It is well known that numerous physical, physiological, and psychological factors play a vital role in shaping the perceived exertion response during exercise (23). It is also established that during single-joint muscle contractions, the intensity of subjective exertion increases as a function of lifting load (9), contraction intensity (18,21), and muscle fatigue (20,26). Results from the scientific literature have revealed some contrary findings that the perceived exertion response during resistance exercise does not differ or differs between men and women (7,17,19,21,28). Specific to submaximal fatiguing contractions involving the upper extremity, O’Connor et al. (17) demonstrated a lower rating of perceived exertion in young women than men, whereas Hunter and Enoka (7) observed that exertional perceptions were similar between men and women. These inconsistent findings occur despite the fact that men exhibited significantly higher values for voluntary force output when expressed in absolute and body mass-corrected units than women (8,20). The use of allometric-modeling in providing a body-mass corrected value for force output has been shown to be a useful procedure in elucidating such potential gender differences (3,12,13,29). In this context, applying the observed relation between body mass and a measure of peak force output provides an advantage for the scaling of various strength measures over traditional ratio-scaling methods. Conversely, women have demonstrated less muscle fatigue during sustained low-level or maximal effort contractions than men (4,5,7,20). Given these findings, the notion of a perceived exertion gender difference during resistance exercise necessitates further investigation. Furthermore, and specific to the lower extremity, very little data are available that document the possibility of a gender-specific perceived exertion response during resistance-type, fatiguing contractions.
One of the methodological challenges of assessing perceived exertion is the interindividual comparison. During fatiguing, repetitive muscle contractions during which perceived exertion is evaluated at discrete time intervals, a different number of ratings commonly occur between individuals. As this experimental design confounds interindividual comparisons on the raw data, appropriate modeling procedures can be used to approximate an individual-specific perceived exertion estimate at fixed intervals during the exercise task. Linear interpolation has been used where perceived exertion is estimated at 10% intervals during a sustained contraction (20). The power function is another acceptable approach that is based on the assumption that sensations grow as a power of a given stimulus (27). Both methods, however, may yield different estimations as: 1) power function modeling will constrain the increase in perceived exertion across repetitions to either a linear or an n-power dependent pattern, and 2) linear interpolation will retain the observed pattern as a linear function or a higher order polynomial across the repetitions to failure. In order to examine the presence of a gender-specific perceived exertion response during fatiguing resistance exercise, the influence of the employed estimation method should also be determined.
The objectives of the present study were to examine: 1) gender differences in the perceived exertion response, and strength, during a single set of continuous dynamic knee extensor contractions; and 2) two different approaches of modeling the perceived exertion response during fatiguing knee-extension exercise. It was hypothesized that: 1) men and women would not differ in the perceived exertion response during a single bout of fatiguing knee extensor exercise (18,21); 2) perceived exertion will demonstrate power function exponents less than 1.0, and will closely approximate a quadratic trend when estimated via linear interpolation (20); and 3) men would inherently possess greater knee extensor strength than women (8,19), when corrected for body mass, and, therefore, demonstrate a greater rate of fatigue (4,5,7).
METHODS AND MATERIALS
Subjects for this study consisted of 15 healthy men (mean age = 25.7 ± 3.9 yr, mean height = 179.5 ± 7.0 cm, mean mass = 79.7 ± 10.6 kg) and 15 healthy women (mean age = 22.4 ± 2.4 yr, mean height = 167.6 ± 5.3 cm, mean mass = 63.5 ± 10.1 kg). Based on previous investigations (18,21) and the within-subject main effect for contraction intensity on the perceived exertion response, pooled across gender, a sample of 6 is determined necessary to demonstrate significant effects (1 − β = 0.80, α = 0.05, η2 = 0.83). All subjects were physically active, as they reported performing various types of routine exercise (for example, jogging, cycling, or other aerobic activity) at least 2 d·wk−1. The subjects were familiar with inertial exercise (i.e., subjects reported previous experience) but were not actively engaging in a regular resistance program involving this exercise at the time of the study. Individuals with a reported history of cardiovascular disease, hypertension, or orthopedic pathology were excluded from participating in this study. The health status of all subjects was ascertained with the Physical Activity Readiness Questionnaire. All subjects provided written informed consent as approved by the Institutional Review Board at Eastern Washington University.
All testing procedures were performed on the same day. Subjects were asked to refrain for exercise activities at least 24 h before their scheduled test. Before the testing procedures, the subjects completed a warm-up period that consisted of submaximal cycling on a Monark ergometer for 5 min. The subjects were instructed to self-select a cycling resistance and pedal cadence that they felt was comfortable and would not induce fatigue. The subjects were then evaluated for their one-repetition maximum (1RM) during inertial knee-extension exercise. After a 2- to 3-min rest period, subjects were then instructed to perform a single set of dynamic knee extensions with a submaximal load to the point of failure.
