GREEN, J. MATT; YANG, ZHANG; LAURENT, CHARLES M.; DAVIS, JON-KYLE; KERR, KELLY; PRITCHETT, ROBERT C.; BISHOP, PHILLIP A.
Ratings of perceived exertion (RPE) offer a valuable determinant of impending fatigue during exercise testing (1) and are widely accepted as an effective tool for prescribing aerobic exercise intensity (2,4,5,9,21). Foster et al. (7,8) have extended the utility of RPE by introducing the concept of session RPE (S-RPE), in which an individual subjectively estimates the overall difficulty of an entire workout after its completion. This differs from the common application of RPE in which acute exertional feelings are estimated during exercise. S-RPE seems to function well as a method for longitudinal evaluation during a training program, with special attention to overtraining syndrome (7). S-RPE as an overall evaluation of a training bout offers similar advantages to use of RPE for assessing intensity. For example, compared with physiological measures such as oxygen consumption (V˙O2), blood lactate ([La]), or heart rate (HR), perceptual measures are more convenient, easily assessed in field settings, require nominal technical expertise to measure or interpret, and require minimal equipment. In some situations, specific physiological mediators may contribute heavily to acute RPE. In other exercise conditions, multiple physiological mediators are operational, with no single variable dominating the perceptual process (15,18,19). Because it is a novel concept, factors potentially influencing S-RPE are not well understood.
Foster et al. (6) compared a subjective response (computed as S-RPE × exercise duration) against a summated HR zone method at a preestablished workload (90% individual anaerobic threshold) among three bouts with different durations (30, 60, and 90 min). The computed variable (S-RPE × duration) increased with duration (means and standard deviations: 30 min (130 ± 57); 60 min (270 ± 63); 90 min: (432 ± 67)). Isolating the subjective component of this computed variable (by dividing by duration) revealed similar mean subjective responses (30 min = 4.3, 60 min = 4.5, 120 min = 4.8) regardless of duration. Consequently, duration accounted for the bulk of differences, suggesting that the globally linked subjective response themselves related most closely with exercise intensity. Although not designed to identify factors mediating S-RPE, that study provided important information regarding the association between S-RPE, exercise duration, and intensity. Sweet et al. (23) found that S-RPE increased systematically with % V˙O2peak during cycling. In Foster et al. (6), S-RPE across a series of cycling interval bouts was sensitive to duration and intensity of the intervals. Whereas the magnitude and duration of intervals were carefully controlled, total work completed was not. Holding work constant while systematically varying intensity, environment, or some other independent measure of interest would extend the understanding of factors associated with S-RPE.
During resistance training, Sweet et al. (22) have shown that S-RPE increased concurrently with relative resistance (% 1RM). Conversely, we have shown that, when resistance bouts are continued to volitional exhaustion, acute as well as S-RPE corresponds more closely with total work (resistance × reps × sets) than % 1RM (Pritchett et al., unpublished data, 2007). The differences between Sweet et al. (22) and our findings (Pritchett et al., unpublished data, 2007) highlight the notion that S-RPE may be altered by multiple factors, most of which have yet to be identified.
Factors contributing to S-RPE are currently not definitive. It is plausible that factors altering cardiac or metabolic demand and, consequently, disrupting homeostasis to different degrees, either acutely or longitudinally, would impact S-RPE. This study examined the effects of heat gain, circulatory adjustment to temperature regulation (HR), and [La] consequent to interval (INT) and constant-load (CON) cycling on S-RPE. Exercise format (constant workload vs interval cycling) was used to manipulate [La], and ambient environment was adjusted to manipulate HR and heat gain. Bouts were completed in hot (~32.5 WBGT) and cool (~21.0 WBGT) environments. Total external work and exercise duration were equated. It was hypothesized that S-RPE would be increased in response to greater heat gain, HR, and [La].
The current study was designed to manipulate [La] by dictating exercise type and to control heat gain (Tre) and HR by manipulating environment. Using interval (vs constant-load) exercise bouts allowed a comparison of S-RPE under conditions of high versus low [La]. Likewise, manipulating the ambient temperature permitted an assessment of the effects of heat gain and circulatory adjustment (HR) on S-RPE. Holding work volume constant among trials ensured that variation in this variable would not function as a confounder. That is, completion of equivalent external work helped eliminate the possibility of differences in total work among trials contributing to S-RPE variation.
