No Sex Difference in Mental Fatigue Effect on High-Level Runners’ Aerobic Performance : Medicine & Science in Sports & Exercise

Journal Logo


No Sex Difference in Mental Fatigue Effect on High-Level Runners’ Aerobic Performance


Author Information
Medicine & Science in Sports & Exercise: October 2020 - Volume 52 - Issue 10 - p 2207-2216
doi: 10.1249/MSS.0000000000002346
  • Free


Professional athletes are commonly exposed to mental stressors including, for example, inhibitory control to deal with dietary and social restrictions, exercise and injury-related discomfort and painful sensation, and emotional regulation during competitions (1). Altogether mental stressors may lead to a psychobiological state of mental fatigue, which is characterized by increased feelings of tiredness and lack of energy, as well as by decreased cognitive performance (2–4). Of note, mental fatigue per se may be deleterious to exercise performance (2,4,5). The reason is that several studies have shown that experimentally induced mental fatigue via prolonged cognitive tasks decrease exercise performance in physically active subjects (4,5). The decrease of exercise performance has not been accompanied by changes in cardiorespiratory and metabolic responses to exercise (2,4–6), but rather by greater perceived effort response to exercise (2,4–6). Therefore, enhancement of perceived effort has been implicated as a key factor responsible for mental fatigue–induced impairment of exercise performance in physically active subjects (4).

A caveat about mental fatigue in the context of exercise physiology and neuroscience is whether mental fatigue per se worsens perceived effort and exercise performance in professional athletes. Few studies have investigated this issue, and so far, available studies have provided conflicting results. Filipas et al. (7) found that a 30-min cognitive task provoked mental fatigue in professional cyclists younger than 23 yr. Oppositely, Martin et al. (1) found that the same cognitive task did not provoke mental fatigue in professional adult cyclists. A possible explanation for the conflicting results may be the fact that chronic exposure to mental stressors related to professional sport practice attenuates mental effects of prolonged cognitive tasks. Thus, longer cognitive tasks are likely required to induce mental fatigue in professional adult athletes.

Another caveat is whether sex influences the effect of a prolonged cognitive task on mental fatigue, perceived effort, and exercise performance in professional athletes. A cognitive task performed during static contraction of elbow flexors provokes greater reduction in time to exhaustion (TTE) in nonathletic women as compared with nonathletic men, supporting greater fatigability in women (8). Nonathletic women are also more likely to suffer from mental disorders (e.g., anxiety) than men are (9). In addition, nonathletic women present with greater increase of perceived stress during low- and high-demand cognitive activities (9). As result, it has been supposed that women activate the brain differently from men in face of mental stressors (9,10). In this sense, nonathletic women have been shown to present with greater activation of the insula and cingulate cortex than men during a high-demand short-duration (8 min) cognitive activity (9). In addition, cingulate cortex remained activated in women beyond the task period (9). Insula is a key integrator of both signals from the periphery and signals from other brain sites, translating afferent signals into sensations (11). Anterior cingulate cortex is thought to be involved in interpretation of signals responsible to generate the perception of effort during exercise (12,13). Thus, prolonged engagement in a demanding cognitive task could generate a sex-specific effect on posttask brain function with possible consequences to the formation of mental fatigue state, perceived effort, and consequently exercise performance. However, to the best of our knowledge, no study has specifically investigated effects of prolonged cognitive tasks in women, independently of being an athlete or not.

Given this background, the aim of our study was to investigate, in professional endurance runners, the effect of prolonged engagement in a demanding cognitive task on psychobiological measurements that characterize the mental fatigue state, as well as on cardiorespiratory, metabolic, and perceptual responses to high-intensity constant-velocity treadmill exercise performed to exhaustion. We hypothesized that women, as compared with men, would show a greater psychobiologic state of mental fatigue after execution of a prolonged and demanding cognitive task. Cardiorespiratory and metabolic responses to exercise would not be affected by the cognitive task in both sexes. On the other hand, the cognitive task would increase perceived effort during exercise in both sexes, such that the maximal effort would be reached earlier, generating a premature disengagement from exercise. In addition, the effect of the cognitive task on perceived effort and exercise tolerance would be greater in women than men.



Thirty-five (17 women and 18 men) Brazilian professional mid- and long-distance endurance runners (800 m to marathon) volunteered to participate in the study. Athletes were recruited in three running teams located in São Paulo. Three volunteers did not complete the study protocol. Moreover, data of a volunteer were considered as outlier for the main study’s outcome (see the Statistical Analysis section for outlier criterion), and so all the data from this subject were excluded from statistical analyses. Thus, final analyses included data of 15 women and 16 men (Table 1). Subjects were informed about risks and discomforts related to the experimental procedures. All subjects provided written informed consent before participating in the study. The study conformed to the Declaration of Helsinki and was approved by the Ethics Committee of the Federal University of São Paulo (no. 0645/2017).

