|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
THE ONSET AND EFFECT OF COGNITIVE FATIGUE ON SIMULATED SPORT
MELANIE BLYTH MOUS SEAU
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE INT
EXERCISE AND SPORT SCIENCES
UNIVERSITY OF FLORIDA
Melanie Blyth Mousseau
This document is dedicated to everyone who has helped me become who I am today, I
am forever grateful.
Although there is only one name directly associated with the following text, there
would be no text without the influence of numerous significant individuals. I would
therefore like to extend my gratitude to those who have made a significant impact on this
manuscript and on my development.
I first would like to thank my dedicated committee members, Dr. Christopher
Janelle, Dr. James Cauraugh, and Dr. Peter Giacobbi, for their time, support, and
insightful feedback. I would especially like to thank my advisor, Dr. Christopher Janelle,
for challenging me to achieve my fullest academic potential.
I also extend my gratitude to Steve Coombes for the countless hours of technical
development dedicated to this proj ect and for the troubleshooting that he and Aaron
Duley have provided me with. I also would like to offer my thanks to Derek Mann,
Jonathan Mousseau, Peter Papadogiannis, and Jason Galea for their ice hockey insight
More personally, I would like to recognize the unconditional support and patience
bestowed upon me by my family and their tolerance of the miles between us. For these
reasons, as well as a million others, I would like to thank my parents, Helene and John,
brother Jonathan, grandmothers, Jen and Helen, and the spirit of my grandfathers, Charlie
and John. Finally, I would like to thank Derek Mann for making me feel truly alive
TABLE OF CONTENTS
ACKNOWLEDGMENT S .............. .................... iv
LIST OF FIGURES ............ ...... ._ ..............viii...
AB STRAC T ................ .............. ix
1 INTRODUCTION ................. ...............1.......... ......
A attention .............. .. ...............2..
Attentional Distractors............... ...............
Visual Attention............... ...............3
Fatigue ............... ...............4....
Physical Fatigue................... ... ..............5
Cognitive Fatigue: A Self-Report Perspective .............. ...............7.....
Cognitive Fatigue: A Physiological Perspective ................ ........................8
Limitations of Previous Research ....__. ................. ...............10. ....
Statement of the Problem. ................. ...............12.._._._ ....
Statement of Purpose ...._.._ ................ ...............12......
H ypotheses.................. .............1
Subjective-Reports .............. ...............13....
T ask Performance ....._.._................. .......__. .......... 1
Behavioral Indices ................. ...............13..._._._ ......
Physiological Reactions............... ............... 1
Si gnificance .............. ...............14....
2 REVIEW OF LITERATURE ................. ......... ...............17. ....
A attention ................... .......... ................. 17....
Attention Metaphors ................. ...............18.................
Attentional Distractors............... ..............2
Visual Attention............... ...............2
Visual Attention in Sport ................. ...............25................
Eye movements in sport ................ .......... .. .......... ...........2
Effectiveness and efficiency in visual search patterns ................ ...............26
Attention: A Summary .............. ...............27....
F ati gue .............. ............... 27....
Conceptualizing Fatigue ................. ...............27.................
Fatigue Defined ................ ..... ....... ..........2
Mental fatigue and related cognitive states .............. ...............33....
Habituation ................. ...............33........ ......
M otivation .............. ...............33....
Boredom ................. ...............34........ .....
Fatigue and Human Performance .............. ...............34....
Exercise-Induced Fatigue and Performance ......__................. .................3 5
Cognitive Fatigue and Performance .............. ...............38....
Early investigations (1943 -1972) ................ ...............39..............
Contemporary investigations (996-2002) ................. ........................41
Limitations in the Literature ......... ................ ...............46 ....
Management of Fatigue .............. ...............47....
Conclusion ........._.._.. ...._... ...............48....
3 M ETHODS .............. ...............50....
Participants .............. ...............50....
Task. ........._._............_. ... ......_. ..._ .... ..........5
Measurement Devices and Dependent Variables .............. ...............53....
Gaze Behavior Measurement and Data Reduction ........._.._.._... ...._.. .........53
Arousal Measurement and Data Reduction............... ...............5
Heart rate ........._.._.. ...._... ...............54....
Skin conductivity............... ..............5
Task Performance: Response Time ...._.._.._ ...... ..... ...._.._._...........5
Task Performance: Response Accuracy .............. ...............55....
Subjective States: Visual Analog Scales .............. ...............55....
Procedure ........._.._.. ...._... ...............56....
Design and Analysis .............. ...............59....
4 RE SULT S .............. ...............61....
Correlates of Subj ective-Reports of Fatigue ................. .............. ......... .....61
T ask Performance ................. ...............62.................
Gaze Behavior ................. ...............62.......... .....
Arousal .............. ........ ...............62
Influence of Time-on-Task ................. .. .......... ... ... ........ ......... 6
Subj ective-Reports, Task Performance, and Physiological Reactions ................64
Gaze Behavior ................. ...............70......... .....
Summary of Results............... ...............7
5 DI SCUS SSION ....._.. ................. ........_.. .........7
Review of the Findings ................. ......._.. ...... .......... ........7
Correlates of Subj ective Ratings of Fatigue ................. ................. ........ 73
Influence of Time-on-Task on Dependent Measures .............. .....................7
Strengths, Limitations, and Directions for Future Research............... ................7
Applied Implications .............. ...............82....
Conclusion ................ ...............84.................
A DEMOGRAPHIC QUESTIONNAIRE............... .............8
B INFORMED CONSENT DOCUMENT .............. ...............88....
C INSTITUTIONAL REVIEW BOARD APPLICATION ................ ............... .....90
D TACTICAL QUESTIONS AND CORRECT RESPONSES .................. ...............92
E PARTICIPANT INSTRUCTIONS .............. ...............99....
F VISUAL ANALOG SCALE (VAS) .............. ...............103....
G DEBRIEFING SCRIPT ................. ...............104................
LIST OF REFERENCE S ................. ...............105................
BIOGRAPHICAL SKETCH ................. ...............111......... ......
LIST OF FIGURES
2-1. Graphical conceptualization of fatigue. .............. ...............32....
4-1. Subj ective fatigue by mean response time. .....__................ ............... ....6
4-2. Subj ective fatigue by heart rate. ........._._. ...._... ...............63....
4-3. Subj ective fatigue by skin conductance level ................. .............................63
4-4. Mean ratings of fatigue, boredom, and motivation by trial block. ................... .........64
4-5. Performance by trial block. ............. ...............65.....
4-6. Mean ratings of arousal by trial block. .............. ...............65....
4-7. Mean fixation duration by viewing. ............. ...............71.....
4-8. Significant relationship between fatigue and lost visual search data. .......................71
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science in Exercise and Sport Sciences
THE ONSET AND EFFECT OF COGNITIVE FATIGUE ON SIMULATED SPORT
Melanie Blyth Mousseau
Chair: Christopher M. Janelle
Major Department: Exercise and Sport Sciences
The sport of ice hockey requires refined anticipation, reaction, strategy, speed, and
decision-making skills within situations that are constantly changing. Competitive
athletic environments characterized by prolonged task exposure, high task difficulty, and
multiple task demands, like that of ice hockey, impose significant information processing
demands on the athlete. Such increases in mental workload can consume as well as
deplete the cognitive system's resources available for task completion, and promote the
development of fatigue.
Fatigue is characterized by a reduction in available resources and a decreased
ability to continue task performance at one's highest potential as a result of engagement
in either mental or physical tasks for a period of time without adequate rest. Specifically,
fatigue has been proposed to mediate attentional allocation, information processing, and
task performance under conditions of elevated and prolonged workload, situations
characteristic of sport competitions. Therefore, the purpose of the current investigation
was to determine the effects of varying levels of cognitive fatigue on ice hockey decision-
making effectiveness and efficiency. More specifically, the aim was to explore
individual decision time and accuracy as a function of time-on-task. A secondary purpose
was to determine if performance changes are related to variations in attentional focus, as
indicated by visual search patterns, and/or to fluctuations in arousal, as indexed by heart
rate and skin conductivity.
Eighteen male athletes (mean age = 20.6, SD = 2.57) with advanced ice hockey
experience participated in the experiment. The testing period consisted of 120 video
sequences, with each scenario being followed by a question concerning a tactical decision
with four multiple-choice responses. Participants were asked to select the most
appropriate response as quickly and as accurately as possible. Heart rate, skin
conductivity, gaze behavior, and subjective ratings of performance states were assessed.
Data supported the notion that time-on-task promotes the development of cognitive
fatigue and fluctuations in performance, as well as arousal. Specifically, fatigue was
related to faster response times and elevated arousal. However, response accuracy and
visual Eixation duration were not significantly related to time-on-task. In sum, the results
suggest that cognitive fatigue is distinctly manifested physiologically, subj ectively, and
behaviorally in athletes. Yet, continued research is necessary to gain a more complex and
accurate understanding of cognitive fatigue within sport. Strengths, limitations, future
directions, and applied implications are addressed.
The ability to accurately and effectively allocate attention is a critical factor in
determining performance outcome. Specifically within athletics, the importance of
appropriately directing and sustaining attention during a sporting competition has been
anecdotally confirmed by athletes for years; "concentration" and "focus" are emphasized
as critical factors in determining sport success. However, the task of appropriately
attending in the sport environment can be as challenging and taxing as the physical
exertion associated with performance. Yet, sport scientists have only recently begun to
explore how attentional fluctuations relate to sport performance.
As the field of sport psychology has progressed, greater consideration has been
directed to understanding the relationship between cognitive processes and athletic
performance. While numerous factors (i.e., strength, cardiovascular endurance,
experience) can contribute to performance fluctuations, recent sport psychology research
has placed an emphasis on understanding how internal factors (i.e., cognitions, emotions)
can influence performance. At present, however, minimal research has progressed to
examine cognitive fatigue as a potential performance mediator. The relationship between
mental fatigue and performance has not been adequately examined in athletics and
therefore fatigue's influence on attention and performance remains unclear. Anecdotal
evidence, however, would support the notion that fatigue does influence performance. It
is a frequent occurrence to hear athletes, coaches, and commentators attribute poor
performance to being mentally taxed/fatigued. For example, following a challenging
semifinal match at the Tennis Masters Cup, both, tennis star Andy Roddick and reporters
attributed the atypical challenge to a combination of mental and physical fatigue.
However, in order to gain a more complete understanding of sport performance, sport
psychology researchers must explore how cognitive fatigue develops in sporting
competitions, and how the resulting fatigue influences performance.
The scientific study of attention has a strong foundation in the general psychology
literature. Over a century ago, William James (1890) markedly defined attention as the
"the taking possession by the mind, in clear and vivid form, of one out of what seem
several simultaneously possible obj ects or trains of thought. It implies withdrawal from
some things in order to deal effectively with others" (p. 403-404). It has been well noted
that successful sport performance is reliant upon an athlete's ability to appropriately
select features of the environment or him/herself to attend to, and then sustain attention
on the relevant information for an appropriate duration (Cox, 1994). For athletes, the
ability to sustain attention and concentrate on selected task related features are often the
most important components of successful performance (Nideffer, 1993). However,
attention can become dispersed or compromised as a result of features of the environment
and/or internal states.
Attending to mass amounts of environmental stimuli can prove to be detrimental
to performance; too much attention is given to irrelevant information and too little
attention is given to significant cues, resulting in a performance decrement. Maintaining
appropriate focus can be a daunting task since nearly anything can serve as a distractor.
Certain conditions (i.e., high emotionality, depleted cognitive resources) increase the
tendency to redirect attentional focus away from the central task and onto irrelevant and
potentially distracting stimuli (Wegner, 1994). Under conditions of depleted resources
(i.e., anxiety, fatigue), distractors compete with relevant cues for already diminished
assets, resulting in the allocation of fewer resources for task performance (Moran, 1996).
The elements that interfere with or redirect attention away from relevant cues can occur
as either external (i.e., crowd noise, temperature) or internal distractors (i.e., one's own
thoughts, subjective feelings of fatigue) (Moran, 1996).
Within the laboratory, shifts in attentional direction to external distracting cues or
irrelevant information can be monitored in a variety of ways. One popular means of
assessing allocation of visual attention is through the use of eye tracking systems. The
location of a visual gaze is typically assumed to index the focus of attention (Duchowski,
2002). When an area of the environment is Eixated upon, the observer is gathering the
most detailed and highest quality information, while gathering relatively less detailed
information from surrounding peripheral areas. In addition to visual Eixation locations,
the search pattern employed will also dictate the efficiency by which information is
extracted for task completion (Williams, 2000). Efficient and successful performance is
often characterized by visual search patterns that involve fewer fixations of longer
duration (Williams, Davids, & Williams, 1999). This approach allows for more
information to be extracted with a single Eixation, rather than requiring several saccadic
movements, during which visual acuity is decreased. The exploration of attention and
attentional lapses has been furthered by the ability to track eye movements and visual
Visual gaze patterns provide an index of attentional focus and how search patterns
and attention can fluctuate under various conditions (i.e., fatigue, anxiety, high
workload). Within a sport performance, attentional focus is often required to be rapidly
redirected, increasing the demands that are placed on the cognitive system and the overall
level of mental workload (Matthews, Davies, Westerman, & Stammers, 2000).
Specifically, competitive environments characterized by prolonged task exposure, high
task difficulty, and multiple task demands evoke an overall greater level of information
processing. Such increases in mental workload monopolize and deplete the cognitive
system's resources available for task completion, thereby initializing the development of
fatigue. Fatigue has been proposed to mediate attentional allocation, information
processing, and task performance under conditions of elevated and prolonged workload
(Matthews & Desmond, 2002).
Generally, fatigue can arise from engagement in both physical and mental activities
of various intensities and duration. The negative results or unpleasant experiences of
prolonged or intense behavior, regardless if the activity is fun or tedious, has been
defined as fatigue (Craig & Cooper, 1992). Unfortunately, Craig and Cooper's broad
conceptualization does not effectively capture the essence or breadth of fatigue. Indeed,
vague conceptualizations of fatigue, like the one above, have plagued the extant literature
on fatigue, regardless of the discipline from which the definitions have emanated.
Recently, Job and Dalziel (2001) proposed a comprehensive definition of fatigue as
"the state of an organism's muscles, viscera, or central nervous system, in which prior
physical activity and/or mental processing, in the absence of sufficient rest, results in
insufficient cellular capacity or systemwide energy to maintain the original levels of
activity and/or processing by using normal resources" (p. 469). In other words, fatigue is
characterized by a reduction in available resources and a decreased ability to continue
task performance at one's highest potential as a result of engagement in either mental or
physical tasks for a period of time without adequate rest. Additionally, this definition
effectively differentiates fatigue from other related constructs, namely habituation and
boredom. In contrast to fatigue, habituation will ensue regardless of rest and is
independent of available energy/resources (Reber, 1995). Similarly, boredom will persist
after rest unlike fatigue and is mediated by subj ective disinterest, rather than by resource
availability. Although there are variable formal definitions of fatigue, Job and Dalziel's
conceptualization most clearly defines fatigue and will therefore be used in this
investigation to operationalize fatigue.
Many investigations have sought to better understand fatigue's influence on human
performance by examining various components of task effectiveness and efficiency. The
fatigue research is diverse and is comprised of studies investigating the effects of both
physical and mental fatigue on the performance of both cognitive and manual tasks.
Although the absence of sufficient capacity or energy to execute a task intuitively
suggests that output/performance would be negatively influenced by a fatigue state, the
empirical findings are somewhat equivocal. The following synopsis highlights selected
findings related to physical and mental fatigue with respect to cognitive performance.