Ten physically active males (N = 10) were recruited for participation. Before data collection, subjects completed a written informed consent outlining participation requirements. On the basis of an alpha level of 0.05, a beta value of 0.80, a standard deviation of 1.6, and an effect size of 2.0 for S-RPE, an a priori power analysis indicated a need for 10 subjects. Procedures were approved by the institutional review board. Each participant arrived at the lab with instructions to be well hydrated, at least 3 h postprandial, and to have abstained from caffeine and alcohol for a minimum of 24 h. Age (yr) was recorded, along with height (cm) and body mass (kg), assessed using a balance scale (Detecto Scales Inc. Brooklyn, NY), with body fat percentage estimated using skinfold calipers (Lange, Cambridge, MD) and a three-site method (chest, abdomen, thigh) (17).
After descriptive data collection, participants completed a cycling trial to determine V˙O2peak. Seat height on a cycle ergometer (Monark, Varberg, Sweden) was adjusted for each individual so that a slight bend was observed in the knee, with the pedal at a 6 o‘clock position. Handlebars were adjusted according to individual preference. Participants were instructed to maintain a cadence of 60 rpm, paced by a metronome (Franz, Country Technology, Gays Mills, WI). Participants were fitted with an appropriately sized, air-cushioned face mask (Vacu-med, Ventura, CA) and a heart rate (HR) monitor transmitter (Polar, Stamford, CT) at the level of the sternum.
As a warm-up, participants pedaled at the required cadence (60 rpm) for 2 min at 0 W. After the warm-up, 30 W was added to the ergometer resistance every minute, with cadence being kept constant (60 rpm). This incremental protocol was followed until participants volitionally terminated exercise because of exhaustion, or until they could no longer maintain the required cadence. Metabolic data (V˙O2, V˙CO2, respiratory exchange ratio, ventilation) were collected using a metabolic measurement system (Vacu-med Vista mini-cpx, silver, Vacu-med, Ventura, CA). Software (Turbofit, Vacu-med, Ventura, CA), designed for use with the metabolic system, was set to report mean metabolic data for 15-s time periods. The system was calibrated before each test with a gas of known composition. A 7-L syringe (Hans Rudolph, Kansas City, MO) was used to calibrate the measurement of ventilation. Ratings of perceived exertion (RPE) were collected during the last 15 s of each stage using the Omni RPE pictorial 0-10 scale (24). The RPE scale was verbally anchored with subjects being told that "0“ corresponded to seated rest and that "9-10“ corresponded to maximal exertion. Participants were instructed to consider overall feelings of exertion when responding. Participants were also encouraged to ask questions for clarity if necessary. Criteria for achievement of V˙O2peak were a) RPE ≥ 9, b) RER ≥ 1.1, c) plateau of V˙O2 with increased workload, and d) > 85% of age-predicted maximum HR (14). A maximal effort for V˙O2peak was achieved if participants met two or more of these four criteria.
Constant-workload cycling (hot and cool).
Within 7 d of the V˙O2peak trial, participants reported to the lab with instructions to be well hydrated, at least 3 h postprandial, and having consumed no caffeine or alcohol for a minimum of 24 h. Participants were fitted with a HR monitor. To assess rectal temperature (Tre), subjects inserted a rectal thermocouple (RET-1, Physitemp, Clifton NJ) 8 cm beyond the rectal sphincter. The thermocouple was interfaced with an Isothermex electronic monitoring system interfaced with a computer system (Isothermex, Columbus Instruments, OH). Ergometer seat height and handlebars were adjusted as described previously.
Using data from the V˙O2peak trial, the power requirement for the follow-up trials was individualized for each participant to achieve the desire percentage (45 and 90%) of V˙O2peak. For constant-workload trials, a power output approximating 45% of V˙O2peak was used. This was determined by calculating 45% of V˙O2peak and by identifying the associated power output from V˙O2 values recorded during the V˙O2peak trial.
