You are on page 1of 8

Available online at www.sciencedirect.


Journal of Exercise Science & Fitness 13 (2015) 123e130

Original article

A hard/heavy intensity is too much: The physiological, affective,

and motivational effects (immediately and 6 months post-training)
of unsupervised perceptually regulated training
Gaynor Parfitt*, Tim Olds, Roger Eston
Alliance for Research in Exercise, Nutrition, and Activity, Sansom Institute for Health Research, University of South Australia, Adelaide, Australia
Received 29 April 2015; revised 27 August 2015; accepted 5 October 2015
Available online 24 November 2015


Background/Objective: There are several practical and theoretical advantages to perceptually regulated training, including its simplicity of use
and potential to influence exercise motivation. The study objective was to examine if perceptually regulated training at ratings of perceived
exertion (RPE) of 13 and 15 resulted in significant increases in aerobic fitness, reductions in metabolic risk factors, and changes in motivational
constructs following an 8-week training program and at follow up 6 months after.
Methods: Following stratified randomization based on estimated aerobic capacity and sex, sedentary volunteers (n ¼ 63; men ¼ 21) were
allocated to one of three groups: RPE 13, RPE 15, and control. The participants completed baseline, post-training, and 6-months post-training
assessments for aerobic fitness, metabolic risk factors, and motivational constructs. The participants' acute exercise training responses (affect,
competence, enjoyment, and work rate) were also recorded.
Results: The data support the fitness and motivational construct hypotheses but not the metabolic risk factor hypothesis. Aerobic fitness increased
from the baseline to post-training in both RPE groups, with the increase maintained 6 months post-training only in the RPE 13 group. Exercise
autonomy increased over the training program, with exercise competence and affect more positive in the RPE 13 group than in the RPE 15.
However, the training programs did not reduce the metabolic risk factors, and attrition levels were high. Unsupervised training at RPE 13 and
RPE 15 did improve fitness, but as hypothesized, this was not maintained in the RPE 15 group 6 months post-training.
Conclusion: The motivational processes associated with RPE 13 regulated exercise (greater competence and autonomy and more positive affect)
potentially explain the maintained fitness in this group.
Copyright © 2015, The Society of Chinese Scholars on Exercise Physiology and Fitness. Published by Elsevier (Singapore) Pte Ltd. This is an
open access article under the CC BY-NC-ND license (

Keywords: aerobic fitness; affect; competence; exercise intensity production; ratings of perceived exertion; self-determination

Introduction the RPE is used to regulate intensity in an exercise training

context, then as the individual becomes fitter, he or she will
Borg's ratings of perceived exertion (RPE) scale1 can be automatically increase the work rate to achieve the RPE
used to regulate exercise intensity.2e4 This method of exer- level.4,5 This mitigates the need to assess and reassess
cise intensity regulation may offer a number of practical and physiological variables, such as heart rate, lactate levels, or
theoretical advantages for exercise training. Pragmatically, if maximal oxygen uptake, to confirm if the RPE-regulated
exercise intensity has produced a training effect. This
makes the method user-friendly, inexpensive, and simple to
* Corresponding author. University of South Australia, GPO Box 2471, administer and use. In programs where exercise intensity is
Adelaide, South Australia 5001, Australia. prescribed based upon a heart rate bandwidth or a percentage
E-mail address: (G. Parfitt).
1728-869X/Copyright © 2015, The Society of Chinese Scholars on Exercise Physiology and Fitness. Published by Elsevier (Singapore) Pte Ltd. This is an open access article under the
CC BY-NC-ND license (
124 G. Parfitt et al. / Journal of Exercise Science & Fitness 13 (2015) 123e130