Inertial Knee-Extension Exercise
Inertial knee-extension exercise was performed on a commercially available apparatus (Pro Leg Extension machine, Power Systems, Inc., Knoxville, TN). After the 5-min cycling period, subjects performed two sets of submaximal (five to seven repetitions) repetitions for warm-up and familiarization purposes with the right leg. The right leg of all subjects was evaluated because the apparatus did not permit testing of the left leg in isolation, without the possibility of assistance from the right leg. Lower-limb dominance was determined for the study for descriptive purposes by asking the subjects which leg they would preferably kick a soccer ball. All subjects reported to be right-leg dominant. The load that was selected for the warm-up procedure was subjectively determined by one investigator and by feedback from the subject. In this regard, all subjects were asked by the same investigator whether the load was sufficiently comfortable for a warm-up. After two sets of submaximal warm-up contractions (approximately five to seven repetitions per set), each subject’s 1RM was determined through trial and error by asking the subject to lift the maximum load they could two times. The loads that were selected for the 1RM testing were arbitrarily determined by the investigator and the subject. The investigator advanced the loads according to the apparent “ease” in which the subjects performed the two repetitions. During the 1RM determination, the incremented load was 4.54 kg. The 1RM was determined when the following criteria were met: subjects successfully performed the first repetition but failed to achieve full knee extension or lower the load in a controlled manner on the second repetition. These subjective criteria were based on repeated pilot testing and were judged by one investigator for all subjects. Subjects were specifically instructed to: 1) lift the load and achieve full knee extension, 2) hold the weight in the knee extended position for approximately 2 s, and 3) to lower the weight in slow and controlled manner. Performance of the knee-extension exercise in this manner was applied as the criteria to determine successful repetitions during all repetitions of the submaximal set. Although a metronome was not used to control the cadence, one investigator continually provided verbal instruction during each repetition for all subjects, according to the specified criteria. In this regard, the investigator indicated to the subjects to “… lift the weight, hold it, now lower slowly, and lift, hold it, lower slowly, and lift, hold it, lower slowly …,” from the first repetition through all successive repetitions. The subjects were required to grasp the metal handles on each side of the chair, in order to mimic the natural performance of the exercise, as designed by the apparatus. All subjects were provided with verbal encouragement by the same investigator in order to facilitate their 1RM. The subjects were provided with a minimum of 2 min of rest in between each 1RM attempt and were permitted additional rest time, if they so requested. During the rest period, the subjects were asked to remain seated and were allowed to freely move the right leg or rest it on a chair placed in front of the subject. The subjects were unaware of the load they were lifting, as a cardboard cover was placed in front of the weight stack in order to remove the influence of this knowledge on their performance. The average number of attempts of 1RM determination for the 30 subjects was 2.6 (range = 2–4).
After establishment of each subjects’ 1RM, they were asked to perform a single set of continuous knee extensions with a load equivalent to 50% of their 1RM to the point of failure. The criteria that were used to determine a successful repetition during 1RM establishment were also applied to each repetition of the single set. The task was discontinued if the subjects were unable to achieve full knee extension, if they were unable to maintain the cadence applied by the investigator, or if they chose to stop volitionally. As indicated earlier, a metronome was not used; a single investigator continually instructed the subject on correct performance of the knee-extension exercise during each repetition. The cadence of the knee-extension repetitions was regulated by the repeated instructions provided by the same investigator for all subjects. Ankle weights were secured to the weight stack of the knee-extension device to approximate the calculated 50% 1RM to the nearest 0.23 kg.