Participants began cycling with a warm-up of 10 min (0 W, 60 rpm). After the warm-up, the individualized power requirement was set, and subjects pedaled for 16 min, maintaining 60 rpm. During this trial, capillary blood samples were taken from the fingertip at 10, 13, 17, 21, 25, and 36 min, using a capillary tube (Analox Inc., Boston, MA). Time points for sampling were selected to match time points corresponding with sampling during INT (described later). Samples were analyzed for lactate concentration, using an enzymatic analyzer (Analox PGM-7, Analox Inc., Boston, MA), which was calibrated before each trial using an 8 mM standard according to the manufacturer‘s instructions. To ensure reliability, each sample was analyzed in duplicate. Two serial samples no greater than 0.2 mM apart were required, with the average of the two used for analysis. HR response, Tre, and acute RPE [RPE overall (RPE-O)] were recorded, using the 0-10 Omni RPE scale (24) at the conclusion of the warm-up (10 min) and in the last 10 s of each min thereafter, until 26 min (i.e., throughout the following 16 min of the CON or INT trials). At 26 min, participants completed a 10-min cool-down (0 W, 60 rpm). HR, RPE, and Tre were also recorded at the conclusion of the 10-min cool-down (at 36 min), at which time the trial was terminated.
The above procedures, including warm-up and cool-down, were completed by all participants in a hot (~32.5°C WBGT, wet bulb = 29.0, dry bulb = 40.2, black globe = 42.6°C) and a cool (~21.0°C WBGT, wet bulb = 19.7, dry bulb = 24.0, black globe = 24.5°C) environment. Hot trials were completed in an environmental chamber. Cool trials were completed in the laboratory space outside the chamber and with central heating/cooling systems used to adjust conditions.
Interval cycling (hot and cool).
For interval cycling trials, participants reported to the lab with instructions to be well hydrated, at least 3 h postprandial, and having consumed no caffeine or alcohol for a minimum of 24 h. Participants were prepped for testing as above with respect to HR and Tre. Ergometer seat height and handlebars were adjusted as before. Lactate analyses were also completed in the same manner.
After participant preparation, interval cycling was initiated with a 10-min warm-up (0 W, 60 rpm) in the same manner as for constant-load exercise trials. At the conclusion of the warm-up, a resistance equivalent to twice the wattage used for constant-load trials was added to the ergometer. Subjects pedaled for 1 min at their individualized interval resistance (as determined in the V˙O2peak trial). At the conclusion of this interval, wattage was returned to 0, and subjects pedaled for 1 min. Eight cycles of this interval/recovery sequence were completed (8 × 1-min intervals with 8 × 1-min recoveries, totaling 26 min). After the final interval/recovery cycle, participants completed a 10-min cool-down (0 W, 60 rpm). For interval trials, blood was sampled at 10, 13, 17, 21, 25, and 36 min, to correspond with the conclusion of the warm-up, the conclusion of intervals 2, 4, 6, and 8, and the conclusion of the cool-down (minute 36). Slight variations resulted in the spacing of blood samples because of the timing of warm-up and completion of intervals. Because timing points for samples were matched, this variation occurred for INT as well as CON. Interval cycling trials were completed by all participants in both hot and cool environments.
Regarding ordering, all four trials were listed in a sequence (interval/hot, interval/cool, constant/hot, constant/cool), with the first participant completing trials in this order. Each participant thereafter began his sequence with the second trial of the preceding participant. This counterbalanced approach controlled for ordering. Participants were given 2-7 d of recovery between trials.
Twenty minutes after the completion of the cool-down for each trial, participants were shown a copy of the 0-10 RPE scale and were asked, "how was your workout?“ with the subjective response recorded as the S-RPE. Participants were allowed to drink cold water ad libitum during this 20-min period; however, no activity was permitted, and participants were not permitted to shower or eat.
Interval and constant-load cycling trials were designed such that total external work completed by each participant was equated across all four trials. External work was equated by holding total wattage and duration constant within CON and within INT. For INT, wattage and duration for warm-up and cool-down were held constant, but the intervals were completed at twice the wattage used for CON, and the recovery (between intervals) was completed at 0 W. This procedure was followed to isolate the potential influence of thermal strain (hot vs cool) as well as lactate concentration (constant vs interval) on S-RPE. Each of these provides a challenge to homeostasis.