of maximal oxygen uptake, these physiological variables outcomes and is hence likely to improve adherence, there is
need to be reassessed to ensure that the prescribed training very little research regarding the longer-term efficacy of a
intensity is modified to meet the physiological adaptation. perceptually regulated training program to increase and
From a theoretical perspective, perceptually regulating maintain aerobic capacity and produce other health-related
exercise intensity can improve the motivation to exercise. exercise benefits. There may be a delicate balancing act be-
According to the self-determination theory,6 autonomy (i.e., tween exercise at an intensity which is pleasant (and hence
control) and competence are two of the three psychological encourages adherence) and exercise at an intensity which
needs required to support the development of intrinsic moti- optimizes health-related benefits. Indeed, as exercise intensity
vation. RPE-regulated exercise affords the individual control increases above the ventilatory threshold, affective responses
of the intensity for a specific RPE level. The individual decline.11
chooses the intensity and is autonomous in changing it to meet Parfitt et al4 support the effect of RPE-13 supervised
his or her individual perception throughout the exercise ses- training on fitness, health variables, and affect (fitness
sion. When exercise intensity is prescribed, this autonomy is increased significantly, mean arterial pressure decreased, and
not present. Further, this control provides a sense of achieve- affect remained positive) immediately following an 8-week
ment and maintains competence throughout the session: the training program. However, that study did not track out-
individual is actively involved and able to increase or decrease comes beyond the 8-week program. We do not know if those
the intensity as necessary. When exercise intensity is pre- effects were maintained at 6 months. The 8-week training
scribed, it may be perceived to be appropriate or otherwise too program was also supervised on a one-to-one basis, and
easy or too hard; neither of the latter perceptions would sup- therefore, the translation and generalizability of RPE 13-
port the development of competence and motivation. In regulated training to more natural gym environments (where
addition, affective responses (feelings of pleasure/displeasure) one-to-one supervision is not available) is unknown. Further-
during exercise are intensity dependent and predict exercise more, we do not know if perceptually regulated training at
behavior in the succeeding 6 months and 12 months.7,8 Wil- RPE 15 (hard/heavy) will improve health-related benefits.
liams' model,9 which integrates self-determination theory and There is evidence that exercise at varying intensities have
hedonic theory (that behavioral decisions are made based upon different effects on health-related outcomes. DiPietro et al20
the pleasure experienced),10 makes a case for self-paced ex- found that vigorous-intensity exercise is more beneficial in
ercise and explains the affective and motivational conse- improving insulin sensitivity, and Moholdt et al21 observed
quences of perceptually regulated exercise. When prescribed that peak oxygen consumption [as measured by percentage
exercise rather than allowed to self-regulate, people have more heart rate maximum (%HRmax)] increased more markedly
variable and less positive affective responses,11 even when the after higher-intensity training in patients with coronary artery
prescribed exercise intensity matches the self-regulated in- disease. However, Midgley et al22 reported that long-interval
tensity.12 Autonomy and affective response are positively training at 100% maximal capacity is extremely efficacious
associated,13 and achievement, ability to cope, and perceptions in improving fitness, albeit, barely sustainable by the
of control explain differences in affect in low-activity adults.14 participants.
Further, in a graded exercise test to volitional exhaustion, if Theoretically, perceptually regulated training clamped at
the exerciser is in control of changes in exercise intensity, either 13 or 15 on the RPE scale (the former indicating
affect during the test is more positive than when he or she is moderate exercise intensity23 and the latter vigorous exercise
not in control.15 intensity24,25) should significantly increase aerobic fitness and
Studies that have used RPE to perceptually regulate exer- reduce cardiac risk factors. Furthermore, as perceptually
cise intensity in a training program include a 4-week study16 regulated training permits the participant to dictate the pace,
with a 2-year follow up17 on residential cardiac patients; intensity, and increments in intensity of the exercise and
studies of breast cancer patients18 and survivors19; and an 8- thereby achieve a sense of autonomy, motivation to exercise
week study of previously sedentary women.4 These studies should become more intrinsic and influence long-term adher-
have specified RPE levels of between 11 and 15 and reported ence to exercise behavior.9 Additionally, self-regulation should
improvements in physical performance, with some evidence to support the maintenance of positive affect, although affect
support the motivational influence of this method of training. should be lower at higher exercise intensities.11 Consequently,
In particular, Parfitt et al4 demonstrated that perceptually affect should be lower during RPE 15 than RPE13 exercise.
regulated exercise corresponding to RPE 13 is perceived to be This study therefore aimed to extend previous research and
“pleasant” during training. Over the 8-week training program compare the effects of an 8-week unsupervised, perceptually
in Parfitt et al's study, the affective responses recorded during regulated treadmill training program at RPE 13 and RPE 15 on
training indicated that the participants felt “good” while fitness and other health measures immediately following and 6
exercising at a physiologically confirmed moderate to months after completion of the program. A secondary purpose
vigorous exercise intensity. of this study was to assess the affective state (participants'
Although evidence and a theoretical explanation exist for pleasure/displeasure) during training and the motivational ef-
why perceptually regulated exercise has beneficial affective fect of training at RPE 13 and RPE 15. It was hypothesized
G. Parfitt et al. / Journal of Exercise Science & Fitness 13 (2015) 123e130 125

