Content uploaded by David Cottrell
Author content
All content in this area was uploaded by David Cottrell
Content may be subject to copyright.
Personal pdf le for
www.thieme.de
For personal use only.
No commercial use, no depositing in repositories.
With compliments of Georg Thieme Verlag
Publisher and Copyright
© 2013 by
Georg Thieme Verlag KG
Rüdigerstraße 14
70469 Stuttgart
ISSN
Reprint with the
permission by
the publisher only
J. Brownsberger, A. Edwards, R. Crowther, D. Cottrell
Impact of Mental Fatigue on
Self-paced Exercise
Int J Sports Med
DOI 10.1055/s-0033-1343402
0172-4622
Physiology & Biochemistry
Brownsberger J et al. Mental Fatigue and Eff ort Perception … Int J Sports Med
accepted after revision
March 04 , 2013
Bibliography
DOI http://dx.doi.org/
10.1055/s-0033-1343402
Published online: 2013
Int J Sports Med
© Georg Thieme
Verlag KG Stuttgart · New York
ISSN 0172-4622
Correspondence
Dr. Andrew Edwards
Institute of Sport & Exercise
Sciences
James Cook University
McGregor Road
4870 Cairns
Australia
Tel.: + 61/7/4042 1653
Fax: + 61/4/4942 1590
andrew.edwards@jcu.edu.au
Key words
●
▶
mental fatigue
●
▶
eff ort perception
●
▶
brain regulation
●
▶
pacing
●
▶
EEG
Impact of Mental Fatigue on Self-paced Exercise
[ 3 ] rather than the brain [ 25 ] . However, recent
studies [ 16 , 26 , 28 ] have suggested that the brain
regulates exercise performance based on aff erent
information from peripheral physiologic recep-
tors that give rise to sensations (e. g. nausea and
thirst among others). This model is supported by
the observation that the brain does not fully
recruit all motor units in the target musculature
even when exercise is performed maximally [ 28 ] ,
suggesting that the brain operates to regulate
physical performance in a process informed by
aff erent-eff erent neuromuscular signalling.
Although a recent study demonstrated that sen-
sations of mental fatigue impeded maximal
endurance performance [ 21 ] , the mechanisms
responsible for this eff ect have not yet been fully
elucidated. It has been suggested that a state of
mental tiredness increases the perceived
demands of a physical task [ 20 ] , which could
explain why individuals tend to perceive physical
tasks to require greater eff ort later in the day [ 2 ] ,
why athletes often attribute poor performance to
mental tiredness [ 4 ] and why military pilots are
reported to experience greater risk when men-
tally fatigued [ 14 ] . Similarly, the challenge of
Introduction
▼
Mental fatigue has recently been demonstrated
to negatively infl uence endurance performance
[ 21 ] and it has been suggested that this fi nding
may be due to an alteration in the perception of
eff ort when individuals are mentally tired [ 20 ] .
However, no studies have yet examined whether
or not mental fatigue infl uences the voluntary
(self-paced or self-regulated) eff ort applied to
exercise challenges, an observation which could
explain the previously observed impairment to
endurance performance [ 21 ] .
It is well known that mental fatigue represents
sensations of tiredness experienced during or
after prolonged periods of cognitive activity
[ 5 , 18 ] . However, as sensations of tiredness often
infl uence decision making, attention, motivation
and the voluntary willingness to resist fatigue
[ 5 ] , it is surprising that studies have rarely con-
sidered the impact of mental fatigue on physical
performance. It is possible the lack of research in
this area is attributable to most researchers sup-
posing that fatigue is a physiological phenome-
non caused by defi ciencies within local muscles
Authors J. Brownsberger
1 , A. Edwards
1 , R. Crowther
1 , D. Cottrell
2
Affi liations
1 Institute of Sport & Exercise Science, James Cook University, Cairns, Australia
2 Psychology, James Cook University, Cairns, Australia
Abstract
▼
The purpose of this study was to examine
whether mental fatigue infl uences the perceived
eff ort required to complete fairly light and hard
eff ort self-paced exercise challenges. 12 partici-
pants completed 2 trials in a randomised cross-
over design. Each participant was required to
complete a time-matched pre-exercise task:
1) a continuous cognitive activity test (EXP con-
dition; n = 12), or 2) a time-matched passive
neutral observation task (CON condition; n = 12).
Following the pre-exercise task, participants
performed 2 consecutive bouts of self-paced
cycling exercise again in randomized order at
fairly light (RPE 11) and hard (RPE 15) eff ort.
Physiological, psychological and EEG indices
were measured throughout both conditions. EXP
participants reported signifi cantly greater sen-
sations of fatigue ( p < 0.01) and demonstrated
greater EEG beta-band activation compared
with CON ( p < 0.01) prior to exercise. Power out-
puts from the exercise bouts were signifi cantly
reduced for EXP in both self-paced: RPE 11
(83 ± 7 vs. 99 ± 7 W; p = 0.005) and RPE 15 (132 ± 9
vs. 143 ± 8 W; p = 0.028) trials. This study demon-
strates that individuals with higher self-reported
sensations of fatigue and elevations of EEG beta
activity in the prefrontal cortex of the brain prior
to exercise produce less work during self-paced
exercise trials than in a control condition, prob-
ably due to an altered perception of eff ort.