TABLE 1 - Subjects’ characteristics.
Women Men P ES
 Weight, kg 52.4 ± 1.0 63.2 ± 1.3 <0.001 2.36
 Height, cm 163.8 ± 1.2 172.7 ± 1.5 <0.001 1.64
 Age, yr 25 ± 1 25 ± 1 0.903 0.04
Athletic performance
 IAAF score, points 867 ± 36 846 ± 49 0.736 0.13
 Weekly training, sessions 7 ± 0 8 ± 0 0.300 0.50
 Distance/training session, km 12 ± 1 12 ± 1 0.951 0.00
 Weekly training distance, km 88 ± 7 96 ± 9 0.488 −0.24
Aerobic fitness
 Running economy
  O2 cost of running, mL·kg−1·km−1 229 ± 4 215 ± 4 0.017 0.90
 Ventilatory threshold
  V˙O2, mL·kg−1·min−1 48.56 ± 0.84 53.71 ± 1.03 0.001 −1.37
  V˙O2, % peak 80 ± 1 74 ± 1 0.005 1.08
  HR, bpm 163 ± 2 161 ± 3 0.642 0.17
  HR, % peak 88 ± 1 84 ± 2 0.049 0.85
  Velocity, km·h−1 13.5 ± 0.3 15.3 ± 0.3 <0.001 −1.82
  Velocity, % peak 72 ± 1 70 ± 1 0.187 0.44
  Perception of effort, AU 11.3 ± 0.5 11.6 ± 0.3 0.554 −1.00
 Respiratory compensation point
  V˙O2, mL·kg−1·min−1 56.09 ± 1.24 64.15 ± 1.05 <0.001 −1.79
  V˙O2, % peak 92 ± 1 88 ± 1 <0.001 1.33
  HR, bpm 176 ± 2 178 ± 2 0.541 −0.22
  HR, % peak 95 ± 0 93 ± 1 0.096 0.63
  Velocity, km·h−1 16.1 ± 0.2 18.5 ± 0.3 <0.001 −2.52
  Velocity, % peak 86 ± 1 85 ± 1 0.376 0.45
  Perception of effort, AU 14.5 ± 0.4 14.9 ± 0.3 0.388 −0.64
 Peak values
  V˙O2, mL·kg−1·min−1 61.29 ± 1.45 73.24 ± 1.37 <0.001 −2.16
  HR, bpm 186 ± 2 191 ± 2 0.050 −0.68
  PV, km·h−1 18.8 ± 0.3 21.9 ± 0.3 <0.001 −2.68
  Perception of effort, AU 19.7 ± 0.1 19.9 ± 0.1 0.177 0.00
   V˙E, L·min−1 112.0 ± 2.4 160.0 ± 4.2 <0.001 −3.53
Data are mean ± SE (women, n = 15; men, n = 16). Athletes’ performance was analyzed according to the 2017 Scoring Tables of Athletics of the IAAF.
ES, effect size calculated as Cohen’s d coefficient; P, P value for independent Student’s t-test.

Study Design

The study was single-blinded, randomized, and crossed over. Each subject visited the laboratory on three different occasions (Fig. 1). The interval between visits was at least 24 h and at most 48 h. All visits for a given subject occurred at the same time of day. In the first visit, we obtained subjects’ personal information. Subjects reported their time in their last official competition (range, 1–24 wk) and their training characteristics in the last 2 wk before the study. Athletes’ performance was analyzed according to the 2017 Scoring Tables of Athletics of the International Association of Athletics Federations (IAAF). Women were asked about the date of their last menstruation. Moreover, subjects were familiarized with and instructed about all procedures and measurements that were conducted in the study. Afterward, an incremental exercise test until voluntary exhaustion was performed to determine subjects’ aerobic fitness. Thirty minutes after the incremental exercise test, subjects were familiarized with the high-intensity constant-velocity treadmill exercise test that was performed in the subsequent experimental visits. In the familiarization visit, this test was not conducted until voluntary exhaustion.

Experimental design. VAS was used to assess the perception of mental fatigue before experimental manipulations, whereas it was used to assess the perceptions of mental fatigue, mental effort, and motivation after experimental manipulations.

During the second and third visits, subjects rested seated for 10 min after arriving at the laboratory. Then, perceptual measures were obtained and a familiarization with the cognitive performance test was performed. Subsequently, one of the two experimental manipulations was administered to the subjects for 45 min in random order. Perceptual measures were obtained again after experimental manipulations administration. Then, subjects were equipped in order to measure cardiorespiratory and metabolic responses to a high-intensity constant-velocity treadmill exercise test carried out until voluntary exhaustion (i.e., TTE test).

All athletes’ trainers collaborated with the study, referring a subject to the laboratory when the subject was at a week of low training load. Subjects received written instructions before the experiments. The instructions were not to perform intense endurance exercise for 24 h and strength exercise for 72 h; not to consume caffeine, alcohol, or supplements for 24 h; to attend the exercise tests with the same running shoes; and to ingest a light meal 1 h before the study visit. Women were at different phases of their menstrual cycle during participation in the study. Eight women were at the follicular phase (up to 14 d after the onset of menstruation), and six women were at the luteal phase (between 14 and 28 d after the onset of menstruation). Experimental visits were performed always at the same menstrual phase in each woman. The real study purpose was not informed to the subjects. Similarly to a previous investigation (6), we informed that the study purpose was to know the effects of two different computer activities on exercise performance, and cardiorespiratory, metabolic, and perceptual responses to high-intensity constant-velocity treadmill exercise. At the end of the last visit, the real purpose of the study was presented to the subject, and we asked her or him not to tell the study purpose to the other participants.

Incremental Exercise Test

The initial treadmill (Master ATL; Inbramed, Porto Alegre, Brazil) velocities were 10 km·h−1 for women and 12 km·h−1 for men. The initial velocity was maintained for 3 min. Thereafter, velocity increased 1 km·h−1·min−1 until voluntary exhaustion, followed by a cool-down of 3 min at 5 km·h−1. The treadmill grade was maintained at 1% throughout the test (14). Pulmonary gas exchange and heart rate (HR) were measured breath-by-breath (Quark CPET; Cosmed, Rome, Italy) and beat-to-beat (s810i; Polar, Kempele, Finland), respectively. Then, both signals were filtered and averaged every 20 s. Perception of effort was registered at the final 10 s of each minute using the 15-point Borg’s scale. All subjects attained at least three criteria to verify the maximal nature of the test (i.e., respiratory exchange ratio ≥1.1, perceived effort ≥18, and peak HR (HRpeak) ≥90% of age-predicted) (15), whenever an oxygen uptake (V˙O2) plateau did not occur. Running economy was considered as the oxygen cost of running at the last minute of the initial velocity. Peak oxygen uptake (V˙O2peak), peak ventilation (V˙Epeak), and HRpeak were considered the highest values obtained during the test (16). The velocity corresponding to the last completed stage was considered as the peak velocity (PV) (16). Ventilatory threshold and respiratory compensation point were calculated (Quark PFT Ergo 10.0e; Cosmed), respectively, by piecewise linear regression of carbon dioxide output as function of V˙O2 and E as a function of carbon dioxide output (17). The velocity corresponding to these physiological landmarks was obtained taking into account the lag (~40 s) in V̇O2 response to the increase in velocity during incremental exercise testing (18).

