Physiology & Biochemistry 453

Perceptual Responses in Free vs. Constant Pace

Exercise

Authors M. Garcin 1, M. Danel 2, V. Billat 3

1
Affiliations University Lille 2, FSSEP, Ronchin, France
2
Centre de Recherches, Décathlon, Villeneuve d’Ascq, France
3
Laboratoire LEPHE, University of Evry, Evry, France

Key words Abstract pace runs were both averaged throughout the en-
" perceived exertion
l ! tire period of exercise and without the last lap.
" estimated time limit
l The purpose of the present investigation was to The results did not show any significant effect of
" velocity
l
study the influence of free versus constant pace free versus constant pace on RPE and ETL. Aver-
on perceived exertion (RPE) and estimated time aged oxygen uptake between free and constant
Limit (ETL). Ten athletes performed a graded test pace runs was not significantly different, where-
aimed to determine maximal oxygen uptake as averaged vV̇O2max, % vV̇O2max and time to ex-
(V̇O2max) and the velocity associated with V̇O2max haustion was significantly higher for free pace
(vV̇O2max), a constant run to exhaustion at 90 % runs only for the entire exercise. Consequently,
vV̇O2max to determine the time and distance to compared to the constant pace run, the free pace
exhaustion at this relative velocity, a free paced one only allowed athletes to finish the run by a
run over the distance to exhaustion set by the sprint which was effective in increasing perfor-
time to exhaustion at 90 % vV̇O2max. Oxygen up- mance, but not to perceive the free pacing run as
take and velocity during constant pace and free being less strenuous than the constant pace one.

Introduction distance or duration they had previously per-

! formed.
Although steady state exercise conditions may During closed loop races (runs performed with a
prevail in long-distance running races such as known endpoint, Saint Clair Gibson et al. [35]),
the marathon, most endurance races are charac- subjects have to appropriately distribute energy
terised by multiple changes of pace and intensity resources so that they are exhausted just at the
throughout the duration of the event. Athletes end of the race. Spontaneous velocity throughout
spontaneously choose to modulate their pace the race is therefore a strategy to minimize the
during the race to avoid becoming overfatigued physiological strain [6]. From a practical view-
before reaching the finishing line [5]. Indeed, best point, athletes seek optimal pace strategy allow-
accepted after revision
performances in middle distance running are ing them to run with both the least strain and
June 27, 2007
usually characterised by relatively large intra- stress in order to achieve optimal performance.
Bibliography event variability in velocity (the range of coeffi- Consequently, it may be of interest to attest
DOI 10.1055/s-2007-989237 cients of variation in velocity is from 1% to 5% whether a small variation in exercise velocity
Published online Nov. 14, 2007
Int J Sports Med 2008; 29:
calculated from the last three world records over would result both in significant changes in oxy-
453 – 459 © Georg Thieme 1500, 3000, 5000 and 10 000 m) [9]. Numerous gen uptake and performance, and also in per-
Verlag KG Stuttgart • New York • studies, which have already dealt with velocity ceived exertion. However, only a few authors
ISSN 0172-4622 variability, have examined different patterns of have dealt with the effect of velocity variability
Correspondence pacing and the physiological differences between on psychological factors such as rating of per-
Dr. Murielle Garcin constant-intensity and variable-intensity effort ceived exertion (RPE, from 6 – 20) in their experi-
FSSEP [2,10,13, 20, 27, 28, 33, 38, 40]. ments [5,12,18, 21, 27].
University Lille 2
9 rue de l’université During constant-intensity effort, the subjects This category rating scale (from 6 – 20) described
59155 Ronchin were asked to maintain the constant pace or by Borg in 1970 [7] is the most commonly applied
France power as long as possible, whereas during varia- one in the field of exercise science [31]. Percep-
Phone: + 33 (0) 3 20 88 73 91
Fax: + 33 (0) 3 20 88 73 63 ble-intensity effort, the subjects were asked to tion of physical exertion includes feelings of ef-
[email protected] run or cycle as fast as possible the same over the fort, strain, discomfort and/or fatigue experi-