Physiology & Biochemistry
Brownsberger J et al. Mental Fatigue and Eff ort Perception … Int J Sports Med
exercise appears to be perceived with greater severity when
individuals experience sensations consistent with mental
fatigue [ 10 , 12 ] . It is consequently plausible that to avoid over-
exertion, a central neural mechanism evokes a voluntary reduc-
tion in self-selected power outputs as a coping mechanism when
individuals are free to self-regulate exercise intensity [ 12 , 26 ] . As
such, reduced power output due to an altered perception of
eff ort could explain why mental fatigue has been reported to
impede maximal endurance performance [ 21 ] .
The combined use of electroencephalogram (EEG) and routine
laboratory performance monitoring systems are rarely utilised
in exercise trials, usually due to the inherent diffi culties of sig-
nal-to-noise issues in EEG measurement during exercise [ 8 ] .
However, due to recent advances in EEG technologies, it is now
possible that EEG data might reveal systematic diff erences in
electro cortical activity between experimental exercise condi-
tions [ 27 ] such as those with or without prior mental fatigue
[ 5 , 20 ] . However, to our knowledge, no studies have yet meas-
ured both physical performance and EEG outputs in the context
of mental fatigue.
Mental tiredness is often self-reported by athletes during train-
ing and by professionals working extended shift hours [ 2 , 30 ] . To
replicate these sensations experimentally, a protocol of sus-
tained vigilance and concentration for 1–3 h is usually employed
[ 5 ] . Such prolonged cognitive activity requires heightened con-
centration and, consequently, is likely to produce elevated beta
wave (13–30 Hz) electrocortical activity in the frontal region of
the brain. Elevated beta wave activity also results from sustained
emotional, attentional and cognitive processes [ 5 , 23 , 24 ] and
this is likely to be responsible for evoking subsequent sensations
of mental fatigue.
The aims of this investigation were to examine 1) whether or not
a condition of elevated mental fatigue infl uences the power out-
put in fairly light and hard eff ort self-paced exercise bouts and
2) to examine selected neural indices (alpha and beta waves) of
electro cortical activity in the pertinent brain region most infl u-
ential in decision-making (i. e., prefrontal cortex: EEG electrode
location F3) [ 23 , 24 ] . A reduction in power output during self-
paced exercise and diff erences in electro cortical neural activa-
tion as a consequence of mental fatigue could provide useful
mechanistic insights to previous observations of impaired
endurance performance [ 21 ] .
Materials and Methods
▼
Participants
12 eligible participants [8 male and 4 female; mean ± SD, age
24 ± 5 yr, height 171 ± 9 cm, weight 71 ± 15 kg, peak oxygen
uptake (V
˙O
2 peak) 56 ± 6 ml · kg − 1 · min − 1 ) were given written
instructions describing all procedures related to the study and
each provided informed consent prior to participation. The
study complies with the established ethical standards for sports
medicine [ 15 ] and was approved by the Research and Ethics
Committee of the University.
All participants were regular exercisers, free of medications (e. g.
antidepressants, antipsychotics) and free of known disease (e. g.
epilepsy, clinical depression) that may infl uence exercise toler-
ance and/or psychological responses [ 21 ] . Each participant com-
pleted all testing sessions over a 4-week period with a 48-h
minimum recovery period between visits to the laboratory.
Experimental design
All participants performed an initial incremental test on a labo-
ratory cycle ergometer (Velotron Dynafi t Pro, RacerMate Inc.,
USA) to assess V
˙O
2 peak with expired air analysis (ML206 Gas
Analyser, AD Instruments, USA). The data acquisition system uti-
lised was a PowerLab 8/35 (AD Instruments, USA). 2 familiariza-
tion sessions were subsequently completed for initial trials of
the cognitively demanding mental fatigue protocol and also to
assess participants’ aptitude to self-pace an exercise bout [ 11 ]
while cycling at a self-regulated pace corresponding to descrip-
tors from Borg’s 6-20 Rating of Perceived Exertion (RPE) scale
[ 6 , 7 , 9 ] . Coeffi cient of variation analysis for test-retest reliability
of power output (W) across 3 familiarisation bouts (RPE 7
CV = 8 %; RPE 13, CV = 5 %; RPE 18, C V = 4 %) demonstrated the
ability of participants to consistently replicate similar perform-
ance outcomes when placed in the same condition [ 7 ] .
Following completion of the familiarization sessions, all (n = 12)
participants completed the 2 trials (EXP and CON conditions) in
an individually randomized order. 48 h prior to each session,
participants were given written instructions to drink 3.5 ml of
water per kg of body weight, sleep for at least 7 h, refrain from
the consumption of alcohol and avoid any vigorous exercise the
day before each visit [ 11 , 17 ] . All testing was conducted in the
morning 2 h after breakfast, for which participants were
instructed to maintain habitual practice, and assessments were
subsequently made of capillary blood glucose concentration to
ensure that glucose was in homeostasis at the start of each trial.