Physical fatigue has been assumed to result in decreased cognitive functioning. To
evaluate this assumption, Davey (1973) examined cognitive task performance following
an exercise bout. Specifically, after cycling for various durations, short-term memory
functioning was assessed through performance on a numeric-sequence identification task.
Results indicated an inverted-U pattern of performance with task performance being
increased after moderate amounts of exercise but hindered after prolonged exercise.
Similarly, Reilly and Smith (1984) investigated the effect of exercise intensity on
cognitive task performance. To induce physical fatigue, active male university students
pedaled on a cycle ergometer at 25, 40, 55, 70, and 85% VOzmax. During the exercise,
participants completed two mental arithmetic tasks for 60 seconds during each loading
interval. Although the tasks were completed while exercising, rather than post-exercise,
the results were similar to those obtained by Davey (1973); performance was better
following some physical activity than when at rest, yet performance deteriorated at the
extreme levels of exercise.
Using a similar procedure, Salmela and Ndoye (1986) examined cognitive
processing fluctuations during progressive exercise. They found that participants reacted
significantly faster to central stimuli than to stimuli in the extremes of the visual field as
exercise intensity and duration increased. The authors proposed that as exercise
increased in intensity, attentional focus narrowed, as was indicated by slower responses
to stimuli presented in the visual periphery.
Collectively, researchers have often assumed that endurance and/or high intensity
exercise will promote a fatigue state. Under this assumption, these studies (Davey, 1973;
Salmela & Ndoye, 1986; Reilly & Smith, 1984; Reilly & Smith, 1986; Davey, Thorpe, &
Williams, 2002) suggest that when the body physically fatigues, cognitive processes will
be influenced by the physiological reaction to exercise. But, as highlighted earlier, fatigue
can develop as a result of cognitive involvement (Bartley & Chute, 1947). Therefore,
how is mental task performance influenced when an individual is cognitively fatigued?
Cognitive Fatigue: A Self-Report Perspective
In contrast to physical fatigue, cognitive fatigue refers to the diminished mental
capacity and resources resulting from the demands placed on the cognitive system
through various types of mental work (i.e., decision making and concentration). Initial
interest and much of the contemporary research on cognitive fatigue has been stimulated
by a desire to understand military and ergonomic issues, while improving working
Bartlett' s (1943) seminal research was designed to better understand how
individuals perform after continuous, prolonged task exposure. Using an aircraft
simulator, behavioral and performance changes were monitored among pilots as time-on-
task progressed. Bartlett found that the longer the pilots were involved in the task, the
greater the decrease in timing accuracy. Likewise, fewer instruments in the display were
attended to. Such variations in performance have direct implications for safety and
efficiency. Specifically, by recognizing that prolonged periods of flying can promote the
development of fatigue and the resulting fatigue can impair performance, pilots and
others involved in related fields can develop schedules that minimize prolonged task
exposure and include periods of sufficient periods of rest.
Bartlett acknowledged that pilots are subj ected to prolonged conditions of sustained
attention and constant information processing, conditions that promote fluctuations in
performance. Recognizing the inherent, cognitively demanding characteristics of flying,
and the often erratic rest patterns of both commercial and military pilots, Morris and
Miller (1996) tested sleep deprived pilots using a flight simulator task. Results indicated
a positive correlation between subj ective reports of fatigue and performance errors.
Again, the negative influence of fatigue on performance was apparent.
Based on previous research and the potentially severe consequences of decreased
performance (i.e., fatal accidents), there has been a recent surge in fatigue research in the
transportation industry. For example, Matthews and Desmond (2002) recently examined
the influence of task-induced fatigue on simulated driving performance. Specifically,
participants were asked to follow another vehicle through a simulated course that was
comprised of both curved and straight roadways for six minutes (control condition). In
the fatigue condition, participants completed the task for 24 minutes, while also engaging
in a signal detection task. Prior to and following each driving session, participants
completed self-report assessments of their levels of fatigue symptoms, emotions,
motivation, effort, and cognition. A post-task comparison of the control and fatigue
groups' self-report and performance scores revealed that those who engaged in the 24-
minute task (fatigue group) indicated that they expended greater effort, yet exhibited
decreased performance. These findings further support the connection between cognitive
fatigue and impaired task completion.
Cognitive Fatigue: A Physiological Perspective
Cognitive fatigue is catalyzed through cognitive processes and often associated
with fluctuations in attention that can be physiologically assessed. In particular, tasks
requiring vigilance or sustained attention, or those that are monotonous have often been
associated with distinct patterns of physiological arousal. Heart rate (HR) and skin
resi stance/conductivity are measures of particular interest among studies investigating the
physiological changes associated with cognitive fatigue.
In their 1972 investigation, Dureman and Boden assessed the effects of four hours
of continuous simulated driving on general performance, subj ective ratings of fatigue,
and physiology (pulse, respiration, skin resistance, and muscular tension in the neck).
The primary finding was that as time-on-task increased, subjective fatigue and skin
resistance increased, while pulse rate and performance decreased. Similarly, Lal and
Craig (2002) recently examined the physiological changes that occurred in relation to a
fatiguing, two hour driving simulation (subj ective assessment of fatigue was used to
ensure fatigue induction). An analysis of pre and post-driving HR indicated a decrease
in HR at the completion of the drive.
Other extant research (Dureman & Boden, 1972; Lal & Craig, 2002; Moolenaar,
Desmond, Mascord, Starmer, Tattam, & Volkerts, 1999; Macchi, Boulos, Ranney,
Simmons, & Campbell, 2002) strongly implicates a triangular relationship between
arousal, fatigue, and performance, with suppressed physiological arousal believed to be
one of the primary factors associated with fatigue and decreased performance.
Specifically, elevated levels of workload arise from prolonged task exposure, high task
complexity/ difficulty, multiple task completion or a combination of the three (all are
scenarios relevant to athletic competition). The elevated workload and cognitive
demands prompt the development of fatigue (Matthews & Desmond, 2002). Therefore,
fewer resources are available for task completion because of the task characteristics
themselves and reduced levels of general arousal associated with fatigue states. In turn,
due to reduced resources and lower levels of arousal, attentional allocation and
information processing capabilities become "sub-optimal," unless resources are sacrificed
from other processes. However, a circular relationship often results; when fatigue
develops resources are compromised and arousal levels are reduced. As such, one must
work "harder" to maintain performance levels. In turn, more resources are needed,
thereby further escalating the development and consequences of fatigue. Because only a
general framework regarding this pattern of events exists, future research is warranted to
better understand the physiological changes associated with cognitive fatigue, and to
determine whether those physiological variations have a direct influence on information
processing, attentional allocation to environmental cues, and eventual performance.
Limitations of Previous Research
As highlighted previously, physiological fluctuations can often redirect attention,
resulting in performance fluctuations. Fatigue (physical and cognitive) has also been
shown to influence physiological states, yet relatively minimal work has examined the
triangular relationship between fatigue, attention, and performance in sport from a
The illusive nature of cognitive fatigue is apparent in the wide variety of
operational definitions of the concept and the overall lack of sport specific literature
addressing the topic. Generally speaking, ambiguous and erroneous definitions are
provided when speaking about fatigue; defining it in terms of its consequences, rather
than its origins (e.g., Brown, 1994). Theoretical inconsistencies have resulted in an
inability to clearly define a state as being fatigue.
An additional shortcoming of previous research is the assumption (e.g., Bartlett,
1943; Davey, 1973; Reilly & Smith, 1984) that fatigue has been induced when not
explicitly monitored (i.e., prolonged exercise, time-on-task). Additionally, in 1947,
Bartley and Chute emphasized that it was inappropriate and inaccurate to assess fatigue
solely through performance indices. However, researchers continue to infer cognitive
fatigue solely through performance scores (Davies, Shackleton, & Parasuraman, 1983).
Contemporary investigations of cognitive fatigue have primarily focused on fatigue
in ergonomic settings, highlighting transportation due to the grave numbers of fatalities
and accidents believed to be the result of fatigued drivers (e.g., Matthews & Desmond,
2002; Lal & Craig, 2002). Within much of the literature, fatigue is referenced as resulting
from a wide variety of factors (i.e., sleepiness, general nighttime impairments, time-on-
task) and is often recognized by fluctuations in performance. However, the assumption
that each type of "fatigue" is equivalent, and that performance scores can indicate the
presence or absence of fatigue, is erroneous. Moreover, boredom and habituation are two
psychological states that arise from similar, yet distinct conditions as fatigue. However,
having similar origins and effects as cognitive fatigue, the confusion between habituation,
boredom, and fatigue exasperates the inherent limitations associated with fatigue
Although there have been numerous investigations that have explored how
muscular fatigue influences both physical and cognitive sport performance, and how
cognitive fatigue and its symptoms are associated with impaired performance in
ergonomic settings, there remains in significant gap within the literature. Specifically,
how cognitive fatigue influences sport related mental tasks has been ignored.
It is apparent that the ability to appropriately and adequately process information is
critical to successful performance. Most sports (e.g., hockey, baseball, football, tennis)
require that participants are able to quickly and continuously use the environment and
prior knowledge to make performance decisions. Other sports impose different time
constraints and are more self-paced (e.g., golf, bowling, dart throwing), yet the cognitive
demands remain intense. The demands on attention and information processing that are
inherent in sport have the potential to induce cognitive fatigue states, yet have been
overlooked in competition scenarios. Therefore, the present investigation was designed
to evaluate the effect of fatigue on attention and performance in a dynamic sport
Statement of the Problem
The sport of ice hockey encompasses anticipation, reaction, strategy, speed, and
decision-making. With the average player moving at speeds approaching 12mph and
attending to obj ects moving at speeds exceeding 90mph in a confined area
(approximately 190 by 90 feet), the ice hockey player must be ready to respond to rapidly
presented cues and situations that are constantly changing. Such physical and mental
demands can promote the development of fatigue, a state characterized by reduced
resources and ability. Despite fatigue's likely influence on performance and how athletes
perform, train, and react, sport scientists have relatively ignored cognitive fatigue.
Indeed, the current sport science literature has not addressed how the cognitive demands
of sport influence the cognitions, behaviors, and physiology of athletes.
Statement of Purpose
The purpose of this investigation was to determine the effects of varying levels of
cognitive fatigue on ice hockey decision-making effectiveness and efficiency. More
specifically, the aim was to explore individual decision time and accuracy as a function
of time-on-task. A secondary purpose was to determine if performance changes were
related to variations in attentional focus, as indicated by visual search patterns, and/or to
fluctuations in arousal, as indexed by heart rate and skin conductivity.
The following hypotheses were based on previous theory and research Eindings
from the Hields of general psychology, human factors, and the sport sciences. The
hypotheses were designed to address the issues concerning cognitive fatigue onset in
sport and resulting performance effects, as well as the physiological and behavioral
correlates of time-on-task.
1. Visual analog ratings of fatigue will increase significantly as time-on-task
increases. Specifically, jinal subjective ratings of fatigue will be significantly
greater than initial fatigue ratings (Dureman & Boden, 1972).
2. Visual analog reports of motivation will not increase as time-on-task increases,
since the task rewards and demands will be maintained across trial blocks
(McMorris & Graydon, 1996).
3. Visual analog reports of boredom will not increase as time-on-task increases, due
to dynamic and engaging task demands (McMorris & Graydon, 1996).
4. Mean response time will significantly increase over the duration of the task.
Therefore, as time-on-task increases, response time will increase, indicating a
positive relationship between task length and performance (Macchi et al., 2002).
5. It is hypothesized that as subjective reports of fatigue increase (as reported on the
visual analog scale) and independent of time-on-task, response time will increase
(Macchi et al., 2002).
6. Response accuracy will decrease significantly across participants as time-on-task
increases, indicating a negative relationship between task duration and performance
accuracy (Lorist, Klein, Nieuwenhuis, De Jong, Mulder, & Meijman, 2000).
7. As subjective reports of fatigue increase (independent of time-on-task), response
accuracy is hypothesized to decrease (Lorist, Klein, Nieuwenhuis, De Jong,
Mulder, & Meijman, 2000).
8. As subjective reports of fatigue increase, mean fixation duration will decrease As
fatigue ratings increase, visual search patterns will become more inefficient, with
decreased fixation duration (Williams et al., 1999; Williams, 2000).
9. As time-on-task increases, the mean fixation duration will decrease (Williams et
al., 1999; Williams, 2000). Specifically, the mean fixation duration for the second
viewing of scenarios (TB 13-24) will be significantly shorter than first (TB 1-12).
10. As visual analog subjective reports of fatigue increase, the number of visual
fixations will increase. As subjective ratings of fatigue increase, more fixations
will be required to extract sufficient information from the visual scene (Williams et
al., 1999; Williams, 2000).
11. As time-on-task increases, the number of visual fixations will increase (Williams et
al., 1999; Williams, 2000). Specifically, the number of visual fixations for the
second viewing of scenarios (TB 13-24) will be significantly greater than first (TB
12. Heart rate will deecrease significantly as subj ective reports of fatigue increase.
Therefore, as participants become more fatigued their arousal levels will decrease
(Dureman & Boden, 1972; Lal & Craig, 2002).
13. As time-on-task increases, heart rate will deecrease (Dureman & Boden, 1972; Lal
& Craig, 2002).
14. .\km1 conductivity will decrease significantly as subjective reports offatigue
increase. It is hypothesized that as participants become more fatigued their arousal
levels will also decrease. (Dureman & Boden, 1972).
15. As time-on-task increases, skin conductivity will deecrease (Dureman & Boden,
Cognitive fatigue is a concern within many performance domains, however, at
present, the cognitive fatigue research is both disj pointed and removed from the sport
sciences. The general fatigue literature continues to be burdened with numerous
incomplete and inaccurate definitions of fatigue, and lacks a formal methodological
framework. Additionally, the influence of cognitive fatigue on performance has been
overlooked in the sport psychology literature. Yet within environments such as athletics
that emphasize performance quality and quantity, it is important to understand the factors
that mediate performance output, so that the occurrence of negative consequences (i.e.,
decreased performance quality) may be reduced.
Within sport, participants are often required to complete several tasks
simultaneously, including attending to relevant internal and external cues effectively.
Over a period of time, this responsibility can be fatiguing (Lal & Craig, 2002). It is
evident that athletes endure demanding physical conditions, while also being presented
with cognitively demanding tasks such as directing and allocating attention, anticipating
upcoming movement sequences, and making decisions. At the present, there is no
evidence to indicate how prolonged decision-making and resulting cognitive fatigue is
manifested physiologically, subjectively, or behaviorally in sport. Therefore, the specific
aim of the current study was to assess changes in athletes' performance, subjective
reports of performance states, and physiology in relation to time-on-task within a
simulated, dynamic, ice hockey scenario.
To achieve this aim, the current investigation implemented a multi-method
approach of assessment, combining the use of both subj ective (i.e., visual analog self
report of current states) and obj ective (i.e., heart rate and response time) measurements.
Within this assessment paradigm, fatigue and performance were examined as separate yet
potentially interacting constructs, with each being assessed independently, a feature of
the design that has been often overlooked in previous investigations. Additionally, this
investigation focused solely on fatigue resulting from prolonged work exposure, a
methodological oversight in previous fatigue research (Holding, 1983). To further
illuminate the onset and influence of fatigue, the two related psychological concepts of
motivation and boredom were also be assessed throughout the duration of the task. With
attention being given to the several methodological and conceptual shortcomings of
previous research, the current study attempted to present a multidimensional, sport-
specific perspective of cognitive fatigue.