Measurement of Perceived Exertion
Perceived exertion was measured with a modified version of the original category-ratio scale (CR-10) developed by Borg (1). The modified CR-10 scale used presently eliminated the numerical rating of 0.5 and changed the verbal descriptors from “weak” and “strong”, to “light” and “hard,” which has been validated in a previous study (18). In order to provide the subjects in the present study with an understanding of their perceptual range, one high and one low anchor was applied (23). Immediately after establishment of the 1RM, the subjects were instructed to link their maximal feelings with the highest scale category; specifically, the investigator instructed the subjects to “think about the feelings in your quadriceps during the contraction, and consider those feelings as maximal.” In doing so, the subjects observed an enlarged copy (27.9 × 43.2 cm) of the modified Borg CR-10 scale while one investigator pointed to the “maximal” descriptor. After a 2-min period of rest, the subjects were instructed to sit quietly and to “think about the feelings in your quadriceps and consider those feelings a zero.” All subjects were instructed on the anchoring procedures before the contractions. Immediately before the 50% 1RM trial, subjects were instructed to “think about the feelings in your quadriceps during the contraction, and give a number from the scale about how your quadriceps feel at the end of each repetition.” The subjects were further instructed that they could provide a number higher than 10 if they so desired (15). In the present study, 15 subjects provided ratings higher than 10, particularly within the terminal portion of the fatiguing set of knee extensions. The subjects performed two repetitions for familiarization purposes before the actual 50% 1RM trial, and were instructed to “think about the feelings in your quadriceps and think of a number from the scale about how your quadriceps feel at the end of each repetition, but do not say the number out loud.” The purpose of this practice trial was to familiarize the subject with lifting the load while concurrently evaluating their effort and generating a number from the CR-10 scale. At least 2 min of rest separated the familiarization trials from the experimental trials.
As a result of the different number of repetitions performed among all the subjects, two different methods were used to allow intersubject comparisons of perceived exertion: linear interpolation and power function modeling.
The linear interpolation procedure involved the estimation of each subjects’ perceived exertion response every 10% of the perceptual continuum during the experimental trial to fatigue (20). The perceived exertion response was calculated for 10% to 100% time points in 10% increments for each subject, via a plot of each subjects’ data. The perceived exertion response at each percent was then estimated linearly between the upper and lower observed values corresponding to the percent value. The interpolation procedure was performed graphically by estimating the repetition number in which each 10% increment occurred for each subject. The perceived exertion response was then estimated between the upper the lower observed ratings about the determined repetition number, assuming linearity between these observed responses. The estimated perceived exertion responses from this procedure were then used for the subsequent statistical analyses.
Power function modeling.
Perceived exertion estimates across the single set to failure were calculated by fitting each subjects’ responses to the following power function: R = kxn, where R is the perceived exertion response, k is the constant of proportionality, x is the percent contraction duration (10–100%, 10% increments), and n is any real number. The natural logarithm (loge) was calculated for the perceived exertion responses and the percent contraction durations for each subject and plotted, and linear regression was applied to estimate the slope (n in the power function) and the y-intercept (k in the power function), (logeR = logek + nlogex). The calculated regression function was then raised to the natural base (e) to obtain the proportionality constant (k). The estimated perceived exertion response for each subject at each percent increment (i.e., 10–100% of repetitions, 10% increments) was then calculated from the derived power functions.
Gender differences for 1RM examined via the three following methods: 1) absolute units (kg), 2) relative units ([mass lifted/body mass] × 100%), and 3) and allometric modeled units. Using the allometric method, the relationship between 1RM mass and body mass was modeled according to the following power function ratio: y = 1RM mass/body massn, where n is the exponent of the power function determined via a log-linear analysis (13). Gender was factored into the regression model as a coded variable (“0” for women, “1” for men) to yield a mass exponent common to both men and women. The natural logarithm (loge) was calculated for the 1RM mass and body mass for each subject; the data from all subjects were then plotted and linear regression was applied in order to estimate the slope (logey = logea + blogeM + cG), where G is gender, M is body mass, and b is the derived exponent. The y-intercept value generated separately for men and women were raised to the natural base (e) to determine the proportionality constant (k) of the power function. An independent t-test was performed separately on each measure (i.e., absolute, relative, and allometric modeled 1RM) to assess gender differences. An adjustment for an inflated Type I error rate was made for the multiple t-tests by the Bonferroni-Dunn inequality (alpha level 0.05/3 = 0.017). A single factor ANCOVA was used to assess gender differences in the number of repetitions performed to failure, with the 1RM mass as the covariate.
A two-factor (duration by gender) ANOVA, with repeated measures on the duration factor, was performed separately on the estimated perceived exertion values from the linear interpolation and power function calculations. After the ANOVA, repeated pair-wise contrasts were performed over successive levels of contraction duration to investigate specific changes across each percent increment as a function of gender. The Bonferroni-Dunn inequality was invoked to ensure the family-wise error rate for the duration main effect (0.05/10 = 0.005) and individual duration by gender interactions (0.05/10 = 0.005). A trend analysis was performed, via the F-test, on the calculated perceived exertion values from the linear interpolation method to examine the linear, quadratic, or cubic change across the duration of the 50% 1RM trial. Regression analyses were further conducted on the interpolated (averaged, N = 30) perceived exertion values to compute the r2, standard error of estimate (SEE) and the residuals for each trend (i.e., linear, quadratic, and cubic). Standardized residuals were subsequently calculated according to the following: esi = ei/SEE, where ei is the computed residual (difference between the interpolated perceived exertion value and the predicted value from each trend) (6). Descriptive data were generated for the calculated power function exponent and proportionality constants (mean ± standard deviations, and 95% confidence intervals) for men and women, separately. A one-sample t-test was used to examine differences between the calculated exponent values versus a test value of 1.0 (indicative of a linear increase in perceived exertion). Gender differences in the exponents were also examined with an independent t-test.