that perceptually regulated training at RPE 15 would lead to (competence ¼ 0.91; interesteenjoyment ¼ 0.84; BREQ-
greater immediate improvements in aerobic fitness and car- 2 ¼ 0.80e0.89).
diovascular health, but that at 6 months, the effects would
persist in the RPE 13 perceptually regulated training group Procedures
and dissipate in the RPE 15 group. Further, it was hypothe-
sized that affective and motivational processes would be more This study required the participants to visit the exercise
positive during and following the RPE 13 perceptually regu- physiology laboratory (temperature, ~24 C; relative humidity,
lated training. ~40%) on three occasions: baseline assessment, post-training,
and 6 months post-training. On the first visit, basic anthro-
Methods pometric data, including height (wall-mounted stadiometer,
Surgical & Medical Products, Adelaide, Australia); body mass
Participants (Tanita BC-418, Tanita Corp., Tokyo, Japan); resting blood
pressure (Dinamap Pro 100 automated sphygmomanometer,
Sixty-three (men ¼ 21) sedentary volunteers (40.8 ± 9.1 GE Medical Systems Information Technologies, Inc., Mil-
years; 83.5 ± 16.6 kg; 169.7 ± 8.9 cm) were recruited from the waukee, WI, USA); and a finger-prick blood sample to test for
local community. The participants were classified as sedentary total cholesterol, high-density lipoprotein (HDL), and blood
if they did not meet the current physical activity guidelines (at glucose levels (CardioChek PA point-of-care device, PTS
least 30 minutes of moderate-intensity activity, 5 or more days Diagnostics, Indianapolis, IN, USA) following overnight
a week, in bouts of at least 10 minutes). The volunteers read fasting were completed. The participants also completed a
and signed an informed consent form approved by the Uni- submaximal graded exercise test on a motorized treadmill (JT-
versity Ethics Committee and completed the Adult Pre- 4000 treadmill, JNB Sports, Kyungkido, South Korea).
exercise Screening System26 prior to participation. The exercise test was based on the BalkeeWare treadmill
graded exercise test protocol. The participants were required
Measures to walk on the treadmill at an initial speed of 4 km/h on a
gradient of 0% for 3 minutes. The speed was then increased to
RPE and affect 5.4 km/h and the gradient to 2%. Every minute thereafter, the
Borg's 6e20 RPE scale1 was used throughout the study. RPE and feeling rating were recorded, and the gradient was
The participants were encouraged to focus on their overall increased by 1% at the same speed of 5.4 km/h until
perception of “whole body” exertion when reporting their the participants reached 80%HRmax (where HRmax ¼
RPE. Two levels of RPE were selected for the training in- 208  0.7  age)33 or 17 on the RPE scale. A submaximal
tensities: RPE 13 and RPE 15. These levels were selected protocol allowing the estimation of maximal oxygen con-
based on previous research,23 as they should elicit intensities sumption (VO2max) was selected to avoid undue physiolog-
that would be classified as moderate and vigorous and lead to ical strain and negative affect.15
fitness gains. During the submaximal test, online respiratory gas was
The participants were also asked to report on their affective measured using a breath-by-breath automatic gas analysis
state at the end of each minute using Hardy and Rejeski's system (Metamax 3B, Cortex Biophysik, Leipzig, Germany).
feeling scale,27 an 11-point, single-item, bipolar scale. The Changes in oxygen uptake were recorded continuously
feeling scale ranges from þ5 to e5, with anchors provided at throughout the tests (MetaSoft 3.1 software, Cortex Bio-
zero and all odd integers (þ5 ¼ very good; þ3 ¼ good; physik). Prior to every test, the system was calibrated against
þ1 ¼ fairly good; 0 ¼ neutral; 1 ¼ fairly bad; 3 ¼ bad; known concentrations of cylinder gases (15% oxygen, 5%
5 ¼ very bad). This scale correlates with other similar carbon dioxide) and a 3 L calibration syringe (for flow vol-
valence scales.28,29 ume) in accordance with the manufacturer's guidelines. Heart
rate was continuously monitored using a wireless chest strap
Motivation telemetry system (T31, Polar Electro, Kempele, Finland). All
The participants completed the perceived competence and physiological outputs were concealed from the participant. We
interesteenjoyment items from the Intrinsic Motivation In- extrapolated from the measured heart rate and oxygen uptake
ventory (IMI)30,31 and the Behavioral Regulation in Exercise in the submaximal protocol to the age-predicted HRmax to
Questionnaire 2 (BREQ-2)32 to assess motivational processes estimate the VO2max.
during training and following the intervention. The IMI was The RPE and feeling scales were mounted on the wall
completed after a training session in Week 1 to assess the directly in front of the participants, and each participant
acute motivational response to training. The BREQ-2 was received standardized instructions on how they should identify
completed at baseline, following training, and 6 months post- and report their overall level of exertion1 and affective
training, and the relative autonomy index was computed to valence.27 Following the test, the participants were block-
directly assess the participants' levels of perceived autonomy randomized and stratified by sex and VO2max to one of
to exercise. Both questionnaires are scored on a Likert scale three conditions: control (n ¼ 21; men ¼ 7), RPE 13-regulated
and exhibit good reliability and validity.30,32 Further, scale training (n ¼ 21; men ¼ 7), or RPE 15-regulated training
reliability was confirmed in the current populations (n ¼ 21; men ¼ 7).
126 G. Parfitt et al. / Journal of Exercise Science & Fitness 13 (2015) 123e130