The unification of the literature from several related, yet often isolated disciplines
form the foundation on which the current investigation is established. This body of
literature and resulting investigation represents a movement towards the greater
comprehension of the prevalence and influence of cognitive fatigue in sport. Through
greater knowledge of the characteristics and mechanisms of cognitive fatigue, especially
attention, athletes can increase self-awareness about cognitions, behaviors, and
physiology during performance and fatigue states. Self-awareness is a primary factor in
the development of self-control and the moderation of potentially negative consequences
(e.g., fatigue and anxiety) (Ravizza, 1993). By recognizing attention related performance
difficulties, athletes can actively engage in strategies to control attention and in turn
minimize the negative influence of fatigue on performance (Gopher, 1992).
REVIEW OF LITERATURE
This chapter presents a review of literature associated with fatigue. Within the
review, a survey of the theoretical and empirical literature associated with attention,
fatigue, and their interactions and associations with performance will be presented. A
concise review of the dominant perspectives of attention and the relevance of attention to
human performance will be presented first, followed by an introduction to the concept of
fatigue. The fatigue-performance relationship will then be addressed in two sections: one
highlighting physical fatigue, and the second focusing on cognitive fatigue, with an
exploration of the measurement techniques frequently used and the common limitations
of the research. The chapter will conclude with a synthesis of the presented material,
with future directions and practical applications being highlighted.
The term "attention" is used frequently in daily conversation and is a topic of
research in many disciplines. In particular, researchers concerned with understanding
human performance have attempted to identify the functions) of attention and how
attention is directed, allocated, and maintained.
The ability to allocate and sustain attention to relevant information is crucial to the
successful completion of both cognitive and motor tasks (Abernethy, 2001). Having such
a profound influence on human behavior, the concept of attention has been a topic of
research and theoretical debate in both psychology and motor behavior, with the first
psychological endeavors beginning in the later half of the 19th century. Continuing from
this point, the study of attention has proceeded through a series of paradigm shifts, each
resulting in the development of several theories, some of which will be discussed below.
James (1890) described attention as "the taking possession by the mind, in clear
and vivid form, of one out of what seem several simultaneously possible obj ects or trains
of thought. It implies withdrawal from some things in order to deal effectively with
others" (p. 403-404). In other words, attention can be viewed as the concentration of
mental activity through the allocation of cognitive resources to select processes so as not
to be interrupted or influenced by irrelevant stimuli (extemal or internal) (Schmid &
Peper, 1993). According to this definition, as well as more contemporary approaches,
attention has been viewed as selective in nature and limited in capacity (Cherry, 1953;
Kahneman, 1973). Specifically, attention enables individuals to focus on a particular
stimulus, task, or sensation, and disregard other incoming information. However, the
amount of information attended to at a time is limited. As the study of attention
progressed and the above concepts expanded on, attention was no longer viewed solely as
consciousness, as purported by James (1890), but through a variety of metaphors
(Femnandez-Duque & Johnson, 1999; Abemnethy, 2001).
Contemporary attention research has been primarily driven by the use of
metaphors, with two dominant perspectives, including notions of attention as a filter and
as a resource (Fernandez-Duque & Johnson, 1999). Broadbent (1958) initially advanced
the concept of attention as a filter through his single-channel theory. Broadbent assumed
that individuals have the ability to allocate attention to a specific task, while filtering
irrelevant and extraneous information from processing. Specifically, when two or more
stimuli occur simultaneously, all enter the sensory buffer (a process which does not
require attention), but only one stimulus can pass through the filter to be processed (an
attention demanding process). According to filter theories, attention is a single resource
that utilizes a filter in the early stages of information processing to prevent irrelevant
information from overloading the system's capacity and reaching long-term memory.
Broadbent' s general concept of a filter has been retained by subsequent researchers (e.g.,
Treisman, 1960) who acknowledge that irrelevant incoming information is blocked from
reaching further stages of information processing and memory by a filter, however the
point at which the filtering of information takes place varies among theorists.
Rather than attention being mediated by a filter, "resource" theorists view attention
as being controlled by its limited-capacity, either as a single resource or a combination of
several resources (Kahneman, 1973; Navon & Gopher, 1979). Kahneman (1973)
proposed that attention is a single, but flexible "resource pool" from which several tasks
may draw simultaneously, until the capacity of the pool is exhausted. Once the task
demands exceed the resources available, performance begins to decline. Using a similar
rationale, but proposing multiple "resource pools," Navon and Gopher (1979) attempted
to explain the successful performance of several tasks simultaneously. In many
situations, several tasks may be performed simultaneously without a noticeable
performance decrement as long as the tasks are drawing resources from different pools,
with the task demands dictating from which pools the resources will be drawn. The
collective premise of this family of theories is that individuals possess limited resources
for attentional processing and when these resources cannot meet the demands of the
task(s) at hand, a decrease in performance is seen, either in efficiency, output, or a
combination of the two (Kahneman, 1973; Desmond & Matthews, 1997). Additionally,
individuals have a specified amount of processing space or capacity that may be divided
among several simultaneous tasks. In other words, there is a limit on the quantity or
amount of attention that can be applied to a single task or several tasks at a time. In
summary, attention is limited in information processing capability, either by lack of
resources, space, or capacity (Abernethy, 2001).
Regardless if one uses a filter or a resource metaphor to describe attention, the
ability to orient attention is undeniably crucial to human performance. While
performance psychology is grounded in theories of cognition, attention, and information
processing, cognitive processes cannot directly be observed in naturalistic or research
settings (Matthews et al., 2000). However, performance variables such as response
accuracy and reaction time can be used to describe unobservable cognitive processes
(Matthews et al., 2000). Monitoring performance fluctuations in various environmental
contexts provides insight into the cognitive processes responsible for such modifications
and is important for understanding and modifying human performance. Through
investigations into human performance, awareness is gained into what factors and
conditions promote successful performance, as well as those that lessen the probability of
accurate and effective task performance as a result of disruptions in attention (e.g.,
distractors) and limitations in information processing and cognitive abilities (e.g., fatigue
and low arousal) (Matthews et al., 2000).
In performance scenarios, not only is it imperative for participants to attend to
appropriate cues, but also to maintain that focus for a sufficient duration (Moran, 1996).
However, distracting elements of the environment often compromise attentional focus,
redirecting attention away from relevant information.
Distractors are factors that prompt shifts in attentional focus, and often will
promote a performance decrement (Moran, 1996). Although an infinite number of
sources can serve as a distractor and are continuously present internally and in the
environment, there is an increased likelihood that they will be attended to in situations
when the cognitive system is already compromised. For example, in conditions of high
emotionality (e.g., high pressure situations) or depleted cognitive resources (e.g., fatigue
states), individuals express a greater tendency to redirect attention away from the central
task and to irrelevant and potentially detrimental stimuli (Wegner, 1994). In these
situations, distractors compete with relevant information for already diminished
resources, resulting in the allocation of even fewer resources to the central task (Moran,
Often in athletics when distractors cause breaks in attention and subsequent
performance deteriorates, the result is described as "choking." Common external
distractors in sport are noise, behaviors and tactics of opponents, in addition to other
various environmental and playing conditions (Moran, 1996). Similarly, physical
sensations (e.g., sweaty palms, tense muscles) may function as internal attentional
distractors for athletes (Nideffer, 1993). Specifically, when such feelings and sensations
are attended to and interpreted negatively, there is an increased probability for a decrease
in performance (Lazarus, 2000).
It is widely assumed that arousal and anxiety have a direct influence on attention
(Abernethy, 2001). Commonly, the responses of anxiety and arousal, coupled with
various physical symptoms redirect attention inward, away from the task relevant cues.
Specifically, anxiety has been empirically illustrated to influence primary task
performance by promoting shifts in attentional allocation and direction (Janelle, Singer,
& Williams, 1999). In a recent investigation, Janelle and colleagues (1999) utilized a
simulated driving task under varying levels of anxiety and task demands to empirically
validate the above assumption. Using a dual-task paradigm, with a central driving task
and a peripheral light detection task, distraction and attentional shifts were assessed
through visual search patterns. Results indicated that in an effort to compensate for a
reduced attentional Hield resulting from elevated levels of anxiety, individuals redirected
foveal fixations to the periphery to identify cues located beyond the central focal area.
During this period of redirection, however information located in the primary focal and
essential to successful primary task performance were disregarded. The findings of
Janelle and colleagues' driving simulation study indicate that anxiety modulates visual
search strategies and patterns, with the resulting modifications negatively impacting
central task performance.
Whereas anxiety and arousal's influence on attention has been explored in some
depth, a related concept, fatigue, has been relatively ignored in both the general and sport
psychology literatures. To gain more insight into the influence of fatigue on cognitive
processes and performance, visual search patterns can be monitored to index attentional
focus, as has been done already in reference to various internal states and external
environments (e.g., Janelle et al., 1999).
Fluctuations in visual attention have customarily been monitored through changes
in overt performance or through self-report (retrospective and concurrent) measures.
However, less subjective measures, such as more sophisticated eye tracking systems have
been developed to indicate where visual attention is being allocated. The basis of the
approach is that when the eye moves, a specific area of the visual display is brought into
finer detail in the fovea, a small, yet sensitive area of the eye. When visual stimuli are
fixated on (i.e., in the fovea), the information extracted is of highest quality and greatest
detail (Duchowski, 2002). Specifically, foveal fixations, areas of interest/point of regard,
and eye movement data can be acquired using a light reflection off the cornea, in
conjunction with a video image of the eye (Duchowski, 2002; Williams et al., 1999). The
ability to track eye movements has provided an additional avenue for the exploration of
attention, and specifically visual attention.
It is assumed that by tracking eye movements (i.e., fixation location, fixation
duration, and search rate), an individual's path of attentional allocation can also be
followed, and the cognitive process of visual attention can be assessed (Duchowski,
2002). Specifically, fixation locations can be used to discern where an individual is
visually attending and what environmental cues are significant in the decision-making
process, without reliance upon subj ective interpretation of where the individual reports
attending. Moreover, the duration and number of fixations made provides an index of the
cognitive demands (i.e., information-processing demands) of the task; the length of visual
fixations, as well as the total number of fixations highlight the importance, as well as the
complexity of each area of information extraction (Williams et al., 1999).
Although it is possible to direct attention to the visual periphery without a change
in visual fixation, it is the definite movements and foveal fixations that indicate the
direction of blatant visual attention and more detailed inspection by the viewer. This is
not to discredit the importance of information presented in the periphery; instead it is
often the material in the periphery that directs subsequent foveal fixations and focus of
attention. The use of concurrent verbal reports, in conjunction with eye movement data,
illustrates the success of and similarity between the two methods in identifying the locus
of visual attention (Williams & Davids, 1997). However, the association can be
moderated by both task characteristics and the performer' s level of expertise.
Specifically, among experts examining 3 versus 3 soccer scenarios, there was a
discrepancy in the focus of attention between self-report and eye movement data, an
inconsistency attributed to the expert' s tendency and ability to use peripheral vision to
extract relevant information from areas other than the fixation location. In sum, a general
association exists between self-report accounts of visual attention and eye movement
data, yet it is possible for skilled performers to maintain a fixation while extracting
important information from the periphery.
It is important, however, to highlight that cognitive processes and information
processing are key determinants, in addition to the visual display, in the orienting of
attention. Because of the involvement of higher-level cognitive processing in directing
attention, it is possible for individuals to voluntarily disconnect their attention from their
foveal fixation (i.e., not "seeing" what one is looking at). Although this limitation
associated with eye tracking is recognized, it is assumed that attention is directed at the
point of fixation. However, it is understood that it is possible for the point of gaze and
direction of attention to be detached. In conclusion, the addition of eye tracking
methodologies to the study of attention has provided a fresh perspective on how attention
can be monitored during task performance with relatively minimal disruption to "normal"
Visual Attention in Sport
Eye movements in sport
During sport performance, athletes primarily implement two styles of eye
movements, pursuit tracking and saccades, to bring information detected in the visual
periphery into the fovea (Williams et al., 1999). Pursuit tracking is a relatively slow and
smooth tracking style often associated with full head movement. Pursuit tracking is
typically utilized when a slow object is being followed through an environment (i.e., the
obj ect is followed continuously from point A to B). Because of the obvious time
requirement to obtain a stable retinal image, the utility and frequency of pursuit tracking
in fast paced sports is minimal and is typically used in slower self-paced tasks. In
contrast, saccades are rapid eye movements that are used to shift the eyes from one
position in the visual space to another (i.e., the fixation "jumps" from point A to B
0I without attending to all points in-between). During the saccadic movement itself (i.e., the
actual movement from point A to B) there is a decreased sensitivity to the visual
environment, prohibiting the extraction of visual information. Therefore, the fixation is
more important for information processing than the movement. Ballistic sport scenarios,
such as an ice hockey game, in which the visual environment is dynamic and under rapid
flux, require frequent saccadic eye movements and fixations for adequate information
In addition to implementing the appropriate eye movements to shift from points in
the display, it is essential that athletes acquire the information from the environment in
the most practical and efficient manner to ensure a rapid and accurate response. The
expert-novice athlete distinction is commonly used to demonstrate the differences
between effective/efficient search patterns and less effective/efficient approaches.
Effectiveness and efficiency in visual search patterns
The ability of an athlete to orient attention appropriately and to efficiently carry out
a movement is directly related to sport performance and outcome (Nideffer, 1993). The
search strategy employed by athletes will dictate the efficiency by which performance
will be carried out (Williams et al., 1999). For example, skilled performers generally use
fewer fixations of a longer duration on highly informative areas of a display in an attempt
to anticipate future actions. This process employed by experts is highly efficient since it
maximizes the utility of the display and the time available (Williams, 2000).
Additionally, elite athletes "expect the unexpected" in the most challenging and difficult
sport specific circumstances, typically orienting attention automatically to the most
important aspects of the visual display. Knowing where to and when to look is a crucial
aspect of successful sport performance. It is therefore imperative that players are able to
recognize the central and most information rich areas of the display and direct their
attention swiftly and appropriately (Williams et al., 1999).
The concept and measurement of visual attention has diverse utility and
application. Within the current investigation, visual gaze patterns and points of fixation
will be used as markers of attentional focus. The purpose of monitoring changes in visual
attention through eye tracking is to observe the potential fluctuations in search efficiency
(i.e., number of fixations and fixation duration) through the stages of development of
cognitive fatigue, a temporary state associated with problems in maintaining task-related
attention and effort during an on-going performance (Brown, 1994). The availability of
attentional resources and ample attentional capacity are central to successful human
performance, however, fluctuations in attention and information processing do occur.
Specifically, fatigue states are associated with reduced attentional capability and
increased distractibility (Moran, 1996; Desmond & Hancock, 2001).
Attention: A Summary
In summary, how attention is directed, allocated, and maintained is of primary
interest to researchers and practioners concerned with understanding human performance.
Various theories and metaphors have been developed to better conceptualize what
attention is and how it influences performances. Yet, regardless of the specific approach
taken (i.e., filter or resource), it is agreed that the amount of information that can be
attended to is limited (Fernandez-Duque & Johnson, 1999). In addition to its limited
nature, attention is also subject to negative disruptions caused by both internal and
external distractors (Moran, 1996). The probability that task irrelevant information will
cause a break in attention is amplified by states of high emotionality and already reduced
resources. Such fluctuations in attentional direction that result from environmental
changes, distractors, or reduced resources can be monitored physiologically through eye
tracking methodology. Specifically, gaze patterns and visual fixations provide obj ective
information about where an individual is "looking," while providing a reasonable
estimate of what they are "seeing." Therefore, by monitoring eye movements
("looking"), it can be inferred what an individual is attending to ("seeing") (Duchowski,
2002). By understanding the purpose of attention, the characteristics of attention, and
how attention can be measured, mediators of sport performance can be explored.