The 2 trials required the participants to complete a 90-min pre-
exercise routine of either 1) a continuous cognitive activity
(CCA) task designed to induce mental fatigue (EXP condition) or
2) a passive neutral observation (PNO) procedure (CON condi-
tion) (20), prior to a 3-min warm up and 2 consecutive self-
paced 10-min bouts of cycling exercise. The two 10-min cycling
bouts were completed at self-selected intensities representative
of fairly light eff ort (RPE 11 – Fairly Light) and hard eff ort (RPE
15 – Hard). Participants were able to self-adjust power output
throughout each exercise bout to sustain the required RPE. An
overview of the main trials is presented in
●
▶
Table 1 . RPE inten-
sities for the 2 trials were deliberately diff erent from those used
in the familiarization sessions, as the purpose of familiarisation
was to introduce the novel concept of pacing exercise bouts in
accordance with transient sensations of eff ort and not exercising
at a pre-assigned power output. The participants were not
instructed as to the specifi c purpose or expected outcome of the
experiment and were encouraged to consider performance for
each component of the protocol as it arose.
Table 1 Each condition was time matched and consisted of pre-exercise
and exercise phases. Both conditions were followed by two 10 min self-paced
exercise bouts of RPE of 11 (Fairly Light) and RPE 15 (Hard). The 2 exercise
bouts were performed in a randomized order. All participants (N = 12) per-
formed both conditions.
90 min
(Experimental
(EXP) or Control
(CON) conditions)
3 min
Warm up
10 min
Bout 1
(RPE 11 or 15)
10 min
Bout 2
(RPE 11 or 15)
PRE-EXERCISE
PHASE EXERCISE PHASE
Physiology & Biochemistry
Brownsberger J et al. Mental Fatigue and Eff ort Perception … Int J Sports Med
Experimental and control conditions
The CCA protocol was designed to induce mental fatigue in
accordance with earlier experiments [ 5 , 21 ] . Briefl y, the partici-
pants were seated in a dimly lit, sound-attenuated room in a
thermo neutral environment (between 20–22 °C, 55 % RH),
0.70 m from a 17 inch PC monitor. In this task, sequences of let-
ters were presented on the computer monitor and participants
were requested to respond to selected letters on a computer key-
board and maintain task vigilance for 90 min. To further promote
the retention of task vigilance across the 90-min period, a $100
cash prize was advertised for the most vigilant participant.
For the PNO protocol in the CON condition, participants watched
2 documentary programmes “World Class Trains – The Al Anda-
lus Express” and “World Class Trains – The Venice Simplon Ori-
ent Express”; Pegasus-Eagle Rock Entertainment, 2004) as a
continuous 90-min sequence under the same viewing condi-
tions as the EXP condition [ 21 ] .
Performance outcomes
Mean power output was recorded from a cycle ergometer for all
RPE exercise bouts in all 5 sessions, and data were analysed
using Velotron coaching software. Heart rates were also recorded
throughout all testing sessions (Polar RS400, New York, USA).
Electroencephalogram recordings and analysis
Prior to each condition, an EEG-cap (Electro-cap International,
Inc., USA) with 16 Ag-AgCl electrodes arranged in the interna-
tional 10-20 system was fi tted to the participants’ head. In
accordance with similar experiments examining fatigue and
decision making [ 23 , 24 ] , electrode sites in the left hemisphere
were selected for detailed analysis in the main area of interest
(frontal region of the cortex: electrode F3) and, for comparison
of localised cortical activation, an electrode from the parietal
region of the left hemisphere cortex (P3) was also analysed [ 8 ] .
The EEG-cap was permeable to air in order to prevent an increase
in heat during cycling. Electrodes were fi lled with Signa-gel
TM
(Parker Laboratories Inc., USA) for optimal signal transduction. A
Biosemi-Active-Two amplifi er (Biosemi Inc., Amsterdam, Neth-
erlands) was used with high and low pass fi lters across a fre-
quency range from 0.16 to 100 Hz. All EEG recordings were
sampled in 30 s epochs [ 23 ] . EEG was processed and analysed
using BESA software (version 5.2 MEGIS software GmbH, Ger-
many) and artefacts due to cardio rhythms, body movement and
eye blinking were eliminated. A power spectrum was calculated
from the EEG signal using Fast Fourier Transformation and
exported in alpha (α) band (8–13 Hz) and beta (β) bands (13–
30 Hz) for analysis [ 27 ] .
EEG outputs sampled across EXP and CON trials were subjected
to detailed analysis at diff erent intervals for comparative pur-
poses between conditions and across time: 1) at 5-min (settling
period of pre-exercise phase), 2) 45-min (at the mid-point of the
pre-exercise phase), 3) 85-min (at the end-point of pre-exercise
phase), 4) during the mid-point of the warm up and 5) the mid-
point of the exercise bouts.
Psychological scales
Visual analogue scales (VAS) have previously been demonstrated
to be a valid and reliable means of conveniently assessing a vari-
ety of psychological constructs [ 20 ] . The practicality and rapid-
ity of VAS measurements were an important consideration in
this experiment so as to not interrupt the sequential, continuous
protocol. Therefore participants were asked to indicate by plac-
ing a mark on a 10 cm line their mood (from negative to posi-
tive), their perceived level of fatigue (from not at all to completely
exhausted) and their motivation to exercise (from very low to
very high) in accordance with other studies [ 1 , 27 ] . All VAS
assessments were made immediately prior to the start of the
protocol, at the conclusion of the 90-min pre-exercise phase and
immediately at the end of the protocol.
Blood sampling
Prior to EXP and CON trials, participants gave a 0.6 μl sample of
whole fresh blood from a fi ngertip for assessment of glucose
concentration (ACCU-CHECK Aviva Blood Glucose Meter System,
Roche Diagnostics, Germany). All trials commenced 2 h after
breakfast to ensure blood glucose levels were within normal
range. This was to ensure similar homeostatic regulation and
postprandial rates of food digestion between conditions to mini-
mise glucose-induced mood disturbance.
