were discussed.

Chapter two will discuss the related literature and research

findings concerning circadian rhythms and circadian dysrhythmia.

This chapter will also discuss shift work and the different types of

shift systems used in industry and air traffic control facilities.

Following this is a review of literature concerning the decrements in

alertness and performance caused by circadian dysrhythmia. Finally,

a review of the literature concerning circadian rhythms and learning

is presented.











CHAPTER 2
RELATED LITERATURE

Circadian Rhythms

Circadian rhythms are cycles which occur on a daily basis, but

they are only part of a larger system of rhythms.

We live in a universe of rhythms. Our galaxy rotates once every
200 years. Sun spots occur every 11 years and tides every 12
1/2 hours. But the most significant rhythm for the inhabitants
of this planet is the earth's rotation every 24 hours.
(Hawkins, 1987, p. 51)

Cycles which are shorter than 24 hours are called "ultradian", and

cycles which are greater than 24 hours are called "circamensual"

(Adan, 1993, p. 146). Organisms which live on our earth have

evolved within this universal, cyclic environment.

Human circadian rhythms are both physiological and

psychological in nature. The study of physiological (or biological)

rhythms is known as chronobiology, and the study of psychological

(and behavioral) rhythms is known as chronopsychology.

The biological rhythms include such variables as hormone

production, body temperature, organ functioning, immune system

cycles, blood cell functioning, and stomach and intestinal tract

functioning (U.S. Congress, Office of Technology Assessment, 1991).

18







19
Each of these variables displays a daily cycle which has its own peak

(acrophase) and low point (nadir).

Psychological and behavioral rhythms include such variables as

alertness, performance, and the habits of an individual. These

rhythms also display an acrophase and nadir. It has been

demonstrated that short-term memory and long-term memory have

separate rhythms which peak at different times of the day (Folkard

& Monk, 1980).

VSome psychological/behavioral rhythms, such as digit

summation ability and short-term memory, and the biological

temperature rhythm tend to correspond and to influence one

another. Other psychological/behavioral rhythms such as alertness

and long-term memory tend to run in a staggered opposition to the

temperature rhythm. The alertness rhythm, for example, tends to

peak between 10 a.m. and noon, while the temperature rhythm

tends to peak between 4 p.m. and 8 p.m. (Hawkins, 1987; U.S.

Congress, Office of Technology Assessment, 1991).

Circadian rhythmicity stems from two causes, "an endogenous

cause, the internal clock, and an exogenous cause, the rhythmic

environment and habits of the individual" (Folkard et al., 1985, p.

33). By the late 1970s, it had become clear that most of the body's

rhythms are controlled by processes deep within the brain (Hawkins,








1987). Here, the most powerful of the entraining agents (or

zeitgebers), the cycle of light and darkness, is detected through the

eyes and influences the body's visceral system and other cyclic

functions. Other rhythms, however, like that of the body

temperature rhythm, are controlled elsewhere and remain the object

of further investigation. Activities, such as meal time, and physical

and social activity, also serve as entraining agents (Hawkins, 1987).

- According to Mistlberger and Rusak (1989), psychological and

biological rhythms have an internal relationship with time-of-day

restraints. Human behavior and physiological processes demonstrate

a temporal structure that matches the 24-hour day-night cycle. The

most obvious daily cycle that humans experience is the cycle of sleep

and wakefulness. Similarly, a myriad of bodily functions, including

endocrine secretions, body temperature regulation, sensory

processing, and cognitive performance, demonstrate a 24-hour

rhythmicity (p. 141).

Historically, circadian rhythms were ascribed to environmental

cues associated with the solar day, but a large number of studies

have confirmed that daily rhythms in a wide variety of species,

including humans, persist under constant environmental conditions.

Because of this, it is acknowledged that daily rhythms are generated







21
internally by the organism and are not simply passive responses to

environmental stimuli (p. 142).

In the absence of environmental cues, dally rhythms are said

to be "free-running." Free-running rhythms approximate the 24-

hour day, but not exactly. This periodicity (time required for one

complete rhythm cycle) is termed "circadian" (circa = about, dies =

day). Humans who are kept in isolation without time cues display a

25-hour sleep-wake cycle. Sleep onsets begin to occur at 25-hour

intervals instead of the normal 24-hour day-night intervals

experienced in ordinary society. Under this free-running cycle, as it

is termed, humans go to sleep later each day than they would under

normal circumstances.

Free-running sleep-wake cycles vary among species from

about 23 to 26 hours and are modified by factors such as ambient

light intensity, or hormone production. But, the fact that free-

running cycles differ from the normal 24-hour cycle is strong

evidence that circadian rhythms are generated internally, and are

not simply responses to dally 24-hour time cues. In fact, individuals

of the same species recorded under laboratory conditions in adjacent

cages demonstrate different periodicities which cause the animals to

slowly drift apart from one another, while sharing the same

environment (Mistlberger & Rusak, 1989, p.142).







22
Other evidence that supports the idea that circadian rhythms are
endogenously generated includes the demonstration that they
can be modified by selective breeding (Bunning, 1973) and gene
mutations (Konopka, 1980) and that they develop normally in
successive generations of organisms kept in light or dark
(Aschoff, 1960; Davis, 1981). Circadian rhythms are thus innate,
rather than learned or imprinted, phenomena. Virtually all
researchers now agree that circadian rhythms are the products
of an internal biological clock mechanism.
(Mistlberger& Rusak, 1989, p. 142)

There are several recognized environmental influences which

serve as entraining agents to rhythmic cycles. Many organisms,

Including humans, find it adaptive to restrict rest and activity to

specific times of the day in order to take advantage of environmental

conditions which are optimal for various functions. By internalizing

the controlling mechanism for initiating rest and activity periods,

organisms are able to anticipate environmental events, such as

temperature, humidity, and changes in light intensity associated with

morning and nightfall, so that they may prepare accordingly.

Organisms accrue advantages by phasing their behaviors with these

environmental events, while anticipating them rather than merely

responding to their constraints. Therefore, some mechanisms must

exist which ensure that the internal timing device of an organism

maintains a synchronous phase relation to the environment

(Mistlberger & Rusak, 1989, p. 146).







23
Light is, as expected, a universal zeitgeber entrainingg agent) for

the activity-rest cycle displayed by most organisms, from single-

celled life forms through mammals. "A recent case report indicates

that circadian rhythms of body temperature and endocrine variables

in an elderly woman could be shifted by appropriately timed

exposure to bright light, independently of the timing of sleep-wake

states" (Mistlberger & Rusak, 1989, p. 145).

There are also several non-photic stimuli that appear to entrain

circadian rhythms. The timing of meals influences circadian cycles

through the anticipatory activity associated with mealtime in

organisms. This effect is very evident in experiments with rats.

Arousal states can also trigger the circadian timing system.

Simple handling of white-footed mice (Rawson, 1960) or
changing the litter of hamsters (Mrosovsky & Hallonquist, 1986)
can phase shift free-running rhythms, suggesting that even brief
arousals may have feedback effects on circadian timing. In
humans recorded in temporal isolation, a daily "anchor sleep" (4
h of forced bed rest at a fixed time every 24 h) can apparently
synchronize daily rhythms (Minors & Waterhouse, 1981), but
whether this synchrony is related to anchor sleep per se or daily
LD (light-dark cycle) or food intake schedules is unclear.
(Mistlberger & Rusak, 1989, p. 146)

Social cues are another well-known entraining agent in many

species, including humans, however, they are not universally

effective. It has been demonstrated that some blind humans display

free-running sleep-wake cycles in diurnal environmental conditions.







24
These individuals are not synchronized by society's social cues. Just

how social cues influence circadian timing is not known, but the

social cues' effects on arousal states are probably the necessary

mediators for the relationship (Mistlberger & Rusak, 1989, p. 146)

Dinges (1989) describes how sleep is affected by the circadian

timekeeping system. According to Dinges, the daily sleep-wake cycle

serves to organize wake behavior into discrete temporal units and to

coordinate and synchronize the internal timing of many biological

rhythms. We cannot know for sure whether early mammals were

diurnal or nocturnal, but there is little doubt that primates, and

homo sapiens in general, evolved as diurnal species who sleep

primarily at night (p. 153).

Dlnges (1989) states that recent sleep research shows us that the

infrastructure of sleep itself might depend less on sleep and more on

a rhythmic process common to sleep and wakefulness. This sleep-

wake cycle is also affected by sometimes contrasting factors.

To begin with, unlike many other physiological parameters such
as body temperature, the sleep cycle is not a unitary process
that varies along a continuum; sleep Is distinct from
wakefulness, sleep onset contrasts with sleep offset, and there
are different stages of sleep. Even more problematic, however,
is the fact that among our species, the initiation and termination
of sleep are influenced as much by soclopsychological factors as
by endogenous biological factors. (p. 156)







25
According to Dinges (1989), research also indicates that the onset

and offset of sleep and wakefulness display a cycle which reflects the

biological temperature cycle. The temperature cycle reaches its low

point at about 4 am and its acrophase at about 4 pm. As

temperature rises, sleep offset generally occurs, and as temperature

decreases, sleep onset generally occurs. The probability for sleep

onset and sleep offset over a 24-hour period in a normal diurnal

cycle follows a pattern:

1. (01:00-07:00) sleep onset probability high; sleep offset

probability low.

2. (07:00-13:00) sleep onset probability low; sleep offset probability

high.

3. (13:00-19:00) sleep onset probability high; sleep offset

probability high.

4. (19:00-01:00) sleep onset probability low; seep offset probability

low.

This pattern of sleep onset/offset propensity seems to indicate that

there are two periods during the cycle where sleep onset is desirable.

Although nocturnal sleep episodes appear in phase with the
circadian nadir in the endogenous core temperature cycle, naps
occur across the typically broad peak (variable acrophase) of the
temperature cycle, suggesting the possibility of a secondary
sleep propensity approximately 180 degrees out of phase with
the circadian cycle in body temperature (Broughton, 1975;
Dinges etal., 1980). (Dinges, 1989, p.155)








Sleep research indicates that long sleep episodes occur when

subjects begin sleep just after the acrophase of the temperature cycle

in the afternoon. If a subject begins sleep at this time, there is a

propensity to sleep until the temperature cycle is on the rise again

the next morning. In a normal diurnal cycle with two sleep episodes,

research indicates that, "longer sleep began just prior to the

temperature minimum, whereas shorter sleep (naps) occurred near

the temperature maximum" (Dinges, 1989, p.159).

The time of sleep onset in humans does not always follow the

natural cycle of tendencies for sleep because chronobiological

influences on our sleep cycles are often superseded by the

sociopsychological factors in our daily lives. Sociopsychological cues,

such as human contact and signals of activities to be carried out, are

important influences on the human sleep-wake cycle. Because of

this, these factors are thought to play a major role as zeitgebers for

the variation in the sleep-wake cycle. However, problems may arise

from social zeitgebers which would not occur from physical

zeitgebers, such as the light-dark cycle. Social zeitgebers can easily

cause a person to attempt a major, abrupt change in the sleep-wake

cycle before internal circadian processes, such as the temperature

rhythm, are able to adjust (Dinges, 1989, p. 159).







The circadian system is very complex and it interacts with the

physiology and psychology of humans and other species. According

to Mistlberger and Rusak (1989), circadian organization is a

pervasive, integral component of mammalian physiological systems,

which affects behavior in a great many ways. The underlying

mechanisms of this influence are themselves complex, involving a

hierarchy of oscillators that interact with each other and with other

regulatory systems. The circadian system is acutely sensitive to

photic cues in the environment which directly influence the

dominant pacemaker of the system, however, behavior may be

equally affected by other environmental variables, by influencing

other elements of the oscillatory hierarchy (p. 150).

Circadian Dysrhvthmia

Circadian dysrhythmia is a condition which describes the

internal disassociation of the biological and/or psychological rhythms

of an individual (Hawkins, 1987). As stated earlier, this condition is

primarily the result of transmeridian flight (Winget et al., 1984) or

rotating shift work (Folkard & Monk, 1979).

Several researchers have shown that many circadian variables

are affected by acute shifts in routine (Akerstedt & Gillberg, 1980;

Gander et al., 1989; Higgins et al., 1975; Monk et al., 1988; Saito et

al., 1992). Most of these studies indicate that, once circadian








rhythms are desynchronized, at least 7 days to several days are

required for resynchronization to occur. Until resynchronization does

occur, an individual will experience biological and psychological

decrements which are reflected in psychomotor and cognitive

performance (U.S. Congress, Office of Technology Assessment, 1991).

Higgins et al. (1975) conducted a study involving 15 male

participants (ages 20 to 28) in order to determine the effects of a

12-hour shift in the wake-sleep cycle on physiological and

psychological circadian rhythms. The results of this study indicated

that the quantity and quality of sleep did not change to a significant

degree after the sleep cycle was altered. The subjective fatigue

index, likewise, indicated that the total fatigue for the awake periods

was not significantly changed. However, the times of greatest fatigue

within days were altered, and complete reversal of the daily pattern

required nine days.

The physiological parameters which made the most rapid

response to stress were also the same parameters which required the

shortest period of time to rephase after the shift in routine. The

physiological parameter requiring the shortest rephasal time was

heart rate, followed in sequence by norepinephrine, epinephrine,

potassium, sodium, internal body temperature, and 17-ketogenic

steroids.








Performance data based on the Civil Aeromedical Institute

Multiple Task Performance Battery suggest that (1) diurnal variation

was present during the preshift period, (2) performance decrements

occurred on the day of the shift following the short sleep period, (3)

performance rated relatively high for most of the day, but became

poor toward the end of the shift for the first three days following the

shift in routine, (4) performance on the fourth through the sixth post

shift days rated at or above average, with relatively small variations

among the five test sessions per day, and (5) performance on the

seventh through ninth post shift days was below average for the

experiment and demonstrated evidence of a return to a diurnal

pattern, which reflected the post shift sleep-wake cycle (Higgins et

aL, 1975, p. 1).

According to Higgins et al. (1975), the implications of the

findings cited in this study include the following:

(1) Individuals making a 12-hour alteration in the wake-sleep
cycle should not perform critical tasks during the first awake
period following the change. (2) After the first full sleep period
following the change, subjects appeared to perform well even
though the physiological and biomedical parameters measured
were still adjusting to the change. (3) For the first week
following the change in the wake-sleep cycle, individuals should
not work longer than 8 hours continuously because performance
deteriorates after that time. After the change, subjects appeared
to fatigue more rapidly toward the end of the awake period than
they did normally. This effect was evident for several days after
the change. (p. 23)








Monk et al. (1988) also conducted a study of the affects of an

acute shift in routine with 8 middle-aged male participants. In this

study, researchers induced jet lag in a laboratory setting by

advancing the routine of the participants by 6 hours. As stated

earlier, the effects of jet lag would be the same as those experienced

by changes in routine due to changing work hours in shift work.

The eight participants in this study were kept in temporal

isolation for the 15-day experiment. After the participants were

entrained to their own habitual routines for five days, each

participant experienced an acute 6-hour phase advance in routine

which was brought about by a shortening of the sixth sleep episode.

The participants were then held to the new phase-advanced routine

for the ten remaining days of the experiment.

Following the shift in routine, the participants demonstrated

significant symptoms of jet lag which were observed in mood,

performance efficiency, sleep, and physiological temperature

rhythms. Some of the variables, such as temperature phase and

percent rapid-eye-movement sleep, showed a monotonic recovery

pattern. However, other variables, such as actual sleep duration,

percent slow-wave sleep, motivation loss, and subjective sleepiness,

demonstrated a zig-zag recovery pattern, which suggest the

interaction of two competing processes (p. 703).








The results of this study indicated that the disruption in the
amplitude of the circadian temperature rhythm was the most

dramatic circadian effect, and this effect lasted for several days

following the phase shift.

For instance, even 6 days after the phase shift, the mean change
in amplitude of each subject's temperature rhythm was 26%
below his own baseline level. This reduction in amplitude may
bespeak a desynchrony within the circadian system. Normally,
the component parts of the circadian system are in the correct
phase relationship to each other, combining to produce robust,
well-defined circadian rhythms. After an acute shift in routine,
however, these component parts do not adjust their temporal
orientation at the same rate, and appropriate phase relationships
are thereby lost. This has been referred to as "internal
dissociation," a state shown to be associated with impairments in
mood and performance efficiency (Wever, 1975; Wever, 1979).
(p.708)

Among the variables showing a zig-zag recovery function were

the subjective alertness/sleepiness rhythm and the motivation loss.

Some of the post-shift days indicated worse effects than those on the

day immediately preceding them (p. 709).

Akerstedt and Gtllberg (1980), in a study that they conducted

with six male subjects in temporal isolation, demonstrated that sleep

termination may be closely related to the sleepiness/alertness

rhythm. In this study, sleep times were displaced to seven different

times of the day through a rate of one sleep condition change per

week. The length of each sleep episode varied according to when it







32
began. The shortest sleep episodes occurred when sleep onset began

in the morning, and the longest sleep episodes occurred when sleep

onset began in the evening. Sleep termination times were related to

specific points on the sleepiness/alertness rhythm.

Gander et al. (1989) conducted a study involving air crews of P-

3 aircraft in order to determine the adjustment of sleep and the

circadian temperature rhythm after flights across nine time zones.

In this study, researchers investigated both westward flight (phase-

delay variation) and eastward flight (phase-advance variation). The

study involved nine Royal Norwegian Air Force volunteers operating

P-3 aircraft during flights across nine time zones. The variables

monitored during the flight included each participant's sleep-wake

pattern and circadian temperature rhythm. Each participant

recorded his own sleep and nap times, rated nightly sleep quality,

and personality inventories, while rectal temperature, heart rate, and

wrist activity were continuously monitored.

Adjustment after the return eastward flight was slower, as

compared to adjustment after the westward flight, for the

readjustment of sleep timing to local time. The eastward flight also

produced greater interindividual variability in the patterns of

adjustment of sleep cycles and temperature rhythms. One of the

participants even exhibited a 15-hour phase-delayed







33
resynchronization, rather than a phase-advanced resynchronization,

of the temperature rhythm after the eastward flight. The

interindividual differences in adjustment of the temperature rhythm

correlated with some of the personality measures, but it was found

that larger phase delays in the temperature rhythm (as measured on

the fifth day after westward flight) were exhibited by participants

who were rated extroverts, and smaller phase delays were exhibited

by evening types (p. 733).

In this study, Gander et al. also found that all of the subjects

showed clear phase delays in the sleep cycle for both sleep onset and

sleep offset on the westward flight (or phase delay), but that there

was greater intersubject variation in the timing of sleep following the

eastward flight (or phase advance). The slower adaptation and

adjustment to eastward flight was due to the variable recovery

methods of the circadian rhythms. After eastward flights, some of

the circadian rhythms tend to resynchronize by delaying, while other

rhythms advance. It appears that extroverts tend to adapt more

rapidly to the delay shift, probably because they are more exposed

to the social routine in the new time zone (p. 742).

Wright et al. (1983) studied 81 male soldiers (ages 18-34) for

five days before and five days after an eastward deployment across

six time zones. Commonly reported jet lag symptoms of tiredness,








sleepiness, weakness, headache, and irritability appeared in the

majority of the subjects. While most of the symptoms had

disappeared or diminished by the fifth day in Germany, tiredness,

sleepiness, and irritability continued.

The cardiorespiratory responses and the perception of effort

during treadmill exercise were completely unaffected by the

conditions experienced by the participants in this study, however,

other variables were affected.

While isometric strength of upper body, legs, and trunk was
unchanged, dynamic strength of arms was significantly reduced
after arrival in Germany in two out of three groups at the slow
contraction speed and even more dramatically in all groups at
the fast velocity. These findings are suggestive of adverse jet-
lag effects on some aspect of muscle contractile capabilities, most
likely motor control and fiber recruitment. Dynamic arm
endurance declined in a comparable fashion. Complete recovery
of arm strength and endurance did not occur within the first 5 d
after deployment. (p. 136)

Salto et al. (1992) conducted a study of the gradual adjustment

of circaseptan-circadian blood pressure and heart rate rhythms of

two adults (36-year-old male; 32-year-old female) and two children

(6-year-old boy; 6-month-old boy) after an eastward flight which

advanced local time by 9 hours. Circadian rhythms are daily

rhythms of about 24 hours, whereas, circaseptan rhythms are

rhythms of about 7-days duration.








The results of this study indicated that gradual changes

continued for four weeks. The adult male's blood pressure rhythm

delayed in order to recover, while the adult female's rhythm

advanced. Approximately four weeks were needed for adjustment of

circaseptan rhythms in the two adults, but it only took two weeks in

the boys.

The adjustment of rhythms during childhood appears to be

faster not only with circaseptans, but also with circadian rhythms.

The results also indicated that the heart rate rhythms adjusted more

rapidly than the blood pressure rhythms, and that the boys'

circadian rhythms adjusted faster than the adults' circadian rhythms.

The older boy's rhythms adjusted more rapidly than the younger

boy's rhythms. "In the younger boy, the adjustment showed great

liability which might suggest that the circadian rhythmicity is still

developing and immature" (p. 73).

The National Aeronautics and Space Administration (NASA) has

developed the NASA Ames Fatigue Countermeasures Program to

compile research concerning fatigue caused by such factors and to

develop countermeasures. According to Rosekind (1993),

More than a decade of research at NASA Ames on pilot fatigue,
sleep, and circadian rhythms has identified new insights into
crew fatigue. Some basic findings include:
Sleep loss and circadian disruption from long-haul flight
operations can result in fatigue, increased sleepiness, and
reduced performance.









While on short-haul trips for 3 to 4 days, pilots take longer to
fall asleep, sleep less, awake earlier, and report lighter and
poorer sleep compared to pre-trip sleep patterns.
Pilots generally report feeling less well during extended duty
periods, but helicopter pilots on short-haul flights for 4 to 5 days
are far more likely to report headaches and back pain than are
commercial short-haul fixed-wing pilots, probably due to the
physical environment of the helicopter flightdeck.
Off-duty time overstates the time available for sleep.
Regulation of duty hours should be considered, much like
flight hours.
Rest periods should occur at the same time on trip days or
progressively later across days.
On layovers, experienced international flight crews sleep
efficiently at selected times or sleep less efficiently but longer
than normal with a preference for sleeping during local night.
However, despite efficient or longer sleep during layovers, the
circadian system is unable to resynchronize and quickly adapt to
rapid, multiple time zone shifts.
A brief in-flight nap is an acute in-flight safety valve to
improve performance and alertness on long-haul flights, but
naps do not affect the cumulative sleep debt in most
crewmembers. (p. 24)

Age has also been shown to have an effect on individuals'

circadian rhythms. Research by Keran and Duchon (as cited in U.S.

Congress, Office of Technology Assessment, 1991, p. 95) has

demonstrated that individuals over the age of 45 may begin to

demonstrate the effects of dysrhythmia and may have trouble

sleeping at night.








Shift Work

Overview

The term "shift work" has many meanings and varies according

to which author or context is used as reference. According to

Akerstedt (1988),

Shift work usually refers to an arrangement of work hours that
uses two or more teams (shifts) to cover the time needed for
production. Whereas two-shift work usually covers only the
daylight hours, three-shift work also covers the night. Shifts are
often changed at -0600 h, 1400 h, and (if a night shift is
included) 2200 h, although many companies employ earlier or
later times. In Europe, the teams usually rotate between the
shifts, whereas in the United States, assignment to a certain shift
is often permanent, at least for a considerable time, before
seniority allows transfers to another shift. Permanent night
work, the watch system at sea, and roster work are other
varieties of work hour arrangements. The latter (roster work) is
similar to conventional shift work but is somewhat more
irregular and "customized" to particular needs, usually in the
service sector. For lack of better terminology we will refer to all
these types of work hour systems as shift work. (p. 17)

Individuals who work rotating shift work, in most cases, are

almost constantly subjected to negative physiological and

psychological effects that are similar to those experienced as a result

of jet lag. The severity of the condition depends upon the type of

rotating schedule that the individual works and the number of hours

worked during each shift.

Schedules which rotate on a weekly basis can cause biological

rhythms to desynchronize, as they try to adjust to each new routine








each week. On the other hand, schedules which rotate rapidly,

covering all shifts each week, allow biological rhythms to remain

diurnal, but can cause desynchronization of the

psychological/behavioral rhythms, as the habits and daily routine of

the individual adjust to accommodate each new work period during

the week (Folkard et al., 1985; Paley & Tepas, 1994; U.S. Congress,

Office of Technology Assessment, 1991). Researchers have indicated

that sleepiness and fatigue which result from shift work are

detrimental to safety in many occupations (Akerstedt, 1988), and

that individuals who work shift work never adjust to it, even after

several years (Akerstedt & Kecklund, 1991).

Work Schedules

Shift work takes a variety of configurations in industry,

government, and the military, and it also varies according to the

tradition of the country or occupation in which it appears. "It has

been estimated, however, that approximately one-fourth of the

working population in industrialized countries is employed on some

kind of shift work system" (Akerstedt, 1988, p. 18).

The different types of rotating schedules in shift work each have

their own advantages and disadvantages. As previously stated,

Folkard et al. (1985), and Monk (1986) indicate that rapidly-rotating

shift systems allow the endogenous circadian system to remain








diurnal, whereas, weekly-rotating systems desynchronize these

physiological circadian rhythms. The rapidly-rotating systems are

also advantageous in that they do not allow a sleep debt to

accumulate because the number of day sleeps is less continuous than

with the weekly-rotating systems. However, decrements in alertness

and increased fatigue are reported by workers under both systems,

indicating that workers' psychological rhythms are negatively

effected, probably due to the changes in routine experienced under

both systems.

Many researchers once thought that permanent night work

might be more conducive to a more efficient adjustment to

sleepiness. But, Folkard et al. (1978) found that the rated alertness

of permanent night nurses remained diurnal, or day-oriented, except

for a minor blunting of the steep fall of the alertness curve during

the night shift. Other researchers (e.g., Dahlgren, 1981; Patkal et al.,

1977) indicate that sleepiness in permanent night work appears to

be phase delayed and exhibits a peak towards the end of the shift.

Melton and Barlanowicz (1986) described the different models of

shift work, such as the phase-advanced 2-2-1 (2 evenings, 2 days, 1

midnight) schedule, and its opposing phase-delayed 1-2-2 model, in

order to explain why certain shift work models are desired by

workers. They indicated that the 2-2-1 rotation is preferred by







40
many controllers because it allows for longer intervals between work

weeks. About an 80-hour break, or 48% of the 7-day week, is

realized between work weeks with this shift rotation. This is

accomplished by compressing 40 hours of work into an 88-hour

period, with short intervals between each shift change (quick-turn-

arounds). Quick-turn-arounds usually take the form of an 8-hour

break between shifts.

A phase-delayed 1-2-2 schedule expands the work week with

longer periods of 18 to 22 hours between shifts (slow-turn-arounds).

These schedules start off with the midnight shift and progress

through later shifts over the week, ending on the evening shift. The

off-duty period at the end of the week is usually about 48-hours

duration.

In the case of either the 2-2-1 or the 1-2-2 rotation, however,

all shift schedules require 40 hours of work in a 5-day period with 8

hours work per shift. It is noteworthy that the expanded 1-2-2

rotation is not known to be in use in any industrial setting. It seems

that workers will accept almost any work schedule that will provide

a long break at the end of the work week. Some nuclear power

plants operate on 12- and 16-hour shifts. Many police officers work

12-hour shifts for 4 days, and then have 4 days off. One of the

perquisites (perks) of seniority in shift work is choice of work








41
schedule, or time off. Workers more often bargain for time off rather

than for monetary compensation (p.3).

Melton and Bartanowicz (1986) also described the weekly-

rotating schedule. This work schedule, the straight five pattern,

allows individuals to work the same shift for five days, have 48

hours off, and then begin the next, usually later, shift. Research by

the Federal Aviation Administration, and others, has demonstrated

that the five mid watches in a row is the wrong type of shift rotation

to use, from the standpoint of circadian rhythmicity. The five

straight mids leads to accumulated sleep loss and fatigue because

day sleeps are cut short due to noise levels, light intensity, and the

phase of the circadian temperature rhythm and other physiological

rhythms. The sleep that is attained is mainly fatigue-induced sleep,

which is usually brief and inadequate (p. 5).

According to Akerstedt (1988), schedules which rotate should do

so with consideration being given to the direction of rotation, if they

are to be beneficial to the health and performance of the worker.

Another important aspect of the shift schedule may be its
direction of rotation. Since the free-running period of the
human sleep/wake cycle averages 25 h, and since it can be
entrained by environmental time cues only within 1-2 h of the
free-running period, phase delays are easier to accomplish than
phase advances (Czeisler et al., 1980). For the rotating shift
worker, this implies that schedules that delay,i.e., rotate
clockwise (morning-afternoon-night) should be preferred to
those that rotate counterclockwise. However, there have been







42
very few practical tests of this theory, particularly in relation to
sleepiness. Still, Czeisler et al. (1982) have demonstrated that a
change from counterclockwise to clockwise rotation, together
with a change from 7-day to 21-day rotation, improved
productivity and wellbeing in three-shift workers. Orth-Gomer
(1983) found that a change in the same direction in police
officers on rapid (1-day) rotation reduced blood pressure and
improved wellbeing. (p. 27)

McAdaragh (1995) conducted a study of 997 air traffic

controllers' (ATC's) work schedules from randomly selected ATC

facilities in order to determine what types of schedules controllers

tend to work. It was determined that 45% of the controllers were

working either phase-advance, compressed schedules of the 2-2-1

variety, or an alternate work schedule which even further

compresses the work week into four 10-hour days. Only 24% of the

controllers were working straight days or straight evenings, whereas,

the remaining 31% were working a variety of none-standard rotating

schedules. None of the controllers were working the 1-2-2 phase-

delayed schedule or anything like the schedule proposed by Czeisler,

which was reported to improve the production and wellbeing of

rotating shift workers.

Sleeniness

According to the U.S. Congress, Office of Technology Assessment

(1991), shift work can cause a variety of negative psychological and

physiological effects. This is attributed to the fact that the shift








worker must fight both the natural, diurnal trend of physiological

circadian rhythms, and the dominant sociocultural attitudes. Because

of this, there are three sources of stress for the shift worker: (1)

disruption of circadian rhythms, (2) disruption of sleep and fatigue,

and (3) social domestic disturbances (p. 89).

The Office of Technology Assessment also indicates that

disruptions to circadian rhythms that interact with loss of sleep and

fatigue can also affect health and performance. "Sleepiness is the

inclination to sleep, whereas fatigue is weariness due to physical and

mental exertion. It is possible to experience one without the other,

but both, either alone or in concert, can have deleterious effects"

(p. 87).

Sleepiness (often referred to as fatigue) has been measured

through a variety of methods. Traditionally, sleepiness was

measured via simple category rating scales or other simple rating

scales. Later methods, using the electroencephalograph (EEG) and the

electrooculograph (EOG), attempted to achieve more objective

measures.

According to Akerstedt (1988), the first attempt to quantify

objective physiological measures of sleepiness was the Multiple Sleep

Latency Test (MSLT). This test expresses sleepiness as the time it

takes for the EEG/EOG signs of sleep to appear in a subject with eyes








closed. This test, and variations of it, remains the only standardized

clinical method of measuring sleepiness. However, it is difficult to

use outside of well-controlled laboratory, or laboratory-like,

environments, and it cannot reflect momentary fluctuations of

sleepiness.

Nevertheless, it appears that other EEG/EOG parameters may be
used for the description of momentary fluctuations of
wakefulness/sleepiness. Thus vast amounts of data support the
proposition that increased Alpha and Theta activity in the EEG,
and also slow eye movement (SEM) activity, are closely
correlated with both subjective and behavioral sleepiness
(Daniel, 1967; O'Hanlon & Beatty, 1977; Torsvall & Akerstedt,
1984; Akerstedt et al., 1984; Lecret & Pottier, 1971; O'Hanlon &
Kelley, 1977;Torsvall & Akerstedt, 1987). In a record study
(Torsvall & Akerstedt, in press), we tried to experimentally
quantify the EEG/EOG changes that characterize severe
sleepiness, i.e., the level of sleepiness at which drowsiness
prevents interaction with the environment and the individual
dozes off. We found after spectral analysis that increases of
~500 and 200% in the power density of alpha and theta activity,
were critical (compared with the values at normal alertness).
Furthermore, field studies have demonstrated that EEG/EOG-
based monitoring of sleepiness Is also feasible in completely
ambulatory subjects under their normal work/leisure/sleep
conditions (Torsvall & Akerstedt, 1987; Torsvall et al.,
unpublished observations). (Akerstedt, 1988, p. 18)

Most studies concerning sleepiness in shift work derive, as

indicated earlier, from subjective questionnaire studies. Several of

these studies, including Wyatt & Marriot, 1953, Thiis-Evensen, 1958,

Mott et al., 1965, Menzel, 1962, Andersen, 1970, and Akerstedt &

Torsvall, 1978 (as cited in Akerstedt, 1988, p. 18) indicate that shift








workers on the whole report greater fatigue than do day workers.

This fatigue is usually greatest on the night shift, intermediate on the

morning shift, and hardly appears at all on the evening shift.

A record study by Torsvall and Akerstedt (as cited in Akerstedt,

1988) used both subjective, self-report measures of sleepiness, and

objective EEG/EOG ratings. The results of this study indicates that

individuals perceive sleepiness coming before it becomes manifested

on objective ratings. "It therefore appears that sleepiness is

perceived by the individual well before he or she is overcome by

sleep. Thus, one should be able to use subjective sleepiness as a

signal or warning that involuntary sleep might ensue" (Akerstedt,

1988, p. 22)

Akerstedt & Torsvall (1978) used an experimental design which

showed that reported fatigue increased upon entering shift work,

and decreased upon leaving it. Also, Akerstedt et al. (1983)

determined that sleepiness has been severe enough to have resulted

in actual incidents of individuals falling asleep during the night shift.

These studies have since been duplicated by other researches.

Folkard and Condon (1987) demonstrated a somewhat unusual

manifestation of night shift sleepiness reported by night shift nurses

which came to be known as "night shift paralysis." This is a rare

phenomenon which seems to occur as a function of sleep loss (due to








shift work) and takes the form of an inability to react to stimuli

which should normally elicit a response. Nurses may not react to a

call from a patient or a question from a colleague, for example.

A similar study by Folkard and Condon was later conducted with

air traffic controllers. This study was designed to examine the

possibility that night shift paralysis may be a problem in this group.

The sample included 435 air traffic controllers from 17 different

countries, who were working a variety of work schedules. The

incidence of paralysis was found to be influenced by four main

factors which, in turn, affect the workers' level of sleep deprivation

and sleepiness. These factors included (a) the time of night, (b) the

number of consecutive night shifts, (c) the requirement to work both

a morning shift and a night shift starting on the same day, and (d)

individual differences in flexibility of sleep habits. "These results

suggest that the incidence of this paralysis may indeed prove to be a

useful 'critical incident' for comparing the level of sleep deprivation

associated with different shift systems or individuals" (Folkard &

Condon, 1987, p. 1353).

Folkard and Condon also described the relationship between the

alertness cycle and the sleep/wake cycle, and how partial sleep

deprivation occurs as a result of shift work.

It is well established that subjectively rated alertness shows a
marked circadian (about 24 h) rhythm that Is at least partially








independent of the normal sleep/wake cycle (Folkard et al.,
1985), and that alertness is low in the early hours of the
morning. Similarly, there is good evidence that the duration of
day sleeps taken between two successive night shifts is
considerably shorter than the duration of normal night sleeps
(Knauth et al., 1980). Thus there will be a cumulative partial
sleep deprivation over successive night shifts which could
account for the increased incidence of this paralysis on the
second or subsequent night shift. Further, there is also evidence
that night sleeps preceding a morning shift are shorter than
normal ones (Knauth et al., 1980). Finally, rigid sleepers, and
especially those who are also evening types, are less likely to be
able to sleep successfully prior to a night shift. (Folkard &
Condon, 1987, p. 1361)

As stated earlier, Pilcher and Huffcut (1996) found, through a

meta-analysis of sleep deprivation studies, that sleep deprivation has

its greatest effect on mood with a lessor effect on cognitive

performance and motor performance, respectively. They also

determined that partial sleep deprivation (sleep loss which occurs

whenever there is a reduction in the usual amounts of sleep obtained

in a 24-hour period), rather than long-term or short-term sleep

deprivation, has a greater negative effect on mood and cognitive

performance. This is Important to shift workers, because partial

sleep deprivation is one result of working Irregular or rotating

schedules. The condition which occurs due to partial sleep

deprivation associated with shift work has been termed "shift-work

insomnia" (Akerstedt & Kecklund, 1991, p. 509).








Other researchers have determined that irregular sleep

schedules are responsible for daytime sleepiness. As stated earlier,

Manber et al. (1996) demonstrated that, even when a full-night's

sleep is achieved by both groups, individuals who sleep at irregular

schedules report a lessor amount of alertness and a greater amount

of sleepiness during the day than do individuals who sleep at regular

sleep-wake schedules. Billiard et al. (1987) support this conclusion

with a study involving French military draftees. These researchers

found that sleep difficulties and irregular sleep-wake schedules were

major factors contributing to excessive daytime somnolence in young

men between the ages of 17 and 22 years of age.

One study conducted by the Federal Aviation Administration's

(FAA's) Civil Aeromedical Institute (CAMI) compared a group of air

traffic controllers on a rapidly-rotating schedule with a group of

controllers on a weekly-rotating schedule. The results indicate that,

during the work week, controllers working the rapidly-rotating

system acquired less sleep than the controllers working the weekly

rotating system, but that this sleep deficit diminished when days off

were taken into consideration. The controllers on the rapidly-

rotating system caught up on their sleep on their days off. But, in

general, both groups reported similar complaints. "Fatigue,

weakness, and somnolence were the most frequent on-duty







49
complaints given by controllers on the mid shift, followed by those

same complaints in decreasing order of frequency on day and

evening shifts" (Saldivar et al., 1977, p. 5).

Individuals who work rapidly-rotating schedules have their

sleep cut short during day sleep primarily because of two reasons.

First, there are higher noise levels usually present in the morning

hours after night work is ended and the normal daytime activities of

society begin. Second, individuals attempting to sleep in the mid- or

late-morning hours are also attempting to sleep while their diurnal

temperature rhythm is rising. Normally, sleep onset occurs in the

evening, when the body temperature rhythm Is on the decline. Day

sleep is, therefore, shortened even if a sleep deficit exists. This in

turn helps lead to sleepiness, especially on the night shift.

"Sleepiness occurring during night work seems to be the result of a

combination of influences: the circadian trough of the alertness

rhythm, and sleep loss/time awake" (Akerstedt, 1988, p. 26).

It has been estimated that most individuals who work shift work

never adapt to it. Akerstedt and Kecklund (1991) completed a two-

year study of rapidly-rotating shift workers at a paper mill. "The

results suggest that sleep has a traitlike character, even when

occurring at daytime after a night shift. This, therefore, suggests that







50
shift work insomnia is a persistent problem--those afflicted tend to

remain so for some time" (p. 509).

According to a review by the U.S. Congress, Office of Technology

Assessment (1991),

Some researchers (Folkard et al., 1985; Knauth et al., 1981)
maintain that, regardless of rotation length, complete
realignment of circadian cycles never actually takes place in
shift work due to the competing influences of cues in the society
telling workers that they are on an abnormal schedule. Even in
permanent night shift work, if the worker does not maintain the
same schedule on days off as on workdays, he or she will rapidly
revert to the natural diurnal orientation (Monk, 1986; Van Loon,
1963). In such circumstances, the circadian system never fully
synchronizes to the night schedule. (p. 91)

Paley and Tepas (1994) conducted a study involving firefighters

working an 8-hour, rotating (2 weeks per shift) schedule. The

results of this study demonstrated that firefighters working this type

of schedule will sleep less than normal and will report lower positive

mood scores, higher negative mood scores, and higher ratings of

sleepiness on the night shift. The results also demonstrated that,

over the two-week course of the shift, the firefighters were unable to

adapt to changes in their sleep schedule (p. 269).

Research over the past two decades including, Akerstedt &

Torsvall, 1981; Rutenfranz et al., 1976; Tepas et al., 1985; Tepas &

Carvalhis, 1990; Williamson & Sanders, 1986 (as cited in Paley &

Tepas, 1994), indicates that, "workers on the afternoon/evening







51
shifts sleep longest, workers on the day shift sleep slightly less, and

night shift workers sleep least" (Paley & Tepas, 1994, p. 270). In

addition to this, Paley and Tepas (1994) also indicate that the partial

sleep loss resulting from working shift work should effect individuals

with some of the same symptoms that have been seen in studies

concerning total sleep loss.

Partial sleep loss occurs whenever there is a reduction in the
usual amounts of sleep obtained in a 24-h period. Johnson and
Naitoh (1974) suggested that the factors involved in determining
the effects of total sleep loss are equally relevant to partial sleep
loss. Therefore, changes in mood state, increased feelings of
fatigue and irritability, inability to concentrate, and periods of
misperception associated with total sleep loss should also occur
from reductions in sleep length or changes in sleep/wakefulness
cycles experienced In night work. (p. 270)

Reinberg et al. (1989) conducted a study concerning shift work

tolerance and the internal desynchronization of circadian rhythms.

It was found that all the participants in the study maintained some

diurnal cycles (including temperature), but displayed a

desynchronlzatlon of heart rhythm, as well as various psychological

and behavioral rhythms, while working a rapidly-rotating shift

system. Individuals who demonstrated intolerance to shift work

complained of drowsiness, fatigue, and lack of attention. "Since an

internal desynchronization can be observed in tolerant shift workers

having no complaint, It is likely that symptoms of intolerance are








related to the subjects' sensitivity to internal desynchronization

rather than to the desynchronization itself" (p. 33).

A report in the U.S. Congress, Office of Technology Assessment

(1991) suggests that the age of an individual may also be a factor in

his/her ability to adjust to different types of shift work. The report

indicates that, as a person ages, the internal clock becomes more

difficult to reset, and sleep becomes more fragile and more easily

disrupted. These effects usually begin to occur after the age of 45.

The report also indicates that surveys of rotating shift workers have

found that workers between the age of 45 and 60 years old had

more difficulty adjusting to afternoon and night shifts than did

younger workers (p. 95).

After reviewing the literature concerning the effects of shift

work, Akerstedt (1988) suggests that sleepiness, which results from

shift work, may be an underestimated potential hazard in many

occupations.

The results reviewed clearly demonstrate that shift work is
associated with increased subjective, behavioral, and
psychological sleepiness. The effects are particularly
pronounced during the night shift, and may terminate In actual
Incidents of falling asleep at work. In some occupations this
clearly constitutes a hazard that may endanger human lives and
have large economic consequences. Furthermore, since night
shift sleepiness affects a large majority of individuals engaged in
it, the threat is quite real and has probably been underestimated
in the past. (p. 30)









Performance

The different types of rotating shift schedules have differential

effects on the individuals working them, based upon the type of

circadian disruption they cause. As stated earlier, rapidly-rotating

work schedules allow the physiological circadian rhythms to remain

diurnal, while disrupting the psychological/behavioral rhythms, thus,

providing a tendency to disorient the individual. Work schedules

which rotate on a weekly basis have social/psychological advantages

for the individual, but they also have a direct effect on the

physiological rhythms because they disrupt these rhythms with a

new routine each week before they have time to adjust.

Physiological rhythms can take from seven to several days to adjust

to each new routine.

Performance rhythms, however, appear to be less effected on

the night shift by rapidly-rotating schedules than by weekly rotating

schedules. According to Folkard et al. (1976), rapidly-rotating shift

systems affect performance differently according to the memory load

involved in the task required of the individual. Jw memory load

task performance shows a high positive correlation with the

physiological temperature rhythm and tends to be poor at night.

However, high memory load task performance is negatively









correlated with body temperature and is best at night for rapidly-

rotating shift workers.

This finding apparently contrasts with some of the earlier

studies concerning the variation in performance during normal

waking hours, which found that performance on most tasks tends to

improve over the day, while short-term memory tasks performance

has a tendency to decrease (p. 21).

More recent research has indicated that the concept of a single
performance rhythm is erroneous and that, like physiological
rhythms, performance rhythms differ not only in their normal
phase but also in the degree to which they are influenced by
exogenous factors (Folkard et al., 1984; Monk et al., 1983).
Indeed, there is evidence that memory-loaded, cognitive tasks,
which are becoming increasingly common in industry, may be
performed particularly well at night provided there is little
adjustment of the individual's circadian rhythms (Folkard et al.,
1976; Folkard & Monk, 1980). Further, the adjustment of this
type of circadian rhythm which peaks at night occurs relatively
rapidly (Hughes & Folkard, 1976; Monk et aL.,1978). Such
adjustment will result in an impairment of night-shift
performance and so suggests that, for this type of task, shift
systems that minimize adjustment (i.e. rapidly-rotating shift
systems) may be preferable. It is thus noteworthy that the only
field of study to have found superior performance on the night
shift concerned the logging of errors of computer operators (a
task with a high memory load) on a rapidly-rotating shift system
(Monk & Embrey, 1981). (Folkard et al., 1985, p. 39)

Folkard points out that, from a performance point of view,

rapidly-rotating shift systems also have the advantage of minimizing

the cumulative sleep debt which can itself impair performance.







55
Sleep debt can also impede performance dramatically as evidenced

by the phenomenon known as night shift paralysis, which prevents

the worker from performing his/her job for several minutes.

According to Folkard (1985), it appears that, unless the shift worker

is engaged in a particularly crucial but relatively simple task, the

advantages of permanent shift systems are outweighed by their

disadvantages. Rapidly-rotating shift systems minimize cumulative

sleep debt and are more advantageous for the performance of

memory-loaded, cognitive tasks which are being performed more

and more by shift workers (p. 39).

It has been observed, however, that individuals who work

rotating work schedules receive lower job performance ratings in

some cases. The U.S. Congress, Office of Technology Assessment

(1991) reports on the findings of a study concerning the performance

ratings of nurses. "Job performance of nurses, as measured by a

questionnaire filled out by supervisors, was found to be lower in

those on a rotating shift than in those on fixed day, afternoon, or

night shifts" (p. 99).

It has also been demonstrated that extended work hours have a

negative effect on performance. Rosa and Bonnet (1993) conducted

the second of two studies which concerned a change from an 8-hour

work schedule to a 12-hour compressed work schedule. After ten








month's adaptation to the new 12-hour shift schedule, workers

expressed decrements in performance and alertness which were

attributable to the extra 4 hours on the shift. The workers also

reported reductions in sleep across the work week which were most

significant on the 12-hour night shifts. Rosa and Bonnet indicate that

these results are consistent with their earlier findings, and that extra

caution should be exercised when scheduling critical activities for

workers on extended shifts, especially during extended night shifts,

such as the 12-hour shift (p. 1177).

Daniel and Potasova (1989), in a study which compared 8-hour

and 12-hour rapidly-rotating shift workers, demonstrated that the

temperature rhythm of the 12-hour shift workers goes into rapid

decline 4 hours earlier than the temperature rhythm of the 8-hour

shift workers. Other results of this study concerned comparisons on

performance. Striking changes in performance were noted during

the circadian rhythm in psychomotor activities, while the

performance curve during more exacting mental activities

demonstrated a more uniform pattern. Some of the lower

achievements by the 12-hour workers could be attributed to

increased fatigue during the longer shift, however, they could

possibly be attributed to lower work requirements being assigned to

these workers (p. 695).









These and other studies tend to demonstrate some differential

effects of shift work schedules, but, regardless of the type of

schedules involved, shift work in general has been shown to cause

social and domestic disruption for shift workers, which in turn can

affect their performance and possibly have costly consequences in

industrial settings.

Work schedules may induce stress by preventing the worker
from fulfilling important family roles (Walker, 1985). Social
companionships, parenting, and sexual partner roles can all be
compromised by work schedules. These effects may be major
and can severely affect mood, motivation, and sleep, therefore
having indirect effects on performance and safety. Marital
problems, excessive domestic load, and community alienation
have all been documented as a result of the strain placed on
workers by work schedules. (U.S. Congress, Office of Technology
Assessment, 1991, p. 93)

Circadian Rhythms and Learning

Instructional design is based on what is known about learning

theory, information technology, systematic analysis, and

management methods. At the lesson and course levels, strategies are

designed so that instruction accommodates the learner. If learning is

to take place during instruction, these strategies must include

planning which considers both the instruction, and the individual

learner. If the learner is not receptive to the instruction for any

reason, learning will not take place. Learning is actually a two-way

process. Instruction is only effective when the instructional








strategies employed are compatible with the learning style and

capabilities of the learner. The learning style and capabilities of the

learner are the individual learner characteristics which are unique to

each learner and must be addressed in instructional planning.

Learner characteristics include such variables as social background,

experiential background, developmental level, motivation, content

knowledge, and learning style. Because these characteristics are

different for each individual, instructional strategies must be

planned to meet each learner's needs.

Chronopsychological research has addressed some other learner

attributes as well. As stated earlier, the human short- and long-term

memory cycles, and the cycle of alertness all indicate peak periods

and low periods at specific times of the day in individuals with

normal, diurnal circadian rhythms. It has also been shown that high

memory-loaded tasks and tasks involving the synthesis of data are

accomplished more effectively at different times of the day for this

very reason (Adan, 1993; Folkard, Minors & Waterhouse, 1985;

Folkard & Monk, 1980). These, and other, chronopsychological

variables operate in rhythmic cycles in all individuals. As Adan

(1993) indicates, chronopsychological parameters are not trivial and

most psychological research has not even addressed time-of-day as a

consideration. She also explains that this could have a been cause for








erroneous experimental results in many cases of past psychological

research.

Monk (1989) indicates that the study of psychological circadian

rhythms uses the physiological temperature rhythm as a reference

for three major reasons:

The first is simply expedience, body temperature being a well-
defined, relatively stable circadian rhythm that can be measured
easily by psychologists who lack medical qualifications or
radioimmunoassay competence. Second, there is a strong
historical tradition, espoused by pioneers such as Kleitman
(Kleitman, 1963) and Colquhoun (Colquhoun, 1971), linking
psychological rhythms to body temperature rhythms. Third,
there are the recent mathematical models of the circadian
system (Borbely, 1982; Kronauer et al., 1982; Wever, 1975)
which all rely crucially upon the concept of an endogenous
circadian oscillator for which the body temperature rhythm is
the major indicator. (p. 163)

Monk also indicates that psychological rhythms are not all the

same and that, "like the physiological circadian rhythms,

psychological rhythms comprise gradual fluctuations over the entire

24 h. Moreover, there are strong intervariable differences" (p. 166).

He also states that psychological rhythms are not just specific to each

individual. Studies by researchers such as Home and Ostberg, 1977,

and Kerkhof, 1980 (as cited in Monk, 1989), have demonstrated that,

even with differences between "morning-type" and "evening-type"

individuals, "for the 80 per cent of people who are neither extreme

'morning type' nor extreme 'evening type', one can make







60
generalizations about psychological circadian rhythms that are valid

for the population as a whole" (Monk, 1989, p. 166). Monk goes on to

say that the link between subjective alertness and the circadian

system makes sense and can be thought of in terms of psychological

rhythms acting as an interface between the physiological circadian

processes and the sleeping and waking behaviors that those

processes influence (p. 166).

Monk's own research in forced desynchronization studies in

1987 (as cited in Monk, 1989) determined the relationship between

the temperature rhythm and the subjective alertness rhythm. Monk

has determined that, when the sleep-wake cycle is driven at a point

which is beyond the range of the endogenous circadian oscillator

controlling the temperature rhythm, the oscillator assumes a period

length which is close to the free-running 25-hour cycle. This causes

two periodicities to be present in the data, that of the sleep-wake

cycle, and that of the temperature rhythm. Examination of the data

has demonstrated that both periodicities influence the subjective

alertness rhythm. A parallelism between the alertness rhythm and

the temperature rhythm exist at the temperature rhythm oscillator

period, and the maximal subjective alertness was found, on the

sleep-wake cycle, to occur from 8 to 10 hours after waking. These

data coincide with the early afternoon peak in alertness which is









usually found. Monk goes on to say that the relationship between

subjective alertness and body temperature can be explained by

postulating control by both the temperature rhythm oscillator and

the sleep-wake mechanism, with the sleep-wake mechanism tending

to have more dominant control under normal, diurnal sleep-wake

patterns (Monk, 1989, p. 167).

In addition to the subjective measures of alertness, there are

also many circadian rhythms which lend themselves to objective

measures of performance efficiency. Some performance rhythms run

parallel to the circadian temperature rhythm and can be measured in

this manner.

Colquhoun (Colquhoun, 1971) studied vigilance tasks (detection
of an infrequent signal), simple addition tasks (adding up 6 two-
digit numbers), and other simple reaction-type tasks, finding
them to parallel (in measures of speed) the circadian
temperature rhythm, even during the process of adjustment to a
change in routine. This observation confirmed earlier studies by
Kleitman (Kleitman, 1963), which used fewer subjects, but still a
rather "simple repetitive" collection of tasks, such as card dealing
and reaction time. Later, a collection of serial search tasks
(Monk, 1979), in which subjects scan through material for
particular targets (e.g. every occurrence of the letter "e" in a
magazine article) again revealed a very good parallelism (in
speed) with the body temperature rhythm....
(Monk, 1989, p. 168)

Monk's research also determined that performance speed

parallels the temperature rhythm. However, when performance

accuracy was measured, performance was found to decline over the









day, reaching the lowest point of accuracy at the same time of the

day as the peak in performance speed. Thus, while performance

speed is increasing over the day, performance accuracy is declining

(Monk, 1989, p.169).

Colquhoun also conducted a study concerning speed in arithmetic

addition. This study concluded that, as the task became more

repetitive and simple, the time-of-day function in performance

speed became more like that of the temperature rhythm. Therefore,

time-of-day functions may change with practice.

For Monk (1989), these findings suggests that it may be wrong

to speak of performance as improving over the day, or to think of

time-of-day factors as representing changes in the brain's capacity to

process information. Monk believes that we should consider changes

in performance as being mediated by changes in strategy over the

day. Individuals probably approach information processing tasks

differently at one time of the day, and at one level of practice, as

compared to another. Measures may indicate better or worse

performance depending upon what type of measure we take (p. 170).

Monk also suggests that research on short-term memory, which

indicates a decline over the day, may also be a result of changes in

information processing strategy, rather than changes in the brain's

capacity to process information. "In several studies (Folkard, 1979;








Folkard & Monk, 1979; Folkard & Monk, 1985) a manipulation that

inhibited subvocalization was also found to eliminate the time-of-day

effect, suggesting that time-of-day differences may also result from

differences in subvocalization strategy" (p. 170).

According to the U.S. Congress, Office of Technology Assessment

(1991), in a diurnal condition, certain psychological rhythms are

evident, though they may be inconsistent.

\ It was found that, if speed of identification was assessed, the
usual relationship with temperature held--that is, subjects
became faster over the course of the day. However, if accuracy
of response became the benchmark, peak performance occurred
in the morning (Craig & Condon, 1985; Monk & Leng, 1986).
Short-term and long-term memory also appear to peak at
different times during the 24-hour cycle (Folkard & Monk,
1980). Motivation can influence performance, too. It has been
shown that when an incentive is offered, such as a significant
sum of money, circadian decrements in performance may be
overcome to some extent (Colquhoun, 1981; Home & Pettitt,
1985). The latter study (Homrne & Pettitt, 1985) indicated,
however, that sleep deprivation and circadian rhythms can
overcome even the strongest incentive influencing performance.
Further complicating the picture is the observation that
individuals are not always accurate judges of their own mood,
alertness, or ability. (p. 47)

Englund (1979) conducted a study of the diurnal function of

reading rate, comprehension and efficiency. The results of this study

indicate that the plotted means for these variables displayed distinct

diurnal patterns. Analysis of the data revealed that comprehension

was superior in the afternoon and early morning when compared to








scores at mid-morning and late at night. Performance speed

indicated a morning rise and afternoon fall. Oral temperature of the

subjects peaked at 15:27, with no differences being found between

males and females. Overall efficiency displayed no significant

differences, indicating a trade-off between speed and accuracy over

the day, which acts to maintain performance efficiency. "These

findings confirm Folkard's observations (Brit. J. Psychol., 1975), while

performance is task dependent, speed and accuracy measures may

represent different diurnal components of performance with

correspondingly different relationships to arousal level" (Englund,

1979, p. 96).

Taking all of this into consideration, Adan (1993) conducted a

literature review in order to classify the most commonly used

psychological tests, and to differentiate these tests on the basis of

variations obtained according to time-of-day factors. She indicates

that the incipient state of chronopsychology stems from two basic

facts. First, there is a great heterogeneity in the results obtained

which is a consequence of the diversity of methods used, the samples

of subjects and the situations in which the various experiments were

conducted. Second, most psychological investigations ignore the

time-of-day factor as both a procedural variable and as a means of

control in experiments. Adan further explains:







65
Daily fluctuation in behavioral parameters cannot be considered
trivial: the total variation detected is of the order of 10 %, and
"over the normal waking day (0800 to 0000) the circadian
variation in performance efficiency can be equivalent in
magnitude to the effect of limiting sleep to 3 hours or ingesting
the legal driving limit of alcohol" (Monk et aL, 1978, p. 166).
(p. 146)

Adan also indicates that other rhythms with different periodicities

(circamensual rhythms = greater than 24 hours, and ultradian

rhythms = less than 24 hours) may mask the results of studies

concerning circadian rhythms and, therefore, should be controlled.

Nevertheless, the classification proposed by Adan (1993) enables

us to differentiate, from a chronopsychological viewpoint, the most

commonly used tests based upon both the required skill of each test,

and the empirical criterion of the resultant circadian function. The

advantage of Adan's classification over traditional classifications is

that coherence is imposed on the heterogeneity of the existing results

for both subjective and objective tests.