Garcin M et al. Perceptual Responses in … Int J Sports Med 2008; 29: 453 – 459
454 Physiology & Biochemistry

enced during exercise [26]. A second perceptually-based scale turbine flowmeter was performed using a 3 – 1 syringe (Quinton
regarding subjective estimation of exhaustion time (estimated Instruments®, Seattle, WA, USA). This analyser has previously
time limit, ETL) has also been used in addition to RPE during ex- been validated over a wide range of exercise intensities [22].
ercise to further understand how the subject is feeling [14,15]. Breath-by-breath V̇O2 was averaged every 15 seconds.
The RPE scale provides an estimation of the intensity of the sig-
nals of exertion associated with exercise, whereas the ETL scale Procedures
deals with a subjective prediction of how long the current exer- Subjects performed three tests until exhaustion on an outdoor
cise level can be maintained. synthetic track 400 m long. The first test was a graded exercise
Most of the authors who studied the effect of velocity variability to determine the maximal oxygen uptake (V̇O2max) and the ve-
on RPE reported no significant difference for RPE between con- locity associated with V̇O2max (vV̇O2max). Initial velocity was set
stant and free pace runs [5,18, 21, 27]. Moreover, the effect of ve- according to previous vV̇O2max values measured two months
locity variability on the subjective prediction of time limit has previously, so that exhaustion occurred for each subject within
never been studied. Therefore, the purpose of the present study 20 min [11]. Each subject began the test at a pace of 10 km • h–1,
was to examine the influence of free versus constant pace run on followed by 1 km • h–1 velocity increments every 2 minutes to ex-
RPE and ETL. haustion. Each stage was separated by a 30-sec rest period in a
Saint Clair Gibson et al. [36] suggested that both the generation standing position. Maximal oxygen uptake was defined as the
of exercise intensity and perceived exertion during an event may highest V̇O2 obtained in two successive 15-sec intervals, and
be controlled by the same regulatory processes in the brain. This vV̇O2max as the minimal running velocity maintained for more
very control mechanism which determines both power output than 1 min that elicited V̇O2max [4]. If, during the last stage, an
and RPE would seem to be utilising the same scalar time param- athlete achieved V̇O2max that was not sustained for at least
eters set by knowledge of the distance to cover and memory of 1 min, the velocity during the previous stage was retained as
similar prior exercise bouts [36]. Consequently, we may suggest his vV̇O2max. If this velocity resulting in fatigue was only sus-
that, during closed loop races, perceived exertion would simi- tained for ‡ 1 min and < 2 min, then vV̇O2max was considered to
larly follow the velocity variations which could occur during free be equal to the velocity during the previous stage plus half the
pace runs. Therefore, it is hypothesised that RPE and ETL are in- velocity increase between the last two stages (i.e., 1 km • h–1/
fluenced by velocity variability, i.e., that athletes perceive the ex- 2 = 0.5 km • h–1) [19].
ercise as being lighter and feel that they can endure a longer Two days later, the subjects performed one constant pace run at
time duration during a free pace run than during a constant pace 90 % vV̇O2max to determine the time to exhaustion and distance
one. to exhaustion (i.e., the maximal distance covered at exhaustion)
at this relative velocity. This velocity was chosen because it is
commonly used in training programmes to improve V̇O2max
Methods [32]. During this run, the subjects were asked to maintain the
! constant pace as long as possible. Maximal performance (i.e.,
Participants time and distance to exhaustion) was indicated to the subjects.
Ten male trained endurance runners (39 ± 10 years; 70 ± 5 kg; Four days later, they performed one free paced run over the dis-
173 ± 4 cm) participated in the study. These subjects were long- tance to exhaustion set by the constant pace run at 90 % vV̇O2max.
distance runners training for the semi-marathon. They were Therefore, they performed their free pace run over the same dis-
chosen to avoid a long-term planned race strategy and to obtain tance to exhaustion to compare the average velocity and the ex-
a stochastic pace (i.e., a variation in pace involving probability haustion time with the constant and free pace trials. For the free
arising from chance). They trained five times per week (70 ± pace run, the runners were asked to run as fast as possible over
20 km • wk–1) and were medically examined before they signed the same distance they had previously covered at 90 % vV̇O2max.
an informed consent form about the purpose and procedures of During the test, the subjects had no knowledge of their veloc-
the experiment. The approval of the Comité Consultatif de Pro- ities.
tection des Personnes pour la Recherche Biomédicale de Lille During the two days separating these tests, the subjects were
was obtained for tests. asked either to rest or to do light training (i.e., 30 min at 60 % of
vV̇O2max). The constant and free pace runs were preceded by a
Materials standardised warming-up period of 20 min (15 min of jogging
Perception was expressed according to two scales: a French at 50 % vV̇O2max and 5 min of stretching). For each test, each sub-
translation [37] of the rating of perceived exertion scale (RPE) ject was verbally encouraged to give maximum effort. During
[7], which consisted of 15 assessments between 6 and 20 (from these runs, the subjects could not use any kind of timing device.
“very very light” to “very very hard”), and a second scale based These exercises were performed on the same running track and
on subjective estimation of time limit (ETL) [15], which con- at the same time of day.
sisted of 20 assessments between 1 and 20 (from “more than 16 Velocity was checked during the incremental and the constant
hours” to less than “2 minutes”). This scale was designed as a exercises by the experimenters. On the track, the athletes fol-
function of the logarithm of the estimated exhaustion time lowed a pacing cyclist travelling at the required velocity. The cy-
(tlim) (ETL = 21 minus 2 n, with n = log2 [tlim] where tlim was clist received audio cues via a Walkman (Sony®, Paris, France),
expressed in min). The validity [15] and the reliability [16] of the cue rhythm determining the velocity needed to cover 20 m.
the ETL scale have previously been attested. Visual marks were set at 20-m intervals along the track (inside
Oxygen uptake (V̇O2) was measured using a portable system the first lane) [5]. The cyclist had to position himself in front of
(Cosmed® K4b2, Rome, Italy). Before each test, the O2 and CO2 each mark at each audio cue to allow the runner to be at the true
analysis systems were calibrated using ambient air and a gas of pace. At the end of each step in the incremental test, the subjects
known O2 and CO2 concentrations. The calibration of the K4b2 were informed of the beginning of a new step. For the constant