Statistical analysis
Data from EXP and CON were analysed using repeated measures
of ANOVA (time × condition) with Greenhouse-Geisser correc-
tion when the assumption of sphericity was violated (SPSS ver-
sion 19.0). Post-hoc Tukey tests of honest signifi cant diff erence
were used to examine diff erences when indicated by ANOVA.
Statistical signifi cance was accepted at p < 0.05. 12 par ticipants
were recruited and all completed the EXP and CON conditions of
the experiment. Sample size analysis was calculated on the
dependent variable of power output (W) in response to the
cycling protocol. This was based on a statistical power (1-beta)
of 90 % and alpha of 0.05. A sample size of 8 was required to gen-
erate a large eff ect ( > 0.8) according to Cohen’s d e ff ect size for
the target population. All data are presented as means ± SD
unless otherwise stated. Diagrams are shown as ± SE.
Results
▼
ANOVA indicated that there was a signifi cant main eff ect
( F (1,11) = 25.2, p < 0.001, η
p 2 = 0.70) of the fatigue manipulations
(EXP vs. CON). The interaction between the fatigue manipula-
tion and the exercise bouts was not signifi cant ( F (1,11) = 1.62,
p = 0.23, η
p 2 = 0.13). Post hoc tests indicated that the EXP condi-
tion produced signifi cantly lower self-selected power outputs
for both the RPE 11 (EXP: 83 ± 7 W vs. CON: 99 ± 7 W; p = 0.005)
and RPE 15 (EXP: 132 ± 9 W vs. CON: 143 ± 8 W; p = 0.028) exer-
cise bouts. There was no evidence of learning eff ects. Individual
responses to the exercise bouts are shown in
●
▶
Fig. 1 .
Mean heart rates were similar between conditions in the 90-min
pre-exercise phase for EXP and CON (40.9 % and 41.2 % of maxi-
mum heart rate respectively) (
●
▶
Table 2 ). In response to the 2
exercise bouts, a small but signifi cant elevation of mean heart
rates was observed for CON compared to EXP for the RPE 11 bout
(4.3 % diff erence between conditions; p < 0.05).
All participants’ maintained blood glucose within the normal
homeostatic range at the commencement of the trials (mean:
5.7 ± 0.46 mmol · L − 1 ). There was no diff erence between trials
( p = 0.91; C V = 1.2 %).
Analysis of EEG α-band cortical activity in the prefrontal cortex
sampling site (electrode F3) for both the 90-min pre-exercise
and exercise phases (
●
▶
Fig. 2a ) did not identify signifi cant dif-
ferences between the EXP and CON conditions ( F (1,11) = 4.31,
p = 0.06, η
p 2 = 0.28). There was however a signifi cant main eff ect
Physiology & Biochemistry
Brownsberger J et al. Mental Fatigue and Eff ort Perception … Int J Sports Med
of testing phase ( F (5,55) = 26.6, p < 0.001, η
p 2 = 0.71) and the inter-
action approached signifi cance ( F (5,55) = 3.23, p = 0.058, η
p 2 = 0.23).
Post hoc analysis indicated that both EXP (RPE 11: p = 0.001; RPE
15: p < 0.001) and CON conditions showed signifi cant (RPE 11:
p = 0.002; RPE 15: p < 0.001) and similar increases in α-band
activity at the onset of exercise compared to the start of the pro-
tocol. There was also a trend for reduced α-power in EXP
( p = 0.09) at the mid-point of the 90-min mental fatigue pre-
exercise task.
In contrast, a separate repeated measures ANOVA revealed a sig-
nifi cant main eff ect of fatigue manipulation ( F (1,11) = 90.8,
p < 0.001, η
p 2 = 0.89), a testing phase main eff ect ( F (5,55) = 15.6,
p < 0.001, η
p 2 = 0.59) and a fatigue manipulation x testing phase
interaction ( F (5,55) = 4.92, p = 0.005, η
p 2 = 0.31) for electro cortical
activity of the β-band of the prefrontal lobe (electrode F3). Post
hoc comparisons indicated no signifi cant diff erence in β-band
activity between the EXP and the CON condition in the pre-exer-
cise phase (5 min: p = 0.76). However, the β-band activity was
elevated in the EXP at the mid-point (45-min; p = 0.005), at the
end of pre-exercise phase (90-min: p = 0.004) and during the
warm-up phase prior to exercise ( p = 0.00002) (
●
▶
Fig. 2b ). Dur-
ing both subsequent exercise bouts, the diff erence in β-band
activity between the EXP and CON condition decreased to non-
signifi cant levels (RPE 11: p = 0.08 RPE 15: p = 0.06).
The ANOVA of the parietal lobe sampling site (electrode P3) did
not demonstrate any signifi cant main eff ects or interactions for
α-band (
●
▶
Fig. 3a ). In the β-band (
●
▶
Fig. 3b ) there was a signifi -
cant main eff ect of testing phase ( F (5,55) = 4.77, p = 0.007,
η
p 2 = 0.30).
Reported motivation to exercise and mood were not signifi -
cantly aff ected by the mentally demanding task compared to the
control task. There was, however, a signifi cant change in reported
mental fatigue. Participants reported signifi cantly greater men-
tal fatigue ( p = 0.001) in the EXP condition at the conclusion of
the 90-min pre-exercise phase compared to CON. Immediately
at the conclusion of the 2 exercise bouts, self-reported fatigue
was similar in the EXP and CON conditions (
●
▶
Fig. 4 ) .