The subjective tests yield data concerning the subjects' own

assessment of his/her state during the test. The two principal types

of tests include self-assessment inventories, and analogico-visual

scales. Patterns of alertness, as measured on the analogico-visual

scales, shows a peak alertness between 11:00 and 14:00, and

somnolence demonstrates an opposed two-phase pattern. The









pattern of drowsiness indicates high scores at 07:00-09:00 and at

21:00-23:00.

Objective tests consist of tasks which are generally normalized,

and are designed to make it difficult for subjects cheat. Subjects are

presented with indicators from which to chose as a criterion of

performance, and these differ as to whether the circadian function

being tested is quantitative or qualitative. Adan's classification has

completely changed the traditional dichotomization of vigilance and

cognitive tasks by subclassifying them into the categories of simple

and complex tasks. Simple tasks differ as to their motor component.

Tasks with a weak motor component include vigilance tasks and

signal detection and/or discrimination tasks. Signal detection and

discrimination tasks demonstrate a 4-hour advance of the optimum

time with respect to that of vigilance tasks. Of the high motor

component tasks, only those requiring manual skill demonstrate

circadian rhythmicity, and peak performance on these tasks occurs

between 12:00 and 14:00.

As with the majority of earlier studies, Adan subdivides complex

tasks into memory tasks and tasks of greater cognitive complexity.

Memory tasks demonstrate circadian variations which are most

pronounced in optimum time, and this is different in each case:

immediate memory 08:00-10:00, working memory 12:00-14:00, and









long-term memory 18:00-20:00. Very little data exist concerning

tasks of greater cognitive complexity, and results are inconclusive

(Adan, 1993, p. 154).

This seems to indicate that, for whatever reason, a rhythmicity

exists. Folkard and Monk (1980) indicate that short-term memory

generally peaks in the morning, and that long-term memory

generally peaks in the afternoon, thus, providing implications for the

timing of instruction according to the type of instructional objectives

involved.

Regardless of whether a direct link can be made between certain

variables and a circadian pacemaker, the aforementioned research

does demonstrate that temporal ordering of these variables does

exists and can be observed and predicted.

Consider the following reported data: the frequency of heart
attacks peaks between 6 a.m. and noon (Muller et al., 1985;
Rocco et al., 1987); Asthma attacks are most prevalent at night
(McFadden, 1988); human babies are born predominantly In the
early morning hours (Glattre & Bjerkedal, 1983; Kaiser &
Halberg, 1962). While these patterns do not necessarily indicate
that the events are driven by the circadian pacemaker, they do
suggest temporal order in the functioning of the human body.
(U.S. Congress, Office of Technology Assessment, 1991, p.41)

One study was conducted in order to determine the effect of the

timing of learning relative to the timing of sleep on memory

retention. Benson and Feinberg (1975) conducted a study concerning

the effects of sleep on memory by comparing the memory retention








of two groups. The two groups were divided in order to examine

retention of initial learning in the morning versus retention of initial

learning at night, at both 8 hours and 24 hours after the initial

learning was accomplished. The results indicated that retention was

superior 8 hours after learning for subjects who learned at night, as

compared to those who learned in the morning. Retention for

subjects who learned at night was equal after 24 hours to that

observed after 8 hours. A rather surprising finding in this study was

that retention scores for subjects who learned in the morning were

superior after 24 hours to those observed at 8 hours after initial

learning (p. 192).

A recent study (Shadmehr & Holcomb, 1997) indicates that it

takes about six hours for the memory of a new skill to move into

long-term memory storage. This study also indicates that practice of

a similar skill just after learning the new skill can interfere with the

long-term memory storage of the new skill, but that practice of a

non-similar skill may not interfere. These findings may have

implications for the surprising results obtained by Benson and

Feinberg.







69
Summary

This chapter has described human biological and psychological

circadian rhythms and how they are controlled by both endogenous

and exogenous factors. This chapter also described the detrimental

effects of circadian dysrhythmia, which are caused by jet lag or

rotating shift work, and the effects of circadian rhythms on learning.

The following chapter describes the methods and procedures used in

this study to collect and analyze the data, in order to address the

hypotheses presented in Chapter 1.












CHAPTER 3
METHODOLOGY

Participants

Thirty-seven participants (N = 37) were self-selected from the

Jacksonville, Florida, Air Route Traffic Control Center (ARTCC).

Although this sample is representative of the population of interest,

this number is less than the desired number of participants for this

type of study and may slightly limit the generalization of the

findings. Participants were recruited on a voluntary basis, so that

two groups (18 rapidly-rotating shift workers, and 19 day/evening

shift workers) were obtained. The population for this study

consisted of six female controllers (three in each group) and thirty-

one male controllers, all 45 years of age or younger. The majority of

the participants were in their late 20s or early 30s.

Instruments

The NovaScan computer-based performance test was the

instrument used to gather data for this study. This test was used to

measure five variables: (a) Dial monitoring reaction time during the

spatial-visualization task (attention allocation), (b) dial monitoring

reaction time during the tracking task (attention allocation),







71
(c) reaction time for transition from the tracking task to the spatial-

visualization task (flexibility in switching resources),

(d) non-transition reaction to object orientation (spatial

visualization), and (e) tracking error during the tracking task

(straight mind/motor coordination).

The NovaScan instrument randomly presents two computer-

based tasks for the participant to perform several times during each

session. Task 1 (spatial visualization) is a cognitive task in which the

participant must orient an airplane by selecting the wing which has

the circle above it like the one below the airplane. The circles are

either open or solid, and one wing has one of these above it, while

the other wing has the other. The participant must respond either

"left" or "right" to select the correct wing by pressing buttons on a

special keyboard. The reaction times for the participant's responses

are recorded in milliseconds.

Task 2 (tracking) is a psychomotor task in which the participant

must use a joy stick to move the computer's cursor left and right to

try to keep it below a horizontally-moving object. The tracking error

at this task is recorded, as the mean number of pixels that the cursor

was off target and outside the range of the area directly below the

moving object during the session.









While the participant is completing these two randomly-

presented tasks during a NovaScan session, he/she must

continuously monitor a dial in the upper right corner of the screen to

determine when the pointer on the dial moves into a solid area.

When this occurs, the participant must respond by pressing another

button on the special keyboard. The participant's responses are

recorded in milliseconds and represent the participant's attention

allocation rating during each of the two tasks.

The reliability measures for the NovaScan variables, obtained

from parametric studies, have been rated via intra-class correlation

as follows: Rotated object (.95), transition to object visualization task

(.94), and tracking error (.91). These measures are calculated by

subtracting the within variance from the between variance, and

dividing by the between variance (O'Donnell, 1992). The attention

allocation variable has only been used in the past for the purpose of

measuring an individual's current attention allocation against his/her

own baseline rating. It has never before been used to compare the

ratings of groups of individuals.

Materials/Apparatus

NovaScan requires a PC with 286 or faster processor, one MB of

RAM, a VGA graphic adapter (black and white or color), a hard-drive

disk (if desired), and DOS 3.3 or higher operating system. A special







73
keyboard is provided for responding. These minimum requirements

were met using two 386 MHz DOS computers. Each participant was

provided with a single 3.5 in floppy disk, which contained the

NovaScan software, so that all NovaScan sessions could be recorded

separately for each individual.

Design

A causal-comparative design was used for this study since the

predictor variable (work schedule) is already differentiated between

the two study groups, and because the cause (circadian dysrhythmia)

is already suspected and the effect (attention allocation) is under

investigation. Self-selected participants were categorized into two

groups. One group of participants was represented by individuals

who work day and evening two-shift schedules, and the other group

only included individuals who work a variety of rapidly-rotating

three-shift schedules which include night shifts.

The variables which were controlled include the following

considerations: (a) The rotating schedules included at least one

midnight shift and one day shift each week prior to the NovaScan

sessions, so that the daily routines of the individuals who worked

them were acutely shifted, (b) all of the participants were no more

than 45 years of age, so that the circadian effects of aging were not a

factor, (c) all testing periods on the NovaScan took place during the







74
day shift hours (8:00 am to 4:00 pm) for each participant, in order to

determine each individual's attention allocation during the hours that

training normally occurs, and (d) all participants indicated that they

were not taking any medications, such as cold remedies or

prescription drugs, which may have an effect on alertness or

attention allocation ratings. Controllers who take such medications

are not authorized to work operational positions in any event, so a

simple acknowledgement to this requirement was all that was

required.

Procedure

Each specialist completed no more than 6 sessions per day on

NovaScan, on no less than 7 separate days, between the hours of

8:00 am and 4:00 pm, until 25 sessions were completed. Parametric

studies for NovaScan indicate that a stable baseline measure for any

individual should be completed by the 20th session, and that massed

practice should be avoided by allowing at least 7 separate days for

practice.

Since the rotating shift workers only work two or three day

shifts each week, the investigator obtained a six-week temporary

duty assignment to the Jacksonville ARTCC in order to have enough

time for the purpose of participant recruitment and data collection.

The day and evening shift participants also kept a similar schedule








for their sessions on NovaScan by taking their sessions on two or

three days of each week.

The procedure described above was necessary because the

controllers could not be scheduled to take their sessions at

predesignated times. The controllers were all volunteer participants,

and they could only be requested to report for their sessions during

their break times if/when they were willing to do so. Because of this

slight design problem, certain trends may have been missed in the

analysis which could have provided more information.

Null Hypotheses

Data were collected and analyzed to address the following:

1. Air traffic controllers who work rapidly-rotating work schedules,

will show no significant difference from controllers who work day-

shift and/or evening-shift schedules on their NovaScan mean

baseline rating of attention allocation for dial monitoring during the

spatial visualization task (Task #1).

2. Air traffic controllers who work rapidly-rotating work schedules

will show no significant difference from controllers who work day-

shift and/or evening-shift schedules on their NovaScan mean

baseline rating of attention allocation for dial monitoring during the

tracking task (Task #2).







76
3. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of

controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of reaction time on dial

monitoring during the spatial visualization task (Task #1).

4. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of

controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of reaction time on dial

monitoring during the tracking task (Task #2).

5. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and Its residuals) which is

not significantly different from the mean learning curve of

controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of reaction time for the transition

from Task #2 to Task #1 (which indicates response flexibility in

switching resources).

6. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of









controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of reaction time on Task #1 non-

transition object orientation (which indicates spatial visualization).

7. Air traffic controllers who work rapidly-rotating work schedules

will indicate a mean learning curve (slope and its residuals) which is

not significantly different from the mean learning curve of

controllers who work day-shift and/or evening-shift schedules for

their NovaScan repeated measures of tracking error on Task #2

(which indicates mind/motor coordination).

Variables

The following criterian variables were measured and analyzed:

1. Attention allocation during Task 1 (spatial visualization).

2. Attention allocation during Task 2 (tracking).

3. Reaction time to transition from Task 2 to Task 1 (response

flexibility in switching resources),

4. Non-transition reaction time to Task 1 (spatial visualization).

5. Tracking error during Task 2 (tracking).

Data Analysis

In order to answer the hypotheses, the data were analyzed so

that the two groups of participants could be compared on two types

of variables: attention allocation, and learning. This was

accomplished in the following manner:








1. The two baseline measures of attention allocation for each

individual were calculated as the average reaction time on dial

monitoring during tasks 1 and 2 from the last 3 sessions taken on

NovaScan. Once all the participants had completed their 25th session

on NovaScan, a t-Test for independent samples was completed for

each measure with p=0.05 defined as statistically significant. The

results of these t-Tests were used to compare the two groups of

participants on each of these two measures and to address

hypotheses 1 and 2.

2. The learning curves for each individual were determined for the

five NovaScan variables: (a) Attention allocation during the spatial

visualization task, (b) attention allocation during the tracking task,

(c) response flexibility in switching resources, (d) non-transition

reaction to spatial visualization, and (e) tracking error during

tracking. The mean learning curve for each variable was determined

for each group to be used to compare the two groups of participants

and to address hypotheses 3 through 7.

A Two-Way General Linear Model, Repeated-Measures Analysis,

which included three statistical test (Group Main Effect, Time Main

Effect, and Group/Time Interaction) was used to analyze the data for

the five variables' mean learning curve slopes and their residuals

with p=0.05 defined as statistically significant, in order to compare







79
the two groups' progress over the sessions. Wherever a Group/Time

Interaction was found to be statistically significant, the Group Main

Effect and Group Time Effect were not reported (per rule). Also, for

each test that was found to be statistically significant, eta squared

(n2) was calculated in order to determine whether the significant

difference was slightly (12=0.01), moderately (-q2=0.06), or highly

(92=0.14) meaningful for the size of the sample used in this study.

According to Cohen (1988), eta squared is the proportion of total

variance in the outcome that is accounted for by the between group

variance. In short, eta squared determines if the significant

difference between the groups is meaningful, and a moderate eta

squared result (12=0.06), or higher, would meet this standard. The

learning curves for the two groups were superimposed and displayed

together for visual comparison on each variable, and each group's

mean learning curves were displayed individually with their error

bands (standard deviations).

Only the last three seconds (time period two) of each trial for

variables 2 (attention allocation during tracking) and 5 (tracking

error during tracking) were used in the calculation for these tasks, in

order to control for the variable state of the tracking task at each of

its initiations. For this same reason, variable 3 (response flexibility







80
in switching resources) was always reported as a transition from the

tracking task to the spatial visualization task, and not the other way

around.

SummarY

Chapter 3 has presented the methodology used in the data

collection process, the participants who volunteered for the study,

and the procedures used to collect the data. This chapter also stated

the hypotheses and variables addressed in the investigation and the

data analysis procedures used to determine the results. The

following chapter presents the research findings which resulted from

these methods and procedures, and Chapter 5 concludes with a

discussion of the results, the theoretical and practical implications of

this research, and recommendations for future research.











CHAPTER 4
RESEARCH FINDINGS


Results

Hypothesis 1

A statistically-significant difference was found between the two

groups' attention allocation baseline measure for dial monitoring

during Task 1 (spatial visualization). Group 1 (Rapidly-Rotating

Shift Workers) demonstrated a better attention allocation rating (a

faster reaction time for dial monitoring) during Task 1 (spatial

visualization) than Group 2 (Day/Evening Shift Workers)

demonstrated. The null hypothesis was rejected.

Hypothesis 2

A statistically-significant difference was found between the two

groups' attention allocation baseline measure for dial monitoring

during Task 2 (tracking). Group 1 (Rapidly-Rotating Shift Workers)

demonstrated a better attention allocation (a faster reaction time for

dial monitoring) during Task 2 (tracking) than Group 2 (Day/Evening

Shift Workers) demonstrated. The null hypothesis was rejected.








Hypothesis 3

A statistically-significant Group/Time Interaction with a small

effect size was found between the two groups' mean learning curves

for Variable 1: dial monitoring (attention allocation) during Task 1

(spatial visualization). Group 1 (Rapidly-Rotating Shift Workers)

demonstrated a decreasing learning curve with an improving

reaction time over the sessions, while Group 2 (Day/Evening Shift

Workers) demonstrated a fairly flat learning curve. Both groups

demonstrated stabilization around their mean baseline measures,

with no Group/Time Interaction between the groups' residuals. The

null hypothesis was rejected.

Hypothesis 4

No statistically-significant Group Main Effect, Time Main Effect,

or Group/Time Interaction was found between the two groups'

learning curve slopes for Variable 2: dial monitoring (attention

allocation) during Task 2 (tracking). The learning curve slopes for

the two groups appear statistically the same (fairly flat) and did not

improve to a statistically-significant degree over time. Both groups

demonstrated a stabilization around their mean baseline measure,

however, the groups' residuals demonstrated a statistically-

significant Group/Time Interaction with a moderate effect size,

indicating a difference in the groups' residuals when compared over









the sessions. The standard deviations around the means of the

slopes were not the same for the groups when compared over time

and the groups did not stabilize around their means at the same rate.

The null hypothesis was rejected.

Hypoothesis 5

No statistically-significant Group Main Effect or Group/Time

Interaction was found between the two groups' learning curves for

Variable 3: reaction time during the transition from Task 2 to Task 1

(response flexibility in switching resources), indicating that the

groups' mean learning curves appear statistically equal. There was,

however, a statistically-significant Time Main Effect with a large

effect size, indicating that both groups' learning curves depict

improved reaction times over the sessions. No error-band effects

were found to be statistically significant, indicating no statistically-

significant change in them over time, and no Group/Time Interaction

between the error bands. The null hypothesis was not rejected.

Hypothesis 6

No statistically-significant Group Main Effect or Group/Time

Interaction was found between the two groups' learning curves for

Variable 4: reaction time during Task 1 (spatial visualization),

indicating that the groups' mean learning curves appear statistically

equal. There was, however, a statistically-significant Time Main









Effect with a large effect size, indicating that both groups' learning

curves depict improved reaction times over time. No error-band

effects were found to be statistically significant, indicating no

statistically-significant change in them over time, and no Group/Time

Interaction between the groups' error bands. The null hypothesis

was not rejected.

Hvoothesis 7

No statistically-significant Group Main Effect or Group/Time

Interaction was found between the two groups' learning curves for

Variable 5: tracking error during Task 2 (tracking), indicating that

the groups' mean learning curves appear statistically equal. There

was, however, a statistically-significant Time Main Effect with a

moderate effect size, indicating that both groups' learning curves

depict improved tracking errors over time. No error-band effects

were found to be statistically significant, indicating no statistically-

significant change in them over time, and no Group/Time Interaction

between the error bands. The null hypothesis was not rejected.

Attention Allocation

Table 1 displays the results of the two comparisons of attention

allocation baseline for the two groups, based on the two t-Tests. The

first group comparison (attention allocation during the spatial

visualization task) indicates that the mean baseline measure for









Table 1. Summary of t-Test Results for Group Mean Attention
Allocation Comparisons in Milliseconds of Reaction Time

Group 1 Group 2

Variable n M SD n M S2


Task 1 18 616.57 107.15 19 778.64 140.91




(t=3.92. p=0.002. rpb=0.31)

Task 2 18 1548.42 55.27 19 1659.16 103.61




(t=4.02. p=0.0001. rpb=0.32)--
*p<0.05


Group 1 (Rapidly-Rotating Shift Workers) is 616.57 milliseconds of

reaction time, while the mean measure for Group 2 (Day/Evening

Shift Workers) is 778.64 milliseconds of reaction time. As

determined by the t-Test, this difference is statistically significant

[t(35)=3.92, p=0.002, rbp=0.31].

The second group comparison (attention allocation during the

tracking task) indicates that the mean baseline measure for Group 1

is 1548.42 milliseconds of reaction time, while the mean measure for

Group 2 is 1659.16 milliseconds of reaction time. As determined by









Table 2. Summary of Learning Curve Analyses


VRBL Fig. Test F Value p Value Eta Souared


1
6/7
6/7
6/7

2 2
2
2
8/9

3 3
3
3
10/11
10/11
10/11

4 4
4
4
12/13
12/13
12/13

5 5
5
5
14/15
14/15
14/15


Slope/GTI
Resdl/GME
ResdL/TME
Resdl/GTI

Slope/GME
Slope/TME
Slope/GTI
Resdl/GTI

Slope/GME
Slope/TME
Slope/GTI
Resdl/GME
Resdl/TME
Resdl/GTI

Slope/GME
Slope/TME
Slope/GTI
Resdl/GME
Resdl/TME
Resdl/GTI

Slope/GME
Slope/TME
Slope/GTI
ResdL/GME
Resdl/TME
Resdl/GTI


2.96
0.37
12.3
1.7

0.01
2.29
1.01
3.96

0.78
43.35
0.9
1.37
0.97
0.88

0.51
72.3
2.05
2.03
1.24
0.63

0.56
12.31
0.50
0.38
1.55
0.64


0.02
0.55
0.0001
0.11

0.94
0.13
0.34
0.05

0.38
0.0001
0.47
0.25
0.43
0.49

0.48
0.0001
0.09
0.16
0.30
0.66

0.46
0.0001
0.85
0.54
0.15
0.73


0.03

0.19





0.09


0.17






0.29






0.10


Abbreviations used in Table 2:


Resdl-Residual; GME=Group Main


Effect: TME=Time Main Effect: GTI=Group/Tlme Interaction
*p<0.05







87
the t-Test, this difference is also statistically significant [t(35)=4.02,

p=0.0001, rpb-0.32].

Learning Curve Comoarison

Table 2 describes a summary of results for the learning curve

analyses and how they relate to the learning curves displayed in

Figures 1 through 15. Figures 1 through 5 display the mean learning

curves for Groups 1 and 2 (superimposed) for each of the five

variables, while Figures 6 through 15 display each group's

independent learning curve for each variable with its error band

(standard deviation).

As indicated in Table 2, there is a statistically-significant

Group/Time Interaction between the groups' learning curves for

Variable 1, dial monitoring reaction time during Task 1 (spatial

visualization). The eta squared value for this interaction is 0.03,

which indicates a slight to moderate effect size. This interaction

implies that the slopes of the two groups' learning curves appear to

be different. Group 1 displays a decreasing slope, while Group 2

displays a fairly flat slope with little improvement over time. The

apparent difference between the two slopes can be seen in Figure 1,

and the statistically-significant difference which was found in the

baselines by the t-Test for the final three sessions is obvious.







88
There is no Group/Time Interaction for the groups' residuals on

Variable 1. The Group Main Effect test (Table 2) for the residuals

indicates that the error bands are statistically the same size for both

groups (Figures 6 and 7), while the Time Main Effect test

demonstrates a statistically-significant effect, indicating that the

error bands for both groups become smaller over the sessions. The

eta squared for the Time Main Effect is 0.19, indicating a large effect

size. It is apparent that the error bands become more stable around

the means over the sessions for both groups, but Group 1 appears to

stabilize slightly more than does Group 2.

The Group/Time Interaction comparison for Variable 2, dial

monitoring reaction time during Task 2 (tracking), produced a

statistically non-significant result, meaning that the slopes appear to

be statistically the same when compared over time. The slope

comparison for Groups 1 and 2 (Table 2) indicates that the slopes

statistically appear to be the same, with no statistically-significant

Group Main Effect (Figure 2). There is also no statistically-significant

Time Main Effect, Indicating that the slopes did not increase or

decrease to a statistically-significant degree over the sessions. The

statistically-significant difference which was found between the two

groups' baselines by the t-Test for this variable is apparent, but not

as obvious as with Variable 1.











- 4 -- Group 2


1000

O A-*


400


0
I 3 5 7 9 11 13 15 17 19 21 23 25
Session


Figure 1. Mean learning curves for dial monitoring reaction time
during Task 1 (Spatial Visualization).


I 3 5 7 9 II 13 15 17 19 21 23 25
1Session

Figure 2. Mean learning curves for dial monitoring reaction time
during Task 2 (Tracking).










2500 --G roup 2
-- 4- Group I




I1500


1000
Soo .- " .. " -.


300



I 3 5 7 9 11 13 15 17 19 21 23 25
Sesston

Figure 3. Mean learning curves for reaction time to transition from
Task 2 (Tracking) to Task 1 (Spatial Visualization).


Figure 4. Mean learning curves for reaction time to Task 1 (Spatial
Visualization).









isa 4Group2
350 Group 2 '




200




50
0
I 3 5 7 9 II 13 15 17 19 21 23 25
Session

Figure 5. Mean learning curves for tracking error during Task 2
(Tracking).


The residual tests (Table 2) for Variable 2 demonstrates a

statistically-significant Group/Time Interaction with a moderate

effect size (=i-0.09) for the residuals (Figures 8 and 9). This indicates

that the size of the error bands decrease at statistically different

rates over time for the two groups.

The statistical analyses (Table 2) for Variables 3 (reaction time

during transition from Task 2 to Task 1), 4 (reaction time during

Task 1), and 5 (tracking error) all had similar results. There was no

Group/Time Interaction between the groups on any of these three

variables, indicating that the slopes of the groups appear statistically

the same over the sessions in each case. As Figures 3, 4, and 5

























I 3 5 7 9 II 13 15 17 19 21 23 25
Sesion

Figure 6. Group 1 mean learning curve with error band (standard
deviation) for dial monitoring reaction time during Task 1 (Spatial
Visualization).


1 3 5 7 9 11 13 15 17 19 21 23 25
L SeSion
Figure 7. Group 2 mean learning curve with error band (standard
deviation) for dial monitoring reaction time during Task 1 (Spatial
Visualization).
























1 3 5 7 9 II 13 15 17 19 21 23 25
Session

Figure 8. Group 1 mean learning curve with error band (standard
deviation) for dial monitoring reaction time during Task 2 (Tracking).







2500




11000
1500 4


31000

500

0
I 3 7 9 II 13 15 17 19 21 23 25
Session

Figure. Group 2 mean learning curve with error band (standard
deviation) for dial monitoring reaction time during Task 2 (Tracking).




Full Text
110
considered time-of-day as a control variable. Due to this, Adan
suspects that at least some research results are erroneous.
Although the results of this study are contrary to expectation,
the fact that statistically-signiflcant differences were found between
the groups lends support to the possibility that the
chronopsychological effects of shift work may affect learning. The
rapidly-rotating shift workers not only demonstrated significantly
better attention allocation baseline measures than did the day and
evening shift workers, they also demonstrated an ability to improve
their attention allocation ability during the learning of a cognitive
task, while the day and evening shift workers did not. These results
add to earlier learning-related research findings in chronopsychology
(Englund, 1979; Folkard & Monk, 1980; Monk, 1989), and speak of a
need for the consideration of the chronopsychological effects of shift
work on learning in instructional theory, and instructional design.
These results also indicate that certain types of shift work may
affect attention on the job. Because shift work schedules may affect
individuals attention allocation ability, research in situation
awareness should consider shift schedule as a control variable in
future studies.


41
schedule, or time off. Workers more often bargain for time off rather
than for monetary compensation (p.3).
Melton and Bartanowicz (1986) also described the weekly-
rotating schedule. This work schedule, the straight five pattern,
allows individuals to work the same shift for five days, have 48
hours off, and then begin the next, usually later, shift. Research by
the Federal Aviation Administration, and others, has demonstrated
that the five mid watches in a row is the wrong type of shift rotation
to use, from the standpoint of circadian rhythmicity. The five
straight mids leads to accumulated sleep loss and fatigue because
day sleeps are cut short due to noise levels, light intensity, and the
phase of the circadian temperature rhythm and other physiological
rhythms. The sleep that is attained is mainly fatigue-induced sleep,
which is usually brief and inadequate (p. 5).
According to Akerstedt (1988), schedules which rotate should do
so with consideration being given to the direction of rotation, if they
are to be beneficial to the health and performance of the worker.
Another important aspect of the shift schedule may be its
direction of rotation. Since the free-running period of the
human sleep/wake cycle averages 25 h, and since it can be
entrained by environmental time cues only within 1-2 h of the
free-running period, phase delays are easier to accomplish than
phase advances (Czeisler et al., 1980). For the rotating shift
worker, this implies that schedules that delay,i.e., rotate
clockwise (morning-afternoon-night) should be preferred to
those that rotate counterclockwise. However, there have been


49
complaints given by controllers on the mid shift, followed by those
same complaints in decreasing order of frequency on day and
evening shifts (Saldivar etal., 1977, p. 5).
Individuals who work rapidly-rotating schedules have their
sleep cut short during day sleep primarily because of two reasons.
First, there are higher noise levels usually present in the morning
hours after night work is ended and the normal daytime activities of
society begin. Second, individuals attempting to sleep in the mid- or
late-morning hours are also attempting to sleep while their diurnal
temperature rhythm is rising. Normally, sleep onset occurs in the
evening, when the body temperature rhythm is on the decline. Day
sleep is, therefore, shortened even if a sleep deficit exists. This in
turn helps lead to sleepiness, especially on the night shift.
Sleepiness occurring during night work seems to be the result of a
combination of influences: the circadian trough of the alertness
rhythm, and sleep loss/time awake (Akerstedt, 1988, p. 26).
It has been estimated that most individuals who work shift work
never adapt to it. Akerstedt and Kecklund (1991) completed a two-
year study of rapidly-rotating shift workers at a paper mill. The
results suggest that sleep has a traitlike character, even when
occurring at daytime after a night shift. This, therefore, suggests that


CHAPTER 2
RELATED LITERATURE
Circadian Rhythms
Circadian rhythms are cycles which occur on a daily basis, but
they are only part of a larger system of rhythms.
We live in a universe of rhythms. Our galaxy rotates once every
200 years. Sun spots occur every 11 years and tides every 12
1/2 hours. But the most significant rhythm for the inhabitants
of this planet is the earths rotation every 24 hours.
(Hawkins, 1987, p. 51)
Cycles which are shorter than 24 hours are called ultradian, and
cycles which are greater than 24 hours are called circamensual
(Adan, 1993, p. 146). Organisms which live on our earth have
evolved within this universal, cyclic environment.
Human circadian rhythms are both physiological and
psychological in nature. The study of physiological (or biological)
rhythms is known as chronobiology, and the study of psychological
(and behavioral) rhythms is known as chronopsychology.
The biological rhythms include such variables as hormone
production, body temperature, organ functioning, immune system
cycles, blood cell functioning, and stomach and intestinal tract
functioning (U.S. Congress, Office of Technology Assessment, 1991).
18


72
While the participant is completing these two randomly-
presented tasks during a NovaScan session, he/she must
continuously monitor a dial in the upper right corner of the screen to
determine when the pointer on the dial moves into a solid area.
When this occurs, the participant must respond by pressing another
button on the special keyboard. The participants responses are
recorded in milliseconds and represent the participants attention
allocation rating during each of the two tasks.
The reliability measures for the NovaScan variables, obtained
from parametric studies, have been rated via intra-class correlation
as follows: Rotated object (.95), transition to object visualization task
(.94), and tracking error (.91). These measures are calculated by
subtracting the within variance from the between variance, and
dividing by the between variance (ODonnell, 1992). The attention
allocation variable has only been used in the past for the purpose of
measuring an individuals current attention allocation against his/her
own baseline rating. It has never before been used to compare the
ratings of groups of individuals.
Materials/Apparatus
NovaScan requires a PC with 286 or faster processor, one MB of
RAM, a VGA graphic adapter (black and white or color), a hard-drive
disk (if desired), and DOS 3.3 or higher operating system. A special


14
learning curves for the participants. The NovaScan instrument was
designed and validated to measure an individuals performance
against his/her own baseline of attention allocation, once it has been
established, in order to determine readiness for duty. It has also
been validated for establishing these ratings of attention allocation,
and for establishing both individual and group learning curves for
the variables which it measures (ODonnell, 1992, p. 19). This was
the first time that this instrument was used to compare the attention
allocation baselines of two groups of individuals who are expected to
differ on this variable according to literature. Evaluation of these
ratings determine if one group displayed a higher mean attention
allocation baseline rating, as determined by NovaScan, than the
other group.
Because NovaScan has been validated for measuring attention
allocation, it has logical (or face) validity for use in this study. It also
has construct validity because it has been validated through research
for its extreme sensitivity to decrements caused by drugs or alcohol
(ODonnell, 1992, p. 26).
The learning curve data were used to make determinations
concerning any possible learning effects between the groups
compared in this study. The criterian variables measured by
NovaScan in this study determined each participants learning curve


3
not reported due to a lack of sufficient data, the data that were
collected indicated a significant negative initial learning effect for the
controllers while they were on the midnight shift relative to when
they were on the day or evening shifts. Pilcher and Huffcut (1996)
found, through a meta-analysis of sleep-deprivation studies, that
sleep deprivation has its greatest effect on mood with a lesser effect
on cognitive performance and motor performance in descending
order. They also determined that partial sleep deprivation (sleep
loss which occurs whenever there is a reduction in the usual amounts
of sleep obtained in a 24-hour period), rather than long-term or
short-term sleep deprivation, has a greater negative effect on mood
and cognitive performance. This Is important to shift workers,
because partial sleep deprivation results from the condition known
as shift-work insomnia which, in turn, results from working
irregular or rotating schedules.
Pilcher and Huffcut also indicate that one clear goal of future
research will be to determine why partial sleep deprivation has such
a pronounced effect on mood and cognitive performance.
For example, partial sleep deprivation may alter certain
circadian rhythm effects on performance and mood. While total
sleep deprivation has been found to interact with circadian
rhythms (Monk et al., 1985; Naitoh et al., 1985), few studies
have investigated the effects of partial sleep deprivation on
circadian rhythms. In addition, partial sleep deprivation may be
similar to fragmented sleep in that subjects in both cases obtain


CHRONOPSYCHOLOGICAL LEARNING EFFECTS
OF RAPIDLY-ROTATING SHIFT WORK
ON DAY-SHIFT ATTENTION
By
RAYMON M. MCADARAGH
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULLFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1999


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
LarryLoesch
Professor of Counselor Education
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Assistant Professor of Foundations of
Education
This dissertation was submitted to the Graduate Faculty of the
College of Education and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
May 1999
Dean, College o^fcducation
Dean, Graduate School


42
very few practical tests of this theory, particularly in relation to
sleepiness. Still, Czeisler et al. (1982) have demonstrated that a
change from counterclockwise to clockwise rotation, together
with a change from 7-day to 21-day rotation, improved
productivity and wellbeing in three-shift workers. Orth-Gomer
(1983) found that a change in the same direction in police
officers on rapid (1-day) rotation reduced blood pressure and
improved wellbeing, (p. 27)
McAdaragh (1995) conducted a study of 997 air traffic
controllers (ATCs) work schedules from randomly selected ATC
facilities in order to determine what types of schedules controllers
tend to work. It was determined that 45% of the controllers were
working either phase-advance, compressed schedules of the 2-2-1
variety, or an alternate work schedule which even further
compresses the work week into four 10-hour days. Only 24% of the
controllers were working straight days or straight evenings, whereas,
the remaining 31% were working a variety of none-standard rotating
schedules. None of the controllers were working the 1-2-2 phase-
delayed schedule or anything like the schedule proposed by Czeisler,
which was reported to improve the production and wellbeing of
rotating shift workers.
Sleepiness
According to the U.S. Congress, Office of Technology Assessment
(1991), shift work can cause a variety of negative psychological and
physiological effects. This is attributed to the fact that the shift


66
V
pattern of drowsiness indicates high scores at 07:00-09:00 and at
21:00-23:00.
Objective tests consist of tasks which are generally normalized,
and are designed to make it difficult for subjects cheat. Subjects are
presented with indicators from which to chose as a criterion of
performance, and these differ as to whether the circadian function
being tested is quantitative or qualitative. Adans classification has
completely changed the traditional dichotomization of vigilance and
cognitive tasks by subclassifying them into the categories of simple
and complex tasks. Simple tasks differ as to their motor component.
Tasks with a weak motor component include vigilance tasks and
signal detection and/or discrimination tasks. Signal detection and
discrimination tasks demonstrate a 4-hour advance of the optimum
time with respect to that of vigilance tasks. Of the high motor
component tasks, only those requiring manual skill demonstrate
circadian rhythmicity, and peak performance on these tasks occurs
between 12:00 and 14:00.
As with the majority of earlier studies, Adan subdivides complex
tasks into memory tasks and tasks of greater cognitive complexity.
Memory tasks demonstrate circadian variations which are most
pronounced in optimum time, and this is different in each case:
immediate memory 08:00-10:00, working memory 12:00-14:00, and


Materials/Apparatus 7 2
Design 73
Procedure. 74
Null Hypotheses 75
Variables 77
Data Analysis 77
Summary. 80
4 RESEARCH FINDINGS 81
Results. 81
Hypothesis 1 81
Hypothesis 2 81
Hypothesis 3 82
Hypothesis 4 82
Hypothesis 5 83
Hypothesis 6 83
Hypothesis 7 84
Attention Allocation 84
Learning Curve Comparisons 87
5 CONCLUSION 98
Introduction 98
Discussion. 101
Implications 109
TheoreticaL 109
PracticaL Ill
Recommendations 112
REFERENCES 118
BIOGRAPHICAL SKETCH 125
IV


117
deficiencies during training may have been involved. It may be
possible to determine this by reviewing the reasons listed for the
errors on the accident reports. Some of the reasons listed, such as
brain lock could be associated with cognitive conflict caused by the
retrieval of inappropriate procedures, or the automation of faulty
mental models or scripts for the situation at hand. Psychologist may
be able to decipher these terms and associate them with possible
cognitive causes. These faulty mental models or scripts could have
developed as the result of an attention decrement during training,
only to be generated later in dynamic control situations through
automatic processing (automation), with little or no conscious
awsareness of them.
Presently, situation awareness is considered to be the primary
variable involved in controller error, but situation awareness is a
broad term and it involves many factors, including training (Endsley,
1995). Further research in situation awareness, including the type
described here, will help to isolate and identify the variables which
affect situation awareness. Once these variables have been
identified, predictive planning measures may be taken to address
them early and avoid the negative consequences of not addressing
them before they become factors in dynamic situations.


115
term memory, and reading rate and comprehension peak at different
times of the day in diurnal situations. Future studies could be
designed to determine how these factors are affected by different
shift systems. For example, groups of individuals on different shift
systems could be given different types of learning tasks
(psychomotor, memorization, comprehension, etc.) at different times
of the day in order to determine if and how the different shift
systems affect the ability of individuals to learn these tasks. This
type of research could help to determine the rhythmic fluctuations
associated with different learning tasks for individuals on different
shift systems. The generalizations derived from this type of research
could then become the learner characteristics predicted to be
associated with individuals on these different shift systems.
The baseline measures for Variables 3 (response flexibility at
switching resources), 4 (reaction time for the spatial visualization
task), and 5 (tracking error during the tracking task) were not
compared between the two groups in this study, because these
variables concern dynamic performance ability. This study was
mainly concerned with the learning-curve data on these variables, in
order to determine if any learning differences could be found to exist
between the groups. As indicated in the results, no learning
differences were indicated. However, these type of performance


30
Monk et al. (1988) also conducted a study of the affects of an
acute shift in routine with 8 middle-aged male participants. In this
study, researchers induced jet lag in a laboratory setting by
advancing the routine of the participants by 6 hours. As stated
earlier, the effects of jet lag would be the same as those experienced
by changes in routine due to changing work hours in shift work.
The eight participants in this study were kept in temporal
isolation for the 15-day experiment. After the participants were
entrained to their own habitual routines for five days, each
participant experienced an acute 6-hour phase advance in routine
which was brought about by a shortening of the sixth sleep episode.
The participants were then held to the new phase-advanced routine
for the ten remaining days of the experiment.
Following the shift in routine, the participants demonstrated
significant symptoms of jet lag which were observed in mood,
performance efficiency, sleep, and physiological temperature
rhythms. Some of the variables, such as temperature phase and
percent rapid-eye-movement sleep, showed a monotonic recovery
pattern. However, other variables, such as actual sleep duration,
percent slow-wave sleep, motivation loss, and subjective sleepiness,
demonstrated a zig-zag recovery pattern, which suggest the
interaction of two competing processes (p. 703).


60
generalizations about psychological circadian rhythms that are valid
for the population as a whole (Monk, 1989, p. 166). Monk goes on to
say that the link between subjective alertness and the circadian
system makes sense and can be thought of in terms of psychological
rhythms acting as an interface between the physiological circadian
processes and the sleeping and waking behaviors that those
processes influence (p. 166).
Monks own research in forced desynchronization studies in
1987 (as cited in Monk, 1989) determined the relationship between
the temperature rhythm and the subjective alertness rhythm. Monk
has determined that, when the sleep-wake cycle is driven at a point
which is beyond the range of the endogenous circadian oscillator
controlling the temperature rhythm, the oscillator assumes a period
length which is close to the free-running 25-hour cycle. This causes
two periodicities to be present in the data, that of the sleep-wake
cycle, and that of the temperature rhythm. Examination of the data
has demonstrated that both periodicities influence the subjective
alertness rhythm. A parallelism between the alertness rhythm and
the temperature rhythm exist at the temperature rhythm oscillator
period, and the maximal subjective alertness was found, on the
sleep-wake cycle, to occur from 8 to 10 hours after waking. These
data coincide with the early afternoon peak in alertness which is


24
These individuals are not synchronized by societys social cues. Just
how social cues influence circadian timing is not known, but the
social cues effects on arousal states are probably the necessary
mediators for the relationship (Mistlberger & Rusak, 1989, p. 146)
Dinges (1989) describes how sleep is affected by the circadian
timekeeping system. According to Dinges, the daily sleep-wake cycle
serves to organize wake behavior into discrete temporal units and to
coordinate and synchronize the internal timing of many biological
rhythms. We cannot know for sure whether early mammals were
diurnal or nocturnal, but there is little doubt that primates, and
homo sapiens in general, evolved as diurnal species who sleep
primarily at night (p. 153).
Dinges (1989) states that recent sleep research shows us that the
infrastructure of sleep itself might depend less on sleep and more on
a rhythmic process common to sleep and wakefulness. This sleep-
wake cycle is also affected by sometimes contrasting factors.
To begin with, unlike many other physiological parameters such
as body temperature, the sleep cycle is not a unitary process
that varies along a continuum; sleep is distinct from
wakefulness, sleep onset contrasts with sleep offset, and there
are different stages of sleep. Even more problematic, however,
is the fact that among our species, the initiation and termination
of sleep are influenced as much by sociopsychological factors as
by endogenous biological factors, (p. 156)


REFERENCES
Adan, A. (1993). Circadian variations in psychological measures:
A new classification. Chronobiologia. 20, 145-162.
Akerstedt, T. (1988). Sleepiness as a consequence of shift work.
Sleep. 11(1). 17-34.
Akerstedt, T & Gillberg, M. (1980). The circadian variation of
experimaentally displace sleep. Sleep. 4(2). 159-169.
Akerstedt, T., & Kecklund, G. (1991). Stability of day and night
sleep: A two-year follow-up of EEG parameters in three-shift
workers. Sleep. 14( 16), 507-510.
Akerstedt, T. & Torsvall, L. (1978). Experimental changes in shift
schedules--their effects on well-being. Ergonomics. 21. 849-856.
Akerstedt, T., Torsvall, L, & Froberg, J. E. (1983). A
questionnaire study of irregular work hours and sleep/wake
disturbances. Sleep Research. 12. 358.
Babkoff, H Caspy, T & Mikulincer, M. (1991). Subjective
sleepiness ratings: The effects of sleep deprivation, circadian
rhythmicity, and cognitive performance. Sleep. 14(6). 534-539.
Benson, K. & Feinberg, I. (1975). Sleep and memory: Retention 8
and 24 hours after initial learning. Psychophysiology. 12(2). 192-
195.
Billiard, M., Alperovitch, A., Perot, C., & Jammes, A. (1987).
Excessive daytime somnolence in young men: Prevalence and
contributing factors. Sleep. 10(4). 297-305.
118


theory and research in the cognitive and behavioral sciences provide
many principles which are used in the design of instruction, and
instructional design addresses many variables in order to accomplish
these instructional objectives. The design process takes into account
learning modalities, styles, and strategies, while giving a great
amount of attention to individual learner characteristics,
instructional settings, instructional media selection and learning
activity development. These variables are all considered during the
design process in order to maximize learning for the target audience.
Research in chronopsychology addresses such variables as short
term and long-term memory, attention and alertness, subjective
fatigue and mood, and cognitive performance issues such as the
ability to synthesize information, all of which express rhythmic
cycles in humans. Even in instructional settings where learners have
regular diurnal circadian rhythms, these variables have great
implications for the design of instruction. But when it is taken into
consideration that many industrial, military, and government
workers work shift work and are susceptible to its consequences, it
becomes apparent that research in chronopsychology offers a wealth
of information and research opportunity to the field of instructional
design. Even though this is the case, at present there is very little, if


50
shift work insomnia is a persistent problem-those afflicted tend to
remain so for some time (p. 509).
According to a review by the U.S. Congress, Office of Technology
Assessment (1991),
Some researchers (Folkard et al 1985; Knauth et al., 1981)
maintain that, regardless of rotation length, complete
realignment of circadian cycles never actually takes place in
shift work due to the competing influences of cues in the society
telling workers that they are on an abnormal schedule. Even in
permanent night shift work, if the worker does not maintain the
same schedule on days off as on workdays, he or she will rapidly
revert to the natural diurnal orientation (Monk, 1986; Van Loon,
1963). In such circumstances, the circadian system never fully
synchronizes to the night schedule, (p. 91)
Paley and Tepas (1994) conducted a study involving firefighters
working an 8-hour, rotating (2 weeks per shift) schedule. The
results of this study demonstrated that firefighters working this type
of schedule will sleep less than normal and will report lower positive
mood scores, higher negative mood scores, and higher ratings of
sleepiness on the night shift. The results also demonstrated that,
over the two-week course of the shift, the firefighters were unable to
adapt to changes in their sleep schedule (p. 269).
Research over the past two decades including, Akerstedt &
Torsvall, 1981; Rutenfranz et al., 1976; Tepas et al., 1985; Tepas &
Carvalhis, 1990; Williamson & Sanders, 1986 (as cited in Paley &
Tepas, 1994), indicates that, workers on the afternoon/evening


109
schedules as compared to controllers who worked day and evening
shift schedules.
The Chi-Square analysis included the following factors. The total
number of controller errors recorded at the Jacksonville ARTCC
facility over the 18-month period was 47. Based on the number of
controllers working each type of schedule, 28.67 of these errors
should have been committed by the day and evening shift workers,
and 18.33 errors should have been committed by the rapidly-
rotating shift workers, if the chances of committing an error were
equal for both groups. The actual number of errors committed by
each group was discovered to be 24 errors by the day and evening
shift workers, and 23 errors by the rapidly-rotating shift workers.
Based on these figures, the result of the Chi-Square analysis was
found to be statistically non-significant. Neither group committed a
higher number of errors considered to be statistically significant.
Implications
Theoretical
Research in chronopsychology has demonstrated that human
psychological and behavioral circadian rhythms are affected by shift
work and transmeridian flight. As Adan (1993) indicates, circadian
effects are not trivial, and most psychological research has not even


20
1987). Here, the most powerful of the entraining agents (or
zeitgebers), the cycle of light and darkness, is detected through the
eyes and influences the bodys visceral system and other cyclic
functions. Other rhythms, however, like that of the body
temperature rhythm, are controlled elsewhere and remain the object
of further investigation. Activities, such as meal time, and physical
and social activity, also serve as entraining agents (Hawkins, 1987).
According to Mistlberger and Rusak (1989), psychological and
biological rhythms have an internal relationship with time-of-day
restraints. Human behavior and physiological processes demonstrate
a temporal structure that matches the 24-hour day-night cycle. The
most obvious daily cycle that humans experience is the cycle of sleep
and wakefulness. Similarly, a myriad of bodily functions, including
endocrine secretions, body temperature regulation, sensory
processing, and cognitive performance, demonstrate a 24-hour
rhythmicity (p. 141).
Historically, circadian rhythms were ascribed to environmental
cues associated with the solar day, but a large number of studies
have confirmed that daily rhythms in a wide variety of species,
including humans, persist under constant environmental conditions.
Because of this, it is acknowledged that daily rhythms are generated


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Lee Mullally, Chair /
Associate Professor of Instruction and
Curriculum
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Jeffrey Hurt
Associate Professor of Instruction and
Curriculum
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy. _
Jti It
Assistant Professor of Instruction and
Curriculum


11
baseline rating of attention allocation for dial monitoring during the
spatial visualization task (Task #1).
2. Air traffic controllers who work rapidly-rotating work schedules
will show no significant difference from controllers who work day-
shift and/or evening-shift schedules on their NovaScan mean
baseline rating of attention allocation for dial monitoring during the
tracking task (Task #2).
3. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of reaction time on dial
monitoring during the spatial visualization task (Task #1).
4. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of reaction time on dial
monitoring during the tracking task (Task #2).
5. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of


LD
1780
1993
UNIVERSITY OF FLORIDA
3 1262 08555 0142


67
long-term memory 18:00-20:00. Very little data exist concerning
tasks of greater cognitive complexity, and results are inconclusive
(Adan, 1993, p. 154).
This seems to indicate that, for whatever reason, a rhythmicity
exists. Folkard and Monk (1980) indicate that short-term memory
generally peaks in the morning, and that long-term memory
generally peaks in the afternoon, thus, providing implications for the
timing of instruction according to the type of instructional objectives
involved.
Regardless of whether a direct link can be made between certain
variables and a circadian pacemaker, the aforementioned research
does demonstrate that temporal ordering of these variables does
exists and can be observed and predicted.
Consider the following reported data: the frequency of heart
attacks peaks between 6 a.m. and noon (Muller et al., 1985;
Rocco et al., 1987); Asthma attacks are most prevalent at night
(McFadden, 1988); human babies are born predominantly in the
early morning hours (Glattre & Bjerkedal, 1983; Kaiser &
Halberg, 1962). While these patterns do not necessarily indicate
that the events are driven by the circadian pacemaker, they do
suggest temporal order in the functioning of the human body.
(U.S. Congress, Office of Techno logy Assessment, 1991, p.41)
One study was conducted in order to determine the effect of the
timing of learning relative to the timing of sleep on memory
retention. Benson and Fein berg (1975) conducted a study concerning
the effects of sleep on memory by comparing the memory retention


31
The results of this study indicated that the disruption in the
amplitude of the circadian temperature rhythm was the most
dramatic circadian effect, and this effect lasted for several days
following the phase shift.
For instance, even 6 days after the phase shift, the mean change
in amplitude of each subjects temperature rhythm was 26%
below his own baseline level. This reduction in amplitude may
bespeak a desynchrony within the circadian system. Normally,
the component parts of the circadian system are in the correct
phase relationship to each other, combining to produce robust,
well-defined circadian rhythms. After an acute shift in routine,
however, these component parts do not adjust their temporal
orientation at the same rate, and appropriate phase relationships
are thereby lost. This has been referred to as internal
dissociation, a state shown to be associated with impairments in
mood and performance efficiency (Wever, 1975; Wever, 1979).
(p.708)
Among the variables showing a zig-zag recovery function were
the subjective alertness/sleepiness rhythm and the motivation loss.
Some of the post-shift days indicated worse effects than those on the
day immediately preceding them (p. 709).
Akerstedt and Gillberg (1980), in a study that they conducted
with six male subjects in temporal isolation, demonstrated that sleep
termination may be closely related to the sleepiness/alertness
rhythm. In this study, sleep times were displaced to seven different
times of the day through a rate of one sleep condition change per
week. The length of each sleep episode varied according to when it


80
in switching resources) was always reported as a transition from the
tracking task to the spatial visualization task, and not the other way
around.
Summary
Chapter 3 has presented the methodology used in the data
collection process, the participants who volunteered for the study,
and the procedures used to collect the data. This chapter also stated
the hypotheses and variables addressed in the investigation and the
data analysis procedures used to determine the results. The
following chapter presents the research findings which resulted from
these methods and procedures, and Chapter 5 concludes with a
discussion of the results, the theoretical and practical implications of
this research, and recommendations for future research.


35
The results of this study indicated that gradual changes
continued for four weeks. The adult males blood pressure rhythm
delayed in order to recover, while the adult females rhythm
advanced. Approximately four weeks were needed for adjustment of
circaseptan rhythms in the two adults, but it only took two weeks in
the boys.
The adjustment of rhythms during childhood appears to be
faster not only with circaseptans, but also with circadian rhythms.
The results also indicated that the heart rate rhythms adjusted more
rapidly than the blood pressure rhythms, and that the boys
circadian rhythms adjusted faster than the adults circadian rhythms.
The older boys rhythms adjusted more rapidly than the younger
boys rhythms. In the younger boy, the adjustment showed great
lability which might suggest that the circadian rhythmicity is still
developing and immature (p. 73).
The National Aeronautics and Space Administration (NASA) has
developed the NASA Ames Fatigue Countermeasures Program to
compile research concerning fatigue caused by such factors and to
develop countermeasures. According to Rosekind (1993),
More than a decade of research at NASA Ames on pilot fatigue,
sleep, and circadian rhythms has identified new insights into
crew fatigue. Some basic findings include:
* Sleep loss and circadian disruption from long-haul flight
operations can result in fatigue, increased sleepiness, and
reduced performance.


61
usually found. Monk goes on to say that the relationship between
subjective alertness and body temperature can be explained by
postulating control by both the temperature rhythm oscillator and
the sleep-wake mechanism, with the sleep-wake mechanism tending
to have more dominant control under normal, diurnal sleep-wake
patterns (Monk, 1989, p. 167).
In addition to the subjective measures of alertness, there are
also many circadian rhythms which lend themselves to objective
measures of performance efficiency. Some performance rhythms run
parallel to the circadian temperature rhythm and can be measured in
this manner.
Colquhoun (Colquhoun, 1971) studied vigilance tasks (detection
of an infrequent signal), simple addition tasks (adding up 6 two-
digit numbers), and other simple reaction-type tasks, finding
them to parallel (in measures of speed) the circadian
temperature rhythm, even during the process of adjustment to a
change in routine. This observation confirmed earlier studies by
Kleitman (Kleitman, 1963), which used fewer subjects, but still a
rather simple repetitive collection of tasks, such as card dealing
and reaction time. Later, a collection of serial search tasks
(Monk, 1979), in which subjects scan through material for
particular targets (e.g. every occurrence of the letter e in a
magazine article) again revealed a very good parallelism (in
speed) with the body temperature rhythm....
(Monk, 1989, p. 168)
Monks research also determined that performance speed
parallels the temperature rhythm. However, when performance
accuracy was measured, performance was found to decline over the


120
Folkard, S. & Monk, T. H. (1979). Shiftwork and performance.
Human Factors. 21, 483-492.
Folkard, S., & Monk, T. H. (1980). Circadian rhythms in human
memory. British lournal of Psychology. 71. 295-307.
Folkard, S., Monk, T. H & Lobban, M. C. (1978). Short and long
term adjustment of circadian rhythms in permanent night nurses.
Ergonomics. 21. 785-799.
Gander, P. H., Myhre, G., Graeber, R. C., Anderson, H. T & Lauber
J. K. (1989, August). Adjustment of sleep and the circadian
temperature rhythm after flights across nine time zones. Aviation.
Space & Environmental Medicine. 60(8). 733-743.
Hawkins, F. H. (1987). Human factors in flight (rev, ed.). England:
Gower Technical Press.
Higgins, E. A Chiles, W. D., McKenzie, J. M Iapietro, P. F., Winget,
C. M., Funkhouser, G. E., Burr, M. J., Vaughan, J. A., & Jennings, A. E.
(1975). The effects of a 12-hour shift in the wake-sleep cycle on
physiological and biomedical responses and on multiple task
performance (Report No. FAA-AM-75-10). Washington, DC: Civil
Aeromedical Institute, Office of Aviation Medicine, Federal Aviation
Administration.
Klien, K. E Wegmann, H. M., Athanassen, G., Hohlweck, H., &
Kuklinski, P. (1976, March). Air operations and circadian
performance rhythms. Aviation. Space, and Environmental Medicine.
132-137.
Knutsson, A., Johnson, B. G., Akerstedt, T., & Orth-Gomer, K.
(1986, July 12). Increased risk of ischaemic heart disease in shift
workers. The lancet. 89-91.
Luna, T. D., French, J., & Mitcha, J. L (1997, January). A study of
air traffic controller shift work. Aviation. Space, and Environmental
Medicine. 6811). 18-23.


26
Sleep research indicates that long sleep episodes occur when
subjects begin sleep just after the acrophase of the temperature cycle
in the afternoon. If a subject begins sleep at this time, there is a
propensity to sleep until the temperature cycle is on the rise again
the next morning. In a normal diurnal cycle with two sleep episodes,
research indicates that, longer sleep began just prior to the
temperature minimum, whereas shorter sleep (naps) occurred near
the temperature maximum (Dinges, 1989, p.159).
The time of sleep onset in humans does not always follow the
natural cycle of tendencies for sleep because chronobiological
influences on our sleep cycles are often superseded by the
sociopsychological factors in our daily lives. Sociopsychological cues,
such as human contact and signals of activities to be carried out, are
important influences on the human sleep-wake cycle. Because of
this, these factors are thought to play a major role as zeitgebers for
the variation in the sleep-wake cycle. However, problems may arise
from social zeitgebers which would not occur from physical
zeitgebers, such as the light-dark cycle. Social zeitgebers can easily
cause a person to attempt a major, abrupt change in the sleep-wake
cycle before internal circadian processes, such as the temperature
rhythm, are able to adjust (Dinges, 1989, p. 159).


36
* While on short-haul trips for 3 to 4 days, pilots take longer to
fall asleep, sleep less, awake earlier, and report lighter and
poorer sleep compared to pre-trip sleep patterns.
* Pilots generally report feeling less well during extended duty
periods, but helicopter pilots on short-haul flights for 4 to 5 days
are far more likely to report headaches and back pain than are
commercial short-haul fixed-wing pilots, probably due to the
physical environment of the helicopter flightdeck.
* Off-duty time overstates the time available for sleep.
* Regulation of duty hours should be considered, much like
flight hours.
* Rest periods should occur at the same time on trip days or
progressively later across days.
* On layovers, experienced international flight crews sleep
efficiently at selected times or sleep less efficiently but longer
than normal with a preference for sleeping during local night.
* However, despite efficient or longer sleep during layovers, the
circadian system is unable to resynchronize and quickly adapt to
rapid, multiple time zone shifts.
* A brief in-flight nap is an acute in-flight safety valve to
improve performance and alertness on long-haul flights, but
naps do not affect the cumulative sleep debt in most
crewmembers, (p. 24)
Age has also been shown to have an effect on individuals
circadian rhythms. Research by Keran and Duchon (as cited in U.S.
Congress, Office of Technology Assessment, 1991, p. 95) has
demonstrated that individuals over the age of 45 may begin to
demonstrate the effects of dysrhythmia and may have trouble
sleeping at night.