Perceptual Measures

Visual analog scale

A 10-cm visual analog scale (VAS) was used to measure subjects’ perception of mental fatigue, mental effort, and motivation. VAS is a practical and valid method that has been used to assess perceptions like those investigated in the present study (3,19,20). Perception of mental fatigue was assessed before and after experimental manipulations, whereas perceptions of mental effort and motivation were assessed only after. Subjects were asked to indicate the perceived level of these feelings placing a mark on the VAS 10-cm line. The left side of the scales was anchored with the descriptor “not at all,” and the right side with the descriptor “maximal.” Mental fatigue, mental effort, and motivation were assessed, respectively, by asking feelings of mental tiredness and lack of energy, mental difficulty to perform the experimental manipulations, and willingness to perform the TTE test.

Brunel mood scale

Brunel Mood Scale (BRUMS) was used to quantify mood profile before and after experimental manipulations. BRUMS was developed by Terry et al. (21) and validated to Portuguese by Rohlfs et al. (22). BRUMS contains 24 items. Subjects were asked to report in a 5-point Likert scale (0, not at all; 1, a little; 2, moderately; 3, quite a bit; and 4, extremely) how they felt in relation to each of the scale’s items. Items are divided into six subscales (anger, confusion, depression, fatigue, tension, and vigor) with four items each, such that each subscale results in a raw score from 0 to 16. Only scores of fatigue and vigor were used in the present study. Previous studies have shown prolonged cognitive tasks increase BRUMS score of fatigue and decrease BRUMS score of vigor (6,23), indicating that BRUMS is sensitive to the effects of mental stress. The same researcher (T.R.L.) assessed both the VAS and BRUMS.

Experimental Manipulations

Stroop task

The Stroop task was used as experimental manipulation to induce mental fatigue (PsychoPy v1.85.6; University of Nottingham, Nottingham, United Kingdom). Previous studies have shown that execution of the Stroop task for 30 to 45 min induces mental fatigue in physically active subjects (3,20). Of note, the longer the task, the greater is the mental fatigue (2,4). A previous study reported that execution of the Stroop task for 30 min did not change indices of mental fatigue in professional adult athletes (1). Thus, herein, we used the Stroop task for 45 min in an attempt to increase the stress stimulus to professional runners. In this task, four words (blue, yellow, red, and green) were presented in Arial font 57, once at a time, in random order, at the center of a computer screen (15 inches). The words were inked with the color blue, yellow, red, or green on a congruent (e.g., word “blue” painted with blue ink) or incongruent manner (e.g., word “blue” painted with red ink). Subjects were instructed to press as quickly and accurately as possible a colored button on the computer keyboard corresponding to the correct response. The word’s ink determined the correct response. If the ink was blue, green, or yellow, subjects should press the button corresponding to the ink color (e.g., if the word “green” appeared inked in yellow, the button yellow should be pressed). If, however, the ink color was red, the button that should be pressed was the button corresponding to word’s meaning, not the ink color (e.g., if the word “blue” appeared inked in red, the button blue should be pressed). Each word stayed on the screen until subjects pressed any answer button. A blank screen with a white cross at the center was displayed for 1000 ms between each word presentation.

Control task

The control task consisted of watching a documentary at the same computer screen used to perform the Stroop task. Subjects had two documentary options to choose (National Geographic, Washington, DC). One of them was “Mankind from Space” and another was “Big, Bigger, Biggest—Series 2, Episode 1: Aircraft.” A researcher (T.R.L. or P.B.S.) stayed all the time in the experimental room, next to the computer, during both the Stroop test and the documentary watching, to maintain subjects’ engagement with the tasks. The researcher only called subjects’ attention in case subjects were feeling asleep, which just occurred during the control experimental manipulation.

Cognitive Performance

Cognitive performance was assessed with a shorter version of the same Stroop task that was used to induce mental fatigue. This short version had 160 trials and lasted 5 to 6 min. Each trial corresponded to one word displayed on the computer screen. Trials were divided into 120 incongruent (word and color not matched) and 40 congruent (word and color matched) trials. Trials were presented in random order. Subjects were instructed to press, as quickly and accurately as possible, the colored button on the computer keyboard corresponding to the correct answer. Cognitive performance was assessed by reaction time, lapses of attention, and accuracy (PsychoPy v1.85.6; University of Nottingham). Reaction time corresponded to the time lag to provide an answer. Lapses of attention correspond to the number of trials with reaction time longer than 2 s (24). Response accuracy corresponded to the proportion of correct answers. The mental fatigue state has been characterized by reaction time increase, lapses of attention increase, and/or accuracy decrease (2).

TTE Test


The TTE test begun with a warm-up of 3 min at 60% of the PV obtained in the incremental exercise test. Thereafter, treadmill velocity immediately increased to 90% of the PV. Treadmill grade was maintained at 1% throughout the test (14). Subjects were instructed to sustain the exercise as much as they could. Time until voluntary exhaustion was recorded with a manual stopwatch. Time recording was initiated when the treadmill achieved the target velocity, and it was stopped when subjects held the safety bar in front of them. Subjects were verbally encouraged throughout the test by a researcher blinded about the experimental manipulation. Only two researchers stayed in the experimental room during the TTE test. The same researchers conducted the TTEs for a given subject, to avoid an effect that different people in the experimental room could exert on subjects’ judgment about their perceived effort (25). The researcher T.R.L. was present in all of the TTEs, and he was accompanied by D.M.O 93% of the TTEs. Previous studies reported that TTE tests are reproducible in trained subjects (coefficient of variation from 6.8% to 28.8%) and are sensitive to mental stressors (2,4,26).