Discussion
▼
The main fi nding of this study was that completion of a 90-min
pre-exercise task of continuous cognitive activity (mental
fatigue task) led to the production of less power output during
self-paced exercise than in a control condition. The participants
also reported sensations of mental tiredness and exhibited
greater β-band activation of the frontal lobe (electrode F3) com-
pared to the CON condition prior to exercise. These results sug-
gest the voluntary regulation of eff ort during exercise may not
only be mediated by aff erent physiological information, but also
by central factors such as negative cognitive associations to sen-
sations of mental fatigue immediately prior to exercise [ 12 , 20 ] .
This fi nding supports previous observations [ 20 , 21 ] that exer-
cise performance is impaired when experiencing greater mental
fatigue and suggests this may be attributable to an altered per-
ception of task diffi culty. As mental tiredness is often self-
reported by athletes during training and by professionals
working extended shift hours [ 2 , 30 ] , it is important to recognise
that mental fatigue potentially impedes self-paced exercise per-
formances. Subsequent intervention studies might examine
procedures to alleviate mental fatigue and the consequent eff ect
on exercise, while also considering experimental designs includ-
ing additional placebo/sham conditions to eliminate the possi-
bility that expectancy is a factor in the observations of this
study.
The sustained elevation of β-band electro cortical activity for
EXP prior to exercise compared to CON (electrode F3) indicates
the 90-min pre-exercise procedure was successful in eliciting
greater attention, information processing and cognitive engage-
ment [ 8 , 18 , 19 , 24 ] . This elevation of cortical excitation appears
to have been localised to the prefrontal region of the cortex
where factors such as attention and decision-making are known
to infl uence EEG activity [ 5 ] . The same level of excitation was
not observed from the parietal region (electrode P3), which was
largely unaff ected by testing phase (
●
▶
Fig. 3a, b ). However, this
is commonly the case in such comparisons. The only signifi cant
diff erence between cortical activation from electrodes F3 and P3
was in β-band activity during the 90-min pre-exercise phase.
This is likely to be a consequence of localised (F3) excitation of
Fig. 1 Individual power outputs for the experimental (EXP) and control
(CON) conditions in response to the 2 bouts of 10-min cycling, self-paced
at individuals’ perceptions of eff ort corresponding to RPE 11 (fairly light)
and RPE 15 (hard) descriptors (n = 12). The order of the 2 exercise bouts
was randomised for each condition and for each individual.
165
155
145
135
125
115
105
95
85
75
0EXP CON
RPE 11
Power output (W)
RPE 15
EXP CON
Table 2 Mean heart rate responses to the 90-min pre-exercise phase and the 2 subsequent self-paced exercise bouts.
Participants (N = 12)
Heart rate responses across pre-exercise and exercise bouts
90-min pre-exercise 10-min RPE 11 bout 10-min RPE-15 bout
Mean HR %HR max Mean HR %HR max Mean HR %HR max
EXP condition 80.3 ± 7.5 40.9 118.3 ± 18.7* 60.4 138.9 ± 16.1 70.9
CON condition 80.8 ± 8.1 41.2 126.8 ± 19.8* 64.7 141.8 ± 15.8 72.3
*signifi cant diff erence between conditions; p < 0.05
Physiology & Biochemistry
Brownsberger J et al. Mental Fatigue and Eff ort Perception … Int J Sports Med
the prefrontal cortex during the mental fatigue task and pro-
vides evidence that the protocol elicited greater regional brain
activation in an area likely to be implicated in decision-making
regarding power output during self-paced exercise. Clearly, as no
other studies have yet examined EEG in response to self-pacing
and self-selected work rates, this observation should be consid-
ered as a preliminary fi nding in need of further substantiation.
As the 90-min pre-exercise phase of the protocol for the CON
condition required only passive observation, it is unsurprising
there was no evidence of shift from α (passive/resting) to β
(active) cortical energy during that 90-min period. This is con-
sistent with aims of the passive task of the CON condition requir-
ing minimal attention and it therefore appears to have acted as
an appropriate control condition [ 19 , 22 ] . It seems plausible that
when participants undertook the EXP condition, the sustained
eff ort of maintaining signifi cant attention for 90-min evoked an
increased use of mental ‘energy’ that then induced mental
fatigue. Self-reported sensations of mental fatigue (
●
▶
Table 3 )
were reported in the EXP condition prior to exercise, and the
combination of these factors with sustained greater β-band acti-
vation could explain the subsequent reduction in self-paced
exercise outputs compared to the CON condition. As such, the
mental consequences of the 90-min mental fatigue task may
have carried forward sensations of tiredness into the self-paced
exercise bouts, thereby infl uencing the magnitude of the eff orts
generated.
In the CON condition, an increase of β-band electro cortical acti-
vation in the prefrontal cortex was observed only after the
90-min of passive television observation (
●
▶
Fig. 2b ). This eff ect
is probably attributable to the greater mental engagement in the
self-paced exercise tasks compared to the observation task,
along with the physiological up-regulation of metabolism [ 13 ] .