15
and baseline rating for the following: (a) Two measures of attention
allocation (Variables 1 and 2), (b) response flexibility in switching
resources during multiple tasks performance (Variable 3), (c) spatial
visualization ability (Variable 4), and (d) straight mind/motor
coordination (Variable 5).
All of these variables have been shown by prior research to be
negatively affected by acute shifts in routine. For example, Monk et
al. (1988), demonstrated that a phase advancement of six hours in
the wake period (induced jet lag in the laboratory) with middle-age
male subjects resulted in several negative effects. These negative
effects included the following: (a) Decreased subjectively-rated
alertness (which corresponds to NovaScan Variables #1 and #2), (b)
decreased performance at search tasks (which corresponds to
NovaScan Variable #3), and (c) decreased performance at manual
dexterity tasks (which corresponds to NovaScan Variable #5).
Higgins et al. (1975) demonstrated that a 12-hour shift in the wake-
sleep cycle of 15 male participants (ages 20 to 28) produced
deficiencies in individuals multiple tasks performance (which
corresponds to NovaScan Variable #3). Luna et al. (1997) found that
Air Force controllers reported greater amounts of confusion (which
corresponds to NovaScan Variable #4) and less vigor on night shifts
than when on evening or day shifts of a rapidly-rotating schedule.


52
related to the subjects sensitivity to internal desynchronization
rather than to the desynchronization itself (p. 33).
A report in the U.S. Congress, Office of Technology Assessment
(1991) suggests that the age of an individual may also be a factor in
his/her ability to adjust to different types of shift work The report
indicates that, as a person ages, the internal clock becomes more
difficult to reset, and sleep becomes more fragile and more easily
disrupted. These effects usually begin to occur after the age of 45.
The report also indicates that surveys of rotating shift workers have
found that workers between the age of 45 and 60 years old had
more difficulty adjusting to afternoon and night shifts than did
younger workers (p. 95).
After reviewing the literature concerning the effects of shift
work, Akerstedt (1988) suggests that sleepiness, which results from
shift work, may be an underestimated potential hazard in many
occupations.
The results reviewed clearly demonstrate that shift work is
associated with increased subjective, behavioral, and
psychological sleepiness. The effects are particularly
pronounced during the night shift, and may terminate in actual
incidents of falling asleep at work. In some occupations this
clearly constitutes a hazard that may endanger human lives and
have large economic consequences. Furthermore, since night
shift sleepiness affects a large majority of individuals engaged in
it, the threat is quite real and has probably been underestimated
in the past. (p. 30)


63
Folkard & Monk, 1979; Folkard & Monk, 1985) a manipulation that
inhibited subvocalization was also found to eliminate the time-of-day
effect, suggesting that time-of-day differences may also result from
differences in subvocalization strategy (p. 170).
According to the U.S. Congress, Office of Technology Assessment
(1991), in a diurnal condition, certain psychological rhythms are
evident, though they may be inconsistent.
\ It was found that, if speed of identification was assessed, the
usual relationship with temperature held-that is, subjects
became faster over the course of the day. However, if accuracy
of response became the benchmark, peak performance occurred
in the morning (Craig & Condon, 1985; Monk& Leng, 1986).
Short-term and long-term memory also appear to peak at
different times during the 24-hour cycle (Folkard & Monk,
1980). Motivation can influence performance, too. It has been
shown that when an incentive is offered, such as a significant
sum of money, circadian decrements in performance may be
overcome to some extent (Colquhoun, 1981; Horne & Pettitt,
1985). The latter study (Horne & Pettitt, 1985) indicated,
however, that sleep deprivation and circadian rhythms can
overcome even the strongest incentive influencing performance.
Further complicating the picture is the observation that
individuals are not always accurate judges of their own mood,
alertness, or ability, (p. 47)
Englund (1979) conducted a study of the diurnal function of
reading rate, comprehension and efficiency. The results of this study
indicate that the plotted means for these variables displayed distinct
diurnal patterns. Analysis of the data revealed that comprehension
was superior in the afternoon and early morning when compared to


88
There is no Group/Time Interaction for the groups residuals on
Variable 1. The Group Main Effect test (Table 2) for the residuals
indicates that the error bands are statistically the same size for both
groups (Figures 6 and 7), while the Time Main Effect test
demonstrates a statistically-signiflcant effect, indicating that the
error bands for both groups become smaller over the sessions. The
eta squared for the Time Main Effect is 0.19, indicating a large effect
size. It is apparent that the error bands become more stable around
the means over the sessions for both groups, but Group 1 appears to
stabilize slightly more than does Group 2.
The Group/Time Interaction comparison for Variable 2, dial
monitoring reaction time during Task 2 (tracking), produced a
statistically non-significant result, meaning that the slopes appear to
be statistically the same when compared over time. The slope
comparison for Groups 1 and 2 (Table 2) indicates that the slopes
statistically appear to be the same, with no statistically-signiflcant
Group Main Effect (Figure 2). There is also no statistically-signiflcant
Time Main Effect, indicating that the slopes did not increase or
decrease to a statistically-signiflcant degree over the sessions. The
statistically-signiflcant difference which was found between the two
groups baselines by the t-Test for this variable is apparent, but not
as obvious as with Variable 1.


48
Other researchers have determined that irregular sleep
schedules are responsible for daytime sleepiness. As stated earlier,
Manber et al. (1996) demonstrated that, even when a full-nights
sleep is achieved by both groups, individuals who sleep at irregular
schedules report a lessor amount of alertness and a greater amount
of sleepiness during the day than do individuals who sleep at regular
sleep-wake schedules. Billiard et al. (1987) support this conclusion
with a study involving French military draftees. These researchers
found that sleep difficulties and irregular sleep-wake schedules were
major factors contributing to excessive daytime somnolence in young
men between the ages of 17 and 22 years of age.
One study conducted by the Federal Aviation Administrations
(FAAs) Civil Aeromedical Institute (CAMI) compared a group of air
traffic controllers on a rapidly-rotating schedule with a group of
controllers on a weekly-rotating schedule. The results indicate that,
during the work week, controllers working the rapidly-rotating
system acquired less sleep than the controllers working the weekly
rotating system, but that this sleep deficit diminished when days off
were taken into consideration. The controllers on the rapidly-
rotating system caught up on their sleep on their days off. But, in
general, both groups reported similar complaints. Fatigue,
weakness, and somnolence were the most frequent on-duty


94
Figure 10. Group 1 mean learning curve with error band (standard
deviation) for reaction time to the transition from Task 2 (Tracking)
to Task 1 (Spatial Visualization).
Figure 11. Group 2 mean learning curve with error band (standard
deviation) for reaction time to the transition from Task 2 (Tracking)
to Task 1 (Spatial Visualization).


107
workers is irregular and lacks routine. If this is the case, it could
also be the case that this type of sleep-wake cycle has a detrimental
effect on alertness which is greater than the effect caused by
rapidly-rotating shift work, due to the more routine changes in the
sleep-wake cycle that the rapidly-rotating shift system provides.
Another possible cause for the comparatively lower daytime
attention allocation ability of the day and evening shift workers
could be that a lower total amount of sleep is obtained by these
workers prior to day shifts. The day and evening shift workers may
have a tendency to become evening types during their evening
shifts and days off. If these workers do not adjust their bedtime to
an earlier hour on the evenings preceding their day shifts, they may
be experiencing a lack of alertness and lower attention allocation
ability during the day shift due to a reduction in total sleep time
during these shifts.
The first day shift for the day and evening shift workers already
has a reduced sleep period prior to its beginning because of the rapid
shift change which occurs at this point each week, during the change
from the last evening shift to the day shift (a quick-turn-around).
Most schedules would only have one or two more day shifts
following this first day shift, and these workers may be maintaining
a fairly late bed-time hour prior to these day shifts, as they do for


119
Coffey, L. C Skipper, J. K Jr & Jung, F. D. (1988). Nurses and
shift work: Effects on job performance and job-related stress, journal
of Advanced Nursing. 13. 245-254.
Cohen, J. (1988) Statistical power analysis for the behavioral
sciences (2nd ed.l Hillsdale, N.J.: Lawrence Erlbaum.
Dahlgren, K. (1981). Adjustment of circadian rhythms and EEG
sleep functions to day and night sleep among permanent night
workers and rotating shift workers. Psychophysiology. 18. 381-391.
Daniel, J & Potasova, A. (1989). Oral temperature and
performance in 8 h and 12 h shifts. Ergonomics. 32(7). 689-696.
Dinges, D. F. (1989). The influence of the human circadian
timekeeping system on sleep. In M. H. Kryger, T. Roth, & W. C.
Dement (Eds.). Principles and practice of sleep medicine (pp. 153-
162). Philadelphia: Sanders.
Endsley, M. R. (1995). Toward a theory of situation awareness in
dynamic systems. Human Factors. 37(1), 32-64.
Englund, C. E. (1979). The diurnal function of reading rate,
comprehension, and efficiency. Chronobiologia.6(2). 96.
Folkard, S., & Condon, R. (1984). Night shift paralysis.
Experientia. 40, 510-520.
Folkard, S & Condon, R. (1987). Night shift paralysis in air
traffic control officers. Ergonomics. 30(9). 1353-1363.
Folkard, S., Knauth, P., Monk, T. H & Rutenfranz, J. (1976). The
effect of memory load on the circadian variation in performance
efficiency under a rapidly rotating shift system. Ergonomics. 19(4).
479-488.
Folkard, S., Minors, D & Waterhouse, J. (1985). Chronobiology
and shift work: Current issues and trends. Chronobiologia. 12. 31-53.


87
the t-Test, this difference is also statistically significant [t(35)=4.02,
p=0.0001, rpb=0.3 2].
Learning Curve Comparison
Table 2 describes a summary of results for the learning curve
analyses and how they relate to the learning curves displayed in
Figures 1 through 15. Figures 1 through 5 display the mean learning
curves for Groups 1 and 2 (superimposed) for each of the five
variables, while Figures 6 through 15 display each groups
independent learning curve for each variable with its error band
(standard deviation).
As indicated in Table 2, there is a statistically-significant
Group/Time Interaction between the groups learning curves for
Variable 1, dial monitoring reaction time during Task 1 (spatial
visualization). The eta squared value for this interaction is 0.03,
which indicates a slight to moderate effect size. This interaction
implies that the slopes of the two groups learning curves appear to
be different. Group 1 displays a decreasing slope, while Group 2
displays a fairly flat slope with little improvement over time. The
apparent difference between the two slopes can be seen in Figure 1,
and the statistically-significant difference which was found in the
baselines by the t-Test for the final three sessions is obvious.


23
Light is, as expected, a universal zeitgeber (entraining agent) for
the activity-rest cycle displayed by most organisms, from single-
celled life forms through mammals. A recent case report indicates
that circadian rhythms of body temperature and endocrine variables
in an elderly woman could be shifted by appropriately timed
exposure to bright light, independently of the timing of sleep-wake
states (Mistlberger & Rusak, 1989, p. 145).
There are also several non-photic stimuli that appear to entrain
circadian rhythms. The timing of meals influences circadian cycles
through the anticipatory activity associated with mealtime in
organisms. This effect is very evident in experiments with rats.
Arousal states can also trigger the circadian timing system.
Simple handling of white-footed mice (Rawson, 1960) or
changing the litter of hamsters (Mrosovsky & Hallonquist, 1986)
can phase shift free-running rhythms, suggesting that even brief
arousals may have feedback effects on circadian timing. In
humans recorded in temporal isolation, a daily anchor sleep (4
h of forced bed rest at a fixed time every 24 h) can apparently
synchronize daily rhythms (Minors & Waterhouse, 1981), but
whether this synchrony is related to anchor sleep per se or daily
LD (light-dark cycle) or food intake schedules is unclear.
(Mistlberger* Rusak, 1989, p. 146)
Social cues are another well-known entraining agent in many
species, including humans, however, they are not universally
effective. It has been demonstrated that some blind humans display
free-running sleep-wake cycles in diurnal environmental conditions.


74
day shift hours (8:00 am to 4:00 pm) for each participant, in order to
determine each individuals attention allocation during the hours that
training normally occurs, and (d) all participants indicated that they
were not taking any medications, such as cold remedies or
prescription drugs, which may have an effect on alertness or
attention allocation ratings. Controllers who take such medications
are not authorized to work operational positions in any event, so a
simple acknowledgement to this requirement was all that was
required.
Procedure
Each specialist completed no more than 6 sessions per day on
NovaScan, on no less than 7 separate days, between the hours of
8:00 am and 4:00 pm, until 25 sessions were completed. Parametric
studies for NovaScan indicate that a stable baseline measure for any
individual should be completed by the 20th session, and that massed
practice should be avoided by allowing at least 7 separate days for
practice.
Since the rotating shift workers only work two or three day
shifts each week, the investigator obtained a six-week temporary
duty assignment to the Jacksonville ARTCC in order to have enough
time for the purpose of participant recruitment and data collection.
The day and evening shift participants also kept a similar schedule


8
any, mention of chronopsychology or its principles in the literature of
instructional design.
Shift work and instructional design are usually found sharing a
common niche in society. Instructional design is most widely used in
industry and the military, where it has its foundations. These are
also areas where a large number of personnel work rotating shift
work and night work. In fact, statistics suggest that approximately
20 percent of the American work force is now engaged in shift work
(Venar et al., 1989).
Shift work and night work have been shown to have a negative
effect on individuals circadian rhythms (the daily physiological and
psychological rhythms of the body and mind) in much the same
manner as transmeridian flight produces the negative state known as
jet lag. The condition which results in an individual in either
situation is known as circadian dysrhythmia, a desynchronization of
the bodys daily rhythmic cycles.
Research indicates that jet lag or shift work affects many
individuals with decrements to health (Vener et al., 1989), and, as
mentioned earlier, subjective sleepiness and fatigue during day and
night shifts, as well as overall decrements in performance. It has
also been demonstrated that irregular sleep-wake schedules are a
cause of alertness disorders (Billiard et al., 1987), and that the timing


43
worker must fight both the natural, diurnal trend of physiological
circadian rhythms, and the dominant sociocultural attitudes. Because
of this, there are three sources of stress for the shift worker: (1)
disruption of circadian rhythms, (2) disruption of sleep and fatigue,
and (3) social domestic disturbances (p. 89).
The Office of Technology Assessment also indicates that
disruptions to circadian rhythms that interact with loss of sleep and
fatigue can also affect health and performance. Sleepiness is the
inclination to sleep, whereas fatigue is weariness due to physical and
mental exertion. It is possible to experience one without the other,
but both, either alone or in concert, can have deleterious effects
(p. 87).
Sleepiness (often referred to as fatigue) has been measured
through a variety of methods. Traditionally, sleepiness was
measured via simple category rating scales or other simple rating
scales. Later methods, using the electroencephalograph (EEG) and the
electrooculograph (EOG), attempted to achieve more objective
measures.
According to Akerstedt (1988), the first attempt to quantify
objective physiological measures of sleepiness was the Multiple Sleep
Latency Test (MSLT). This test expresses sleepiness as the time it
takes for the EEG/EOG signs of sleep to appear in a subject with eyes


71
(c) reaction time for transition from the tracking task to the spatial-
visualization task (flexibility in switching resources),
(d) non-transition reaction to object orientation (spatial
visualization), and (e) tracking error during the tracking task
(straight mind/motor coordination).
The NovaScan instrument randomly presents two computer-
based tasks for the participant to perform several times during each
session. Task 1 (spatial visualization) is a cognitive task in which the
participant must orient an airplane by selecting the wing which has
the circle above it like the one below the airplane. The circles are
either open or solid, and one wing has one of these above it, while
the other wing has the other. The participant must respond either
left or right to select the correct wing by pressing buttons on a
special keyboard. The reaction times for the participants responses
are recorded in milliseconds.
Task 2 (tracking) is a psychomotor task in which the participant
must use a joy stick to move the computers cursor left and right to
try to keep it below a horizontally-moving object. The tracking error
at this task is recorded, as the mean number of pixels that the cursor
was off target and outside the range of the area directly below the
moving object during the session.


9
of sleep termination may be closely related to the
sleepiness/alertness rhythm (Akerstedt & Gillberg, 1980). This line
of research implies that shift workers, such as air traffic controllers,
may suffer from an alertness disorder that could cause an attention
decrement, if such a decrement were to be identified in controllers
who are engaged in rotating shift work during day shift hours (when
training sessions are scheduled at many facilities), it could imply a
need to amend future instructional strategies to address this variable
when considering instructional activities and media selection.
Research concerning diurnal psychological rhythms also
demonstrates implications for instructional design for normal day-
shift workers. According to Folkard and Monk (1980), short-term
and long-term memory rhythms demonstrate implications for the
timing of instruction depending on the type of instructional objective.
Englund (1979) has demonstrated a diurnal function of reading rate,
comprehension, and efficiency which implies a relationship with
memory, body temperature, and general activity cycles of circadian
rhythms. Reading rate and comprehension peak at different times of
the day, while reading efficiency is maintained.
According to Adan (1993), analysis of the psychological or
behavioral variables of circadian rhythms (chronopsychology) is
recent and still developing. She indicates that most psychological


13
only be considered as a probable cause because no other social or
psychological differences among the participants were controlled.
The first group of participants only included controllers who
work rapidly-rotating schedules which are usually composed of day
shifts, evening shifts, and night shifts. To be included in this study,
schedules for participants in this group included at least one
midnight shift and one day shift each week so that behavioral and
psychological circadian rhythms could be expected to be dysrhythmic
in the individuals working them. According to the literature, this
group should be characterized by diurnal biological rhythms and
desynchronous behavioral and psychological rhythms. Therefore,
only desynchronous behavioral and psychological circadian rhythms
were suspected as a probable cause for any decrement to day-shift
attention that was determined. No determinations were made
concerning biological circadian rhythms.
The second group only included controllers who work day-shift
and evening-shift rotating work schedules. According to the
literature, this group is expected to be characterized by diurnal
circadian rhythms, with rotating work and leisure time.
Delimitations
The instrument used in this study measured variables which
were used to determine attention allocation baseline ratings and


103
decreasing slope, while Group 2 displayed a fairly flat slope
(Figure 1). However, as Table 2 indicates, the eta squared for the
difference between the two groups on this measure is only t^=0.03,
indicating less than a moderately-meaningful difference (rp=0.06).
Because of this, further research should be conducted to determine if
this effect can be duplicated with a more meaningful outcome. The
error bands for both slopes (Figures 6 & 7) stabilized at a statistically
significant degree over the sessions, with Group 1 demonstrating an
apparently higher degree of stabilization over time than did Group 2.
In the second measure (attention allocation during tracking),
neither Group 1 or 2 appears to have improved over the sessions
(Figure 2), but Group 1 reached a baseline measure which is better
than that of Group 2, as indicated by the t-Test for Variable 2. As
with the first measure, both groups stabilized around their mean
baseline measures, however, they stabilized at different rates which
were found to be statistically significant by the Group/Time
Interaction between the the groups residuals.
As can be seen in figures 6 through 9, the groups residuals
stabilize around their individual group means on both measures of
attention allocation. This is strong evidence that the individual
differences within each group disappear over the sessions, and that


33
resynchronization, rather than a phase-advanced resynchronization,
of the temperature rhythm after the eastward flight. The
interindividual differences in adjustment of the temperature rhythm
correlated with some of the personality measures, but it was found
that larger phase delays in the temperature rhythm (as measured on
the fifth day after westward flight) were exhibited by participants
who were rated extroverts, and smaller phase delays were exhibited
by evening types (p. 733).
In this study, Gander et al. also found that ail of the subjects
showed clear phase delays in the sleep cycle for both sleep onset and
sleep offset on the westward flight (or phase delay), but that there
was greater intersubject variation in the timing of sleep following the
eastward flight (or phase advance). The slower adaptation and
adjustment to eastward flight was due to the variable recovery
methods of the circadian rhythms. After eastward flights, some of
the circadian rhythms tend to resynchronize by delaying, while other
rhythms advance. It appears that extroverts tend to adapt more
rapidly to the delay shift, probably because they are more exposed
to the social routine in the new time zone (p. 742).
Wright et al. (1983) studied 81 male soldiers (ages 18-34) for
five days before and five days after an eastward deployment across
six time zones. Commonly reported jet lag symptoms of tiredness.


Shift Work
37
Overview
The term shift work has many meanings and varies according
to which author or context is used as reference. According to
Akerstedt (1988),
Shift work usually refers to an arrangement of work hours that
uses two or more teams (shifts) to cover the time needed for
production. Whereas two-shift work usually covers only the
daylight hours, three-shift work also covers the night. Shifts are
often changed at -0600 h, 1400 h, and (if a night shift is
included) 2200 h, although many companies employ earlier or
later times. In Europe, the teams usually rotate between the
shifts, whereas in the United States, assignment to a certain shift
is often permanent, at least for a considerable time, before
seniority allows transfers to another shift. Permanent night
work, the watch system at sea, and roster work are other
varieties of work hour arrangements. The latter (roster work) is
similar to conventional shift work but is somewhat more
irregular and customized to particular needs, usually in the
service sector. For lack of better terminology we will refer to all
these types of work hour systems as shift work. (p. 17)
Individuals who work rotating shift work, in most cases, are
almost constantly subjected to negative physiological and
psychological effects that are similar to those experienced as a result
of jet lag. The severity of the condition depends upon the type of
rotating schedule that the individual works and the number of hours
worked during each shift.
Schedules which rotate on a weekly basis can cause biological
rhythms to desynchronize, as they try to adjust to each new routine


5
nation-wide, or about 0.53 operational errors per 100,000 facility
activities in the air traffic system (U.S. Department of Transportation,
Federal Aviation Administration, 1997, p. 6).
Some operational errors result in aircraft accidents. In 1996,
there were 38 large air carrier, 12 commuter, 88 air taxi, 1,911
general aviation, and 182 rotor craft accidents (U.S. Department of
Transportation, Federal Aviation Administration, 1997, p. 4). Some
of these accidents were the result of controller error. Based on what
is known about chronopsychology and circadian dysrhythmia, it is
very possible that at least some of the so-called controller errors are
the result of a failure on the part of controllers to follow procedures,
either because of degraded performance while working, or because
of a failure to fully acquire information concerning these procedures
during training. These controllers may not be assimilating some
information during training sessions due to a degraded attention
allocation and a negative learning effect which are brought on by the
dysrhythmic condition.
Need for the Study
Air traffic controllers need to be in a high state of alertness
while performing their duties, and while undergoing training on new
procedures or during refresher training of prior learning. Reddings
(1992) analysis of operational errors and workload in air traffic


85
Table 1. Summary of t-Test Results for Group Mean Attention
Allocation Comparisons in Milliseconds of Reaction Time
Group 1 Group 2
Variable
n
M
S2
n
M
Task 1
18
616.57
107.15
19
778.64
140.91
(t=3.92, o=0.002, rpb=0.31)
Task 2
18
1548.42
55.27
19
1659.16
103.61
(t=4.02. p=0.0001. rob=0.32)
*p<0.05
Group 1 (Rapidly-Rotating Shift Workers) is 616.57 milliseconds of
reaction time, while the mean measure for Group 2 (Day/Evening
Shift Workers) is 778.64 milliseconds of reaction time. As
determined by the t-Test, this difference is statistically significant
[t(35)=3.92, p=0.002,rbp=0.31].
The second group comparison (attention allocation during the
tracking task) indicates that the mean baseline measure for Group 1
is 1548.42 milliseconds of reaction time, while the mean measure for
Group 2 is 1659.16 milliseconds of reaction time. As determined by


79
the two groups progress over the sessions. Wherever a Group/Time
Interaction was found to be statistically significant, the Group Main
Effect and Group Time Effect were not reported (per rule). Also, for
each test that was found to be statistically significant, eta squared
(n2) was calculated in order to determine whether the significant
difference was slightly (n2=0.01), moderately (n2=0.06), or highly
(>12=0.14) meaningful for the size of the sample used in this study.
According to Cohen (1988), eta squared is the proportion of total
variance in the outcome that is accounted for by the between group
variance. In short, eta squared determines if the significant
difference between the groups is meaningful, and a moderate eta
squared result (r)2=0.06), or higher, would meet this standard. The
learning curves for the two groups were superimposed and displayed
together for visual comparison on each variable, and each groups
mean learning curves were displayed individually with their error
bands (standard deviations).
Only the last three seconds (time period two) of each trial for
variables 2 (attention allocation during tracking) and 5 (tracking
error during tracking) were used in the calculation for these tasks, in
order to control for the variable state of the tracking task at each of
its initiations. For this same reason, variable 3 (response flexibility


Hypothesis 3
82
A statistically-signiflcant Group/Time Interaction with a small
effect size was found between the two groups mean learning curves
for Variable 1: dial monitoring (attention allocation) during Task 1
(spatial visualization). Group 1 (Rapidly-Rotating Shift Workers)
demonstrated a decreasing learning curve with an improving
reaction time over the sessions, while Group 2 (Day/Evening Shift
Workers) demonstrated a fairly flat learning curve. Both groups
demonstrated stabilization around their mean baseline measures,
with no Group/Time Interaction between the groups residuals. The
null hypothesis was rejected.
Hypothesis 4
No statistically-signiflcant Group Main Effect, Time Main Effect,
or Group/Time Interaction was found between the two groups
learning curve slopes for Variable 2: dial monitoring (attention
allocation) during Task 2 (tracking). The learning curve slopes for
the two groups appear statistically the same (fairly flat) and did not
improve to a statistically-signiflcant degree over time. Both groups
demonstrated a stabilization around their mean baseline measure,
however, the groups residuals demonstrated a statistically-
signiflcant Group/Time Interaction with a moderate effect size,
indicating a difference in the groups residuals when compared over


122
ODonnell, R. D. (19921. The NovaScan test paradigm: Theoretical
basis and validation. Dayton, Oh: NTI, Inc.
Paley, M. J., & Tepas, D. I. (1994). Fatigue and the shiftworker:
Firefighters working on a rotating shift schedule. Human Factors.
36(2), 269-284.
Patkai, P., Akerstedt, T., & Pettersson, K. (1977). Field studies of
shift work: I. Temporal patterns in psychophysiological activation in
permanent night workers. Ergonomics. 20. 611-619.
Pilcher, J. J., & Huffcutt, A. I. (1996). Effects of sleep deprivation
on performance: A meta-analysis. Sleep. 19(4). 318-326.
Rankin, M. L, Latham, G., Peters, R. D., & Penetar, D. M. (1989).
The effects of 48 hours total sleep deprivation on human physiology,
mood, and memory. Proceedings of the Human Factors Society 33 rd
Anual Meeting. 625-629.
Redding, E. R. (1992). Analysis of operational errors and
workload in air traffic control. Proceedings of the Human Factors
Society 36th Annual Meeting. Training: Analyzing Workload-
Modeling Expertise, and Maintaining Motivation. 2. 1321-1325.
Reinberg, A., Motohashi, Y., Bourdeleau, P., Touitou, Y., Nouguier,
J., Nouguier, J., Levi, F., & Nicolai, A (1989). Internal
desynchronization of circadian rhythms and tolerance of shift work.
Chronobiologia. 16. 21-34.
Ribak, J., Ashkenazi, I. E., Klepfish, A., Avgar, D Tall, J., Kallner,
B & Noyman, Y. (1983, December). Diurnal rhythmicity and air force
flight accidents due to pilot error. Aviation. Space, and Environmental
Medicine. 1096-1099.
Rosa, R. R., & Bonnet, M. H. (1993). Performance and alertness on
8 h and 12 h rotating shifts at a natural gas utility. Ergonomics.
3£(10), 1177-1193.


56
months adaptation to the new 12-hour shift schedule, workers
expressed decrements in performance and alertness which were
attributable to the extra 4 hours on the shift. The workers also
reported reductions in sleep across the work week which were most
significant on the 12-hour night shifts. Rosa and Bonnet indicate that
these results are consistent with their earlier findings, and that extra
caution should be exercised when scheduling critical activities for
workers on extended shifts, especially during extended night shifts,
such as the 12-hour shift (p. 1177).
Daniel and Potasova (1989), in a study which compared 8-hour
and 12-hour rapidly-rotating shift workers, demonstrated that the
temperature rhythm of the 12-hour shift workers goes into rapid
decline 4 hours earlier than the temperature rhythm of the 8-hour
shift workers. Other results of this study concerned comparisons on
performance. Striking changes in performance were noted during
the circadian rhythm in psychomotor activities, while the
performance curve during more exacting mental activities
demonstrated a more uniform pattern. Some of the lower
achievements by the 12-hour workers could be attributed to
increased fatigue during the longer shift, however, they could
possibly be attributed to lower work requirements being assigned to
these workers (p. 695).


108
other nights of the week. All of the other days of the week may be
late-evening-oriented for these individuals.
During the data collection phase of this study, a record was kept
for the participants bedtimes, but this record only recorded the
participants bedtimes for the nights prior to their NovaScan sessions.
Since these sessions only took place two, three, or possibly four days
of the week, this record is inappropriate for determining if irregular
sleep-wake cycles were practiced by the participants. However, this
record does indicate that the day and evening shift workers
bedtimes on the nights prior to their NovaScan sessions ranged from
9:00 pm to 3:00 am, with an average bedtime of about 11:30 pm,
during the course of the investigation. Because of the findings of this
study, a more complete and accurate measure of this variable should
be taken in future research of this type.
Another point of discussion concerns the possible relationship
between controller error and controller shift system type. During the
data collection phase of this study, the author gained access to the
Jacksonville ARTCC controller error statistics for the 18-month period
preceding the investigation. The author conducted a Chi-Square
analysis to determine if a higher frequency of controller errors were
committed by controllers who worked rapidly-rotating shift


84
Effect with a large effect size, indicating that both groups learning
curves depict improved reaction times over time. No error-band
effects were found to be statistically significant, indicating no
statistically-signiflcant change in them over time, and no Group/Time
Interaction between the groups error bands. The null hypothesis
was not rejected.
Hypo thesis 7
No statistically-signiflcant Group Main Effect or Group/Time
Interaction was found between the two groups learning curves for
Variable 5: tracking error during Task 2 (tracking), indicating that
the groups mean learning curves appear statistically equal. There
was, however, a statistically-signiflcant Time Main Effect with a
moderate effect size, indicating that both groups learning curves
depict improved tracking errors over time. No error-band effects
were found to be statistically significant, indicating no statistically-
signiflcant change in them over time, and no Group/Time Interaction
between the error bands. The null hypothesis was not rejected.
Attention Allocation
Table 1 displays the results of the two comparisons of attention
allocation baseline for the two groups, based on the two t-Tests. The
first group comparison (attention allocation during the spatial
visualization task) indicates that the mean baseline measure for


Table 2. Summary of Learning Curve Analyses
86
VRBL
F2.
Test
F Value
n Value
Eta Sauared
1
1
Slope/GTI
2.96
0.02
0.03
6/7
Resdl/GME
0.37
0.55
6/7
Resdl/TME
12.3
0.0001
0.19
6/7
Resdl/GTl
1.7
0.11
2
2
Slope/GME
0.01
0.94
2
Slope/TME
2.29
0.13
2
Slope/GTI
1.01
0.34
8/9
Resdl/GTI
3.96
0.05
0.09
3
3
Slope/GME
0.78
0.38
3
Slope/TME
43.35
0.0001
0.17
3
Slope/GTI
0.9
0.47
10/11
Resdl/GME
1.37
0.25
10/11
Resdl/TME
0.97
0.43
10/11
Resdl/GTI
0.88
0.49
4
4
Slope/GME
0.51
0.48
4
Slope/TME
72.3
0.0001
0.29
4
Slope/GTI
2.05
0.09
12/13
Resdl/GME
2.03
0.16
12/13
Resdl/TME
1.24
0.30
12/13
Resdl/GTI
0.63
0.66
5
5
Slope/GME
0.56
0.46
5
Slope/TME
12.31
0.0001
0.10
5
Slope/GTI
0.50
0.85
14/15
Resdl/GME
0.38
0.54
14/15
Resdl/TME
1.55
0.15
14/15
Resdl/GTI
0.64
0.73
Abbreviations used in Table 2: Resdl=Residual; GME=Group Main
Effect; TME=Time Main Effect: GTI=Group/Time Interaction
*p<0.05


CHAPTER 5
CONCLUSION
Introduction
Air traffic controllers tend to prefer phase-advanced rapidly-
rotating work schedules because of the greater breaks they provide
between work weeks. Phase-advanced schedules rotate
counterclockwise beginning with the evening shift. The controller
may work one, two, or three of these evening shifts, depending upon
the individualized schedule that he or she works. This type of
schedule then allows only an 8-hour break before the controller
must return for a day shift, and at the end of the last day shift, the
controller encounters another 8-hour quick-turn-around before
returning to work the night shift. Individual controllers work a
variety of combinations of these three shifts, but the rotation pattern
remains the same. By compressing the work schedule during the
week in this manner, controllers realize an 80-hour break between
work weeks.
Although controllers prefer to work rapidly-rotating schedules,
another reason that controllers generally do not work steady,
nonrotating schedules is that the FAA discourages this practice.
98


73
keyboard is provided for responding. These minimum requirements
were met using two 386 MHz DOS computers. Each participant was
provided with a single 3.5 in floppy disk, which contained the
NovaScan software, so that all NovaScan sessions could be recorded
separately for each individual.
Design
A causal-comparative design was used for this study since the
predictor variable (work schedule) is already differentiated between
the two study groups, and because the cause (circadian dysrhythmia)
is already suspected and the effect (attention allocation) is under
investigation. Self-selected participants were categorized into two
groups. One group of participants was represented by individuals
who work day and evening two-shift schedules, and the other group
only included individuals who work a variety of rapidly-rotating
three-shift schedules which include night shifts.
The variables which were controlled include the following
considerations: (a) The rotating schedules included at least one
midnight shift and one day shift each week prior to the NovaScan
sessions, so that the daily routines of the individuals who worked
them were acutely shifted, (b) all of the participants were no more
than 45 years of age, so that the circadian effects of aging were not a
factor, (c) all testing periods on the NovaScan took place during the


39
diurnal, whereas, weekly-rotating systems desynchronize these
physiological circadian rhythms. The rapidly-rotating systems are
also advantageous in that they do not allow a sleep debt to
accumulate because the number of day sleeps is less continuous than
with the weekly-rotating systems. However, decrements in alertness
and increased fatigue are reported by workers under both systems,
indicating that workers psychological rhythms are negatively
effected, probably due to the changes in routine experienced under
both systems.
Many researchers once thought that permanent night work
might be more conducive to a more efficient adjustment to
sleepiness. But, Folkard et al. (1978) found that the rated alertness
of permanent night nurses remained diurnal, or day-oriented, except
for a minor blunting of the steep fall of the alertness curve during
the night shift. Other researchers (e.g., Dahlgren, 1981; Patkai et al.,
1977) indicate that sleepiness in permanent night work appears to
be phase delayed and exhibits a peak towards the end of the shift.
Melton and Barlanowicz (1986) described the different models of
shift work, such as the phase-advanced 2-2-1 (2 evenings, 2 days, 1
midnight) schedule, and its opposing phase-delayed 1-2-2 model, in
order to explain why certain shift work models are desired by
workers. They indicated that the 2-2-1 rotation is preferred by


121
Manber, R., Bootzin, R Aceba, C., & Carskadon, M. (1996). The
effects of regularizing sleep-wake schedules on daytime sleepiness.
Sleep. 19(5). 432-441.
McAdaragh, R. M. (1995, Spring). Human circadian rhythms and
the shift work practices of air traffic controllers. The journal of
Aviation/Aerospace Education & Research. 5(3). 7-15.
Melton, C. E. (1985). Physiological responses to unvarying
(steadv) and 2-2-1 shifts: Miami International Flight Service Station
(Report No. FAA-AM-85). Washington, DC: Civil Aeromedical
Institute, Office of Aviation Medicine, Federal Aviation
Administration.
Melton, C. E., & Bartanowicz, R. S. (1986). Biological rhythms and
rotating shift work: Some considerations for air traffic controllers and
managers (Reprt No. DOT/FAA/AM-86/2). Washington, DC: Civil
Aeromedical Institute, Office of Aviation Medicine, Federal Aviation
Administration.
Mistlberger, R., & Rusak, B. (1989). Mechanisms and models of
the circadian timekeeping system. In M. H. Kryger, T. Roth, & W. C.
Dement (eds.). Principles and practice of sleep medicine (pp. 141-
151). Philadelphia: Sanders.
Monk, T. H. (1986). Advantages and disadvantages of rapidly
rotating shift schedules--a circadian viewpoint. Human Factors. 28.
553-557.
Monk, T. H. (1989). Circadian rhythms in subjective activation,
mood, and performance efficiency. In M. H. Kryger, T. Roth, & W. C.
Dement (Eds.). Principles and practice of sleep medicine (pp. 163-
172). Philadelphia: Sanders.
Monk, T. H Moline, M. L, & Graeber, R. C. (1988, August).
Inducing jet lag in the laboratory: Patterns of adjustment to an acute
shift in routine. Aviation. Space, and Environmental Medicine. 59(8).
703-709.


104
the differences between the groups baseline measures really are
group differences.
It is also noteworthy that the mean reaction times for both
groups are higher during Task 2 than they are during Task 1. Task 2
(tracking), which is a psychomotor activity, appears to require more
concentration than does Task 1 (spatial visualization), which is a
cognitive task. Both groups demonstrated a lower attention
allocation ability while performing the psychomotor task, and neither
group improved to a statistically-significant degree on this task over
the sessions.
The attention allocation ratings for the groups during the
cognitive task (spatial visualization) indicate differential results. As
with the psychomotor task, Group 2 again demonstrated no
statistically-significant improvement over the sessions. However,
Group 1 does improve attention allocation during this task.
The analyses of the learning curves for Variables 3, 4, and 5
demonstrate no statistically-significant learning differences between
the two groups. Both groups improved their reaction times for
response flexibility at switching resources (Variable 3: changing
from Task 2 to Task 1), and at object orientation (Variable 4: spatial
visualization). Both groups also improved their tracking error during


51
shifts sleep longest, workers on the day shift sleep slightly less, and
night shift workers sleep least (Paley & Tepas, 1994, p. 270). In
addition to this, Paley and Tepas (1994) also indicate that the partial
sleep loss resulting from working shift work should effect individuals
with some of the same symptoms that have been seen in studies
concerning total sleep loss.
Partial sleep loss occurs whenever there is a reduction in the
usual amounts of sleep obtained in a 24-h period. Johnson and
Naitoh (1974) suggested that the factors involved in determining
the effects of total sleep loss are equally relevant to partial sleep
loss. Therefore, changes in mood state, increased feelings of
fatigue and irritability, inability to concentrate, and periods of
misperception associated with total sleep loss should also occur
from reductions in sleep length or changes in sleep/wakefulness
cycles experienced in night work. (p. 270)
Reinberg et al. (1989) conducted a study concerning shift work
tolerance and the internal desynchronization of circadian rhythms.
It was found that all the participants in the study maintained some
diurnal cycles (including temperature), but displayed a
desynchronization of heart rhythm, as well as various psychological
and behavioral rhythms, while working a rapidly-rotating shift
system. Individuals who demonstrated intolerance to shift work
complained of drowsiness, fatigue, and lack of attention. Since an
internal desynchronization can be observed in tolerant shift workers
having no complaint, it is likely that symptoms of intolerance are


102
Group 2 (Day/Evening Shift Workers) on both measures: Variable 1
(dial monitoring during the spatial visualization task), and Variable 2
(dial monitoring during the tracking task). The difference between
the two groups means on Variable 1 was 162.07 milliseconds of
reaction time, and the difference on Variable 2 was 110.74
milliseconds of reaction time. Although this difference appears to be
small, it is important that there is a significant difference between
the two groups because attention allocation is of primary concern to
the tasks involved in radar control. These tasks include a high
degree of concentration on changing traffic situations, and the
constant necessity to solve current and potential traffic conflicts.
Because of this, dynamic situation awareness research should make
note of the differences found in this study and take these results into
consideration in future studies. Educational psychology research
should consider conducting studies to determine if small differences
in attention allocation such as these are significant enough to affect
the learning process during instruction.
The analyses of the learning curves for these two variables
(Table 2) indicate some interesting findings (Figures 1, 2, 6, 7, 8 & 9).
In both measures of attention allocation baseline, Group 1 performed
better than did Group 2. In the first measure (attention allocation
during spatial visualization), Group 1 improved over time with a


92
Figure 6. Group 1 mean learning curve with error band (standard
deviation) for dial monitoring reaction time during Task 1 (Spatial
Visualization).
Figure 7. Group 2 mean learning curve with error band (standard
deviation) for dial monitoring reaction time during Task 1 (Spatial
Visualization).


CHAPTER 3
METHODOLOGY
Participants
Thirty-seven participants (N = 37) were self-selected from the
Jacksonville, Florida, Air Route Traffic Control Center (ARTCC).
Although this sample is representative of the population of interest,
this number is less than the desired number of participants for this
type of study and may slightly limit the generalization of the
findings. Participants were recruited on a voluntary basis, so that
two groups (18 rapidly-rotating shift workers, and 19 day/evening
shift workers) were obtained. The population for this study
consisted of six female controllers (three in each group) and thirty-
one male controllers, all 45 years of age or younger. The majority of
the participants were in their late 20s or early 30s.
Instruments
The NovaScan computer-based performance test was the
instrument used to gather data for this study. This test was used to
measure five variables: (a) Dial monitoring reaction time during the
spatial-visualization task (attention allocation), (b) dial monitoring
reaction time during the tracking task (attention allocation),
70


task. No differences were found between the groups learning curves
for the three other variables. It is suggested that day and evening
shift workers may practice irregular sleep-wake schedules and/or
accumulate a greater sleep debt during the work week than do
rapidly-rotating shift workers.


2
Babkoff et al. (1991) determined that subjective sleepiness
ratings of individuals during periods of sleep deprivation are
dependent upon the phase of each individuals circadian cycle.
Manber et al. (1996) demonstrated that, even when a normal full-
nights sleep is achieved by both groups, individuals who sleep at
irregular schedules report a lesser amount of alertness and a greater
amount of sleepiness during the day than do individuals who sleep at
regular sleep-wake schedules. Billiard et al. (1987), in a study
involving French military draftees, found that sleep difficulties and
irregular sleep-wake schedules were major factors contributing to
excessive daytime somnolence in young men between the ages of 17
and 22 years of age.
Circadian dysrhythmia has also been shown to cause
performance decrements in affected individuals (Higgins, et al., 1975;
Monk et al., 1988), and dysrhythmia and sleep deprivation have
been shown to produce a learning effect (Rankin et al., 1989). Luna
et al. (1997), in a study of air traffic controllers working a rapidly
rotating schedule, found that controllers on the midnight shift of a
forward rapidly-rotating schedule appeared to be falling asleep and
reported increased confusion and fatigue. In a telephone
conversation concerning this study on March 31,1997, Dr. French
advised the author that, although the performance data collected was


59
erroneous experimental results in many cases of past psychological
research.
Monk (1989) indicates that the study of psychological circadian
rhythms uses the physiological temperature rhythm as a reference
for three major reasons:
The first is simply expedience, body temperature being a well-
defined, relatively stable circadian rhythm that can be measured
easily by psychologists who lack medical qualifications or
radioimmunoassay competence. Second, there is a strong
historical tradition, espoused by pioneers such as Kleitman
(Kleitman, 1963) and Colquhoun (Colquhoun, 1971), linking
psychological rhythms to body temperature rhythms. Third,
there are the recent mathematical models of the circadian
system (Borbely, 1982; Kronauer et al., 1982; Wever, 1975)
which all rely crucially upon the concept of an endogenous
circadian oscillator for which the body temperature rhythm is
the major indicator, (p. 163)
Monk also indicates that psychological rhythms are not all the
same and that, like the physiological circadian rhythms,
psychological rhythms comprise gradual fluctuations over the entire
24 h. Moreover, there are strong intervariable differences (p. 166).
He also states that psychological rhythms are not just specific to each
individual. Studies by researchers such as Horne and Ostberg, 1977,
and Kerkhof, 1980 (as cited in Monk, 1989), have demonstrated that,
even with differences between morning-type and evening-type
individuals, for the 80 per cent of people who are neither extreme
morning type nor extreme evening type, one can make


16
The difference between this study and the previously cited
research is that this study determined objective measures by
NovaScan in a natural setting for all the variables tested, whereas,
the previous research has only reported subjective ratings in natural
or laboratory settings, or objective ratings determined in contrived
laboratory environments.
Assumptions
It was assumed that the participants recruited for this study
had no significant cognitive differences and that they were free of
any attention or learning decrements not related to dysrhythmia.
Air traffic controllers possess unique cognitive skills and abilities.
These cognitive attributes become unique to controllers through a
weeding-out process and through cognitive training. Controllers
undergo pre-employment cognitive and psychological testing which
has been developed to eliminate individuals who do not possess the
necessary mental attributes required of air traffic control work in
general. The training that controllers obtain continues to eliminate
individuals who do not possess the particular skills and abilities that
are required, and to develop those skills and abilities in the
individuals who do possess them. The cognitive variables that were
investigated in this study are some of the same cognitive variables
that are developed through controller training.


32
began. The shortest sleep episodes occurred when sleep onset began
in the morning, and the longest sleep episodes occurred when sleep
onset began in the evening. Sleep termination times were related to
specific points on the sleepiness/alertness rhythm.
Gander et al. (1989) conducted a study involving air crews of P-
3 aircraft in order to determine the adjustment of sleep and the
circadian temperature rhythm after flights across nine time zones.
In this study, researchers investigated both westward flight (phase-
delay variation) and eastward flight (phase-advance variation). The
study involved nine Royal Norwegian Air Force volunteers operating
P-3 aircraft during flights across nine time zones. The variables
monitored during the flight included each participants sleep-wake
pattern and circadian temperature rhythm. Each participant
recorded his own sleep and nap times, rated nightly sleep quality,
and personality inventories, while rectal temperature, heart rate, and
wrist activity were continuously monitored.
Adjustment after the return eastward flight was slower, as
compared to adjustment after the westward flight, for the
readjustment of sleep timing to local time. The eastward flight also
produced greater interindividual variability in the patterns of
adjustment of sleep cycles and temperature rhythms. One of the
participants even exhibited a 15-hour phase-delayed


99
Traffic is generally light on the night shift and it is felt that
controllers permanently on this shift would experience a
deterioration of proficiency in handling heavier traffic loads, should
they be called upon to do so. Also, social isolation from professional
colleagues would keep these controllers from attending training,
briefings, and conferences (Melton, 1985).
The hypotheses addressed by this study were designed to
examine whether rapidly-rotating shift work may have an
objectively-measurable effect on attention allocation during the day-
shift hours (8:00 am to 4:00 pm), when training is most likely to
occur in air traffic control facilities. They were also designed to
determine if this type of shift work may cause any learning effects
which could be measured via learning curve analysis. This study
was deemed necessary because previous studies have demonstrated
that air traffic controllers who work rapidly-rotating shift schedules
report above average ratings of subjectively-rated sleepiness and a
lack of alertness during these hours. If objective measures of
attention or learning deficiencies could be demonstrated in
controllers who work rapidly-rotating shift systems, training
treatments could be suggested or devised during training
development in order to accommodate these deficiencies, as with


46
shift work) and takes the form of an inability to react to stimuli
which should normally elicit a response. Nurses may not react to a
call from a patient or a question from a colleague, for example.
A similar study by Folkard and Condon was later conducted with
air traffic controllers. This study was designed to examine the
possibility that night shift paralysis may be a problem in this group.
The sample included 435 air traffic controllers from 17 different
countries, who were working a variety of work schedules. The
incidence of paralysis was found to be influenced by four main
factors which, in turn, affect the workers level of sleep deprivation
and sleepiness. These factors included (a) the time of night, (b) the
number of consecutive night shifts, (c) the requirement to work both
a morning shift and a night shift starting on the same day, and (d)
individual differences in flexibility of sleep habits. These results
suggest that the incidence of this paralysis may indeed prove to be a
useful critical incident for comparing the level of sleep deprivation
associated with different shift systems or individuals (Folkard &
Condon, 1987, p. 1353).
Folkard and Condon also described the relationship between the
alertness cycle and the sleep/wake cycle, and how partial sleep
deprivation occurs as a result of shift work.
It is well established that subjectively rated alertness shows a
marked circadian (about 24 h) rhythm that is at least partially


CHAPTER 1
INTRODUCTION
The occupation of air traffic controller carries with it a great
responsibility in terms of both life and property, in that the
controllers job requires him/her to maintain the safe and
expeditious flow of air traffic in his/her area of responsibility.
McAdaragh (1995) has demonstrated that many controllers work
rotating work schedules that have been shown to induce circadian
dysrhythmia, a misalignment or desynchronization of an individuals
biological and/or psychological daily rhythmic cycles (Hawkins,
1987). Chronopsychological research (the study of behavioral and
psychological rhythms) indicates that these type of work schedules
produce a high degree of subjective fatigue in these individuals
during midnight and day shifts, with the lowest degree being
reported on the evening shifts (Melton, 1985; Monk et al 1988;
Saldivar et aL, 1977; Smith et al., 1971 ). The term shift-work
insomnia has been used to describe the condition caused by
circadian dysrhythmia which reduces the total amount of sleep
achieved by affected individuals who work shift work (Akerstedt &
Kecklund, 1991).
1


65
Daily fluctuation in behavioral parameters cannot be considered
trivial: the total variation detected is of the order of 10 %, and
over the normal waking day (0800 to 0000) the circadian
variation in performance efficiency can be equivalent in
magnitude to the effect of limiting sleep to 3 hours or ingesting
the legal driving limit of alcohol (Monk et al., 1978, p. 166).
(p. 146)
Adan also indicates that other rhythms with different periodicities
(circamensual rhythms = greater than 24 hours, and ultradian
rhythms = less than 24 hours) may mask the results of studies
concerning circadian rhythms and, therefore, should be controlled.
Nevertheless, the classification proposed by Adan (1993) enables
us to differentiate, from a chronopsychological viewpoint, the most
commonly used tests based upon both the required skill of each test,
and the empirical criterion of the resultant circadian function. The
advantage of Adans classification over traditional classifications is
that coherence is imposed on the heterogeneity of the existing results
for both subjective and objective tests.
The subjective tests yield data concerning the subjects own
assessment of his/her state during the test. The two principal types
of tests include self-assessment inventories, and analogico-visual
scales. Patterns of alertness, as measured on the analogico-visual
scales, shows a peak alertness between 11:00 and 14:00, and
somnolence demonstrates an opposed two-phase pattern. The


12
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of reaction time for the transition
from Task #2 to Task #1 (which indicates response flexibility in
switching resources).
6. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of reaction time on Task #1 non
transition object orientation (which indicates spatial visualization).
7. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of tracking error on Task #2
(which indicates mind/motor coordination).
Limitations
Work schedule, the predictor variable, was the only variable
used to differentiate the two groups of participants. Because of this,
work schedule was considered as the suspected cause for any
detriments that were found in this study. Circadian rhythmicity can


34
sleepiness, weakness, headache, and irritability appeared in the
majority of the subjects. While most of the symptoms had
disappeared or diminished by the fifth day in Germany, tiredness,
sleepiness, and irritability continued.
The cardiorespiratory responses and the perception of effort
during treadmill exercise were completely unaffected by the
conditions experienced by the participants in this study, however,
other variables were affected.
While isometric strength of upper body, legs, and trunk was
unchanged, dynamic strength of arms was significantly reduced
after arrival in Germany in two out of three groups at the slow
contraction speed and even more dramatically in all groups at
the fast velocity. These findings are suggestive of adverse jet-
lag effects on some aspect of muscle contractile capabilities, most
likely motor control and fiber recruitment. Dynamic arm
endurance declined in a comparable fashion. Complete recovery
of arm strength and endurance did not occur within the first 5 d
after deployment, (p. 136)
Saito et al. (1992) conducted a study of the gradual adjustment
of circaseptan-circadian blood pressure and heart rate rhythms of
two adults (36-year-old male; 32-year-old female) and two children
(6-year-old boy; 6-month-old boy) after an eastward flight which
advanced local time by 9 hours. Circadian rhythms are daily
rhythms of about 24 hours, whereas, circaseptan rhythms are
rhythms of about 7-days duration.


68
of two groups. The two groups were divided in order to examine
retention of initial learning in the morning versus retention of initial
learning at night, at both 8 hours and 24 hours after the initial
learning was accomplished. The results indicated that retention was
superior 8 hours after learning for subjects who learned at night, as
compared to those who learned in the morning. Retention for
subjects who learned at night was equal after 24 hours to that
observed after 8 hours. A rather surprising finding in this study was
that retention scores for subjects who learned in the morning were
superior after 24 hours to those observed at 8 hours after initial
learning (p. 192).
A recent study (Shadmehr & Holcomb, 1997) indicates that it
takes about six hours for the memory of a new skill to move into
long-term memory storage. This study also indicates that practice of
a similar skill just after learning the new skill can interfere with the
long-term memory storage of the new skill, but that practice of a
non-similar skill may not interfere. These findings may have
implications for the surprising results obtained by Benson and
Fein berg.


91
Figure 5. Mean learning curves for tracking error during Task 2
(Tracking).
The residual tests (Table 2) for Variable 2 demonstrates a
statistically-significant Group/Time Interaction with a moderate
effect size (n=0.09) for the residuals (Figures 8 and 9). This indicates
that the size of the error bands decrease at statistically different
rates over time for the two groups.
The statistical analyses (Table 2) for Variables 3 (reaction time
during transition from Task 2 to Task 1), 4 (reaction time during
Task 1), and 5 (tracking error) all had similar results. There was no
Group/Time Interaction between the groups on any of these three
variables, indicating that the slopes of the groups appear statistically
the same over the sessions in each case. As Figures 3, 4, and 5


CHAPTER 4
RESEARCH FINDINGS
Results
Hypothesis 1
A statistically-significant difference was found between the two
groups attention allocation baseline measure for dial monitoring
during Task 1 (spatial visualization). Group 1 (Rapidly-Rotating
Shift Workers) demonstrated a better attention allocation rating (a
faster reaction time for dial monitoring) during Task 1 (spatial
visualization) than Group 2 (Day/Evening Shift Workers)
demonstrated. The null hypothesis was rejected.
Hypothesis 2
A statistically-significant difference was found between the two
groups attention allocation baseline measure for dial monitoring
during Task 2 (tracking). Group 1 (Rapidly-Rotating Shift Workers)
demonstrated a better attention allocation (a faster reaction time for
dial monitoring) during Task 2 (tracking) than Group 2 (Day/Evening
Shift Workers) demonstrated. The null hypothesis was rejected.
81


44
closed. This test, and variations of it, remains the only standardized
clinical method of measuring sleepiness. However, it is difficult to
use outside of well-controlled laboratory, or laboratory-like,
environments, and it cannot reflect momentary fluctuations of
sleepiness.
Nevertheless, it appears that other EEG/EOG parameters may be
used for the description of momentary fluctuations of
wakefulness/sleepiness. Thus vast amounts of data support the
proposition that increased Alpha and Theta activity in the EEG,
and also slow eye movement (SEM) activity, are closely
correlated with both subjective and behavioral sleepiness
(Daniel, 1967; OHanlon & Beatty, 1977; Torsvall & Akerstedt,
1984; Akerstedt et al., 1984; Lecret & Pottier, 1971; OHanlon &
Kelley, 1977;Torsvall & Akerstedt, 1987). In a record study
(Torsvall & Akerstedt, in press), we tried to experimentally
quantify the EEG/EOG changes that characterize severe
sleepiness, i.e., the level of sleepiness at which drowsiness
prevents interaction with the environment and the individual
dozes off. We found after spectral analysis that increases of
~500 and 200% in the power density of alpha and theta activity,
were critical (compared with the values at normal alertness).
Furthermore, field studies have demonstrated that EEG/EOG-
based monitoring of sleepiness is also feasible in completely
ambulatory subjects under their normal work/leisure/sleep
conditions (Torsvall & Akerstedt, 1987; Torsvall et al.,
unpublished observations). (Akerstedt, 1988, p. 18)
Most studies concerning sleepiness in shift work derive, as
indicated earlier, from subjective questionnaire studies. Several of
these studies, including Wyatt & Marriot, 1953, Thiis-Evensen, 1958,
Mott et al., 1965, Menzel, 1962, Andersen, 1970, and Akerstedt &
Torsvall, 1978 (as cited in Akerstedt, 1988, p. 18) indicate that shift


93
Figure 8. Group 1 mean learning curve with error band (standard
deviation) for dial monitoring reaction time during Task 2 (Tracking).
Figure 9. Group 2 mean learning curve with error band (standard
deviation) for dial monitoring reaction time during Task 2 (Tracking).


123
Rosekind, M. R., Gander, P. H., Miller, D. L, Gregory, K. B.,
McNally, K. L, Smith, R. M., & Lebacqz, J. V. (1993). NASA Ames
Fatigue Countermeasures Program. FAA Aviation Safety lournal.
3(1), 20-25.
Saito, Y. Z., Cornelissen, G., Sonkowsky, R., Saito, Y. K., Saito, Y. I.,
Saito, J., Wu, J., Hillman, D., Wang, Z., Hata, Y., Tamura, K., Halberg, E
& Halberg, F. (1992). Gradual adjustment of circaseptan-circadian
blood pressure and heart rate rhythms after a trans-9-meridian
flight. Chronobiologia. 19. 67-74.
Saldivar, J. T., Hoffman, S. M & Melton, C. E. (1977, February).
Sleen in air traffic controllers (Report No. FAA-AM-77-5).
Washington, DC: Civil Aeromedical Institute, Office of Aviation
Medicine, Federal Aviation Administration.
Shadmehr, R. & Holcomb, H. (1997, August 8). Neural correlates
of motor memory consolidation. Science. 277. 821-825.
Smith, R. C., Melton, C. E., & McKenzie, J. M. (1971). Affect
adjective check list assessment of mood variations in air traffic
controllers (Report No. FAA-AM-71-21). Washington, DC: Civil
Aeromedical Institute, Office of Aviation Medicine, Federal Aviation
Administration.
U. S. Congress, Office of Technology Assessment (1991,
September). Biological Rhythms: Implications for the Worker. OTA-
BA-463. Washington, DC: U. S. Government Printing Office.
U.S. Department of Transportation, Federal Aviation
Administration (1997, April). Administrators Fact Book. Washington,
DC: ABC-100.
U.S. Department of Transportation, Federal Aviation
Administration, Air Traffic Rules and Procedures Service (1996,
February 29). Facility Operations and Administration: 7110.3M.
Washington, DC: U.S. Government Printing Office.