Basic descriptive characteristics were computed for the participants. S-RPE was compared using a 2 (trials) × 2 (environments) repeated-measures ANOVA with Fisher‘s LSD post hoc tests. Temperature gain (Tre increase) was computed as the difference in the highest versus lowest Tre observed and was labeled ΔTre. ΔTre was compared using a 2 (trials) × 2 (environments) repeated-measures ANOVA with Fisher‘s LSD post hoc tests. Lactate responses were compared using a 2 (trials) × 2 (environments) × 6 (time points) repeated-measures ANOVA with Fisher‘s LSD post hoc tests where appropriate. Of 240 potential samples, only three lactate samples were lost. For these cells, the mean value of the remaining nine participants was entered for that specific trial with respect to the interval versus constant and environment. For all analyses, results were considered significant at alpha ≤ 0.05.
HR responses among trials were compared at warm-up (at the 10-min point), cool-down (at 36 min), and by calculating a mean exercise HR (HR mean for minutes 11-26). HR at 10 min, mean exercise HR (HRmean), and HR at 36 min were entered into separate 2 (environments) × 2 (trials) ANOVA. When main effects were found, Fisher‘s LSD post hoc tests were used to compare CON versus INT and COOL versus HOT for 10 min, HRmean, and 36 min. This analysis was chosen because HR in the current study was evaluated as an indicator of overall cardiovascular strain, and responses at specific time points were not viewed as being as important as the global response. HR analysis, excluding 10 min of warm-up and a 10 min cool-down, permitted a more specific assessment of the response to heat, as this was the only independent variable modified in the initial portion of any trial. Analysis of the mean HR was used because during the INT trial, HR each minute cycled up and then down. Therefore, the mean HR was considered the most representative for the exercise period. Analysis at 36 min after cool-down permitted an assessment of acute recovery after the cycling trials, and it also permitted an indirect comparison of the differences in overall disruption in homeostasis among trials (because of environment and exercise type). These analyses provided a more meaningful evaluation of HR response in relation to S-RPE than an analysis of at specific time points. Acute RPE-O estimations were compared using 2 (trials) × 2 (environments) × 6 (time points) repeated-measures ANOVA with Fisher‘s LSD post hoc tests. Data points entered in the analysis were those corresponding with lactate samples (10, 13, 17, 21, 25, and 36 min).
Descriptive data are presented in Table 1. There was a significant main effect for S-RPE for environments, with HOT being significantly greater than COOL (Wilks‘ lambda: F1,9 = 18.6, P = 0.002). The main effect for exercise type tended towards a greater S-RPE for interval (INT) versus constant load (CON), with a P value approaching significance (Wilks‘ lambda: F1,9 = 4.32, P = 0.07). Follow-up tests showed COOL S-RPE significantly lower than HOT S-RPE for INT as well as CON. Additionally, INT S-RPE was significantly higher than CON S-RPE during HOT, with no significant difference (P = 0.22) (INT vs CON) during COOL. S-RPE responses are displayed in Figure 1.
Main effects for [La] were significantly higher for INT versus CON (P = 0.01), with no significant main effect for HOT versus COOL (P = 0.49). Follow-up tests for INT versus CON showed no significant difference for [La] at 10 min (HOT, P = 0.07; COOL, P = 0.41) but significantly higher [La] at 13, 17, 21, 25, and 36 min for INT versus CON within HOT and within COOL trials (Fig. 2). ΔTre showed a significant main effect (P = 0.001) for HOT versus COOL but no significant main effect for INT versus CON (P = 0.71). Follow-up tests for ΔTre showed significantly greater heat gain for HOT INT than COOL INT (P = 0.008) and significantly greater heat gain for HOT CON than COOL CON (P = 0.003).
FIGURE 2-Lactate (mM...Image Tools
There was a main effect for HOT versus COOL for HR at 10 min, HRmean, and 36 min. There was no main effect for trial. t-tests showed significantly greater HR for HOT versus COOL for 10 min, HRmean, and 36 min within CON as well as INT (Fig. 3).
FIGURE 3-HR (bpm) be...Image Tools
Main effects for acute RPE-O were: HOT versus COOL (P = 0.003) and INT versus CON (P = 0.07). For RPE-O at specific time points, INT-HOT was significantly greater than INT-COOL at 10, 13, 17, 21, 25, and 36 min. CON- HOT was significantly greater than CON-COOL at 10, 13, 17, and 25 min (Table 2). RPE-O was significantly higher for INT versus CON at 10, 21, and 25 min (HOT only). RPE-O was significantly higher for INT (vs CON) at 25 and 36 min (COOL only) (Table 2).