Cardiorespiratory, metabolic, and perceived effort responses

V˙O2 and ventilation (E) were measured breath-by-breath (Quark CPET; Cosmed). Aberrant values were filtered (2 SDs of a 15-breath moving average), and then reduced to 1-min averages in a customized Excel spreadsheet (Microsoft, Redmond, WA). Before each test, O2 and CO2 analyzers were calibrated according to the manufacturer’s specifications using ambient air and gases with a known concentration (16% O2 and 5% CO2). The flowmeter was calibrated using a 3-L syringe. HR was measured beat-to-beat (s810i; Polar). R-R intervals were extracted, filtered and corrected for artifacts (Kubios HRV Standard 3.2.0; Kubios, Kuopio, Finland), and then reduced to 1-min averages in a customized Excel spreadsheet (Microsoft). If subjects could not sustain a complete minute near to exhaustion in the TTE test, the average of the data corresponding to the time sustained was calculated. Because of technical problems, V˙O2 was not obtained in two women and one man, E in one woman, and HR in two men. The perception of effort was registered at the last 10 s of each exercise minute or immediately after exhaustion, if subjects did not complete a full minute. V˙O2, E, HR, and perception of effort data were analyzed as 25%, 50%, 75%, and 100% of isotime. The 100% isotime was considered the last complete minute of the shorter TTE test (Stroop task or documentary) for given subject. To obtain the values of 25%, 50%, and 75% of isotime, V˙O2, E, HR, and perception of effort data were linearly interpolated from 0 (last minute of warm-up) to 100% of isotime at every 5% within each condition for a given subject (Origin 2020; OriginLab, Northampton, MA). A capillary blood sample (25 μL) was collected from an earlobe in a heparinized and calibrated capillary before and 3 min after the TTE test. Each blood sample was stored in an Eppendorf containing 50 μL of 1% NaF (i.e., anticoagulant) and frozen at −20°C until analysis of blood lactate concentration ([BL]; YSI 1500 SPORT; Yellow Springs Instruments, Yellow Springs, OH).

Perception of effort was measured with the 15-point Borg’s scale. The scale was displayed in a large panel in front of the treadmill. Standardized written instructions were presented and orally explained to the subjects according to previous recommendations (27). Briefly, subjects were instructed to rate how hard was the exercise with regard to command their legs and to breathe (27). This question was also displayed in a large panel above the 15-point Borg’s scale. Subjects were oriented to read the verbal expressions and then report the number corresponding to their feeling at a given moment. In addition, they were informed that there was no right or wrong response, and were oriented not to underestimate or overestimate their feelings of exertion. A memory-based approach was used before the incremental exercise test to anchor subjects’ perception with the scale’s verbal expressions (i.e., some points in the scale were anchored with the perception of effort derived from their training paces). On the other hand, before the TTE test, the reference to anchor subjects’ perception was based on the incremental exercise test (i.e., maximal effort corresponded to the perception experienced at exhaustion). It is important to say that the aforementioned perception of effort definition and anchoring has been shown to be sensitive to physiological and psychological experimental manipulations (27). The same researcher (T. R. L.) assessed perceived effort.

Statistical Analysis

The sample size calculation took into account the interaction effect between sex and experimental manipulation (i.e., condition) on the time to complete the TTE test in preliminary data (seven women and six men). Thirty subjects would be necessary to find a P value lower than 0.05 with 0.80 of power (G-Power; Dusseldorf University, Dusseldorf, Germany), considering a partial eta squared (η2p) of 0.07 (effect size f = 0.27), correlation among repeat measures of 0.50, and nonsphericity correction of 1.0. However, taking into account a possible sample loss of ~15%, we aimed to recruit 35 subjects.

Results are presented as mean ± SE. Shapiro–Wilk’s test was used to ascertain data distribution. Independent Student’s t-test was used to compare the groups concerning subjects’ characteristics, absolute and relative changes in TTE test performance, and order of the experimental manipulations on change in TTE test performance. Two-way repeated-measures ANOVA was used to investigate the effect of sex and condition on the perception of mental effort, motivation, cognitive performance, TTE test performance, and data at exhaustion of the TTE test. Three-way repeated-measures ANOVA was used to investigate the effect of sex, time and condition on the perception of mental fatigue, and scores of fatigue and vigor in the BRUMS, as well as on cardiorespiratory, metabolic, and perceived effort responses at isotimes during the TTE test. Homogeneity of variance assumption was ascertained with Levene’s test, followed by Welch’s correction when necessary. Greenhouse–Geisser’s correction was used to adjust ANOVA results, whenever sphericity was violated in the Mauchly’s test. Fisher’s post hoc was used when significant F values were found. Effect sizes for Student’s t-test and ANOVA results were calculated as Cohen’s d coefficient and η2p, respectively (28). The interquartile range (IQR) method was used to determine outliers in the absolute change of the TTE test time (29). This method consisted of calculating lower (Q1) and upper (Q3) quartiles, and IQR (Q3 minus Q1) of absolute changes (i.e., experimental data minus control data). Outliers were considered as values of absolute change lower or upper than 1.5 times the IQR. Statistical significance was set at P < 0.05. Statistical analyses were conducted in the software Statistica 12 (Statsoft, EUA).


Subjects’ Characteristics

There was no difference between sexes in the IAAF score, number of weekly training sessions, distance covered per training session, and total distance of weekly training (Table 1). Running economy was worse in women than men. V˙O2 at ventilatory threshold and respiratory compensation point were lower in women, but %V˙O2peak at ventilatory threshold and respiratory compensation point were higher in women as compared with men. There were no differences between sexes in HR and perception of effort at ventilatory threshold and respiratory compensation point. The %HRpeak at ventilatory threshold was higher in women as compared with men, but it was not different between sexes at respiratory compensation point. The velocity at ventilatory threshold and respiratory compensation point was lower in women, but there was no difference between sexes in the %PV at ventilatory threshold and respiratory compensation point. V˙O2peak, PV, and Epeak were lower in women, whereas HRpeak and peak perceived effort were similar between sexes.