Exercise training intervention Results

Following the stratified randomization based on estimated
aerobic capacity and sex, the participants were allocated to the Descriptive data across the three groups following the
control, RPE 13, and RPE 15 groups. The control group par- stratified randomization are displayed in Table 1. The groups
ticipants were instructed to continue their normal physical were statistically different in terms of total cholesterol at
activity habits (before the study began) for 8 weeks. The baseline.
training groups were invited to attend a gym for up to three
occasions per week for 30 minutes each time for 8 weeks. In Intention-to-treat analysis
the first session, the participants were shown how to adjust the
speed and gradient on the treadmill and were reminded that The repeated measures intention-to-treat analysis for the
they should make any adjustments necessary to ensure that estimated VO2max resulted in a time (F(2,120) ¼ 5.39,
they perceived the intensity to be at RPE 13 (somewhat hard) p < 0.01, h2p ¼ 0.08) main effect and a time-by-group
or RPE 15 (hard/heavy) on the Borg 6e20 scale, depending on interaction that approached significance (F(4,120) ¼ 2.31,
group randomization. In subsequent sessions, staff in the gym p ¼ 0.06, h2p ¼ 0.07). Figure 1 illustrates the changes in the
reminded the participants to perceptually regulate the intensity estimated VO2max from baseline to 6 months following the
(RPE 13 or RPE 15) and were available in the gym to answer intervention across the three groups.
any questions. During an exercise session in Week 1 and Week A time main effect was recorded for mean arterial pressure
4, acute data (affect, heart rate, treadmill speed, and gradient) (F(1.6,96.3) ¼ 18.9, p < 0.01, h2p ¼ 0.24), due to reductions in
were recorded every 5 minutes, and perceived competence and pressure from baseline (94 ± 11) to 8 weeks and 6 months
enjoyment (assessed with the IMI) were measured after the post-intervention (89 ± 10 and 89 ± 9, respectively). Further,
session. the statistical difference at baseline for cholesterol remained
All gym sessions were completed using motorized tread- throughout the study with a group main effect (F(1,60) ¼ 4.43,
mills (97T Integrity treadmill, Life Fitness, Rosemont, IL, p < 0.05, h2p ¼ 0.13) and with cholesterol lower in the RPE
USA). The display screen of the treadmill (time, distance, 13 group (4.9 ± 1.2) than in the control (5.8 ± 0.8) and RPE 15
speed, and gradient) was masked from the participant with a (5.4 ± 1.2) groups. No other intention-to-treat analyses were
laminated copy of the RPE scale. The participants had no prior significant.
experience with perceptual scaling prior to participation. The
sessions were not individually supervised. Each participant Per-protocol analysis
received standardized written and verbal instructions on how
to report the “overall” feelings of exertion.1 All physiological Over the duration of the study, attrition levels were high,
outputs were concealed from the participant. with complete data gathered from only 34 participants
(control ¼ 14, RPE 13 ¼ 11, RPE 15 ¼ 9). Twelve participants
Data analysis in the RPE 15 group did not complete the study; 11 of the
original 21 failed to complete the 8-week training program.
From the recorded treadmill speed and gradients chosen by Seventeen participants in the RPE 13 group completed
the participants, indirect measures of VO2 were calculated training, and 11 attended the 6-months post-training assess-
using the American College for Sports Medicine's metabolic ment. The participants who completed the training program
equation.34 The statistical software package SPSS 21.0 for attended the gym for training between 10 and 24 times (RPE
Windows (SPSS Inc., Chicago, IL, USA) was used to analyze 13 ¼ 19.4 ± 5.7 sessions, RPE 15 ¼ 17.5 ± 3.7 sessions).
the data. For all analyses, the alpha was set at p < 0.05. The per-protocol analyses resulted in a significant interac-
The data were initially analyzed using an intention-to-treat tion for the estimated VO2max (F(3.4,52.4) ¼ 3.19, p < 0.05,
analysis to examine the efficacy of the intervention. In cases of h2p ¼ 0.17). The RPE 13 group improved from baseline and
missing data, the “last observation carried forward” method
was adopted. Per-protocol analyses were conducted for par- Table 1
ticipants who had completed the study as intended. To Descriptive data.
examine the acute affective (feeling scale), motivational Controla RPE 13a RPE 15a
(competence, interesteenjoyment), and physiological (work
Height (cm) 169.5 (9.6) 169.5 (7.5) 170.1 (8.6)
rate and %HRmax) responses recorded during two of the Mass (kg) 80.6 (17.2) 83.2 (25.6) 86 (15)
training sessions, an additional RPE group (RPE 13, RPE 15) BMI (kg/m2) 28 (5.7) 28.7 (7.9) 29.5 (3.3)
by time (5 minutes, 10 minutes, 15 minutes, 20 minutes, 25 VO2max (mL/kg/min) 30 (5.9) 31.5 (7.5) 31 (5.7)
minutes, 30 minutes) mixed model analysis of variance Cholesterol (mM/L)* 5.6 (0.6) 4.8 (1.2) 5.3 (1.2)
(ANOVA) or t tests were conducted. Where sphericity was HDL 1.6 (0.5) 1.3 (0.5) 1.4 (0.5)
HDL/cholesterol ratio (mM/L) 0.3 (0.1) 0.29 (0.12) 0.26 (0.1)
violated, the GreenhouseeGeisser epsilon was used. Effect Blood glucose (mM/L) 4.7 (0.5) 4.9 (0.7) 5.1 (0.4)
sizes associated with F statistics (ANOVAs) were expressed as
*p < 0.05 at baseline.
partial eta squared (h2p) defined as small (0.01), medium BMI ¼ body mass index; HDL ¼ high-density lipoprotein; RPE ¼ rating of
(0.06), and large (0.14).35 perceived exertion; VO2max ¼ maximal aerobic capacity.
All data are presented as mean (standard deviation).
G. Parfitt et al. / Journal of Exercise Science & Fitness 13 (2015) 123e130 127