45
workers on the whole report greater fatigue than do day workers.
This fatigue is usually greatest on the night shift, intermediate on the
morning shift, and hardly appears at all on the evening shift.
A record study by Torsvall and Akerstedt (as cited in Akerstedt,
1988) used both subjective, self-report measures of sleepiness, and
objective EEG/EOG ratings. The results of this study indicates that
individuals perceive sleepiness coming before it becomes manifested
on objective ratings. It therefore appears that sleepiness is
perceived by the individual well before he or she is overcome by
sleep. Thus, one should be able to use subjective sleepiness as a
signal or warning that involuntary sleep might ensue (Akerstedt,
1988, p. 22)
Akerstedt & Torsvall (1978) used an experimental design which
showed that reported fatigue increased upon entering shift work,
and decreased upon leaving it. Also, Akerstedt et al. (1983)
determined that sleepiness has been severe enough to have resulted
in actual incidents of individuals falling asleep during the night shift.
These studies have since been duplicated by other researches.
Folkard and Condon (1987) demonstrated a somewhat unusual
manifestation of night shift sleepiness reported by night shift nurses
which came to be known as night shift paralysis. This is a rare
phenomenon which seems to occur as a function of sleep loss (due to


112
learners characteristics, which should include an individuals
chronopsychological abilities to learn different tasks at different
times of the day.
If the results of this study are supported by future research
findings, practitioners may classify the comparable lack of attention
allocation ability (which does not improve during the learning of a
cognitive task) on the part of the day and evening shift workers as a
learner characteristic typical of this shift system. An instructional
intervention, such as Interactive Computer-Based Training (I-CBT),
interactive class discussion, or interactive small-group activities
could then be developed to address this learner characteristic so that
the trainee could be actively involved in the instructional unit,
thereby directing his/her attention to the training task and
maximizing the learners ability to learn, in spite of this deficiency.
Recommendations
Since earlier studies (Melton, 1985; Monk et al., 1988; Saldivar et
al., 1977; Smith et al., 1971) indicate that controllers who work
rapidly-rotating schedules report a high level of subjective fatigue
and a lack of alertness on the day shift, the results of this study seem
to indicate that subjective alertness ratings may not necessarily be
good indicators of attention allocation ability. In order to determine
the relationship between subjective ratings of alertness and objective


76
3. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of reaction time on dial
monitoring during the spatial visualization task (Task #1).
4. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of reaction time on dial
monitoring during the tracking task (Task #2).
5. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of reaction time for the transition
from Task #2 to Task #1 (which indicates response flexibility in
switching resources).
6. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of


113
measures of attention allocation ability, future research should
compare these variables for individuals working various shift
systems.
Also, as discussed earlier, future research of this type should
keep a continuous record of the participants sleep-wake cycles. The
bedtimes and wake-up times should be logged for each participant
during the entire course of an investigation, in order to determine
factors such as irregular sleep-wake cycles and total sleep obtained.
Chronopsychology is a fairly new field of study, and, as Adan
(1993) points out, there is a great deal of heterogeneity in the results
of its research due to the diversity of methods that have been used.
One step has been taken to remedy this situation in the classification
which Adan has provided to impose coherence on the heterogeneity
of the existing research results for both the subjective and objective
test used in chronopsychoiogical research. Although this is very
beneficial to the field, more work needs to be completed in order to
more clearly define psychological and performance rhythms and
their responses to various shift systems, so that practitioners in
education, and other fields may gain the benefit of the use of
predictive planning.
Industry can clearly benefit through a greater understanding of
chronopsychology, because workers performance on the job


96
Figure 14. Group 1 mean learning curve with error band (standard
deviation) for tracking error during Task 2 (Tracking).
Figure 15. Group 2 mean learning curve with error band (standard
deviation) for tracking error during Task 2 (Tracking).


97
illustrate, there are no Group Main Effects, indicating that the slopes
for both groups look similar in each case. There is, however a
statistically-significant Time Main Effect for all three variables,
meaning that both groups improved their reaction times over the
sessions on each variable. The eta squared value for this effect on
Variables 3 and 4 is large at 0.17 and 0.29 respectively, indicating a
large effect size. The eta squared value for Variable 5 is 0.10,
indicating between a moderate and large effect size.
The analyses of the residuals (Table 2) for Variables 3 (Figures
10 and 11), 4 (Figures 12 and 13), and 5 (Figures 14 and 15) all
indicate no statistically-significant difference between the two
groups. This means that there was no Group/Time Interaction
between the two groups error bands. It also means that the error
bands are statistically equal in size for the two groups on each
variable, and that the size of the error bands did not change to a
statistically-significant degree for either group over the sessions.


4
at least some sleep. Since sleep fragmentation has been shown
to significantly decrease performance and mood (Bonnet, 1986;
Bonnet, 1989), it is possible that the effects of partial sleep
deprivation more closely resemble those of sleep fragmentation
than those of total sleep deprivation. Furthermore, partial sleep
deprivation could have a unique effect on certain psychological
variables. Decreased interest and attention, for example, are
thought to be two prominent variables related to total sleep
deprivation (Meddis, 1982) and could be investigated with
partial sleep deprivation.... In sum, the effects of partial sleep
deprivation need to be more thoroughly investigated,
particularly since partial sleep loss is a relatively common
condition in our society, (p.324)
Statement of the Problem
The Federal Aviation Administration (FAA) defines an
Operational Error as; An occurrence attributable to an element of
the air traffic system which: (1) Results in less than the applicable
separation minimum between two or more aircraft, or between an
aircraft and terrain or obstacles as required by FAA 7110.65, Air
Traffic Control, and supplemental Instructions. Obstacles include
vehicles, equipment, and personnel on runways; or (2) Aircraft lands
or departs on a runway closed to aircraft operations after receiving
air traffic authorization (U.S. Department of Transportation, Federal
Aviation Administration, Air Traffic Rules and Procedures Service,
1996, p. 5-1-1). Each year there are a number of operational errors
in the air traffic system, many of which are attributed to controller
error. In 1996, there were 792 operational errors


29
Performance data based on the Civil Aeromedical Institute
Multiple Task Performance Battery suggest that (1) diurnal variation
was present during the preshift period, (2) performance decrements
occurred on the day of the shift following the short sleep period, (3)
performance rated relatively high for most of the day, but became
poor toward the end of the shift for the first three days following the
shift in routine, (4) performance on the fourth through the sixth post
shift days rated at or above average, with relatively small variations
among the five test sessions per day, and (5) performance on the
seventh through ninth post shift days was below average for the
experiment and demonstrated evidence of a return to a diurnal
pattern, which reflected the post shift sleep-wake cycle (Higgins et
aL, 1975, p. 1).
According to Higgins et al. (1975), the implications of the
findings cited in this study include the following:
(1) Individuals making a 12-hour alteration in the wake-sleep
cycle should not perform critical tasks during the first awake
period following the change. (2) After the first full sleep period
following the change, subjects appeared to perform well even
though the physiological and biomedical parameters measured
were still adjusting to the change. (3) For the first week
following the change in the wake-sleep cycle, individuals should
not work longer than 8 hours continuously because performance
deteriorates after that time. After the change, subjects appeared
to fatigue more rapidly toward the end of the awake period than
they did normally. This effect was evident for several days after
the change, (p. 23)


10
research ignores the time-of-day factor as both a procedural variable
and as a means of control in experiments. This oversight can cause
erroneous results in that behavioral parameters are not trivial and
can be equivalent in magnitude to the effects experienced from
limiting sleep to three hours or from ingesting the legal driving limit
of alcohol.
The fact that chronopsychological parameters are significant in
diurnal circadian conditions is ground enough to consider addressing
these psychological rhythms when designing instruction. By so
doing, designers may take advantage of learners abilities to
accomplish different objectives at different times of the day.
Furthermore, the complex relationship of these same psychological
variables and their resulting learner characteristics, when influenced
by a variety of work schedule rotations, can and should be the focus
of further research. Only then can the typical psychological
principles concerning the chronopsychological effects of each type of
schedule become understood. Only then can instruction be designed
to take advantage of these same variables for shift workers.
Null Hypotheses
1. Air traffic controllers who work rapidly-rotating work schedules,
will show no significant difference from controllers who work day-
shift and/or evening-shift schedules on their NovaScan mean


55
Sleep debt can also impede performance dramatically as evidenced
by the phenomenon known as night shift paralysis, which prevents
the worker from performing his/her job for several minutes.
According to Folkard (1985), it appears that, unless the shift worker
is engaged in a particularly crucial but relatively simple task, the
advantages of permanent shift systems are outweighed by their
disadvantages. Rapidly-rotating shift systems minimize cumulative
sleep debt and are more advantageous for the performance of
memory-loaded, cognitive tasks which are being performed more
and more by shift workers (p. 39).
It has been observed, however, that individuals who work
rotating work schedules receive lower job performance ratings in
some cases. The U.S. Congress, Office of Technology Assessment
(1991) reports on the findings of a study concerning the performance
ratings of nurses. Job performance of nurses, as measured by a
questionnaire filled out by supervisors, was found to be lower in
those on a rotating shift than in those on fixed day, afternoon, or
night shifts (p. 99).
It has also been demonstrated that extended work hours have a
negative effect on performance. Rosa and Bonnet (1993) conducted
the second of two studies which concerned a change from an 8-hour
work schedule to a 12-hour compressed work schedule. After ten


64
scores at mid-morning and late at night. Performance speed
indicated a morning rise and afternoon fall. Oral temperature of the
subjects peaked at 15:27, with no differences being found between
males and females. Overall efficiency displayed no significant
differences, indicating a trade-off between speed and accuracy over
the day, which acts to maintain performance efficiency. These
findings confirm Folkards observations (Brit. J. Psychol., 1975), while
performance is task dependent, speed and accuracy measures may
represent different diurnal components of performance with
correspondingly different relationships to arousal level (Englund,
1979, p. 96).
Taking all of this into consideration, Adan (1993) conducted a
literature review in order to classify the most commonly used
psychological tests, and to differentiate these tests on the basis of
variations obtained according to time-of-day factors. She indicates
that the incipient state of chronopsychology stems from two basic
facts. First, there is a great heterogeneity in the results obtained
which is a consequence of the diversity of methods used, the samples
of subjects and the situations in which the various experiments were
conducted. Second, most psychological investigations ignore the
time-of-day factor as both a procedural variable and as a means of
control in experiments. Adan further explains:


17
Summary
In the previous pages, the problem statement and need for the
study were identified. The problem is that, each year many
controller errors lead to aircraft accidents. This study is necessary
because many air traffic controllers work rotating shift work and are
susceptible to the adverse effects of circadian dysrhythmia, a
condition which causes decrements in alertness and performance
ability. These decrements could be a factor in controller learning
deficiencies during training sessions, and/or deficiencies in situation
awareness during dynamic air traffic control situations. In order to
address these factors, the hypotheses to be answered by this study
were proposed, and the underlying limitations, delimitations and
assumptions of the investigation were discussed.
Chapter two will discuss the related literature and research
findings concerning circadian rhythms and circadian dysrhythmia.
This chapter will also discuss shift work and the different types of
shift systems used in industry and air traffic control facilities.
Following this is a review of literature concerning the decrements in
alertness and performance caused by circadian dysrhythmia. Finally,
a review of the literature concerning circadian rhythms and learning
is presented.


400
200
0
I 3 5 7 9 II 13 15 17 19 21 23 25
Session
Figure 1. Mean learning curves for dial monitoring reaction time
during Task 1 (Spatial Visualization).
Figure 2. Mean learning curves for dial monitoring reaction time
during Task 2 (Tracking).


124
Vener, K.J., Szabo, S & Moore, J. G. (1989). The effects of shift
work on gastrointestinal (GI) function: A review. Chronobiologia. 16.
421-439.
Winget, C. M., DeRoshia, C. W., Markley, C. L, & Holley, D. C.
(1984). A review of human physiological and performance changes
associated with desynchronosis of biological rhythms. Aviation.
Snace. & Environmental Medicine. 55. 1085-1096.
Wright, J. E Vogel, J. A., Sampson, J. B., Knapik, J. J., Patton, J. F
& Daniels, W. L. (1983, February). Effects of travel across time zones
(Jet-Lag) on exercise capacity and performance. Aviation. Space, and
Environmental Medicine. 132-137.


78
1. The two baseline measures of attention allocation for each
individual were calculated as the average reaction time on dial
monitoring during tasks 1 and 2 from the last 3 sessions taken on
NovaScan. Once all the participants had completed their 25 th session
on NovaScan, a t-Test for independent samples was completed for
each measure with p=0.05 defined as statistically significant. The
results of these t-Tests were used to compare the two groups of
participants on each of these two measures and to address
hypotheses 1 and 2.
2. The learning curves for each individual were determined for the
five NovaScan variables: (a) Attention allocation during the spatial
visualization task, (b) attention allocation during the tracking task,
(c) response flexibility in switching resources, (d) non-transition
reaction to spatial visualization, and (e) tracking error during
tracking. The mean learning curve for each variable was determined
for each group to be used to compare the two groups of participants
and to address hypotheses 3 through 7.
A Two-Way General Linear Model, Repeated-Measures Analysis,
which included three statistical test (Group Main Effect, Time Main
Effect, and Group/Time Interaction) was used to analyze the data for
the five variables mean learning curve slopes and their residuals
with p=0.05 defined as statistically significant, in order to compare


114
determines industrial efficiency, safety, and profit. Since a large
percentage of industrial employees work shift work, factors such as
efficiency, safety, and profit need to be understood in relation to the
chronopsychological effects of shift work, so that interventions may
be devised in the work place and in training for the purpose of
making improvements.
Because the effects of shift work are experienced by workers in
terms of cognitive ability or strategy, and performance at various
times of the day, further research should be conducted in order to
determine the resulting psychological tendencies, or learner
characteristics, associated with different shift systems. It is also
recommended that instructional designers and training specialists
are made aware of the results of chronopsychological research in
both their initial training phases, and in the field. These
practitioners should be encouraged to view the chronopsychological
tendencies which result from shift work, as learner characteristics
which need to be addressed in instruction, in order to help ensure
effective learning during training.
One type of chronopsychological research which could be
conducted in order to help establish some chronopsychological
learner characteristics would be based upon the findings of prior
research mentioned earlier. It is known that short-term and long-


6
controllers has demonstrated that a loss of situation awareness Is the
primary cause of controller error, and Endsley (1995) has described
a model of situation awareness which includes training as a
contributing factor to situation awareness in dynamic situations.
Controllers currently work a variety of shift-work schedules, but
in many facilities they are most likely to receive training during
administrative hours, while they are working day shifts. These
training sessions usually take the form of a group presentation in a
classroom setting and do not address the individual differences
among the learners chronopsychological rhythms. Based on the
aforementioned research, controllers who work dysrhythmia-
inducing work schedules should be experiencing, to at least some
degree, decrements in performance, somnolence, and most probably
a negative chronopsychological learning effect (a negative learning
effect due to a decreased interest or ability to concentrate and focus
their attention) while attending these training sessions. These
decrements can be expected to be present in these controllers due to
irregular sleep/wakefulness cycles, sleep loss (due to shift work
insomnia), and fatigue, which are all induced by rotating or irregular
work schedules.
Training involves instruction, and instruction is designed with
the goal of accomplishing the objectives of the training. Learning


Discussion
101
The procedure used in this investigation was necessary because
of the problems involved in recruiting volunteers for such an
enduring study at a busy air traffic control facility. The participants
could not be randomly assigned to the two groups under
investigation because the controllers were already assigned to work
schedules through a seniority system. The controllers with the most
seniority have more choice in what type of schedule they work,
while controllers with less seniority generally get stuck with what is
left and are typically assigned to work schedules with little choice in
the matter. Because of this, certain personality traits may be a factor
in which controllers work which schedules.
Another slight design problem which must be acknowledged is
that the testing intervals for the participants were not equal. Some
controllers kept fairly regular intervals between testing periods over
their 25 sessions, while other controllers kept irregular testing
intervals. These slight design problems may somewhat limit the
generalization of the results.
The findings of the attention allocation analyses indicate a
rejection of the null hypothesis for both hypotheses 1 and 2.
Contrary to expectation, Group 1 (Rapidly-Rotating Shift Workers)
demonstrated a better attention allocation baseline measure than did


Abstract of Dissertation Presented to the Graduate School
Of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CHRONOPSYCHOLOGICAL LEARNING EFFECTS
OF RAPIDLY-ROTATING SHIFT WORK
ON DAY-SHIFT ATTENTION
By
Raymon M. McAdaragh
May 1999
Chairman: Lee Mullally, Ph.D.
Major Department: Instruction and Curriculum
The purpose of this study was to determine if an objectively-
measurable day-shift attention decrement or learning deficiency
exists among rapidly-rotating shift workers. Research using
subjective rating scales indicates that this type of shift work is
associated with a lack of alertness during day-shift hours. An
attention decrement may cause learning deficiencies during day-shift
training sessions, or problems with dynamic situation awareness on
the job.
For this study, 37 air traffic controllers (aged under 46 years)
were recruited from the Jacksonville Air Route Traffic Control Center
as volunteer participants. Group 1 consisted of 18 rapidly-rotating
v


53
Performance
The different types of rotating shift schedules have differential
effects on the individuals working them, based upon the type of
circadian disruption they cause. As stated earlier, rapidly-rotating
work schedules allow the physiological circadian rhythms to remain
diurnal, while disrupting the psychological/behavioral rhythms, thus,
providing a tendency to disorient the individual. Work schedules
which rotate on a weekly basis have social/psychological advantages
for the individual, but they also have a direct effect on the
physiological rhythms because they disrupt these rhythms with a
new routine each week before they have time to adjust.
Physiological rhythms can take from seven to several days to adjust
to each new routine.
Performance rhythms, however, appear to be less effected on
the night shift by rapidly-rotating schedules than by weekly rotating
schedules. According to Folkard et al. (1976), rapidly-rotating shift
systems affect performance differently according to the memory load
/
involved in the task required of the individual. Low memory load
task performance shows a high positive correlation with the
physiological temperature rhythm and tends to be poor at night.
However, high memory load task performance is negatively


95
Figure 12. Group 1 mean learning curve with error band (standard
deviation) for reaction time to Task 1 (Spatial Visualization).
Figure 13. Group 2 mean learning curve with error band (standard
deviation for reaction time to Task 1 (Spatial Visualization).


47
independent of the normal sleep/wake cycle (Folkard et al.,
1985), and that alertness is low in the early hours of the
morning. Similarly, there is good evidence that the duration of
day sleeps taken between two successive night shifts is
considerably shorter than the duration of normal night sleeps
(Knauth et al., 1980). Thus there will be a cumulative partial
sleep deprivation over successive night shifts which could
account for the increased incidence of this paralysis on the
second or subsequent night shift. Further, there is also evidence
that night sleeps preceding a morning shift are shorter than
normal ones (Knauth et al., 1980). Finally, rigid sleepers, and
especially those who are also evening types, are less likely to be
able to sleep successfully prior to a night shift. (Folkard &
Condon,1987,p. 1361)
As stated earlier, Pilcher and Huffcut (1996) found, through a
meta-analysis of sleep deprivation studies, that sleep deprivation has
its greatest effect on mood with a lessor effect on cognitive
performance and motor performance, respectively. They also
determined that partial sleep deprivation (sleep loss which occurs
whenever there is a reduction in the usual amounts of sleep obtained
in a 24-hour period), rather than long-term or short-term sleep
deprivation, has a greater negative effect on mood and cognitive
performance. This is important to shift workers, because partial
sleep deprivation is one result of working irregular or rotating
schedules. The condition which occurs due to partial sleep
deprivation associated with shift work has been termed shift-work
insomnia (Akerstedt & Kecklund, 1991, p. 509).


83
the sessions. The standard deviations around the means of the
slopes were not the same for the groups when compared over time
and the groups did not stabilize around their means at the same rate.
The null hypothesis was rejected.
Hypothesis 5
No statistically-signiflcant Group Main Effect or Group/Time
Interaction was found between the two groups learning curves for
Variable 3: reaction time during the transition from Task 2 to Task 1
(response flexibility in switching resources), indicating that the
groups mean learning curves appear statistically equal. There was,
however, a statistically-signiflcant Time Main Effect with a large
effect size, indicating that both groups learning curves depict
improved reaction times over the sessions. No error-band effects
were found to be statistically significant, indicating no statistically-
signiflcant change in them over time, and no Group/Time Interaction
between the error bands. The null hypothesis was not rejected.
Hypothesis 6
No statistically-signiflcant Group Main Effect or Group/Time
Interaction was found between the two groups learning curves for
Variable 4: reaction time during Task 1 (spatial visualization),
indicating that the groups mean learning curves appear statistically
equal. There was, however, a statistically-signiflcant Time Main


BIOGRAPHICAL SKETCH
Raymon Michael McAdaragh was born February 19, 1951, in
Springfield, Ohio. He spent four years and eight months in the United
States Army from 1969 to 1974, serving as an air traffic controller
and a flight simulator instrument ground instructor. He completed
and received his first Bachelor of Science (B.S.) degree in biology
from Christopher Newport University in 1980.
Raymon began his current career as an air traffic control
specialist with the Federal Aviation Administration (FAA) in 1981.
In 1991, while working with the FAA, he completed his second
Bachelor of Science degree in Applied Science and Technology
(B.S.A.S.T.) through Thomas A Edison State College, Trenton, New
Jersey. After completing this second Bachelors degree, he began
graduate school at Embry-Riddle Aeronautical University, and
completed a Master of Aeronautical Science (M.AS.) degree in 1994.
In August of 1994, he was admitted to the University of Florida
where he completed his Ph.D. in educational technology.
125


75
for their sessions on NovaScan by taking their sessions on two or
three days of each week.
The procedure described above was necessary because the
controllers could not be scheduled to take their sessions at
predesignated times. The controllers were all volunteer participants,
and they could only be requested to report for their sessions during
their break times if/when they were willing to do so. Because of this
slight design problem, certain trends may have been missed in the
analysis which could have provided more information.
Null Hypotheses
Data were collected and analyzed to address the following:
1. Air traffic controllers who work rapidly-rotating work schedules,
will show no significant difference from controllers who work day-
shift and/or evening-shift schedules on their NovaScan mean
baseline rating of attention allocation for dial monitoring during the
spatial visualization task (Task #1).
2. Air traffic controllers who work rapidly-rotating work schedules
will show no significant difference from controllers who work day-
shift and/or evening-shift schedules on their NovaScan mean
baseline rating of attention allocation for dial monitoring during the
tracking task (Task #2).


ACKNOWLEDGMENTS
My most sincere appreciation goes to Dr. Lee Mullally, the chair
of my dissertation committee, for his scholarly support and guidance
in the preparation of this manuscript. Despite his busy schedule, he
spent a great deal of time reading drafts of my dissertation and
providing me with valuable and timely feedback. He was a very
supportive and encouraging advisor.
I would also like to thank Dr. Edward Wolfe for his time and
effort in providing me with the guidance and help that I needed in
the data analysis portion of this study. He contributed many hours
of research and guidance so that the correct analysis procedures
would be used to describe the data in a logical and meaningful way.
My appreciation also goes out to my other committee members
for their input and support. Dr. Sebastian Foti, Dr. Larry Loesch, and
Dr. Jeff Hurt were always there when I needed them.
Finally, 1 would like to thank my wife, Carol, who was always
helpful and supportive, and my two sons, Jeffrey and Eric, who, with
their mother, were very patient and understanding of my busy
schedule at school and at work. I plan to spend much more time
with them from now on.
ii


90
Figure 3. Mean learning curves for reaction time to transition from
Task 2 (Tracking) to Task 1 (Spatial Visualization).
Figure 4. Mean learning curves for reaction time to Task 1 (Spatial
Visualization).


58
strategies employed are compatible with the learning style and
capabilities of the learner. The learning style and capabilities of the
learner are the individual learner characteristics which are unique to
each learner and must be addressed in instructional planning.
Learner characteristics include such variables as social background,
experiential background, developmental level, motivation, content
knowledge, and learning style. Because these characteristics are
different for each individual, instructional strategies must be
planned to meet each learners needs.
Chronopsychological research has addressed some other learner
attributes as well. As stated earlier, the human short- and long-term
memory cycles, and the cycle of alertness all indicate peak periods
and low periods at specific times of the day in individuals with
normal, diurnal circadian rhythms. It has also been shown that high
memory-loaded tasks and tasks involving the synthesis of data are
accomplished more effectively at different times of the day for this
very reason (Adan, 1993; Folkard, Minors & Waterhouse, 1985;
Folkard & Monk, 1980). These, and other, chronopsychological
variables operate in rhythmic cycles in all individuals. As Adan
(1993) indicates, chronopsychological parameters are not trivial and
most psychological research has not even addressed time-of-day as a
consideration. She also explains that this could have a been cause for


106
These differential results may be due to differences between the
two groups in the timing of their sleep-wake cycles, or due to group
differences in the amount of total sleep obtained. Either of these
factors, or both, may be the cause.
Although the rapidly-rotating shift workers have a rotating
sleep-wake cycle, this cycle is kept to a fairly routine change over
the course of the week. These workers have two short breaks during
shift changes, known as quick-turn-arounds, when they change from
the evening shift to the day shift, and when they change from the
day shift to the midnight shift. Workers are routinely forced to
obtain their sleep during these two short breaks.
Unlike the rapidly-rotating shift workers, day and evening shift
workers have only one quick-turn-around each week, which takes
place when they change from the evening shift to the day shift.
These workers are forced to obtain their sleep at the same time each
week during this shift change, but they may be varying their sleep
schedule more erratically than the rapidly-rotating shift workers on
other nights of the week.
According to Manber et al. (1996), individuals who sleep at
irregular schedules (schedules which vary by two hours or more)
experience a lack of alertness during the daytime. It may be that the
type of sleep-wake cycle practiced by the day and evening shift


Practical
ill
Instructional designers and instructors, alike, should be made
aware of the diurnal chronopsychological learning effects which have
been reported in the literature of chronopsychology. Based on this
research (e.g. Englund, 1979; Folkard & Monk, 1980; Monk, 1989;
Adan, 1993), these effects have direct implications to the timing of
instruction based upon the desired instructional objective.
Practitioners should also be made aware of the negative and
positive effects of different shift systems on learning, at different
times of the day. With this knowledge, practitioners will become
capable of making use of the advantages of the various shift systems,
while avoiding their disadvantages, when developing and delivering
instruction. As more information concerning the chronopsychological
learning effects of shift work becomes available, practitioners may
become capable of using the type of shift system worked by an
individual to identify the probable learner characteristics associated
with that shift system. Practitioners may then develop instructional
interventions to address these various learner characteristics.
Performance on the job can be affected by both performance
ability, which is influenced by an individuals circadian rhythmic
phases at the time of work, and by training, which is affected by
instructional design. Instructional design includes the analysis of a


22
Other evidence that supports the idea that circadian rhythms are
endogenously generated includes the demonstration that they
can be modified by selective breeding (Bunning, 1973) and gene
mutations (Konopka, 1980) and that they develop normally in
successive generations of organisms kept in light or dark
(Aschoff, 1960; Davis, 1981). Circadian rhythms are thus innate,
rather than learned or imprinted, phenomena. Virtually all
researchers now agree that circadian rhythms are the products
of an internal biological clock mechanism.
(Mistlberger & Rusak, 1989, p. 142)
There are several recognized environmental influences which
serve as entraining agents to rhythmic cycles. Many organisms,
including humans, find it adaptive to restrict rest and activity to
specific times of the day in order to take advantage of environmental
conditions which are optimal for various functions. By internalizing
the controlling mechanism for initiating rest and activity periods,
organisms are able to anticipate environmental events, such as
temperature, humidity, and changes in light intensity associated with
morning and nightfall, so that they may prepare accordingly.
Organisms accrue advantages by phasing their behaviors with these
environmental events, while anticipating them rather than merely
responding to their constraints. Therefore, some mechanisms must
exist which ensure that the internal timing device of an organism
maintains a synchronous phase relation to the environment
(Mistlberger & Rusak, 1989, p. 146).


TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS U
ABSTRACT V
CHAPTERS
1 INTRODUCTION 1
Statement of the Problem 4
Need for the Study 5
Null Hypotheses 10
Limitations 12
Delimitations. 13
Assumptions 16
Summary 17
2 RELATED LITERATURE 18
Circadian Rhythms 18
Circadian Dysrhythmia. 27
Shiftwork. 37
Overview 37
Work Schedules 38
Sleepiness 42
Performance 53
Circadian Rhythms and Learning 5 7
Summary. 69
3 METHODOLOGY. 70
Participants 70
Instruments 70
Hi


25
According to Dinges (1989), research also indicates that the onset
and offset of sleep and wakefulness display a cycle which reflects the
biological temperature cycle. The temperature cycle reaches its low
point at about 4 am and its acrophase at about 4 pm. As
temperature rises, sleep offset generally occurs, and as temperature
decreases, sleep onset generally occurs. The probability for sleep
onset and sleep offset over a 24-hour period in a normal diurnal
cycle follows a pattern:
1. (01:00-07:00) sleep onset probability high; sleep offset
probability low.
2. (07:00-13:00) sleep onset probability low; sleep offset probability
high.
3. (13:00-19:00) sleep onset probability high; sleep offset
probability high.
4. (19:00-01:00) sleep onset probability low; seep offset probability
low.
This pattern of sleep onset/offset propensity seems to indicate that
there are two periods during the cycle where sleep onset is desirable.
Although nocturnal sleep episodes appear in phase with the
circadian nadir in the endogenous core temperature cycle, naps
occur across the typically broad peak (variable acronhase) of the
temperature cycle, suggesting the possibility of a secondary
sleep propensity approximately 180 degrees out of phase with
the circadian cycle in body temperature (Broughton, 1975;
Dingesetal., 1980). (Dinges, 1989, p.155)


19
Each of these variables displays a daily cycle which has its own peak
(acrophase) and low point (nadir).
Psychological and behavioral rhythms include such variables as
alertness, performance, and the habits of an individual. These
rhythms also display an acrophase and nadir. It has been
demonstrated that short-term memory and long-term memory have
separate rhythms which peak at different times of the day (Folkard
& Monk, 1980).
ISome psychological/behavioral rhythms, such as digit
summation ability and short-term memory, and the biological
temperature rhythm tend to correspond and to influence one
another, pther psychological/behavioral rhythms such as alertness
and long-term memory tend to run in a staggered opposition to the
temperature rhythm. The alertness rhythm, for example, tends to
peak between 10 a.m. and noon, while the temperature rhythm
tends to peak between 4 p.m. and 8 p.m. (Hawkins, 1987; U.S.
Congress, Office of Technology Assessment, 1991).
Circadian rhythmicity stems from two causes, an endogenous
cause, the internal clock, and an exogenous cause, the rhythmic
environment and habits of the individual (Folkard et al., 1985, p.
33). By the late 1970s, it had become clear that most of the bodys
rhythms are controlled by processes deep within the brain (Hawkins,


105
the tracking task (Variable 5), and there were no Group/Time
Interactions between the groups.
As can be seen in figures 10 through 15, the size of the residuals
around each groups mean remains constant over the sessions for
these variables. Because of this, there is no stabilization around the
groups means over the sessions, as with the two measures of
attention allocation. This is because variables 3, 4, and 5 are learning
tasks within the NovaScan testing paradigm, and the individual
differences within each group remain constant over the sessions
during the learning of these tasks.
Based on these results, it appears that rapidly-rotating shift
workers have a better ability at attention allocation than do day and
evening shift workers, while completing both cognitive and
psychomotor tasks. It also appears that rapidly-rotating shift
workers can significantly improve their attention allocation while
learning a cognitive task, while day and evening shift workers can
not, although the effect size for this result is only slightly meaningful.
Finally, it appears that neither rapidly-rotating shift workers nor day
and evening shift workers have the ability to improve their attention
allocation to a statistically-significant degree while completing a
psychomotor task, and these two groups stabilize around their
baseline measures for this type of task at differential rates.



PAGE 1

&+521236<&+2/2*,&$/ /($51,1* ())(&76 2) 5$3,'/<527$7,1* 6+,)7 :25. 21 '$<6+,)7 $77(17,21 %\ 5$<021 0 0&$'$5$*+ $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8//),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

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

PAGE 3

7$%/( 2) &217(176 3DJH $&.12:/('*0(176 8 $%675$&7 9 &+$37(56 ,1752'8&7,21 6WDWHPHQW RI WKH 3UREOHP 1HHG IRU WKH 6WXG\ 1XOO +\SRWKHVHV /LPLWDWLRQV 'HOLPLWDWLRQV $VVXPSWLRQV 6XPPDU\ 5(/$7(' /,7(5$785( &LUFDGLDQ 5K\WKPV &LUFDGLDQ '\VUK\WKPLD 6KLIWZRUN 2YHUYLHZ :RUN 6FKHGXOHV 6OHHSLQHVV 3HUIRUPDQFH &LUFDGLDQ 5K\WKPV DQG /HDUQLQJ 6XPPDU\ 0(7+2'2/2*< 3DUWLFLSDQWV ,QVWUXPHQWV +L

PAGE 4

0DWHULDOV$SSDUDWXV 'HVLJQ 3URFHGXUH 1XOO +\SRWKHVHV 9DULDEOHV 'DWD $QDO\VLV 6XPPDU\ 5(6($5&+ ),1',1*6 5HVXOWV +\SRWKHVLV +\SRWKHVLV +\SRWKHVLV +\SRWKHVLV +\SRWKHVLV +\SRWKHVLV +\SRWKHVLV $WWHQWLRQ $OORFDWLRQ /HDUQLQJ &XUYH &RPSDULVRQV &21&/86,21 ,QWURGXFWLRQ 'LVFXVVLRQ ,PSOLFDWLRQV 7KHRUHWLFD/ 3UDFWLFD/ ,OO 5HFRPPHQGDWLRQV 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ ,9

PAGE 5

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f ZHUH UHFUXLWHG IURP WKH -DFNVRQYLOOH $LU 5RXWH 7UDIILF &RQWURO &HQWHU DV YROXQWHHU SDUWLFLSDQWV *URXS FRQVLVWHG RI UDSLGO\URWDWLQJ Y

PAGE 6

WKUHHVKLIW ZRUNHUV PDOH IHPDOHf DQG *URXS FRQVLVWHG RI GD\HYHQLQJ WZRVKLIW ZRUNHUV PDOH IHPDOHf (DFK SDUWLFLSDQW FRPSOHWHG VHVVLRQV RQ WKH 1RYD6FDQ FRPSXWHUEDVHG SHUIRUPDQFH WHVW WR GHWHUPLQH WZR EDVHOLQH PHDVXUHV RI DWWHQWLRQ DOORFDWLRQ RQH GXULQJ D VSDWLDOYLVXDOL]DWLRQ WDVN DQG WKH RWKHU GXULQJ D WUDFNLQJ WDVNf 7KH PHDQ EDVHOLQH PHDVXUHV IRU WKH JURXSV ZHUH FRPSDUHG XVLQJ WWHVWV 7KH PHDQ OHDUQLQJ FXUYHV IRU WKH JURXSV ZHUH FRPSDUHG RQ ILYH YDULDEOHV DWWHQWLRQ DOORFDWLRQ GXULQJ WKH VSDWLDOYLVXDOL]DWLRQ WDVN DWWHQWLRQ DOORFDWLRQ GXULQJ WKH WUDFNLQJ WDVN UHVSRQVH IOH[LELOLW\ LQ VZLWFKLQJ UHVRXUFHV QRQWUDQVLWLRQ UHDFWLRQ WR VSDWLDO YLVXDOL]DWLRQ DQG WUDFNLQJ HUURU GXULQJ WUDFNLQJf 7KH OHDUQLQJ FXUYHVf VORSHV DQG UHVLGXDOV ZHUH FRPSDUHG XVLQJ D WZRZD\ JHQHUDO OLQHDU PRGHO UHSHDWHGPHDVXUHV DQDO\VLV ZKLFK LQFOXGHG WKUHH VWDWLVWLFDO WHVWV JURXS PDLQ HIIHFW WLPH PDLQ HIIHFW DQG JURXSWLPH LQWHUDFWLRQf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

PAGE 7

WDVN 1R GLIIHUHQFHV ZHUH IRXQG EHWZHHQ WKH JURXSVf OHDUQLQJ FXUYHV IRU WKH WKUHH RWKHU YDULDEOHV ,W LV VXJJHVWHG WKDW GD\ DQG HYHQLQJ VKLIW ZRUNHUV PD\ SUDFWLFH LUUHJXODU VOHHSZDNH VFKHGXOHV DQGRU DFFXPXODWH D JUHDWHU VOHHS GHEW GXULQJ WKH ZRUN ZHHN WKDQ GR UDSLGO\URWDWLQJ VKLIW ZRUNHUV

PAGE 8

&+$37(5 ,1752'8&7,21 7KH RFFXSDWLRQ RI DLU WUDIILF FRQWUROOHU FDUULHV ZLWK LW D JUHDW UHVSRQVLELOLW\ LQ WHUPV RI ERWK OLIH DQG SURSHUW\ LQ WKDW WKH FRQWUROOHUfV MRE UHTXLUHV KLPKHU WR PDLQWDLQ WKH VDIH DQG H[SHGLWLRXV IORZ RI DLU WUDIILF LQ KLVKHU DUHD RI UHVSRQVLELOLW\ 0F$GDUDJK f KDV GHPRQVWUDWHG WKDW PDQ\ FRQWUROOHUV ZRUN URWDWLQJ ZRUN VFKHGXOHV WKDW KDYH EHHQ VKRZQ WR LQGXFH FLUFDGLDQ G\VUK\WKPLD D PLVDOLJQPHQW RU GHV\QFKURQL]DWLRQ RI DQ LQGLYLGXDOfV ELRORJLFDO DQGRU SV\FKRORJLFDO GDLO\ UK\WKPLF F\FOHV +DZNLQV f &KURQRSV\FKRORJLFDO UHVHDUFK WKH VWXG\ RI EHKDYLRUDO DQG SV\FKRORJLFDO UK\WKPVf LQGLFDWHV WKDW WKHVH W\SH RI ZRUN VFKHGXOHV SURGXFH D KLJK GHJUHH RI VXEMHFWLYH IDWLJXH LQ WKHVH LQGLYLGXDOV GXULQJ PLGQLJKW DQG GD\ VKLIWV ZLWK WKH ORZHVW GHJUHH EHLQJ UHSRUWHG RQ WKH HYHQLQJ VKLIWV 0HOWRQ 0RQN HW DOf 6DOGLYDU HW D/ 6PLWK HW DO f 7KH WHUP fVKLIWZRUN LQVRPQLDf KDV EHHQ XVHG WR GHVFULEH WKH FRQGLWLRQ FDXVHG E\ FLUFDGLDQ G\VUK\WKPLD ZKLFK UHGXFHV WKH WRWDO DPRXQW RI VOHHS DFKLHYHG E\ DIIHFWHG LQGLYLGXDOV ZKR ZRUN VKLIW ZRUN $NHUVWHGW t .HFNOXQG f

PAGE 9

%DENRII HW DO f GHWHUPLQHG WKDW VXEMHFWLYH VOHHSLQHVV UDWLQJV RI LQGLYLGXDOV GXULQJ SHULRGV RI VOHHS GHSULYDWLRQ DUH GHSHQGHQW XSRQ WKH SKDVH RI HDFK LQGLYLGXDOfV FLUFDGLDQ F\FOH 0DQEHU HW DO f GHPRQVWUDWHG WKDW HYHQ ZKHQ D QRUPDO IXOO QLJKWfV VOHHS LV DFKLHYHG E\ ERWK JURXSV LQGLYLGXDOV ZKR VOHHS DW LUUHJXODU VFKHGXOHV UHSRUW D OHVVHU DPRXQW RI DOHUWQHVV DQG D JUHDWHU DPRXQW RI VOHHSLQHVV GXULQJ WKH GD\ WKDQ GR LQGLYLGXDOV ZKR VOHHS DW UHJXODU VOHHSZDNH VFKHGXOHV %LOOLDUG HW DO f LQ D VWXG\ LQYROYLQJ )UHQFK PLOLWDU\ GUDIWHHV IRXQG WKDW VOHHS GLIILFXOWLHV DQG LUUHJXODU VOHHSZDNH VFKHGXOHV ZHUH PDMRU IDFWRUV FRQWULEXWLQJ WR H[FHVVLYH GD\WLPH VRPQROHQFH LQ \RXQJ PHQ EHWZHHQ WKH DJHV RI DQG \HDUV RI DJH &LUFDGLDQ G\VUK\WKPLD KDV DOVR EHHQ VKRZQ WR FDXVH SHUIRUPDQFH GHFUHPHQWV LQ DIIHFWHG LQGLYLGXDOV +LJJLQV HW DO 0RQN HW DO f DQG G\VUK\WKPLD DQG VOHHS GHSULYDWLRQ KDYH EHHQ VKRZQ WR SURGXFH D OHDUQLQJ HIIHFW 5DQNLQ HW DO f /XQD HW DO f LQ D VWXG\ RI DLU WUDIILF FRQWUROOHUV ZRUNLQJ D UDSLGO\ URWDWLQJ VFKHGXOH IRXQG WKDW FRQWUROOHUV RQ WKH PLGQLJKW VKLIW RI D IRUZDUG UDSLGO\URWDWLQJ VFKHGXOH DSSHDUHG WR EH IDOOLQJ DVOHHS DQG UHSRUWHG LQFUHDVHG FRQIXVLRQ DQG IDWLJXH ,Q D WHOHSKRQH FRQYHUVDWLRQ FRQFHUQLQJ WKLV VWXG\ RQ 0DUFK 'U )UHQFK DGYLVHG WKH DXWKRU WKDW DOWKRXJK WKH SHUIRUPDQFH GDWD FROOHFWHG ZDV

PAGE 10

QRW UHSRUWHG GXH WR D ODFN RI VXIILFLHQW GDWD WKH GDWD WKDW ZHUH FROOHFWHG LQGLFDWHG D VLJQLILFDQW QHJDWLYH LQLWLDO OHDUQLQJ HIIHFW IRU WKH FRQWUROOHUV ZKLOH WKH\ ZHUH RQ WKH PLGQLJKW VKLIW UHODWLYH WR ZKHQ WKH\ ZHUH RQ WKH GD\ RU HYHQLQJ VKLIWV 3LOFKHU DQG +XIIFXW f IRXQG WKURXJK D PHWDDQDO\VLV RI VOHHSGHSULYDWLRQ VWXGLHV WKDW VOHHS GHSULYDWLRQ KDV LWV JUHDWHVW HIIHFW RQ PRRG ZLWK D OHVVHU HIIHFW RQ FRJQLWLYH SHUIRUPDQFH DQG PRWRU SHUIRUPDQFH LQ GHVFHQGLQJ RUGHU 7KH\ DOVR GHWHUPLQHG WKDW SDUWLDO VOHHS GHSULYDWLRQ VOHHS ORVV ZKLFK RFFXUV ZKHQHYHU WKHUH LV D UHGXFWLRQ LQ WKH XVXDO DPRXQWV RI VOHHS REWDLQHG LQ D KRXU SHULRGf UDWKHU WKDQ ORQJWHUP RU VKRUWWHUP VOHHS GHSULYDWLRQ KDV D JUHDWHU QHJDWLYH HIIHFW RQ PRRG DQG FRJQLWLYH SHUIRUPDQFH 7KLV ,V LPSRUWDQW WR VKLIW ZRUNHUV EHFDXVH SDUWLDO VOHHS GHSULYDWLRQ UHVXOWV IURP WKH FRQGLWLRQ NQRZQ DV fVKLIWZRUN LQVRPQLDf ZKLFK LQ WXUQ UHVXOWV IURP ZRUNLQJ LUUHJXODU RU URWDWLQJ VFKHGXOHV 3LOFKHU DQG +XIIFXW DOVR LQGLFDWH WKDW RQH FOHDU JRDO RI IXWXUH UHVHDUFK ZLOO EH WR GHWHUPLQH ZK\ SDUWLDO VOHHS GHSULYDWLRQ KDV VXFK D SURQRXQFHG HIIHFW RQ PRRG DQG FRJQLWLYH SHUIRUPDQFH )RU H[DPSOH SDUWLDO VOHHS GHSULYDWLRQ PD\ DOWHU FHUWDLQ FLUFDGLDQ UK\WKP HIIHFWV RQ SHUIRUPDQFH DQG PRRG :KLOH WRWDO VOHHS GHSULYDWLRQ KDV EHHQ IRXQG WR LQWHUDFW ZLWK FLUFDGLDQ UK\WKPV 0RQN HW DO 1DLWRK HW DO f IHZ VWXGLHV KDYH LQYHVWLJDWHG WKH HIIHFWV RI SDUWLDO VOHHS GHSULYDWLRQ RQ FLUFDGLDQ UK\WKPV ,Q DGGLWLRQ SDUWLDO VOHHS GHSULYDWLRQ PD\ EH VLPLODU WR IUDJPHQWHG VOHHS LQ WKDW VXEMHFWV LQ ERWK FDVHV REWDLQ

PAGE 11

DW OHDVW VRPH VOHHS 6LQFH VOHHS IUDJPHQWDWLRQ KDV EHHQ VKRZQ WR VLJQLILFDQWO\ GHFUHDVH SHUIRUPDQFH DQG PRRG %RQQHW %RQQHW f LW LV SRVVLEOH WKDW WKH HIIHFWV RI SDUWLDO VOHHS GHSULYDWLRQ PRUH FORVHO\ UHVHPEOH WKRVH RI VOHHS IUDJPHQWDWLRQ WKDQ WKRVH RI WRWDO VOHHS GHSULYDWLRQ )XUWKHUPRUH SDUWLDO VOHHS GHSULYDWLRQ FRXOG KDYH D XQLTXH HIIHFW RQ FHUWDLQ SV\FKRORJLFDO YDULDEOHV 'HFUHDVHG LQWHUHVW DQG DWWHQWLRQ IRU H[DPSOH DUH WKRXJKW WR EH WZR SURPLQHQW YDULDEOHV UHODWHG WR WRWDO VOHHS GHSULYDWLRQ 0HGGLV f DQG FRXOG EH LQYHVWLJDWHG ZLWK SDUWLDO VOHHS GHSULYDWLRQ ,Q VXP WKH HIIHFWV RI SDUWLDO VOHHS GHSULYDWLRQ QHHG WR EH PRUH WKRURXJKO\ LQYHVWLJDWHG SDUWLFXODUO\ VLQFH SDUWLDO VOHHS ORVV LV D UHODWLYHO\ FRPPRQ FRQGLWLRQ LQ RXU VRFLHW\ Sf 6WDWHPHQW RI WKH 3UREOHP 7KH )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ )$$f GHILQHV DQ 2SHUDWLRQDO (UURU DV f$Q RFFXUUHQFH DWWULEXWDEOH WR DQ HOHPHQW RI WKH DLU WUDIILF V\VWHP ZKLFK f 5HVXOWV LQ OHVV WKDQ WKH DSSOLFDEOH VHSDUDWLRQ PLQLPXP EHWZHHQ WZR RU PRUH DLUFUDIW RU EHWZHHQ DQ DLUFUDIW DQG WHUUDLQ RU REVWDFOHV DV UHTXLUHG E\ )$$ $LU 7UDIILF &RQWURO DQG VXSSOHPHQWDO ,QVWUXFWLRQV 2EVWDFOHV LQFOXGH YHKLFOHV HTXLSPHQW DQG SHUVRQQHO RQ UXQZD\V RU f $LUFUDIW ODQGV RU GHSDUWV RQ D UXQZD\ FORVHG WR DLUFUDIW RSHUDWLRQV DIWHU UHFHLYLQJ DLU WUDIILF DXWKRUL]DWLRQf 86 'HSDUWPHQW RI 7UDQVSRUWDWLRQ )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ $LU 7UDIILF 5XOHV DQG 3URFHGXUHV 6HUYLFH S f (DFK \HDU WKHUH DUH D QXPEHU RI RSHUDWLRQDO HUURUV LQ WKH DLU WUDIILF V\VWHP PDQ\ RI ZKLFK DUH DWWULEXWHG WR FRQWUROOHU HUURU ,Q WKHUH ZHUH RSHUDWLRQDO HUURUV

PAGE 12

QDWLRQZLGH RU DERXW RSHUDWLRQDO HUURUV SHU IDFLOLW\ DFWLYLWLHV LQ WKH DLU WUDIILF V\VWHP 86 'HSDUWPHQW RI 7UDQVSRUWDWLRQ )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ S f 6RPH RSHUDWLRQDO HUURUV UHVXOW LQ DLUFUDIW DFFLGHQWV ,Q WKHUH ZHUH ODUJH DLU FDUULHU FRPPXWHU DLU WD[L JHQHUDO DYLDWLRQ DQG URWRU FUDIW DFFLGHQWV 86 'HSDUWPHQW RI 7UDQVSRUWDWLRQ )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ S f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fV f DQDO\VLV RI RSHUDWLRQDO HUURUV DQG ZRUNORDG LQ DLU WUDIILF

PAGE 13

FRQWUROOHUV KDV GHPRQVWUDWHG WKDW D ORVV RI VLWXDWLRQ DZDUHQHVV ,V WKH SULPDU\ FDXVH RI FRQWUROOHU HUURU DQG (QGVOH\ f KDV GHVFULEHG D PRGHO RI VLWXDWLRQ DZDUHQHVV ZKLFK LQFOXGHV WUDLQLQJ DV D FRQWULEXWLQJ IDFWRU WR VLWXDWLRQ DZDUHQHVV LQ G\QDPLF VLWXDWLRQV &RQWUROOHUV FXUUHQWO\ ZRUN D YDULHW\ RI VKLIWZRUN VFKHGXOHV EXW LQ PDQ\ IDFLOLWLHV WKH\ DUH PRVW OLNHO\ WR UHFHLYH WUDLQLQJ GXULQJ DGPLQLVWUDWLYH KRXUV ZKLOH WKH\ DUH ZRUNLQJ GD\ VKLIWV 7KHVH WUDLQLQJ VHVVLRQV XVXDOO\ WDNH WKH IRUP RI D JURXS SUHVHQWDWLRQ LQ D FODVVURRP VHWWLQJ DQG GR QRW DGGUHVV WKH LQGLYLGXDO GLIIHUHQFHV DPRQJ WKH OHDUQHUVf FKURQRSV\FKRORJLFDO UK\WKPV %DVHG RQ WKH DIRUHPHQWLRQHG UHVHDUFK FRQWUROOHUV ZKR ZRUN G\VUK\WKPLD LQGXFLQJ ZRUN VFKHGXOHV VKRXOG EH H[SHULHQFLQJ WR DW OHDVW VRPH GHJUHH GHFUHPHQWV LQ SHUIRUPDQFH VRPQROHQFH DQG PRVW SUREDEO\ D QHJDWLYH FKURQRSV\FKRORJLFDO OHDUQLQJ HIIHFW D QHJDWLYH OHDUQLQJ HIIHFW GXH WR D GHFUHDVHG LQWHUHVW RU DELOLW\ WR FRQFHQWUDWH DQG IRFXV WKHLU DWWHQWLRQf ZKLOH DWWHQGLQJ WKHVH WUDLQLQJ VHVVLRQV 7KHVH GHFUHPHQWV FDQ EH H[SHFWHG WR EH SUHVHQW LQ WKHVH FRQWUROOHUV GXH WR LUUHJXODU VOHHSZDNHIXOQHVV F\FOHV VOHHS ORVV GXH WR VKLIW ZRUN LQVRPQLDf DQG IDWLJXH ZKLFK DUH DOO LQGXFHG E\ URWDWLQJ RU LUUHJXODU ZRUN VFKHGXOHV 7UDLQLQJ LQYROYHV LQVWUXFWLRQ DQG LQVWUXFWLRQ LV GHVLJQHG ZLWK WKH JRDO RI DFFRPSOLVKLQJ WKH REMHFWLYHV RI WKH WUDLQLQJ /HDUQLQJ

PAGE 14

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n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

PAGE 15

DQ\ PHQWLRQ RI FKURQRSV\FKRORJ\ RU LWV SULQFLSOHV LQ WKH OLWHUDWXUH RI LQVWUXFWLRQDO GHVLJQ 6KLIW ZRUN DQG LQVWUXFWLRQDO GHVLJQ DUH XVXDOO\ IRXQG VKDULQJ D FRPPRQ QLFKH LQ VRFLHW\ ,QVWUXFWLRQDO GHVLJQ LV PRVW ZLGHO\ XVHG LQ LQGXVWU\ DQG WKH PLOLWDU\ ZKHUH LW KDV LWV IRXQGDWLRQV 7KHVH DUH DOVR DUHDV ZKHUH D ODUJH QXPEHU RI SHUVRQQHO ZRUN URWDWLQJ VKLIW ZRUN DQG QLJKW ZRUN ,Q IDFW VWDWLVWLFV VXJJHVW WKDW DSSUR[LPDWHO\ SHUFHQW RI WKH $PHULFDQ ZRUN IRUFH LV QRZ HQJDJHG LQ VKLIW ZRUN 9HQDU HW DO f 6KLIW ZRUN DQG QLJKW ZRUN KDYH EHHQ VKRZQ WR KDYH D QHJDWLYH HIIHFW RQ LQGLYLGXDOVf FLUFDGLDQ UK\WKPV WKH GDLO\ SK\VLRORJLFDO DQG SV\FKRORJLFDO UK\WKPV RI WKH ERG\ DQG PLQGf LQ PXFK WKH VDPH PDQQHU DV WUDQVPHULGLDQ IOLJKW SURGXFHV WKH QHJDWLYH VWDWH NQRZQ DV fMHW ODJf 7KH FRQGLWLRQ ZKLFK UHVXOWV LQ DQ LQGLYLGXDO LQ HLWKHU VLWXDWLRQ LV NQRZQ DV FLUFDGLDQ G\VUK\WKPLD D GHV\QFKURQL]DWLRQ RI WKH ERG\fV GDLO\ UK\WKPLF F\FOHV 5HVHDUFK LQGLFDWHV WKDW MHW ODJ RU VKLIW ZRUN DIIHFWV PDQ\ LQGLYLGXDOV ZLWK GHFUHPHQWV WR KHDOWK 9HQHU HW DO f DQG DV PHQWLRQHG HDUOLHU VXEMHFWLYH VOHHSLQHVV DQG IDWLJXH GXULQJ GD\ DQG QLJKW VKLIWV DV ZHOO DV RYHUDOO GHFUHPHQWV LQ SHUIRUPDQFH ,W KDV DOVR EHHQ GHPRQVWUDWHG WKDW LUUHJXODU VOHHSZDNH VFKHGXOHV DUH D FDXVH RI DOHUWQHVV GLVRUGHUV %LOOLDUG HW DO f DQG WKDW WKH WLPLQJ

PAGE 16

RI VOHHS WHUPLQDWLRQ PD\ EH FORVHO\ UHODWHG WR WKH VOHHSLQHVVDOHUWQHVV UK\WKP $NHUVWHGW t *LOOEHUJ f 7KLV OLQH RI UHVHDUFK LPSOLHV WKDW VKLIW ZRUNHUV VXFK DV DLU WUDIILF FRQWUROOHUV PD\ VXIIHU IURP DQ DOHUWQHVV GLVRUGHU WKDW FRXOG FDXVH DQ DWWHQWLRQ GHFUHPHQW LI VXFK D GHFUHPHQW ZHUH WR EH LGHQWLILHG LQ FRQWUROOHUV ZKR DUH HQJDJHG LQ URWDWLQJ VKLIW ZRUN GXULQJ GD\ VKLIW KRXUV ZKHQ WUDLQLQJ VHVVLRQV DUH VFKHGXOHG DW PDQ\ IDFLOLWLHVf LW FRXOG LPSO\ D QHHG WR DPHQG IXWXUH LQVWUXFWLRQDO VWUDWHJLHV WR DGGUHVV WKLV YDULDEOH ZKHQ FRQVLGHULQJ LQVWUXFWLRQDO DFWLYLWLHV DQG PHGLD VHOHFWLRQ 5HVHDUFK FRQFHUQLQJ GLXUQDO SV\FKRORJLFDO UK\WKPV DOVR GHPRQVWUDWHV LPSOLFDWLRQV IRU LQVWUXFWLRQDO GHVLJQ IRU QRUPDO GD\ VKLIW ZRUNHUV $FFRUGLQJ WR )RONDUG DQG 0RQN f VKRUWWHUP DQG ORQJWHUP PHPRU\ UK\WKPV GHPRQVWUDWH LPSOLFDWLRQV IRU WKH WLPLQJ RI LQVWUXFWLRQ GHSHQGLQJ RQ WKH W\SH RI LQVWUXFWLRQDO REMHFWLYH (QJOXQG f KDV GHPRQVWUDWHG D GLXUQDO IXQFWLRQ RI UHDGLQJ UDWH FRPSUHKHQVLRQ DQG HIILFLHQF\ ZKLFK LPSOLHV D UHODWLRQVKLS ZLWK PHPRU\ ERG\ WHPSHUDWXUH DQG JHQHUDO DFWLYLW\ F\FOHV RI FLUFDGLDQ UK\WKPV 5HDGLQJ UDWH DQG FRPSUHKHQVLRQ SHDN DW GLIIHUHQW WLPHV RI WKH GD\ ZKLOH UHDGLQJ HIILFLHQF\ LV PDLQWDLQHG $FFRUGLQJ WR $GDQ f DQDO\VLV RI WKH SV\FKRORJLFDO RU EHKDYLRUDO YDULDEOHV RI FLUFDGLDQ UK\WKPV FKURQRSV\FKRORJ\f LV UHFHQW DQG VWLOO GHYHORSLQJ 6KH LQGLFDWHV WKDW PRVW SV\FKRORJLFDO