Garcin M et al. Perceptual Responses in … Int J Sports Med 2008; 29: 453 – 459
 Physiology & Biochemistry 455

pace run, the cyclist riding in front of the subject imposed the Statistical analysis
velocity, whereas during the free pace run, the cyclist rode next Results are presented as mean (M) ± standard deviation (SD) val-
to the subject so that the latter could impose the velocity. For the ues.
graded and constant run exercises, exhaustion was defined During the free-pace runs, the time (t) required to cover 20 m
when the subject was unable to sustain the velocity, i.e., when was recorded. The running velocity between two successive in-
the runner was more than 5 m behind the cyclist, whereas dur- terval marks was thus computed as v (m • sec–1) = 20 (m)/t (sec).
ing the free pace run, subjects were informed by external experi- Consequently, 143.1 ± 46.4 velocity values were collected during
menters at the end of the run that their target distance had been free pace runs.
covered. Each subject was verbally encouraged to give maximum Oxygen uptake and velocity during constant pace and free pace
effort for each test. Moreover, two experimenters independently runs were averaged over the entire period of exercise and with-
measured the time required to complete 20 m in order to check out the last lap (i.e., the last 400 m). Statistical differences for
the pacer’s and runner’s velocity with a chronometer (Digisport averaged V̇O2, velocities and exhaustion times between the con-
Instruments®, Seyssins, France). Since it was operated manually, stant and free pace runs were tested with a Student’s t-test for
the registration of the two operators was checked for similarity. paired data.
During previous training sessions, the subjects were familiarised As there is a large interindividual variability of time to exhaus-
with both scales and a copy of the scales was provided to each tion or distance to exhaustion [4], the 2-minute fixed measure-
subject. Instructions on the scales were read by the subjects ment period could correspond to a higher or lower percentage of
and the scales were explained to each of them before every exer- maximal distance (i.e., distance to exhaustion) per subject.
cise, as recommended by Noble et al. [25], to help participants to Moreover, this 2-minute fixed measurement period may not
link their full exercise stimulus range with their full ratings of have corresponded to the same distance between exercise mo-
perceived exertion response range [17]. dality for a given subject. Consequently, in order to compare
Instructions for the scales were given straight after each other. RPE and ETL between constant and free pace runs, we expressed
These scales were written on a board fixed on the back of the ex- the running distance as a percentage of distance to exhaustion.
perimenter who rode in front of the subject during the incre- For the constant and free pace runs, RPE and ETL values were cal-
mental and constant exercises, and who rode next to the subject culated for each participant at 20, 40, 60, and 80 % distance to ex-
during the free exercise so that subjects could read each of these haustion by regression analysis.
scales. The subjects were asked “How heavy and strenuous does Thereafter, statistical significance according to RPE and ETL val-
the exercise feel to you?” [8] and “How long would you be able to ues was studied by means of a two-way repeated factor ANOVA
perform an exercise at this intensity to exhaustion?”. During with two repeated factors (distance and pace) and if necessary
both exercises, ratings were collected by a second experimenter completed by the Tukey post hoc test.
who rode next to the runner. For the incremental exercise, sub- Velocity variability was presented as the coefficient of variation
jects had to give ratings corresponding to their sensations during (100 × SD/mean velocity calculated per 20 m-interval). This val-
the last 15 sec of each stage. They had to point to a value on the ue corresponded to the part of SD relative to the average velocity
perceived exertion scales and the ratings were collected during (i.e., to a percentage of average velocity).
the 30-sec rest. For the constant and free pace exercises, the pro- Data were analysed with Sigma Stat® (Jandel, Germany). For all
cedure was the same but subjects expressed the perceived exer- analysis, the level for significance was set a priori at .05.
tion values with their fist (= 10 points) or their fingers (each
one = 1 point) every 2 min up to the end of exercise. This 2-min-
ute time-delay allowed subjects to appraise their feeling of exer- Results
tion and give a number on both scales. The order of RPE and ETL !
was the same during both exercises for each subject but was Maximal oxygen uptake and vV̇O2max were 55.0 ± 7.5 ml • min–1 •
counterbalanced between subjects in order to eliminate any ef- kg–1 and 17.2 ± 2.0 km • h–1, respectively. Physiological and per-
fect of the order in which the scales were presented on percep- formance values are presented in l " Table 1. Averaged velocities