The effect of the modeling procedure on the estimated perceived exertion values was examined with a two-factor (duration by model) ANOVA with repeated measures. After the ANOVA, repeated pair-wise contrasts were performed over successive levels of contraction duration to investigate specific changes across each percent increment as a function of model. The Bonferroni-Dunn inequality was used to ensure the family-wise error rate of 0.05 for the duration main effect (0.05/9 = 0.006) and individual duration by model interactions (0.05/9 = 0.006). Regression analyses were also performed on the interpolated perceived exertion values for each subject to calculate the r2 and SEE values for each trend (i.e., linear, quadratic, and cubic). These values were then examined with a two-factor ANOVA (gender by trend), with post hoc repeated pair-wise contrasts and the Bonferroni-Dunn inequality to ensure the family-wise error rate of 0.05 for the trend main effect (0.05/2 = 0.025). All statistical analyses were performed with the SPSS for Windows software program, version 11.0 (SPSS Inc., Chicago, IL).
Knee-extension strength and fatigue.
The results demonstrated that during the 1RM testing, men lifted significantly more mass (78.17 ± 12.85 kg) than the women (52.01 ± 10.96 kg) (t28 = 6.0, P < 0.05). When normalized to body mass, men were shown to lift significantly more mass (98.52% ± 13.24%) than women (82.07% ± 12.11%) (t28 = 3.56, P < 0.05). The derived regression model, via allometric scaling, was y = 0.614 + 0.801M + 0.231G (r2 = 0.73, SEE = 0.15), yielding the following equations for men and women, respectively: y = 0.845 + 0.801M (k = 2.33), and y = 0.614 + 0.801M (k = 1.85). The power function ratios demonstrated a significant difference (t28 = 4.47, P < 0.05) as men yielded higher values (2.35 ± 0.31) than women (1.87 ± 0.28). After adjustment for the family-wise error rate of 0.05, this comparison was found to be statistically significant. Figure 1 illustrates the scatterplot of 1RM strength (kg) versus body mass (kg). The number of repetitions performed to failure was observed to be 12.13 ± 2.42 for men and 13.13 ± 2.0 for women. The results demonstrated no significant differences in the number of repetitions performed when adjusted for 1RM mass (adjusted mean ± standard error; men: 11.93 ± 0.73, women: 13.27 ± 0.73; F1,27 = 1.20, P = 0.28, η2 = 0.04, 1 − β = 0.19).
Perceived exertion interpolation.
The results showed a significant increase in perceived exertion across the duration of the knee-extension repetitions to failure (F9,252 = 583.17, P < 0.001, η2 = 0.95), with no significant gender main effect observed (F1,28 = 1.27, P = 0.27, η2 = 0.04, 1 − β = 0.19). A significant duration by gender interaction was also observed (F9,252 = 5.84, P < 0.001, η2 = 0.17) (Table 1). Specifically, significant interactions were noted between duration levels 50–60% (F1,28 = 6.20, P = 0.02, η2 = 0.18, 1 − β = 0.67), 70–80% (F1,28 = 6.92, P = 0.01, η2 = 0.20, 1 − β = 0.72), and 80–90% (F1,28 = 7.04, P = 0.01, η2 = 0.20, 1 − β = 0.73) as perceived exertion appeared to increase significantly more in women than men. After adjustment for the family-wise error rate, however, these effects were not found to be statistically significant. The results of the trend analysis revealed that the increase in perceived exertion was significantly linear (F1,29 = 902.9, P < 0.001, η2 = 0.97), quadratic (F1,29 = 5.20, P = 0.03, η2 = 0.15), and cubic (F1,29 = 44.54, P < 0.001, η2 = 0.61). Figure 2 illustrates the interpolated perceived exertion means and standard deviations at each increment in exercise duration, and the cubic (third-order polynomial) fit separately to the men and women. The calculated regression equations for each trend are as follows: linear: y = 0.968x + 1.256; quadratic: y = 0.017x2 + 0.780x + 1.634; cubic: y = −0.014x3 + 0.254x2 + −0.313x + 2.865, where y = perceived exertion response and x = percent contraction duration. Figure 3 illustrates the computed residuals for the linear, quadratic and cubic trends, and demonstrated that all residuals were less than two standard deviations about the mean, suggesting the presence of no outliers and, therefore, appropriate fits of each trend to the data (6).