37 p < 0.05, h2p ¼ 0.11) and group (F(1,30) ¼ 4.45, p < 0.05,
Control (ITT) h2p ¼ 0.13) main effects. Affect declined over the 30-minute
training session (1.9 ± 1.5 to 1.2 ± 1.8) and was more positive
VO2max (mL/kg/min)

RPE 13 (ITT)
33 RPE 15 (ITT) in the RPE 13 group (2.1 ± 1.6) than in the RPE 15 group
Control (PP) (1 ± 1.6). There were no significant differences ( p < 0.05)
RPE 13 (PP)
between the actual work rate (expressed as %VO2max
29 RPE 15 (PP)
computed from the speed and gradient of the treadmill)34 and
27 the percentage of the maximum heart rate during training
(RPE 13 ¼ 74% ± 11% VO2max, 79% ± 9.5% HRmax; RPE
Baseline Post-training 6 mo post- 15 ¼ 78% ± 15% VO2max, 83% ± 10% HRmax).
training Analysis of the IMI subscales following the training session
Figure 1. Maximal aerobic capacity means for the intention-to-treat and per- indicated that competence was significantly ( p < 0.05) higher
protocol condition-by-time analyses. RPE ¼ rating of perceived exertion; in the RPE 13 group (40.6 ± 6.9) than in the RPE 15 group
VO2max ¼ maximal aerobic capacity. (34.4 ± 10.1). The scores for interesteenjoyment were not
maintained the improvement, while the RPE 15 group initially significantly different (RPE 13 ¼ 40.1 ± 6.5, RPE
improved from baseline to 8 weeks but then returned to 15 ¼ 34.8 ± 12.2).
baseline levels (Figure 1). This interaction reflected the nature
of the approaching interaction in the intention-to-treat ana- Discussion
lyses. A significant time main effect for mean arterial pressure
(F(2,62) ¼ 18.5, p < 0.01, h2p ¼ 0.37) and group main effect The objective of the study was to examine if perceptually
for cholesterol (F(1,27) ¼ 6.86, p < 0.01, h2p ¼ 0.34) reflected regulated training at RPE 13 and RPE 15 significantly
the same pattern of results as in the intention-to-treat analyses. enhanced aerobic fitness, reduced metabolic risk factors, and
Analysis of the BREQ-2 indicated a time main effect increased exercise autonomy following an 8-week training
(F(2,62) ¼ 5.4, p < 0.01, h2p ¼ 0.15) for the relative autonomy program and at 6-months follow up. The data partially support
index, with scores increasing from baseline (43.3 ± 24.1) to 8 the hypotheses, with significant improvement in aerobic fitness
weeks (49.3 ± 19.7) and no further change 6 months post- maintained in the RPE 13 training group but a return to
training (50.8 ± 22.8). The condition-by-time interaction baseline post-training in the RPE 15 group. Exercise auton-
was not significant, although the time main effect appears to omy also increased over the training program, with acute
be largely due to the increase in the relative autonomy index in motivational responses (exercise competence and affect) more
the RPE 13 and RPE 15 participants (Table 2). positive in the RPE 13 group than in the RPE 15. However, the
RPE 15 training did not lead to significantly greater immediate
Acute responses during training session improvements in aerobic fitness or reduce metabolic risk
factors. There was a reduction in mean arterial pressure, but
A repeated measures ANOVA on the affective responses this was common in all three groups, with no differences
during training resulted in significant time (F(1.7,52.3) ¼ 3.8, between them.

Table 2
Means (standard deviations) for the group-by-time analyses of the per-protocol analysis.
Control RPE 13 RPE 15
Baselinea Post-training 6 mo Baselinea Post-training 6 mo Baselinea Post-training 6 mo
post-training post-training post-training
VO2max (mL/kg/min)d,* 28.8 (4.9) 29.2 (4.4) 29.2 (4.5) 33.0 (7.7) 35.4 (7.2) 35.7 (6.9) 29.7 (5.2) 32.9 (5.8) 29.3 (4.3)
BMI (kg/m) 28.0 (5.2) 27.8 (5.0) 27.3 (4.6) 25.7 (5.8) 25.8 (5.8) 25.7 (5.8) 28.3 (3.5) 28.2 (3.5) 28.3 (3.6)
Cholesterol (mM/L)b,** 5.8 (0.7) 6.3 (0.7) 6.0 (0.8) 4.7 (1.2) 4.7 (1.1) 4.8 (1.1) 5.3 (1.2) 5.3 (1.0) 5.7 (1.1)
HDL (mM/L) 1.5 (0.4) 1.5 (0.5) 1.5 (0.3) 1.3 (0.4) 1.3 (0.4) 1.3 (0.5) 1.3 (0.5) 1.2 (0.4) 1.3 (0.6)
HDL/cholesterol ratio 0.26 (0.09) 0.24 (0.07) 0.25 (0.06) 0.29 (0.10) 0.28 (0.10) 0.28 (0.10) 0.27 (0.10) 0.23 (0.08) 0.25 (0.12)
Blood glucose (mM/L) 4.8 (0.5) 4.5 (0.6) 4.6 (0.5) 4.7 (0.5) 4.6 (0.5) 4.7 (0.4) 5.0 (0.5) 4.8 (0.3) 4.8 (0.6)
Mean arterial pressure 95.7 (8.5) 89.1 (7.8) 86.7 (6.9) 96.1 (12.0) 86.7 (11.6) 88.1 (12.2) 93.4 (10.8) 86.4 (10.7) 91.6 (6.9)
Relative autonomy indexc,* 38.6 (15.6) 41.6 (15.0) 41.9 (19.9) 49.0 (24.4) 56.7 (18.2) 62.4 (19.4) 43.8 (24.4) 52.3 (25.0) 50.7 (26.4)
*p < 0.05.
**p < 0.01.
BMI ¼ body mass index; HDL ¼ high-density lipoprotein; RPE ¼ rating of perceived exertion; VO2max ¼ maximal aerobic capacity.
These baseline means differ from those presented in Table 1 as they are computed from only the participants who completed all three data points.
Group main effect.
Time main effect.
Group-by-time interaction.
128 G. Parfitt et al. / Journal of Exercise Science & Fitness 13 (2015) 123e130