tual responses. and exhaustion times were only significantly different between
constant and free paces for the entire exercise (p < .001, l" Table

1). Averaged V̇O2 remained the same between the two exercise
" Table 1). The percentage of vV̇O
conditions (p > .05, l 2max was

Table 1 Mean and standard deviations values of distance limit (dlim), velocities, exhaustion times (tlim) and oxygen uptake (V̇O2) during constant and free pace
runs, for entire exercise and without the last lap (400 m), for endurance-trained runners (n = 10)

dlim (m) Constant Free pace tlim at constant tlim at free V̇O2 at V̇O2 at free pace
pace (km • h–1) (km • h–1) pace (sec) pace (sec) constant pace (ml • min–1 • kg–1)
(ml • min–1 • kg–1)
For entire 2871 ± 930 15.5 ± 1.8 * 16.2 ± 1.8 657 ± 154 * 630 ± 161 51.9 ± 6.9 52.7 ± 7.5
exercise
Without the 2471 ± 930 15.5 ± 1.8 15.4 ± 1.7 564 ± 158 569 ± 169 51.0 ± 7.0 52.2 ± 7.6
last lap

Constant pace was run at 90 % of the velocity associated with maximal oxygen uptake. * Significantly statistically different, p < .001

Garcin M et al. Perceptual Responses in … Int J Sports Med 2008; 29: 453 – 459
456 Physiology & Biochemistry

locity variations occurred during the race, which ended with a

Table 2 Ratings of perceived exertion (RPE) and estimated time limit (ETL)
according to type of exercise (constant vs. free pace runs) and percentage of
final sprint during the last lap.
distance limit (20, 40, 60, 80% dlim) in male trained endurance runners
(n = 10)
Discussion
RPE ETL
!
mean SD mean SD
The main findings of this study were that RPE, ETL and averaged
Constant pace run
V̇O2 were not influenced by the variability of the velocity,
20% dlim 13.9 1.0 14.7 1.6
whereas performances (exhaustion times and velocities) were
40% dlim 15.9 0.7 16.6 1.4
different for the free versus constant pacing during an exhaust-
60% dlim 17.0 0.9 17.7 1.4
80% dlim 17.8 1.0 18.5 1.5 ing run over a distance to exhaustion at 90 % vV̇O2max. This result
Free pace run means that athletes perceived the free pace exercise as being as
20% dlim 14.2 1.1 14.3 1.4 strenuous as the constant pace one. They also felt that they could
40% dlim 15.9 0.9 16.3 1.1 endure the free pace exercise as long as the constant pace one,
60% dlim 16.9 1.0 17.6 1.1 for a faster averaged velocity during the free pace run.
80% dlim 17.6 1.1 18.4 1.2 Similarly, Liedl et al. [21], Palmer et al. [27], Billat et al. [5] and
Significant differences for RPE or ETL between all distances, p £ .002 Kang et al. [18] reported no significant difference for RPE be-
tween constant and free pace runs. Only Edwards et al. [12]
found the opposite results, but this study was only carried out
significantly higher for the free pace run compared to the con- with 3 subjects. A possible explanation is that during free pace
stant pace run (94.5 ± 2.9 % vs. 90 %, p < .001, respectively), runs, short bouts of low- and high-intensity work alternated rap-
whereas there was no significant difference when the compari- idly during the race (velocity was calculated every 20 m, i.e.,
son was made without the last lap (89.6 ± 1.9 % vV̇O2max for the every 3 – 5 sec), whereas RPE values were collected only every
free pace run, p > .05). 2 min up to the end of the race. Moreover, according to Noakes
The results of the two-way repeated factor ANOVA with two re- et al. [23], pacing strategy would occur at a subconscious level
peated factors (distance and pace) showed that only distance in an apparently oscillatory manner, whereas the ratings of per-
had a significant effect on RPE and ETL ([F(3.27) = 151.6], and ceived exertion would change more regularly with activity.
[F(3.27) = 155.4], p < .001, respectively). The Tukey post hoc test In the same way, during endurance cycling, Liedl et al. [21] re-
showed significant differences for RPE or ETL between all distan- ported similar RPE values during constant effort at mean power
ces (p £ .002, l
" Table 2). equal to 78% V̇O2max compared to variable power alternating on-
In free-pace runs, velocity variability was 3.3 ± 0.9 % and ly ± 5% of mean power every 5 min during 1 h. It may be hypoth-
3.0 ± 0.7 % for the entire race and without the last lap, respec- esised that greater variations in exercise intensity would allow
tively. An example in a typical subject for V̇O2 and velocity dur- lower RPE to be observed during low-intensity phases. However,
ing constant and free pace runs is presented in l " Fig. 1. Low ve- Palmer et al. [27] and Kang et al. [18] did not find any RPE differ-