Fatigue is a term commonly used to describe a negative result or unpleasant
experience of a wide range of behaviors such as long distance running, lifting weights,
prolonged driving, monitoring a visual display, or solving logic problems. Sustaining
these activities, either positive or negative, for a few seconds to several weeks, can result
in the development of fatigue (Craig & Cooper, 1992). Specifically, during the
performance of cognitive and motor tasks, attentional demands are placed on the human
system, creating an amount of cognitive workload (Matthews et al., 2000). The
information processing required to successfully execute such tasks necessitates cognitive
"energy"/ "resources," yet after periods of prolonged task exposure, high task difficulty,
or multitasking, it is likely that such resources may become insufficient for optimal
performance. That is, the cognitive resources have been already consumed by other task
requirements and environmental demands. The minimization of available resources
prompts the development of a fatigue state that continues to reduce the overall working
capacity of the system, making the system (both the physical and cognitive components)
more susceptible to stress (Schoinpflug, 1983). Therefore, in conditions of elevated or
prolonged workload, fatigue becomes a mediator between information processing and
task performance (Matthews & Desmond, 2002). In environments, such as sport, that
require attention and alertness to successfully execute multiple tasks, understanding how
performers respond to the numerous attentional demands under stable and variable
conditions is essential for reducing the occurrence of negative performance consequences
(Lal & Craig, 2002).
To better understand what fatigue is and how it influences performance, a
working/operational definition of fatigue will be provided. Currently, numerous common
definitions of fatigue exist and a few of the most relevant and frequently referenced
examples will be briefly introduced because the current number of definitions is
extensive. Next, the "essential features" of fatigue will be described and contrasted with
the characteristics of related constructs. Finally, a contemporary view of the state of
empirical knowledge concerning fatigue will be provided. The discussion will highlight
the influence of physical fatigue on performance, while also focusing on cognitive fatigue
and its influence on performance as indicated by subj ective reports, physiology, and other
Within the common vocabulary fatigue has a broad connotation, with individuals
using the term to refer to feelings of being tired, overworked, and unmotivated. Having
such a broad range of reference in everyday language has the potential to create
confusion in the scientific literature. It is therefore essential to operationally
conceptualize and define fatigue. Regrettably, a universally accepted definition of fatigue
is lacking and the ambiguous and complex nature of fatigue is perpetuated by the lack of
a formal definition. As a result, a standard measurement of fatigue and fatigue states is
lacking (Fairclough, 2001). With the discrepancies pertaining to the conceptualization
and definition of fatigue, some researchers, even as early as the 1920s, proposed the
abandonment of such a concept of fatigue (Muscio, 1921). Muscio (1921) argued that no
test would validly be able to assess fatigue since there is not a definite observable
criterion except for that provided by the test itself. However, advancements have been
made in the conceptualization of fatigue so that abandonment of the concept is
Fatigue may be generated through two primary modes of exertion, either physical
or mental, each resulting in subj ective feelings of tiredness, with perhaps unique
influences on performance (Matthews et al., 2000). Additionally, fatigue can be either
active or passive in nature (Desmond & Hancock, 2001). During active fatigue, a
constant, inescapable demand is placed on attention. When fatigue develops during
situations of continuous demand, available attentional resources and the frequency by
which external sources are sampled is reduced. In contrast, passive fatigue is the result of
chronic understimulation (Desmond & Hancock, 2001). Although fatigue can be further
reduced and categorized from the general to the specific, it is essential that an
overarching definition of fatigue be presented.
Although all definitions of fatigue attempt to operationalize and clarify the same
construct, disparity among and omissions within many of the definitions of fatigue
remain. Craig and Cooper (1992) broadly define fatigue as "the weariness that accrues
from applying oneself to a task over a period of time" (p. 289), regardless of the setting
or the task. Recognizing the broad scope of this definition, the authors proceeded to limit
fatigue as a result of "the effects that stem from the continued exercise of an activity" (p.
290-291), thereby distinguishing fatigue from the tiredness encountered from eating,
drinking, or lack of sleeping. However, Craig and Cooper' s definition lacks sufficient
specificity and depth by failing to address how rest influences fatigue and what
constitutes "weariness" and therefore cannot be considered an adequate depiction of the
concept of fatigue.
Similar to Craig and Cooper, Brown (1994) failed to make reference to the cause or
nature of fatigue, defining psychological fatigue as a "subj ectively experienced
disinclination to continue the task" (p. 298). As a result of this lack of causal specificity,
such a definition could be interpreted as a lack of motivation, rather than fatigue (Job &
Dalziel, 2001). Conversely, Hancock and Verwey (1997) significantly narrowed the
scope of fatigue, defining it as an "individual's multi-dimensional physiological-
cognitive state associated with stimulus repetition which results in prolonged residence
beyond a zone of performance comfort" (p.497). By only highlighting stimulus
repetition as a cause of fatigue, other potential fatiguing agents such as constant
attentional demands, decision-making demands, or behavior repetition are disregarded.
Although the above definitions describe the same construct, a definite lack of consistency
exists among them; some definitions fail to recognize the two types of fatigue (physical
[peripheral] and mental [central]), while others fail to capture the essence of the concept
of fatigue, often being either too vague or too specific. The limitations that are inherent in
the presented definitions reflect a common problem within the fatigue literature that has
plagued the empirical investigation of fatigue.
In an attempt to clarify the ambiguity that has characterized the term "fatigue," Job
and Dalziel (2001) have recently set forth a series of "essential features" that all
definitions of fatigue should include. First, fatigue must be viewed as a hypothetical
construct; it is a state of the individual, not a feature of his/her behavior or a performance
outcome. In other words, reduced performance effectiveness or efficiency is not fatigue;
however, fatigue may lead to such results. An adequate definition should also identify
the cause, not solely the result of the state. Similarly, fatigue should include a description
of the conditions that arise in either the muscles or central nervous system that contribute
to the onset of fatigue. Next, definitions of fatigue should avoid extremely technical
language, so that the description fits within the general population's logical conception of
fatigue. Finally, the definition should be adequate so that fatigue can be distinguished
from other related phenomenon. Job and Dalziel (2001) integrated each of the critical
features and consequently defined fatigue as "the state of an organism's muscles, viscera,
or central nervous system, in which prior physical activity and/or mental processing, in
the absence of sufficient rest, results in insufficient cellular capacity or systemwide
energy to maintain the original levels of activity and/or processing by using normal
resources" (p. 469). Figure 2-1 presents a graphical conceptualization of fatigue, based
on the definition provided by Job and Dalziel (2001). The Eigure emphasizes the
significance of previous work and insufficient rest in the promotion of fatigue, which
results in a diminished ability to perform at previous standards and a reduction in
available resources, with a probable decrement in overt performance.
Figure 2-1. Graphical conceptualization of fatigue.
Although it is difficult to assert the precise influence of fatigue on performance
considering the prevalence of variable definitions and related constructs (e.g.,
habituation, motivation, boredom), it is assumed that information processing efficiency
and effectiveness may be compromised by fatigue (Matthews et al., 2000). In summary,
fatigue is the result of previous activity without sufficient rest that leads to inadequate
levels of energy or available resources to continue the activity with original levels of
effort. Additionally, the resulting fatigue state has the potential to increase the
probability for distractors to disrupt task-focused attention (Moran, 1996), redirect
attention to the subj ective symptoms of the state (Matthews et al., 2000), and hinder other
factors related to successful task performance.
Mental fatigue and related cognitive states
While Job and Dalziel (2001) conceptualized fatigue in a coherent and
comprehensive manner, how can fatigue be differentiated from related constructs? To
recapitulate, fatigue is a state (either physical or mental) that is the result of activity
without adequate rest, resulting in insufficient means to continue performing using
In comparison to fatigue, habituation is a form of learning that results from
repeated exposure (either successive or intermittent) to a stimulus and is characterized by
a decreased reaction to the particular stimulus (Job & Dalziel, 2001). Therefore, the
states of fatigue and habituation can be discriminated according to two primary
character sti cs. First, fatigue results not only from repeated stimulus contact, but also
from a reduction in energy or capacity to perform. Second, habituation to a stimulus will
continue after sufficient rest, whereas fatigue will dissipate after an adequate rest period.
Therefore, habituation can be distinguished from fatigue based on its cause and
Motivation is the drive that initiates, directs, and sustains behavior and is based on
a plethora of factors, ranging from rewards and punishment to perceived ability (Petri,
1991; Job & Dalziel, 2001). Although fatigue influences behavior patterns and
persistence, it is solely based on performance capacity, previous activity, and rest, rather
than on a range of internal and external motivators.
Boredom is an individual's emotional reaction to an event, stimulus, or
environment that is perceived to be repetitive, unvaried, predictable, or monotonous
(Davies et al., 1983). Both boredom and fatigue may arise from the presence of a
repetitive behavior or stimulus. However, boredom is influenced by previous exposure to
a behavior or stimulus, independent of a rest period, whereas a rest period will modulate
fatigue. Simply defined, boredom is a subjective feeling of disinterest, independent of
time on task, energy, and rest (Job & Dalziel, 2001).
Although habituation, motivation, boredom, and fatigue are similar in origin and
influence, and are often used interchangeably, it is necessary to stress the distinctions
between them. Additionally it is important to highlight the unique combination of
characteristics that result in the individual state of fatigue.
Fatigue and Human Performance
Fatigue is a common, yet complex phenomenon that puzzles researchers and can
impede human performance. In many performance scenarios, particularly sport, the
human body is physically taxed. How this exertion influences the performance of
cognitive and motor tasks will be briefly reviewed with a focus on physical fatigue's
influence on psychological variables (i.e., information processing and cognition).
Although physical fatigue influences muscular exertion and performance (see Bompa,
1999 for a review), it also impacts cognitive functioning, an aspect of its influence that is
often overlooked in sport. The following section will provide a brief review of the
relevant literature related to exercise-induced (physical) fatigue and cognitive
performance. Additionally, because performance demands extend to include cognitively
demanding tasks, such as maintaining attention and decision-making, the review will also
focus on how cognitive fatigue is related to performance, subj ective, and physiological
Exercise-Induced Fatigue and Performance
In sports, fatigue is public enemy number one! Those athletes who cannot
effectively cope with fatigue have a high probability of performing poorly and
loosing the game, race, or match. Fatigue also affects the ability to stay focused,
resulting in technical and tactical mistakes and throwing or shooting inaccuracies.
This explains why, toward the end of a game or match, more mistakes are visible.
(Bompa, 2000, p.149)
Bompa (2000) both emphatically and clearly portrays the prevalence and influence
of fatigue in sport. In athletics, a decline in task performance is often attributed to the
onset of fatigue; individuals can no longer perform at the same level due to a perceived
reduction in ability, resources, or capacity because of previous performance. However,
human performance researchers and sport scientists are both concerned with the validity
of such statements and interested in the mechanisms that may promote such a
performance decrement. Therefore, empirical investigations have examined the influence
of physical fatigue/exertion on the performance of cognitive and motor tasks.
Physical exertion and cognitive processes are not independent; exercise can
modulate arousal. Specifically, when engaging in any type of physical activity, arousal
levels (e.g., heart rate) increase. The resulting alterations in arousal levels can either
facilitate or hinder cognitive functioning and performance (Zaichkowsky & Baltzell,
2001). Selected studies that have examined the exercise induced arousal/fatigue-
performance relationship will be reviewed.
Under the assumption that individuals make incorrect decisions when physically
fatigued, Davey (1973) examined cognitive functioning following exercise. A post-
activity protocol was used to assess the short-term memory functioning of participants
following cycling bouts of consistent resistance, but of variable duration. Performance
on a numeric-sequence identification task was the dependent measure. An analysis of
pre- and post-exercise mental task performance scores revealed an inverted-U pattern of
performance. The inverted-U hypothesis (Yerkes & Dodson, 1908) has frequently been
used to frame both the positive and negative effects of arousal on performance. Within
the inverted-U paradigm, performance is predicted to increase as physical arousal
increases until an optimal arousal threshold is reached. If arousal proceeds to increase
beyond this upper limit, a reduction in performance will result. Specifically within this
investigation, the inverted-U pattern emerged, with exercise improving task performance
after 30 second and after two-minute bouts of cycling. However, performance was
hampered after 10 minutes of exercise (the threshold within the hypothesis). The
resulting performance pattern supports the contention that physical exertion/fatigue has a
similar effect on cognitive performance as general arousal, enhancing performance to a
degree, but also hindering performance if the optimal level of exercise/exertion is
To replicate and extend the exercise-induced fatigue research, Reilly and Smith
conducted a pair of studies to investigate the effect of exercise intensity on both cognitive
(1984) and psychomotor (1986) task performance. In both protocols, active male
university students pedaled on a cycle ergometer at 25, 40, 55, 70, and 85% VOzmaX.
During the exercise, participants completed two mental arithmetic (1984) or pursuit rotor
(1986) tasks for 60 seconds during each loading interval. Although the tasks were
completed while exercising, rather than post-exercise, the results were similar to those
obtained by Davey (1973); performance was better at 25% VOzmax than at rest, and
performance deteriorated at 70 and 85% VOzmax. In the two experiments, moderate
physical work (44% and 38% VOzmaX TOSpectively) was associated with peak
performance, a pattern indicative of the inverted-U. In summary, Reilly and Smith
illustrated that both cognitive and psychomotor task performance are affected similarly
by physical activity and fatigue.
Using a similar procedure as the preceding studies, Salmela and Ndoye (1986)
examined cognitive alterations during progressive exercise. The authors hypothesized
that as exercise increased in intensity, attentional focus would narrow, as indicated by a
slower response to stimuli presented beyond the central visual Hield (i.e., located in the
periphery). To test this hypothesis, participants cycled at increasingly difficult levels of
resistance and responded verbally to Hyve lights presented in various areas of their visual
Hield. Salmela and Ndoye found support for their hypothesis; reactions to central stimuli
were significantly faster than to stimuli in the extremes of the visual field as exercise
progressed. Additionally, the inverted-U pattern of performance was once again
apparent, with performance being initially enhanced by activity, but as the level and
intensity of exercise increased beyond an "optimal" level, the activity hampered task
In summary, the above investigations are representative of the maj ority of the
literature examining the exercise-cognition association. Within the literature, as in the
selected sample, various forms of activity and exertion are used to induce what are
described to be fatigue states (although fatigue is not explicitly measured). Results
indicate that prolonged/intense exercise and the resulting fatigue negatively influence
performance on a variety of cognitive and motor tasks. Although the results support the
inverted-U hypothesis in describing the relationship between exercise and cognitive
performance, caution must be exercised not to blindly accept this unidimensional,
simplistic explanation. Further research is necessary to produce a more complex and
accurate depiction of the influence of exercise and ensuing fatigue on performance,
specifically assessing the development of fatigue, taking into account task characteristics
(both physical and mental), the fitness level of the population involved, the point at which
assessment is done (post versus during exertion), and the subjective appraisal of the
situation and activity by the participant. Clearly, the study of the effects of exercise on
cognition and human performance is central to understanding human behavior and
performance, however, the body is not the only resource taxed during performance.
Cognitive Fatigue and Performance
As alluded to, performance demands are not limited to physical stress and strain;
attention and information processing exert similar demands and can stimulate the
development of cognitive fatigue. Early research into the effects of cognitive fatigue
approached the relationship between mental activity and performance as though mental
and physical activity had a similar origin and influence on performance, attempting to
find straightforward decrements in cognitive performance (Craig & Cooper, 1992).
Unfortunately, little support was found for reduced mental output (e.g., decreases in
performance quality) after repeated exposure to a task (Craig & Cooper, 1992).
The measurement of mental activity and subsequent cognitive alterations is unique,
in that a maj ority of the processes are not directly observable (Matthews et al., 2000).