PAGE 17

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f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

PAGE 18

EDVHOLQH UDWLQJ RI DWWHQWLRQ DOORFDWLRQ IRU GLDO PRQLWRULQJ GXULQJ WKH VSDWLDO YLVXDOL]DWLRQ WDVN 7DVN f $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO VKRZ QR VLJQLILFDQW GLIIHUHQFH IURP FRQWUROOHUV ZKR ZRUN GD\ VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV RQ WKHLU 1RYD6FDQ PHDQ EDVHOLQH UDWLQJ RI DWWHQWLRQ DOORFDWLRQ IRU GLDO PRQLWRULQJ GXULQJ WKH WUDFNLQJ WDVN 7DVN f $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI UHDFWLRQ WLPH RQ GLDO PRQLWRULQJ GXULQJ WKH VSDWLDO YLVXDOL]DWLRQ WDVN 7DVN f $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI UHDFWLRQ WLPH RQ GLDO PRQLWRULQJ GXULQJ WKH WUDFNLQJ WDVN 7DVN f $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI

PAGE 19

FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI UHDFWLRQ WLPH IRU WKH WUDQVLWLRQ IURP 7DVN WR 7DVN ZKLFK LQGLFDWHV UHVSRQVH IOH[LELOLW\ LQ VZLWFKLQJ UHVRXUFHVf $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI UHDFWLRQ WLPH RQ 7DVN QRQn WUDQVLWLRQ REMHFW RULHQWDWLRQ ZKLFK LQGLFDWHV VSDWLDO YLVXDOL]DWLRQf $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI WUDFNLQJ HUURU RQ 7DVN ZKLFK LQGLFDWHV PLQGPRWRU FRRUGLQDWLRQf /LPLWDWLRQV :RUN VFKHGXOH WKH SUHGLFWRU YDULDEOH ZDV WKH RQO\ YDULDEOH XVHG WR GLIIHUHQWLDWH WKH WZR JURXSV RI SDUWLFLSDQWV %HFDXVH RI WKLV ZRUN VFKHGXOH ZDV FRQVLGHUHG DV WKH VXVSHFWHG FDXVH IRU DQ\ GHWULPHQWV WKDW ZHUH IRXQG LQ WKLV VWXG\ &LUFDGLDQ UK\WKPLFLW\ FDQ

PAGE 20

RQO\ EH FRQVLGHUHG DV D SUREDEOH FDXVH EHFDXVH QR RWKHU VRFLDO RU SV\FKRORJLFDO GLIIHUHQFHV DPRQJ WKH SDUWLFLSDQWV ZHUH FRQWUROOHG 7KH ILUVW JURXS RI SDUWLFLSDQWV RQO\ LQFOXGHG FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ VFKHGXOHV ZKLFK DUH XVXDOO\ FRPSRVHG RI GD\ VKLIWV HYHQLQJ VKLIWV DQG QLJKW VKLIWV 7R EH LQFOXGHG LQ WKLV VWXG\ VFKHGXOHV IRU SDUWLFLSDQWV LQ WKLV JURXS LQFOXGHG DW OHDVW RQH PLGQLJKW VKLIW DQG RQH GD\ VKLIW HDFK ZHHN VR WKDW EHKDYLRUDO DQG SV\FKRORJLFDO FLUFDGLDQ UK\WKPV FRXOG EH H[SHFWHG WR EH G\VUK\WKPLF LQ WKH LQGLYLGXDOV ZRUNLQJ WKHP $FFRUGLQJ WR WKH OLWHUDWXUH WKLV JURXS VKRXOG EH FKDUDFWHUL]HG E\ GLXUQDO ELRORJLFDO UK\WKPV DQG GHV\QFKURQRXV EHKDYLRUDO DQG SV\FKRORJLFDO UK\WKPV 7KHUHIRUH RQO\ GHV\QFKURQRXV EHKDYLRUDO DQG SV\FKRORJLFDO FLUFDGLDQ UK\WKPV ZHUH VXVSHFWHG DV D SUREDEOH FDXVH IRU DQ\ GHFUHPHQW WR GD\VKLIW DWWHQWLRQ WKDW ZDV GHWHUPLQHG 1R GHWHUPLQDWLRQV ZHUH PDGH FRQFHUQLQJ ELRORJLFDO FLUFDGLDQ UK\WKPV 7KH VHFRQG JURXS RQO\ LQFOXGHG FRQWUROOHUV ZKR ZRUN GD\VKLIW DQG HYHQLQJVKLIW URWDWLQJ ZRUN VFKHGXOHV $FFRUGLQJ WR WKH OLWHUDWXUH WKLV JURXS LV H[SHFWHG WR EH FKDUDFWHUL]HG E\ GLXUQDO FLUFDGLDQ UK\WKPV ZLWK URWDWLQJ ZRUN DQG OHLVXUH WLPH 'HOLPLWDWLRQV 7KH LQVWUXPHQW XVHG LQ WKLV VWXG\ PHDVXUHG YDULDEOHV ZKLFK ZHUH XVHG WR GHWHUPLQH DWWHQWLRQ DOORFDWLRQ EDVHOLQH UDWLQJV DQG

PAGE 21

OHDUQLQJ FXUYHV IRU WKH SDUWLFLSDQWV 7KH 1RYD6FDQ LQVWUXPHQW ZDV GHVLJQHG DQG YDOLGDWHG WR PHDVXUH DQ LQGLYLGXDOfV SHUIRUPDQFH DJDLQVW KLVKHU RZQ EDVHOLQH RI DWWHQWLRQ DOORFDWLRQ RQFH LW KDV EHHQ HVWDEOLVKHG LQ RUGHU WR GHWHUPLQH UHDGLQHVV IRU GXW\ ,W KDV DOVR EHHQ YDOLGDWHG IRU HVWDEOLVKLQJ WKHVH UDWLQJV RI DWWHQWLRQ DOORFDWLRQ DQG IRU HVWDEOLVKLQJ ERWK LQGLYLGXDO DQG JURXS OHDUQLQJ FXUYHV IRU WKH YDULDEOHV ZKLFK LW PHDVXUHV 2f'RQQHOO S f 7KLV ZDV WKH ILUVW WLPH WKDW WKLV LQVWUXPHQW ZDV XVHG WR FRPSDUH WKH DWWHQWLRQ DOORFDWLRQ EDVHOLQHV RI WZR JURXSV RI LQGLYLGXDOV ZKR DUH H[SHFWHG WR GLIIHU RQ WKLV YDULDEOH DFFRUGLQJ WR OLWHUDWXUH (YDOXDWLRQ RI WKHVH UDWLQJV GHWHUPLQH LI RQH JURXS GLVSOD\HG D KLJKHU PHDQ DWWHQWLRQ DOORFDWLRQ EDVHOLQH UDWLQJ DV GHWHUPLQHG E\ 1RYD6FDQ WKDQ WKH RWKHU JURXS %HFDXVH 1RYD6FDQ KDV EHHQ YDOLGDWHG IRU PHDVXULQJ DWWHQWLRQ DOORFDWLRQ LW KDV ORJLFDO RU IDFHf YDOLGLW\ IRU XVH LQ WKLV VWXG\ ,W DOVR KDV FRQVWUXFW YDOLGLW\ EHFDXVH LW KDV EHHQ YDOLGDWHG WKURXJK UHVHDUFK IRU LWV H[WUHPH VHQVLWLYLW\ WR GHFUHPHQWV FDXVHG E\ GUXJV RU DOFRKRO 2f'RQQHOO S f 7KH OHDUQLQJ FXUYH GDWD ZHUH XVHG WR PDNH GHWHUPLQDWLRQV FRQFHUQLQJ DQ\ SRVVLEOH OHDUQLQJ HIIHFWV EHWZHHQ WKH JURXSV FRPSDUHG LQ WKLV VWXG\ 7KH FULWHULDQ YDULDEOHV PHDVXUHG E\ 1RYD6FDQ LQ WKLV VWXG\ GHWHUPLQHG HDFK SDUWLFLSDQWfV OHDUQLQJ FXUYH

PAGE 22

DQG EDVHOLQH UDWLQJ IRU WKH IROORZLQJ Df 7ZR PHDVXUHV RI DWWHQWLRQ DOORFDWLRQ 9DULDEOHV DQG f Ef UHVSRQVH IOH[LELOLW\ LQ VZLWFKLQJ UHVRXUFHV GXULQJ PXOWLSOH WDVNV SHUIRUPDQFH 9DULDEOH f Ff VSDWLDO YLVXDOL]DWLRQ DELOLW\ 9DULDEOH f DQG Gf VWUDLJKW PLQGPRWRU FRRUGLQDWLRQ 9DULDEOH f $OO RI WKHVH YDULDEOHV KDYH EHHQ VKRZQ E\ SULRU UHVHDUFK WR EH QHJDWLYHO\ DIIHFWHG E\ DFXWH VKLIWV LQ URXWLQH )RU H[DPSOH 0RQN HW DO f GHPRQVWUDWHG WKDW D SKDVH DGYDQFHPHQW RI VL[ KRXUV LQ WKH ZDNH SHULRG LQGXFHG MHW ODJ LQ WKH ODERUDWRU\f ZLWK PLGGOHDJH PDOH VXEMHFWV UHVXOWHG LQ VHYHUDO QHJDWLYH HIIHFWV 7KHVH QHJDWLYH HIIHFWV LQFOXGHG WKH IROORZLQJ Df 'HFUHDVHG VXEMHFWLYHO\UDWHG DOHUWQHVV ZKLFK FRUUHVSRQGV WR 1RYD6FDQ 9DULDEOHV DQG f Ef GHFUHDVHG SHUIRUPDQFH DW VHDUFK WDVNV ZKLFK FRUUHVSRQGV WR 1RYD6FDQ 9DULDEOH f DQG Ff GHFUHDVHG SHUIRUPDQFH DW PDQXDO GH[WHULW\ WDVNV ZKLFK FRUUHVSRQGV WR 1RYD6FDQ 9DULDEOH f +LJJLQV HW DO f GHPRQVWUDWHG WKDW D KRXU VKLIW LQ WKH ZDNH VOHHS F\FOH RI PDOH SDUWLFLSDQWV DJHV WR f SURGXFHG GHILFLHQFLHV LQ LQGLYLGXDOVf PXOWLSOH WDVNV SHUIRUPDQFH ZKLFK FRUUHVSRQGV WR 1RYD6FDQ 9DULDEOH f /XQD HW DO f IRXQG WKDW $LU )RUFH FRQWUROOHUV UHSRUWHG JUHDWHU DPRXQWV RI FRQIXVLRQ ZKLFK FRUUHVSRQGV WR 1RYD6FDQ 9DULDEOH f DQG OHVV YLJRU RQ QLJKW VKLIWV WKDQ ZKHQ RQ HYHQLQJ RU GD\ VKLIWV RI D UDSLGO\URWDWLQJ VFKHGXOH

PAGE 23

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

PAGE 24

6XPPDU\ ,Q WKH SUHYLRXV SDJHV WKH SUREOHP VWDWHPHQW DQG QHHG IRU WKH VWXG\ ZHUH LGHQWLILHG 7KH SUREOHP LV WKDW HDFK \HDU PDQ\ FRQWUROOHU HUURUV OHDG WR DLUFUDIW DFFLGHQWV 7KLV VWXG\ LV QHFHVVDU\ EHFDXVH PDQ\ DLU WUDIILF FRQWUROOHUV ZRUN URWDWLQJ VKLIW ZRUN DQG DUH VXVFHSWLEOH WR WKH DGYHUVH HIIHFWV RI FLUFDGLDQ G\VUK\WKPLD D FRQGLWLRQ ZKLFK FDXVHV GHFUHPHQWV LQ DOHUWQHVV DQG SHUIRUPDQFH DELOLW\ 7KHVH GHFUHPHQWV FRXOG EH D IDFWRU LQ FRQWUROOHU OHDUQLQJ GHILFLHQFLHV GXULQJ WUDLQLQJ VHVVLRQV DQGRU GHILFLHQFLHV LQ VLWXDWLRQ DZDUHQHVV GXULQJ G\QDPLF DLU WUDIILF FRQWURO VLWXDWLRQV ,Q RUGHU WR DGGUHVV WKHVH IDFWRUV WKH K\SRWKHVHV WR EH DQVZHUHG E\ WKLV VWXG\ ZHUH SURSRVHG DQG WKH XQGHUO\LQJ OLPLWDWLRQV GHOLPLWDWLRQV DQG DVVXPSWLRQV RI WKH LQYHVWLJDWLRQ ZHUH GLVFXVVHG &KDSWHU WZR ZLOO GLVFXVV WKH UHODWHG OLWHUDWXUH DQG UHVHDUFK ILQGLQJV FRQFHUQLQJ FLUFDGLDQ UK\WKPV DQG FLUFDGLDQ G\VUK\WKPLD 7KLV FKDSWHU ZLOO DOVR GLVFXVV VKLIW ZRUN DQG WKH GLIIHUHQW W\SHV RI VKLIW V\VWHPV XVHG LQ LQGXVWU\ DQG DLU WUDIILF FRQWURO IDFLOLWLHV )ROORZLQJ WKLV LV D UHYLHZ RI OLWHUDWXUH FRQFHUQLQJ WKH GHFUHPHQWV LQ DOHUWQHVV DQG SHUIRUPDQFH FDXVHG E\ FLUFDGLDQ G\VUK\WKPLD )LQDOO\ D UHYLHZ RI WKH OLWHUDWXUH FRQFHUQLQJ FLUFDGLDQ UK\WKPV DQG OHDUQLQJ LV SUHVHQWHG

PAGE 25

&+$37(5 5(/$7(' /,7(5$785( &LUFDGLDQ 5K\WKPV &LUFDGLDQ UK\WKPV DUH F\FOHV ZKLFK RFFXU RQ D GDLO\ EDVLV EXW WKH\ DUH RQO\ SDUW RI D ODUJHU V\VWHP RI UK\WKPV :H OLYH LQ D XQLYHUVH RI UK\WKPV 2XU JDOD[\ URWDWHV RQFH HYHU\ \HDUV 6XQ VSRWV RFFXU HYHU\ \HDUV DQG WLGHV HYHU\ KRXUV %XW WKH PRVW VLJQLILFDQW UK\WKP IRU WKH LQKDELWDQWV RI WKLV SODQHW LV WKH HDUWKfV URWDWLRQ HYHU\ KRXUV +DZNLQV S f &\FOHV ZKLFK DUH VKRUWHU WKDQ KRXUV DUH FDOOHG fXOWUDGLDQf DQG F\FOHV ZKLFK DUH JUHDWHU WKDQ KRXUV DUH FDOOHG fFLUFDPHQVXDOf $GDQ S f 2UJDQLVPV ZKLFK OLYH RQ RXU HDUWK KDYH HYROYHG ZLWKLQ WKLV XQLYHUVDO F\FOLF HQYLURQPHQW +XPDQ FLUFDGLDQ UK\WKPV DUH ERWK SK\VLRORJLFDO DQG SV\FKRORJLFDO LQ QDWXUH 7KH VWXG\ RI SK\VLRORJLFDO RU ELRORJLFDOf UK\WKPV LV NQRZQ DV FKURQRELRORJ\ DQG WKH VWXG\ RI SV\FKRORJLFDO DQG EHKDYLRUDOf UK\WKPV LV NQRZQ DV FKURQRSV\FKRORJ\ 7KH ELRORJLFDO UK\WKPV LQFOXGH VXFK YDULDEOHV DV KRUPRQH SURGXFWLRQ ERG\ WHPSHUDWXUH RUJDQ IXQFWLRQLQJ LPPXQH V\VWHP F\FOHV EORRG FHOO IXQFWLRQLQJ DQG VWRPDFK DQG LQWHVWLQDO WUDFW IXQFWLRQLQJ 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f

PAGE 26

(DFK RI WKHVH YDULDEOHV GLVSOD\V D GDLO\ F\FOH ZKLFK KDV LWV RZQ SHDN DFURSKDVHf DQG ORZ SRLQW QDGLUf 3V\FKRORJLFDO DQG EHKDYLRUDO UK\WKPV LQFOXGH VXFK YDULDEOHV DV DOHUWQHVV SHUIRUPDQFH DQG WKH KDELWV RI DQ LQGLYLGXDO 7KHVH UK\WKPV DOVR GLVSOD\ DQ DFURSKDVH DQG QDGLU ,W KDV EHHQ GHPRQVWUDWHG WKDW VKRUWWHUP PHPRU\ DQG ORQJWHUP PHPRU\ KDYH VHSDUDWH UK\WKPV ZKLFK SHDN DW GLIIHUHQW WLPHV RI WKH GD\ )RONDUG t 0RQN f ,6RPH SV\FKRORJLFDOEHKDYLRUDO UK\WKPV VXFK DV GLJLW VXPPDWLRQ DELOLW\ DQG VKRUWWHUP PHPRU\ DQG WKH ELRORJLFDO WHPSHUDWXUH UK\WKP WHQG WR FRUUHVSRQG DQG WR LQIOXHQFH RQH DQRWKHU SWKHU SV\FKRORJLFDOEHKDYLRUDO UK\WKPV VXFK DV DOHUWQHVV DQG ORQJWHUP PHPRU\ WHQG WR UXQ LQ D VWDJJHUHG RSSRVLWLRQ WR WKH WHPSHUDWXUH UK\WKP 7KH DOHUWQHVV UK\WKP IRU H[DPSOH WHQGV WR SHDN EHWZHHQ DP DQG QRRQ ZKLOH WKH WHPSHUDWXUH UK\WKP WHQGV WR SHDN EHWZHHQ SP DQG SP +DZNLQV 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f &LUFDGLDQ UK\WKPLFLW\ VWHPV IURP WZR FDXVHV fDQ HQGRJHQRXV FDXVH WKH LQWHUQDO FORFN DQG DQ H[RJHQRXV FDXVH WKH UK\WKPLF HQYLURQPHQW DQG KDELWV RI WKH LQGLYLGXDOf )RONDUG HW DO S f %\ WKH ODWH V LW KDG EHFRPH FOHDU WKDW PRVW RI WKH ERG\fV UK\WKPV DUH FRQWUROOHG E\ SURFHVVHV GHHS ZLWKLQ WKH EUDLQ +DZNLQV

PAGE 27

f +HUH WKH PRVW SRZHUIXO RI WKH HQWUDLQLQJ DJHQWV RU ]HLWJHEHUVf WKH F\FOH RI OLJKW DQG GDUNQHVV LV GHWHFWHG WKURXJK WKH H\HV DQG LQIOXHQFHV WKH ERG\fV YLVFHUDO V\VWHP DQG RWKHU F\FOLF IXQFWLRQV 2WKHU UK\WKPV KRZHYHU OLNH WKDW RI WKH ERG\ WHPSHUDWXUH UK\WKP DUH FRQWUROOHG HOVHZKHUH DQG UHPDLQ WKH REMHFW RI IXUWKHU LQYHVWLJDWLRQ $FWLYLWLHV VXFK DV PHDO WLPH DQG SK\VLFDO DQG VRFLDO DFWLYLW\ DOVR VHUYH DV HQWUDLQLQJ DJHQWV +DZNLQV f f§ $FFRUGLQJ WR 0LVWOEHUJHU DQG 5XVDN f SV\FKRORJLFDO DQG ELRORJLFDO UK\WKPV KDYH DQ LQWHUQDO UHODWLRQVKLS ZLWK WLPHRIGD\ UHVWUDLQWV +XPDQ EHKDYLRU DQG SK\VLRORJLFDO SURFHVVHV GHPRQVWUDWH D WHPSRUDO VWUXFWXUH WKDW PDWFKHV WKH KRXU GD\QLJKW F\FOH 7KH PRVW REYLRXV GDLO\ F\FOH WKDW KXPDQV H[SHULHQFH LV WKH F\FOH RI VOHHS DQG ZDNHIXOQHVV 6LPLODUO\ D P\ULDG RI ERGLO\ IXQFWLRQV LQFOXGLQJ HQGRFULQH VHFUHWLRQV ERG\ WHPSHUDWXUH UHJXODWLRQ VHQVRU\ SURFHVVLQJ DQG FRJQLWLYH SHUIRUPDQFH GHPRQVWUDWH D KRXU UK\WKPLFLW\ S f +LVWRULFDOO\ FLUFDGLDQ UK\WKPV ZHUH DVFULEHG WR HQYLURQPHQWDO FXHV DVVRFLDWHG ZLWK WKH VRODU GD\ EXW D ODUJH QXPEHU RI VWXGLHV KDYH FRQILUPHG WKDW GDLO\ UK\WKPV LQ D ZLGH YDULHW\ RI VSHFLHV LQFOXGLQJ KXPDQV SHUVLVW XQGHU FRQVWDQW HQYLURQPHQWDO FRQGLWLRQV %HFDXVH RI WKLV LW LV DFNQRZOHGJHG WKDW GDLO\ UK\WKPV DUH JHQHUDWHG

PAGE 28

LQWHUQDOO\ E\ WKH RUJDQLVP DQG DUH QRW VLPSO\ SDVVLYH UHVSRQVHV WR HQYLURQPHQWDO VWLPXOL S f ,Q WKH DEVHQFH RI HQYLURQPHQWDO FXHV GDLO\ UK\WKPV DUH VDLG WR EH fIUHHUXQQLQJf )UHHUXQQLQJ UK\WKPV DSSUR[LPDWH WKH KRXU GD\ EXW QRW H[DFWO\ 7KLV SHULRGLFLW\ WLPH UHTXLUHG IRU RQH FRPSOHWH UK\WKP F\FOHf LV WHUPHG fFLUFDGLDQf FLUFD DERXW GLHV GD\f +XPDQV ZKR DUH NHSW LQ LVRODWLRQ ZLWKRXW WLPH FXHV GLVSOD\ D KRXU VOHHSZDNH F\FOH 6OHHS RQVHWV EHJLQ WR RFFXU DW KRXU LQWHUYDOV LQVWHDG RI WKH QRUPDO KRXU GD\QLJKW LQWHUYDOV H[SHULHQFHG LQ RUGLQDU\ VRFLHW\ 8QGHU WKLV IUHHUXQQLQJ F\FOH DV ,W LV WHUPHG KXPDQV JR WR VOHHS ODWHU HDFK GD\ WKDQ WKH\ ZRXOG XQGHU QRUPDO FLUFXPVWDQFHV )UHHUXQQLQJ VOHHSZDNH F\FOHV YDU\ DPRQJ VSHFLHV IURP DERXW WR KRXUV DQG DUH PRGLILHG E\ IDFWRUV VXFK DV DPELHQW OLJKW LQWHQVLW\ RU KRUPRQH SURGXFWLRQ %XW WKH IDFW WKDW IUHH UXQQLQJ F\FOHV GLIIHU IURP WKH QRUPDO KRXU F\FOH LV VWURQJ HYLGHQFH WKDW FLUFDGLDQ UK\WKPV DUH JHQHUDWHG LQWHUQDOO\ DQG DUH QRW VLPSO\ UHVSRQVHV WR GDLO\ KRXU WLPH FXHV ,Q IDFW LQGLYLGXDOV RI WKH VDPH VSHFLHV UHFRUGHG XQGHU ODERUDWRU\ FRQGLWLRQV LQ DGMDFHQW FDJHV GHPRQVWUDWH GLIIHUHQW SHULRGLFLWLHV ZKLFK FDXVH WKH DQLPDOV WR VORZO\ GULIW DSDUW IURP RQH DQRWKHU ZKLOH VKDULQJ WKH VDPH HQYLURQPHQW 0LVWOEHUJHU t 5XVDN Sf

PAGE 29

2WKHU HYLGHQFH WKDW VXSSRUWV WKH LGHD WKDW FLUFDGLDQ UK\WKPV DUH HQGRJHQRXVO\ JHQHUDWHG LQFOXGHV WKH GHPRQVWUDWLRQ WKDW WKH\ FDQ EH PRGLILHG E\ VHOHFWLYH EUHHGLQJ %XQQLQJ f DQG JHQH PXWDWLRQV .RQRSND f DQG WKDW WKH\ GHYHORS QRUPDOO\ LQ VXFFHVVLYH JHQHUDWLRQV RI RUJDQLVPV NHSW LQ OLJKW RU GDUN $VFKRII 'DYLV f &LUFDGLDQ UK\WKPV DUH WKXV LQQDWH UDWKHU WKDQ OHDUQHG RU LPSULQWHG SKHQRPHQD 9LUWXDOO\ DOO UHVHDUFKHUV QRZ DJUHH WKDW FLUFDGLDQ UK\WKPV DUH WKH SURGXFWV RI DQ LQWHUQDO ELRORJLFDO FORFN PHFKDQLVP 0LVWOEHUJHU t 5XVDN S f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t 5XVDN S f

PAGE 30

/LJKW LV DV H[SHFWHG D XQLYHUVDO ]HLWJHEHU HQWUDLQLQJ DJHQWf IRU WKH DFWLYLW\UHVW F\FOH GLVSOD\HG E\ PRVW RUJDQLVPV IURP VLQJOH FHOOHG OLIH IRUPV WKURXJK PDPPDOV f$ UHFHQW FDVH UHSRUW LQGLFDWHV WKDW FLUFDGLDQ UK\WKPV RI ERG\ WHPSHUDWXUH DQG HQGRFULQH YDULDEOHV LQ DQ HOGHUO\ ZRPDQ FRXOG EH VKLIWHG E\ DSSURSULDWHO\ WLPHG H[SRVXUH WR EULJKW OLJKW LQGHSHQGHQWO\ RI WKH WLPLQJ RI VOHHSZDNH VWDWHVf 0LVWOEHUJHU t 5XVDN S f 7KHUH DUH DOVR VHYHUDO QRQSKRWLF VWLPXOL WKDW DSSHDU WR HQWUDLQ FLUFDGLDQ UK\WKPV 7KH WLPLQJ RI PHDOV LQIOXHQFHV FLUFDGLDQ F\FOHV WKURXJK WKH DQWLFLSDWRU\ DFWLYLW\ DVVRFLDWHG ZLWK PHDOWLPH LQ RUJDQLVPV 7KLV HIIHFW LV YHU\ HYLGHQW LQ H[SHULPHQWV ZLWK UDWV $URXVDO VWDWHV FDQ DOVR WULJJHU WKH FLUFDGLDQ WLPLQJ V\VWHP 6LPSOH KDQGOLQJ RI ZKLWHIRRWHG PLFH 5DZVRQ f RU FKDQJLQJ WKH OLWWHU RI KDPVWHUV 0URVRYVN\ t +DOORQTXLVW f FDQ SKDVH VKLIW IUHHUXQQLQJ UK\WKPV VXJJHVWLQJ WKDW HYHQ EULHI DURXVDOV PD\ KDYH IHHGEDFN HIIHFWV RQ FLUFDGLDQ WLPLQJ ,Q KXPDQV UHFRUGHG LQ WHPSRUDO LVRODWLRQ D GDLO\ fDQFKRU VOHHSf K RI IRUFHG EHG UHVW DW D IL[HG WLPH HYHU\ Kf FDQ DSSDUHQWO\ V\QFKURQL]H GDLO\ UK\WKPV 0LQRUV t :DWHUKRXVH f EXW ZKHWKHU WKLV V\QFKURQ\ LV UHODWHG WR DQFKRU VOHHS SHU VH RU GDLO\ /' OLJKWGDUN F\FOHf RU IRRG LQWDNH VFKHGXOHV LV XQFOHDU 0LVWOEHUJHUr 5XVDN S f 6RFLDO FXHV DUH DQRWKHU ZHOONQRZQ HQWUDLQLQJ DJHQW LQ PDQ\ VSHFLHV LQFOXGLQJ KXPDQV KRZHYHU WKH\ DUH QRW XQLYHUVDOO\ HIIHFWLYH ,W KDV EHHQ GHPRQVWUDWHG WKDW VRPH EOLQG KXPDQV GLVSOD\ IUHHUXQQLQJ VOHHSZDNH F\FOHV LQ GLXUQDO HQYLURQPHQWDO FRQGLWLRQV

PAGE 31

7KHVH LQGLYLGXDOV DUH QRW V\QFKURQL]HG E\ VRFLHW\fV VRFLDO FXHV -XVW KRZ VRFLDO FXHV LQIOXHQFH FLUFDGLDQ WLPLQJ LV QRW NQRZQ EXW WKH VRFLDO FXHVf HIIHFWV RQ DURXVDO VWDWHV DUH SUREDEO\ WKH QHFHVVDU\ PHGLDWRUV IRU WKH UHODWLRQVKLS 0LVWOEHUJHU t 5XVDN S f 'LQJHV f GHVFULEHV KRZ VOHHS LV DIIHFWHG E\ WKH FLUFDGLDQ WLPHNHHSLQJ V\VWHP $FFRUGLQJ WR 'LQJHV WKH GDLO\ VOHHSZDNH F\FOH VHUYHV WR RUJDQL]H ZDNH EHKDYLRU LQWR GLVFUHWH WHPSRUDO XQLWV DQG WR FRRUGLQDWH DQG V\QFKURQL]H WKH LQWHUQDO WLPLQJ RI PDQ\ ELRORJLFDO UK\WKPV :H FDQQRW NQRZ IRU VXUH ZKHWKHU HDUO\ PDPPDOV ZHUH GLXUQDO RU QRFWXUQDO EXW WKHUH LV OLWWOH GRXEW WKDW SULPDWHV DQG KRPR VDSLHQV LQ JHQHUDO HYROYHG DV GLXUQDO VSHFLHV ZKR VOHHS SULPDULO\ DW QLJKW S f 'LQJHV f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f

PAGE 32

$FFRUGLQJ WR 'LQJHV f UHVHDUFK DOVR LQGLFDWHV WKDW WKH RQVHW DQG RIIVHW RI VOHHS DQG ZDNHIXOQHVV GLVSOD\ D F\FOH ZKLFK UHIOHFWV WKH ELRORJLFDO WHPSHUDWXUH F\FOH 7KH WHPSHUDWXUH F\FOH UHDFKHV LWV ORZ SRLQW DW DERXW DP DQG LWV DFURSKDVH DW DERXW SP $V WHPSHUDWXUH ULVHV VOHHS RIIVHW JHQHUDOO\ RFFXUV DQG DV WHPSHUDWXUH GHFUHDVHV VOHHS RQVHW JHQHUDOO\ RFFXUV 7KH SUREDELOLW\ IRU VOHHS RQVHW DQG VOHHS RIIVHW RYHU D KRXU SHULRG LQ D QRUPDO GLXUQDO F\FOH IROORZV D SDWWHUQ f VOHHS RQVHW SUREDELOLW\ KLJK VOHHS RIIVHW SUREDELOLW\ ORZ f VOHHS RQVHW SUREDELOLW\ ORZ VOHHS RIIVHW SUREDELOLW\ KLJK f VOHHS RQVHW SUREDELOLW\ KLJK VOHHS RIIVHW SUREDELOLW\ KLJK f VOHHS RQVHW SUREDELOLW\ ORZ VHHS RIIVHW SUREDELOLW\ ORZ 7KLV SDWWHUQ RI VOHHS RQVHWRIIVHW SURSHQVLW\ VHHPV WR LQGLFDWH WKDW WKHUH DUH WZR SHULRGV GXULQJ WKH F\FOH ZKHUH VOHHS RQVHW LV GHVLUDEOH $OWKRXJK QRFWXUQDO VOHHS HSLVRGHV DSSHDU LQ SKDVH ZLWK WKH FLUFDGLDQ QDGLU LQ WKH HQGRJHQRXV FRUH WHPSHUDWXUH F\FOH QDSV RFFXU DFURVV WKH W\SLFDOO\ EURDG SHDN YDULDEOH DFURQKDVHf RI WKH WHPSHUDWXUH F\FOH VXJJHVWLQJ WKH SRVVLELOLW\ RI D VHFRQGDU\ VOHHS SURSHQVLW\ DSSUR[LPDWHO\ GHJUHHV RXW RI SKDVH ZLWK WKH FLUFDGLDQ F\FOH LQ ERG\ WHPSHUDWXUH %URXJKWRQ 'LQJHVHWDO f 'LQJHV Sf

PAGE 33

6OHHS UHVHDUFK LQGLFDWHV WKDW ORQJ VOHHS HSLVRGHV RFFXU ZKHQ VXEMHFWV EHJLQ VOHHS MXVW DIWHU WKH DFURSKDVH RI WKH WHPSHUDWXUH F\FOH LQ WKH DIWHUQRRQ ,I D VXEMHFW EHJLQV VOHHS DW WKLV WLPH WKHUH LV D SURSHQVLW\ WR VOHHS XQWLO WKH WHPSHUDWXUH F\FOH LV RQ WKH ULVH DJDLQ WKH QH[W PRUQLQJ ,Q D QRUPDO GLXUQDO F\FOH ZLWK WZR VOHHS HSLVRGHV UHVHDUFK LQGLFDWHV WKDW fORQJHU VOHHS EHJDQ MXVW SULRU WR WKH WHPSHUDWXUH PLQLPXP ZKHUHDV VKRUWHU VOHHS QDSVf RFFXUUHG QHDU WKH WHPSHUDWXUH PD[LPXPf 'LQJHV Sf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f

PAGE 34

7KH FLUFDGLDQ V\VWHP LV YHU\ FRPSOH[ DQG LW LQWHUDFWV ZLWK WKH SK\VLRORJ\ DQG SV\FKRORJ\ RI KXPDQV DQG RWKHU VSHFLHV $FFRUGLQJ WR 0LVWOEHUJHU DQG 5XVDN f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f &LUFDGLDQ 'YVUKYWKPLD &LUFDGLDQ G\VUK\WKPLD LV D FRQGLWLRQ ZKLFK GHVFULEHV WKH LQWHUQDO GLVDVVRFLDWLRQ RI WKH ELRORJLFDO DQGRU SV\FKRORJLFDO UK\WKPV RI DQ LQGLYLGXDO +DZNLQV f $V VWDWHG HDUOLHU WKLV FRQGLWLRQ LV SULPDULO\ WKH UHVXOW RI WUDQVPHULGLDQ IOLJKW :LQJHW HW DO f RU URWDWLQJ VKLIW ZRUN )RONDUG t 0RQN f 6HYHUDO UHVHDUFKHUV KDYH VKRZQ WKDW PDQ\ FLUFDGLDQ YDULDEOHV DUH DIIHFWHG E\ DFXWH VKLIWV LQ URXWLQH $NHUVWHGW t *LOOEHUJ *DQGHU HW DO +LJJLQV HW DO 0RQN HW DO 6DLWR HW DO f 0RVW RI WKHVH VWXGLHV LQGLFDWH WKDW RQFH FLUFDGLDQ

PAGE 35

UK\WKPV DUH GHV\QFKURQL]HG DW OHDVW GD\V WR VHYHUDO GD\V DUH UHTXLUHG IRU UHV\QFKURQL]DWLRQ WR RFFXU 8QWLO UHV\QFKURQL]DWLRQ GRHV RFFXU DQ LQGLYLGXDO ZLOO H[SHULHQFH ELRORJLFDO DQG SV\FKRORJLFDO GHFUHPHQWV ZKLFK DUH UHIOHFWHG LQ SV\FKRPRWRU DQG FRJQLWLYH SHUIRUPDQFH 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f +LJJLQV HW DO f FRQGXFWHG D VWXG\ LQYROYLQJ PDOH SDUWLFLSDQWV DJHV WR f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

PAGE 36

3HUIRUPDQFH GDWD EDVHG RQ WKH &LYLO $HURPHGLFDO ,QVWLWXWH 0XOWLSOH 7DVN 3HUIRUPDQFH %DWWHU\ VXJJHVW WKDW f GLXUQDO YDULDWLRQ ZDV SUHVHQW GXULQJ WKH SUHVKLIW SHULRG f SHUIRUPDQFH GHFUHPHQWV RFFXUUHG RQ WKH GD\ RI WKH VKLIW IROORZLQJ WKH VKRUW VOHHS SHULRG f SHUIRUPDQFH UDWHG UHODWLYHO\ KLJK IRU PRVW RI WKH GD\ EXW EHFDPH SRRU WRZDUG WKH HQG RI WKH VKLIW IRU WKH ILUVW WKUHH GD\V IROORZLQJ WKH VKLIW LQ URXWLQH f SHUIRUPDQFH RQ WKH IRXUWK WKURXJK WKH VL[WK SRVW VKLIW GD\V UDWHG DW RU DERYH DYHUDJH ZLWK UHODWLYHO\ VPDOO YDULDWLRQV DPRQJ WKH ILYH WHVW VHVVLRQV SHU GD\ DQG f SHUIRUPDQFH RQ WKH VHYHQWK WKURXJK QLQWK SRVW VKLIW GD\V ZDV EHORZ DYHUDJH IRU WKH H[SHULPHQW DQG GHPRQVWUDWHG HYLGHQFH RI D UHWXUQ WR D GLXUQDO SDWWHUQ ZKLFK UHIOHFWHG WKH SRVW VKLIW VOHHSZDNH F\FOH +LJJLQV HW D/ S f $FFRUGLQJ WR +LJJLQV HW DO f WKH LPSOLFDWLRQV RI WKH ILQGLQJV FLWHG LQ WKLV VWXG\ LQFOXGH WKH IROORZLQJ f ,QGLYLGXDOV PDNLQJ D KRXU DOWHUDWLRQ LQ WKH ZDNHVOHHS F\FOH VKRXOG QRW SHUIRUP FULWLFDO WDVNV GXULQJ WKH ILUVW DZDNH SHULRG IROORZLQJ WKH FKDQJH f $IWHU WKH ILUVW IXOO VOHHS SHULRG IROORZLQJ WKH FKDQJH VXEMHFWV DSSHDUHG WR SHUIRUP ZHOO HYHQ WKRXJK WKH SK\VLRORJLFDO DQG ELRPHGLFDO SDUDPHWHUV PHDVXUHG ZHUH VWLOO DGMXVWLQJ WR WKH FKDQJH f )RU WKH ILUVW ZHHN IROORZLQJ WKH FKDQJH LQ WKH ZDNHVOHHS F\FOH LQGLYLGXDOV VKRXOG QRW ZRUN ORQJHU WKDQ KRXUV FRQWLQXRXVO\ EHFDXVH SHUIRUPDQFH GHWHULRUDWHV DIWHU WKDW WLPH $IWHU WKH FKDQJH VXEMHFWV DSSHDUHG WR IDWLJXH PRUH UDSLGO\ WRZDUG WKH HQG RI WKH DZDNH SHULRG WKDQ WKH\ GLG QRUPDOO\ 7KLV HIIHFW ZDV HYLGHQW IRU VHYHUDO GD\V DIWHU WKH FKDQJH S f

PAGE 37

0RQN HW DO f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f

PAGE 38

7KH UHVXOWV RI WKLV VWXG\ LQGLFDWHG WKDW WKH GLVUXSWLRQ LQ WKH DPSOLWXGH RI WKH FLUFDGLDQ WHPSHUDWXUH UK\WKP ZDV WKH PRVW GUDPDWLF FLUFDGLDQ HIIHFW DQG WKLV HIIHFW ODVWHG IRU VHYHUDO GD\V IROORZLQJ WKH SKDVH VKLIW )RU LQVWDQFH HYHQ GD\V DIWHU WKH SKDVH VKLIW WKH PHDQ FKDQJH LQ DPSOLWXGH RI HDFK VXEMHFWfV WHPSHUDWXUH UK\WKP ZDV b EHORZ KLV RZQ EDVHOLQH OHYHO 7KLV UHGXFWLRQ LQ DPSOLWXGH PD\ EHVSHDN D GHV\QFKURQ\ ZLWKLQ WKH FLUFDGLDQ V\VWHP 1RUPDOO\ WKH FRPSRQHQW SDUWV RI WKH FLUFDGLDQ V\VWHP DUH LQ WKH FRUUHFW SKDVH UHODWLRQVKLS WR HDFK RWKHU FRPELQLQJ WR SURGXFH UREXVW ZHOOGHILQHG FLUFDGLDQ UK\WKPV $IWHU DQ DFXWH VKLIW LQ URXWLQH KRZHYHU WKHVH FRPSRQHQW SDUWV GR QRW DGMXVW WKHLU WHPSRUDO RULHQWDWLRQ DW WKH VDPH UDWH DQG DSSURSULDWH SKDVH UHODWLRQVKLSV DUH WKHUHE\ ORVW 7KLV KDV EHHQ UHIHUUHG WR DV fLQWHUQDO GLVVRFLDWLRQf D VWDWH VKRZQ WR EH DVVRFLDWHG ZLWK LPSDLUPHQWV LQ PRRG DQG SHUIRUPDQFH HIILFLHQF\ :HYHU :HYHU f Sf $PRQJ WKH YDULDEOHV VKRZLQJ D ]LJ]DJ UHFRYHU\ IXQFWLRQ ZHUH WKH VXEMHFWLYH DOHUWQHVVVOHHSLQHVV UK\WKP DQG WKH PRWLYDWLRQ ORVV 6RPH RI WKH SRVWVKLIW GD\V LQGLFDWHG ZRUVH HIIHFWV WKDQ WKRVH RQ WKH GD\ LPPHGLDWHO\ SUHFHGLQJ WKHP S f $NHUVWHGW DQG *LOOEHUJ f LQ D VWXG\ WKDW WKH\ FRQGXFWHG ZLWK VL[ PDOH VXEMHFWV LQ WHPSRUDO LVRODWLRQ GHPRQVWUDWHG WKDW VOHHS WHUPLQDWLRQ PD\ EH FORVHO\ UHODWHG WR WKH VOHHSLQHVVDOHUWQHVV UK\WKP ,Q WKLV VWXG\ VOHHS WLPHV ZHUH GLVSODFHG WR VHYHQ GLIIHUHQW WLPHV RI WKH GD\ WKURXJK D UDWH RI RQH VOHHS FRQGLWLRQ FKDQJH SHU ZHHN 7KH OHQJWK RI HDFK VOHHS HSLVRGH YDULHG DFFRUGLQJ WR ZKHQ LW

PAGE 39

EHJDQ 7KH VKRUWHVW VOHHS HSLVRGHV RFFXUUHG ZKHQ VOHHS RQVHW EHJDQ LQ WKH PRUQLQJ DQG WKH ORQJHVW VOHHS HSLVRGHV RFFXUUHG ZKHQ VOHHS RQVHW EHJDQ LQ WKH HYHQLQJ 6OHHS WHUPLQDWLRQ WLPHV ZHUH UHODWHG WR VSHFLILF SRLQWV RQ WKH VOHHSLQHVVDOHUWQHVV UK\WKP *DQGHU HW DO f FRQGXFWHG D VWXG\ LQYROYLQJ DLU FUHZV RI 3 DLUFUDIW LQ RUGHU WR GHWHUPLQH WKH DGMXVWPHQW RI VOHHS DQG WKH FLUFDGLDQ WHPSHUDWXUH UK\WKP DIWHU IOLJKWV DFURVV QLQH WLPH ]RQHV ,Q WKLV VWXG\ UHVHDUFKHUV LQYHVWLJDWHG ERWK ZHVWZDUG IOLJKW SKDVH GHOD\ YDULDWLRQf DQG HDVWZDUG IOLJKW SKDVHDGYDQFH YDULDWLRQf 7KH VWXG\ LQYROYHG QLQH 5R\DO 1RUZHJLDQ $LU )RUFH YROXQWHHUV RSHUDWLQJ 3 DLUFUDIW GXULQJ IOLJKWV DFURVV QLQH WLPH ]RQHV 7KH YDULDEOHV PRQLWRUHG GXULQJ WKH IOLJKW LQFOXGHG HDFK SDUWLFLSDQWfV VOHHSZDNH SDWWHUQ DQG FLUFDGLDQ WHPSHUDWXUH UK\WKP (DFK SDUWLFLSDQW UHFRUGHG KLV RZQ VOHHS DQG QDS WLPHV UDWHG QLJKWO\ VOHHS TXDOLW\ DQG SHUVRQDOLW\ LQYHQWRULHV ZKLOH UHFWDO WHPSHUDWXUH KHDUW UDWH DQG ZULVW DFWLYLW\ ZHUH FRQWLQXRXVO\ PRQLWRUHG $GMXVWPHQW DIWHU WKH UHWXUQ HDVWZDUG IOLJKW ZDV VORZHU DV FRPSDUHG WR DGMXVWPHQW DIWHU WKH ZHVWZDUG IOLJKW IRU WKH UHDGMXVWPHQW RI VOHHS WLPLQJ WR ORFDO WLPH 7KH HDVWZDUG IOLJKW DOVR SURGXFHG JUHDWHU LQWHULQGLYLGXDO YDULDELOLW\ LQ WKH SDWWHUQV RI DGMXVWPHQW RI VOHHS F\FOHV DQG WHPSHUDWXUH UK\WKPV 2QH RI WKH SDUWLFLSDQWV HYHQ H[KLELWHG D KRXU SKDVHGHOD\HG

PAGE 40

UHV\QFKURQL]DWLRQ UDWKHU WKDQ D SKDVHDGYDQFHG UHV\QFKURQL]DWLRQ RI WKH WHPSHUDWXUH UK\WKP DIWHU WKH HDVWZDUG IOLJKW 7KH LQWHULQGLYLGXDO GLIIHUHQFHV LQ DGMXVWPHQW RI WKH WHPSHUDWXUH UK\WKP FRUUHODWHG ZLWK VRPH RI WKH SHUVRQDOLW\ PHDVXUHV EXW LW ZDV IRXQG WKDW ODUJHU SKDVH GHOD\V LQ WKH WHPSHUDWXUH UK\WKP DV PHDVXUHG RQ WKH ILIWK GD\ DIWHU ZHVWZDUG IOLJKWf ZHUH H[KLELWHG E\ SDUWLFLSDQWV ZKR ZHUH UDWHG H[WURYHUWV DQG VPDOOHU SKDVH GHOD\V ZHUH H[KLELWHG E\ HYHQLQJ W\SHV S f ,Q WKLV VWXG\ *DQGHU HW DO DOVR IRXQG WKDW DLO RI WKH VXEMHFWV VKRZHG FOHDU SKDVH GHOD\V LQ WKH VOHHS F\FOH IRU ERWK VOHHS RQVHW DQG VOHHS RIIVHW RQ WKH ZHVWZDUG IOLJKW RU SKDVH GHOD\f EXW WKDW WKHUH ZDV JUHDWHU LQWHUVXEMHFW YDULDWLRQ LQ WKH WLPLQJ RI VOHHS IROORZLQJ WKH HDVWZDUG IOLJKW RU SKDVH DGYDQFHf 7KH VORZHU DGDSWDWLRQ DQG DGMXVWPHQW WR HDVWZDUG IOLJKW ZDV GXH WR WKH YDULDEOH UHFRYHU\ PHWKRGV RI WKH FLUFDGLDQ UK\WKPV $IWHU HDVWZDUG IOLJKWV VRPH RI WKH FLUFDGLDQ UK\WKPV WHQG WR UHV\QFKURQL]H E\ GHOD\LQJ ZKLOH RWKHU UK\WKPV DGYDQFH ,W DSSHDUV WKDW H[WURYHUWV WHQG WR DGDSW PRUH UDSLGO\ WR WKH GHOD\ VKLIW SUREDEO\ EHFDXVH WKH\ DUH PRUH H[SRVHG WR WKH VRFLDO URXWLQH LQ WKH QHZ WLPH ]RQH S f :ULJKW HW DO f VWXGLHG PDOH VROGLHUV DJHV f IRU ILYH GD\V EHIRUH DQG ILYH GD\V DIWHU DQ HDVWZDUG GHSOR\PHQW DFURVV VL[ WLPH ]RQHV &RPPRQO\ UHSRUWHG MHW ODJ V\PSWRPV RI WLUHGQHVV

PAGE 41

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f 6DLWR HW DO f FRQGXFWHG D VWXG\ RI WKH JUDGXDO DGMXVWPHQW RI FLUFDVHSWDQFLUFDGLDQ EORRG SUHVVXUH DQG KHDUW UDWH UK\WKPV RI WZR DGXOWV \HDUROG PDOH \HDUROG IHPDOHf DQG WZR FKLOGUHQ \HDUROG ER\ PRQWKROG ER\f DIWHU DQ HDVWZDUG IOLJKW ZKLFK DGYDQFHG ORFDO WLPH E\ KRXUV &LUFDGLDQ UK\WKPV DUH GDLO\ UK\WKPV RI DERXW KRXUV ZKHUHDV FLUFDVHSWDQ UK\WKPV DUH UK\WKPV RI DERXW GD\V GXUDWLRQ

PAGE 42

7KH UHVXOWV RI WKLV VWXG\ LQGLFDWHG WKDW JUDGXDO FKDQJHV FRQWLQXHG IRU IRXU ZHHNV 7KH DGXOW PDOHfV EORRG SUHVVXUH UK\WKP GHOD\HG LQ RUGHU WR UHFRYHU ZKLOH WKH DGXOW IHPDOHfV UK\WKP DGYDQFHG $SSUR[LPDWHO\ IRXU ZHHNV ZHUH QHHGHG IRU DGMXVWPHQW RI FLUFDVHSWDQ UK\WKPV LQ WKH WZR DGXOWV EXW LW RQO\ WRRN WZR ZHHNV LQ WKH ER\V 7KH DGMXVWPHQW RI UK\WKPV GXULQJ FKLOGKRRG DSSHDUV WR EH IDVWHU QRW RQO\ ZLWK FLUFDVHSWDQV EXW DOVR ZLWK FLUFDGLDQ UK\WKPV 7KH UHVXOWV DOVR LQGLFDWHG WKDW WKH KHDUW UDWH UK\WKPV DGMXVWHG PRUH UDSLGO\ WKDQ WKH EORRG SUHVVXUH UK\WKPV DQG WKDW WKH ER\Vf FLUFDGLDQ UK\WKPV DGMXVWHG IDVWHU WKDQ WKH DGXOWVf FLUFDGLDQ UK\WKPV 7KH ROGHU ER\fV UK\WKPV DGMXVWHG PRUH UDSLGO\ WKDQ WKH \RXQJHU ER\fV UK\WKPV f,Q WKH \RXQJHU ER\ WKH DGMXVWPHQW VKRZHG JUHDW ODELOLW\ ZKLFK PLJKW VXJJHVW WKDW WKH FLUFDGLDQ UK\WKPLFLW\ LV VWLOO GHYHORSLQJ DQG LPPDWXUHf S f 7KH 1DWLRQDO $HURQDXWLFV DQG 6SDFH $GPLQLVWUDWLRQ 1$6$f KDV GHYHORSHG WKH 1$6$ $PHV )DWLJXH &RXQWHUPHDVXUHV 3URJUDP WR FRPSLOH UHVHDUFK FRQFHUQLQJ IDWLJXH FDXVHG E\ VXFK IDFWRUV DQG WR GHYHORS FRXQWHUPHDVXUHV $FFRUGLQJ WR 5RVHNLQG f 0RUH WKDQ D GHFDGH RI UHVHDUFK DW 1$6$ $PHV RQ SLORW IDWLJXH VOHHS DQG FLUFDGLDQ UK\WKPV KDV LGHQWLILHG QHZ LQVLJKWV LQWR FUHZ IDWLJXH 6RPH EDVLF ILQGLQJV LQFOXGH r 6OHHS ORVV DQG FLUFDGLDQ GLVUXSWLRQ IURP ORQJKDXO IOLJKW RSHUDWLRQV FDQ UHVXOW LQ IDWLJXH LQFUHDVHG VOHHSLQHVV DQG UHGXFHG SHUIRUPDQFH

PAGE 43

r :KLOH RQ VKRUWKDXO WULSV IRU WR GD\V SLORWV WDNH ORQJHU WR IDOO DVOHHS VOHHS OHVV DZDNH HDUOLHU DQG UHSRUW OLJKWHU DQG SRRUHU VOHHS FRPSDUHG WR SUHWULS VOHHS SDWWHUQV r 3LORWV JHQHUDOO\ UHSRUW IHHOLQJ OHVV ZHOO GXULQJ H[WHQGHG GXW\ SHULRGV EXW KHOLFRSWHU SLORWV RQ VKRUWKDXO IOLJKWV IRU WR GD\V DUH IDU PRUH OLNHO\ WR UHSRUW KHDGDFKHV DQG EDFN SDLQ WKDQ DUH FRPPHUFLDO VKRUWKDXO IL[HGZLQJ SLORWV SUREDEO\ GXH WR WKH SK\VLFDO HQYLURQPHQW RI WKH KHOLFRSWHU IOLJKWGHFN r 2IIGXW\ WLPH RYHUVWDWHV WKH WLPH DYDLODEOH IRU VOHHS r 5HJXODWLRQ RI GXW\ KRXUV VKRXOG EH FRQVLGHUHG PXFK OLNH IOLJKW KRXUV r 5HVW SHULRGV VKRXOG RFFXU DW WKH VDPH WLPH RQ WULS GD\V RU SURJUHVVLYHO\ ODWHU DFURVV GD\V r 2Q OD\RYHUV H[SHULHQFHG LQWHUQDWLRQDO IOLJKW FUHZV VOHHS HIILFLHQWO\ DW VHOHFWHG WLPHV RU VOHHS OHVV HIILFLHQWO\ EXW ORQJHU WKDQ QRUPDO ZLWK D SUHIHUHQFH IRU VOHHSLQJ GXULQJ ORFDO QLJKW r +RZHYHU GHVSLWH HIILFLHQW RU ORQJHU VOHHS GXULQJ OD\RYHUV WKH FLUFDGLDQ V\VWHP LV XQDEOH WR UHV\QFKURQL]H DQG TXLFNO\ DGDSW WR UDSLG PXOWLSOH WLPH ]RQH VKLIWV r $ EULHI LQIOLJKW QDS LV DQ DFXWH LQIOLJKW VDIHW\ YDOYH WR LPSURYH SHUIRUPDQFH DQG DOHUWQHVV RQ ORQJKDXO IOLJKWV EXW QDSV GR QRW DIIHFW WKH FXPXODWLYH VOHHS GHEW LQ PRVW FUHZPHPEHUV S f $JH KDV DOVR EHHQ VKRZQ WR KDYH DQ HIIHFW RQ LQGLYLGXDOVf FLUFDGLDQ UK\WKPV 5HVHDUFK E\ .HUDQ DQG 'XFKRQ DV FLWHG LQ 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW S f KDV GHPRQVWUDWHG WKDW LQGLYLGXDOV RYHU WKH DJH RI PD\ EHJLQ WR GHPRQVWUDWH WKH HIIHFWV RI G\VUK\WKPLD DQG PD\ KDYH WURXEOH VOHHSLQJ DW QLJKW

PAGE 44

6KLIW :RUN 2YHUYLHZ 7KH WHUP fVKLIW ZRUNf KDV PDQ\ PHDQLQJV DQG YDULHV DFFRUGLQJ WR ZKLFK DXWKRU RU FRQWH[W LV XVHG DV UHIHUHQFH $FFRUGLQJ WR $NHUVWHGW f 6KLIW ZRUN XVXDOO\ UHIHUV WR DQ DUUDQJHPHQW RI ZRUN KRXUV WKDW XVHV WZR RU PRUH WHDPV VKLIWVf WR FRYHU WKH WLPH QHHGHG IRU SURGXFWLRQ :KHUHDV WZRVKLIW ZRUN XVXDOO\ FRYHUV RQO\ WKH GD\OLJKW KRXUV WKUHHVKLIW ZRUN DOVR FRYHUV WKH QLJKW 6KLIWV DUH RIWHQ FKDQJHG DW K K DQG LI D QLJKW VKLIW LV LQFOXGHGf K DOWKRXJK PDQ\ FRPSDQLHV HPSOR\ HDUOLHU RU ODWHU WLPHV ,Q (XURSH WKH WHDPV XVXDOO\ URWDWH EHWZHHQ WKH VKLIWV ZKHUHDV LQ WKH 8QLWHG 6WDWHV DVVLJQPHQW WR D FHUWDLQ VKLIW LV RIWHQ SHUPDQHQW DW OHDVW IRU D FRQVLGHUDEOH WLPH EHIRUH VHQLRULW\ DOORZV WUDQVIHUV WR DQRWKHU VKLIW 3HUPDQHQW QLJKW ZRUN WKH ZDWFK V\VWHP DW VHD DQG URVWHU ZRUN DUH RWKHU YDULHWLHV RI ZRUN KRXU DUUDQJHPHQWV 7KH ODWWHU URVWHU ZRUNf LV VLPLODU WR FRQYHQWLRQDO VKLIW ZRUN EXW LV VRPHZKDW PRUH LUUHJXODU DQG fFXVWRPL]HGf WR SDUWLFXODU QHHGV XVXDOO\ LQ WKH VHUYLFH VHFWRU )RU ODFN RI EHWWHU WHUPLQRORJ\ ZH ZLOO UHIHU WR DOO WKHVH W\SHV RI ZRUN KRXU V\VWHPV DV VKLIW ZRUN S f ,QGLYLGXDOV ZKR ZRUN URWDWLQJ VKLIW ZRUN LQ PRVW FDVHV DUH DOPRVW FRQVWDQWO\ VXEMHFWHG WR QHJDWLYH SK\VLRORJLFDO DQG SV\FKRORJLFDO HIIHFWV WKDW DUH VLPLODU WR WKRVH H[SHULHQFHG DV D UHVXOW RI MHW ODJ 7KH VHYHULW\ RI WKH FRQGLWLRQ GHSHQGV XSRQ WKH W\SH RI URWDWLQJ VFKHGXOH WKDW WKH LQGLYLGXDO ZRUNV DQG WKH QXPEHU RI KRXUV ZRUNHG GXULQJ HDFK VKLIW 6FKHGXOHV ZKLFK URWDWH RQ D ZHHNO\ EDVLV FDQ FDXVH ELRORJLFDO UK\WKPV WR GHV\QFKURQL]H DV WKH\ WU\ WR DGMXVW WR HDFK QHZ URXWLQH