Fig. 1 Example in a typical subject for oxygen up-

take (V̇O2) and velocity during constant and free
pace runs. V̇O2 is indicated by empty circles and
black dots, velocity by empty and black squares, for
constant and free pace runs, respectively.

Garcin M et al. Perceptual Responses in … Int J Sports Med 2008; 29: 453 – 459
 Physiology & Biochemistry 457

ences either, despite larger variations in exercise intensity limitations, gained from previous similar exercises. The total ex-
(140 min of exercise at average exercise intensity equal to 58 ± ercise load and time the athlete’s body can tolerate the given
11 % peak power output compared to five repeated 20-min peri- metabolic level is known. Having this information and using
ods of variable intensity from 35 to 77% peak power output in- psychophysiological feedback, the athletes are able to arrange
terspersed with four 10-min periods of work at constant power their degree of exertion (i.e., exercise velocity) in order to avoid
output in the study by Palmer et al. [27], and 30 min exercise at premature fatigue prior to the completion of the event [39].
average exercise intensity equal to 67 ± 3% maximal heart rate This optimal adjustment of metabolic rate during heavy exercise
vs. a Spinning® workout at 60 to 80 % maximal heart rate in the using a feedback control system has recently been further devel-
study by Kang et al. [18]). In the latter studies, exercise intensity oped by St Clair Gibson and Noakes [34]. These authors sug-
varied every 2 min and every 4 to 8 min, respectively. Probably a gested that before and continuously during exercise, the brain
protocol comprising both greater variations in exercise intensity performs subconscious calculations of the metabolic cost re-
and duration would bring about RPE and ETL differences. quired to complete a given exercise task, and then computes
In this case, a change in a particular perceptual state could occur, how this will be influenced by the prevailing environmental con-
which we would become aware of, and only if the quanta of RPE ditions and the current physical state. This allows the selection
was different from the previous RPE level, would we feel that a of an optimum pacing strategy that will allow completion of
change of effort perception occurred [36]. Such a protocol has the task in the most efficient way while maintaining internal ho-
been proposed by Yaspelkis et al. [41] (30 min cycling at 45% moeostasis and a metabolic and physiological reserve capacity.
V̇O2max followed by 6 repeated 16-min periods of alternate cy- This is true for maximal intensity endurance exercises but also
cling at 75 and 45% – 8 min each). However, this protocol would for closed loop activity which is defined when either the dis-
be a simulation far from real conditions in competition. tance or time required to complete the activity is known before
The lack of influence of velocity variability on RPE and ETL was the exercise begins [34].
maybe linked with the experimental protocol. Indeed, the exper- For most subjects, velocity varied during the race and increased
imental design was inherently not counterbalanced in terms of " Fig. 1). According to the St Clair Gibson and
during the last lap (l
order of trials (the constant pace run at 90 % vV̇O2max was always Noakes’ theory, the initial pace chosen at the beginning of the
performed following the incremental test and before the free free race was subconsciously calculated on the basis of prior ex-
paced run in order to impose the distance to exhaustion set by perience. However, as the trial progressed, metabolic and other
the constant paced one). Therefore, the RPE values during the changes could have initiated subconscious integrative calcula-
free paced run may have been influenced by learning factors tions establishing that the initial pace would require not enough
from the previous trials. However, in order to limit this possible or too strong sensations of fatigue to maintain homoeostasis at