The increases in aerobic fitness in the RPE 13 and RPE 15 there was no dropout. As previously described, Parfitt et al
groups following the 8 weeks of training were approximately used a 1:1 closely supervised training protocol and only one
7% and 11%, respectively. A greater increase in the RPE 15 training intensity (RPE 13), both of which likely influenced
group had been hypothesized, as the exercise intensity elicited the dropout statistic. The higher dropout and lower adherence
at RPE 15 should be greater than that at RPE 13 and would to the exercise program, particularly in the RPE 15 group, may
therefore provide a more potent training stimulus to improve be attributable to the lower affect and lower perceived
aerobic fitness. According to the average speed and gradient of competence. Training at RPE 15 was perceived to be signifi-
the treadmill, the relative exercise intensities produced for the cantly less pleasant than training at RPE 13, and the partici-
RPE 13 and RPE 15 groups were 74% and 79% VO2max, pants in the RPE 15 group also felt significantly less
respectively. These work rates match previous observations2 competent during trainingdboth of which influence intrinsic
and most likely explain the slightly greater (although statisti- motivation.6
cally nonsignificant) improvement in aerobic fitness in the The psychological data provide some support for the
RPE 15 group. The improvement in aerobic fitness in the RPE rationale that RPE-regulated training can influence motiva-
13 group (7%) is smaller than that observed by Parfitt et al4 tional processes. The significant increase in the relative au-
(17%)dwho applied a similar RPE 13-regulated treadmill tonomy index over the study period, particularly in the RPE 13
protocolddespite the higher elicited %VO2max during RPE participants, is testament to the fact that the participants
13 training in the current study (74% vs. 64%). However, a moved from being largely externally regulated to exercise to
critical difference of Parfitt et al's study was their consistent being more intrinsically regulated. While it is highly unlikely
1:1 supervision throughout each of the three 30-minute ses- that the participants would become truly intrinsically moti-
sions/wk for 8 weeks. This likely explains the 17% improve- vated (because the byproducts of exercise, such as breath-
ment in %VO2max in their study. The participants in the lessness and sweating, are not in themselves enjoyable),
current study were guided to complete three training sessions a allowing the participants to control what they were doing
week but were required to attend the gym for exercise on only enabled them to move along the self-determination contin-
one of these sessions. Further, the supervision was minimal: uum.6 This, along with the higher competence during training,
gym staff were available to support the individuals only if presumably enabled those who remained in the study to
requested to, which more closely reflect a natural gym, where continue to exercise over the 6 months post-training to
there is minimal supervision, unless booked in advance. maintain their levels of aerobic fitness.
A decline in aerobic fitness in the RPE 15 group 6 months Based upon the principles of hedonics10 and in support of
following training, compared with the post-training VO2max Williams's model,9 perceptions during training in Week 1 of
in the RPE 13 group, which was sustained after 6 months, was the program could have contributed to the decision to com-
hypothesized. While not measured in this study, it is likely that plete (or not) the training program. According to hedonic
a level of aerobic activity was maintained by the participants theory and the peakeend rule,37 behavioral decisions are made
in the RPE 13 group, accounting for the difference, particu- based upon how pleasant or unpleasant an experience is,
larly as they reported significantly more positive affect during particularly the peak and end experiences. A subsequent
the RPE 13-regulated training sessions compared with the analysis (that coded the participants based on program
RPE15 group. The acute affective response during exercise completion rather than training group) was run to examine any
predicts exercise behavior at 6 months and 12 months.7,8 differences between completers and dropouts. The data
Indeed, a one-unit difference in the feeling scale (the revealed a significant (F(1.83,55) ¼ 3.89, p < 0.05, h2p ¼ 0.12)
average difference recorded in this study) has been equated to outcome (completed, dropped out) by time (5 minutes, 10
at least 38 minutes a week of more moderate to vigorous minutes, 15 minutes, 20 minutes, 25 minutes, 30 minutes)
physical activity 6 months later.11 interaction for affect across the 30-minute training session.
Although the total cholesterol was lower at baseline in the Those who dropped out of the study (irrespective of RPE
RPE 13 group, the cholesterol values did not change for the training group) had a significantly lower affective response in
control group or for either RPE training group. The HDL the last 15 minutes of the 30-minute RPE-regulated training
values also remained similar for each group at each time point, session.
with HDL fractions (HDL/total cholesterol) also remaining
similar and within the desirable range for each group.36 It is Limitations
probable that the lack of change in cholesterol and HDL is
attributable to the quality of the overall sustained intensity and The study had some limitations, the most significant of
adherence to the exercise program, the insufficient duration of which was the level of attrition. Participants from all three
the exercise program (<150 min/wk), or the fact that the groups dropped out across the duration of the study, the most
cholesterol values were already within the desirable range and from the RPE 15 group. Although six (two from the RPE 15
would be less susceptible to change through an exercise pro- group) moved away, the combination of study demands and
gram. These explanations may also apply to the lack of change loss of interest potentially contributed to the high level of
in blood glucose levels. attrition. The unsupervised nature of the training does have
A significant observation in this study is the high level of high ecological validity and potentially highlights the tradeoff
dropout, particularly in comparison with Parfitt et al,4 where between level of supervision, initial compliance with training,
G. Parfitt et al. / Journal of Exercise Science & Fitness 13 (2015) 123e130 129