PAGE 45

HDFK ZHHN 2Q WKH RWKHU KDQG VFKHGXOHV ZKLFK URWDWH UDSLGO\ FRYHULQJ DOO VKLIWV HDFK ZHHN DOORZ ELRORJLFDO UK\WKPV WR UHPDLQ GLXUQDO EXW FDQ FDXVH GHV\QFKURQL]DWLRQ RI WKH SV\FKRORJLFDOEHKDYLRUDO UK\WKPV DV WKH KDELWV DQG GDLO\ URXWLQH RI WKH LQGLYLGXDO DGMXVW WR DFFRPPRGDWH HDFK QHZ ZRUN SHULRG GXULQJ WKH ZHHN )RONDUG HW D/ 3DOH\ t 7HSDV 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f 5HVHDUFKHUV KDYH LQGLFDWHG WKDW VOHHSLQHVV DQG IDWLJXH ZKLFK UHVXOW IURP VKLIW ZRUN DUH GHWULPHQWDO WR VDIHW\ LQ PDQ\ RFFXSDWLRQV $NHUVWHGW f DQG WKDW LQGLYLGXDOV ZKR ZRUN VKLIW ZRUN QHYHU DGMXVW WR LW HYHQ DIWHU VHYHUDO \HDUV $NHUVWHGW t .HFNOXQG f :RUN 6FKHGXOHV 6KLIW ZRUN WDNHV D YDULHW\ RI FRQILJXUDWLRQV LQ LQGXVWU\ JRYHUQPHQW DQG WKH PLOLWDU\ DQG LW DOVR YDULHV DFFRUGLQJ WR WKH WUDGLWLRQ RI WKH FRXQWU\ RU RFFXSDWLRQ LQ ZKLFK LW DSSHDUV f,W KDV EHHQ HVWLPDWHG KRZHYHU WKDW DSSUR[LPDWHO\ RQHIRXUWK RI WKH ZRUNLQJ SRSXODWLRQ LQ LQGXVWULDOL]HG FRXQWULHV LV HPSOR\HG RQ VRPH NLQG RI VKLIW ZRUN V\VWHPf $NHUVWHGW S f 7KH GLIIHUHQW W\SHV RI URWDWLQJ VFKHGXOHV LQ VKLIW ZRUN HDFK KDYH WKHLU RZQ DGYDQWDJHV DQG GLVDGYDQWDJHV $V SUHYLRXVO\ VWDWHG )RONDUG HW DO f DQG 0RQN f LQGLFDWH WKDW UDSLGO\URWDWLQJ VKLIW V\VWHPV DOORZ WKH HQGRJHQRXV FLUFDGLDQ V\VWHP WR UHPDLQ

PAGE 46

GLXUQDO ZKHUHDV ZHHNO\URWDWLQJ V\VWHPV GHV\QFKURQL]H WKHVH SK\VLRORJLFDO FLUFDGLDQ UK\WKPV 7KH UDSLGO\URWDWLQJ V\VWHPV DUH DOVR DGYDQWDJHRXV LQ WKDW WKH\ GR QRW DOORZ D VOHHS GHEW WR DFFXPXODWH EHFDXVH WKH QXPEHU RI GD\ VOHHSV LV OHVV FRQWLQXRXV WKDQ ZLWK WKH ZHHNO\URWDWLQJ V\VWHPV +RZHYHU GHFUHPHQWV LQ DOHUWQHVV DQG LQFUHDVHG IDWLJXH DUH UHSRUWHG E\ ZRUNHUV XQGHU ERWK V\VWHPV LQGLFDWLQJ WKDW ZRUNHUVf SV\FKRORJLFDO UK\WKPV DUH QHJDWLYHO\ HIIHFWHG SUREDEO\ GXH WR WKH FKDQJHV LQ URXWLQH H[SHULHQFHG XQGHU ERWK V\VWHPV 0DQ\ UHVHDUFKHUV RQFH WKRXJKW WKDW SHUPDQHQW QLJKW ZRUN PLJKW EH PRUH FRQGXFLYH WR D PRUH HIILFLHQW DGMXVWPHQW WR VOHHSLQHVV %XW )RONDUG HW DO f IRXQG WKDW WKH UDWHG DOHUWQHVV RI SHUPDQHQW QLJKW QXUVHV UHPDLQHG GLXUQDO RU GD\RULHQWHG H[FHSW IRU D PLQRU EOXQWLQJ RI WKH VWHHS IDOO RI WKH DOHUWQHVV FXUYH GXULQJ WKH QLJKW VKLIW 2WKHU UHVHDUFKHUV HJ 'DKOJUHQ 3DWNDL HW DO f LQGLFDWH WKDW VOHHSLQHVV LQ SHUPDQHQW QLJKW ZRUN DSSHDUV WR EH SKDVH GHOD\HG DQG H[KLELWV D SHDN WRZDUGV WKH HQG RI WKH VKLIW 0HOWRQ DQG %DUODQRZLF] f GHVFULEHG WKH GLIIHUHQW PRGHOV RI VKLIW ZRUN VXFK DV WKH SKDVHDGYDQFHG HYHQLQJV GD\V PLGQLJKWf VFKHGXOH DQG LWV RSSRVLQJ SKDVHGHOD\HG PRGHO LQ RUGHU WR H[SODLQ ZK\ FHUWDLQ VKLIW ZRUN PRGHOV DUH GHVLUHG E\ ZRUNHUV 7KH\ LQGLFDWHG WKDW WKH URWDWLRQ LV SUHIHUUHG E\

PAGE 47

PDQ\ FRQWUROOHUV EHFDXVH LW DOORZV IRU ORQJHU LQWHUYDOV EHWZHHQ ZRUN ZHHNV $ERXW DQ KRXU EUHDN RU b RI WKH GD\ ZHHN LV UHDOL]HG EHWZHHQ ZRUN ZHHNV ZLWK WKLV VKLIW URWDWLRQ 7KLV LV DFFRPSOLVKHG E\ FRPSUHVVLQJ KRXUV RI ZRUN LQWR DQ KRXU SHULRG ZLWK VKRUW LQWHUYDOV EHWZHHQ HDFK VKLIW FKDQJH TXLFNWXUQ DURXQGVf 4XLFNWXUQDURXQGV XVXDOO\ WDNH WKH IRUP RI DQ KRXU EUHDN EHWZHHQ VKLIWV $ SKDVHGHOD\HG VFKHGXOH H[SDQGV WKH ZRUN ZHHN ZLWK ORQJHU SHULRGV RI WR KRXUV EHWZHHQ VKLIWV VORZWXUQDURXQGVf 7KHVH VFKHGXOHV VWDUW RII ZLWK WKH PLGQLJKW VKLIW DQG SURJUHVV WKURXJK ODWHU VKLIWV RYHU WKH ZHHN HQGLQJ RQ WKH HYHQLQJ VKLIW 7KH RIIGXW\ SHULRG DW WKH HQG RI WKH ZHHN LV XVXDOO\ DERXW KRXUV GXUDWLRQ ,Q WKH FDVH RI HLWKHU WKH RU WKH URWDWLRQ KRZHYHU DOO VKLIW VFKHGXOHV UHTXLUH KRXUV RI ZRUN LQ D GD\ SHULRG ZLWK KRXUV ZRUN SHU VKLIW ,W LV QRWHZRUWK\ WKDW WKH H[SDQGHG URWDWLRQ LV QRW NQRZQ WR EH LQ XVH LQ DQ\ LQGXVWULDO VHWWLQJ ,W VHHPV WKDW ZRUNHUV ZLOO DFFHSW DOPRVW DQ\ ZRUN VFKHGXOH WKDW ZLOO SURYLGH D ORQJ EUHDN DW WKH HQG RI WKH ZRUN ZHHN 6RPH QXFOHDU SRZHU SODQWV RSHUDWH RQ DQG KRXU VKLIWV 0DQ\ SROLFH RIILFHUV ZRUN KRXU VKLIWV IRU GD\V DQG WKHQ KDYH GD\V RII 2QH RI WKH SHUTXLVLWHV SHUNVf RI VHQLRULW\ LQ VKLIW ZRUN LV FKRLFH RI ZRUN

PAGE 48

VFKHGXOH RU WLPH RII :RUNHUV PRUH RIWHQ EDUJDLQ IRU WLPH RII UDWKHU WKDQ IRU PRQHWDU\ FRPSHQVDWLRQ Sf 0HOWRQ DQG %DUWDQRZLF] f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f $FFRUGLQJ WR $NHUVWHGW f VFKHGXOHV ZKLFK URWDWH VKRXOG GR VR ZLWK FRQVLGHUDWLRQ EHLQJ JLYHQ WR WKH GLUHFWLRQ RI URWDWLRQ LI WKH\ DUH WR EH EHQHILFLDO WR WKH KHDOWK DQG SHUIRUPDQFH RI WKH ZRUNHU $QRWKHU LPSRUWDQW DVSHFW RI WKH VKLIW VFKHGXOH PD\ EH LWV GLUHFWLRQ RI URWDWLRQ 6LQFH WKH IUHHUXQQLQJ SHULRG RI WKH KXPDQ VOHHSZDNH F\FOH DYHUDJHV K DQG VLQFH LW FDQ EH HQWUDLQHG E\ HQYLURQPHQWDO WLPH FXHV RQO\ ZLWKLQ K RI WKH IUHHUXQQLQJ SHULRG SKDVH GHOD\V DUH HDVLHU WR DFFRPSOLVK WKDQ SKDVH DGYDQFHV &]HLVOHU HW DO f )RU WKH URWDWLQJ VKLIW ZRUNHU WKLV LPSOLHV WKDW VFKHGXOHV WKDW GHOD\LH URWDWH FORFNZLVH PRUQLQJDIWHUQRRQQLJKWf VKRXOG EH SUHIHUUHG WR WKRVH WKDW URWDWH FRXQWHUFORFNZLVH +RZHYHU WKHUH KDYH EHHQ

PAGE 49

YHU\ IHZ SUDFWLFDO WHVWV RI WKLV WKHRU\ SDUWLFXODUO\ LQ UHODWLRQ WR VOHHSLQHVV 6WLOO &]HLVOHU HW DO f KDYH GHPRQVWUDWHG WKDW D FKDQJH IURP FRXQWHUFORFNZLVH WR FORFNZLVH URWDWLRQ WRJHWKHU ZLWK D FKDQJH IURP GD\ WR GD\ URWDWLRQ LPSURYHG SURGXFWLYLW\ DQG ZHOOEHLQJ LQ WKUHHVKLIW ZRUNHUV 2UWK*RPHU f IRXQG WKDW D FKDQJH LQ WKH VDPH GLUHFWLRQ LQ SROLFH RIILFHUV RQ UDSLG GD\f URWDWLRQ UHGXFHG EORRG SUHVVXUH DQG LPSURYHG ZHOOEHLQJ S f 0F$GDUDJK f FRQGXFWHG D VWXG\ RI DLU WUDIILF FRQWUROOHUVf $7&fVf ZRUN VFKHGXOHV IURP UDQGRPO\ VHOHFWHG $7& IDFLOLWLHV LQ RUGHU WR GHWHUPLQH ZKDW W\SHV RI VFKHGXOHV FRQWUROOHUV WHQG WR ZRUN ,W ZDV GHWHUPLQHG WKDW b RI WKH FRQWUROOHUV ZHUH ZRUNLQJ HLWKHU SKDVHDGYDQFH FRPSUHVVHG VFKHGXOHV RI WKH YDULHW\ RU DQ DOWHUQDWH ZRUN VFKHGXOH ZKLFK HYHQ IXUWKHU FRPSUHVVHV WKH ZRUN ZHHN LQWR IRXU KRXU GD\V 2QO\ b RI WKH FRQWUROOHUV ZHUH ZRUNLQJ VWUDLJKW GD\V RU VWUDLJKW HYHQLQJV ZKHUHDV WKH UHPDLQLQJ b ZHUH ZRUNLQJ D YDULHW\ RI QRQHVWDQGDUG URWDWLQJ VFKHGXOHV 1RQH RI WKH FRQWUROOHUV ZHUH ZRUNLQJ WKH SKDVH GHOD\HG VFKHGXOH RU DQ\WKLQJ OLNH WKH VFKHGXOH SURSRVHG E\ &]HLVOHU ZKLFK ZDV UHSRUWHG WR LPSURYH WKH SURGXFWLRQ DQG ZHOOEHLQJ RI URWDWLQJ VKLIW ZRUNHUV 6OHHSLQHVV $FFRUGLQJ WR WKH 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f VKLIW ZRUN FDQ FDXVH D YDULHW\ RI QHJDWLYH SV\FKRORJLFDO DQG SK\VLRORJLFDO HIIHFWV 7KLV LV DWWULEXWHG WR WKH IDFW WKDW WKH VKLIW

PAGE 50

ZRUNHU PXVW ILJKW ERWK WKH QDWXUDO GLXUQDO WUHQG RI SK\VLRORJLFDO FLUFDGLDQ UK\WKPV DQG WKH GRPLQDQW VRFLRFXOWXUDO DWWLWXGHV %HFDXVH RI WKLV WKHUH DUH WKUHH VRXUFHV RI VWUHVV IRU WKH VKLIW ZRUNHU f GLVUXSWLRQ RI FLUFDGLDQ UK\WKPV f GLVUXSWLRQ RI VOHHS DQG IDWLJXH DQG f VRFLDO GRPHVWLF GLVWXUEDQFHV S f 7KH 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW DOVR LQGLFDWHV WKDW GLVUXSWLRQV WR FLUFDGLDQ UK\WKPV WKDW LQWHUDFW ZLWK ORVV RI VOHHS DQG IDWLJXH FDQ DOVR DIIHFW KHDOWK DQG SHUIRUPDQFH f6OHHSLQHVV LV WKH LQFOLQDWLRQ WR VOHHS ZKHUHDV IDWLJXH LV ZHDULQHVV GXH WR SK\VLFDO DQG PHQWDO H[HUWLRQ ,W LV SRVVLEOH WR H[SHULHQFH RQH ZLWKRXW WKH RWKHU EXW ERWK HLWKHU DORQH RU LQ FRQFHUW FDQ KDYH GHOHWHULRXV HIIHFWVf S f 6OHHSLQHVV RIWHQ UHIHUUHG WR DV IDWLJXHf KDV EHHQ PHDVXUHG WKURXJK D YDULHW\ RI PHWKRGV 7UDGLWLRQDOO\ VOHHSLQHVV ZDV PHDVXUHG YLD VLPSOH FDWHJRU\ UDWLQJ VFDOHV RU RWKHU VLPSOH UDWLQJ VFDOHV /DWHU PHWKRGV XVLQJ WKH HOHFWURHQFHSKDORJUDSK ((*f DQG WKH HOHFWURRFXORJUDSK (2*f DWWHPSWHG WR DFKLHYH PRUH REMHFWLYH PHDVXUHV $FFRUGLQJ WR $NHUVWHGW f WKH ILUVW DWWHPSW WR TXDQWLI\ REMHFWLYH SK\VLRORJLFDO PHDVXUHV RI VOHHSLQHVV ZDV WKH 0XOWLSOH 6OHHS /DWHQF\ 7HVW 06/7f 7KLV WHVW H[SUHVVHV VOHHSLQHVV DV WKH WLPH LW WDNHV IRU WKH ((*(2* VLJQV RI VOHHS WR DSSHDU LQ D VXEMHFW ZLWK H\HV

PAGE 51

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f DFWLYLW\ DUH FORVHO\ FRUUHODWHG ZLWK ERWK VXEMHFWLYH DQG EHKDYLRUDO VOHHSLQHVV 'DQLHO 2f+DQORQ t %HDWW\ 7RUVYDOO t $NHUVWHGW $NHUVWHGW HW DO /HFUHW t 3RWWLHU 2f+DQORQ t .HOOH\ 7RUVYDOO t $NHUVWHGW f ,Q D UHFRUG VWXG\ 7RUVYDOO t $NHUVWHGW LQ SUHVVf ZH WULHG WR H[SHULPHQWDOO\ TXDQWLI\ WKH ((*(2* FKDQJHV WKDW FKDUDFWHUL]H VHYHUH VOHHSLQHVV LH WKH OHYHO RI VOHHSLQHVV DW ZKLFK GURZVLQHVV SUHYHQWV LQWHUDFWLRQ ZLWK WKH HQYLURQPHQW DQG WKH LQGLYLGXDO GR]HV RII :H IRXQG DIWHU VSHFWUDO DQDO\VLV WKDW LQFUHDVHV RI a DQG b LQ WKH SRZHU GHQVLW\ RI DOSKD DQG WKHWD DFWLYLW\ ZHUH FULWLFDO FRPSDUHG ZLWK WKH YDOXHV DW QRUPDO DOHUWQHVVf )XUWKHUPRUH ILHOG VWXGLHV KDYH GHPRQVWUDWHG WKDW ((*(2* EDVHG PRQLWRULQJ RI VOHHSLQHVV LV DOVR IHDVLEOH LQ FRPSOHWHO\ DPEXODWRU\ VXEMHFWV XQGHU WKHLU QRUPDO ZRUNOHLVXUHVOHHS FRQGLWLRQV 7RUVYDOO t $NHUVWHGW 7RUVYDOO HW DO XQSXEOLVKHG REVHUYDWLRQVf $NHUVWHGW S f 0RVW VWXGLHV FRQFHUQLQJ VOHHSLQHVV LQ VKLIW ZRUN GHULYH DV LQGLFDWHG HDUOLHU IURP VXEMHFWLYH TXHVWLRQQDLUH VWXGLHV 6HYHUDO RI WKHVH VWXGLHV LQFOXGLQJ :\DWW t 0DUULRW 7KLLV(YHQVHQ 0RWW HW DO 0HQ]HO $QGHUVHQ DQG $NHUVWHGW t 7RUVYDOO DV FLWHG LQ $NHUVWHGW S f LQGLFDWH WKDW VKLIW

PAGE 52

ZRUNHUV RQ WKH ZKROH UHSRUW JUHDWHU IDWLJXH WKDQ GR GD\ ZRUNHUV 7KLV IDWLJXH LV XVXDOO\ JUHDWHVW RQ WKH QLJKW VKLIW LQWHUPHGLDWH RQ WKH PRUQLQJ VKLIW DQG KDUGO\ DSSHDUV DW DOO RQ WKH HYHQLQJ VKLIW $ UHFRUG VWXG\ E\ 7RUVYDOO DQG $NHUVWHGW DV FLWHG LQ $NHUVWHGW f XVHG ERWK VXEMHFWLYH VHOIUHSRUW PHDVXUHV RI VOHHSLQHVV DQG REMHFWLYH ((*(2* UDWLQJV 7KH UHVXOWV RI WKLV VWXG\ LQGLFDWHV WKDW LQGLYLGXDOV SHUFHLYH VOHHSLQHVV FRPLQJ EHIRUH LW EHFRPHV PDQLIHVWHG RQ REMHFWLYH UDWLQJV f,W WKHUHIRUH DSSHDUV WKDW VOHHSLQHVV LV SHUFHLYHG E\ WKH LQGLYLGXDO ZHOO EHIRUH KH RU VKH LV RYHUFRPH E\ VOHHS 7KXV RQH VKRXOG EH DEOH WR XVH VXEMHFWLYH VOHHSLQHVV DV D VLJQDO RU ZDUQLQJ WKDW LQYROXQWDU\ VOHHS PLJKW HQVXHf $NHUVWHGW S f $NHUVWHGW t 7RUVYDOO f XVHG DQ H[SHULPHQWDO GHVLJQ ZKLFK VKRZHG WKDW UHSRUWHG IDWLJXH LQFUHDVHG XSRQ HQWHULQJ VKLIW ZRUN DQG GHFUHDVHG XSRQ OHDYLQJ LW $OVR $NHUVWHGW HW DO f GHWHUPLQHG WKDW VOHHSLQHVV KDV EHHQ VHYHUH HQRXJK WR KDYH UHVXOWHG LQ DFWXDO LQFLGHQWV RI LQGLYLGXDOV IDOOLQJ DVOHHS GXULQJ WKH QLJKW VKLIW 7KHVH VWXGLHV KDYH VLQFH EHHQ GXSOLFDWHG E\ RWKHU UHVHDUFKHV )RONDUG DQG &RQGRQ f GHPRQVWUDWHG D VRPHZKDW XQXVXDO PDQLIHVWDWLRQ RI QLJKW VKLIW VOHHSLQHVV UHSRUWHG E\ QLJKW VKLIW QXUVHV ZKLFK FDPH WR EH NQRZQ DV fQLJKW VKLIW SDUDO\VLVf 7KLV LV D UDUH SKHQRPHQRQ ZKLFK VHHPV WR RFFXU DV D IXQFWLRQ RI VOHHS ORVV GXH WR

PAGE 53

VKLIW ZRUNf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f OHYHO RI VOHHS GHSULYDWLRQ DQG VOHHSLQHVV 7KHVH IDFWRUV LQFOXGHG Df WKH WLPH RI QLJKW Ef WKH QXPEHU RI FRQVHFXWLYH QLJKW VKLIWV Ff WKH UHTXLUHPHQW WR ZRUN ERWK D PRUQLQJ VKLIW DQG D QLJKW VKLIW VWDUWLQJ RQ WKH VDPH GD\ DQG Gf LQGLYLGXDO GLIIHUHQFHV LQ IOH[LELOLW\ RI VOHHS KDELWV f7KHVH UHVXOWV VXJJHVW WKDW WKH LQFLGHQFH RI WKLV SDUDO\VLV PD\ LQGHHG SURYH WR EH D XVHIXO fFULWLFDO LQFLGHQWf IRU FRPSDULQJ WKH OHYHO RI VOHHS GHSULYDWLRQ DVVRFLDWHG ZLWK GLIIHUHQW VKLIW V\VWHPV RU LQGLYLGXDOVf )RONDUG t &RQGRQ S f )RONDUG DQG &RQGRQ DOVR GHVFULEHG WKH UHODWLRQVKLS EHWZHHQ WKH DOHUWQHVV F\FOH DQG WKH VOHHSZDNH F\FOH DQG KRZ SDUWLDO VOHHS GHSULYDWLRQ RFFXUV DV D UHVXOW RI VKLIW ZRUN ,W LV ZHOO HVWDEOLVKHG WKDW VXEMHFWLYHO\ UDWHG DOHUWQHVV VKRZV D PDUNHG FLUFDGLDQ DERXW Kf UK\WKP WKDW LV DW OHDVW SDUWLDOO\

PAGE 54

LQGHSHQGHQW RI WKH QRUPDO VOHHSZDNH F\FOH )RONDUG HW DO f DQG WKDW DOHUWQHVV LV ORZ LQ WKH HDUO\ KRXUV RI WKH PRUQLQJ 6LPLODUO\ WKHUH LV JRRG HYLGHQFH WKDW WKH GXUDWLRQ RI GD\ VOHHSV WDNHQ EHWZHHQ WZR VXFFHVVLYH QLJKW VKLIWV LV FRQVLGHUDEO\ VKRUWHU WKDQ WKH GXUDWLRQ RI QRUPDO QLJKW VOHHSV .QDXWK HW DO f 7KXV WKHUH ZLOO EH D FXPXODWLYH SDUWLDO VOHHS GHSULYDWLRQ RYHU VXFFHVVLYH QLJKW VKLIWV ZKLFK FRXOG DFFRXQW IRU WKH LQFUHDVHG LQFLGHQFH RI WKLV SDUDO\VLV RQ WKH VHFRQG RU VXEVHTXHQW QLJKW VKLIW )XUWKHU WKHUH LV DOVR HYLGHQFH WKDW QLJKW VOHHSV SUHFHGLQJ D PRUQLQJ VKLIW DUH VKRUWHU WKDQ QRUPDO RQHV .QDXWK HW DO f )LQDOO\ ULJLG VOHHSHUV DQG HVSHFLDOO\ WKRVH ZKR DUH DOVR HYHQLQJ W\SHV DUH OHVV OLNHO\ WR EH DEOH WR VOHHS VXFFHVVIXOO\ SULRU WR D QLJKW VKLIW )RONDUG t &RQGRQS f $V VWDWHG HDUOLHU 3LOFKHU DQG +XIIFXW f IRXQG WKURXJK D PHWDDQDO\VLV RI VOHHS GHSULYDWLRQ VWXGLHV WKDW VOHHS GHSULYDWLRQ KDV LWV JUHDWHVW HIIHFW RQ PRRG ZLWK D OHVVRU HIIHFW RQ FRJQLWLYH SHUIRUPDQFH DQG PRWRU SHUIRUPDQFH UHVSHFWLYHO\ 7KH\ DOVR GHWHUPLQHG WKDW SDUWLDO VOHHS GHSULYDWLRQ VOHHS ORVV ZKLFK RFFXUV ZKHQHYHU WKHUH LV D UHGXFWLRQ LQ WKH XVXDO DPRXQWV RI VOHHS REWDLQHG LQ D KRXU SHULRGf UDWKHU WKDQ ORQJWHUP RU VKRUWWHUP VOHHS GHSULYDWLRQ KDV D JUHDWHU QHJDWLYH HIIHFW RQ PRRG DQG FRJQLWLYH SHUIRUPDQFH 7KLV LV LPSRUWDQW WR VKLIW ZRUNHUV EHFDXVH SDUWLDO VOHHS GHSULYDWLRQ LV RQH UHVXOW RI ZRUNLQJ LUUHJXODU RU URWDWLQJ VFKHGXOHV 7KH FRQGLWLRQ ZKLFK RFFXUV GXH WR SDUWLDO VOHHS GHSULYDWLRQ DVVRFLDWHG ZLWK VKLIW ZRUN KDV EHHQ WHUPHG fVKLIWZRUN LQVRPQLDf $NHUVWHGW t .HFNOXQG S f

PAGE 55

2WKHU UHVHDUFKHUV KDYH GHWHUPLQHG WKDW LUUHJXODU VOHHS VFKHGXOHV DUH UHVSRQVLEOH IRU GD\WLPH VOHHSLQHVV $V VWDWHG HDUOLHU 0DQEHU HW DO f GHPRQVWUDWHG WKDW HYHQ ZKHQ D IXOOQLJKWfV VOHHS LV DFKLHYHG E\ ERWK JURXSV LQGLYLGXDOV ZKR VOHHS DW LUUHJXODU VFKHGXOHV UHSRUW D OHVVRU DPRXQW RI DOHUWQHVV DQG D JUHDWHU DPRXQW RI VOHHSLQHVV GXULQJ WKH GD\ WKDQ GR LQGLYLGXDOV ZKR VOHHS DW UHJXODU VOHHSZDNH VFKHGXOHV %LOOLDUG HW DO f VXSSRUW WKLV FRQFOXVLRQ ZLWK D VWXG\ LQYROYLQJ )UHQFK PLOLWDU\ GUDIWHHV 7KHVH UHVHDUFKHUV IRXQG WKDW VOHHS GLIILFXOWLHV DQG LUUHJXODU VOHHSZDNH VFKHGXOHV ZHUH PDMRU IDFWRUV FRQWULEXWLQJ WR H[FHVVLYH GD\WLPH VRPQROHQFH LQ \RXQJ PHQ EHWZHHQ WKH DJHV RI DQG \HDUV RI DJH 2QH VWXG\ FRQGXFWHG E\ WKH )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQfV )$$fVf &LYLO $HURPHGLFDO ,QVWLWXWH &$0,f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f)DWLJXH ZHDNQHVV DQG VRPQROHQFH ZHUH WKH PRVW IUHTXHQW RQGXW\

PAGE 56

FRPSODLQWV JLYHQ E\ FRQWUROOHUV RQ WKH PLG VKLIW IROORZHG E\ WKRVH VDPH FRPSODLQWV LQ GHFUHDVLQJ RUGHU RI IUHTXHQF\ RQ GD\ DQG HYHQLQJ VKLIWVf 6DOGLYDU HWDO S f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f6OHHSLQHVV RFFXUULQJ GXULQJ QLJKW ZRUN VHHPV WR EH WKH UHVXOW RI D FRPELQDWLRQ RI LQIOXHQFHV WKH FLUFDGLDQ WURXJK RI WKH DOHUWQHVV UK\WKP DQG VOHHS ORVVWLPH DZDNHf $NHUVWHGW S f ,W KDV EHHQ HVWLPDWHG WKDW PRVW LQGLYLGXDOV ZKR ZRUN VKLIW ZRUN QHYHU DGDSW WR LW $NHUVWHGW DQG .HFNOXQG f FRPSOHWHG D WZR \HDU VWXG\ RI UDSLGO\URWDWLQJ VKLIW ZRUNHUV DW D SDSHU PLOO f7KH UHVXOWV VXJJHVW WKDW VOHHS KDV D WUDLWOLNH FKDUDFWHU HYHQ ZKHQ RFFXUULQJ DW GD\WLPH DIWHU D QLJKW VKLIW 7KLV WKHUHIRUH VXJJHVWV WKDW

PAGE 57

VKLIW ZRUN LQVRPQLD LV D SHUVLVWHQW SUREOHPWKRVH DIIOLFWHG WHQG WR UHPDLQ VR IRU VRPH WLPHf S f $FFRUGLQJ WR D UHYLHZ E\ WKH 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f 6RPH UHVHDUFKHUV )RONDUG HW DOf .QDXWK HW DO f PDLQWDLQ WKDW UHJDUGOHVV RI URWDWLRQ OHQJWK FRPSOHWH UHDOLJQPHQW RI FLUFDGLDQ F\FOHV QHYHU DFWXDOO\ WDNHV SODFH LQ VKLIW ZRUN GXH WR WKH FRPSHWLQJ LQIOXHQFHV RI FXHV LQ WKH VRFLHW\ WHOOLQJ ZRUNHUV WKDW WKH\ DUH RQ DQ DEQRUPDO VFKHGXOH (YHQ LQ SHUPDQHQW QLJKW VKLIW ZRUN LI WKH ZRUNHU GRHV QRW PDLQWDLQ WKH VDPH VFKHGXOH RQ GD\V RII DV RQ ZRUNGD\V KH RU VKH ZLOO UDSLGO\ UHYHUW WR WKH QDWXUDO GLXUQDO RULHQWDWLRQ 0RQN 9DQ /RRQ f ,Q VXFK FLUFXPVWDQFHV WKH FLUFDGLDQ V\VWHP QHYHU IXOO\ V\QFKURQL]HV WR WKH QLJKW VFKHGXOH S f 3DOH\ DQG 7HSDV f FRQGXFWHG D VWXG\ LQYROYLQJ ILUHILJKWHUV ZRUNLQJ DQ KRXU URWDWLQJ ZHHNV SHU VKLIWf VFKHGXOH 7KH UHVXOWV RI WKLV VWXG\ GHPRQVWUDWHG WKDW ILUHILJKWHUV ZRUNLQJ WKLV W\SH RI VFKHGXOH ZLOO VOHHS OHVV WKDQ QRUPDO DQG ZLOO UHSRUW ORZHU SRVLWLYH PRRG VFRUHV KLJKHU QHJDWLYH PRRG VFRUHV DQG KLJKHU UDWLQJV RI VOHHSLQHVV RQ WKH QLJKW VKLIW 7KH UHVXOWV DOVR GHPRQVWUDWHG WKDW RYHU WKH WZRZHHN FRXUVH RI WKH VKLIW WKH ILUHILJKWHUV ZHUH XQDEOH WR DGDSW WR FKDQJHV LQ WKHLU VOHHS VFKHGXOH S f 5HVHDUFK RYHU WKH SDVW WZR GHFDGHV LQFOXGLQJ $NHUVWHGW t 7RUVYDOO 5XWHQIUDQ] HW DO 7HSDV HW DO 7HSDV t &DUYDOKLV :LOOLDPVRQ t 6DQGHUV DV FLWHG LQ 3DOH\ t 7HSDV f LQGLFDWHV WKDW fZRUNHUV RQ WKH DIWHUQRRQHYHQLQJ

PAGE 58

VKLIWV VOHHS ORQJHVW ZRUNHUV RQ WKH GD\ VKLIW VOHHS VOLJKWO\ OHVV DQG QLJKW VKLIW ZRUNHUV VOHHS OHDVWf 3DOH\ t 7HSDV S f ,Q DGGLWLRQ WR WKLV 3DOH\ DQG 7HSDV f DOVR LQGLFDWH WKDW WKH SDUWLDO VOHHS ORVV UHVXOWLQJ IURP ZRUNLQJ VKLIW ZRUN VKRXOG HIIHFW LQGLYLGXDOV ZLWK VRPH RI WKH VDPH V\PSWRPV WKDW KDYH EHHQ VHHQ LQ VWXGLHV FRQFHUQLQJ WRWDO VOHHS ORVV 3DUWLDO VOHHS ORVV RFFXUV ZKHQHYHU WKHUH LV D UHGXFWLRQ LQ WKH XVXDO DPRXQWV RI VOHHS REWDLQHG LQ D K SHULRG -RKQVRQ DQG 1DLWRK f VXJJHVWHG WKDW WKH IDFWRUV LQYROYHG LQ GHWHUPLQLQJ WKH HIIHFWV RI WRWDO VOHHS ORVV DUH HTXDOO\ UHOHYDQW WR SDUWLDO VOHHS ORVV 7KHUHIRUH FKDQJHV LQ PRRG VWDWH LQFUHDVHG IHHOLQJV RI IDWLJXH DQG LUULWDELOLW\ LQDELOLW\ WR FRQFHQWUDWH DQG SHULRGV RI PLVSHUFHSWLRQ DVVRFLDWHG ZLWK WRWDO VOHHS ORVV VKRXOG DOVR RFFXU IURP UHGXFWLRQV LQ VOHHS OHQJWK RU FKDQJHV LQ VOHHSZDNHIXOQHVV F\FOHV H[SHULHQFHG LQ QLJKW ZRUN S f 5HLQEHUJ HW DO f FRQGXFWHG D VWXG\ FRQFHUQLQJ VKLIW ZRUN WROHUDQFH DQG WKH LQWHUQDO GHV\QFKURQL]DWLRQ RI FLUFDGLDQ UK\WKPV ,W ZDV IRXQG WKDW DOO WKH SDUWLFLSDQWV LQ WKH VWXG\ PDLQWDLQHG VRPH GLXUQDO F\FOHV LQFOXGLQJ WHPSHUDWXUHf EXW GLVSOD\HG D GHV\QFKURQL]DWLRQ RI KHDUW UK\WKP DV ZHOO DV YDULRXV SV\FKRORJLFDO DQG EHKDYLRUDO UK\WKPV ZKLOH ZRUNLQJ D UDSLGO\URWDWLQJ VKLIW V\VWHP ,QGLYLGXDOV ZKR GHPRQVWUDWHG LQWROHUDQFH WR VKLIW ZRUN FRPSODLQHG RI GURZVLQHVV IDWLJXH DQG ODFN RI DWWHQWLRQ f6LQFH DQ LQWHUQDO GHV\QFKURQL]DWLRQ FDQ EH REVHUYHG LQ WROHUDQW VKLIW ZRUNHUV KDYLQJ QR FRPSODLQW LW LV OLNHO\ WKDW V\PSWRPV RI LQWROHUDQFH DUH

PAGE 59

UHODWHG WR WKH VXEMHFWVf VHQVLWLYLW\ WR LQWHUQDO GHV\QFKURQL]DWLRQ UDWKHU WKDQ WR WKH GHV\QFKURQL]DWLRQ LWVHOIf S f $ UHSRUW LQ WKH 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f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f $IWHU UHYLHZLQJ WKH OLWHUDWXUH FRQFHUQLQJ WKH HIIHFWV RI VKLIW ZRUN $NHUVWHGW f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f

PAGE 60

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f UDSLGO\URWDWLQJ VKLIW V\VWHPV DIIHFW SHUIRUPDQFH GLIIHUHQWO\ DFFRUGLQJ WR WKH PHPRU\ ORDG LQYROYHG LQ WKH WDVN UHTXLUHG RI WKH LQGLYLGXDO /RZ PHPRU\ ORDG WDVN SHUIRUPDQFH VKRZV D KLJK SRVLWLYH FRUUHODWLRQ ZLWK WKH SK\VLRORJLFDO WHPSHUDWXUH UK\WKP DQG WHQGV WR EH SRRU DW QLJKW +RZHYHU KLJK PHPRU\ ORDG WDVN SHUIRUPDQFH LV QHJDWLYHO\

PAGE 61

FRUUHODWHG ZLWK ERG\ WHPSHUDWXUH DQG LV EHVW DW QLJKW IRU UDSLGO\ URWDWLQJ VKLIW ZRUNHUV 7KLV ILQGLQJ DSSDUHQWO\ FRQWUDVWV ZLWK VRPH RI WKH HDUOLHU VWXGLHV FRQFHUQLQJ WKH YDULDWLRQ LQ SHUIRUPDQFH GXULQJ QRUPDO ZDNLQJ KRXUV ZKLFK IRXQG WKDW SHUIRUPDQFH RQ PRVW WDVNV WHQGV WR LPSURYH RYHU WKH GD\ ZKLOH VKRUWWHUP PHPRU\ WDVNV SHUIRUPDQFH KDV D WHQGHQF\ WR GHFUHDVH S f 0RUH UHFHQW UHVHDUFK KDV LQGLFDWHG WKDW WKH FRQFHSW RI D VLQJOH SHUIRUPDQFH UK\WKP LV HUURQHRXV DQG WKDW OLNH SK\VLRORJLFDO UK\WKPV SHUIRUPDQFH UK\WKPV GLIIHU QRW RQO\ LQ WKHLU QRUPDO SKDVH EXW DOVR LQ WKH GHJUHH WR ZKLFK WKH\ DUH LQIOXHQFHG E\ H[RJHQRXV IDFWRUV )RONDUG HW DO 0RQN HW DO f ,QGHHG WKHUH LV HYLGHQFH WKDW PHPRU\ORDGHG FRJQLWLYH WDVNV ZKLFK DUH EHFRPLQJ LQFUHDVLQJO\ FRPPRQ LQ LQGXVWU\ PD\ EH SHUIRUPHG SDUWLFXODUO\ ZHOO DW QLJKW SURYLGHG WKHUH LV OLWWOH DGMXVWPHQW RI WKH LQGLYLGXDOfV FLUFDGLDQ UK\WKPV )RONDUG HW DO )RONDUG t 0RQN f )XUWKHU WKH DGMXVWPHQW RI WKLV W\SH RI FLUFDGLDQ UK\WKP ZKLFK SHDNV DW QLJKW RFFXUV UHODWLYHO\ UDSLGO\ +XJKHV t )RONDUG 0RQN HW DOf 6XFK DGMXVWPHQW ZLOO UHVXOW LQ DQ LPSDLUPHQW RI QLJKWVKLIW SHUIRUPDQFH DQG VR VXJJHVWV WKDW IRU WKLV W\SH RI WDVN VKLIW V\VWHPV WKDW PLQLPL]H DGMXVWPHQW LH UDSLGO\URWDWLQJ VKLIW V\VWHPVf PD\ EH SUHIHUDEOH ,W LV WKXV QRWHZRUWK\ WKDW WKH RQO\ ILHOG RI VWXG\ WR KDYH IRXQG VXSHULRU SHUIRUPDQFH RQ WKH QLJKW VKLIW FRQFHUQHG WKH ORJJLQJ RI HUURUV RI FRPSXWHU RSHUDWRUV D WDVN ZLWK D KLJK PHPRU\ ORDGf RQ D UDSLGO\URWDWLQJ VKLIW V\VWHP 0RQNt (PEUH\ f )RONDUG HW DO S f )RONDUG SRLQWV RXW WKDW IURP D SHUIRUPDQFH SRLQW RI YLHZ UDSLGO\URWDWLQJ VKLIW V\VWHPV DOVR KDYH WKH DGYDQWDJH RI PLQLPL]LQJ WKH FXPXODWLYH VOHHS GHEW ZKLFK FDQ LWVHOI LPSDLU SHUIRUPDQFH -

PAGE 62

6OHHS GHEW FDQ DOVR LPSHGH SHUIRUPDQFH GUDPDWLFDOO\ DV HYLGHQFHG E\ WKH SKHQRPHQRQ NQRZQ DV QLJKW VKLIW SDUDO\VLV ZKLFK SUHYHQWV WKH ZRUNHU IURP SHUIRUPLQJ KLVKHU MRE IRU VHYHUDO PLQXWHV $FFRUGLQJ WR )RONDUG f LW DSSHDUV WKDW XQOHVV WKH VKLIW ZRUNHU LV HQJDJHG LQ D SDUWLFXODUO\ FUXFLDO EXW UHODWLYHO\ VLPSOH WDVN WKH DGYDQWDJHV RI SHUPDQHQW VKLIW V\VWHPV DUH RXWZHLJKHG E\ WKHLU GLVDGYDQWDJHV 5DSLGO\URWDWLQJ VKLIW V\VWHPV PLQLPL]H FXPXODWLYH VOHHS GHEW DQG DUH PRUH DGYDQWDJHRXV IRU WKH SHUIRUPDQFH RI PHPRU\ORDGHG FRJQLWLYH WDVNV ZKLFK DUH EHLQJ SHUIRUPHG PRUH DQG PRUH E\ VKLIW ZRUNHUV S f ,W KDV EHHQ REVHUYHG KRZHYHU WKDW LQGLYLGXDOV ZKR ZRUN URWDWLQJ ZRUN VFKHGXOHV UHFHLYH ORZHU MRE SHUIRUPDQFH UDWLQJV LQ VRPH FDVHV 7KH 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f UHSRUWV RQ WKH ILQGLQJV RI D VWXG\ FRQFHUQLQJ WKH SHUIRUPDQFH UDWLQJV RI QXUVHV f-RE SHUIRUPDQFH RI QXUVHV DV PHDVXUHG E\ D TXHVWLRQQDLUH ILOOHG RXW E\ VXSHUYLVRUV ZDV IRXQG WR EH ORZHU LQ WKRVH RQ D URWDWLQJ VKLIW WKDQ LQ WKRVH RQ IL[HG GD\ DIWHUQRRQ RU QLJKW VKLIWVf S f ,W KDV DOVR EHHQ GHPRQVWUDWHG WKDW H[WHQGHG ZRUN KRXUV KDYH D QHJDWLYH HIIHFW RQ SHUIRUPDQFH 5RVD DQG %RQQHW f FRQGXFWHG WKH VHFRQG RI WZR VWXGLHV ZKLFK FRQFHUQHG D FKDQJH IURP DQ KRXU ZRUN VFKHGXOH WR D KRXU FRPSUHVVHG ZRUN VFKHGXOH $IWHU WHQ

PAGE 63

PRQWKf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f 'DQLHO DQG 3RWDVRYD f LQ D VWXG\ ZKLFK FRPSDUHG KRXU DQG KRXU UDSLGO\URWDWLQJ VKLIW ZRUNHUV GHPRQVWUDWHG WKDW WKH WHPSHUDWXUH UK\WKP RI WKH KRXU VKLIW ZRUNHUV JRHV LQWR UDSLG GHFOLQH KRXUV HDUOLHU WKDQ WKH WHPSHUDWXUH UK\WKP RI WKH KRXU VKLIW ZRUNHUV 2WKHU UHVXOWV RI WKLV VWXG\ FRQFHUQHG FRPSDULVRQV RQ SHUIRUPDQFH 6WULNLQJ FKDQJHV LQ SHUIRUPDQFH ZHUH QRWHG GXULQJ WKH FLUFDGLDQ UK\WKP LQ SV\FKRPRWRU DFWLYLWLHV ZKLOH WKH SHUIRUPDQFH FXUYH GXULQJ PRUH H[DFWLQJ PHQWDO DFWLYLWLHV GHPRQVWUDWHG D PRUH XQLIRUP SDWWHUQ 6RPH RI WKH ORZHU DFKLHYHPHQWV E\ WKH KRXU ZRUNHUV FRXOG EH DWWULEXWHG WR LQFUHDVHG IDWLJXH GXULQJ WKH ORQJHU VKLIW KRZHYHU WKH\ FRXOG SRVVLEO\ EH DWWULEXWHG WR ORZHU ZRUN UHTXLUHPHQWV EHLQJ DVVLJQHG WR WKHVH ZRUNHUV S f

PAGE 64

7KHVH DQG RWKHU VWXGLHV WHQG WR GHPRQVWUDWH VRPH GLIIHUHQWLDO HIIHFWV RI VKLIW ZRUN VFKHGXOHV EXW UHJDUGOHVV RI WKH W\SH RI VFKHGXOHV LQYROYHG VKLIW ZRUN LQ JHQHUDO KDV EHHQ VKRZQ WR FDXVH VRFLDO DQG GRPHVWLF GLVUXSWLRQ IRU VKLIW ZRUNHUV ZKLFK LQ WXUQ FDQ DIIHFW WKHLU SHUIRUPDQFH DQG SRVVLEO\ KDYH FRVWO\ FRQVHTXHQFHV LQ LQGXVWULDO VHWWLQJV :RUN VFKHGXOHV PD\ LQGXFH VWUHVV E\ SUHYHQWLQJ WKH ZRUNHU IURP IXOILOOLQJ LPSRUWDQW IDPLO\ UROHV :DONHU f 6RFLDO FRPSDQLRQVKLSV SDUHQWLQJ DQG VH[XDO SDUWQHU UROHV FDQ DOO EH FRPSURPLVHG E\ ZRUN VFKHGXOHV 7KHVH HIIHFWV PD\ EH PDMRU DQG FDQ VHYHUHO\ DIIHFW PRRG PRWLYDWLRQ DQG VOHHS WKHUHIRUH KDYLQJ LQGLUHFW HIIHFWV RQ SHUIRUPDQFH DQG VDIHW\ 0DULWDO SUREOHPV H[FHVVLYH GRPHVWLF ORDG DQG FRPPXQLW\ DOLHQDWLRQ KDYH DOO EHHQ GRFXPHQWHG DV D UHVXOW RI WKH VWUDLQ SODFHG RQ ZRUNHUV E\ ZRUN VFKHGXOHV 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW S f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

PAGE 65

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f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t :DWHUKRXVH )RONDUG t 0RQN f 7KHVH DQG RWKHU FKURQRSV\FKRORJLFDO YDULDEOHV RSHUDWH LQ UK\WKPLF F\FOHV LQ DOO LQGLYLGXDOV $V $GDQ f LQGLFDWHV FKURQRSV\FKRORJLFDO SDUDPHWHUV DUH QRW WULYLDO DQG PRVW SV\FKRORJLFDO UHVHDUFK KDV QRW HYHQ DGGUHVVHG WLPHRIGD\ DV D FRQVLGHUDWLRQ 6KH DOVR H[SODLQV WKDW WKLV FRXOG KDYH D EHHQ FDXVH IRU

PAGE 66

HUURQHRXV H[SHULPHQWDO UHVXOWV LQ PDQ\ FDVHV RI SDVW SV\FKRORJLFDO UHVHDUFK 0RQN f LQGLFDWHV WKDW WKH VWXG\ RI SV\FKRORJLFDO FLUFDGLDQ UK\WKPV XVHV WKH SK\VLRORJLFDO WHPSHUDWXUH UK\WKP DV D UHIHUHQFH IRU WKUHH PDMRU UHDVRQV 7KH ILUVW LV VLPSO\ H[SHGLHQFH ERG\ WHPSHUDWXUH EHLQJ D ZHOO GHILQHG UHODWLYHO\ VWDEOH FLUFDGLDQ UK\WKP WKDW FDQ EH PHDVXUHG HDVLO\ E\ SV\FKRORJLVWV ZKR ODFN PHGLFDO TXDOLILFDWLRQV RU UDGLRLPPXQRDVVD\ FRPSHWHQFH 6HFRQG WKHUH LV D VWURQJ KLVWRULFDO WUDGLWLRQ HVSRXVHG E\ SLRQHHUV VXFK DV .OHLWPDQ .OHLWPDQ f DQG &ROTXKRXQ &ROTXKRXQ f OLQNLQJ SV\FKRORJLFDO UK\WKPV WR ERG\ WHPSHUDWXUH UK\WKPV 7KLUG WKHUH DUH WKH UHFHQW PDWKHPDWLFDO PRGHOV RI WKH FLUFDGLDQ V\VWHP %RUEHO\ .URQDXHU HW DO :HYHU f ZKLFK DOO UHO\ FUXFLDOO\ XSRQ WKH FRQFHSW RI DQ HQGRJHQRXV FLUFDGLDQ RVFLOODWRU IRU ZKLFK WKH ERG\ WHPSHUDWXUH UK\WKP LV WKH PDMRU LQGLFDWRU S f 0RQN DOVR LQGLFDWHV WKDW SV\FKRORJLFDO UK\WKPV DUH QRW DOO WKH VDPH DQG WKDW fOLNH WKH SK\VLRORJLFDO FLUFDGLDQ UK\WKPV SV\FKRORJLFDO UK\WKPV FRPSULVH JUDGXDO IOXFWXDWLRQV RYHU WKH HQWLUH K 0RUHRYHU WKHUH DUH VWURQJ LQWHUYDULDEOH GLIIHUHQFHVf S f +H DOVR VWDWHV WKDW SV\FKRORJLFDO UK\WKPV DUH QRW MXVW VSHFLILF WR HDFK LQGLYLGXDO 6WXGLHV E\ UHVHDUFKHUV VXFK DV +RUQH DQG 2VWEHUJ DQG .HUNKRI DV FLWHG LQ 0RQN f KDYH GHPRQVWUDWHG WKDW HYHQ ZLWK GLIIHUHQFHV EHWZHHQ fPRUQLQJW\SHf DQG fHYHQLQJW\SHf LQGLYLGXDOV fIRU WKH SHU FHQW RI SHRSOH ZKR DUH QHLWKHU H[WUHPH fPRUQLQJ W\SHf QRU H[WUHPH fHYHQLQJ W\SHf RQH FDQ PDNH

PAGE 67

JHQHUDOL]DWLRQV DERXW SV\FKRORJLFDO FLUFDGLDQ UK\WKPV WKDW DUH YDOLG IRU WKH SRSXODWLRQ DV D ZKROHf 0RQN S f 0RQN JRHV RQ WR VD\ WKDW WKH OLQN EHWZHHQ VXEMHFWLYH DOHUWQHVV DQG WKH FLUFDGLDQ V\VWHP PDNHV VHQVH DQG FDQ EH WKRXJKW RI LQ WHUPV RI SV\FKRORJLFDO UK\WKPV DFWLQJ DV DQ LQWHUIDFH EHWZHHQ WKH SK\VLRORJLFDO FLUFDGLDQ SURFHVVHV DQG WKH VOHHSLQJ DQG ZDNLQJ EHKDYLRUV WKDW WKRVH SURFHVVHV LQIOXHQFH S f 0RQNfV RZQ UHVHDUFK LQ IRUFHG GHV\QFKURQL]DWLRQ VWXGLHV LQ DV FLWHG LQ 0RQN f GHWHUPLQHG WKH UHODWLRQVKLS EHWZHHQ WKH WHPSHUDWXUH UK\WKP DQG WKH VXEMHFWLYH DOHUWQHVV UK\WKP 0RQN KDV GHWHUPLQHG WKDW ZKHQ WKH VOHHSZDNH F\FOH LV GULYHQ DW D SRLQW ZKLFK LV EH\RQG WKH UDQJH RI WKH HQGRJHQRXV FLUFDGLDQ RVFLOODWRU FRQWUROOLQJ WKH WHPSHUDWXUH UK\WKP WKH RVFLOODWRU DVVXPHV D SHULRG OHQJWK ZKLFK LV FORVH WR WKH IUHHUXQQLQJ KRXU F\FOH 7KLV FDXVHV WZR SHULRGLFLWLHV WR EH SUHVHQW LQ WKH GDWD WKDW RI WKH VOHHSZDNH F\FOH DQG WKDW RI WKH WHPSHUDWXUH UK\WKP ([DPLQDWLRQ RI WKH GDWD KDV GHPRQVWUDWHG WKDW ERWK SHULRGLFLWLHV LQIOXHQFH WKH VXEMHFWLYH DOHUWQHVV UK\WKP $ SDUDOOHOLVP EHWZHHQ WKH DOHUWQHVV UK\WKP DQG WKH WHPSHUDWXUH UK\WKP H[LVW DW WKH WHPSHUDWXUH UK\WKP RVFLOODWRU SHULRG DQG WKH PD[LPDO VXEMHFWLYH DOHUWQHVV ZDV IRXQG RQ WKH VOHHSZDNH F\FOH WR RFFXU IURP WR KRXUV DIWHU ZDNLQJ 7KHVH GDWD FRLQFLGH ZLWK WKH HDUO\ DIWHUQRRQ SHDN LQ DOHUWQHVV ZKLFK LV

PAGE 68

XVXDOO\ IRXQG 0RQN JRHV RQ WR VD\ WKDW WKH UHODWLRQVKLS EHWZHHQ VXEMHFWLYH DOHUWQHVV DQG ERG\ WHPSHUDWXUH FDQ EH H[SODLQHG E\ SRVWXODWLQJ FRQWURO E\ ERWK WKH WHPSHUDWXUH UK\WKP RVFLOODWRU DQG WKH VOHHSZDNH PHFKDQLVP ZLWK WKH VOHHSZDNH PHFKDQLVP WHQGLQJ WR KDYH PRUH GRPLQDQW FRQWURO XQGHU QRUPDO GLXUQDO VOHHSZDNH SDWWHUQV 0RQN S f ,Q DGGLWLRQ WR WKH VXEMHFWLYH PHDVXUHV RI DOHUWQHVV WKHUH DUH DOVR PDQ\ FLUFDGLDQ UK\WKPV ZKLFK OHQG WKHPVHOYHV WR REMHFWLYH PHDVXUHV RI SHUIRUPDQFH HIILFLHQF\ 6RPH SHUIRUPDQFH UK\WKPV UXQ SDUDOOHO WR WKH FLUFDGLDQ WHPSHUDWXUH UK\WKP DQG FDQ EH PHDVXUHG LQ WKLV PDQQHU &ROTXKRXQ &ROTXKRXQ f VWXGLHG YLJLODQFH WDVNV GHWHFWLRQ RI DQ LQIUHTXHQW VLJQDOf VLPSOH DGGLWLRQ WDVNV DGGLQJ XS WZR GLJLW QXPEHUVf DQG RWKHU VLPSOH UHDFWLRQW\SH WDVNV ILQGLQJ WKHP WR SDUDOOHO LQ PHDVXUHV RI VSHHGf WKH FLUFDGLDQ WHPSHUDWXUH UK\WKP HYHQ GXULQJ WKH SURFHVV RI DGMXVWPHQW WR D FKDQJH LQ URXWLQH 7KLV REVHUYDWLRQ FRQILUPHG HDUOLHU VWXGLHV E\ .OHLWPDQ .OHLWPDQ f ZKLFK XVHG IHZHU VXEMHFWV EXW VWLOO D UDWKHU fVLPSOH UHSHWLWLYHf FROOHFWLRQ RI WDVNV VXFK DV FDUG GHDOLQJ DQG UHDFWLRQ WLPH /DWHU D FROOHFWLRQ RI VHULDO VHDUFK WDVNV 0RQN f LQ ZKLFK VXEMHFWV VFDQ WKURXJK PDWHULDO IRU SDUWLFXODU WDUJHWV HJ HYHU\ RFFXUUHQFH RI WKH OHWWHU fHf LQ D PDJD]LQH DUWLFOHf DJDLQ UHYHDOHG D YHU\ JRRG SDUDOOHOLVP LQ VSHHGf ZLWK WKH ERG\ WHPSHUDWXUH UK\WKP 0RQN S f 0RQNfV UHVHDUFK DOVR GHWHUPLQHG WKDW SHUIRUPDQFH VSHHG SDUDOOHOV WKH WHPSHUDWXUH UK\WKP +RZHYHU ZKHQ SHUIRUPDQFH DFFXUDF\ ZDV PHDVXUHG SHUIRUPDQFH ZDV IRXQG WR GHFOLQH RYHU WKH

PAGE 69

GD\ UHDFKLQJ WKH ORZHVW SRLQW RI DFFXUDF\ DW WKH VDPH WLPH RI WKH GD\ DV WKH SHDN LQ SHUIRUPDQFH VSHHG 7KXV ZKLOH SHUIRUPDQFH VSHHG LV LQFUHDVLQJ RYHU WKH GD\ SHUIRUPDQFH DFFXUDF\ LV GHFOLQLQJ 0RQN Sf ?]ROTXKRXQ DOVR FRQGXFWHG D VWXG\ FRQFHUQLQJ VSHHG LQ DULWKPHWLF DGGLWLRQ 7KLV VWXG\ FRQFOXGHG WKDW DV WKH WDVN EHFDPH PRUH UHSHWLWLYH DQG VLPSOH WKH WLPHRIGD\ IXQFWLRQ LQ SHUIRUPDQFH VSHHG EHFDPH PRUH OLNH WKDW RI WKH WHPSHUDWXUH UK\WKP 7KHUHIRUH WLPHRIGD\ IXQFWLRQV PD\ FKDQJH ZLWK SUDFWLFH )RU 0RQN f WKHVH ILQGLQJV VXJJHVWV WKDW LW PD\ EH ZURQJ WR VSHDN RI SHUIRUPDQFH DV LPSURYLQJ RYHU WKH GD\ RU WR WKLQN RI WLPHRIGD\ IDFWRUV DV UHSUHVHQWLQJ FKDQJHV LQ WKH EUDLQfV FDSDFLW\ WR SURFHVV LQIRUPDWLRQ 0RQN EHOLHYHV WKDW ZH VKRXOG FRQVLGHU FKDQJHV LQ SHUIRUPDQFH DV EHLQJ PHGLDWHG E\ FKDQJHV LQ VWUDWHJ\ RYHU WKH GD\ ,QGLYLGXDOV SUREDEO\ DSSURDFK LQIRUPDWLRQ SURFHVVLQJ WDVNV GLIIHUHQWO\ DW RQH WLPH RI WKH GD\ DQG DW RQH OHYHO RI SUDFWLFH DV FRPSDUHG WR DQRWKHU 0HDVXUHV PD\ LQGLFDWH EHWWHU RU ZRUVH SHUIRUPDQFH GHSHQGLQJ XSRQ ZKDW W\SH RI PHDVXUH ZH WDNH S f 0RQN DOVR VXJJHVWV WKDW UHVHDUFK RQ VKRUWWHUP PHPRU\ ZKLFK LQGLFDWHV D GHFOLQH RYHU WKH GD\ PD\ DOVR EH D UHVXOW RI FKDQJHV LQ LQIRUPDWLRQ SURFHVVLQJ VWUDWHJ\ UDWKHU WKDQ FKDQJHV LQ WKH EUDLQfV FDSDFLW\ WR SURFHVV LQIRUPDWLRQ f,Q VHYHUDO VWXGLHV )RONDUG