influence of familiarisation during the last test, the subjects had this exercise intensity. As a result of these subconscious calcula-
no knowledge of their RPE values collected during the constant tions, power output might be successively increased and de-
pace run. creased till the end of the race. Finally, as the last lap ap-
Finally, this lack of significant difference for RPE and ETL be- proached, these calculations would establish that the intensity
tween constant and free pace runs could be due to the low could be increased for the end of the race. The athletes com-
power, which depends, in part, on the relatively small sample pleted the free pace run by a final sprint during the last lap and
size in this study. decreased their running exhaustion times compared to the
As in the study by Kang et al. [18], the lack of discrepancy in aver- constant paced one. Only this final sprint was effective in in-
aged V̇O2 values between constant and free pace runs may be at- creasing performance. Consequently, compared to the constant
tributable to the fact that each higher velocity did not last long pacing run, the free pacing one performed over the same dis-
enough, was not of sufficient magnitude, and was always fol- tance to exhaustion allowed athletes to be at a significantly
lowed by a period of milder intensity. Consequently, these low higher % vV̇O2max.
velocity variations (less than 4%) did not increase the oxygen As presented above, Saint Clair Gibson et al. [36] suggested that
cost during free pace runs. The higher intensity phase (a sprint) before and continuously during exercise, the brain performs
probably did not last long enough to allow an increase in V̇O2. subconscious calculations (in part from knowledge of the end-
The same result was obtained when analysing data without the point and memory of pacing strategy from prior events) to allow
last lap. Robertson [30] and Shephard et al. [37] have shown that the athlete to set an appropriate pacing strategy to achieve opti-
RPE varies with the %V̇O2max. Consequently, as for V̇O2, RPE val- mal performance. Albertus et al. [1] have completed this theory
ues were not significantly different between constant and free adding that RPE at the onset of the exercise bout would be deter-
pace runs. As suggested by Liedl et al. [21], small variations in mined by the expected duration or distance of exercise. More-
power (3.3 % in our study and ± 5% in the Liedl et al. study) as re- over, the RPE values would increase according to this expected
gards to a relatively high mean power output are physiologically duration or distance and/or by the duration or distance of exer-
sustainable and would result in nonsignificant increases in V̇O2 cise that remains [24]. Therefore, anticipating the maximal RPE
or RPE mean values. Moreover, such small variations in exercise that the individual will tolerate, the brain centre responsible for
velocity would be effective in our study in enhancing time trial the generation of the RPE then increases that RPE according to
competitive performance. the percentage of the total exercise duration or distance that
Several authors have hypothesised that a learned subconscious has been completed (or the percentage of duration or distance
anticipatory/regulation system exists, known as “teleoanticipa- that remains) [24].
tion” originating from the central nervous system [39]. This sub- Indeed, other authors like Rejeski and Ribisl [29], and Baden et
conscious feedback mechanism serves to decrease efferent out- al. [3] have also shown that RPE values may be influenced by
put from the motor cortex. Prior to the beginning of a given exer- the expected running duration or distance. Contrary to our hy-
cise, it is hypothesised that the athlete’s central nervous system pothesis, as the result of the present study showed that RPE and
is aware of the athlete’s fitness level, endurance capacity and ETL were not influenced by the velocity variability, the expected

Garcin M et al. Perceptual Responses in … Int J Sports Med 2008; 29: 453 – 459
458 Physiology & Biochemistry