Perceived exertion power function.
The average exponent and proportionality constant, respectively, of the power function (mean ± standard deviation) was found to be 0.57 ± 0.16 (95% C.I. = 0.49–0.65) and 9.78 ± 1.24 (95% C.I. = 9.10–10.47) for the men, and 0.72 ± 0.16 (95% C.I. = 0.63–0.81) and 10.20 ± 1.15 (95% C.I. = 9.56–10.84) for the women. The exponents were found to be significantly different between men and women (t28 = −2.07, P = 0.047), and were found to be statistically different than a test value of 1.0 for men (t14 = −6.70, P < 0.001) and women (t14 = −10.74, P < 0.001). The perceived exertion values predicted from the generated power functions for each subject revealed a significant duration main effect (F9,252 = 930.48, P < 0.001, η2 = 0.97), a significant duration by gender interaction (F9,252 = 6.58, P < 0.01, η2 = 0.19), and no significant gender main effect (F1,28 = 0.59, P = 0.45, η2 = 0.02, 1 − β = 0.12). Interactions were noted across each successive level of contraction duration beyond level 30% (Fig. 4), as perceived exertion appeared to increase significantly more in women than men. After family-wise adjustment for the a priori alpha level, however, these interactions were not found to be statistically significant.
In terms of the two different estimation models, the results demonstrated a significant duration main effect (F9,261 = 715.64, P < 0.001, η2 = 0.96), and a duration by model interaction (F9,261 = 46.75, P < 0.001, η2 = 0.62), with no significant model main effect (F1,29 = 0.14, P = 0.71, η2 = 0.005, 1 − β = 0.07) (Fig. 5). Specific interactions were observed between durations 10% and 20% (F1,29 = 43.55, P < 0.001, η2 = 0.60), and 20% and 30% (F1,29 = 29.33, P < 0.001, η2 = 0.50), as perceived exertion increased significantly more across these levels with the power function modeling approach than linear interpolation. Significant interactions were also noted between durations 50% and 60% (F1,29 = 31.03, P < 0.001, η2 = 0.52), 60% and 70% (F1,29 = 42.21, P < 0.001, η2 = 0.59), and 70% and 80% (F1,29 = 14.18, P = 0.001, η2 = 0.33), as perceived exertion estimated from the linear interpolation method increased significantly more across these levels than the power function approach. After adjustment for the a priori alpha level, these interactions were found to be statistically significant.
Descriptive data for the r2 and SEE values for each perceived exertion trend are summarized in Table 2. The results of the gender by trend analysis on the r2 values revealed a significant trend main effect (F2,56 = 3.75, P = 0.03, η2 = 0.12) and no gender main effect (F1,28 = 3.56, P = 0.07, η2 = 0.11, 1 − β = 0.44) or trend by gender interaction (F2,56 = 0.60, P = 0.56, η2 = 0.02, 1 − β = 0.14). Repeated pair-wise contrasts demonstrated a significantly higher r2 value for the cubic trend, as compared with the linear trend (F1,28 = 36.75, P < 0.001, η2 = 0.57). The cubic and quadratic trends were not found to be significantly different. The SEE values were found to exhibit a significant trend main effect (F2,56 = 23.46, P < 0.001, η2 = 0.47) and no gender main effect (F1,28 = 0.04, P = 0.85, η2 = 0.001, 1 − β = 0.05) or trend by gender interaction (F2,56 = 0.10, P = 0.90, η2 = 0.004, 1 − β = 0.07). The repeated pair-wise contrasts demonstrated significantly greater SEE values for the linear and quadratic trends, as compared with the cubic trend (linear by cubic contrast:F1,28 = 51.22, P < 0.001, η2 = 0.65; quadratic by cubic contrast:F1,28 = 20.37, P < 0.001, η2 = 0.42). The SEE values for the linear and quadratic trends were not found to be significantly different.
The major findings of this study demonstrated that modeling the perceived exertion response during the 50% 1RM trial with either method (i.e., linear interpolation or power function) resulted in no significant differences between men and women. Although the results of the overall test suggest duration by gender interactions, adjustment of the inflated alpha level with repeated contrasts demonstrated these differences were not statistically significant. The findings also show that men are able to lift a greater amount of mass during a single-leg, knee-extension exercise than women, when corrected for body mass. However, the ability to perform a maximal number of dynamic knee-extension repetitions with the same relative load (i.e., 50% 1RM) did not differ between genders. Modeling the perceived exertion responses via linear interpolation and the power function approach appeared to generate discrepant values at the beginning (10–30% of the duration) and within the middle to late part (50–80% of the duration) of the 50% 1RM trial. In this regard, the estimated perceived exertion response increased more rapidly via the power function prediction at the beginning of the trial but increased more through linear interpolation within the latter portion of the trial.