and long-term adherence. The 1:1 supervision adopted by 2. Eston RG, Davies BL, Williams JG. Use of perceived effort ratings to
Parfitt et al4 resulted in initial compliance but theoretically control exercise intensity in young healthy adults. Eur J Appl Physiol
Occup Physiol. 1987;56:222e224.
could have developed supervisor dependence rather than au- 3. Goosey-Tolfrey V, Lenton V, Goddard J, et al. Regulating intensity using
tonomy (not assessed in that study). Future research could perceived exertion in spinal cord-injured participants. Med Sci Sports
explore the use of perceptually regulated training and a Exerc. 2010;42:608e613.
tapering of supervision to maximize compliance with training 4. Parfitt G, Evans H, Eston RG. Perceptually-regulated training at RPE13 is
and long-term behavior change. Maximal aerobic fitness was pleasant and improves physical health. Med Sci Sports Exerc.
estimated in the study to avoid the negative affective response 5. Coquart JB, Tourny-Chollet C, Lemaitre F, et al. Relevance of the measure
produced in a maximal test,15 but it is acknowledged that there of perceived exertion for the rehabilitation of obese patients. Ann Phys
is error when extrapolating to an individual's HRmax.33 As a Rehabil Med. 2012;55:623e640.
result, fitness may have been under- or overestimated for some 6. Deci EL, Ryan RM. Intrinsic Motivation and Self-determination in
individuals, although across the study this should have been a Human Behavior. New York: Plenum; 1985.
7. Williams DM, Dunsiger S, Ciccolo JT, et al. Acute affective response to a
systematic error. The participants were not asked to keep a moderate-intensity exercise stimulus predicts physical activity participa-
record of what they did over the 6-month post-training period, tion 6 and 12 months later. Psychol Sport Exerc. 2008;9:231e245.
and it is therefore unclear whether the maintenance of fitness 8. Williams DM, Dunsiger S, Jennings EG, et al. Does affective valence
in the RPE 13 group was due to continued volitional exercise during and immediately following a 10-min walk predict concurrent and
over the period. Subsequent research could objectively future physical activity? Ann Behav Med. 2012;44:43e51.
9. Williams D. Exercise, affect and adherence: an integrated model and a
monitor behavior both during the training period and after it to case for self-paced exercise. J Sport Ex Psychol. 2008;30:471e496.
check compliance and the long-term impact of the training on 10. Kahneman D. Objective happiness. In: Kahneman D, ed. Wellbeing:
exercise behavior. Foundations of Hedonic Psychology. New York: Russell-Sage; 1999:3e25.
11. Ekkekakis P, Parfitt G, Petruzzello SJ. The pleasure and displeasure
Conclusion people feel when they exercise at different intensities: decennial update
and progress towards a tripartite rationale for exercise intensity pre-
scription. Sports Med. 2011;41:641e671.
The results support that RPE 13 and RPE 15-regulated 12. Hamlyn-Williams C, Freeman P, Parfitt G. Acute affective responses to
exercise improves fitness over an 8-week training period; prescribed and self-selected exercise sessions in adolescent girls: an
however, the improvement in the study was maintained only in observational study. BMC Sports Sci Med Rehabil. 2014;6:35.
the RPE 13 group at 6 months post-training. Acute affective 13. Lutz R, Lochbaum M, Turnbow K. The role of relative autonomy in post-
exercise affect responding. J Sport Behav. 2003;26:137e155.
responses were more positive and competence was higher in 14. Rose E, Parfitt G. Exercise experience influences affective and motiva-
the RPE 13 group than in the RPE 15 group during and tional outcomes of prescribed and self-selected intensity exercise. Scan J
following exercise. These motivational process responses and Med Sci Sports. 2012;22:265e277.
the changes in relative autonomy theoretically explain the 15. Evans H, Parfitt G, Eston R. Use of a perceptually-regulated test to
continued exercise behavior in the RPE 13 group. The training measure maximal oxygen uptake is valid and feels better. Eur J Sport Sci.
did not reduce metabolic risk factors, either because the ex- 16. Ilarraza H, Myers J, Kottman W, et al. An evaluation of training responses
ercise dose was not high enough (the improvements in fitness using self-regulation in a residential rehabilitation program. J Cardiopulm
although significant were not high) or because the data were Rehabil. 2004;24:27e33.
within normal ranges. 17. Boesch C, Myers J, Habersaat A, et al. Maintenance of exercise capacity
and physical activity patterns 2 years after cardiac rehabilitation. J Car-
diopulm Rehabil. 2005;25:14e21.
Conflicts of interest 18. Nikander R, Siev€anen H, Ojala K, et al. Effect of a vigorous aerobic
regimen on physical performance in breast cancer patients - a randomized
The authors declare no conflicts of interest. controlled pilot trial. Acta Oncol. 2007;46:181e186.
19. Matthews CE, Wilcox S, Hanby CL, et al. Evaluation of a 12-week home-
Funding/support based walking intervention for breast cancer survivors. Support Care
Cancer. 2007;15:203e211.
20. DiPietro L, Dziura J, Yeckel CW, et al. Exercise and improved insulin
This research was funded by a research development grant sensitivity in older women: evidence of the enduring benefits of higher
from the Sansom Institute for Health Research, Division of intensity training. J Appl Physiol. 2006;100:142e149.
Health Sciences, University of South Australia, Adelaide, 21. Moholdt T, Madssen E, Rognmo Ø, et al. The higher the better? Interval
Australia. training intensity in coronary heart disease. J Sci Med Sport.
22. Midgley AW, McNaughton LR, Wilkinson M. Is there an optimal training
Acknowledgments intensity for enhancing the maximal oxygen uptake of distance runners?:
Empirical research findings, current opinions, physiological rationale and
The authors would like to thank Belinda Norton, the practical recommendations. Sports Med. 2006;36:117e132.
research assistant who collected the data for the study. 23. Scherr J, Wolfarth B, Christle JW, et al. Associations between Borg's
rating of perceived exertion and physiological measures of exercise in-
tensity. Eur J Appl Physiol. 2013;113:147e155.
References 24. Morris M, Lamb KL, Hayton J, et al. The validity and reliability of pre-
dicting maximal oxygen uptake from a treadmill-based sub-maximal
1. Borg GAV. Borg's Perceived Exertion and Pain Scales. Leeds: Human perceptually regulated exercise test. Eur J Appl Physiol.
Kinetics; 1998. 2010;109:983e988.
130 G. Parfitt et al. / Journal of Exercise Science & Fitness 13 (2015) 123e130