running distance could be an important determinant of RPE. 13 Foster C, Snyder AC, Thompson NN, Green MA, Folley M, Schrager M. Ef-
fect of pacing strategy on cycle time trial performance. Med Sci Sports
Therefore, as in the study by Albertus et al. [1], our results sup-
Exerc 1993; 25: 383 – 388
port a dissociation between RPE and work rate because changes 14 Garcin M, Billat V. Perceived exertion scales attest to both intensity
in velocity were not followed by changes in RPE and ETL values. and exercise duration. Percept Mot Skills 2001; 93: 661 – 671
This last comment is, however, in conflict with Saint Clair Gibson 15 Garcin M, Vandewalle H, Monod H. A new rating scale of perceived ex-
et al. [36], who suggested that both the control of exercise inten- ertion based on subjective estimation of exhaustion time. Int J Sports
Med 1999; 20: 40 – 43
sity and the perceived exertion during an event would be con-
16 Garcin M, Wolff M, Bejma T. Reliability of rating scales of perceived ex-
trolled by the same regulatory processes in the brain. Future re- ertion and heart rate during progressive and maximal constant load
search is needed to explore how both changes in exercise inten- exercises till exhaustion in physical education students. Int J Sports
sity and perceived exertion are processed. Med 2003; 24: 285 – 290
17 Gearhart RF Jr, Becque MD, Hutchins MD, Palm CM. Comparison of
memory and combined exercise and memory anchoring procedures
on ratings of perceived exertion during short duration, near-peak-in-
Conclusion tensity cycle ergometer exercise. Percept Mot Skills 2004; 99: 775 –
! 784
The results of the present study demonstrated that subjects’ RPE 18 Kang J, Chaloupka EC, Mastrangelo MA, Hoffman JR, Ratamess NA,
O’Connor E. Metabolic and perceptual responses during Spinning cycle
and ETL values did not reflect the alterations in exercise veloc-
exercise. Med Sci Sports Exerc 2005; 37: 853 – 859
ities. Moreover, it would seem that these variations did not have 19 Kuipers H, Keiser HA. Overtraining in elite athletes – review and direc-
an effect on perceptual responses during exercise as long as the tions for the future. Sports Med 1998; 6: 79 – 92
average V̇O2 was kept the same. Probably a protocol comprising 20 Leger LA, Ferguson RJ. Effect of pacing on oxygen uptake and peak lac-
both greater variations in exercise intensity and exercise dura- tate for a mile run. Eur J Appl Physiol 1974; 32: 251 – 257
21 Liedl MA, Swain DP, Branch JD. Physiological effects of constant versus
tion would bring about RPE and ETL differences.
variable power during endurance cycling. Med Sci Sports Exerc 1999;
31: 1472 – 1477
22 Mc Laughlin JE, King GA, Howley ET, Bassett DR, Ainsworth BE. Valida-
Acknowledgements tion of the Cosmed K4 b2 portable metabolic system. Int J Sports Med
! 2001; 22: 280 – 284
23 Noakes TD, St Clair Gibson A, Lambert EV. From catastrophe to com-
This study was supported by grants from the Caisse Centrale des
plexity: a novel model of integrative central neural regulation of effort
Activités Sociales d’Electricité et Gaz de France. The authors and fatigue during exercise in humans: summary and conclusions. Br J
gratefully acknowledge Mr. John Hall for his expert advice in Sports Med 2005; 39: 120 – 124
the revision of the manuscript. 24 Noakes TD, Snow RJ, Febbraio MA. Linear relationship between the per-
ception of effort and the duration of constant load exercise that re-
mains. J Appl Physiol 2004; 96: 1571 – 1573
References 25 Noble BJ, Metz KF, Pandolf KB, Bell CW, Cafarelli E, Sime WE. Perceived
1 Albertus Y, Tucker R, St Clair Gibson A, Lambert E, Hampson D, Noakes T. exertion during walking and running-II. Med Sci Sports Exerc 1973;
Effect of distance feedback on pacing strategy and perceived exertion 5: 116 – 120
during cycling. Med Sci Sports Exerc 2005; 37: 461 – 468 26 Noble BJ, Robertson RJ. Physiological and psychological mediators. In:
2 Ariyoshi M, Yamaji K, Shephard RJ. Influence of running pace upon per- Borg G (ed). Perceived Exertion. Champaign: Human Kinetics, 1996:
formance effects upon treadmill endurance time and oxygen cost. Eur 105 – 197
J Appl Physiol 1979; 41: 83 – 91 27 Palmer GS, Borghouts LB, Noakes TD, Hawley AJ. Metabolic and per-
3 Baden DA, McLean TL, Tucker R, Noakes TD, St Clair Gibson A. Effect of formance responses to constant-load vs. variable-intensity exercise