The hypotheses related to the perceived exertion findings were partially supported in the present investigation. The increase in perceived exertion during the fatigue task was not different between genders, although the results demonstrated that the statistical power values were less than an optimal value of 0.80, which may have been due to small sample sizes. In a recent investigation, O’Connor et al. (17) observed that women rated their perceived exertion lower than men, via Borg’s 6–20 scale, during repeated eccentric elbow-flexion contractions using a relative weight (80% concentric 1RM). However, all subjects in that study performed a predetermined equal number of repetitions at three different submaximal levels, such that the endurance limit was not necessarily achieved for all individuals. As young adult women have demonstrated significantly longer submaximal endurance limits than men (4,5,7,10), the lower ratings for women observed by O’Connor et al. (17) may have been reflective of lower physiological stress and, therefore, lower perceptual effort. Furthermore, Hunter and Enoka (7) found that during a sustained isometric elbow flexor contraction at 20% of the maximal voluntary contraction (MVC) to failure, perceived exertion did not differ between men and women at the beginning, middle, and end of the contraction. Specific to dynamic resistance exercise, Tomporowski (28) observed that previously sedentary men and women rated similar levels of perceived exertion after leg-extension exercise. During brief, nonfatiguing contractions, nonsignificant gender differences were also observed for isometric and isokinetic knee extensions (21). In considering the results of the present study, the notion of a gender-specific perceived exertion response during resistance exercise remains inconclusive.
The growth in the perceived exertion response during fatiguing knee-extension exercise was clearly shown to follow a nonlinear pattern. Similar results were also observed by Stevens and Cain (26) during a sustained handgrip task, in which a power function exponent of 0.57 was demonstrated. The negatively accelerating power function exponents of 0.57 and 0.72 for men and women, respectively, is suggestive of higher perceptual sensitivity within the initial portion of the exercise task, and lower perceptual sensitivity toward the terminal phase (2). This result is consistent with a previous study (20) demonstrating a more rapid increase in perceived exertion within the first 30% of a submaximal, isometric knee extension, to which EMG activity of the vastus lateralis and vastus medialis muscles followed a similar pattern. This fact may be related to an inherent ceiling effect of the scale during these types of contractions (15). In addition, the cumulative effect of developing muscle fatigue during the repeated contractions may have heightened perceptual sensations early in the task when discharge of thermo- and chemo-sensitive Type III and IV afferent receptors (24) may have been initiating. This may be particularly important as the quadriceps femoris muscle was continually loaded throughout the exercise task, via the concentric, isometric, and eccentric phases, which would have minimized recovery between repetitions. Although Steven’s psychophysical model (27) supports a perceived exertion response that follows a power function, the cubic trend observed in the present study differs from the hypothesized quadratic pattern. The patterns depicted in Figures 3 and 5 (linear interpolation curve) suggest that the cubic function retains the plateau effect near the terminal portion of the task (>70% contraction duration) while demonstrating an accelerating growth over the earlier phase (<70% contraction duration). The analyses of the present study also demonstrated that the cubic trend may be a better model than the quadratic model, due to the significantly lower error estimates of the third order trend. Furthermore, upon visual inspection of the data once plotted, the cubic fit proved to be a better representation of the perceived exertion response, particularly because the duration by model ANOVA results demonstrated opposite changes in perceived exertion at the beginning and terminal phases of the exercise task when compared with the power function model. However, this notion is speculative at present as the r2 values were not significantly different between these two trends. The physiological mechanisms contributing to this pattern remain unresolved at the present time, although preliminary evidence suggests that markers of muscle usage, such as EMG, blood lactate, and electroencephalographic patterns, may provide insight into the link with perceived exertion (9,16,20).
Knee extensor strength and fatigue.