The role of catecholamines for example, in the fi ght-or-fl ight
response is known to up-regulate preparedness for physical
challenges [ 13 , 29 ] . An abrupt increase of generalised EEG activ-
ity would therefore be anticipated at the onset of exercise
[ 8 , 24 , 27 ] .
Despite the observation of signifi cant diff erences in β-band EEG
outputs (F3) during the mental fatigue task in this study, it is
important to note that EEG recordings in response to exercise
are often problematic and prone to interference from external
noise [ 8 ] . Standard EEG artefact elimination control measures
were applied in this study [ 27 ] , and so while signifi cant pre-
exercise diff erences were noted, other more reliable brain indi-
ces such as fMRI may reveal systematic diff erences during
exercise that were not identifi ed in this study [ 5 ] . Due to the
limitations of EEG it is therefore not possible to draw a defi nitive
conclusion as to whether or not diff erences exist in regional
brain activation during exercise when mentally fatigued. Per-
formance and self-pacing are compromised, but the mechanism
for this eff ect requires further elucidation.
As the performance outcome (e. g. power output) for both exer-
cise bouts was impeded in the EXP condition (RPE 11 bout;
p = 0.005 and RPE 15 bout; p = 0.028) compared to CON (
●
▶
Fig. 1 ),
it is likely that the 90 min of attention requiring tasks prior to
exercise carried residual eff ects of mental fatigue into the subse-
quent self-paced exercise bouts. This observation indicates that
the brain not only considers aff erent peripheral physiologic
Fig. 2 a , b Log transformed α- and β-band power
(μV
2 ) for position F3 in both the 90-min pre-exer-
cise and exercise phases of the protocol. EXP = ex-
perimental condition (n = 12) and CON = control
condition (n = 12). *signifi cant between condition
eff ect ( p < 0.05), **signifi cant between condi-
tion eff ect ( p < 0.01). # signifi cant time eff ect for
intra-condition time eff ect ( p < 0.05), ## signifi cant
time eff ect intra-condition time eff ect ( p < 0.01).
Diagrams shown as ± SE.
1.8
1.7
1.6
1.5
Alpha power (log µV
2
)
1.4
1.3
1.2
1.1
1
0Start Mid End Warm up RPE11
EXP
#
##
CON
RPE15
Exercise phase90-min pre-exercise phase
0.8
0.7
Beta power (log µV
2
)
0.6
0.5
0Start Mid End Warm up RPE11
EXP
#
**
**
*
CON
RPE15
Exercise phase90-min pre-exercise phase
a
b
Physiology & Biochemistry
Brownsberger J et al. Mental Fatigue and Eff ort Perception … Int J Sports Med
information such as muscle soreness, changes in pH or respira-
tory discomfort [ 25 ] when determining appropriate eff ort in
self-paced exercise, but also considers psychological factors such
as perceived tiredness [ 11 ] . This supports the previously pro-
posed concept that alterations to the perception of eff ort when
mentally fatigued infl uences performance attenuation and the
eff ect of brain regulation over exercise performance [ 21 , 25 ] .
Heart rates between conditions were similar in the pre-exercise
phase of the experiment (EXP: 80.3 ± 7.5 vs. CON: 80.8 ± 8.1 b/
min), indicating that underlying physiological stress was similar
for that period (~40 % of maximal heart rate) (
●
▶
Table 2 ). Conse-
quently, sensations of fatigue derived from this pre-exercise
period would not have had a physical cause. This strongly sug-
gests that any subsequent performance impediment in the self-
paced tasks were the result of mental and not physical factors
[ 20 ] . Analysis of mean heart rates for both subsequent exercise
bouts indicated they were all performed within the normal
range observed for the specifi ed RPE descriptors (RPE 11:
60–65 % of maximum heart rate) and RPE 15: 70–75 % of maxi-
mum heart rate) [ 25 ] . Yet the CON condition produced slightly
elevated mean heart rates and signifi cantly so for the lower
intensity exercise bout (
●
▶
Table 2 ; p < 0.05). This observation
supports the fi nding of greater power outputs for CON across
both RPE 11 and RPE 15 bouts (
●
▶
Fig. 1 ), which demonstrates
that heart rate responds to changes in physical work, such that
the greater muscular work (power output) achieved in the CON
condition evoked a subsequent requirement for greater cardio-
vascular support [ 25 ] .
Task motivation was unaff ected by the 90-min mental fatigue
condition in this experiment. This indicates that the monetary
incentive (prize) off ered for the best performance in the mental
fatigue task was eff ective in eliciting sustained motivation on
the attention-demanding activity and that participants did not
disengage from it. Mental disengagement from the experiment
could have indicated that boredom [ 5 ] rather than mental
fatigue was the most plausible explanation for the observed
eff ect on self-paced exercise. As motivation was well maintained
it seems that this was unlikely in our experiment.
Fig. 3 a , b Log transformed α- and β-band power
(μV
2 ) for position P3 in both the 90-min pre-exer-
cise and exercise phases of the protocol. EXP = ex-
perimental condition (n = 12) and CON = control
condition (n = 12). Diagrams shown as ± SE.
1.8
1.7
1.6
1.5
1.4
Alpha power (log µV2)Beta power (log µV2)
1.3
1.2
1.1
0.9
0.8
0.7
0.6
0.5
0Start
0Start Mid
90-min pre-exercise phase
End Warm up RPE15
EXP CON
RPE11
Exercise phase
Mid
90-min pre-exercise phase
End Warm up RPE15
EXP CON
RPE11
Exercise phase
1
a
b
Fig. 4 Mean rated a fatigue, b mood and c moti-
vation to exercise at pre-test, mid-test (following
the 90 min pre-exercise phase) and post exercise.