anticipation during unknown or unexpected exercise duration on rat- in trained cyclists. J Appl Physiol 1999; 87: 1186 – 1196
ing of perceived exertion, affect, and physiological function. Br J Sports 28 Palmer GS, Noakes TD, Hawley AJ. Effects of steady-state versus sto-
Med 2005; 39: 742 – 746 chastic exercise on subsequent cycling performance. Med Sci Sports
4 Billat V, Koralsztein JP. Significance of the velocity at V̇O2max and time Exerc 1997; 29: 684 – 687
to exhaustion at this velocity. Sports Med 1996; 22: 90 – 108 29 Rejeski W, Ribisl P. Expected task duration and perceived effort: an at-
5 Billat V, Slawinski J, Danel M, Koralsztein JP. Effect of free versus con- tributional analysis. J Sport Psychol 1980; 2: 227 – 236
stant pace on performance and oxygen kinetics in running. Med Sci 30 Robertson RJ. Central signals of perceived exertion during dynamic ex-
Sports Exerc 2001; 33: 2082 – 2088 ercise. Med Sci Sports Exerc 1982; 14: 390 – 396
6 Billat V, Wesfreid E, Kapfer C, Koralsztein JP, Meyer Y. Nonlinear dynam- 31 Robertson RJ. Development of the perceived exertion knowledge base:
ics of heart rate and oxygen uptake in exhaustive 10 000 runs: influ- an interdisciplinary process. Int J Sport Psychol 2001; 32: 189 – 196
ence of constant vs. freely paced. J Physiol Sci 2006; 56: 1 – 9 32 Robinson DM, Robinson SM, Hume PA, Hopkins WG. Training intensity
7 Borg GV. Perceived exertion as an indicator of somatic stress. Scand J of elite male distance runners. Med Sci Sports Exerc 1991; 23: 1078 –
Rehabil Med 1970; 2: 92 – 98 1082
8 Borg GV. Borg’s Perceived Exertion and Pain Scales. Champaign: Hu- 33 Robinson S, Robinson DL, Mountjoy RJ, Bullard RW. Influence of fatigue
man Kinetics, 1998: 106 on the efficiency of men during exhausting run. J Appl Physiol 1958;
9 Cottin F, Papelier Y, Durbin F, Koralsztein JP, Billat V. Effect of fatigue on 12: 197 – 201
spontaneous velocity variations in human middle-distance running: 34 St Clair Gibson A, Noakes TD. Evidence for complex system integration
use of short-term Fourier transformation. Eur J Appl Physiol 2002; and dynamic neural regulation of skeletal muscle recruitment during
87: 17 – 27 exercise in humans. Br J Sports Med 2004; 38: 797 – 806
10 De Koning JJ, Bobbert MF, Foster C. Determination of optimal pacing 35 St Clair Gibson A, Lambert ML, Noakes TD. Neural control of force out-
strategy in track cycling with an energy flow model. J Sci Med Sport put during maximal and submaximal exercise. Sports Med 2001; 31:
1999; 2: 266 – 277 637 – 650
11 Demarle AP, Slawinski JJ, Laffite LP, Bocquet VG, Koralsztein JP, Billat V. 36 St Clair Gibson A, Lambert EV, Rauch LHG, Tucker R, Baden DA, Foster C,
Decrease of O2 deficit is potential factor in increased time to exhaus- Noakes TD. The role of information processing between the brain and
tion after specific endurance training. J Appl Physiol 2001; 90: 947 – peripheral physiological systems in pacing and perception of effort.
953 Sports Med 2006; 36: 706 – 722
12 Edwards RH, Melcher A, Hesser CM, Wigertz O, Ekelund LG. Physiological 37 Shephard R, Vandewalle H, Gil V, Bouhlel E, Monod H. Respiratory, mus-
correlates of perceived exertion in continuous and intermittent exer- cular and overall perceptions of effort: the influence of hypoxia and
cise with the same average power output. Eur J Clin Invest 1972; 2: muscle mass. Med Sci Sports Exerc 1992; 24: 556 – 567
108 – 114

Garcin M et al. Perceptual Responses in … Int J Sports Med 2008; 29: 453 – 459
 Physiology & Biochemistry 459

38 Staab JS, Agnew JW, Siconolfi SF. Metabolic and performance responses 40 Wais M. Comparaison des relations „distance/temps“ établies dans
to uphill and downhill running in distance runners. Med Sci Sports des courses à allure libre ou imposée. Sciences & Motricité 2003; 48:
Exerc 1992; 24: 124 – 127 99 – 117
39 Ulmer HV. Concept of an extracellular regulation of muscular metabol- 41 Yaspelkis BB, Patterson JG, Anderla PA, Ding Z, Ivy, JL. Carbohydrate sup-
ic rate during heavy exercise in humans by psychophysiolgical feed- plementation spares muscle glycogen during variable-intensity exer-
back. Experientia 1996; 52: 416 – 420 cise. J Appl Physiol 1993; 75: 1477 – 1485

Garcin M et al. Perceptual Responses in … Int J Sports Med 2008; 29: 453 – 459