Foster et al. (7,8) have developed the concept of S-RPE, extending the utility of subjective assessment of exercise. S-RPE permits a rating of the perceived global difficulty of an exercise bout after its completion. Despite its potential utility, attributional aspects and factors possibly influencing S-RPE are not well understood, because of its novel nature. This study examined effects of heat gain (ΔTre), circulatory adjustment (HR), and [La] consequent to interval (INT) and constant-load (CON) cycling, as well as hot (HOT) versus cool (COOL) environment conditions on S-RPE. Ten males completed four cycling bouts, with external work equated among trials.
Current results indicate S-RPE was responsive to added stress introduced by a hot environment. The HOT condition resulted in increased S-RPE for a constant total work volume, whether exercise involved constant-load (~45% V˙O2peak) cycling or repeated, 1-min interval bouts (~90% V˙O2peak) (Fig. 1). Compared with COOL, HOT resulted in a magnified circulatory strain according to a significantly higher HR response (10 min, HRmean, 36 min) (Fig. 3). Higher HR under HOT was consistent for CON as well as INT. Similarly, HOT trials resulted in greater heat gain (ΔTre) for both CON and INT. Whereas elevations in HR and Tre were predictable for HOT compared with COOL, equating external work among trials helped isolate potential influence of environment and consequently HR and Tre on S-RPE. Amplified physiological responses (HR, ΔTre), coupled with greater S-RPE, indicate that heat gain could partially account for greater S-RPE. However, similar to acute RPE, it should be noted that an association with a physiological measure should not be interpreted as causal. Acute RPE has been coupled with various physiological measures such as [La] (3,15,16,20,21) and ventilation (3,15,18). However, it is accepted that no single factor universally dominates acute exertional perceptions (15,19). Because of its subjective nature, it is logical that S-RPE would also be influenced by multiple factors, with these results supporting this notion. A consistent link between HR and ΔTre suggests that overall heat gain and circulatory adjustment (HR) may be among the factors associated with, and potentially mediating, S-RPE. Conversely, there was no main effect for [La] (HOT vs COOL). Observing greater S-RPE with similar [La] suggests that [La] may not be a factor contributing to S-RPE in the current paradigm. Additional studies are warranted exploring the potential influence of [La] on S-RPE.
Comparing INT versus CON conditions with total overall work revealed a consistently higher S-RPE for INT (Fig. 1); however, only the difference observed in the HOT condition reached significance. With S-RPE proposed to reflect global exertion for the entire session, it would be expected that an exercise bout resulting in greater disruption in homeostasis would result in intensified S-RPE. With [La] reflecting greater glycolytic turnover, significantly greater [La] at all time points (except 10 min) indicates a greater disruption of homeostasis during INT than CON (Fig. 2). [La] systematically increased from time point to time point, except during cool-down (from 25 to 36 min) during INT, with a less dramatic increase followed by a subtle plateau during CON (Fig. 2). This pattern, anticipated on the basis of [La] kinetics observed during previous studies in our lab (10,12), was consistent independent of environment.
Whereas acute RPE has been linked with [La] (3,15,16,20,21), other studies show that perceived exertion and [La] diverge (10,13), refuting a consistent linkage in some exercise situations. Current results do not indicate that a global subjective measure after exercise (S-RPE) is heavily influenced by the [La] response during the exercise bout.
S-RPE was significantly higher for INT versus CON only for HOT, suggesting a correspondence with exercise [La] only when bouts were completed in a hot environment. When cycling was performed in a cool environment, elevated [La] resulted in a tendency towards greater S-RPE (CON = 4.3 ± 1.3, INT = 5.1 ± 2.0, P = 0.07); however, the difference failed to reach statistical significance.
Unlike [La], no significant differences were found for HR between CON and INT (Fig. 3). HR reflects cardiovascular demand, and it is not surprising that no overall differences were found, as total external work between trials was equated. Because HR also reflects metabolic demand, it is possible that S-RPE could reflect overall aerobic metabolic demand. However, V˙O2 measurements, which were not included in this study, would be necessary to confidently make this conclusion. It should also be noted that no measures were taken to control for any possible metabolic drift that may have occurred across the trials. Although controlling drift is possible, this would likely not have permitted tight control of external work.