It can be reasonably surmised that the ability of men to lift more absolute mass than women is a function of their greater body mass. However, correction of the absolute mass lifted by body mass, via ratio- and allometric-scaling, still yielded significantly higher values for the men. With respect to strength values that are scaled as a ratio of body mass, previous investigations have observed either greater values for men (19) or no differences (12). Based on biological similitude theory and dimensional analysis the scaling of physiological variables, such as strength, to body mass (2/3) has been suggested to be a viable approach for examining gender differences (13). It should be noted that the present investigation applied the observed (i.e., statistical) relation into the scaling procedure. Similar body mass exponent values to those determined in the present study (0.801) have been observed by: Vanderburgh and Dooman (29) for powerlifting (exponents = 0.63–0.87), Davies and Dalsky (3) for isokinetic knee extensor peak torque (exponents = 0.67 for women and 0.72 for men), and Neder et al. (12) for isokinetic knee extensor peak torque (exponents = 0.913 for women and 1.054 for men). Although the present study did not incorporate more specific measures of muscle mass, the greater strength values for the men may be attributed to a greater cross-sectional area of existing fiber types (11) and a greater proportion of fast-twitch muscle fibers (25) in the quadriceps femoris muscle. It is also suggested that greater “muscularity” (defined as the ratio of skeletal muscle to adipose tissue-free mass) in men compared with women, may also account for observed strength differences, although this relation has yet to be determined (30).
Findings within the scientific literature suggest that women have an apparent fatigue resistance advantage during sustained contractions (4,5,7). During a low-level contraction of the elbow flexors (20% MVC), Hunter and Enoka (7) observed a 118% longer endurance time in women than men; however, after adjustment of the endurance time for the maximal voluntary contraction via ANCOVA, no gender differences were present. Similar results were observed in a recent study, as a covariance adjustment of isokinetic quadriceps femoris muscle fatigue for peak torque resulted in no significant gender differences during 30 maximal effort knee-extension repetitions (22). These findings are contrary to the present investigation, as the covariance analysis did not reveal significant gender differences in muscle fatigue. When matched for absolute adductor pollicis muscle strength, Fulco et al. (4,5) found that men experienced a faster decline than women during intermittent submaximal contractions. However, Maughan et al. (10) and Ng et al. (14) observed no gender difference in isometric knee extensor endurance time for contractions at 50% and 80% MVC, and 30% and 50% MVC, respectively. These latter studies concur with the results of the present investigation and are contrary to the hypothesized gender difference, as the number of repetitions performed to failure did not differ between men and women. These conflicting outcomes underscore the importance of continuing studies to investigate task specific patterns of muscle fatigue between men and women.
The results from the present investigation demonstrated that perceived exertion during fatiguing knee extensor contractions did not differ significantly between men and women, despite the fact that men were able to lift a greater amount of mass than women when corrected for body mass. It should be noted that this outcome may be reflective of the finding that low statistical power for the gender main effect was observed. From an experimental perspective, a modeling approach is necessary in order to compare the perceived exertion responses throughout a given task that is performed to volitional failure. As a result, the perceived exertion response throughout the exercise task was fit to a power or cubic polynomial function. What remains to be answered in the present study, however, is which model is more accurate. It is, therefore, suggested that future approaches should attempt to model the physiological processes that occur concomitantly with these ratings throughout a given task. To this end, the results from this study suggest that although either modeling approach appears to be appropriate, concurrent validation to physiological measures is warranted.
This work was funded through a Faculty Research Grant at Eastern Washington University awarded by the Office of Grants and Research Development.
1. Borg, G. A. V. Psychophysical bases of perceived exertion. Med. Sci. Sports Exerc. 14: 377–381, 1982.
2. Cooper, D. F., G. Grimby, D. A. Jones, and R. H. T. Edwards. Perception of effort in isometric and dynamic muscular contraction. Eur. J. Appl. Physiol. 41: 173–180, 1979.
3. Davies, M. J., G. P. Dalsky. Normalizing strength for body size differences in older adults. Med. Sci. Sports. Exerc. 29: 713–717, 1997.
4. Fulco, C. S., P. B. Rock, S. R. Muza, et al. Slower fatigue and faster recovery of the adductor pollicis muscle in women matched for strength with men. Acta Physiol. Scand. 167: 233–239, 1999.
5. Fulco, C. S., P. B. Rock, S. R. Muza, et al. Gender alters impact of hypobaric hypoxia on adductor pollicis muscle performance. J. Appl. Physiol. 91: 100–108, 2001.
6. Glantz, S. A., B. K. Slinker. Primer of Applied Regression and Analysis of Variance. New York: McGraw-Hill, Inc., 1980, pp. 110–180.
7. Hunter, S. K., and R. M. Enoka. Sex differences in the fatigability of arm muscles depends on absolute force during isometric contractions. J. Appl. Physiol. 91: 2686–2694. 2001.
8. Kanehisa, H., H. Okuyama, S. Ikegawa, and T. Fukanaga. Sex differences in force generation capacity during repeated maximal knee extensions. Eur. J. Appl. Physiol. 73: 557–562, 1996.