Effect of experimental manipulations on perceptual measures

Perception of mental effort was higher in the Stroop task (women, 6.0 ± 0.7 cm; men, 6.2 ± 0.6 cm) than control (women, 4.1 ± 0.8 cm; men, 3.6 ± 0.7 cm), independently of sex (sex effect: P = 0.911, η2p = 0.000; condition effect: P < 0.001, η2p = 0.453; interaction: P = 0.428, η2p = 0.022). Perception of mental fatigue and score of fatigue increased from pre to post in both experimental manipulations, without differences between women and men (Table 2). Noteworthy, perception of mental fatigue at post-Stroop was greater than at post-documentary. Score of vigor decreased from pre to post in both experimental manipulations, without difference between women and men. In addition, women reported lower score of vigor than in men, independently of the moment and manipulation. Motivation to perform the TTE was similar after Stroop task and control. Nevertheless, motivation to perform the TTE was lower in women than men, independently of the manipulation (Stroop task: 3.8 ± 0.6 cm for women vs 4.2 ± 0.7 cm for control; Stroop task: 6.4 ± 0.5 cm for men vs 6.2 ± 0.6 cm for control; sex effect: P = 0.003, η2p = 0.263; condition effect: P = 0.779, η2p = 0.003; interaction: P = 0.475, η2p = 0.018).

TABLE 2 - Effect of experimental manipulations on perceptual measures and cognitive performance.
Perceptual Measures Cognitive Performance
Mental Fatigue, cm Fatigue Score, AU Vigor Score, AU Reaction Time, s Lapses of Attention Accuracy, %
Women, control
 Pre 2.3 ± 0.4 3.7 ± 0.7 7.5 ± 0.9
 Post 5.3 ± 0.6 5.7 ± 0.9 5.6 ± 0.7 0.995 ± 0.051 3 ± 1 89 ± 4
Women, Stroop task
 Pre 2.5 ± 0.4 3.5 ± 0.6 7.0 ± 0.6
 Post 6.8 ± 0.6* 7.7 ± 1.0* 4.7 ± 0.8 0.998 ± 0.054 4 ± 1 88 ± 4
Men, control
 Pre 1.6 ± 0.4 2.7 ± 0.6 9.9 ± 0.6
 Post 4.3 ± 0.6 4.4 ± 0.8 8.0 ± 0.6 0.901 ± 0.050 2 ± 1 94 ± 4
Men, Stroop task
 Pre 1.7 ± 0.4 2.3 ± 0.6 10.3 ± 0.6
 Post 5.8 ± 0.6* 5.9 ± 0.9* 7.4 ± 0.7 0.936 ± 0.050 3 ± 1 92 ± 4
ANOVA, P values (η2p)
 Sex (S) 0.059 (0.117) 0.077 (0.104) 0.003 (0.270) 0.253 (0.045) 0.666 (0.007) 0.408 (0.024)
 Condition (C) 0.018 (0.179) 0.119 (0.082) 0.341 (0.031) 0.526 (0.014) 0.060 (0.117) 0.027 (0.158)
 Time (T) <0.001 (0.734) <0.001 (0.515) <0.001 (0.566)
 S × C 0.952 (0.000) 0.766 (0.003) 0.460 (0.019) 0.591 (0.010) 0.848 (0.001) 0.499 (0.016)
 S × T 0.812 (0.002) 0.651 (0.007) 0.714 (0.005)
 C × T 0.012 (0.200) 0.007 (0.223) 0.192 (0.058)
 S × C × T 0.881 (0.001) 0.823 (0.002) 0.549 (0.012)
Data are mean ± SE (women, n = 15; men, n = 16). Capital letters separated by multiplication symbol represent interaction between main effects. Perceptual measures were obtained with VAS (mental fatigue) and BRUMS (scores of fatigue and vigor). Cognitive performance is referent to incongruent responses in the short Stroop task. Lapses of attention are number of trials with reaction time longer than 2 s.
*Higher than control independent of sex (perception of mental fatigue: P < 0.001 in the Fisher’s post hoc, Cohen’s d coefficient = 0.68; score of fatigue: P = 0.002 in the Fisher’s post hoc, Cohen’s d coefficient d = 0.51).

Effect of experimental manipulations on cognitive performance

Reaction time in incongruent responses was similar between sexes and experimental manipulations (Table 2). Independently of sex, lapses of attention tended to be higher and accuracy in incongruent responses was lower after the Stroop task as compared with the control manipulation. On the other hand, sex and experimental manipulation did not influence performance of congruent responses (all, P > 0.05).

Effect of experimental manipulations on cardiorespiratory, metabolic, and perceived effort responses to the TTE test

Women had lower V˙O2 and E than did men during the TTE test and at exhaustion (Figs. 2A, B). V˙O2, E, and HR were not affected by the Stroop task along the TTE and at exhaustion (Figs. 2A–C). Women had lower [BL] as compared with men (Fig. 2D). [BL] was higher post-TTE test in the Stroop task as compared with control only in men. The Stroop task, as compared with control, increased perceived effort at 100% (Fig. 3A) of isotime in women and at 25% of isotime in men (Fig. 3B). Of note, maximal perceived effort was attained earlier in both sexes (Figs. 3A, B).