This potential limitation however, has not prevented exploration into cognitive fatigue's
influence on human performance, but has encouraged the use of multiple modes of
assessment. This section will begin with an exploration into what cognitive fatigue is,
followed by an overview of contemporary research. Within the overview, methods for
assessing cognitive fatigue, namely, self-report and physiological indices will be
highlighted. The section will conclude with a synopsis of where the current literature
stands and what questions remain unanswered.
Fatigue is a term that is used to refer to a variety of sensations and effects
encountered in the laboratory, everyday life, work, and sport. Fatigue resulting from
prolonged mental activity has two distinct categories: task specific and generalized
fatigue (Holding, 1983). Task specific mentalfatigue results when an individual is tired
of performing a specific task, and changing activities will alleviate the fatigue. In
contrast, generalized fatigue is a broad state of tiredness that is not related to one specific
task; therefore the switching of tasks will not alleviate generalized fatigue. With fatigue
having such broad foundation and implications, this section will be limited to a
discussion of task specific cognitive fatigue and its relationship with performance.
Early investigations (1943-1972)
Seminal research conducted by Bartlett (1943) began the investigation into the
fatigue-performance association. Bartlett used the Cambridge Cockpit, a simplistic
aircraft simulation, to assess changes in behavior and overall performance as time-on-task
increased. Pilots were asked to respond to information presented on several displays, a
task with little physical involvement. As time-on-task progressed, a decrease in timing
accuracy was found, although responses were otherwise appropriate. Additionally, a
reduction in the number of instruments attended to by the pilots was noted; the pilots
attended to the most important aspects of the display while ignoring other relevant
information, an indicator of a reduced attentional Hield as a result of prolonged activity.
Although, Bartlett' s work was conducted a half a century ago, it set the stage for more
recent research, drawing attention to the relative breakdown in coordination of behavior
as time-on-task/fatigue increases; as fatigue develops, overt performance may not
decrease, although the efficiency of task completion may be compromised (Bartlett,
Rather than manipulating fatigue induction solely through time-on-task, Hockey
(1970) used sleep deprivation in conjunction with a prolonged task to fatigue participants.
In the 1970 experiment, both sleep deprived and rested participants completed the dual-
task of tracking a target and responding to light signals. During the initial phases of the
experiment, both groups responded more quickly to centrally presented signals than to
those presented in the periphery. However, as time-on-task increased, already fatigued
(sleep deprived) participants began to respond more slowly to centrally presented signals,
minimizing the difference in response time between signals presented in the periphery
and those in the central visual field. Rested participants did not duplicate the response
pattern of their fatigued counterparts, but rather maintained a faster response time to
centrally presented stimuli as compared to peripheral stimuli. The decreased sensitivity
to centrally presented stimuli by fatigued participants indicates that the cognitive
processes of either response selection or attentional selectivity are hampered by fatigue.
The practical application of fatigue research was further advance by Dureman and
Boden (1972). In their 1972 investigation, Dureman and Boden combined the realism of
a driving simulator with physiological data collection to depict a more complete picture
of what occurs subj ectively, overtly, and physiologically as a result of four hours of
continuous driving. Results indicated that the drive successfully induced progressive
feelings of fatigue that paralleled overt performance errors. In conjunction with
performance errors, a decrease in arousal was noted, with lower heart rate and increased
skin resistance as time-on-task increased. This pattern of performance and arousal was
markedly different when arousal and motivation were stimulated through an electric
shock when a steering error was made, a change that highlights the importance of optimal
arousal for successful performance and how fatigue may compromise the maintenance of
an optimal level of arousal.
Contemporary investigations (1996-2002)
Contemporary investigations of cognitive fatigue have primarily focused on fatigue
in ergonomic settings, highlighting transportation due to the grave number of fatalities
and accidents believed to be the result of fatigued drivers/pilots (e.g., Desmond &
Matthews, 1997; Matthews & Desmond, 2002; Lal & Craig, 2002). Within this
literature, fatigue is examined as a general construct resulting from factors such as
sleepiness, general nighttime impairments, and time-on-task. The present review will
however focus on the influence of time-on-task, with an ergonomic orientation.
The link between fatigue and performance decrements is not a new concept.
However, within the last decade there has been a surge in the investigation of fatigue
within the transportation industry. Similar to Bartlett' s (1943) early investigation, Morris
and Miller (1996) utilized a simulated flight paradigm to examine the relationship
between time-on-task and performance, while incorporating the measurement of
physiological reactivity. Ten partially sleep deprived Air Force pilots flew eight, two-
segment legs in a flight simulator. During the flight, performance and physiology were
monitored and altitude, velocity, and speed error scores were calculated to measure
performance over time. Electrooculographic (EOG) data was used to monitor blink
characteristics (i.e., amplitude, rate, duration, and closure rate) and saccade information
(i.e., velocity and rate). In conjunction with physiological and performance indices,
subj ective ratings of fatigue, workload, and sleepiness were also collected using a three-
scale survey consisting of the Crew Status Check Card, the Sleep Survey Form developed
by the USAF School of Aerospace Medicine (SAM Form 202 and SAM Form 154) and
the Stanford Sleepiness Scale (SSS; Hoddes, Zarcone, Smythe, Phillips, & Dement,
1973) to assess fatigue and workload, sleep habits from the previous evening, and current
sensations of "sleepiness," respectively. Results indicated a positive correlation between
subjective reports of fatigue and error. Additionally, blink amplitude was found to be the
best predictor of changes in error; smaller blink amplitude corresponded with increases in
error. It was also found, through manual inspection, that saccade rate apparently
decreased across epochs as indicated through the pilot' s crosscheck of instruments.
Morris and Miller' s investigation revealed that measurable, negative performance
changes occur in fatigued pilots. Additionally, these fatigue symptoms and performance
correlates can be measured physiologically.
To better understand and subsequently manage fatigue in the transportation field,
Lal and Craig (2002) monitored the physiological changes that occurred during a driving
simulation. Thirty-five nonprofessional drivers, who were self-described to be sleep
deprived, performed a driving simulation task until physical signs of fatigue were
apparent (approximately two hours). Heart rate was recorded prior to and following the
"drive." Additionally, during the drive, video was used to capture the driver's face and
overt physical signs of fatigue. The video was used in conjunction with EOG and an
electroencephalogram (EEG). To assess self-report assessment of fatigue, the "fatigue
state question" scale was created (and validated) to measure fatigue levels before and
after the task. The most relevant finding to the current investigation is the decrease in
heart rate post-race, as compared to pre-race. It is this decreased physiological arousal
that is believed to be one of the contributing factors to performance decrements in a
variety of settings, however future research is warranted to better understand the
physiological changes associated with fatigue.
To extend earlier findings pertaining to the physiological correlates of fatigue and
how fatigue effects may be moderated, Moolenaar and colleagues (1999) examined the
influence of the stimulant ephedrine and a placebo on driving performance and
physiological reactivity to a four hour, 3-way divided attention task (pursuit tracking,
peripheral target detection, and random visual signal detection) designed to mimic the
demands of driving. Subj ective reports indicated that the participants who received the
placebo did become fatigued as time-on-task increased, in comparison to the
experimental group. Additionally, the control group experienced a progressive
deceleration in heart rate over the four-hour drive. Although subj ective and physiological
differences were noted between the groups, performance was only significantly different
for pursuit tracking; as time-on-task increased, control subj ect' s tracking performance
decreased, as indicated by an increase in tracking delay. In contrast, the experimental
group's performance increased on the tracking task as time-on-task increased.
Surprisingly, and counter to related research (e.g., Hockey, 1970), the control (fatigue)
group did not express a decreased sensitivity to peripheral targets. In summary, the direct
relationship between subj ective fatigue and lower arousal was supported, whereas
performance was not necessarily hindered by fatigue or lower arousal.
In a similar investigation, Macchi and colleagues (2002) conducted a direct
comparison between fatigued (no nap) and rested (nap) professional long-haul drivers.
The purpose of the study was to assess the effects of an afternoon nap on perceptions of
fatigue, performance, and arousal. After a three-hour nap and a two-hour "night-driving"
session in a simulator, the participants who were rested were significantly less sleepy and
fatigued, had faster reaction times, exhibited less performance variability, and had higher
levels of arousal than prior to the rest. It appears that the nap provided a "sufficient"
p eri od of re st, and was ab le to mi ni mize fati gue and its negative effe cts/con sequence s.
In another recent transportation investigation, Matthews and Desmond (2002)
examined the influence of task-induced fatigue on simulated driving performance. In a
two-part investigation, fatigue effects were hypothesized to be the result of reduced
attentional resources and ineffective effort regulation. Two tasks (control and fatigue
inducing [FI]) were used in a repeated measures design to contrast fatigued and non-
fatigued driving performance. In both conditions, participants were instructed to follow a
lead car on a stimulated track at a constant speed of 30mph. For the control task, this was
the participant' s only objective for six minutes while on both curved (higher workload)
and straight (lower workload) roadways. In contrast, for the 24-minute, FI task,
participants were required to complete a secondary signal detection task in addition to the
central driving task on both curved and straight roadways. Fatigue symptoms, emotions,
motivation, effort, and cognitions were subj ectively assessed before and after each drive.
Analysis of the subj ective reports revealed differences between the two groups, including
higher effort ratings during the FI task. Although there was an elevated level of
perceived effort exerted during the FI task, results indicated a decrease in performance as
fatigue developed. Specifically, there was an increase in heading error (difference
between front of car and roadway) and a decrease in signal detection on straight-aways,
compared with curved road segments as fatigue levels increased. According to the
performance data, in situations perceived to require low effort (straight-aways), fatigued
individuals ineffectively regulated effort, rather than had insufficient resources available
for task performance. This finding emphasizes the importance of understanding the task
demands when predicting and understanding fatigue's effect on performance. In
summary, the data confirms that task-directed effort or lack thereof is responsible for
driving performance while fatigued. Additionally, the authors suggest increasing
motivation prior to and during task execution, rather than attempting to minimize
attentional demands, is more effective for decreasing the negative influence of fatigue on
The implications for fatigue research to improve performance have been well
recognized by the transportation industry, yet has been relatively ignored within sport and
exercise sciences. It is apparent through the reviewed research that cognitive fatigue
influences and is influenced by human output, both directly and indirectly. Specifically,
it is cognitive effort (e.g., allocating attention, decision-making) that has the potential to
deplete resources to prompt a fatigue state. However, once a fatigue state develops it is
often paired with lowered arousal, decreased environmental sensitivity, slowed reaction
time, and overall impaired performance. Understanding how this procedure develops and
is expressed in a sporting context could provide useful information pertaining to
individual cognitive involvement in training and gaming situations.
Limitations in the Literature
This chapter has presented a survey of relevant literature and theory associated with
the performance-fatigue relationship, however, insufficient definitions, theories, and
methodologies continue to plague the literature, leaving many unanswered questions. In
much of the reviewed literature (e.g., Bartlett, 1943; Dureman & Boden, 1972), the word
fatigue is used in the title of the article or the abstract, but is not specifically revisited in
the content of the article. This common oversight within the literature makes the
assumption that all tasks can and will create a fatigue state for individuals at the same
rate, an incorrect assumption; it is incorrect for researchers (e.g., Bartlett, 1943; Davey,
1973; Reilly & Smith, 1984) to assume that a fatigue state has developed as the result of
prolonged exercise or time-on-task when it has not been specifically assessed.
Additionally, this oversight perpetuates the confusion associated with the term fatigue.
Although the term fatigue was used in the title or briefly in the introduction, it was used
oftentimes interchangeably with vigilance, drowsiness, boredom, and other related but
independent constructs. Such ambiguity among researchers emphasizes the importance
of the implementation of an effective operational definition.
However, caution must be exercised when defining fatigue. Although there have
been numerous attempts to conceptualize fatigue, one common misconception is that
fatigue is defined in terms of its consequences and is not recognized as a state of being
(Brown, 1994). Research that has taken this approach is circular in nature and does not
adequately address the concept of fatigue. Although this conceptual inconsistency has
plagued the literature, Job and Dalziel (2001) have provided a solid example of what
future definitions of fatigue should incorporate to accurately depict the state.
Conceptual and methodological limitations discussed above have minimized the
applicability and generalizations of many of the fatigue research's findings. However,
recognition of such limitations will promote more valid investigations. The more
knowledge gained about fatigue, the better it and its negative consequences can be
Management of Fatigue
Matthews and Desmond's (2002) results point to the notion that individuals have
the ability to offset/minimize some of the deficits in information-processing that occur as
a result of fatigue by simply modifying effort or strategy. As reviewed above, the effects
of cognitive fatigue on performance are linked to changes in effort and/or resource
availability (Dureman & Boden, 1972; Matthews & Desmond, 2002). Not only do
recent empirical investigations indicate the potential for individuals to override fatigue
effects, but also this phenomenon is well noted anecdotally. "Fatigue can be quickly
forgotten in a state of emergency or an excess of enthusiasm" (Holding, 1983, p. 145).
Through alterations in strategy and/or effort, performance may be maintained. However,
Brown (1994) cautions that although a decrease in performance or task effectiveness is
not guaranteed if work continues after the onset of a fatigue state, task efficiency will be
Although ambiguity remains regarding how output is influenced as fatigue sets in,
some certainty exists that as fatigue increases, response variability increases and
performance efficiency decreases (Bartlett, 1943; Craig & Cooper, 1992; Brown, 1994).
In summary, the influence of fatigue on performance efficiency and effectiveness has the
potential to be moderated by an individual's motivation and/or ability to exert additional
effort required for task maintenance, as well as through effective attentional allocation.
The ability for athletes to effectively manage their attention and effort is central not only
performance, but also to the minimization of injury and the prevention of burnout.
However, the benefits of fatigue management cannot be obtained unless more sport-
specific cognitive fatigue research is conducted.
Attention is a vital component of human performance. Additionally, the
appropriate allocation of attentional resources and cognitive effort is indicative of
successful and even elite performance. Unfortunately, there are many internal and
external factors that can compromise the efficiency and effectiveness of attentional
allocation. One such variable is fatigue, either physical or cognitive. Within sport,
cognitive fatigue is often the result of athletes concentrating maximally on the task at
hand and constantly having to make quick and accurate decisions based on environmental
stimuli (Bompa, 1999).
Anecdotally, coaches and athletes have linked losses in concentration, impaired
decision making, lack of focus, and other mental breakdowns to fatigue. The sources of
fatigue in sport are numerous and inherent given the physical and mental demands placed
on the athlete. In most sporting contexts, players are required to search a complex and
dynamic visual display, to attend to relevant cues while ignoring irrelevant information,
and then make the most appropriate decision based upon the newly acquired and stored
information. The process hopefully culminates in an efficient and effective series of
movements (McMorris & Graydon, 1996). However, performing such activities over an
extended period of time with minimal rest can result in fatigue. One objective of the
present investigation is to reduce some of the ambiguity associated with cognitive
fatigue, specifically cognitive fatigue in sport. Through the use of subj ective reports of
fatigue, physiological patterning, visual search patterns, and performance indices, insight
will be gained regarding the onset, symptoms, and influence of cognitive fatigue in sport,
while making advancements in the knowledge base of sport science and the general
human performance literature.
Herein the methodology employed to empirically investigate the onset and effects
of cognitive fatigue during a sporting simulation is described. An hour-long hockey
decision-making task was used to induce fatigue and monitor performance changes in
simulation situations (Reilly, 1996). Through the assessment of fatigue, boredom,
motivation, performance, and physiological reactivity, a more detailed depiction of
cognitive fatigue's development and influence in sport has been gained. Specific
variables of interest and importance are reviewed below.