PAGE 70

)RONDUG t 0RQN )RONDUG t 0RQN f D PDQLSXODWLRQ WKDW LQKLELWHG VXEYRFDOL]DWLRQ ZDV DOVR IRXQG WR HOLPLQDWH WKH WLPHRIGD\ HIIHFW VXJJHVWLQJ WKDW WLPHRIGD\ GLIIHUHQFHV PD\ DOVR UHVXOW IURP GLIIHUHQFHV LQ VXEYRFDOL]DWLRQ VWUDWHJ\f S f $FFRUGLQJ WR WKH 86 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW f LQ D GLXUQDO FRQGLWLRQ FHUWDLQ SV\FKRORJLFDO UK\WKPV DUH HYLGHQW WKRXJK WKH\ PD\ EH LQFRQVLVWHQW ? ,W ZDV IRXQG WKDW LI VSHHG RI LGHQWLILFDWLRQ ZDV DVVHVVHG WKH XVXDO UHODWLRQVKLS ZLWK WHPSHUDWXUH KHOGWKDW LV VXEMHFWV EHFDPH IDVWHU RYHU WKH FRXUVH RI WKH GD\ +RZHYHU LI DFFXUDF\ RI UHVSRQVH EHFDPH WKH EHQFKPDUN SHDN SHUIRUPDQFH RFFXUUHG LQ WKH PRUQLQJ &UDLJ t &RQGRQ 0RQNt /HQJ f 6KRUWWHUP DQG ORQJWHUP PHPRU\ DOVR DSSHDU WR SHDN DW GLIIHUHQW WLPHV GXULQJ WKH KRXU F\FOH )RONDUG t 0RQN f 0RWLYDWLRQ FDQ LQIOXHQFH SHUIRUPDQFH WRR ,W KDV EHHQ VKRZQ WKDW ZKHQ DQ LQFHQWLYH LV RIIHUHG VXFK DV D VLJQLILFDQW VXP RI PRQH\ FLUFDGLDQ GHFUHPHQWV LQ SHUIRUPDQFH PD\ EH RYHUFRPH WR VRPH H[WHQW &ROTXKRXQ +RUQH t 3HWWLWW f 7KH ODWWHU VWXG\ +RUQH t 3HWWLWW f LQGLFDWHG KRZHYHU WKDW VOHHS GHSULYDWLRQ DQG FLUFDGLDQ UK\WKPV FDQ RYHUFRPH HYHQ WKH VWURQJHVW LQFHQWLYH LQIOXHQFLQJ SHUIRUPDQFH )XUWKHU FRPSOLFDWLQJ WKH SLFWXUH LV WKH REVHUYDWLRQ WKDW LQGLYLGXDOV DUH QRW DOZD\V DFFXUDWH MXGJHV RI WKHLU RZQ PRRG DOHUWQHVV RU DELOLW\ S f (QJOXQG f FRQGXFWHG D VWXG\ RI WKH GLXUQDO IXQFWLRQ RI UHDGLQJ UDWH FRPSUHKHQVLRQ DQG HIILFLHQF\ 7KH UHVXOWV RI WKLV VWXG\ LQGLFDWH WKDW WKH SORWWHG PHDQV IRU WKHVH YDULDEOHV GLVSOD\HG GLVWLQFW GLXUQDO SDWWHUQV $QDO\VLV RI WKH GDWD UHYHDOHG WKDW FRPSUHKHQVLRQ ZDV VXSHULRU LQ WKH DIWHUQRRQ DQG HDUO\ PRUQLQJ ZKHQ FRPSDUHG WR

PAGE 71

VFRUHV DW PLGPRUQLQJ DQG ODWH DW QLJKW 3HUIRUPDQFH VSHHG LQGLFDWHG D PRUQLQJ ULVH DQG DIWHUQRRQ IDOO 2UDO WHPSHUDWXUH RI WKH VXEMHFWV SHDNHG DW ZLWK QR GLIIHUHQFHV EHLQJ IRXQG EHWZHHQ PDOHV DQG IHPDOHV 2YHUDOO HIILFLHQF\ GLVSOD\HG QR VLJQLILFDQW GLIIHUHQFHV LQGLFDWLQJ D WUDGHRII EHWZHHQ VSHHG DQG DFFXUDF\ RYHU WKH GD\ ZKLFK DFWV WR PDLQWDLQ SHUIRUPDQFH HIILFLHQF\ f7KHVH ILQGLQJV FRQILUP )RONDUGfV REVHUYDWLRQV %ULW 3V\FKRO f ZKLOH SHUIRUPDQFH LV WDVN GHSHQGHQW VSHHG DQG DFFXUDF\ PHDVXUHV PD\ UHSUHVHQW GLIIHUHQW GLXUQDO FRPSRQHQWV RI SHUIRUPDQFH ZLWK FRUUHVSRQGLQJO\ GLIIHUHQW UHODWLRQVKLSV WR DURXVDO OHYHOf (QJOXQG S f 7DNLQJ DOO RI WKLV LQWR FRQVLGHUDWLRQ $GDQ f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

PAGE 72

'DLO\ IOXFWXDWLRQ LQ EHKDYLRUDO SDUDPHWHUV FDQQRW EH FRQVLGHUHG WULYLDO WKH WRWDO YDULDWLRQ GHWHFWHG LV RI WKH RUGHU RI b DQG fRYHU WKH QRUPDO ZDNLQJ GD\ WR f WKH FLUFDGLDQ YDULDWLRQ LQ SHUIRUPDQFH HIILFLHQF\ FDQ EH HTXLYDOHQW LQ PDJQLWXGH WR WKH HIIHFW RI OLPLWLQJ VOHHS WR KRXUV RU LQJHVWLQJ WKH OHJDO GULYLQJ OLPLW RI DOFRKROf 0RQN HW DO S f S f $GDQ DOVR LQGLFDWHV WKDW RWKHU UK\WKPV ZLWK GLIIHUHQW SHULRGLFLWLHV FLUFDPHQVXDO UK\WKPV JUHDWHU WKDQ KRXUV DQG XOWUDGLDQ UK\WKPV OHVV WKDQ KRXUVf PD\ PDVN WKH UHVXOWV RI VWXGLHV FRQFHUQLQJ FLUFDGLDQ UK\WKPV DQG WKHUHIRUH VKRXOG EH FRQWUROOHG 1HYHUWKHOHVV WKH FODVVLILFDWLRQ SURSRVHG E\ $GDQ f HQDEOHV XV WR GLIIHUHQWLDWH IURP D FKURQRSV\FKRORJLFDO YLHZSRLQW WKH PRVW FRPPRQO\ XVHG WHVWV EDVHG XSRQ ERWK WKH UHTXLUHG VNLOO RI HDFK WHVW DQG WKH HPSLULFDO FULWHULRQ RI WKH UHVXOWDQW FLUFDGLDQ IXQFWLRQ 7KH DGYDQWDJH RI $GDQfV FODVVLILFDWLRQ RYHU WUDGLWLRQDO FODVVLILFDWLRQV LV WKDW FRKHUHQFH LV LPSRVHG RQ WKH KHWHURJHQHLW\ RI WKH H[LVWLQJ UHVXOWV IRU ERWK VXEMHFWLYH DQG REMHFWLYH WHVWV 7KH VXEMHFWLYH WHVWV \LHOG GDWD FRQFHUQLQJ WKH VXEMHFWVf RZQ DVVHVVPHQW RI KLVKHU VWDWH GXULQJ WKH WHVW 7KH WZR SULQFLSDO W\SHV RI WHVWV LQFOXGH VHOIDVVHVVPHQW LQYHQWRULHV DQG DQDORJLFRYLVXDO VFDOHV 3DWWHUQV RI DOHUWQHVV DV PHDVXUHG RQ WKH DQDORJLFRYLVXDO VFDOHV VKRZV D SHDN DOHUWQHVV EHWZHHQ DQG DQG VRPQROHQFH GHPRQVWUDWHV DQ RSSRVHG WZRSKDVH SDWWHUQ 7KH

PAGE 73

9 SDWWHUQ RI GURZVLQHVV LQGLFDWHV KLJK VFRUHV DW DQG DW 2EMHFWLYH WHVWV FRQVLVW RI WDVNV ZKLFK DUH JHQHUDOO\ QRUPDOL]HG DQG DUH GHVLJQHG WR PDNH LW GLIILFXOW IRU VXEMHFWV FKHDW 6XEMHFWV DUH SUHVHQWHG ZLWK LQGLFDWRUV IURP ZKLFK WR FKRVH DV D FULWHULRQ RI SHUIRUPDQFH DQG WKHVH GLIIHU DV WR ZKHWKHU WKH FLUFDGLDQ IXQFWLRQ EHLQJ WHVWHG LV TXDQWLWDWLYH RU TXDOLWDWLYH $GDQf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

PAGE 74

ORQJWHUP PHPRU\ 9HU\ OLWWOH GDWD H[LVW FRQFHUQLQJ WDVNV RI JUHDWHU FRJQLWLYH FRPSOH[LW\ DQG UHVXOWV DUH LQFRQFOXVLYH $GDQ S f 7KLV VHHPV WR LQGLFDWH WKDW IRU ZKDWHYHU UHDVRQ D UK\WKPLFLW\ H[LVWV )RONDUG DQG 0RQN f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f $VWKPD DWWDFNV DUH PRVW SUHYDOHQW DW QLJKW 0F)DGGHQ f KXPDQ EDELHV DUH ERUQ SUHGRPLQDQWO\ LQ WKH HDUO\ PRUQLQJ KRXUV *ODWWUH t %MHUNHGDO .DLVHU t +DOEHUJ f :KLOH WKHVH SDWWHUQV GR QRW QHFHVVDULO\ LQGLFDWH WKDW WKH HYHQWV DUH GULYHQ E\ WKH FLUFDGLDQ SDFHPDNHU WKH\ GR VXJJHVW WHPSRUDO RUGHU LQ WKH IXQFWLRQLQJ RI WKH KXPDQ ERG\ 86 &RQJUHVV 2IILFH RI 7HFKQR ORJ\ $VVHVVPHQW Sf 2QH VWXG\ ZDV FRQGXFWHG LQ RUGHU WR GHWHUPLQH WKH HIIHFW RI WKH WLPLQJ RI OHDUQLQJ UHODWLYH WR WKH WLPLQJ RI VOHHS RQ PHPRU\ UHWHQWLRQ %HQVRQ DQG )HLQ EHUJ f FRQGXFWHG D VWXG\ FRQFHUQLQJ WKH HIIHFWV RI VOHHS RQ PHPRU\ E\ FRPSDULQJ WKH PHPRU\ UHWHQWLRQ

PAGE 75

RI WZR JURXSV 7KH WZR JURXSV ZHUH GLYLGHG LQ RUGHU WR H[DPLQH UHWHQWLRQ RI LQLWLDO OHDUQLQJ LQ WKH PRUQLQJ YHUVXV UHWHQWLRQ RI LQLWLDO OHDUQLQJ DW QLJKW DW ERWK KRXUV DQG KRXUV DIWHU WKH LQLWLDO OHDUQLQJ ZDV DFFRPSOLVKHG 7KH UHVXOWV LQGLFDWHG WKDW UHWHQWLRQ ZDV VXSHULRU KRXUV DIWHU OHDUQLQJ IRU VXEMHFWV ZKR OHDUQHG DW QLJKW DV FRPSDUHG WR WKRVH ZKR OHDUQHG LQ WKH PRUQLQJ 5HWHQWLRQ IRU VXEMHFWV ZKR OHDUQHG DW QLJKW ZDV HTXDO DIWHU KRXUV WR WKDW REVHUYHG DIWHU KRXUV $ UDWKHU VXUSULVLQJ ILQGLQJ LQ WKLV VWXG\ ZDV WKDW UHWHQWLRQ VFRUHV IRU VXEMHFWV ZKR OHDUQHG LQ WKH PRUQLQJ ZHUH VXSHULRU DIWHU KRXUV WR WKRVH REVHUYHG DW KRXUV DIWHU LQLWLDO OHDUQLQJ S f $ UHFHQW VWXG\ 6KDGPHKU t +ROFRPE f LQGLFDWHV WKDW LW WDNHV DERXW VL[ KRXUV IRU WKH PHPRU\ RI D QHZ VNLOO WR PRYH LQWR ORQJWHUP PHPRU\ VWRUDJH 7KLV VWXG\ DOVR LQGLFDWHV WKDW SUDFWLFH RI D VLPLODU VNLOO MXVW DIWHU OHDUQLQJ WKH QHZ VNLOO FDQ LQWHUIHUH ZLWK WKH ORQJWHUP PHPRU\ VWRUDJH RI WKH QHZ VNLOO EXW WKDW SUDFWLFH RI D QRQVLPLODU VNLOO PD\ QRW LQWHUIHUH 7KHVH ILQGLQJV PD\ KDYH LPSOLFDWLRQV IRU WKH VXUSULVLQJ UHVXOWV REWDLQHG E\ %HQVRQ DQG )HLQ EHUJ

PAGE 76

6XPPDU\n 7KLV FKDSWHU KDV GHVFULEHG KXPDQ ELRORJLFDO DQG SV\FKRORJLFDO FLUFDGLDQ UK\WKPV DQG KRZ WKH\ DUH FRQWUROOHG E\ ERWK HQGRJHQRXV DQG H[RJHQRXV IDFWRUV 7KLV FKDSWHU DOVR GHVFULEHG WKH GHWULPHQWDO HIIHFWV RI FLUFDGLDQ G\VUK\WKPLD ZKLFK DUH FDXVHG E\ MHW ODJ RU URWDWLQJ VKLIW ZRUN DQG WKH HIIHFWV RI FLUFDGLDQ UK\WKPV RQ OHDUQLQJ 7KH IROORZLQJ FKDSWHU GHVFULEHV WKH PHWKRGV DQG SURFHGXUHV XVHG LQ WKLV VWXG\ WR FROOHFW DQG DQDO\]H WKH GDWD LQ RUGHU WR DGGUHVV WKH K\SRWKHVHV SUHVHQWHG LQ &KDSWHU

PAGE 77

&+$37(5 0(7+2'2/2*< 3DUWLFLSDQWV 7KLUW\VHYHQ SDUWLFLSDQWV 1 f ZHUH VHOIVHOHFWHG IURP WKH -DFNVRQYLOOH )ORULGD $LU 5RXWH 7UDIILF &RQWURO &HQWHU $57&&f $OWKRXJK WKLV VDPSOH LV UHSUHVHQWDWLYH RI WKH SRSXODWLRQ RI LQWHUHVW WKLV QXPEHU LV OHVV WKDQ WKH GHVLUHG QXPEHU RI SDUWLFLSDQWV IRU WKLV W\SH RI VWXG\ DQG PD\ VOLJKWO\ OLPLW WKH JHQHUDOL]DWLRQ RI WKH ILQGLQJV 3DUWLFLSDQWV ZHUH UHFUXLWHG RQ D YROXQWDU\ EDVLV VR WKDW WZR JURXSV UDSLGO\URWDWLQJ VKLIW ZRUNHUV DQG GD\HYHQLQJ VKLIW ZRUNHUVf ZHUH REWDLQHG 7KH SRSXODWLRQ IRU WKLV VWXG\ FRQVLVWHG RI VL[ IHPDOH FRQWUROOHUV WKUHH LQ HDFK JURXSf DQG WKLUW\ RQH PDOH FRQWUROOHUV DOO \HDUV RI DJH RU \RXQJHU 7KH PDMRULW\ RI WKH SDUWLFLSDQWV ZHUH LQ WKHLU ODWH V RU HDUO\ V ,QVWUXPHQWV 7KH 1RYD6FDQ FRPSXWHUEDVHG SHUIRUPDQFH WHVW ZDV WKH LQVWUXPHQW XVHG WR JDWKHU GDWD IRU WKLV VWXG\ 7KLV WHVW ZDV XVHG WR PHDVXUH ILYH YDULDEOHV Df 'LDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ WKH VSDWLDOYLVXDOL]DWLRQ WDVN DWWHQWLRQ DOORFDWLRQf Ef GLDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ WKH WUDFNLQJ WDVN DWWHQWLRQ DOORFDWLRQf

PAGE 78

Ff UHDFWLRQ WLPH IRU WUDQVLWLRQ IURP WKH WUDFNLQJ WDVN WR WKH VSDWLDO YLVXDOL]DWLRQ WDVN IOH[LELOLW\ LQ VZLWFKLQJ UHVRXUFHVf Gf QRQWUDQVLWLRQ UHDFWLRQ WR REMHFW RULHQWDWLRQ VSDWLDO YLVXDOL]DWLRQf DQG Hf WUDFNLQJ HUURU GXULQJ WKH WUDFNLQJ WDVN VWUDLJKW PLQGPRWRU FRRUGLQDWLRQf 7KH 1RYD6FDQ LQVWUXPHQW UDQGRPO\ SUHVHQWV WZR FRPSXWHU EDVHG WDVNV IRU WKH SDUWLFLSDQW WR SHUIRUP VHYHUDO WLPHV GXULQJ HDFK VHVVLRQ 7DVN VSDWLDO YLVXDOL]DWLRQf LV D FRJQLWLYH WDVN LQ ZKLFK WKH SDUWLFLSDQW PXVW RULHQW DQ DLUSODQH E\ VHOHFWLQJ WKH ZLQJ ZKLFK KDV WKH FLUFOH DERYH LW OLNH WKH RQH EHORZ WKH DLUSODQH 7KH FLUFOHV DUH HLWKHU RSHQ RU VROLG DQG RQH ZLQJ KDV RQH RI WKHVH DERYH LW ZKLOH WKH RWKHU ZLQJ KDV WKH RWKHU 7KH SDUWLFLSDQW PXVW UHVSRQG HLWKHU fOHIWf RU fULJKWf WR VHOHFW WKH FRUUHFW ZLQJ E\ SUHVVLQJ EXWWRQV RQ D VSHFLDO NH\ERDUG 7KH UHDFWLRQ WLPHV IRU WKH SDUWLFLSDQWfV UHVSRQVHV DUH UHFRUGHG LQ PLOOLVHFRQGV 7DVN WUDFNLQJf LV D SV\FKRPRWRU WDVN LQ ZKLFK WKH SDUWLFLSDQW PXVW XVH D MR\ VWLFN WR PRYH WKH FRPSXWHUfV FXUVRU OHIW DQG ULJKW WR WU\ WR NHHS LW EHORZ D KRUL]RQWDOO\PRYLQJ REMHFW 7KH WUDFNLQJ HUURU DW WKLV WDVN LV UHFRUGHG DV WKH PHDQ QXPEHU RI SL[HOV WKDW WKH FXUVRU ZDV RII WDUJHW DQG RXWVLGH WKH UDQJH RI WKH DUHD GLUHFWO\ EHORZ WKH PRYLQJ REMHFW GXULQJ WKH VHVVLRQ

PAGE 79

:KLOH WKH SDUWLFLSDQW LV FRPSOHWLQJ WKHVH WZR UDQGRPO\ SUHVHQWHG WDVNV GXULQJ D 1RYD6FDQ VHVVLRQ KHVKH PXVW FRQWLQXRXVO\ PRQLWRU D GLDO LQ WKH XSSHU ULJKW FRUQHU RI WKH VFUHHQ WR GHWHUPLQH ZKHQ WKH SRLQWHU RQ WKH GLDO PRYHV LQWR D VROLG DUHD :KHQ WKLV RFFXUV WKH SDUWLFLSDQW PXVW UHVSRQG E\ SUHVVLQJ DQRWKHU EXWWRQ RQ WKH VSHFLDO NH\ERDUG 7KH SDUWLFLSDQWfV UHVSRQVHV DUH UHFRUGHG LQ PLOOLVHFRQGV DQG UHSUHVHQW WKH SDUWLFLSDQWfV DWWHQWLRQ DOORFDWLRQ UDWLQJ GXULQJ HDFK RI WKH WZR WDVNV 7KH UHOLDELOLW\ PHDVXUHV IRU WKH 1RYD6FDQ YDULDEOHV REWDLQHG IURP SDUDPHWULF VWXGLHV KDYH EHHQ UDWHG YLD LQWUDFODVV FRUUHODWLRQ DV IROORZV 5RWDWHG REMHFW f WUDQVLWLRQ WR REMHFW YLVXDOL]DWLRQ WDVN f DQG WUDFNLQJ HUURU f 7KHVH PHDVXUHV DUH FDOFXODWHG E\ VXEWUDFWLQJ WKH ZLWKLQ YDULDQFH IURP WKH EHWZHHQ YDULDQFH DQG GLYLGLQJ E\ WKH EHWZHHQ YDULDQFH 2f'RQQHOO f 7KH DWWHQWLRQ DOORFDWLRQ YDULDEOH KDV RQO\ EHHQ XVHG LQ WKH SDVW IRU WKH SXUSRVH RI PHDVXULQJ DQ LQGLYLGXDOfV FXUUHQW DWWHQWLRQ DOORFDWLRQ DJDLQVW KLVKHU RZQ EDVHOLQH UDWLQJ ,W KDV QHYHU EHIRUH EHHQ XVHG WR FRPSDUH WKH UDWLQJV RI JURXSV RI LQGLYLGXDOV 0DWHULDOV$SSDUDWXV 1RYD6FDQ UHTXLUHV D 3& ZLWK RU IDVWHU SURFHVVRU RQH 0% RI 5$0 D 9*$ JUDSKLF DGDSWHU EODFN DQG ZKLWH RU FRORUf D KDUGGULYH GLVN LI GHVLUHGf DQG '26 RU KLJKHU RSHUDWLQJ V\VWHP $ VSHFLDO

PAGE 80

NH\ERDUG LV SURYLGHG IRU UHVSRQGLQJ 7KHVH PLQLPXP UHTXLUHPHQWV ZHUH PHW XVLQJ WZR 0+] '26 FRPSXWHUV (DFK SDUWLFLSDQW ZDV SURYLGHG ZLWK D VLQJOH LQ IORSS\ GLVN ZKLFK FRQWDLQHG WKH 1RYD6FDQ VRIWZDUH VR WKDW DOO 1RYD6FDQ VHVVLRQV FRXOG EH UHFRUGHG VHSDUDWHO\ IRU HDFK LQGLYLGXDO 'HVLJQ $ FDXVDOFRPSDUDWLYH GHVLJQ ZDV XVHG IRU WKLV VWXG\ VLQFH WKH SUHGLFWRU YDULDEOH ZRUN VFKHGXOHf LV DOUHDG\ GLIIHUHQWLDWHG EHWZHHQ WKH WZR VWXG\ JURXSV DQG EHFDXVH WKH FDXVH FLUFDGLDQ G\VUK\WKPLDf LV DOUHDG\ VXVSHFWHG DQG WKH HIIHFW DWWHQWLRQ DOORFDWLRQf LV XQGHU LQYHVWLJDWLRQ 6HOIVHOHFWHG SDUWLFLSDQWV ZHUH FDWHJRUL]HG LQWR WZR JURXSV 2QH JURXS RI SDUWLFLSDQWV ZDV UHSUHVHQWHG E\ LQGLYLGXDOV ZKR ZRUN GD\ DQG HYHQLQJ WZRVKLIW VFKHGXOHV DQG WKH RWKHU JURXS RQO\ LQFOXGHG LQGLYLGXDOV ZKR ZRUN D YDULHW\ RI UDSLGO\URWDWLQJ WKUHHVKLIW VFKHGXOHV ZKLFK LQFOXGH QLJKW VKLIWV 7KH YDULDEOHV ZKLFK ZHUH FRQWUROOHG LQFOXGH WKH IROORZLQJ FRQVLGHUDWLRQV Df 7KH URWDWLQJ VFKHGXOHV LQFOXGHG DW OHDVW RQH PLGQLJKW VKLIW DQG RQH GD\ VKLIW HDFK ZHHN SULRU WR WKH 1RYD6FDQ VHVVLRQV VR WKDW WKH GDLO\ URXWLQHV RI WKH LQGLYLGXDOV ZKR ZRUNHG WKHP ZHUH DFXWHO\ VKLIWHG Ef DOO RI WKH SDUWLFLSDQWV ZHUH QR PRUH WKDQ \HDUV RI DJH VR WKDW WKH FLUFDGLDQ HIIHFWV RI DJLQJ ZHUH QRW D IDFWRU Ff DOO WHVWLQJ SHULRGV RQ WKH 1RYD6FDQ WRRN SODFH GXULQJ WKH

PAGE 81

GD\ VKLIW KRXUV DP WR SPf IRU HDFK SDUWLFLSDQW LQ RUGHU WR GHWHUPLQH HDFK LQGLYLGXDOfV DWWHQWLRQ DOORFDWLRQ GXULQJ WKH KRXUV WKDW WUDLQLQJ QRUPDOO\ RFFXUV DQG Gf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

PAGE 82

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f $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO VKRZ QR VLJQLILFDQW GLIIHUHQFH IURP FRQWUROOHUV ZKR ZRUN GD\ VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV RQ WKHLU 1RYD6FDQ PHDQ EDVHOLQH UDWLQJ RI DWWHQWLRQ DOORFDWLRQ IRU GLDO PRQLWRULQJ GXULQJ WKH WUDFNLQJ WDVN 7DVN f

PAGE 83

$LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI UHDFWLRQ WLPH RQ GLDO PRQLWRULQJ GXULQJ WKH VSDWLDO YLVXDOL]DWLRQ WDVN 7DVN f $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI UHDFWLRQ WLPH RQ GLDO PRQLWRULQJ GXULQJ WKH WUDFNLQJ WDVN 7DVN f $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI UHDFWLRQ WLPH IRU WKH WUDQVLWLRQ IURP 7DVN WR 7DVN ZKLFK LQGLFDWHV UHVSRQVH IOH[LELOLW\ LQ VZLWFKLQJ UHVRXUFHVf $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI

PAGE 84

FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI UHDFWLRQ WLPH RQ 7DVN QRQn WUDQVLWLRQ REMHFW RULHQWDWLRQ ZKLFK LQGLFDWHV VSDWLDO YLVXDOL]DWLRQf $LU WUDIILF FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ ZRUN VFKHGXOHV ZLOO LQGLFDWH D PHDQ OHDUQLQJ FXUYH VORSH DQG LWV UHVLGXDOVf ZKLFK LV QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKH PHDQ OHDUQLQJ FXUYH RI FRQWUROOHUV ZKR ZRUN GD\VKLIW DQGRU HYHQLQJVKLIW VFKHGXOHV IRU WKHLU 1RYD6FDQ UHSHDWHG PHDVXUHV RI WUDFNLQJ HUURU RQ 7DVN ZKLFK LQGLFDWHV PLQGPRWRU FRRUGLQDWLRQf 9DULDEOHV 7KH IROORZLQJ FULWHULDQ YDULDEOHV ZHUH PHDVXUHG DQG DQDO\]HG $WWHQWLRQ DOORFDWLRQ GXULQJ 7DVN VSDWLDO YLVXDOL]DWLRQf $WWHQWLRQ DOORFDWLRQ GXULQJ 7DVN WUDFNLQJf 5HDFWLRQ WLPH WR WUDQVLWLRQ IURP 7DVN WR 7DVN UHVSRQVH IOH[LELOLW\ LQ VZLWFKLQJ UHVRXUFHVf 1RQWUDQVLWLRQ UHDFWLRQ WLPH WR 7DVN VSDWLDO YLVXDOL]DWLRQf 7UDFNLQJ HUURU GXULQJ 7DVN WUDFNLQJf 'DWD $QDO\VLV ,Q RUGHU WR DQVZHU WKH K\SRWKHVHV WKH GDWD ZHUH DQDO\]HG VR WKDW WKH WZR JURXSV RI SDUWLFLSDQWV FRXOG EH FRPSDUHG RQ WZR W\SHV RI YDULDEOHV DWWHQWLRQ DOORFDWLRQ DQG OHDUQLQJ 7KLV ZDV DFFRPSOLVKHG LQ WKH IROORZLQJ PDQQHU

PAGE 85

7KH WZR EDVHOLQH PHDVXUHV RI DWWHQWLRQ DOORFDWLRQ IRU HDFK LQGLYLGXDO ZHUH FDOFXODWHG DV WKH DYHUDJH UHDFWLRQ WLPH RQ GLDO PRQLWRULQJ GXULQJ WDVNV DQG IURP WKH ODVW VHVVLRQV WDNHQ RQ 1RYD6FDQ 2QFH DOO WKH SDUWLFLSDQWV KDG FRPSOHWHG WKHLU WK VHVVLRQ RQ 1RYD6FDQ D W7HVW IRU LQGHSHQGHQW VDPSOHV ZDV FRPSOHWHG IRU HDFK PHDVXUH ZLWK S GHILQHG DV VWDWLVWLFDOO\ VLJQLILFDQW 7KH UHVXOWV RI WKHVH W7HVWV ZHUH XVHG WR FRPSDUH WKH WZR JURXSV RI SDUWLFLSDQWV RQ HDFK RI WKHVH WZR PHDVXUHV DQG WR DGGUHVV K\SRWKHVHV DQG 7KH OHDUQLQJ FXUYHV IRU HDFK LQGLYLGXDO ZHUH GHWHUPLQHG IRU WKH ILYH 1RYD6FDQ YDULDEOHV Df $WWHQWLRQ DOORFDWLRQ GXULQJ WKH VSDWLDO YLVXDOL]DWLRQ WDVN Ef DWWHQWLRQ DOORFDWLRQ GXULQJ WKH WUDFNLQJ WDVN Ff UHVSRQVH IOH[LELOLW\ LQ VZLWFKLQJ UHVRXUFHV Gf QRQWUDQVLWLRQ UHDFWLRQ WR VSDWLDO YLVXDOL]DWLRQ DQG Hf WUDFNLQJ HUURU GXULQJ WUDFNLQJ 7KH PHDQ OHDUQLQJ FXUYH IRU HDFK YDULDEOH ZDV GHWHUPLQHG IRU HDFK JURXS WR EH XVHG WR FRPSDUH WKH WZR JURXSV RI SDUWLFLSDQWV DQG WR DGGUHVV K\SRWKHVHV WKURXJK $ 7ZR:D\ *HQHUDO /LQHDU 0RGHO 5HSHDWHG0HDVXUHV $QDO\VLV ZKLFK LQFOXGHG WKUHH VWDWLVWLFDO WHVW *URXS 0DLQ (IIHFW 7LPH 0DLQ (IIHFW DQG *URXS7LPH ,QWHUDFWLRQf ZDV XVHG WR DQDO\]H WKH GDWD IRU WKH ILYH YDULDEOHVf PHDQ OHDUQLQJ FXUYH VORSHV DQG WKHLU UHVLGXDOV ZLWK S GHILQHG DV VWDWLVWLFDOO\ VLJQLILFDQW LQ RUGHU WR FRPSDUH

PAGE 86

WKH WZR JURXSVf SURJUHVV RYHU WKH VHVVLRQV :KHUHYHU D *URXS7LPH ,QWHUDFWLRQ ZDV IRXQG WR EH VWDWLVWLFDOO\ VLJQLILFDQW WKH *URXS 0DLQ (IIHFW DQG *URXS 7LPH (IIHFW ZHUH QRW UHSRUWHG SHU UXOHf $OVR IRU HDFK WHVW WKDW ZDV IRXQG WR EH VWDWLVWLFDOO\ VLJQLILFDQW HWD VTXDUHG Qf ZDV FDOFXODWHG LQ RUGHU WR GHWHUPLQH ZKHWKHU WKH VLJQLILFDQW GLIIHUHQFH ZDV VOLJKWO\ Q f PRGHUDWHO\ Q f RU KLJKO\ f PHDQLQJIXO IRU WKH VL]H RI WKH VDPSOH XVHG LQ WKLV VWXG\ $FFRUGLQJ WR &RKHQ f HWD VTXDUHG LV WKH SURSRUWLRQ RI WRWDO YDULDQFH LQ WKH RXWFRPH WKDW LV DFFRXQWHG IRU E\ WKH EHWZHHQ JURXS YDULDQFH ,Q VKRUW HWD VTXDUHG GHWHUPLQHV LI WKH VLJQLILFDQW GLIIHUHQFH EHWZHHQ WKH JURXSV LV PHDQLQJIXO DQG D PRGHUDWH HWD VTXDUHG UHVXOW Uf f RU KLJKHU ZRXOG PHHW WKLV VWDQGDUG 7KH OHDUQLQJ FXUYHV IRU WKH WZR JURXSV ZHUH VXSHULPSRVHG DQG GLVSOD\HG WRJHWKHU IRU YLVXDO FRPSDULVRQ RQ HDFK YDULDEOH DQG HDFK JURXSfV PHDQ OHDUQLQJ FXUYHV ZHUH GLVSOD\HG LQGLYLGXDOO\ ZLWK WKHLU HUURU EDQGV VWDQGDUG GHYLDWLRQVf 2QO\ WKH ODVW WKUHH VHFRQGV WLPH SHULRG WZRf RI HDFK WULDO IRU YDULDEOHV DWWHQWLRQ DOORFDWLRQ GXULQJ WUDFNLQJf DQG WUDFNLQJ HUURU GXULQJ WUDFNLQJf ZHUH XVHG LQ WKH FDOFXODWLRQ IRU WKHVH WDVNV LQ RUGHU WR FRQWURO IRU WKH YDULDEOH VWDWH RI WKH WUDFNLQJ WDVN DW HDFK RI LWV LQLWLDWLRQV )RU WKLV VDPH UHDVRQ YDULDEOH UHVSRQVH IOH[LELOLW\

PAGE 87

LQ VZLWFKLQJ UHVRXUFHVf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

PAGE 88

&+$37(5 5(6($5&+ ),1',1*6 5HVXOWV +\SRWKHVLV $ VWDWLVWLFDOO\VLJQLILFDQW GLIIHUHQFH ZDV IRXQG EHWZHHQ WKH WZR JURXSVf DWWHQWLRQ DOORFDWLRQ EDVHOLQH PHDVXUH IRU GLDO PRQLWRULQJ GXULQJ 7DVN VSDWLDO YLVXDOL]DWLRQf *URXS 5DSLGO\5RWDWLQJ 6KLIW :RUNHUVf GHPRQVWUDWHG D EHWWHU DWWHQWLRQ DOORFDWLRQ UDWLQJ D IDVWHU UHDFWLRQ WLPH IRU GLDO PRQLWRULQJf GXULQJ 7DVN VSDWLDO YLVXDOL]DWLRQf WKDQ *URXS 'D\(YHQLQJ 6KLIW :RUNHUVf GHPRQVWUDWHG 7KH QXOO K\SRWKHVLV ZDV UHMHFWHG +\SRWKHVLV $ VWDWLVWLFDOO\VLJQLILFDQW GLIIHUHQFH ZDV IRXQG EHWZHHQ WKH WZR JURXSVf DWWHQWLRQ DOORFDWLRQ EDVHOLQH PHDVXUH IRU GLDO PRQLWRULQJ GXULQJ 7DVN WUDFNLQJf *URXS 5DSLGO\5RWDWLQJ 6KLIW :RUNHUVf GHPRQVWUDWHG D EHWWHU DWWHQWLRQ DOORFDWLRQ D IDVWHU UHDFWLRQ WLPH IRU GLDO PRQLWRULQJf GXULQJ 7DVN WUDFNLQJf WKDQ *URXS 'D\(YHQLQJ 6KLIW :RUNHUVf GHPRQVWUDWHG 7KH QXOO K\SRWKHVLV ZDV UHMHFWHG

PAGE 89

+\SRWKHVLV $ VWDWLVWLFDOO\VLJQLIOFDQW *URXS7LPH ,QWHUDFWLRQ ZLWK D VPDOO HIIHFW VL]H ZDV IRXQG EHWZHHQ WKH WZR JURXSVf PHDQ OHDUQLQJ FXUYHV IRU 9DULDEOH GLDO PRQLWRULQJ DWWHQWLRQ DOORFDWLRQf GXULQJ 7DVN VSDWLDO YLVXDOL]DWLRQf *URXS 5DSLGO\5RWDWLQJ 6KLIW :RUNHUVf GHPRQVWUDWHG D GHFUHDVLQJ OHDUQLQJ FXUYH ZLWK DQ LPSURYLQJ UHDFWLRQ WLPH RYHU WKH VHVVLRQV ZKLOH *URXS 'D\(YHQLQJ 6KLIW :RUNHUVf GHPRQVWUDWHG D IDLUO\ IODW OHDUQLQJ FXUYH %RWK JURXSV GHPRQVWUDWHG VWDELOL]DWLRQ DURXQG WKHLU PHDQ EDVHOLQH PHDVXUHV ZLWK QR *URXS7LPH ,QWHUDFWLRQ EHWZHHQ WKH JURXSVf UHVLGXDOV 7KH QXOO K\SRWKHVLV ZDV UHMHFWHG +\SRWKHVLV 1R VWDWLVWLFDOO\VLJQLIOFDQW *URXS 0DLQ (IIHFW 7LPH 0DLQ (IIHFW RU *URXS7LPH ,QWHUDFWLRQ ZDV IRXQG EHWZHHQ WKH WZR JURXSVf OHDUQLQJ FXUYH VORSHV IRU 9DULDEOH GLDO PRQLWRULQJ DWWHQWLRQ DOORFDWLRQf GXULQJ 7DVN WUDFNLQJf 7KH OHDUQLQJ FXUYH VORSHV IRU WKH WZR JURXSV DSSHDU VWDWLVWLFDOO\ WKH VDPH IDLUO\ IODWf DQG GLG QRW LPSURYH WR D VWDWLVWLFDOO\VLJQLIOFDQW GHJUHH RYHU WLPH %RWK JURXSV GHPRQVWUDWHG D VWDELOL]DWLRQ DURXQG WKHLU PHDQ EDVHOLQH PHDVXUH KRZHYHU WKH JURXSVf UHVLGXDOV GHPRQVWUDWHG D VWDWLVWLFDOO\ VLJQLIOFDQW *URXS7LPH ,QWHUDFWLRQ ZLWK D PRGHUDWH HIIHFW VL]H LQGLFDWLQJ D GLIIHUHQFH LQ WKH JURXSVf UHVLGXDOV ZKHQ FRPSDUHG RYHU

PAGE 90

WKH VHVVLRQV 7KH VWDQGDUG GHYLDWLRQV DURXQG WKH PHDQV RI WKH VORSHV ZHUH QRW WKH VDPH IRU WKH JURXSV ZKHQ FRPSDUHG RYHU WLPH DQG WKH JURXSV GLG QRW VWDELOL]H DURXQG WKHLU PHDQV DW WKH VDPH UDWH 7KH QXOO K\SRWKHVLV ZDV UHMHFWHG +\SRWKHVLV 1R VWDWLVWLFDOO\VLJQLIOFDQW *URXS 0DLQ (IIHFW RU *URXS7LPH ,QWHUDFWLRQ ZDV IRXQG EHWZHHQ WKH WZR JURXSVf OHDUQLQJ FXUYHV IRU 9DULDEOH UHDFWLRQ WLPH GXULQJ WKH WUDQVLWLRQ IURP 7DVN WR 7DVN UHVSRQVH IOH[LELOLW\ LQ VZLWFKLQJ UHVRXUFHVf LQGLFDWLQJ WKDW WKH JURXSVf PHDQ OHDUQLQJ FXUYHV DSSHDU VWDWLVWLFDOO\ HTXDO 7KHUH ZDV KRZHYHU D VWDWLVWLFDOO\VLJQLIOFDQW 7LPH 0DLQ (IIHFW ZLWK D ODUJH HIIHFW VL]H LQGLFDWLQJ WKDW ERWK JURXSVf OHDUQLQJ FXUYHV GHSLFW LPSURYHG UHDFWLRQ WLPHV RYHU WKH VHVVLRQV 1R HUURUEDQG HIIHFWV ZHUH IRXQG WR EH VWDWLVWLFDOO\ VLJQLILFDQW LQGLFDWLQJ QR VWDWLVWLFDOO\ VLJQLIOFDQW FKDQJH LQ WKHP RYHU WLPH DQG QR *URXS7LPH ,QWHUDFWLRQ EHWZHHQ WKH HUURU EDQGV 7KH QXOO K\SRWKHVLV ZDV QRW UHMHFWHG +\SRWKHVLV 1R VWDWLVWLFDOO\VLJQLIOFDQW *URXS 0DLQ (IIHFW RU *URXS7LPH ,QWHUDFWLRQ ZDV IRXQG EHWZHHQ WKH WZR JURXSVf OHDUQLQJ FXUYHV IRU 9DULDEOH UHDFWLRQ WLPH GXULQJ 7DVN VSDWLDO YLVXDOL]DWLRQf LQGLFDWLQJ WKDW WKH JURXSVf PHDQ OHDUQLQJ FXUYHV DSSHDU VWDWLVWLFDOO\ HTXDO 7KHUH ZDV KRZHYHU D VWDWLVWLFDOO\VLJQLIOFDQW 7LPH 0DLQ

PAGE 91

(IIHFW ZLWK D ODUJH HIIHFW VL]H LQGLFDWLQJ WKDW ERWK JURXSVf OHDUQLQJ FXUYHV GHSLFW LPSURYHG UHDFWLRQ WLPHV RYHU WLPH 1R HUURUEDQG HIIHFWV ZHUH IRXQG WR EH VWDWLVWLFDOO\ VLJQLILFDQW LQGLFDWLQJ QR VWDWLVWLFDOO\VLJQLIOFDQW FKDQJH LQ WKHP RYHU WLPH DQG QR *URXS7LPH ,QWHUDFWLRQ EHWZHHQ WKH JURXSVf HUURU EDQGV 7KH QXOO K\SRWKHVLV ZDV QRW UHMHFWHG +\SR WKHVLV 1R VWDWLVWLFDOO\VLJQLIOFDQW *URXS 0DLQ (IIHFW RU *URXS7LPH ,QWHUDFWLRQ ZDV IRXQG EHWZHHQ WKH WZR JURXSVf OHDUQLQJ FXUYHV IRU 9DULDEOH WUDFNLQJ HUURU GXULQJ 7DVN WUDFNLQJf LQGLFDWLQJ WKDW WKH JURXSVf PHDQ OHDUQLQJ FXUYHV DSSHDU VWDWLVWLFDOO\ HTXDO 7KHUH ZDV KRZHYHU D VWDWLVWLFDOO\VLJQLIOFDQW 7LPH 0DLQ (IIHFW ZLWK D PRGHUDWH HIIHFW VL]H LQGLFDWLQJ WKDW ERWK JURXSVf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f LQGLFDWHV WKDW WKH PHDQ EDVHOLQH PHDVXUH IRU

PAGE 92

7DEOH 6XPPDU\ RI W7HVW 5HVXOWV IRU *URXS 0HDQ $WWHQWLRQ $OORFDWLRQ &RPSDULVRQV LQ 0LOOLVHFRQGV RI 5HDFWLRQ 7LPH *URXS *URXS 9DULDEOH Q 0 6 Q 0 7DVN W R USE f 7DVN W S URE f rS *URXS 5DSLGO\5RWDWLQJ 6KLIW :RUNHUVf LV PLOOLVHFRQGV RI UHDFWLRQ WLPH ZKLOH WKH PHDQ PHDVXUH IRU *URXS 'D\(YHQLQJ 6KLIW :RUNHUVf LV PLOOLVHFRQGV RI UHDFWLRQ WLPH $V GHWHUPLQHG E\ WKH W7HVW WKLV GLIIHUHQFH LV VWDWLVWLFDOO\ VLJQLILFDQW >Wf S UES @ 7KH VHFRQG JURXS FRPSDULVRQ DWWHQWLRQ DOORFDWLRQ GXULQJ WKH WUDFNLQJ WDVNf LQGLFDWHV WKDW WKH PHDQ EDVHOLQH PHDVXUH IRU *URXS LV PLOOLVHFRQGV RI UHDFWLRQ WLPH ZKLOH WKH PHDQ PHDVXUH IRU *URXS LV PLOOLVHFRQGV RI UHDFWLRQ WLPH $V GHWHUPLQHG E\

PAGE 93

7DEOH 6XPPDU\ RI /HDUQLQJ &XUYH $QDO\VHV 95%/ ) 7HVW ) 9DOXH Q 9DOXH (WD 6DXDUHG 6ORSH*7, 5HVGO*0( 5HVGO70( 5HVGO*7O 6ORSH*0( 6ORSH70( 6ORSH*7, 5HVGO*7, 6ORSH*0( 6ORSH70( 6ORSH*7, 5HVGO*0( 5HVGO70( 5HVGO*7, 6ORSH*0( 6ORSH70( 6ORSH*7, 5HVGO*0( 5HVGO70( 5HVGO*7, 6ORSH*0( 6ORSH70( 6ORSH*7, 5HVGO*0( 5HVGO70( 5HVGO*7, $EEUHYLDWLRQV XVHG LQ 7DEOH 5HVGO 5HVLGXDO *0( *URXS 0DLQ (IIHFW 70( 7LPH 0DLQ (IIHFW *7, *URXS7LPH ,QWHUDFWLRQ rS

PAGE 94

WKH W7HVW WKLV GLIIHUHQFH LV DOVR VWDWLVWLFDOO\ VLJQLILFDQW >Wf S USE @ /HDUQLQJ &XUYH &RPSDULVRQ 7DEOH GHVFULEHV D VXPPDU\ RI UHVXOWV IRU WKH OHDUQLQJ FXUYH DQDO\VHV DQG KRZ WKH\ UHODWH WR WKH OHDUQLQJ FXUYHV GLVSOD\HG LQ )LJXUHV WKURXJK )LJXUHV WKURXJK GLVSOD\ WKH PHDQ OHDUQLQJ FXUYHV IRU *URXSV DQG VXSHULPSRVHGf IRU HDFK RI WKH ILYH YDULDEOHV ZKLOH )LJXUHV WKURXJK GLVSOD\ HDFK JURXSfV LQGHSHQGHQW OHDUQLQJ FXUYH IRU HDFK YDULDEOH ZLWK LWV HUURU EDQG VWDQGDUG GHYLDWLRQf $V LQGLFDWHG LQ 7DEOH WKHUH LV D VWDWLVWLFDOO\VLJQLILFDQW *URXS7LPH ,QWHUDFWLRQ EHWZHHQ WKH JURXSVf OHDUQLQJ FXUYHV IRU 9DULDEOH GLDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ 7DVN VSDWLDO YLVXDOL]DWLRQf 7KH HWD VTXDUHG YDOXH IRU WKLV LQWHUDFWLRQ LV ZKLFK LQGLFDWHV D VOLJKW WR PRGHUDWH HIIHFW VL]H 7KLV LQWHUDFWLRQ LPSOLHV WKDW WKH VORSHV RI WKH WZR JURXSVf OHDUQLQJ FXUYHV DSSHDU WR EH GLIIHUHQW *URXS GLVSOD\V D GHFUHDVLQJ VORSH ZKLOH *URXS GLVSOD\V D IDLUO\ IODW VORSH ZLWK OLWWOH LPSURYHPHQW RYHU WLPH 7KH DSSDUHQW GLIIHUHQFH EHWZHHQ WKH WZR VORSHV FDQ EH VHHQ LQ )LJXUH DQG WKH VWDWLVWLFDOO\VLJQLILFDQW GLIIHUHQFH ZKLFK ZDV IRXQG LQ WKH EDVHOLQHV E\ WKH W7HVW IRU WKH ILQDO WKUHH VHVVLRQV LV REYLRXV

PAGE 95

7KHUH LV QR *URXS7LPH ,QWHUDFWLRQ IRU WKH JURXSVf UHVLGXDOV RQ 9DULDEOH 7KH *URXS 0DLQ (IIHFW WHVW 7DEOH f IRU WKH UHVLGXDOV LQGLFDWHV WKDW WKH HUURU EDQGV DUH VWDWLVWLFDOO\ WKH VDPH VL]H IRU ERWK JURXSV )LJXUHV DQG f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f SURGXFHG D VWDWLVWLFDOO\ QRQVLJQLILFDQW UHVXOW PHDQLQJ WKDW WKH VORSHV DSSHDU WR EH VWDWLVWLFDOO\ WKH VDPH ZKHQ FRPSDUHG RYHU WLPH 7KH VORSH FRPSDULVRQ IRU *URXSV DQG 7DEOH f LQGLFDWHV WKDW WKH VORSHV VWDWLVWLFDOO\ DSSHDU WR EH WKH VDPH ZLWK QR VWDWLVWLFDOO\VLJQLIOFDQW *URXS 0DLQ (IIHFW )LJXUH f 7KHUH LV DOVR QR VWDWLVWLFDOO\VLJQLIOFDQW 7LPH 0DLQ (IIHFW LQGLFDWLQJ WKDW WKH VORSHV GLG QRW LQFUHDVH RU GHFUHDVH WR D VWDWLVWLFDOO\VLJQLIOFDQW GHJUHH RYHU WKH VHVVLRQV 7KH VWDWLVWLFDOO\VLJQLIOFDQW GLIIHUHQFH ZKLFK ZDV IRXQG EHWZHHQ WKH WZR JURXSVf EDVHOLQHV E\ WKH W7HVW IRU WKLV YDULDEOH LV DSSDUHQW EXW QRW DV REYLRXV DV ZLWK 9DULDEOH

PAGE 96

, ,, 6HVVLRQ )LJXUH 0HDQ OHDUQLQJ FXUYHV IRU GLDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ 7DVN 6SDWLDO 9LVXDOL]DWLRQf )LJXUH 0HDQ OHDUQLQJ FXUYHV IRU GLDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ 7DVN 7UDFNLQJf

PAGE 97

)LJXUH 0HDQ OHDUQLQJ FXUYHV IRU UHDFWLRQ WLPH WR WUDQVLWLRQ IURP 7DVN 7UDFNLQJf WR 7DVN 6SDWLDO 9LVXDOL]DWLRQf )LJXUH 0HDQ OHDUQLQJ FXUYHV IRU UHDFWLRQ WLPH WR 7DVN 6SDWLDO 9LVXDOL]DWLRQf

PAGE 98

)LJXUH 0HDQ OHDUQLQJ FXUYHV IRU WUDFNLQJ HUURU GXULQJ 7DVN 7UDFNLQJf 7KH UHVLGXDO WHVWV 7DEOH f IRU 9DULDEOH GHPRQVWUDWHV D VWDWLVWLFDOO\VLJQLILFDQW *URXS7LPH ,QWHUDFWLRQ ZLWK D PRGHUDWH HIIHFW VL]H Q f IRU WKH UHVLGXDOV )LJXUHV DQG f 7KLV LQGLFDWHV WKDW WKH VL]H RI WKH HUURU EDQGV GHFUHDVH DW VWDWLVWLFDOO\ GLIIHUHQW UDWHV RYHU WLPH IRU WKH WZR JURXSV 7KH VWDWLVWLFDO DQDO\VHV 7DEOH f IRU 9DULDEOHV UHDFWLRQ WLPH GXULQJ WUDQVLWLRQ IURP 7DVN WR 7DVN f UHDFWLRQ WLPH GXULQJ 7DVN f DQG WUDFNLQJ HUURUf DOO KDG VLPLODU UHVXOWV 7KHUH ZDV QR *URXS7LPH ,QWHUDFWLRQ EHWZHHQ WKH JURXSV RQ DQ\ RI WKHVH WKUHH YDULDEOHV LQGLFDWLQJ WKDW WKH VORSHV RI WKH JURXSV DSSHDU VWDWLVWLFDOO\ WKH VDPH RYHU WKH VHVVLRQV LQ HDFK FDVH $V )LJXUHV DQG

PAGE 99

)LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU GLDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ 7DVN 6SDWLDO 9LVXDOL]DWLRQf )LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU GLDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ 7DVN 6SDWLDO 9LVXDOL]DWLRQf

PAGE 100

)LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU GLDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ 7DVN 7UDFNLQJf )LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU GLDO PRQLWRULQJ UHDFWLRQ WLPH GXULQJ 7DVN 7UDFNLQJf

PAGE 101

)LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU UHDFWLRQ WLPH WR WKH WUDQVLWLRQ IURP 7DVN 7UDFNLQJf WR 7DVN 6SDWLDO 9LVXDOL]DWLRQf )LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU UHDFWLRQ WLPH WR WKH WUDQVLWLRQ IURP 7DVN 7UDFNLQJf WR 7DVN 6SDWLDO 9LVXDOL]DWLRQf

PAGE 102

)LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU UHDFWLRQ WLPH WR 7DVN 6SDWLDO 9LVXDOL]DWLRQf )LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQ IRU UHDFWLRQ WLPH WR 7DVN 6SDWLDO 9LVXDOL]DWLRQf

PAGE 103

)LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU WUDFNLQJ HUURU GXULQJ 7DVN 7UDFNLQJf )LJXUH *URXS PHDQ OHDUQLQJ FXUYH ZLWK HUURU EDQG VWDQGDUG GHYLDWLRQf IRU WUDFNLQJ HUURU GXULQJ 7DVN 7UDFNLQJf

PAGE 104

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f IRU 9DULDEOHV )LJXUHV DQG f )LJXUHV DQG f DQG )LJXUHV DQG f DOO LQGLFDWH QR VWDWLVWLFDOO\VLJQLILFDQW GLIIHUHQFH EHWZHHQ WKH WZR JURXSV 7KLV PHDQV WKDW WKHUH ZDV QR *URXS7LPH ,QWHUDFWLRQ EHWZHHQ WKH WZR JURXSVf HUURU EDQGV ,W DOVR PHDQV WKDW WKH HUURU EDQGV DUH VWDWLVWLFDOO\ HTXDO LQ VL]H IRU WKH WZR JURXSV RQ HDFK YDULDEOH DQG WKDW WKH VL]H RI WKH HUURU EDQGV GLG QRW FKDQJH WR D VWDWLVWLFDOO\VLJQLILFDQW GHJUHH IRU HLWKHU JURXS RYHU WKH VHVVLRQV

PAGE 105

&+$37(5 &21&/86,21 ,QWURGXFWLRQ $LU WUDIILF FRQWUROOHUV WHQG WR SUHIHU SKDVHDGYDQFHG UDSLGO\ URWDWLQJ ZRUN VFKHGXOHV EHFDXVH RI WKH JUHDWHU EUHDNV WKH\ SURYLGH EHWZHHQ ZRUN ZHHNV 3KDVHDGYDQFHG VFKHGXOHV URWDWH FRXQWHUFORFNZLVH EHJLQQLQJ ZLWK WKH HYHQLQJ VKLIW 7KH FRQWUROOHU PD\ ZRUN RQH WZR RU WKUHH RI WKHVH HYHQLQJ VKLIWV GHSHQGLQJ XSRQ WKH LQGLYLGXDOL]HG VFKHGXOH WKDW KH RU VKH ZRUNV 7KLV W\SH RI VFKHGXOH WKHQ DOORZV RQO\ DQ KRXU EUHDN EHIRUH WKH FRQWUROOHU PXVW UHWXUQ IRU D GD\ VKLIW DQG DW WKH HQG RI WKH ODVW GD\ VKLIW WKH FRQWUROOHU HQFRXQWHUV DQRWKHU KRXU TXLFNWXUQDURXQG EHIRUH UHWXUQLQJ WR ZRUN WKH QLJKW VKLIW ,QGLYLGXDO FRQWUROOHUV ZRUN D YDULHW\ RI FRPELQDWLRQV RI WKHVH WKUHH VKLIWV EXW WKH URWDWLRQ SDWWHUQ UHPDLQV WKH VDPH %\ FRPSUHVVLQJ WKH ZRUN VFKHGXOH GXULQJ WKH ZHHN LQ WKLV PDQQHU FRQWUROOHUV UHDOL]H DQ KRXU EUHDN EHWZHHQ ZRUN ZHHNV $OWKRXJK FRQWUROOHUV SUHIHU WR ZRUN UDSLGO\URWDWLQJ VFKHGXOHV DQRWKHU UHDVRQ WKDW FRQWUROOHUV JHQHUDOO\ GR QRW ZRUN VWHDG\ QRQURWDWLQJ VFKHGXOHV LV WKDW WKH )$$ GLVFRXUDJHV WKLV SUDFWLFH