Effect of experimental manipulations on cardiorespiratory responses at isotimes and exhaustion in the TTE test, as well as [BL] pretest and 3 min posttest. A hundred percent of isotime is the last complete minute of the shorter TTE test for a given subject (Stroop task or control). A, V˙O2 (women, n = 13; men, n = 15). †Lower in women than men at exhaustion (sex effect: P < 0.001, η2p = 0.481; condition effect: P = 0.860, η2p = 0.001; interaction: P = 0.355, η2p = 0.033). B, E (women, n = 14; men, n = 16). †Lower in women than men at exhaustion (sex effect: P < 0.001, η2p = 0.755; condition effect: P = 0.072, η2p = 0.111; interaction: P = 0.714, η2p = 0.005). C, HR (women, n = 15; men, n = 14). There was no sex (P = 0.369, η2p = 0.030), condition effect (P = 0.272, η2p = 0.044), or interaction (P = 0.590, η2p = 0.011) effects at exhaustion on HR. D, [BL] (women, n = 15; men, n = 16). *Higher than control (P = 0.021, Cohen’s d coefficient = 0.39). Data are presented as mean and SE. Three-way (at isotime) and two-way (at exhaustion) repeated-measures ANOVA, followed by Fisher’s post hoc when necessary, was used to analyze the data.
Effect of experimental manipulations on perception of effort at isotimes and exhaustion in the TTE test. A hundred percent of isotime is the last complete minute of the shorter TTE test for a given subject (Stroop task or control). A, Perception of effort at isotimes and at exhaustion in women (n = 15). *Higher than control (P < 0.001, Cohen’s d coefficient = 0.74). B, Perception of effort at isotimes and exhaustion in men (n = 16). *Higher than control (P = 0.022, Cohen’s d coefficient = 0.22). At exhaustion, there were no sex (P = 0.203, η2p = 0.055), condition (P = 0.564, η2p = 0.012), or interaction (P = 0.564, η2p = 0.012) effects on perception of effort. Data are presented as mean and SE. Three-way (at isotime) and two-way (at exhaustion) repeated-measures ANOVA, followed by Fisher’s post hoc when necessary, was used to analyze the data. #Shorter in Stroop task (P = 0.043 and η2p = 0.133 in the two-way repeated-measures ANOVA).

Effect of experimental manipulations on TTE test performance

The Stroop task shortened TTE versus control similarly in women and men (−6% ± 6% and −3% ± 5% in women and men, respectively; P = 0.732; Cohen’s d coefficient = 0.124; Figs. 4A, B). Order of experimental manipulations did not alter the effect of Stroop task on TTE test performance (P = 0.938; Cohen’s d coefficient = 0.028). Performance impairment was similar between subjects who started the first session with the control manipulation (−28 ± 86 s) and subjects who started with the Stroop task (−26 ± 58 s).

Effect of experimental manipulations on TTE test performance. Data are presented as mean and SE (women, n = 15; men, n = 16). A, Duration of TTE test in women (triangle) and men (circle) at control (filled symbol) and Stroop task (empty symbol). Two-way repeated-measures ANOVA, followed by Fisher’s post hoc when necessary, was used to analyze these data. B, Absolute change on TTE duration after Stroop task as compared with control. Independent Student’s t-test was used to compare results between women and men.


Our results indicate that prolonged engagement in the Stroop task generated a psychobiological state of mental fatigue in professional endurance runners, but this was not greater in women in comparison to men as we had hypothesized. The mental fatigue generated by the Stroop task did not alter exercise-induced cardiorespiratory responses in both sexes. In addition, attainment of maximal effort was anticipated in both sexes and exercise tolerance was shortened.

Effect of experimental manipulations on perceptual measures and cognitive performance

It has been suggested that perceptual measures and cognitive performance should be interpreted in an integrated manner to ensure that induction of mental fatigue was successfully achieved (4). In the present study, the Stroop task and the documentary similarly reduced vigor perception, which is similar to previous studies (6,19,23). Fatigue perception increased after both manipulations. Other studies have also found that vigilance tasks with low information load may induce some level of mental stress (3,30). Nevertheless, the increase of fatigue perception in the present study was greater after the Stroop task than watching the documentary. In addition, more lapses of attention and lower accuracy in incongruent responses were found in the cognitive test carried out after the Stroop task as compared with the control manipulation, indicating worse cognitive performance after the Stroop task. Altogether, these findings indicate that a state of mental fatigue was induced after the prolonged Stroop task and that Stroop effects on indicators of mental fatigue were greater than documentary effects in the professional endurance runners. Of note, experimental manipulations did not change the motivation to perform the TTE test. This finding strengthens an interpretation presented later that the reduction in the TTE test performance was possibly attributed to an anticipated attainment of maximal perceived effort rather than lack of motivation.

Although a mental fatigue state was induced in endurance athletes in the present study, men and women presented similar decrease of vigor, increase of fatigue perception, increase of mental effort, and decrease of cognitive performance with the Stroop task. Therefore, our data indicate that the Stroop task–induced mental fatigue was similar in men and women, which is contrary to our hypothesis. We defined our hypothesis taking into consideration available evidence that women activate the anterior cingulate cortex more than men (9). However, these results were found during a cognitive task of only 8 min (9). Moreover, the cognitive task used to investigate patterns of brain activation in women and men involved performing math operations and providing verbal responses (9). The executive functions needed to perform math operations with verbal responses might be, to some extent, different from performing a congruent and incongruent color–word task without verbal responses (10,31). Therefore, brain activation might have been somewhat different in our study in comparison with the aforementioned one. Another possibility is that brain activation was different between sexes early in the task and became similar afterward. In addition, activation of areas such as insula and posterior cingulate cortex, which are known to be involved in mental fatigue (32), could have played a major role in the present study.

Another explanation for the lack of effect of sex on indicators of mental fatigue may be related to the similarity in the competitive level and training characteristics of women and men. Competitions and training are tasks that involve inhibitory control of aversive sensations such as pain, dyspnea, and thermal discomfort, as well as negative thoughts (1). Inhibitory control is an important self-regulating behavior to ensure that the goal of winning/finishing a competition or running a planned training session is met (1,33). For instance, Cona et al. (34) have investigated the cognitive performance of 30 runners before an ultramarathon. The authors found greater inhibitory control in the athletes with better running performance. Athletes’ lifestyle requires constant self-regulatory behaviors, such as, dietary monitoring, avoidance of alcohol intake, and smoking, and strictly follow training programs (1). Such self-regulatory behaviors could strengthen inhibitory control (1). Because women and men participating in the present study had similar characteristics of athletic performance and training volume, it could be speculated that men and women had similar mental fatigue owing to a similar level of self-regulation in daily training and competitions.