To determine an adequate sample size, the power table for single group repeated
measures designs (Barcikowski & Robey, 1985) was referenced. The desired power was
determined to be 0.80 with a medium effect size and alpha level set at the 0.05 level.
Based on these requirements, a sample size of 14 was determined to be adequate.
Eighteen male athletes (mean age = 20.6, SD = 2.57) with advanced ice hockey
experience (high school level or beyond) and an average of 10.03 (SD = 4.41) of playing
years of hockey involvement participated in the experiment. Participation was limited to
males because the simulation was only available with male players. Prior to
participation, participants were asked to restrict caffeine and food intake for the four
hours prior to testing and to avoid alcohol for 24 hours prior to their scheduled session
due to the potential interaction between food and beverage consumption and fatigue (Lal
& Craig, 2002) (see Appendix A).
Video clips demonstrating actual on-ice play extracted from EA Sports NHL 2001
(Electronic Arts, Redwood, CA) for Sony PlayStation 2 (Sony Computer Entertainment
America, San Mateo, CA) using the Pinnacle Video Editing Suite (Pinnacle Systems,
Inc., Mountain View, CA) were used in the simulation. Scenarios were created using the
replay function of the simulation and altering the angle so that an "on-ice" perspective
was created. Attention was given to creating scenarios that were different enough to
minimize boredom, yet similar in the level of complexity. The teams used in the
simulation footage were numbered players dressed in Team USA home and away jerseys.
The level of play was set to "pro" and the speed of the simulation was "moderate" in an
attempt to create scenarios that are most similar to the actual playing intensity and skill
level of the participants. Scenarios include defensively and offensively oriented
perspectives that players from either position are qualified and able to "read."
To establish the validity of the scenarios, as well as the most appropriate response
to each question, a CD-Rom of the simulation was provided to 7 hockey coaches. The
coaches were provided with each question and four possible responses and asked to
identify the most appropriate response. If the coach felt that none of the responses were
applicable, they were asked to supply what they believed to be the most appropriate
response. Coach response concordance demonstrated reliability (R = .87). A total of 60
clips, lasting 6.3 to 14.4 seconds in duration were selected for inclusion in the simulation.
A Sharp Notevision LCD Projector (Model XG-NV2U, Tokyo, Japan) projected
the resulting hockey footage from a Falcon Northwest Mach V XP 2800 computer
(Falcon Northwest Computer Systems, Ashland, Oregon). The testing period consisted of
12 trial blocks (TB) presented twice, once in the first half and once in the second half, for
a total of 24 trial blocks. Blocks consist of five video sequences, with each video clip
being followed by a question concerning a scenario-relevant decision with four multiple-
choice responses. Participants were asked to select the most appropriate response as
quickly and as accurately as possible. Each trial block was approximately 2.5 minutes in
duration, however the actual block lengths varied based on individual response times due
to individual differences in the time required by participants to respond to each question.
Trial blocks were presented randomly across participants, as was the order of the clips
within each block. A program written in LabVIEW (6.0, National Instruments, Austin,
TX) will control the timing and presentation of information.
Prior to the beginning of each trial bock and following TB 24, three separate visual
analog scales (VAS) were presented to monitor changes in fatigue, motivation, and
boredom over the duration of the experiment. Although this assessment interrupted the
central task, this pause is not a period of sufficient rest, and therefore did not decrease or
eliminate a fatigue state (Job & Dalziel, 2001). Throughout the testing period both visual
search patterns and arousal were monitored to index the onset and effect of cognitive
Although the methodology is grounded in the assumption that time-on-task
promotes the development of fatigue, it is inappropriate and inaccurate to presume as
time-on-task increases, fatigue increases proportionally (Sirevaag & Stern, 2000;
Holding, 1983). Therefore, time-on-task was used to induce fatigue, yet fatigue
development was also monitored through successive reports of the subj ective experience
of fatigue and fluctuations in physiology.
Measurement Devices and Dependent Variables
Gaze Behavior Measurement and Data Reduction
Eye movements were monitored and recorded using the Applied Science
Laboratories 5000 Series Model 504 Eye Movement System with Eyehead Integration
(ASL, Bedford, MA) in conjunction with a Dell Optiplex GX250 2.53 GHz computer
(Dell Inc., Austin, TX). The Model 504 is a complete video eye tracking system with a
remote optics eye camera that allows for a moderate amount of head movement
(approximately one square foot), and therefore does not limit a seated individual's normal
range of head movement. The Model 504 uses pupil-corneal reflection to measure point
of gaze related to the scenarios recorded by a scene camera. A head mounted magnetic
sensor and transmitter was used to monitor the head's location in space. The
participant' s visual gaze patterns are displayed on the image acquired from the remote
scene camera as a set of cross hairs. The intersection of the cross hairs represents the
coordinates of the pupil position and corneal reflex of the "dominant" eye.
The system's accuracy is to 0.50 visual angle with a visual range of 500
horizontally and 400 vertically. Point of gaze coordinates are sampled at 60HZ.
Calibration of the system through a nine-point reference grid prior to testing ensured that
the points of gaze correspond to the appropriate elements of the visual display. The
numbers in the calibration grid were arranged in a three by three matrix, with the upper
left corner labeled point 1, the middle point 5, and the lower right point 9. During the
calibration process the coordinates of the visual fixation were recorded as the participant
progresses through the nine points.
For this investigation, fixation duration and total number of fixations per scenario
were of primary interest. A fixation occurs when the eye remains stationary on any point
in the display for a period of 100ms of longer. Eyenal software (ASL, Bedford, MA) was
used offline to identify eye fixation and scan patterns per segment (i.e., clip) of data. The
total number of fixations per trial block and the mean fixation duration of each trial block
was used to represent gaze behavior for each trial block.
Arousal Measurement and Data Reduction
In conjunction with the ASL Model 504 eye tracking system and Dell Optiplex
GX250 2.53GHz computer, AcqKnowledge (3.7.3, Biopac Systems Inc., Santa Barbara,
CA) was used to manage the data collection through integration with a program written in
LabVIEW. Specifically, the LabVIEW software triggered the collection of heart rate and
skin conductivity through a Biopac MPl50 system (Biopac Systems, Inc., Santa Barbara,
CA) and signaled video clip onset and offset, as well as indicated trial block durations.
Additionally, the LabVIEW software synchronized response time and eye movement data
with arousal levels for each video clip scenario. Data was reduced offline using the
AcqKnowledge software. Specifically, the software calculated the mean, standard
deviation, and maximum values of heart rate (beats per minute (BPM)) and skin
conductance levels (micromho (pmho)).
Heart rate was measured using pre-gelled disposable snap electrodes placed on the
interior of both the left and right forearms, with a ground electrode on the participant' s
non-dominant forearm. The arms were prepared using rubbing alcohol. The Biopac
amplifier bandpass filter was set to 1 35 Hz with a Gain of 5000 Hz.
Skin conductivity was measured using nonpolarized silver-silver chloride (Ag-Ag
Cl) electrodes placed on the medial edge of the non-dominant hand. The hand and
electrodes were prepared using Omni-prep with distilled water and ECI electrode paste,
respectively. A Biopac Electrodermal activity amplifier (GSR100B) provided a
continuous 0.5 V through the electrodes. The amplifier was set to detect a range of 0-50
Clmhos, with a bandpass filter set from DC to 1.0 Hz. The signal was amplified 500 times
by the Biopac amplifier.
Task Performance: Response Time
The time taken from question onset until response submission constituted the total
response time (RT) in milliseconds (ms) for each scenario (trial). The LabVIEW
program running on a Falcon Northwest Mach V XP 2800 computer both triggered the
timing mechanism and stored the response times after each trial.
Task Performance: Response Accuracy
To obtain a measure of decision-making effectiveness, the LabVIEW program
running on a Falcon Northwest Mach V XP 2800 computer was used to record participant
responses and score the responses according to the master list of responses. The numbers
of correct responses per trial block constituted the response accuracy (RA) score, with
maximum score per block being five and minimum score being zero.
Subjective States: Visual Analog Scales
To monitor fluctuations in subj ective states related to performance, three visual
analog scales were used. Visual analog scales are brief, simple scales that place minimal
cognitive demands on the respondent and minimally disrupt the completion of other tasks
(a primary rational for selection in this study). The scales are 100 unit horizontal lines
without unit markers and that contain anchor words/phrases at each end to identify the
maximal and minimal levels of the mood/state being assessed (Brown, 1994). Descriptive
anchors were selected to accurately capture how the participants were feeling at a given
instance and have been used in other fatigue/performance related studies successfully
(Brunier & Graydon, 1996). Brown (1994) cites that using a bipolar analog scale is a
"sophisticated" method for assessing the subj ective symptoms of fatigue that is not
influenced by a potentially diminished ability or tendency to report such symptoms when
fatigued, as noted through the use of other self-report assessments. Additionally, a pre-
assessment training session on how to use the VAS increases the reliability and validity
of the measure (Gift, 1989). Because the primary task was computer based, a program
written in LabVIEW modified the VAS so it could be administered electronically. Scores
on the scale range from zero to 100, with the scores increasing in value from left to right
along the scale. The score/value that corresponded to where the participant placed the
cursor was the subj ective state score (boredom, motivation, or fatigue) that the participant
received for that trial block. The LabVIEW program recorded subscale scores.
Upon entering the laboratory, participants were informed that during the session
they would be answering questions related to their ice hockey background and their
experiences in the experimental session. Additionally, participants were told that they
would be viewing ice hockey scenarios, which would be followed by tactical decisions
related to the previously viewed video clip. Participants were informed that their
response time and accuracy would be calculated after the session to determine how they
performed in comparison to other ice hockey players, and that they would be notified
about their performance at a later date. It was then explained that eye movements and
physiological responses would be monitored during the task through several non-invasive
sensors. Following explanation of the tasks and their involvement, participants
completed an informed consent document (see Appendix B) approved by the Institutional
Review Board (see Appendix C) and the sport history demographic questionnaire.
After the completion of the informational material, participants were fitted with
heart rate and skin conductivity sensors in the preparation area and then seated in the
testing area 1.5m away from a 1.5 m x 2 m projection screen. The laboratory was
darkened to enhance the image of the pupil obtained from the eye tracking system and to
enhance the quality of the video projection. Once seated comfortably, the eye tracking
system was calibrated.
Following successful calibration of the eye tracking system, the participants was
reminded that their goal was to respond as quickly and as accurately as possible to the
presented questions, because their final score would be based on both measures and
would determine their ranking in the "competition" (McMorris & Graydon, 1996). The
competition incentive was provided to increase motivation and the probability that the
participants would exert more effort than if no incentives were provided. Feedback was
not provided in an effort to minimize the habituation effects associated with
repeated/prolonged testing. In addition, the location of the relevant cues varied among
the clips in order to enhance involvement in the task and minimize boredom, an approach
similar to that taken by McMorris & Graydon (1996).
Following verbal explanation of the task, participants were guided through a
practice session. During the demonstration, task guidelines and directions were provided
(see Appendix E for complete instructions). Specifically, participants were shown, and
given the opportunity to familiarize themselves with the interactive components they
would encounter during the testing session. The first feature participants were exposed to
were the three Visual Analog Scales (VAS) that were used to measure subjective
experiences of fatigue, motivation, and boredom (Appendix F). Participants were
provided with a training session on how to use the VAS and were specifically instructed
how to select, change, and submit responses when presented with the scales, as well as
provided with explicit operational definitions for the anchors. The experimenter
demonstrated how to use the provided computer mouse to indicate how they feel at that
moment by placing a mark in the box the appropriate distance from the corresponding
descriptor. Following the brief explanation and demonstration, participants had the
opportunity to practice selecting and submitting VAS responses.
Once participants demonstrated competency using the VAS, the next component of
the experiment was presented. Specifically, a sample hockey scenario followed by four
response selections were shown. Participants were informed that this video-question
sequence would comprise the maj ority of the testing session and their obj ective was to
select the most appropriate response as quickly and accurately as possible. Only one
selection per question could be submitted, but responses could be changed an unlimited
number of times prior to clicking the continue button. Participants had a maximum of
one minute to submit a final response. Once again, the participants were provided with
the opportunity to practice selecting, changing, and submitting responses. At the
completion of the practice period lasting approximately five minutes, the actual testing
Following completion of the twenty-fourth trial block and subsequent VAS, the
experimental session ended and the experimenter returned to the laboratory to remove the
sensors and to explain the purpose of the investigation to the participant (see Appendix G
for debriefing script). The duration of the testing sessions was approximately 60 minutes,
a period comparable to an ice hockey competition.
Design and Analysis
The design was a repeated measures time series, with all participants engaging in
all 24 trial blocks, a design typical of fatigue research that examines output (Campbell &
Stanley, 1969). A common limitation associated with time series designs is that
alternative explanations, other than the factor under investigation, could be responsible
for the observed changes. The current investigation however minimized the potential
influence of extraneous variables by simultaneously assessing factors, in addition to
fatigue, which may be related to performance fluctuations (i.e., response time, boredom,
motivation, visual search patterns, and arousal).
Data were examined for outliers, defined to deviate from the mean by at least three
standard deviations, an area which comprises 99.74% of the distribution (Herzberg,
1989). Data that exceeded this criterion were considered to be missing data, variables
with considerable missing data (at least one-third of the data set) were not considered for
further analysis. A Kolmogorov-Smirnov test was conducted to assess the distribution of
the data, and determine if the assumption of normality was met. A Pearson-product-
moment correlation was used to determine the magnitude and direction of the relationship
between subjective reports of fatigue and performance indices (i.e., response time,
response accuracy), gaze behavior (i.e., fixation duration), and arousal (i.e., heart rate,
skin conductance levels). A repeated measures multivariate analysis of variance
(MANOVA) was used to evaluate changes in subj ective ratings (i.e., fatigue, motivation,
boredom), performance (i.e., response time, response accuracy), gaze behavior (i.e.,
fixation duration), and arousal (i.e., heart rate, skin conductance levels) over the 24 trial
blocks, across participants. Univariate analyses of variance (ANOVA) were used to
evaluate simple effects and significant simple effects were followed with Tukey's
Honestly Significant Difference (HSD) follow-up analyses for planned comparisons,
controlling for Type I Error violations possible when conducting multiple comparisons.
Results for this investigation are presented in the following chapter. The first
section will present the correlations between subj ective cognitive fatigue and
performance, gaze behavior, and arousal. The second section will assess the influence of
time-on-task on subj ective reports of fatigue, boredom, and motivation, as well as time-
on-task with the obj ective measures of performance, behavioral indices, and physiology.
Across dependent measures, data that exceeded three standard deviations from the
mean were deemed outliers and were scored as missing data. A total of six response time
values, nine mean Eixations values, and one heart rate value were deemed outliers and
thus discarded from further analysis. Additionally, the dependent measure of visual
Eixations was discarded from further analysis due to excessive lost data within and
between participants. Kolmogorov-Smirnov tests confirmed that the data within blocks
of the remaining dependent measures did not deviate significantly from the normal
distribution (p-values ranged from 0. 11 to 1.00), therefore raw data was considered
appropriate for use in subsequent analyses.
Correlates of Subjective-Reports of Fatigue
Pearson product-moment correlation coefficients were computed to assess the
magnitude and direction of the relationship between subj ective reports of fatigue and
performance indices, gaze behavior, and arousal.
A significant negative correlation was found between fatigue and mean response
time (r = -.39, p < .001), indicating that as participants reported higher levels of
subj ective fatigue they also responded faster to the tactical decision-making task. The
significant association is displayed in Figure 4-1. No relationship was found between
fatigue and response accuracy (r = -.06, p = 0.22).
10.00- q. -
** . *'
,, ** .. .