Lack of a difference for heat gain for CON versus INT is also explained by equal work volume between trials. With no difference in HR, heat gain, or total work, the greater S-RPE for INT (HOT only) cannot be attributed to these factors. Alternatively, differences seen for the HOT condition seen between INT and CON could be attributed to elevated [La] (Fig. 2). If so, it is necessary to reemphasize that this only occurred for HOT, and elevated [La] was not associated with significantly greater S-RPE for COOL. Collectively, the current results confirm that a global subjective measure (S-RPE) may be mediated by different factors in different situations, a similar contention to that made regarding acute RPE (15,19).
With external work equated, the effects of exercise type (INT vs CON) and environment (HOT vs COOL) on S-RPE were isolated. S-RPE responded to magnified heat load, with less convincing evidence of an S-RPE response to amplified internal disruption (i.e., elevated [La]) resulting from interval versus constant-load cycling. Foster et al. (6) have identified a link between work completed and S-RPE for aerobic exercise. We have also found a close association of S-RPE with total work during strength training bouts completed to failure at 60% versus 90% 1RM (Pritchett et al., unpublished data, 2007). In the current study, the absence of an identified significant difference for S-RPE between INT and CON for COOL supports the link between total work and S-RPE. In contrast, INT versus CON for HOT, as well as comparisons between HOT and COOL, indicate that total work and S-RPE will not always correspond if other stressors are present.
Table 2 shows that acute RPE-O was significantly greater for HOT than for COOL at all time points for INT and at 10, 13, 17, 21, and 25 min for CON. This suggests that elevated subjective measures embedded within a bout are associated with greater S-RPE when comparing hot and cool environments. Elevated RPE-O during HOT reflects acute awareness of the added strain associated with exercise in the heat, with current results also indicating a global rating in response to the additional strain. To decrease the potential that participants would bias global ratings from acute ratings, S-RPE was requested 20 min after bouts. It is possible that this time frame may not entirely eliminate the residual effects of acute RPE on S-RPE. This warrants further investigation.
Compared with CON, for INT RPE-O was significantly greater for HOT (10, 21, and 25 min) and for COOL (25 and 36 min) (Table 2). Acute RPE drifts across extended-duration constant-load cycling (13,23) and throughout repeated interval cycling (10). It can be observed in Table 2 that RPE in the current study followed this pattern, increasing more across time for INT compared with CON, and the effects of such a drift likely accounted, at least in part, for acute RPE-O differences for INT versus CON. Acute RPE has been linked with various physiological markers, yet no single mediator dominates RPE in all exercise situations (15,19). Dissimilar physiological responses for HR, [La], and ΔTre, and acute RPE-O responses among trials, concurrent with an equal volume of work, offer introductory evidence that S-RPE is also dependent on different factors in different exercise situations.
In conclusion, this study explored the influence of thermal (HOT vs COOL) and metabolic stress (INT vs CON) and the resulting physiological strain (HR, ΔTre, and [La]) on S-RPE. More specifically, the results show that S-RPE was raised consequent to a hotter environment, with amplified values attributed to greater circulatory adjustment (HR) and greater heat gain (ΔTre). Increased S-RPE resulting from interval (vs constant load) exercise was limited to a hot environment. It is plausible that greater internal disruption ([La]) was partially responsible for greater S-RPE, even though [La] response indicated greater internal disruption for INT in a cool environment, with the S-RPE difference tending to be greater (although not significantly so) for INT. Collectively, the results suggest that S-RPE may be similar to acute RPE in that no single factor is dominant and that the exercise paradigm and environment alter the mediating factors. S-RPE varied significantly among trials, even with total work volume equated, which discounts a tight link between the examined factors. Conversely, S-RPE seems responsive to other factors (environment and exercise type), which may amplify the overall difficulty of an exercise bout. Because greater disruption of homeostasis may require greater recovery duration, S-RPE may consequently prove useful in evaluating recovery across exercise sessions. More work is warranted to determine the precision of this measure and to further elaborate the physiological as well as psychological factors influencing S-RPE.
The authors wish to extend their appreciation to the College of Education at The University of Alabama for financial support of this project, offered through the Faculty Research Grant Program.
1. American College of Sports Medicine. Guidelines for exercise testing and prescription
, 6th ed. Philadelphia, PA: Lippincott Williams and Wilkins, 2006.
2. Ceci, R., and P. Hassmen. Self-monitored exercise at three different RPE intensities in treadmill vs. field running. Med. Sci. Sports Exerc.