Effect of experimental manipulations on cardiorespiratory, metabolic, and perceived effort responses to the TTE test

Mental fatigue generated by the Stroop task did not alter V˙O2, V˙E, and HR, responses throughout the TTE test as compared with the control condition. Previous studies have also showed no effect of mental fatigue on cardiorespiratory and metabolic responses during TTE tests (4,6,20,23). Another study reported lower HR and [BL] at exhaustion after mental fatigue induction (6). Lower cardiorespiratory and metabolic values at exhaustion could be explained by shorter exercise duration after mental fatigue induction. Although cardiorespiratory responses were unchanged during the TTE test, perception of effort during the TTE test was higher after the Stroop task than after the control condition. Indeed, a well-established effect in the literature is that mental fatigue generated by prolonged involvement with demanding cognitive activity increases perceived effort during exercise (1,4,6,7,20,23). Of note, however, this result had not been found by Martin et al. (1) in professional endurance athletes. Unlike the study by Martin et al. (1) that used 30 min, we used 45 min of cognitive effort and successfully altered the perceived effort during exercise. Thus, our results suggest that longer engagement with a cognitive activity may be required to change the perception of effort during exercise in professional endurance athletes.

The perception of effort formation during exercise involves processing sensory signals by the brain (4,27). Thus, any change in perceived effort could be determined by changes either in sensory signals or in central processing. Some authors argue that the sensory signals underling the perceived effort formation come from afferent receptors located in skeletal muscles and cardiorespiratory system (4,27), whereas other authors believe that perceived effort formation relies on corollary discharges associated with central motor command that originate in the premotor, motor, and supplementary motor areas (4,27,35). All of these sensory signals are also involved on cardiorespiratory responses to exercise (36). Given that cardiorespiratory responses to exercise were unchanged by mental fatigue, we speculate that the change in perceived effort was mediated by a change in the central processing rather than in afferent signaling.

Effect of experimental manipulations on TTE test performance

In the present study, mental fatigue worsened endurance performance similarly in women and men. Traditionally, TTE test performance has been considered to be determined by an incapacity of the fatigued neuromuscular system to produce the speed/power required by the task, despite a maximal voluntary effort (37). However, in 2010, a seminal work showed that the maximal power of the locomotor muscle at the exhaustion moment during the TTE test is significantly higher than the power required by the task (38). This finding indicated that neuromuscular fatigue cannot per se be considered a key determinant of the TTE test performance (38), which supported an alternative explanation that was named psychobiological model for endurance exercise performance (6,38,39). According to the psychobiological model (6,38,39), whose theoretical background comes from the motivational intensity theory (40), exhaustion during a TTE test can be explained as a conscious decision to stop exercising based on the perception of effort and motivation (i.e., maximal effort that a person is willing to exert in order to satisfy a motive). Thus, exercise is interrupted when the perceived effort equals or exceeds the maximal effort that a person wishes to exert. In the present study, there was no difference between interventions in the motivation to perform the TTE test. However, perceived effort during exercise was greater when subjects were mentally fatigued. Thus, it is likely that earlier achievement of maximal effort in women and men provoked earlier disengagement to perform the TTE test shortening exercise duration.


A prolonged cognitive task provoked mental fatigue, anticipated attainment of maximal perceived effort, and worsened aerobic performance in professional runners with no sex differences. Although we did not contrast athletes with nonathletes, our results suggest that being an athlete may somehow prevent women from developing greater mental fatigue and suffering more from its underlying effects compared with men.

The authors are grateful to the athletes who participated in the study and to the coaches Claudio Castilho, Luiz Cândido, and Clodoaldo Lopes from Esporte Cluble Pinheiros, São Paulo Futebol Clube, and Centro Olímpico de Treinamento e Pesquisa, who kindly referred athletes to participate in the study. The study was funded by the Coordination for the Improvement of Higher Education Personnel. Diogo Machado de Oliveira received a scholarship from Coordination for the Improvement of Higher Education Personnel (Award No. 88882430773/201901).

The authors have no conflict of interest to declare. The results of the present study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation and do not constitute endorsement by the American College of Sports Medicine.