0.00 25.00 50.00 75.00 100.00
Figure 4-1. Subjective fatigue by mean response time.
Fatigue and mean Eixation duration (r = .01, p = .89) were not significantly related.
A significant positive correlation was found between fatigue and heart rate (r = .30,
p < .001), indicating that higher levels of subj ective fatigue were associated with elevated
heart rate. Additionally, a significant positive correlation was found between fatigue and
skin conductance levels (r = .40, p < .001), indicating that higher ratings of fatigue were
associated with greater skin conductivity. The associations are displayed in Figures 4-2
and 4-3 respectively.
.' .' ** *
...~ .. *
Figure 4-2. Subjective fatigue by heart rate.
0.00 25.00 50.00 75.00
Figure 4-3. Subjective fatigue by skin conductance level.
Influence of Time-on-Task
Subjective-Reports, Task Performance, and Physiological Reactions
A repeated measures multivariate analysis of variance (RM MANOVA) was
conducted to determine if there was a significant trial block effect across the dependent
measures of subj ective reports, task performance, and arousal. A significant trial block
difference was found (Wilk' s Lambda = 0.22, F (161, 2291) = 3.59, p < .001). Follow-up
univariate Analysis of Variance (ANOVA) revealed significant trial block differences for
subj ective fatigue (F (23, 345) = 17.49, p < .001), boredom (F (23, 345) = 21.77, p <
.001), motivation (F (23, 345) = 19.65, p < .001), response time (F (23, 345) = 11.32, p <
.001), response accuracy (F (23, 345) = 1.70, p = .02), and skin conductance (F (23, 345)
= 2.21, p = .001). No significant differences were found for heart rate (F (23, 345) =
0.63, p = .911). Figures 4-4, 4-5, and 4-6 represent the respective trends and Table 4-1
presents the descriptive statistics. Note in Figure 4-4 that motivation scores were plotted
as the difference from 100 to depict the actual trend of the data.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
~Boredom + Fatigue --Motivation
Figure 4-4. Mean ratings of fatigue, boredom, and motivation by trial block.
4 *~ 14
S3.5 *;~ A 12 .
SResponse Accuracy + Response Time
Figure 4-5. Performance by trial block.
5.8 t5h 73
c 5.6 e\ 72ifI
8 ~~5.4 71
U'5 69 a
S 4.8 68
-tSkin Conductance Heart Rate
Figure 4-6. Mean ratings of arousal by trial block.
Tukey's HSD follow-up analyses were conducted to determine the location of
significant differences within dependent measures. Results revealed a general trend of
significantly lower levels of fatigue being reported in the earlier trial blocks when
compared with the later trial blocks. For example, significantly less fatigue was reported
in TB 5 (M~ = 23.43, SD = 14.47), than in TB 22 (M~= 51.02, SD = 31.66). A similar
pattern of significance was evident with respect to boredom; significantly lower levels of
boredom were reported in earlier trial blocks than in later trial blocks. For example,
significantly lower levels of boredom were reported in TB 3 (M~= 14.34, SD = 9.60) than
in TB 24 (M~ = 57.25, SD = 35.60).
Self-reported motivation revealed an opposite pattern; significantly higher levels of
motivation were reported in the earlier trial blocks as compared to the later trial blocks.
For example, participants reported being significantly more motivated in TB 2 (M~=
14.72, SD = 12.22) than in TB 20 (M~= 45.98, SD = 32.14). Additionally, response times
were significantly faster in the later trial block than in the earlier trial blocks. For
example, participants responded significantly faster in TB 19 (M~= 8.76 sec, SD = 2.99)
than in TB 1 (M~ = 14.41 sec, SD = 4.80).
Significant differences were also noted in response accuracy between TB 23 (M~=
3.89, SD = 1.23) and TB 1 (M~= 2.39, SD = 1.14), TB 10 (M~= 2.44, SD = 1.20), as well
as TB 12 (M~= 2.44, SD = 1.50), indicating that participants had significantly better
performance accuracy in trial block 23 compared with that of trial blocks 1, 10, and 12.
Additionally, a significant difference in mean skin conductance levels was noted between
TB 5 (M~= 5.08, SD = 2.30) and TB 22 (M~= 5.92, SD = 2.92), indicating significantly
higher arousal in the later block.
Due to the high correlations between fatigue and boredom as well as motivation (r
=.74, p <.001 and r = .77, p < .001, respectively), in addition to the significant variability
of boredom and motivation across trial blocks, two repeated measures multivariate
Table 4-1. Dependent measure descriptive statistics M, (SD), and n by trial block.
Table 4-1. Continued
Table 4-1. Continued Trial Block
analysis of covariance (RM MANCOVA) were conducted to account for the variance
explained by boredom and motivation, respectively.
A significant trial block difference was found, controlling for boredom (Wilk' s
Lambda = 0.39, F (115, 1565) = 2.88, p <.001). A follow-up univariate ANCOVA
revealed significant trial block differences for subj ective fatigue (F (23, 322) = 8.71, p <
.001), and response time (F (23, 322) = 6.51, p < .001). No significant differences were
found for response accuracy (F (23, 322) = 0.82, p = .711), heart rate (F (23, 322) = 1.39,
p = .111), or skin conductance (F (23, 322) = 1.93, p = .249).
A significant trial block difference was also found when controlling for motivation
(Wilk' s Lambda = 0.52, F (115, 1565) = 1.92, p <.001). A follow-up univariate
ANCOVA similarly revealed significant trial block differences for subj ective fatigue (F
(23, 322) = 4. 12, p < .001), and response time (F (23, 322) = 4.00, p < .001). No
significant differences were found for response accuracy (F (23, 322) = 0.95, p = .528),
heart rate (F (23, 322) = 1.37, p = .120), or skin conductance (F (23, 322) = 0.76, p =
To compare mean fixation durations across viewings, overall mean fixation
duration was calculated for trial blocks 1 -12 (M~= 0.47, SD = 0. 19), as well as for trial
blocks 13 24 (M~= 0.44, SD = 0.15) (see Figure 4-7). A paired-samples t-test was
conducted to compare overall mean fixation durations between the first and second
viewings of the tactical scenarios. No significant differences were found for fixation
duration (t (10) = 0.95, p = .37). However, a significant relationship was found between
subjective fatigue and the length of lost gaze data (sec) (r = .34, p < .001) (see Figure 4-
1 st Viewing
O 2nd Viewing
Mean fixation duration by viewing.
10 -- *
2 a -I
+ Gaze Inss + Fatigue
Figure 4-8. Significant relationship between fatigue and lost visual search data.
Summary of Results
The onset and influence of cognitive fatigue was assessed through self-report
measures (i.e., VAS), task performance (i.e., response time, response accuracy), gaze
behavior (i.e., mean fixation duration), and physiological arousal (i.e., heart rate, skin
conductance). These data support the notion that time-on-task promotes the development
of cognitive fatigue and fluctuations in performance, as well as arousal. However,
counter to original hypotheses, subj ective reports of cognitive fatigue were not associated
with decreased performance, inefficient visual search patterns, or suppressed arousal.
Fatigue was however, related to faster response times and elevated arousal.
Anecdotally, cognitive fatigue has been associated with reduced energy and
resources necessary to perform at one's highest potential, resulting in increased response
times and obstructed decision-making capabilities (Silva & Stevens, 2002). Yet, the
pervasive construct of cognitive fatigue has been insufficiently examined in sport
populations. The primary purpose of this project was to empirically investigate the onset
and influence of cognitive fatigue on multiple performance-relevant dimensions in the
context of a simulated ice hockey decision-making task. Time-on-task was predicted to
induce cognitive fatigue, a condition that would be characterized by elevated subj ective
perceptions of fatigue, suppressed arousal, and inefficient visual search patterns.
However, the general prediction that cognitive fatigue would influence performance both
directly and indirectly by promoting systematic changes in attention, arousal, and
response patterns was not supported. Specific Endings are discussed, strengths and
limitations of the investigation are addressed, directions for future research are proposed,
and applied implications of these Eindings are suggested.
Review of the Findings
Correlates of Subjective Ratings of Fatigue
Increased ratings of subj ective fatigue were expected to be associated with
decreased performance in both response accuracy and response time. Although response
accuracy was not related to fatigue, response time was associated with fatigue. As
participants became more fatigued, their response times decreased, a finding counter to
the original hypothesis. This finding may be attributable to the participant's prior
exposure to the scenarios during the earlier trial blocks (e.g., TB 1- TB 12). That is,
although participants were less fatigued during earlier trial blocks, they required more
time to respond to the tactical decision due to lack of familiarity with the presented
situation, in contrast to when the scenarios were presented the second time (TB 13 TB
24). Therefore, it is possible that the originally hypothesized relationship may have been
obtained if the scenarios in the later trial blocks were novel, rather than repetitive.
Another probable alternative explanation is that as participants became more fatigued,
bored, and/or less motivated, their primary goal shifted from task performance to task
completion. However, this shift was unrelated to performance accuracy, suggesting the
presence of a learning effect.
Although no relationship was found between fatigue and mean fixation duration,
subjective fatigue was correlated with the length of lost gaze data. This unexpected
finding is particularly interesting considering that the lost visual search data may be
attributed to variations in blink magnitude, closed eyes, and changes in posture, all of
which can result in the inability of the eye tracking unit to accurately capture the retina
and/or corneal reflex while tracking visual fixations. Such causal factors could also be
indices of fatigue, both directly (e.g., napping) or indirectly (e.g., slouching in the seat),
although it can only be hypothesized that they are based on the above correlational
It was hypothesized that cognitive fatigue would be associated with suppressed
arousal. However, higher levels of cognitive fatigue were associated with elevated heart
rate and skin conductivity. In contrast to the current study, previous research (e.g.,
Dureman & Boden, 1972; Lal & Craig, 2002) that has supported the fatigue-lower
arousal relationship have examined physiological changes in arousal after several hours
of engagement in long-haul driving simulations. The lack of replication of earlier
research here suggests that the arousal deflating effects of fatigue may not be present or
recognizable until prolonged exposure to cognitive task demands has elapsed.
Additionally, unlike previous studies, the current design simulated the highly physical
task of playing ice hockey, rather than a more passive task of continuous driving. By
viewing and vividly mentally recreating scenarios individuals are able to elicit a
physiological response with a patterning similar to that evoked during actual task
completion, albeit to a lesser magnitude (Mann, Mousseau, & Janelle, in review; Lang,
1979). Therefore, participants in the current investigation may have exhibited greater
arousal in response to active mental engagement in the simulation of a highly arousing
task, an ice hockey game. Consequently, the increased arousal evident in this
investigation may be more strongly associated with the task characteristics, rather than
cognitive fatigue. Future research is needed to better discern the physiological patterning
of arousal associated with cognitive fatigue resulting from a sport-related task.
Influence of Time-on-Task on Dependent Measures
In general, time-on-task influenced subj ective states, performance, and arousal.
Most notably, time-on-task, as predicted, was successful at producing elevated levels of
cognitive fatigue. However, initial predictions pertaining to the pattern of change across
trial blocks for boredom, motivation, response time, response accuracy, heart rate, skin
conductance, and gaze behavior were not supported.
Performance incentives and dynamic video scenarios were implemented to
maximize participant involvement and motivation, while minimizing boredom
(McMorris & Graydon, 1996). Despite this, boredom began to increase and motivation
decrease following trial block 14 (see Tables 4-2 and 4-3). It is likely that after viewing
two blocks of "repeated" scenarios (TB 13 and 14) that participants had already viewed
within blocks 1 tol2, they began to withdraw and disengage themselves from the task as
indicated by their subj ective reports. Independently controlling for boredom and
motivation, MANCOVA supported this contention, revealing each variable's influence
on response accuracy and skin conductance. However, fatigue and response time were
not influenced by boredom or motivation.
It was further hypothesized that time-on-task would promote decreased
performance, observable through increased response time and decreased response
accuracy. Counter to initial hypotheses, response time decreased as time-on-task
increased (see Table 4-4). As noted previously, this pattern may have been due to the
participant' s familiarity with the repeated tactical scenarios, or withdrawal from the task.
Interestingly, a negative relationship between time-on-task and performance accuracy
was not evident, indicating that participants did not sacrifice accuracy for speed in
Additionally, it was hypothesized that time-on-task would elicit suppressed arousal.
However, heart rate did not fluctuate as a function of time-on-task. In contrast, skin
conductivity was higher in later, rather than earlier trials. The discrepancy in arousal
measures in reference to time-on-task, yet significant association between heart rate and
skin conductance with perceptions of cognitive fatigue, suggests that further empirical
investigation is warranted to determine the degree and direction of the triangular
relationship between arousal, task engagement duration, and subj ective fatigue.
Due to the excessive loss of visual fixation data, an incomplete assessment of
visual search efficiency was not possible, however, a partial analysis of visual fixation
duration was completed. In the paired-sample comparison of the mean fixation durations
for viewings one (TB 1 -12) and two (TB 13 -24), it is evident that the mean fixation
duration decreased as time-on-task increased, however the difference of .02 seconds was
not significant, possibly due to the small sample with useable data (N= 11). However,
in accord with the original hypotheses, trends suggest that as the period of cognitive task
engagement lengthens and fatigue develops, visual search patterns become more
inefficient, requiring more total fixations to be made to extract relevant information,
thereby not maximizing the utility of the display (Williams, 2000). These trends require
further empirical investigation.
Strengths, Limitations, and Directions for Future Research
The current investigation adopted an explicit and definitive operational definition
of cognitive fatigue, examined the construct from multiple perspectives (i.e., subjectively,
behaviorally, and physiologically), and extended the research to sport applications. An
additional strength of this study's design is that the continued cognitive involvement of
players in the actual competitive scenario was accounted for with the creation of a 60-
minute tactical cognitive task. Although it can be argued that sport specific elements of
ice hockey such as, line changes and period breaks, could moderate the effects of fatigue,
it is important to return to Job and Dalziel's (2001) definition of fatigue which
emphasizes insufficient rest in the development and persistence of fatigue. Therefore, the
question becomes, can such minimal, in-game breaks be considered sufficient to heed the
development of fatigue? During such breaks players are given physical rest, yet they are
required to "keep their heads in the game" in preparation for their next shift or the next
period. In sum, by maximizing the similarities between the actual and laboratory tasks
the current investigation was able to assess for the development and onset of cognitive
fatigue with temporal sensitivity.
Despite attempts to maximize ecological validity, the attempt to provide control
when comparing visual search patterns between the first twelve trial blocks and the last
twelve, resulted in the same scenarios being presented in both conditions. Although
performance feedback was not provided to the players, the repeated scenarios may have
prompted the development of boredom, as well as reduced response time as a result of the
players' familiarity with the content. Therefore, it is recommended that future
researchers, when implementing similar protocol, use distinct and novel scenarios
throughout the task, while controlling for content. As such, more accurate comparisons
of performance over time may be made while controlling for extraneous variables,
Similarly, much of the fatigue research has generally failed to account for related
psychological phenomena, namely boredom and motivation. Such factors were
specifically assessed here, revealing the relatedness of such variables to fatigue, as well
as their influence on performance related constructs (e.g., response accuracy, skin
As with most phenomena, fatigue does not occur in isolation, but maintains a
reciprocal relationship with a variety of other factors. Therefore, it would be beneficial to
empirically examine a host of other mood, personality/individual differences,
experience/expertise, and competition-related variables.