3. Demello, J., K. J. Cureton, J. Robin, and E. Boineau. Ratings of perceived exertion at lactate threshold in trained and untrained men and women. Med. Sci. Sports Exerc.
4. Dunbar, C. C., R. J. Robertson, R. Baun, M. J. Blandin, K. Metz, and R. Burdett. The validity of regulating exercise intensity by ratings of perceived exertion. Med. Sci. Sports Exerc.
5. Eston, R. G., B. L. Davies, and J. G. Williams. Use of perceived effort ratings to control exercise intensity in young healthy adults. Eur. J. Appl. Physiol.
6. Foster, C., J. A. Florhaug, J. Franklin, et al. A new approach to monitoring exercise training. J. Strength Cond. Res.
7. Foster, F. Monitoring training in athletes with reference to overtraining syndrome. Med. Sci. Sports Exerc.
8. Foster, C., E. Daines, L. Hector, A. C. Snyder, and R. Welsh. Athletic performance in relation to training load. Wisc. Med. J.
9. Glass, S. C., R. G. Knowlton, and M. D. Becque. Accuracy of RPE from graded exercise to establish exercise training intensity. Med. Sci. Sports Exerc.
10. Green, J. M., J. R. McLester, P. J. Wickwire, R. C. Pritchett, and R. G. Lomax. RPE association with lactate and heart rate during high intensity interval cycling. Med. Sci. Sports Exerc.
11. Green, J. M., T. R. Crews, A. M. Bosak, and W. W. Peveler. Overall and differentiated ratings of perceived exertion at the respiratory compensation threshold: effects of gender and mode. Eur. J. Appl. Physiol. 89:445-450, 2003.
12. Green, J. M., P. A. Bishop, H. E. Heath, R. G. Lomax, J. R. McLester, and I. H. Muir. Effects of high and low blood lactate concentrations on sweat lactate response. Int. J. Sports Med.
13. Green, J. M., J. R. McLester, T. R. Crews, P. J. Wickwire, R. C. Pritchett, and R. C. Redden. RPE dissociates from blood lactate during extended cycling. Eur. J. Appl. Physiol.
14. Maud, P. J., and C. Foster. Physiological Assessment of Human Fitness
. Champaign, IL: Human Kinetics, 1995.
15. Mihevic, P. M. Sensory cues for perceived exertion: a review. Med. Sci. Sports Exerc.
16. Noble, B. J., G. Borg, I. Jacobs, R. Ceci, and P. Kaiser. A category-ratio perceived exertion scale: relationship to blood and muscle lactates and heart rate. Med. Sci. Sports Exerc.
17. Pollock, M. L., D. H. Schmidt, and A. S. Jackson. Measurement of cardiorespiratory fitness and body composition in the clinical setting. Clin. Ther.
18. Robertson, R. J. Central signals of perceived exertion during exercise. Med. Sci. Sports Exerc.
19. Robertson, R. J., and B. J. Noble. Perception of physical exertion: methods, mediators, and applications. In: Exercise and Sports Science Reviews
, J. O. Holloszy (Eds.). Baltimore, MD: Williams and Wilkins, pp. 407-452, 1997.
20. Seip, R. L., D. S. Snead, E. F. Pierce, P. Stein, and A. Weltman. Perceptual responses and blood lactate concentration: effect of training state. Med. Sci. Sports Exerc.
21. Stoudemire, N. M., L. Wideman, K. A. Pass, C. L. McGinnes, G. A. Gaesser, and A. Weltman. The validity of regulating blood lactate concentration during running by ratings of perceived exertion. Med. Sci. Sports Exerc.
22. Sweet, T. W., C. Foster, M. R. McGuigan, and G. Brice. Quantitation of resistance training using the session rating of perceived exertion method. J. Strength Cond. Res.
23. Utter, A. C., J. Kang, D. C. Nieman, et al. Effect of carbohydrate ingestion and hormonal responses on ratings of perceived exertion during prolonged cycling and running. Eur. J. Appl. Physiol.
24. Utter, A. C., J. Robertson, J. M. Green, R. Suminski, S. R. McAnulty, and D. C. Nieman. Validation of the adult OMNI scale of perceived exertion for walking/running exercise. Med. Sci. Sports Exerc.