1. Martin K, Staiano W, Menaspa P, et al. Superior inhibitory control and resistance to mental fatigue in professional road cyclists. PLoS One. 2016;11(7):e0159907.
2. Pageaux B, Lepers R. The effects of mental fatigue on sport-related performance. Prog Brain Res. 2018;240:291–315.
3. Smith MR, Chai R, Nguyen HT, Marcora SM, Coutts AJ. Comparing the effects of three cognitive tasks on indicators of mental fatigue. J Psychol. 2019;12(8):1–25.
4. Van Cutsem J, Marcora S, De Pauw K, Bailey S, Meeusen R, Roelands B. The effects of mental fatigue on physical performance: a systematic review. Sports Med. 2017;47(8):1569–88.
5. Pageaux B, Lepers R. Fatigue induced by physical and mental exertion increases perception of effort and impairs subsequent endurance performance. Front Physiol. 2016;7:587.
6. Marcora SM, Staiano W, Manning V. Mental fatigue impairs physical performance in humans. J Appl Physiol. 2009;106(3):857–64.
7. Filipas L, Gallo G, Pollastri L, La Torre A. Mental fatigue impairs time trial performance in sub-elite under 23 cyclists. PLoS One. 2019;14(6):e0218405.
8. Yoon T, Keller ML, De-Lap BS, Harkins A, Lepers R, Hunter SK. Sex differences in response to cognitive stress during a fatiguing contraction. J Appl Physiol (1985). 2009;107(5):1486–96.
9. Wang J, Korczykowski M, Rao H, et al. Gender difference in neural response to psychological stress. Soc Cogn Affect Neurosci. 2007;2(3):227–39.
10. Bell EC, Willson MC, Wilman AH, Dave S, Silverstone PH. Males and females differ in brain activation during cognitive tasks. Neuroimage. 2006;30(2):529–38.
11. Craig AD. How do you feel? Interoception: the sense of the physiological condition of the body. Nat Rev Neurosci. 2002;3(8):655–66.
12. Williamson JW, McColl R, Mathews D, Mitchell JH, Raven PB, Morgan WP. Hypnotic manipulation of effort sense during dynamic exercise: cardiovascular responses and brain activation. J Appl Physiol. 2001;90(4):1392–9.
13. Fontes EB, Okano AH, De Guio F, et al. Brain activity and perceived exertion during cycling exercise: an fMRI study. Br J Sports Med. 2015;49(8):556–60.
14. Jones AM, Doust JH. A 1% treadmill grade most accurately reflects the energetic cost of outdoor running. J Sports Sci. 1996;14(4):321–7.
15. Howley ET, Bassett DR Jr, Welch HG. Criteria for maximal oxygen uptake: review and commentary. Med Sci Sports Exerc. 1995;27(9):1292–301.
16. Nakamoto FP, Ivamoto RK, Andrade MDS, de Lira CAB, Silva BM, da Silva AC. Effect of intermittent hypoxic training followed by intermittent hypoxic exposure on aerobic capacity of long distance runners. J Strength Cond Res. 2016;30(6):1708–20.
17. Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol. 1986;60(6):2020–7.
18. Whipp BJ, Davis JA, Torres F, Wasserman K. A test to determine parameters of aerobic function during exercise. J Appl Physiol. 1981;50(1):217–21.
19. Smith MR, Coutts AJ, Merlini M, Deprez D, Lenoir M, Marcora SM. Mental fatigue impairs soccer-specific physical and technical performance. Med Sci Sports Exerc. 2016;48(2):267–76.
20. Pageaux B, Marcora SM, Rozand V, Lepers R. Mental fatigue induced by prolonged self-regulation does not exacerbate central fatigue during subsequent whole-body endurance exercise. Front Hum Neurosci. 2015;9(67).
21. Terry PC, Lane AM, Fogarty GJ. Construct validity of the Profile of Mood States—Adolescents for use with adults. Psychol Sport Exerc. 2003;4(2):125–39.
22. Rohlfs ICPM, Rotta TM, Luft CDB, Andrade A, Krebs RJ, Carvalho T. Brunel Mood Scale (BRUMS): an instrument for early detection of overtraining syndrome. Braz J Sport Med. 2008;14(3):176–81.
23. Pageaux B, Marcora SM, Lepers R. Prolonged mental exertion does not alter neuromuscular function of the knee extensors. Med Sci Sports Exerc. 2013;45(12):2254–64.
24. Wang C, Trongnetrpunya A, Samuel IB, Ding M, Kluger BM. Compensatory neural activity in response to cognitive fatigue. J Neurosci. 2016;36(14):3919–24.
25. Robertson RJ, Noble BJ. Perception of physical exertion: methods, mediators, and applications. Exerc Sport Sci Rev. 1997;25(1):407–52.
26. Laursen PB, Francis GT, Abbiss CR, Newton MJ, Nosaka K. Reliability of time-to-exhaustion versus time-trial running tests in runners. Med Sci Sports Exerc. 2007;39(8):1374–9.
27. Pageaux B. Perception of effort in exercise science: definition, measurement and perspectives. Eur J Sport Sci. 2016;16(8):885–94.
28. Cohen J. Statistical Power Analysis for the Behavioral Sciences. ed ed. New York: Lawrence Erlbaum Associates; 1988.
29. Hoaglin DC, Iglewicz B, Tukey JW. Performance of some resistant rules for outlier labeling. J Am Stat Assoc. 1986;81(396):991–9.
30. Lorist MM, Faber LG. Consideration of the influence of mental fatigue on controlled and automatic cognitive processes and related neuromodulatory effects. In: Cognitive Fatigue: Multidisciplinary Perspectives on Current Research and Future Applications. Washington, DC: American Psychological Association; 2011. pp. 105–26.
31. Carrillo E, Moya-Albiol L, Gonzalez-Bono E, Salvador A, Ricarte J, Gomez-Amor J. Gender differences in cardiovascular and electrodermal responses to public speaking task: the role of anxiety and mood states. Int J Psychophysiol. 2001;42(3):253–64.
32. Ishii A, Tanaka M, Watanabe Y. Neural mechanisms of mental fatigue. Rev Neurosci. 2014;25(4):469–79.
33. Pageaux B, Lepers R, Dietz KC, Marcora SM. Response inhibition impairs subsequent self-paced endurance performance. Eur J Appl Physiol. 2014;114(5):1095–105.
34. Cona G, Cavazzana A, Paoli A, Marcolin G, Grainer A, Bisiacchi PS. It’s a matter of mind! Cognitive functioning predicts the athletic performance in ultra-marathon runners. PLoS One. 2015;10(7):e0132943.
35. Marcora S. Perception of effort during exercise is independent of afferent feedback from skeletal muscles, heart, and lungs. J Appl Physiol. 2009;106(6):2060–2.
36. Nobrega ACL, O’Leary D, Silva BM, Marongiu E, Piepoli MF, Crisafulli A. Neural regulation of cardiovascular response to exercise: role of central command and peripheral afferents. Biomed Res Int. 2014;2014:478965.
37. Hepple RT. The role of O2 supply in muscle fatigue. Can J Appl Physiol. 2002;27(1):56–69.
38. Marcora SM, Staiano W. The limit to exercise tolerance in humans: mind over muscle? Eur J Appl Physiol. 2010;109(4):763–70.
39. Staiano W, Bosio A, de Morree HM, Rampinini E, Marcora S. The cardinal exercise stopper: muscle fatigue, muscle pain or perception of effort? Prog Brain Res. 2018;240:175–200.
40. Wright RA. Refining the prediction of effort: Brehm’s distinction between potential motivation and motivation intensity. Soc Pers Psychol Compass. 2008;2(2):682–701.


Copyright © 2020 by the American College of Sports Medicine