Diagrams shown as ± SE.
10
Rated Fatigue (max 10)
9
8
7
6
5
4
3
2
1
0Pre-test Mid-test Post-
exercise
Pre-test Mid-test Post-
exercise
Pre-test Mid-test Post-
exercise
10
Rated Mood (max 10)
9
8
7
6
5
4
3
2
1
0
10
Rated Motivation (max 10)
9
8
7
6
5
4
3
2
1
0
EXP CON
ab c
Physiology & Biochemistry
Brownsberger J et al. Mental Fatigue and Eff ort Perception … Int J Sports Med
At the conclusion of the 2 exercise bouts, it was evident that
fatigue had improved in the EXP trial relative to mid-test
(
●
▶
Table 3 ). Improvements to mood states and other positive
associations with fatigue sensations are often reported after
submaximal exercise [ 3 ] . To our knowledge, however, no studies
have examined mood state or fatigue levels after an exercise
bout specifi cally examining responses from individuals who are
experiencing sensations of mental fatigue. Nevertheless, despite
the post-exercise improvement in psychological state, power
outputs for EXP had already been impeded. The psychological
state of the individual prior to exercise is therefore an important
consideration for subsequent successful performance [ 21 ] .
In conclusion, this study demonstrates mental fatigue negatively
infl uences the performance of self-paced exercise at 2 diff erent
intensities of eff ort. This is probably due to an altered perception
of eff ort mediated by a sustained period (90 min) of pre-exercise
mental activity [ 5 , 20 ] . Signifi cant localised β-band activity from
the prefrontal cortex accompanied sensations of mental fatigue
prior to exercise performance. These factors may have contrib-
uted to the impaired performances in the 2 self-paced exercise
bouts. However, the mechanisms of mental fatigue remain
equivocal. Nevertheless, the fi ndings of this study support previ-
ous observations that mental fatigue negatively infl uences
endurance performance [ 20 , 21 ] , and it seems likely that this is,
at least in part, due to an altered perception of eff ort.
The practical signifi cance of the observations from this study
indicate that mental fatigue experienced prior to the completion
of self-paced physical tasks is likely to compromise performance.
This was evident by a 19 % diff erence in performance in the fairly
light eff ort bout (RPE 11) and an 8 % diff erence between condi-
tions in the hard eff ort bout (RPE 15), both of which were statis-
tically signifi cant observations, although the eff ect was more
consistent across individuals in the lighter eff ort trial (
●
▶
Fig. 1 ).
Nevertheless, these observations may lead other researchers to
further explore this novel concept and could have important
implications for the completion of tasks where safety is an issue,
or in any circumstances (e. g. racing) where performance impair-
ment is meaningful. Further work is required to ascertain the
signifi cance of these fi ndings to diff erent population groups and
in response to other modes of exercise.
Disclosure statements: 1. All authors contributed equally to
planning, data collection, analysis and assembly of this manu-
script. The corresponding author is the team leader and project
supervisor. 2. This project was not subject to a grant and was
fi nancially supported by discretionary budget of the Institute of
Sport & Exercise Science for incidental consumables and minor
costs. 3. The authors of this manuscript have no competing
interests.
References
1 Ahearn E P . The use of visual analog scales in mood disorders: A critical
review . J Psychiatr Res 1998 ; 31 : 569 – 579
2 Åkerstedt T , Kecklund G , Johansson S . Shift work and mortality . Chrono-
biol Int 2004 ; 21 : 1055 – 1061
3 Bassett D J , Howley E T . Maximal oxygen uptake: “classical” versus
“contemporary” viewpoints . Med Sci Sports Exerc 1997 ; 29 : 591 – 603
4 Berger B G , Motl R W . Exercise and mood: A selective review and syn-
thesis of research employing the profi le of mood states . J Appl Psychol
2000 ; 12 : 69 – 92
5 Boksem MA S , Meijman T F , Lorist M M . E ff ects of mental fatigue on
attention: An ERP study . Cognit Brain Res 2005 ; 25 : 107 – 116
6 Borg G A . Psychophysical bases of perceived exertion . Med Sci Sports
Exerc 1982 ; 14 : 377 – 381
7 Chen M , Fan X , Moe S . Criterion-related validity of the Borg ratings
of perceived exertion scale in healthy individuals: a meta-analysis . J
Sports Sci 2002 ; 20 : 873 – 899
8 Crabbe J B , Dishman R K . Brain electrocortical activity during and after
exercise: a quantitative synthesis . Psychophysiol 2004 ; 41 : 563 – 574
9 Dunbar C C , Robertson R J , Baun R , Blandin M F , Metz K , Burdett R , Goss
F L . The validity of regulating exercise intensity by ratings of perceived
exertion . Med Sci Sports Exerc 1992 ; 24 : 94 – 99
10 Edwards A M , Bentley M B , Mann M E , Seaholme T S . Self-pacing in inter-
val training: a teleoanticipatory approach . Psychophysiol 2011 ; 48 :
136 – 141
11 Edwards A M , Claxton D B , Challis N V , Chapman J H , Fysh M L . O x y g e n
uptake kinetics determined by PRBS techniques diff erentiate elite
endurance runners from elite sprinters . Int J Sports Med 1999 ; 20 : 1 – 6
12 Edwards A M , Polman RC J . Pacing in sport and exercise: a psycho-
physiological perspective . New York : Nova Scientifi c Publishers , 2012 ;