25. Coquart JB, Garcin M, Parfitt G, et al. Prediction of maximal or peak oxygen 31. McAuley E, Wraith S, Duncan TE. Self-efficacy, perceptions of success,
uptake from ratings of perceived exertion. Sports Med. 2014;44:563e578. and intrinsic motivation for exercise. J Appl Soc Psychol.
26. Exercise and Sports Science Australia, Fitness Australia, Sports Medicine 1991;21:139e155.
Australia. Adult Pre-exercise Screening Tool. Albion, Australia: Exercise 32. Markland D, Tobin V. A modification to the behavioural regulation in
and Sports Science Australia; 2011. Available at: exercise questionnaire to include an assessment of amotivation. J Sport
wp/wp-content/uploads/Screen-tool-version-v1.1.pdf. Accessed 10.04.14. Exerc Pyschol. 2004;26:191e196.
27. Hardy CJ, Rejeski WJ. Not what, but how one feels: the measurement of 33. Tanaka H, Monahan KG, Seals DS. Age-predicted maximal heart rate
affect during exercise. J Sport Exerc Psychol. 1989;11:304e317. revisited. J Am Coll Cardiol. 2001;37:153e156.
28. Lang PJ. Behavioural treatment and bio-behavioural assessment: com- 34. American College of Sports Medicine. ACSM's Resources Manual for
puter applications. In: Sodowski JB, Johnson JH, Williams TA, eds. Guidelines for Exercise Testing and Prescription. 6th ed. Philadelphia:
Technology in Mental Health Care Delivery Systems. New Jersey: Ablex; Lippincott Williams & Wilkins; 2009.
1980:119e137. 35. Cohen J. Statistical Power Analysis for the Behavioural Sciences. 2nd ed.
29. Van Landuyt LM, Ekkekakis P, Hall EE, et al. Throwing the mountains Hillsdale, NJ: Erlbaum; 1988.
into the lakes: on the perils of nomothetic conceptions of the exercise- 36. Pollock ML, Wilmore JH. Exercise in Health and Disease. Philadelphia:
affect relationship. J Sport Exerc Psychol. 2000;22:208e234. Saunders; 1990.
30. McAuley E, Duncan T, Tammen VV. Psychometric properties of the 37. Kahneman D. A perspective on judgement and choice: mapping bounded
Intrinsic Motivation Inventory in a competitive sport setting: a confir- rationality. Am Psychol. 2003;58:697e720.
matory factor analysis. Res Q Exerc Sport. 1989;60:48e58.