PAGE 106

7UDIILF LV JHQHUDOO\ OLJKW RQ WKH QLJKW VKLIW DQG LW LV IHOW WKDW FRQWUROOHUV SHUPDQHQWO\ RQ WKLV VKLIW ZRXOG H[SHULHQFH D GHWHULRUDWLRQ RI SURILFLHQF\ LQ KDQGOLQJ KHDYLHU WUDIILF ORDGV VKRXOG WKH\ EH FDOOHG XSRQ WR GR VR $OVR VRFLDO LVRODWLRQ IURP SURIHVVLRQDO FROOHDJXHV ZRXOG NHHS WKHVH FRQWUROOHUV IURP DWWHQGLQJ WUDLQLQJ EULHILQJV DQG FRQIHUHQFHV 0HOWRQ f 7KH K\SRWKHVHV DGGUHVVHG E\ WKLV VWXG\ ZHUH GHVLJQHG WR H[DPLQH ZKHWKHU UDSLGO\URWDWLQJ VKLIW ZRUN PD\ KDYH DQ REMHFWLYHO\PHDVXUDEOH HIIHFW RQ DWWHQWLRQ DOORFDWLRQ GXULQJ WKH GD\ VKLIW KRXUV DP WR SPf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

PAGE 107

RWKHU OHDUQHU FKDUDFWHULVWLFV ZKLFK QHHG WR EH DGGUHVVHG IRU WKH SXUSRVH RI PD[LPL]LQJ OHDUQLQJ 7KLV VWXG\ ZDV DOVR GHHPHG QHFHVVDU\ EHFDXVH UHFHQW UHVHDUFK 5HGGLQJ f GHWHUPLQHG WKDW D ODFN RI VLWXDWLRQ DZDUHQHVV LV WKH SULPDU\ FDXVH RI FRQWUROOHU HUURU DQG (QGVOH\ f GHYHORSHG D PRGHO RI VLWXDWLRQ DZDUHQHVV ZKLFK LQFOXGHV WUDLQLQJ DV D FRQWULEXWLQJ IDFWRU WR VLWXDWLRQ DZDUHQHVV LQ G\QDPLF VLWXDWLRQV (DFK \HDU D QXPEHU RI FRQWUROOHU HUURUV VRPH RI ZKLFK OHDG WR DLUFUDIW DFFLGHQWV PD\ RFFXU EHFDXVH FRQWUROOHUV DUH DIIHFWHG E\ FLUFDGLDQ G\VUK\WKPLD D PLVDOLJQPHQW RI SK\VLRORJLFDO DQGRU SV\FKRORJLFDO GDLO\ UK\WKPVf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

PAGE 108

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f GHPRQVWUDWHG D EHWWHU DWWHQWLRQ DOORFDWLRQ EDVHOLQH PHDVXUH WKDQ GLG

PAGE 109

*URXS 'D\(YHQLQJ 6KLIW :RUNHUVf RQ ERWK PHDVXUHV 9DULDEOH GLDO PRQLWRULQJ GXULQJ WKH VSDWLDO YLVXDOL]DWLRQ WDVNf DQG 9DULDEOH GLDO PRQLWRULQJ GXULQJ WKH WUDFNLQJ WDVNf 7KH GLIIHUHQFH EHWZHHQ WKH WZR JURXSVf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f LQGLFDWH VRPH LQWHUHVWLQJ ILQGLQJV )LJXUHV t f ,Q ERWK PHDVXUHV RI DWWHQWLRQ DOORFDWLRQ EDVHOLQH *URXS SHUIRUPHG EHWWHU WKDQ GLG *URXS ,Q WKH ILUVW PHDVXUH DWWHQWLRQ DOORFDWLRQ GXULQJ VSDWLDO YLVXDOL]DWLRQf *URXS LPSURYHG RYHU WLPH ZLWK D

PAGE 110

GHFUHDVLQJ VORSH ZKLOH *URXS GLVSOD\HG D IDLUO\ IODW VORSH )LJXUH f +RZHYHU DV 7DEOH LQGLFDWHV WKH HWD VTXDUHG IRU WKH GLIIHUHQFH EHWZHHQ WKH WZR JURXSV RQ WKLV PHDVXUH LV RQO\ WA LQGLFDWLQJ OHVV WKDQ D PRGHUDWHO\PHDQLQJIXO GLIIHUHQFH US f %HFDXVH RI WKLV IXUWKHU UHVHDUFK VKRXOG EH FRQGXFWHG WR GHWHUPLQH LI WKLV HIIHFW FDQ EH GXSOLFDWHG ZLWK D PRUH PHDQLQJIXO RXWFRPH 7KH HUURU EDQGV IRU ERWK VORSHV )LJXUHV t f VWDELOL]HG DW D VWDWLVWLFDOO\ VLJQLILFDQW GHJUHH RYHU WKH VHVVLRQV ZLWK *URXS GHPRQVWUDWLQJ DQ DSSDUHQWO\ KLJKHU GHJUHH RI VWDELOL]DWLRQ RYHU WLPH WKDQ GLG *URXS ,Q WKH VHFRQG PHDVXUH DWWHQWLRQ DOORFDWLRQ GXULQJ WUDFNLQJf QHLWKHU *URXS RU DSSHDUV WR KDYH LPSURYHG RYHU WKH VHVVLRQV )LJXUH f EXW *URXS UHDFKHG D EDVHOLQH PHDVXUH ZKLFK LV EHWWHU WKDQ WKDW RI *URXS DV LQGLFDWHG E\ WKH W7HVW IRU 9DULDEOH $V ZLWK WKH ILUVW PHDVXUH ERWK JURXSV VWDELOL]HG DURXQG WKHLU PHDQ EDVHOLQH PHDVXUHV KRZHYHU WKH\ VWDELOL]HG DW GLIIHUHQW UDWHV ZKLFK ZHUH IRXQG WR EH VWDWLVWLFDOO\ VLJQLILFDQW E\ WKH *URXS7LPH ,QWHUDFWLRQ EHWZHHQ WKH WKH JURXSVf UHVLGXDOV $V FDQ EH VHHQ LQ ILJXUHV WKURXJK WKH JURXSVf UHVLGXDOV VWDELOL]H DURXQG WKHLU LQGLYLGXDO JURXS PHDQV RQ ERWK PHDVXUHV RI DWWHQWLRQ DOORFDWLRQ 7KLV LV VWURQJ HYLGHQFH WKDW WKH LQGLYLGXDO GLIIHUHQFHV ZLWKLQ HDFK JURXS GLVDSSHDU RYHU WKH VHVVLRQV DQG WKDW

PAGE 111

WKH GLIIHUHQFHV EHWZHHQ WKH JURXSVf EDVHOLQH PHDVXUHV UHDOO\ DUH JURXS GLIIHUHQFHV ,W LV DOVR QRWHZRUWK\ WKDW WKH PHDQ UHDFWLRQ WLPHV IRU ERWK JURXSV DUH KLJKHU GXULQJ 7DVN WKDQ WKH\ DUH GXULQJ 7DVN 7DVN WUDFNLQJf ZKLFK LV D SV\FKRPRWRU DFWLYLW\ DSSHDUV WR UHTXLUH PRUH FRQFHQWUDWLRQ WKDQ GRHV 7DVN VSDWLDO YLVXDOL]DWLRQf ZKLFK LV D FRJQLWLYH WDVN %RWK JURXSV GHPRQVWUDWHG D ORZHU DWWHQWLRQ DOORFDWLRQ DELOLW\ ZKLOH SHUIRUPLQJ WKH SV\FKRPRWRU WDVN DQG QHLWKHU JURXS LPSURYHG WR D VWDWLVWLFDOO\VLJQLILFDQW GHJUHH RQ WKLV WDVN RYHU WKH VHVVLRQV 7KH DWWHQWLRQ DOORFDWLRQ UDWLQJV IRU WKH JURXSV GXULQJ WKH FRJQLWLYH WDVN VSDWLDO YLVXDOL]DWLRQf LQGLFDWH GLIIHUHQWLDO UHVXOWV $V ZLWK WKH SV\FKRPRWRU WDVN *URXS DJDLQ GHPRQVWUDWHG QR VWDWLVWLFDOO\VLJQLILFDQW LPSURYHPHQW RYHU WKH VHVVLRQV +RZHYHU *URXS GRHV LPSURYH DWWHQWLRQ DOORFDWLRQ GXULQJ WKLV WDVN 7KH DQDO\VHV RI WKH OHDUQLQJ FXUYHV IRU 9DULDEOHV DQG GHPRQVWUDWH QR VWDWLVWLFDOO\VLJQLILFDQW OHDUQLQJ GLIIHUHQFHV EHWZHHQ WKH WZR JURXSV %RWK JURXSV LPSURYHG WKHLU UHDFWLRQ WLPHV IRU UHVSRQVH IOH[LELOLW\ DW VZLWFKLQJ UHVRXUFHV 9DULDEOH FKDQJLQJ IURP 7DVN WR 7DVN f DQG DW REMHFW RULHQWDWLRQ 9DULDEOH VSDWLDO YLVXDOL]DWLRQf %RWK JURXSV DOVR LPSURYHG WKHLU WUDFNLQJ HUURU GXULQJ

PAGE 112

WKH WUDFNLQJ WDVN 9DULDEOH f DQG WKHUH ZHUH QR *URXS7LPH ,QWHUDFWLRQV EHWZHHQ WKH JURXSV $V FDQ EH VHHQ LQ ILJXUHV WKURXJK WKH VL]H RI WKH UHVLGXDOV DURXQG HDFK JURXSfV PHDQ UHPDLQV FRQVWDQW RYHU WKH VHVVLRQV IRU WKHVH YDULDEOHV %HFDXVH RI WKLV WKHUH LV QR VWDELOL]DWLRQ DURXQG WKH JURXSVf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

PAGE 113

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f LQGLYLGXDOV ZKR VOHHS DW LUUHJXODU VFKHGXOHV VFKHGXOHV ZKLFK YDU\ E\ WZR KRXUV RU PRUHf H[SHULHQFH D ODFN RI DOHUWQHVV GXULQJ WKH GD\WLPH ,W PD\ EH WKDW WKH W\SH RI VOHHSZDNH F\FOH SUDFWLFHG E\ WKH GD\ DQG HYHQLQJ VKLIW

PAGE 114

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fHYHQLQJ W\SHVf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f 0RVW VFKHGXOHV ZRXOG RQO\ KDYH RQH RU WZR PRUH GD\ VKLIWV IROORZLQJ WKLV ILUVW GD\ VKLIW DQG WKHVH ZRUNHUV PD\ EH PDLQWDLQLQJ D IDLUO\ ODWH EHGWLPH KRXU SULRU WR WKHVH GD\ VKLIWV DV WKH\ GR IRU

PAGE 115

RWKHU QLJKWV RI WKH ZHHN $OO RI WKH RWKHU GD\V RI WKH ZHHN PD\ EH ODWHHYHQLQJRULHQWHG IRU WKHVH LQGLYLGXDOV 'XULQJ WKH GDWD FROOHFWLRQ SKDVH RI WKLV VWXG\ D UHFRUG ZDV NHSW IRU WKH SDUWLFLSDQWVf EHGWLPHV EXW WKLV UHFRUG RQO\ UHFRUGHG WKH SDUWLFLSDQWVf EHGWLPHV IRU WKH QLJKWV SULRU WR WKHLU 1RYD6FDQ VHVVLRQV 6LQFH WKHVH VHVVLRQV RQO\ WRRN SODFH WZR WKUHH RU SRVVLEO\ IRXU GD\V RI WKH ZHHN WKLV UHFRUG LV LQDSSURSULDWH IRU GHWHUPLQLQJ LI LUUHJXODU VOHHSZDNH F\FOHV ZHUH SUDFWLFHG E\ WKH SDUWLFLSDQWV +RZHYHU WKLV UHFRUG GRHV LQGLFDWH WKDW WKH GD\ DQG HYHQLQJ VKLIW ZRUNHUVf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

PAGE 116

VFKHGXOHV DV FRPSDUHG WR FRQWUROOHUV ZKR ZRUNHG GD\ DQG HYHQLQJ VKLIW VFKHGXOHV 7KH &KL6TXDUH DQDO\VLV LQFOXGHG WKH IROORZLQJ IDFWRUV 7KH WRWDO QXPEHU RI FRQWUROOHU HUURUV UHFRUGHG DW WKH -DFNVRQYLOOH $57&& IDFLOLW\ RYHU WKH PRQWK SHULRG ZDV %DVHG RQ WKH QXPEHU RI FRQWUROOHUV ZRUNLQJ HDFK W\SH RI VFKHGXOH RI WKHVH HUURUV VKRXOG KDYH EHHQ FRPPLWWHG E\ WKH GD\ DQG HYHQLQJ VKLIW ZRUNHUV DQG HUURUV VKRXOG KDYH EHHQ FRPPLWWHG E\ WKH UDSLGO\ URWDWLQJ VKLIW ZRUNHUV LI WKH FKDQFHV RI FRPPLWWLQJ DQ HUURU ZHUH HTXDO IRU ERWK JURXSV 7KH DFWXDO QXPEHU RI HUURUV FRPPLWWHG E\ HDFK JURXS ZDV GLVFRYHUHG WR EH HUURUV E\ WKH GD\ DQG HYHQLQJ VKLIW ZRUNHUV DQG HUURUV E\ WKH UDSLGO\URWDWLQJ VKLIW ZRUNHUV %DVHG RQ WKHVH ILJXUHV WKH UHVXOW RI WKH &KL6TXDUH DQDO\VLV ZDV IRXQG WR EH VWDWLVWLFDOO\ QRQVLJQLILFDQW 1HLWKHU JURXS FRPPLWWHG D KLJKHU QXPEHU RI HUURUV FRQVLGHUHG WR EH VWDWLVWLFDOO\ VLJQLILFDQW ,PSOLFDWLRQV 7KHRUHWLFDO 5HVHDUFK LQ FKURQRSV\FKRORJ\ KDV GHPRQVWUDWHG WKDW KXPDQ SV\FKRORJLFDO DQG EHKDYLRUDO FLUFDGLDQ UK\WKPV DUH DIIHFWHG E\ VKLIW ZRUN DQG WUDQVPHULGLDQ IOLJKW $V $GDQ f LQGLFDWHV FLUFDGLDQ HIIHFWV DUH QRW WULYLDO DQG PRVW SV\FKRORJLFDO UHVHDUFK KDV QRW HYHQ

PAGE 117

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t 0RQN 0RQN f DQG VSHDN RI D QHHG IRU WKH FRQVLGHUDWLRQ RI WKH FKURQRSV\FKRORJLFDO HIIHFWV RI VKLIW ZRUN RQ OHDUQLQJ LQ LQVWUXFWLRQDO WKHRU\ DQG LQVWUXFWLRQDO GHVLJQ 7KHVH UHVXOWV DOVR LQGLFDWH WKDW FHUWDLQ W\SHV RI VKLIW ZRUN PD\ DIIHFW DWWHQWLRQ RQ WKH MRE %HFDXVH VKLIW ZRUN VFKHGXOHV PD\ DIIHFW LQGLYLGXDOVf DWWHQWLRQ DOORFDWLRQ DELOLW\ UHVHDUFK LQ VLWXDWLRQ DZDUHQHVV VKRXOG FRQVLGHU VKLIW VFKHGXOH DV D FRQWURO YDULDEOH LQ IXWXUH VWXGLHV

PAGE 118

3UDFWLFDO LOO ,QVWUXFWLRQDO GHVLJQHUV DQG LQVWUXFWRUV DOLNH VKRXOG EH PDGH DZDUH RI WKH GLXUQDO FKURQRSV\FKRORJLFDO OHDUQLQJ HIIHFWV ZKLFK KDYH EHHQ UHSRUWHG LQ WKH OLWHUDWXUH RI FKURQRSV\FKRORJ\ %DVHG RQ WKLV UHVHDUFK HJ (QJOXQG )RONDUG t 0RQN 0RQN $GDQ f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fV FLUFDGLDQ UK\WKPLF SKDVHV DW WKH WLPH RI ZRUN DQG E\ WUDLQLQJ ZKLFK LV DIIHFWHG E\ LQVWUXFWLRQDO GHVLJQ ,QVWUXFWLRQDO GHVLJQ LQFOXGHV WKH DQDO\VLV RI D

PAGE 119

OHDUQHUfV FKDUDFWHULVWLFV ZKLFK VKRXOG LQFOXGH DQ LQGLYLGXDOfV FKURQRSV\FKRORJLFDO DELOLWLHV WR OHDUQ GLIIHUHQW WDVNV DW GLIIHUHQW WLPHV RI WKH GD\ ,I WKH UHVXOWV RI WKLV VWXG\ DUH VXSSRUWHG E\ IXWXUH UHVHDUFK ILQGLQJV SUDFWLWLRQHUV PD\ FODVVLI\ WKH FRPSDUDEOH ODFN RI DWWHQWLRQ DOORFDWLRQ DELOLW\ ZKLFK GRHV QRW LPSURYH GXULQJ WKH OHDUQLQJ RI D FRJQLWLYH WDVNf RQ WKH SDUW RI WKH GD\ DQG HYHQLQJ VKLIW ZRUNHUV DV D OHDUQHU FKDUDFWHULVWLF W\SLFDO RI WKLV VKLIW V\VWHP $Q LQVWUXFWLRQDO LQWHUYHQWLRQ VXFK DV ,QWHUDFWLYH &RPSXWHU%DVHG 7UDLQLQJ ,&%7f LQWHUDFWLYH FODVV GLVFXVVLRQ RU LQWHUDFWLYH VPDOOJURXS DFWLYLWLHV FRXOG WKHQ EH GHYHORSHG WR DGGUHVV WKLV OHDUQHU FKDUDFWHULVWLF VR WKDW WKH WUDLQHH FRXOG EH DFWLYHO\ LQYROYHG LQ WKH LQVWUXFWLRQDO XQLW WKHUHE\ GLUHFWLQJ KLVKHU DWWHQWLRQ WR WKH WUDLQLQJ WDVN DQG PD[LPL]LQJ WKH OHDUQHUfV DELOLW\ WR OHDUQ LQ VSLWH RI WKLV GHILFLHQF\ 5HFRPPHQGDWLRQV 6LQFH HDUOLHU VWXGLHV 0HOWRQ 0RQN HW DO 6DOGLYDU HW DO 6PLWK HW DO f LQGLFDWH WKDW FRQWUROOHUV ZKR ZRUN UDSLGO\URWDWLQJ VFKHGXOHV UHSRUW D KLJK OHYHO RI VXEMHFWLYH IDWLJXH DQG D ODFN RI DOHUWQHVV RQ WKH GD\ VKLIW WKH UHVXOWV RI WKLV VWXG\ VHHP WR LQGLFDWH WKDW VXEMHFWLYH DOHUWQHVV UDWLQJV PD\ QRW QHFHVVDULO\ EH JRRG LQGLFDWRUV RI DWWHQWLRQ DOORFDWLRQ DELOLW\ ,Q RUGHU WR GHWHUPLQH WKH UHODWLRQVKLS EHWZHHQ VXEMHFWLYH UDWLQJV RI DOHUWQHVV DQG REMHFWLYH

PAGE 120

PHDVXUHV RI DWWHQWLRQ DOORFDWLRQ DELOLW\ IXWXUH UHVHDUFK VKRXOG FRPSDUH WKHVH YDULDEOHV IRU LQGLYLGXDOV ZRUNLQJ YDULRXV VKLIW V\VWHPV $OVR DV GLVFXVVHG HDUOLHU IXWXUH UHVHDUFK RI WKLV W\SH VKRXOG NHHS D FRQWLQXRXV UHFRUG RI WKH SDUWLFLSDQWVf VOHHSZDNH F\FOHV 7KH EHGWLPHV DQG ZDNHXS WLPHV VKRXOG EH ORJJHG IRU HDFK SDUWLFLSDQW GXULQJ WKH HQWLUH FRXUVH RI DQ LQYHVWLJDWLRQ LQ RUGHU WR GHWHUPLQH IDFWRUV VXFK DV LUUHJXODU VOHHSZDNH F\FOHV DQG WRWDO VOHHS REWDLQHG &KURQRSV\FKRORJ\ LV D IDLUO\ QHZ ILHOG RI VWXG\ DQG DV $GDQ f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f SHUIRUPDQFH RQ WKH MRE

PAGE 121

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

PAGE 122

WHUP PHPRU\ DQG UHDGLQJ UDWH DQG FRPSUHKHQVLRQ SHDN DW GLIIHUHQW WLPHV RI WKH GD\ LQ GLXUQDO VLWXDWLRQV )XWXUH VWXGLHV FRXOG EH GHVLJQHG WR GHWHUPLQH KRZ WKHVH IDFWRUV DUH DIIHFWHG E\ GLIIHUHQW VKLIW V\VWHPV )RU H[DPSOH JURXSV RI LQGLYLGXDOV RQ GLIIHUHQW VKLIW V\VWHPV FRXOG EH JLYHQ GLIIHUHQW W\SHV RI OHDUQLQJ WDVNV SV\FKRPRWRU PHPRUL]DWLRQ FRPSUHKHQVLRQ HWFf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f UHDFWLRQ WLPH IRU WKH VSDWLDO YLVXDOL]DWLRQ WDVNf DQG WUDFNLQJ HUURU GXULQJ WKH WUDFNLQJ WDVNf ZHUH QRW FRPSDUHG EHWZHHQ WKH WZR JURXSV LQ WKLV VWXG\ EHFDXVH WKHVH YDULDEOHV FRQFHUQ G\QDPLF SHUIRUPDQFH DELOLW\ 7KLV VWXG\ ZDV PDLQO\ FRQFHUQHG ZLWK WKH OHDUQLQJFXUYH GDWD RQ WKHVH YDULDEOHV LQ RUGHU WR GHWHUPLQH LI DQ\ OHDUQLQJ GLIIHUHQFHV FRXOG EH IRXQG WR H[LVW EHWZHHQ WKH JURXSV $V LQGLFDWHG LQ WKH UHVXOWV QR OHDUQLQJ GLIIHUHQFHV ZHUH LQGLFDWHG +RZHYHU WKHVH W\SH RI SHUIRUPDQFH

PAGE 123

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

PAGE 124

GHILFLHQFLHV GXULQJ WUDLQLQJ PD\ KDYH EHHQ LQYROYHG ,W PD\ EH SRVVLEOH WR GHWHUPLQH WKLV E\ UHYLHZLQJ WKH UHDVRQV OLVWHG IRU WKH HUURUV RQ WKH DFFLGHQW UHSRUWV 6RPH RI WKH UHDVRQV OLVWHG VXFK DV fEUDLQ ORFNf FRXOG EH DVVRFLDWHG ZLWK FRJQLWLYH FRQIOLFW FDXVHG E\ WKH UHWULHYDO RI LQDSSURSULDWH SURFHGXUHV RU WKH DXWRPDWLRQ RI IDXOW\ PHQWDO PRGHOV RU VFULSWV IRU WKH VLWXDWLRQ DW KDQG 3V\FKRORJLVW PD\ EH DEOH WR GHFLSKHU WKHVH WHUPV DQG DVVRFLDWH WKHP ZLWK SRVVLEOH FRJQLWLYH FDXVHV 7KHVH IDXOW\ PHQWDO PRGHOV RU VFULSWV FRXOG KDYH GHYHORSHG DV WKH UHVXOW RI DQ DWWHQWLRQ GHFUHPHQW GXULQJ WUDLQLQJ RQO\ WR EH JHQHUDWHG ODWHU LQ G\QDPLF FRQWURO VLWXDWLRQV WKURXJK DXWRPDWLF SURFHVVLQJ DXWRPDWLRQf ZLWK OLWWOH RU QR FRQVFLRXV DZVDUHQHVV RI WKHP 3UHVHQWO\ VLWXDWLRQ DZDUHQHVV LV FRQVLGHUHG WR EH WKH SULPDU\ YDULDEOH LQYROYHG LQ FRQWUROOHU HUURU EXW VLWXDWLRQ DZDUHQHVV LV D EURDG WHUP DQG LW LQYROYHV PDQ\ IDFWRUV LQFOXGLQJ WUDLQLQJ (QGVOH\ f )XUWKHU UHVHDUFK LQ VLWXDWLRQ DZDUHQHVV LQFOXGLQJ WKH W\SH GHVFULEHG KHUH ZLOO KHOS WR LVRODWH DQG LGHQWLI\ WKH YDULDEOHV ZKLFK DIIHFW VLWXDWLRQ DZDUHQHVV 2QFH WKHVH YDULDEOHV KDYH EHHQ LGHQWLILHG SUHGLFWLYH SODQQLQJ PHDVXUHV PD\ EH WDNHQ WR DGGUHVV WKHP HDUO\ DQG DYRLG WKH QHJDWLYH FRQVHTXHQFHV RI QRW DGGUHVVLQJ WKHP EHIRUH WKH\ EHFRPH IDFWRUV LQ G\QDPLF VLWXDWLRQV

PAGE 125

5()(5(1&(6 $GDQ $ f &LUFDGLDQ YDULDWLRQV LQ SV\FKRORJLFDO PHDVXUHV $ QHZ FODVVLILFDWLRQ &KURQRELRORJLD $NHUVWHGW 7 f 6OHHSLQHVV DV D FRQVHTXHQFH RI VKLIW ZRUN 6OHHS f $NHUVWHGW 7f t *LOOEHUJ 0 f 7KH FLUFDGLDQ YDULDWLRQ RI H[SHULPDHQWDOO\ GLVSODFH VOHHS 6OHHS f $NHUVWHGW 7 t .HFNOXQG f 6WDELOLW\ RI GD\ DQG QLJKW VOHHS $ WZR\HDU IROORZXS RI ((* SDUDPHWHUV LQ WKUHHVKLIW ZRUNHUV 6OHHS f $NHUVWHGW 7 t 7RUVYDOO / f ([SHULPHQWDO FKDQJHV LQ VKLIW VFKHGXOHVWKHLU HIIHFWV RQ ZHOOEHLQJ (UJRQRPLFV $NHUVWHGW 7 7RUVYDOO / t )UREHUJ ( f $ TXHVWLRQQDLUH VWXG\ RI LUUHJXODU ZRUN KRXUV DQG VOHHSZDNH GLVWXUEDQFHV 6OHHS 5HVHDUFK %DENRII +f &DVS\ 7f t 0LNXOLQFHU 0 f 6XEMHFWLYH VOHHSLQHVV UDWLQJV 7KH HIIHFWV RI VOHHS GHSULYDWLRQ FLUFDGLDQ UK\WKPLFLW\ DQG FRJQLWLYH SHUIRUPDQFH 6OHHS f %HQVRQ t )HLQEHUJ f 6OHHS DQG PHPRU\ 5HWHQWLRQ DQG KRXUV DIWHU LQLWLDO OHDUQLQJ 3V\FKRSK\VLRORJ\ f %LOOLDUG 0 $OSHURYLWFK $ 3HURW & t -DPPHV $ f ([FHVVLYH GD\WLPH VRPQROHQFH LQ \RXQJ PHQ 3UHYDOHQFH DQG FRQWULEXWLQJ IDFWRUV 6OHHS f

PAGE 126

&RIIH\ / &f 6NLSSHU .f -Uf t -XQJ ) f 1XUVHV DQG VKLIW ZRUN (IIHFWV RQ MRE SHUIRUPDQFH DQG MREUHODWHG VWUHVV MRXUQDO RI $GYDQFHG 1XUVLQJ &RKHQ f 6WDWLVWLFDO SRZHU DQDO\VLV IRU WKH EHKDYLRUDO VFLHQFHV QG HGO +LOOVGDOH 1/DZUHQFH (UOEDXP 'DKOJUHQ f $GMXVWPHQW RI FLUFDGLDQ UK\WKPV DQG ((* VOHHS IXQFWLRQV WR GD\ DQG QLJKW VOHHS DPRQJ SHUPDQHQW QLJKW ZRUNHUV DQG URWDWLQJ VKLIW ZRUNHUV 3V\FKRSK\VLRORJ\ 'DQLHO -f t 3RWDVRYD $ f 2UDO WHPSHUDWXUH DQG SHUIRUPDQFH LQ K DQG K VKLIWV (UJRQRPLFV f 'LQJHV ) f 7KH LQIOXHQFH RI WKH KXPDQ FLUFDGLDQ WLPHNHHSLQJ V\VWHP RQ VOHHS ,Q 0 + .U\JHU 7 5RWK t : & 'HPHQW (GVf 3ULQFLSOHV DQG SUDFWLFH RI VOHHS PHGLFLQH SS f 3KLODGHOSKLD 6DQGHUV (QGVOH\ 0 5 f 7RZDUG D WKHRU\ RI VLWXDWLRQ DZDUHQHVV LQ G\QDPLF V\VWHPV +XPDQ )DFWRUV f (QJOXQG & ( f 7KH GLXUQDO IXQFWLRQ RI UHDGLQJ UDWH FRPSUHKHQVLRQ DQG HIILFLHQF\ &KURQRELRORJLDf )RONDUG 6 t &RQGRQ 5 f 1LJKW VKLIW SDUDO\VLV ([SHULHQWLD )RONDUG 6f t &RQGRQ 5 f 1LJKW VKLIW SDUDO\VLV LQ DLU WUDIILF FRQWURO RIILFHUV (UJRQRPLFV f )RONDUG 6 .QDXWK 3 0RQN 7 +f t 5XWHQIUDQ] f 7KH HIIHFW RI PHPRU\ ORDG RQ WKH FLUFDGLDQ YDULDWLRQ LQ SHUIRUPDQFH HIILFLHQF\ XQGHU D UDSLGO\ URWDWLQJ VKLIW V\VWHP (UJRQRPLFV f )RONDUG 6 0LQRUV 'f t :DWHUKRXVH f &KURQRELRORJ\ DQG VKLIW ZRUN &XUUHQW LVVXHV DQG WUHQGV &KURQRELRORJLD

PAGE 127

)RONDUG 6 t 0RQN 7 + f 6KLIWZRUN DQG SHUIRUPDQFH +XPDQ )DFWRUV )RONDUG 6 t 0RQN 7 + f &LUFDGLDQ UK\WKPV LQ KXPDQ PHPRU\ %ULWLVK ORXUQDO RI 3V\FKRORJ\ )RONDUG 6 0RQN 7 +f t /REEDQ 0 & f 6KRUW DQG ORQJ WHUP DGMXVWPHQW RI FLUFDGLDQ UK\WKPV LQ fSHUPDQHQWf QLJKW QXUVHV (UJRQRPLFV *DQGHU 3 + 0\KUH *UDHEHU 5 & $QGHUVRQ + 7f t /DXEHU $XJXVWf $GMXVWPHQW RI VOHHS DQG WKH FLUFDGLDQ WHPSHUDWXUH UK\WKP DIWHU IOLJKWV DFURVV QLQH WLPH ]RQHV $YLDWLRQ 6SDFH t (QYLURQPHQWDO 0HGLFLQH f +DZNLQV ) + f +XPDQ IDFWRUV LQ IOLJKW UHY HGf (QJODQG *RZHU 7HFKQLFDO 3UHVV +LJJLQV ( $f &KLOHV : 0F.HQ]LH 0f ,DSLHWUR 3 ) :LQJHW & 0 )XQNKRXVHU ( %XUU 0 9DXJKDQ $ t -HQQLQJV $ ( f 7KH HIIHFWV RI D KRXU VKLIW LQ WKH ZDNHVOHHS F\FOH RQ SK\VLRORJLFDO DQG ELRPHGLFDO UHVSRQVHV DQG RQ PXOWLSOH WDVN SHUIRUPDQFH 5HSRUW 1R )$$$0f :DVKLQJWRQ '& &LYLO $HURPHGLFDO ,QVWLWXWH 2IILFH RI $YLDWLRQ 0HGLFLQH )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ .OLHQ (f :HJPDQQ + 0 $WKDQDVVHQ +RKOZHFN + t .XNOLQVNL 3 0DUFKf $LU RSHUDWLRQV DQG FLUFDGLDQ SHUIRUPDQFH UK\WKPV $YLDWLRQ 6SDFH DQG (QYLURQPHQWDO 0HGLFLQH .QXWVVRQ $ -RKQVRQ % $NHUVWHGW 7 t 2UWK*RPHU -XO\ f ,QFUHDVHG ULVN RI LVFKDHPLF KHDUW GLVHDVH LQ VKLIW ZRUNHUV 7KH ODQFHW /XQD 7 )UHQFK t 0LWFKD / -DQXDU\f $ VWXG\ RI DLU WUDIILF FRQWUROOHU VKLIW ZRUN $YLDWLRQ 6SDFH DQG (QYLURQPHQWDO 0HGLFLQH f

PAGE 128

0DQEHU 5 %RRW]LQ 5f $FHED & t &DUVNDGRQ 0 f 7KH HIIHFWV RI UHJXODUL]LQJ VOHHSZDNH VFKHGXOHV RQ GD\WLPH VOHHSLQHVV 6OHHS f 0F$GDUDJK 5 0 6SULQJf +XPDQ FLUFDGLDQ UK\WKPV DQG WKH VKLIW ZRUN SUDFWLFHV RI DLU WUDIILF FRQWUROOHUV 7KH MRXUQDO RI $YLDWLRQ$HURVSDFH (GXFDWLRQ t 5HVHDUFK f 0HOWRQ & ( f 3K\VLRORJLFDO UHVSRQVHV WR XQYDU\LQJ VWHDGYf DQG VKLIWV 0LDPL ,QWHUQDWLRQDO )OLJKW 6HUYLFH 6WDWLRQ 5HSRUW 1R )$$$0f :DVKLQJWRQ '& &LYLO $HURPHGLFDO ,QVWLWXWH 2IILFH RI $YLDWLRQ 0HGLFLQH )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ 0HOWRQ & ( t %DUWDQRZLF] 5 6 f %LRORJLFDO UK\WKPV DQG URWDWLQJ VKLIW ZRUN 6RPH FRQVLGHUDWLRQV IRU DLU WUDIILF FRQWUROOHUV DQG PDQDJHUV 5HSUW 1R '27)$$$0f :DVKLQJWRQ '& &LYLO $HURPHGLFDO ,QVWLWXWH 2IILFH RI $YLDWLRQ 0HGLFLQH )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ 0LVWOEHUJHU 5 t 5XVDN % f 0HFKDQLVPV DQG PRGHOV RI WKH FLUFDGLDQ WLPHNHHSLQJ V\VWHP ,Q 0 + .U\JHU 7 5RWK t : & 'HPHQW HGVf 3ULQFLSOHV DQG SUDFWLFH RI VOHHS PHGLFLQH SS f 3KLODGHOSKLD 6DQGHUV 0RQN 7 + f $GYDQWDJHV DQG GLVDGYDQWDJHV RI UDSLGO\ URWDWLQJ VKLIW VFKHGXOHVD FLUFDGLDQ YLHZSRLQW +XPDQ )DFWRUV 0RQN 7 + f &LUFDGLDQ UK\WKPV LQ VXEMHFWLYH DFWLYDWLRQ PRRG DQG SHUIRUPDQFH HIILFLHQF\ ,Q 0 + .U\JHU 7 5RWK t : & 'HPHQW (GVf 3ULQFLSOHV DQG SUDFWLFH RI VOHHS PHGLFLQH SS f 3KLODGHOSKLD 6DQGHUV 0RQN 7 +f 0ROLQH 0 / t *UDHEHU 5 & $XJXVWf ,QGXFLQJ MHW ODJ LQ WKH ODERUDWRU\ 3DWWHUQV RI DGMXVWPHQW WR DQ DFXWH VKLIW LQ URXWLQH $YLDWLRQ 6SDFH DQG (QYLURQPHQWDO 0HGLFLQH f

PAGE 129

2f'RQQHOO 5 7KH 1RYD6FDQ WHVW SDUDGLJP 7KHRUHWLFDO EDVLV DQG YDOLGDWLRQ 'D\WRQ 2K 17, ,QF 3DOH\ 0 t 7HSDV f )DWLJXH DQG WKH VKLIWZRUNHU )LUHILJKWHUV ZRUNLQJ RQ D URWDWLQJ VKLIW VFKHGXOH +XPDQ )DFWRUV f 3DWNDL 3 $NHUVWHGW 7 t 3HWWHUVVRQ f )LHOG VWXGLHV RI VKLIW ZRUN 7HPSRUDO SDWWHUQV LQ SV\FKRSK\VLRORJLFDO DFWLYDWLRQ LQ SHUPDQHQW QLJKW ZRUNHUV (UJRQRPLFV 3LOFKHU t +XIIFXWW $ f (IIHFWV RI VOHHS GHSULYDWLRQ RQ SHUIRUPDQFH $ PHWDDQDO\VLV 6OHHS f 5DQNLQ 0 / /DWKDP 3HWHUV 5 t 3HQHWDU 0 f 7KH HIIHFWV RI KRXUV WRWDO VOHHS GHSULYDWLRQ RQ KXPDQ SK\VLRORJ\ PRRG DQG PHPRU\ 3URFHHGLQJV RI WKH +XPDQ )DFWRUV 6RFLHW\ UG $QXDO 0HHWLQJ 5HGGLQJ ( 5 f $QDO\VLV RI RSHUDWLRQDO HUURUV DQG ZRUNORDG LQ DLU WUDIILF FRQWURO 3URFHHGLQJV RI WKH +XPDQ )DFWRUV 6RFLHW\ WK $QQXDO 0HHWLQJ 7UDLQLQJ $QDO\]LQJ :RUNORDG 0RGHOLQJ ([SHUWLVH DQG 0DLQWDLQLQJ 0RWLYDWLRQ 5HLQEHUJ $ 0RWRKDVKL < %RXUGHOHDX 3 7RXLWRX < 1RXJXLHU 1RXJXLHU /HYL ) t 1LFRODL $ f ,QWHUQDO GHV\QFKURQL]DWLRQ RI FLUFDGLDQ UK\WKPV DQG WROHUDQFH RI VKLIW ZRUN &KURQRELRORJLD 5LEDN $VKNHQD]L ( .OHSILVK $ $YJDU 'f 7DOO .DOOQHU %f t 1R\PDQ < 'HFHPEHUf 'LXUQDO UK\WKPLFLW\ DQG DLU IRUFH IOLJKW DFFLGHQWV GXH WR SLORW HUURU $YLDWLRQ 6SDFH DQG (QYLURQPHQWDO 0HGLFLQH 5RVD 5 5 t %RQQHW 0 + f 3HUIRUPDQFH DQG DOHUWQHVV RQ K DQG K URWDWLQJ VKLIWV DW D QDWXUDO JDV XWLOLW\ (UJRQRPLFV ef

PAGE 130

5RVHNLQG 0 5 *DQGHU 3 + 0LOOHU / *UHJRU\ % 0F1DOO\ / 6PLWK 5 0 t /HEDFT] 9 f 1$6$ $PHV )DWLJXH &RXQWHUPHDVXUHV 3URJUDP )$$ $YLDWLRQ 6DIHW\ ORXUQDO f 6DLWR < = &RUQHOLVVHQ 6RQNRZVN\ 5 6DLWR < 6DLWR < 6DLWR :X +LOOPDQ :DQJ = +DWD < 7DPXUD +DOEHUJ (f t +DOEHUJ ) f *UDGXDO DGMXVWPHQW RI FLUFDVHSWDQFLUFDGLDQ EORRG SUHVVXUH DQG KHDUW UDWH UK\WKPV DIWHU D WUDQVPHULGLDQ IOLJKW &KURQRELRORJLD 6DOGLYDU 7 +RIIPDQ 6 0f t 0HOWRQ & ( )HEUXDU\f 6OHHQ LQ DLU WUDIILF FRQWUROOHUV 5HSRUW 1R )$$$0f :DVKLQJWRQ '& &LYLO $HURPHGLFDO ,QVWLWXWH 2IILFH RI $YLDWLRQ 0HGLFLQH )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ 6KDGPHKU 5 t +ROFRPE + $XJXVW f 1HXUDO FRUUHODWHV RI PRWRU PHPRU\ FRQVROLGDWLRQ 6FLHQFH 6PLWK 5 & 0HOWRQ & ( t 0F.HQ]LH 0 f $IIHFW DGMHFWLYH FKHFN OLVW DVVHVVPHQW RI PRRG YDULDWLRQV LQ DLU WUDIILF FRQWUROOHUV 5HSRUW 1R )$$$0f :DVKLQJWRQ '& &LYLO $HURPHGLFDO ,QVWLWXWH 2IILFH RI $YLDWLRQ 0HGLFLQH )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ 8 6 &RQJUHVV 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW 6HSWHPEHUf %LRORJLFDO 5K\WKPV ,PSOLFDWLRQV IRU WKH :RUNHU 27$ %$ :DVKLQJWRQ '& 8 6 *RYHUQPHQW 3ULQWLQJ 2IILFH 86 'HSDUWPHQW RI 7UDQVSRUWDWLRQ )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ $SULOf $GPLQLVWUDWRUV )DFW %RRN :DVKLQJWRQ '& $%& 86 'HSDUWPHQW RI 7UDQVSRUWDWLRQ )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ $LU 7UDIILF 5XOHV DQG 3URFHGXUHV 6HUYLFH )HEUXDU\ f )DFLOLW\ 2SHUDWLRQV DQG $GPLQLVWUDWLRQ 0 :DVKLQJWRQ '& 86 *RYHUQPHQW 3ULQWLQJ 2IILFH

PAGE 131

9HQHU .6]DER 6f t 0RRUH f 7KH HIIHFWV RI VKLIW ZRUN RQ JDVWURLQWHVWLQDO *,f IXQFWLRQ $ UHYLHZ &KURQRELRORJLD :LQJHW & 0 'H5RVKLD & : 0DUNOH\ & / t +ROOH\ & f $ UHYLHZ RI KXPDQ SK\VLRORJLFDO DQG SHUIRUPDQFH FKDQJHV DVVRFLDWHG ZLWK GHV\QFKURQRVLV RI ELRORJLFDO UK\WKPV $YLDWLRQ 6QDFH t (QYLURQPHQWDO 0HGLFLQH :ULJKW (f 9RJHO $ 6DPSVRQ % .QDSLN 3DWWRQ )f t 'DQLHOV : / )HEUXDU\f (IIHFWV RI WUDYHO DFURVV WLPH ]RQHV -HW/DJf RQ H[HUFLVH FDSDFLW\ DQG SHUIRUPDQFH $YLDWLRQ 6SDFH DQG (QYLURQPHQWDO 0HGLFLQH

PAGE 132

%,2*5$3+,&$/ 6.(7&+ 5D\PRQ 0LFKDHO 0F$GDUDJK ZDV ERUQ )HEUXDU\ LQ 6SULQJILHOG 2KLR +H VSHQW IRXU \HDUV DQG HLJKW PRQWKV LQ WKH 8QLWHG 6WDWHV $UP\ IURP WR VHUYLQJ DV DQ DLU WUDIILF FRQWUROOHU DQG D IOLJKW VLPXODWRU LQVWUXPHQW JURXQG LQVWUXFWRU +H FRPSOHWHG DQG UHFHLYHG KLV ILUVW %DFKHORU RI 6FLHQFH %6f GHJUHH LQ ELRORJ\ IURP &KULVWRSKHU 1HZSRUW 8QLYHUVLW\ LQ 5D\PRQ EHJDQ KLV FXUUHQW FDUHHU DV DQ DLU WUDIILF FRQWURO VSHFLDOLVW ZLWK WKH )HGHUDO $YLDWLRQ $GPLQLVWUDWLRQ )$$f LQ ,Q ZKLOH ZRUNLQJ ZLWK WKH )$$ KH FRPSOHWHG KLV VHFRQG %DFKHORU RI 6FLHQFH GHJUHH LQ $SSOLHG 6FLHQFH DQG 7HFKQRORJ\ %6$67f WKURXJK 7KRPDV $ (GLVRQ 6WDWH &ROOHJH 7UHQWRQ 1HZ -HUVH\ $IWHU FRPSOHWLQJ WKLV VHFRQG %DFKHORUfV GHJUHH KH EHJDQ JUDGXDWH VFKRRO DW (PEU\5LGGOH $HURQDXWLFDO 8QLYHUVLW\ DQG FRPSOHWHG D 0DVWHU RI $HURQDXWLFDO 6FLHQFH 0$6f GHJUHH LQ ,Q $XJXVW RI KH ZDV DGPLWWHG WR WKH 8QLYHUVLW\ RI )ORULGD ZKHUH KH FRPSOHWHG KLV 3K' LQ HGXFDWLRQDO WHFKQRORJ\

PAGE 133

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

PAGE 134

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ /DUU\/RHVFK 3URIHVVRU RI &RXQVHORU (GXFDWLRQ FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ $VVLVWDQW 3URIHVVRU RI )RXQGDWLRQV RI (GXFDWLRQ 7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI (GXFDWLRQ DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 0D\ 'HDQ &ROOHJH RAIFGXFDWLRQ 'HDQ *UDGXDWH 6FKRRO

PAGE 135

/' 81,9(56,7< 2) )/25,'$


27
The circadian system is very complex and it interacts with the
physiology and psychology of humans and other species. According
to Mistlberger and Rusak (1989), circadian organization is a
pervasive, integral component of mammalian physiological systems,
which affects behavior in a great many ways. The underlying
mechanisms of this influence are themselves complex, involving a
hierarchy of oscillators that interact with each other and with other
regulatory systems. The circadian system is acutely sensitive to
photic cues in the environment which directly influence the
dominant pacemaker of the system, however, behavior may be
equally affected by other environmental variables, by influencing
other elements of the oscillatory hierarchy (p. 150).
Circadian Dvsrhvthmia
Circadian dysrhythmia is a condition which describes the
internal disassociation of the biological and/or psychological rhythms
of an individual (Hawkins, 1987). As stated earlier, this condition is
primarily the result of transmeridian flight (Winget et al., 1984) or
rotating shift work (Folkard & Monk, 1979).
Several researchers have shown that many circadian variables
are affected by acute shifts in routine (Akerstedt & Gillberg, 1980;
Gander et al., 1989; Higgins et al., 1975; Monk et al., 1988; Saito et
al., 1992). Most of these studies indicate that, once circadian


77
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of reaction time on Task #1 non
transition object orientation (which indicates spatial visualization).
7. Air traffic controllers who work rapidly-rotating work schedules
will indicate a mean learning curve (slope and its residuals) which is
not significantly different from the mean learning curve of
controllers who work day-shift and/or evening-shift schedules for
their NovaScan repeated measures of tracking error on Task #2
(which indicates mind/motor coordination).
Variables
The following criterian variables were measured and analyzed:
1. Attention allocation during Task 1 (spatial visualization).
2. Attention allocation during Task 2 (tracking).
3. Reaction time to transition from Task 2 to Task 1 (response
flexibility in switching resources),
4. Non-transition reaction time to Task 1 (spatial visualization).
5. Tracking error during Task 2 (tracking).
Data Analysis
In order to answer the hypotheses, the data were analyzed so
that the two groups of participants could be compared on two types
of variables: attention allocation, and learning. This was
accomplished in the following manner:


57
These and other studies tend to demonstrate some differential
effects of shift work schedules, but, regardless of the type of
schedules involved, shift work in general has been shown to cause
social and domestic disruption for shift workers, which in turn can
affect their performance and possibly have costly consequences in
industrial settings.
Work schedules may induce stress by preventing the worker
from fulfilling important family roles (Walker, 1985). Social
companionships, parenting, and sexual partner roles can all be
compromised by work schedules. These effects may be major
and can severely affect mood, motivation, and sleep, therefore
having indirect effects on performance and safety. Marital
problems, excessive domestic load, and community alienation
have all been documented as a result of the strain placed on
workers by work schedules. (U.S. Congress, Office of Technology
Assessment, 1991, p. 93)
Circadian Rhythms and Learning
Instructional design is based on what is known about learning
theory, information technology, systematic analysis, and
management methods. At the lesson and course levels, strategies are
designed so that instruction accommodates the learner. If learning is
to take place during instruction, these strategies must include
planning which considers both the instruction, and the individual
learner. If the learner is not receptive to the instruction for any
reason, learning will not take place. Learning is actually a two-way
process. Instruction is only effective when the instructional


28
rhythms are desynchronized, at least 7 days to several days are
required for resynchronization to occur. Until resynchronization does
occur, an individual will experience biological and psychological
decrements which are reflected in psychomotor and cognitive
performance (U.S. Congress, Office of Technology Assessment, 1991).
Higgins et al. (1975) conducted a study involving 15 male
participants (ages 20 to 28) in order to determine the effects of a
12-hour shift in the wake-sleep cycle on physiological and
psychological circadian rhythms. The results of this study indicated
that the quantity and quality of sleep did not change to a significant
degree after the sleep cycle was altered. The subjective fatigue
index, likewise, indicated that the total fatigue for the awake periods
was not significantly changed. However, the times of greatest fatigue
within days were altered, and complete reversal of the daily pattern
required nine days.
The physiological parameters which made the most rapid
response to stress were also the same parameters which required the
shortest period of time to rephase after the shift in routine. The
physiological parameter requiring the shortest rephasal time was
heart rate, followed in sequence by norepinephrine, epinephrine,
potassium, sodium, internal body temperature, and 17-ketogenic
steroids.


54
correlated with body temperature and is best at night for rapidly-
rotating shift workers.
This finding apparently contrasts with some of the earlier
studies concerning the variation in performance during normal
waking hours, which found that performance on most tasks tends to
improve over the day, while short-term memory tasks performance
has a tendency to decrease (p. 21).
More recent research has indicated that the concept of a single
performance rhythm is erroneous and that, like physiological
rhythms, performance rhythms differ not only in their normal
phase but also in the degree to which they are influenced by
exogenous factors (Folkard et al., 1984; Monk et al., 1983).
Indeed, there is evidence that memory-loaded, cognitive tasks,
which are becoming increasingly common in industry, may be
performed particularly well at night provided there is little
adjustment of the individuals circadian rhythms (Folkard et al.,
1976; Folkard & Monk, 1980). Further, the adjustment of this
type of circadian rhythm which peaks at night occurs relatively
rapidly (Hughes & Folkard, 1976; Monk et al.,1978). Such
adjustment will result in an impairment of night-shift
performance and so suggests that, for this type of task, shift
systems that minimize adjustment (i.e. rapidly-rotating shift
systems) may be preferable. It is thus noteworthy that the only
field of study to have found superior performance on the night
shift concerned the logging of errors of computer operators (a
task with a high memory load) on a rapidly-rotating shift system
(Monk& Embrey, 1981). (Folkard et al., 1985, p. 39)
Folkard points out that, from a performance point of view,
rapidly-rotating shift systems also have the advantage of minimizing
the cumulative sleep debt which can itself impair performance.
J


Summary'
69
This chapter has described human biological and psychological
circadian rhythms and how they are controlled by both endogenous
and exogenous factors. This chapter also described the detrimental
effects of circadian dysrhythmia, which are caused by jet lag or
rotating shift work, and the effects of circadian rhythms on learning.
The following chapter describes the methods and procedures used in
this study to collect and analyze the data, in order to address the
hypotheses presented in Chapter 1.


40
many controllers because it allows for longer intervals between work
weeks. About an 80-hour break, or 48% of the 7-day week, is
realized between work weeks with this shift rotation. This is
accomplished by compressing 40 hours of work into an 88-hour
period, with short intervals between each shift change (quick-turn-
arounds). Quick-turn-arounds usually take the form of an 8-hour
break between shifts.
A phase-delayed 1-2-2 schedule expands the work week with
longer periods of 18 to 22 hours between shifts (slow-turn-arounds).
These schedules start off with the midnight shift and progress
through later shifts over the week, ending on the evening shift. The
off-duty period at the end of the week is usually about 48-hours
duration.
In the case of either the 2-2-1 or the 1-2-2 rotation, however,
all shift schedules require 40 hours of work in a 5-day period with 8
hours work per shift. It is noteworthy that the expanded 1-2-2
rotation is not known to be in use in any industrial setting. It seems
that workers will accept almost any work schedule that will provide
a long break at the end of the work week. Some nuclear power
plants operate on 12- and 16-hour shifts. Many police officers work
12-hour shifts for 4 days, and then have 4 days off. One of the
perquisites (perks) of seniority in shift work is choice of work


21
internally by the organism and are not simply passive responses to
environmental stimuli (p. 142).
In the absence of environmental cues, daily rhythms are said
to be free-running. Free-running rhythms approximate the 24-
hour day, but not exactly. This periodicity (time required for one
complete rhythm cycle) is termed circadian (circa = about, dies =
day). Humans who are kept in isolation without time cues display a
25-hour sleep-wake cycle. Sleep onsets begin to occur at 25-hour
intervals instead of the normal 24-hour day-night intervals
experienced in ordinary society. Under this free-running cycle, as It
is termed, humans go to sleep later each day than they would under
normal circumstances.
Free-running sleep-wake cycles vary among species from
about 23 to 26 hours and are modified by factors such as ambient
light intensity, or hormone production. But, the fact that free-
running cycles differ from the normal 24-hour cycle is strong
evidence that circadian rhythms are generated internally, and are
not simply responses to daily 24-hour time cues. In fact, individuals
of the same species recorded under laboratory conditions in adjacent
cages demonstrate different periodicities which cause the animals to
slowly drift apart from one another, while sharing the same
environment (Mistlberger & Rusak, 1989, p.142).


116
baseline measures may prove to be of importance in future research
which is concerned with dynamic situation awareness. If differences
in dynamic performance are demonstrated between groups of
individuals working different shift systems, it could be concluded
that some shift systems have a chronopsychological effect on
dynamic situation awareness. Because of this, it is recommended
that these baseline measures are compared between groups who
work different shift systems in future research concerning dynamic
situation awareness.
Finally, it is recommended that research be conducted with
controller error statistics, using large sample si2es, in order to
determine if higher frequencies of errors are associated with
different shift systems. For example, a Chi-Square frequency
analysis could be conducted to determine controller error rates based
on controller shift system type for all ATC radar control facilities
nation-wide. The data used for this study could include all of the
controller errors committed at these facilities for the past three to
five years, depending on the availability of records. If a statistically-
significant frequency of errors is discovered to be associated with
any particular type of shift system, further analysis could then be
conducted to determine if the primary factor involved in the errors
was degraded job performance during the time of the error, or if


38
each week. On the other hand, schedules which rotate rapidly,
covering all shifts each week, allow biological rhythms to remain
diurnal, but can cause desynchronization of the
psychological/behavioral rhythms, as the habits and daily routine of
the individual adjust to accommodate each new work period during
the week (Folkard et aL, 1985; Paley & Tepas, 1994; U.S. Congress,
Office of Technology Assessment, 1991). Researchers have indicated
that sleepiness and fatigue which result from shift work are
detrimental to safety in many occupations (Akerstedt, 1988), and
that individuals who work shift work never adjust to it, even after
several years (Akerstedt & Kecklund, 1991).
Work Schedules
Shift work takes a variety of configurations in industry,
government, and the military, and it also varies according to the
tradition of the country or occupation in which it appears. It has
been estimated, however, that approximately one-fourth of the
working population in industrialized countries is employed on some
kind of shift work system (Akerstedt, 1988, p. 18).
The different types of rotating schedules in shift work each have
their own advantages and disadvantages. As previously stated,
Folkard et al. (1985), and Monk (1986) indicate that rapidly-rotating
shift systems allow the endogenous circadian system to remain


62
day, reaching the lowest point of accuracy at the same time of the
day as the peak in performance speed. Thus, while performance
speed is increasing over the day, performance accuracy is declining
(Monk, 1989, p.169).
\zolquhoun also conducted a study concerning speed in arithmetic
addition. This study concluded that, as the task became more
repetitive and simple, the time-of-day function in performance
speed became more like that of the temperature rhythm. Therefore,
time-of-day functions may change with practice.
For Monk (1989), these findings suggests that it may be wrong
to speak of performance as improving over the day, or to think of
time-of-day factors as representing changes in the brains capacity to
process information. Monk believes that we should consider changes
in performance as being mediated by changes in strategy over the
day. Individuals probably approach information processing tasks
differently at one time of the day, and at one level of practice, as
compared to another. Measures may indicate better or worse
performance depending upon what type of measure we take (p. 170).
Monk also suggests that research on short-term memory, which
indicates a decline over the day, may also be a result of changes in
information processing strategy, rather than changes in the brains
capacity to process information. In several studies (Folkard, 1979;


100
other learner characteristics which need to be addressed for the
purpose of maximizing learning.
This study was also deemed necessary because recent research
(Redding, 1992) determined that a lack of situation awareness is the
primary cause of controller error, and Endsley (1995) developed a
model of situation awareness which includes training as a
contributing factor to situation awareness in dynamic situations.
Each year, a number of controller errors, some of which lead to
aircraft accidents, may occur because controllers are affected by
circadian dysrhythmia (a misalignment of physiological and/or
psychological daily rhythms). If controllers who work certain shift
systems possess an attention or learning deficiency, which is a
resulting effect of the shift system they work, certain information or
procedures may be misinterpreted or overlooked during training
sessions. This could later contribute to at least some of the errors
which occur in dynamic control situations, while other controller
errors may occur due to the effects of circadian dysrhythmia which
are manifest during dynamic control situations. If training
treatments are devised to accommodate attention or learning
deficiencies during training, at least some controller errors may be
eliminated in the future.


three-shift workers (15 male/3 female), and Group 2 consisted of 19
day/evening two-shift workers (16 male/3 female).
Each participant completed 25 sessions on the NovaScan
computer-based performance test to determine two baseline
measures of attention allocation (one during a spatial-visualization
task and the other during a tracking task). The mean baseline
measures for the groups were compared using t-tests. The mean
learning curves for the groups were compared on five variables
(attention allocation during the spatial-visualization task, attention
allocation during the tracking task, response flexibility in switching
resources, non-transition reaction to spatial visualization, and
tracking error during tracking). The learning curves slopes and
residuals were compared using a two-way general linear model,
repeated-measures analysis, which included three statistical tests
(group main effect, time main effect, and group-time interaction).
The t-tests determined that the rapidly-rotating shift workers
have a significantly better attention allocation ability than the
day/evening shift workers on both variables measured. The learning
curve comparisons indicated that the rapidly-rotating shift workers
improved their attention allocation ability during the spatial
visualization task, while the day/evening shift workers did not, and
that the two groups stabilized around their means at different rates
on the learning curve for attention allocation during the tracking
vi