For example, future studies should consider the moderating influence of various
mood states on the perception and influence of cognitive fatigue. It is assumed that mood
can affect both the speed, as well as the efficiency of information processing, and thereby
influence performance (Matthews, 1992). Specifically, mood has been associated with
competitive sport performance; positive moods (e.g., happy, energetic, and enthusiastic)
were correlated with better performance (Totterdell, 1999). Additionally, pre-
competition mood states have been suggested to influence appraisals of an ensuing
performance (Mellalieu, 2003). For example, negative mood states may prompt greater
attention to be directed to developing fatigue symptoms than a positive mood state, since
negative mood states tend to signal that a situation is atypical/challenging and requires
additional energy and effort in order to be resolved (Schwarz, 2000). Therefore, more in-
depth pre and posttest assessment of mood states (e.g., POMS; McNair, Lorr, &
Droppleman, 1992; PANAS; Watson, Clarke, & Tellegen, 1988), as well as continued in-
task assessment of mood, may be able to provide a more detailed indication of the
potential intervening role of various mood states (both positive and negative; Mellalieu,
2003) in the development, perception, and influence of cognitive fatigue in sport.
Additionally, potentially significant individual differences, such as self-esteem,
have been shown to have performance implications (e.g., Di Paula & Campbell, 2002)
and may further moderate cognitive fatigue. Specifically, high self-esteem, the belief that
one is capable and successful, is often associated with attempts to bring about success
and confirm one's sense of self-value (Sommer & Baumeister, 2002). In contrast, low
self-esteem is correlated with low levels of confidence in one's skills, abilities, and
overall worth. Moreover, individuals with low self-esteem generally have lower
expectations for their own performance and success. Di Paula and Campbell (2002)
found that when faced with potential failure, individuals high in self-esteem were more
persistent at overcoming the adversity than those low in self-esteem. Specifically, high
self-esteem was associated with greater motivation to work harder when the potential
reward/level of success is significant because individuals with high self-esteem are
"motivated to achieve success', rather than "avoid failure," as is typical of individuals
with low self-esteem. Therefore, the recognition and influence of fatigue, as well as
persistence in a fatigue state, may be moderated by an individual's level of self-esteem.
The amount of experience or level of expertise achieved by an athlete may also
influence the effects of cognitive fatigue. When addressing the development of expertise,
Ericsson, Krampe, and Tesch-Roimer (1993) emphasize the significance of rest between
deliberate practice sessions, yet they underscore the need for participants to exert
maximum effort during practice. However, such high levels of exertion can only be
maintained for a limited duration. Additionally, if sufficient rest does not follow such
intense, deliberate practice, Ericsson and colleagues (1993) suggest that exhaustion and
prolonged cognitive fatigue will result. In spite of this, experience and continued practice
can prompt adaptation and quicker recovery. Therefore, future research should consider
the level of expertise/experience (i.e., degree of adaptation) as a moderating factor in the
onset, development, and influence of cognitive fatigue in sport.
Furthermore, future investigations should explore how various competition-related
variables may influence the development of fatigue. For example, the addition of
magnified performance pressures (e.g., tied or championship games) increase the number
of stressors that must be dealt with, reducing the amount of energy and resources
available for task completion and the management of fatigue (Henschen, 2000).
Additionally, performance-relevant factors of feedback and reinforcement can influence
an athlete' s level of enj oyment, mood, and perceptions of control, and therefore may have
a moderating influence on fatigue (Henschen, 2000). Consequently, future research
should explore how fatigue develops in the presence of other performance-relevant
stressors through the variation of task constraints and performance feedback (e.g., time
Preliminary evidence suggests that there may be interactive effects between
physical and cognitive fatigue, with physical fatigue moderated by psychological states,
namely cognitive fatigue (Holding, 1983). Therefore, following the development of a
better understanding of the construct of cognitive fatigue in isolation, the next step would
be to examine the interactive effects of cognitive and physical fatigue. By constructing
three experimental groups, namely, (1) cognitive fatigue only, (2) physical fatigue only,
and (3) the combination of cognitive and physical fatigue, the independent and interactive
effects of fatigue could be delineated.
A similar design may also be applied to better understanding the influence of
practicing in blocks within a training session to reduce the development of fatigue (both
physical and mental). By isolating mental practice (e.g., observational learning, mental
imagery, reviewing of film), the overall cognitive demands are reduced, allowing the
athlete to engage in more effective training, than physical practice alone, which requires
resources to be divided among both cognitive and physical tasks simultaneously (Wulf &
Shea, 2002). Additionally, the intervals between blocks, as well as the alternation of
demands, may provide sufficient rest and recovery, minimizing fatigue. Therefore, future
investigations should explore the effectiveness of alternating between physical and
mental practice to reduce the development of fatigue, as well as monotony, a promoter of
fatigue (Wulf & Shea, 2002; Henschen, 2000).
Although the current investigation's findings are preliminary, the potential practical
implications of these, as well as the results of future investigations, are promising.
Specifically, in the current investigation performance (i.e., response accuracy) was not
influenced by cognitive fatigue or time-on-task yet, fluctuations in arousal were evident.
Such changes suggest that participants were working harder and less efficiently (i.e.,
increased arousal) later in the session, a pattern that cannot be maintained without
eventual consequence (Henschen, 2000). Therefore, by understanding the effects of
cognitive fatigue, athletes may become better equipped to maximize their performance
potentials through modifications in training, and an increased ability to manage cognitive
Through greater understanding of the construct of cognitive fatigue, it may be
possible to reorganize training and practice regimens, as well as coaching philosophies.
In the most basic sense, coaches and athletes can become more aware of how players are
affected by the cognitive demands associated with practice and competition. Specific
consideration may be given to how intermissions should be handled, as well as attention
directed to preparations both the night prior to competition, as well as immediately
preceding performance, to ensure that each player has had a period of sufficient rest
necessary to eliminate any resonating cognitive fatigue (Macchi et al., 2002).
Additionally, by providing sufficient rest during practice, competitions, and the off-
season it may be possible to avoid athlete burnout, which often results from excessive
physical and mental training demands (Henschen, 2000). Understanding the effects of
cognitive fatigue on performance can open the doors for related training practices and
learning how to cope when confronted with a fatigued state.
By manipulating or inducing organismic constraints (e.g., fatigue) during practice
sessions, players may learn to become more attentive to fatigue symptoms and be taught
to successfully inhibit the development and/or influence of fatigue. Specifically, training
programs may include biofeedback techniques to increase awareness of the physiological
changes associated with cognitive fatigue. That is, through practice and awareness
training, athletes become capable of adapting to fatiguing conditions (Ericsson et al.,
Additionally, practice sessions could alternate between physical and mental
training. Wulf and Shea (2002) recommend division of skills into physical practice and
observational learning sessions to reduce the amount of cognitive resources used. In turn,
training session can be prolonged, while further increasing the effectiveness and
efficiency of practice.
The maintenance of appropriate attention is not only important for task
performance (Nideffer, 1993), but also for athlete safety. Specifically, reduced peripheral
attention or misdirected attention, attentional biases commonly associated with stressors
(e.g., fatigue), have been suggested to reduce an athlete's ability to respond maximally to
potentially threatening or dangerous stimuli (Williams, Tonyman, & Andersen, 1991;
Wegner, 1994). Therefore, an indirect benefit of coping with and early identification of
fatigue may be injury reduction. By effectively coping with or avoiding the development
of fatigue, athletes will be less susceptible to distraction, and thereby able to maintain
appropriate focus and awareness, potentially reducing the likelihood of incurring an
injury as the result of misdirected or reduced peripheral attention.
Again, it is important to emphasize that the findings of this study are preliminary,
yet they open the door for numerous future investigations and hold promise for
performance enhancement applications.
Cognitive fatigue is a concern within virtually any performance domain; however,
minimal research has examined the construct within a sport context. Specifically, prior to
the current investigation, there has been no evidence to indicate how prolonged decision-
making and resulting cognitive fatigue may be manifested physiologically, subj ectively,
or behaviorally in athletes. Therefore, the present study assessed changes in athletes'
performance, subj ective reports of performance states, and physiology in relation to time-
on-task within a simulated, dynamic, ice hockey scenario, using a multi-method
approach. Results indicated that a sport simulation and cognitive tactical task were
sufficient in promoting the development of cognitive fatigue. Though the findings
counter the original hypotheses, illustrating increased arousal and decreased response
time under prolonged task-exposure, it is suggested that cognitive fatigue is distinctly
manifested physiologically, subjectively, and behaviorally in athletes. In sum, although
continued research is necessary to further advance the current understanding of the onset
and influence of cognitive fatigue in sport, the current investigation was successful in
revealing preliminary evidence concerning the relationship among subjective perceptions
of cognitive fatigue, time-on-task, performance, attention, and arousal.
E pLat a66peo:
E~ye Xohop: Ao \you meuap Xoppeyztim hevaeo (yhaocea op
Ape \you Xpppenly, Texvy avy, pLe~tqutov (omep Z7E XoZ)vZEP op xnpecypIxxxov)?
IQ VEL, S0oe Z7E pLE~Iqurtov 7pmE a Spomaoy, m6E E ~EX?
2. IXE HOXKE\Y BAXKTPOYNA
Hxygear AEmEh o@ Xogn~TEZrIOV (E.Y., XohhcyturE):
a) Tozah # yIEapo sitaytvy: p) RAme 178 poXKEy yampe XopLTEZE6 Iv (MovZp
X) Amepays # o~ Iouop npayrgXe6 mOEErly SUpIvy IoZ)p hame ceacov o~ sitay:
6) Amepays # o~ XopLneratove Re~p mOEEK SUpIvy IoZ)p hame ceacov o~ sitay:
E) Amepays # o# ollt)To sep ~nepto6 Septvy youp hame ceacov o~ witay:
~) Amepays olltz h kv~yql Sptvy youp hame ceacov o~ witay:
y) Haume \you pXeetms6 avy, amuap80 op moov avy, Xpaym~ovolmeo
I~ \ye, dehace herz Z7E 3 pLoor peXEVZ (tyV~h)vy Z7E cEaCIov & hEmEh o~ nhy):
a) Ov amepaye, loow gavy, Iopoop o# 01sx So \you yeT eaX7 vtmz?
(3) Axx~potLahy purly ow gavy, Iopoop o# 0sen St6 \you 7pumie haoo vtygT?
X) Axx~potLahy purly ow gavy, Iopoop o# 0sen So \you yeT Z7E vigs anptop to a
6) RoohG yougovatGEp \youpoel a pLopvivy sepoov OP a vigs oml ?
E) Hame \you XovougeSG Xaggetve (t.e., co6a, XoggEE, ZEa, X70oyolmE) Iv Z7E haoo 4
f) Hame \you XovougeSG avy, aulloho tv Z7E haoo 24 Iloupo?
g) O~EV moao Z7E haoor ZIgE nlz \you aze a pLeah?
h) Howo gavy, Iopoop sep mOEE, ov amepaye, So \you wiay, stio yagLeo?
i) Please rate your overall video game playing ability:
1 2 3 4
j) Please rate your familiarity with ice hockey video games:
1 2 3 4 5
Never Played Some Experience
INFORMED CONSENT DOCUMENT
Protocol Title: The onset and effect of cognitive fatigue on simulated sport performance.
You are being asked to participate in a research study. Before you give your consent to
volunteer, it is important that you read the following information and ask as many
questions as necessary to be sure you understand what you will be asked to do.
Melanie Mousseau, B.S.
University of Florida
Department of Exercise and Sport Sciences
Christopher Janelle, Ph.D.
University of Florida
Department of Exercise and Sport Sciences
Purpose of the research study: To determine the effects of mental fatigue on sport
What you will be asked to do in the study: You will be asked by the Primary
Investigator listed above to:
-Meet in the Motor Behavior and Human Performance Laboratory at the
University of Florida located in the Florida Gym room 132 at a predetermined
-Read and complete this document, stating that you understand what will be
required of you as a participant in this study and are willing to participate.
-Complete a brief questionnaire regarding how you feel.
-Complete a demographic questionnaire. You have the right to withhold any
information that you do not wish to share.
-Learn how to interact with the simulation task. Sit in front of a screen and wear a
headband that communicates with the eye tracking system.
-View a series of hockey scenarios, answer multiple-choice questions pertaining to
the scenarios, and rate how you feel during the task.
Time required: 90 minutes maximum.
Risks and Discomforts: There are no anticipated risks or discomforts associated with
Benefits: Potential benefits include a better understanding of your mental fatigue
thresholds and what you can do recognize and counteract the potential negative
performance effects of fatigue. Additionally, you will gain insight into how you make
decisions during a game.
Compensation: You will not be compensated for your participation; participation is
Confidentiality: Your identity will be kept confidential to the extent provided by law;
any information obtained through this testing that can be identified with you will remain
Voluntary Nature of Participation: Your participation in this study is completely
voluntary. Your choice of whether or not to participate will not influence your future
relations with the University of Florida. If you decide to participate, you are free to
withdraw your consent and to stop your participation at any time without penalty or loss
of benefits to which you are entitled.
Questions about the Study: If you have any questions about the study or results, please
ask or contact:
MELANIE MOUSEAU, BS, Graduate Assistant Sport and Fitness Program,
Department of Exercise and Sport Sciences, Florida Gym Room 121, 392-0580 xl369,
CHRISTOPHER M. JANELLE, Ph.D.; Director Performance Psychology Laboratory,
College of Health and Human Performance, Florida Gym Room 132E, 392-9575 xl270,
Whom to Contact about your Rights as a Research Participant:
UJFIRB Office, Box 112250, University of Florida, Gainesville, FL 32611-2250; phone
As a representative of this study, I have explained the purpose, the procedures, the
benefits, and the risks that are involved in this research study.
Primary Investigator: Date:
I have read the procedure described above. I voluntarily agree to participate in the
procedure and I have received a copy of this description.
INSTITUTIONAL REVIEW BOARD APPLICATION
1. TITLE OF PROTOCOL: The onset and effect of cognitive fatigue on simulated
2. PRINCIPAL INVESTIGATOR(s):
MELANIE MOUSEAU, BS, Graduate Assistant Sport and Fitness Program,
Department of Exercise and Sport Sciences, Florida Gym Room 121, 392-0580 xl369,
3. SUPERVISOR (IF PI IS STUDENT):
CHRISTOPHER M. JANELLE, Ph.D.; Director Performance Psychology Laboratory,
College of Health and Human Performance, Florida Gym Room 132E, 392-9575 xl270,
4. DATES OF PROPOSED PROTOCOL:
From: January 2003 To: July 2003
5. SOURCE OF FUNDING FOR THE PROTOCOL: None
6. SCIENTIFIC PURPOSE OF THE INVESTIGATION: To determine the effects of
cognitive fatigue on sport performance as indicated by response time and accuracy.
7. DESCRIBE THE RESEARCH METHODOLOGY IN NON-TECHNICAL
LANGUAGE. Consent from participants will be obtained prior to inclusion. An initial
assessment of fatigue, boredom, and motivation will be taken using a self-report visual
analog scale (VAS) representation of the constructs of interest. Participants will then
complete a demographic questionnaire. Following completion of the demographic
questionnaire, participants will be introduced to the task and given a period of time to
acquire familiarity with the testing process and demands of the task. Specifically,
participants will be seated in front of a large proj section screen on which simulated ice
hockey scenarios will be presented. These scenarios will be brief with a duration ranging
from 10 to 30 seconds. At the completion of each scenario, a short question related to the
scenario will appear on the screen, along with response options (e.g., The right winger
should ... A) Pass Left B) Skate with the puck C) Pass Right D) Pass Center). During
this skill acquisition period, participants will also become familiar with the computerized
VAS used to assess fatigue, boredom, and motivation during the task. Once the
participant has become familiar with the testing procedure. An eye tracking system will
be calibrated to record their eye movements and the actual testing session will begin. The
testing session will not exceed 60 minutes and will included 15 Trial Blocks consisting of