49 – 70
13 French D N , Kraemer W J , Volek J S , Spiering B A , Judelson D A , Hoff man J R ,
Maresh C M . Anticipatory responses of catecholamines on muscle force
production . J Appl Physiol 2007 ; 102 : 94 – 102
14 Goode J H . Are pilots at risk of accidents due to fatigue? J Safety Res
2003 ; 34 : 309 – 313
15 Harriss D J , Atkinson G . Update – Ethical standards in sport and exercise
science research . Int J Sports Med 2011 ; 32 : 819 – 821
16 Lambert E V , St Clair Gibson A , Noakes T D . Complex systems model of
fatigue: integrative homoeostatic control of peripheral physiological
systems during exercise in humans . Br J Sports Med 2005 ; 39 : 52 – 62
17 Lander P J , Edwards A M , Butterly R J . Self-paced exercise is less physi-
cally challenging than enforced constant pace exercise of the same
intensity: infl uence of complex central metabolic control . Br J Sports
Med 2009 ; 43 : 789 – 795
18 Lorist M M , Bezdan E , Caat M , Span M M , Roerdin JBTM , Maurits N M .
The infl uence of mental fatigue and motivation on neural network
dynamics; an EEG coherence study . Brain Res 2009 ; 1270 : 95 – 106
19 Lorist M M , Klein M , Nieuwenhuis S , de Jong R , Mulder G , Meijman T F .
Mental fatigue and task control: Planning and preparation . Psycho-
physiology 2000 ; 37 : 614 – 625
20 Marcora S , Staiano W . The limit to exercise tolerance in humans: mind
over muscle? Eur J Appl Physiol 2010 ; 109 : 763 – 770
21 Marcora S M , Staiano W , Manning V . Mental fatigue impairs physical
performance in humans . J Appl Physiol 2009 ; 106 : 857 – 864
22 Miller R . Theory of the normal waking EEG: From single neurones
to waveforms in the alpha, beta and gamma frequency ranges . Int J
Psychophysiol 2007 ; 64 : 18 – 23
23 Moraes H , Ferreira C , Deslandes A , Cagy M , Pompeu F , Ribeiro P , Piedade
R . Beta and alpha electroencephalographic activity changes after acute
exercise . Arqde Neuro-Psiquiat 2007 ; 65 : 637 – 641
24 Nielsen B , Hyldig T , Bidstrup F , González-Alonso J , Christoff ersen GR J .
Brain activity and fatigue during prolonged exercise in the heat .
P fl ügers Arch – Eur J Physiol 2001 ; 442 : 41 – 48
Self-perceptions
0–10 VAS scales
Test occasions for EXP and CON conditions
Pre-test Mid-test (90-min) Post-exercise
EXP CON EXP CON EXP CON
fatigue (n = 12) 2.1 ± 2.2 2.2 ± 2.1 6.5 ± 1.6 ##** 4.2 ± 2.3 4.8 ± 2.4 # 4.5 ± 2.4
mood (n = 12) 7.2 ± 1.5 7.3 ± 1.2 5.6 ± 1.9 6.4 ± 1.8 7.1 ± 1.8 7.4 ± 1.2
motivation (n = 12) 6.0 ± 2.8 5.9 ± 2.4 5.4 ± 2.2 4.8 ± 2.4 6.5 ± 2.4 7.0 ± 1.9
# = signifi cant time eff ect from pre-test; p < 0.05, ## signifi cant time eff ect from pre-test; p < 0.01, * signifi cant diff erence between
conditions; p < 0.01
Table 3 Psychological indices
of perceived fatigue and mood
at baseline (pre-test), mid-test
(immediately following the
90-min pre-exercise phase), and
post-exercise intervals.
Physiology & Biochemistry
Brownsberger J et al. Mental Fatigue and Eff ort Perception … Int J Sports Med
28 Swart J , Lamberts R P , Lambert M I , Clairgibson A , Lambert E V , Skowno J ,
Noakes T D . Exercising with reserve: evidence that the central nervous
system regulates prolonged exercise performance . Br J Sports Med
2009 ; 43 : 782 – 788
29 Watt M J , Stellingwerff T , Heigenhauser GJ F , Spriet L L . E ff ects of plasma
adrenaline on hormone-sensitive lipase at rest and during moder-
ate exercise in human skeletal muscle . J Physiol 2003 ; 550 : 325 – 332
30 Yeung R R . The acute eff ects of exercise on mood state . J Psychosom
Res 1996 ; 40 : 123 – 141
25 Noakes T D . Time to move beyond a brainless exercise physiology: the
evidence for complex regulation of human exercise performance . Appl
Physiol Nutr Metab 2011 ; 36 : 23 – 35
26 Noakes T D , Snow R J , Febbraio M A . Linear relationship between the
perception of eff ort and the duration of constant load exercise that
remains . J Appl Physiol 2004 ; 96 : 1571 – 1573
27 Schnieder S , Askew C D , Diehl J , Mierau A , Kleinert J , Abel T , Carnahan
H . EEG activity and mood in health orientated runners after diff erent
exercise intensities . Physiol Behav 2009 ; 96 : 709 – 716