9. Legally, K. M., R. J. Robertson, K. I. Gallagher, et al. Perceived exertion, electromyography, and blood lactate during acute bouts of resistance exercise. Med. Sci. Sports Exerc. 34: 552–559, 2002.
10. Maughan, R. J., M. Harmon, J. B. Leiper, D. Sale, and A. Delman. Endurance capacity of untrained males and women in isometric and dynamic muscular contractions. Eur. J. Appl. Physiol. 55: 395–400, 1986.
11. Miller, A. E., J. D. Macdougall, M. A. Tarnopolsky, and D. G. Sale. Gender differences in strength and muscle fiber characteristics. Eur. J. Appl. Physiol. 66: 254–262, 1993.
12. Neder, J. A., L. E. Nery, A. C. Silva, S. Andreoni, and B. J. Whipp. Maximal aerobic power and leg muscle mass and strength related to age in non-athletic males and females. Eur. J. Appl. Physiol. 79: 522–530, 1999.
13. Nevill, A. M., and R. L. Holder. Scaling, normalizing, and per ratio standards: an allometric modeling approach. J. Appl. Physiol. 79: 1027–1031, 1995.
14. Ng, A. V., J. C. Agre, P. Hanson, M. S. Harrington, and F. J. Nagle. Influence of muscle length and force on endurance and pressor responses to isometric exercise. J. Appl. Physiol. 76: 2561–2569, 1994.
15. Noble, B. J., R. J. Robertson. Perceived Exertion. Champaign, IL: Human Kinetics, 1996, pp. 4–71.
16. Nybo, L., and B. Nielsen. Perceived exertion is associated with an altered brain activity during exercise with progressive hyperthermia. J. Appl. Physiol. 91: 2017–2023, 2001.
17. O’Connor, P. J., M. S. Poudevigne, and J. D. Pasley. Perceived exertion responses to novel elbow flexor eccentric action in women and men. Med. Sci. Sports Exerc. 34: 862–868, 2002.
18. Pincivero, D. M., A. J. Coelho, and W. Erikson. Perceived exertion during isometric quadriceps
contractions: a comparison between men and women. J. Sports Med. Phys. Fitness 40: 319–326, 2000.
19. Pincivero, D. M., W. S. Gear, R. L. Sterner, and R. G. Karunakara. Gender differences in the relationship between quadriceps
work and fatigue during high intensity exercise. J. Strength Cond. Res. 14: 202–206, 2000.
20. Pincivero, D. M., and W. S. Gear. Neuromuscular activation and perceived exertion during a high intensity, steady-state contraction to failure. Muscle Nerve 23: 514–520, 2000.
21. Pincivero, D. M., A. J. Coelho, R. M. Campy, Y. Salfetnokov, and A. Bright. The effects of voluntary contraction intensity and gender on perceived exertion during isokinetic quadriceps
exercise. Eur. J. Appl. Physiol. 84: 221–226, 2001.
22. Pincivero, D. M., C. B. Gandaio, and Y. Ito. Gender-specific knee extensor torque, flexor torque, and muscle fatigue responses during maximal effort contractions. Eur. J. Appl. Physiol. 89: 134–141, 2003.
23. Robertson, R. J., and B. J. Noble. Perception of physical exertion: methods, mediators, and applications. In: Exercise and Sport Science Reviews, J. O. Holloszy (Ed.). Baltimore: Williams & Wilkins, 1997, pp. 407–452.
24. Rotto, D. M., and M. P. Kaufman. Effect of metabolic products of muscular contraction on discharge of group III and IV afferents. J. Appl. Physiol. 64: 2306–2313, 1988.
25. Simoneau, J. A., and C. Bouchard. Human variation in skeletal muscle fiber-type proportion and enzyme activities. Am. J. Physiol. 257: E567–E572, 1989.
26. Stevens, J. C., and W. S. Cain. Effort in isometric muscular contractions related to force level and duration. Percept. Psychophys. 8: 240–244, 1970.
27. Stevens, S. S. On the psychophysical law. Psychol. Rev. 64: 153–181, 1957.
28. Tomporowski, P. D. Men’s and women’s perceptions of effort during progressive-resistance strength training. Percept. Mot. Skills 92: 368–372, 2001.
29. Vanderburgh, P. M., and C. Dooman. Considering body mass differences, who are the world’s strongest women? Med. Sci. Sports Exerc. 32: 197–201, 2000.
30. Wang, Z., M. Heo, R. C. Lee, D. P. Kotler, R. T. Withers, and S. B. Heymsfield. Muscularity in adult humans: proportion of adipose tissue-free body mass as skeletal muscle. Am. J. Hum. Biol. 13: 612–619, 2001.