Journal Logo


Central Regulation and Neuromuscular Fatigue during Exercise of Different Durations


Author Information
Medicine & Science in Sports & Exercise: June 2016 - Volume 48 - Issue 6 - p 1024-1032
doi: 10.1249/MSS.0000000000000867


An important question in exercise physiology is how skeletal muscle recruitment and force production are regulated during self-paced exercise. A recently proposed hypothesis is that exercise performance during time trials (TT) is regulated by sensory feedback to the central nervous system to ensure that an individual’s critical threshold of peripheral fatigue (measured as a reduction in the capacity of the muscle beyond the neuromuscular junction to produce maximal force) is never exceeded (2,3). This hypothesis predicts that high levels of peripheral fatigue regulate exercise performance by limiting the extent of skeletal muscle recruitment through afferent feedback from the exercising muscles to the central nervous system (2,3). However, it is not known when this critical threshold is reached or if it even exists for various exercise durations (24).

Neuromuscular fatigue can be measured as a reduction in maximal voluntary contraction (MVC) force. Both peripheral fatigue and central fatigue can contribute to neuromuscular fatigue. Peripheral fatigue is defined as the reduction in force originating from sites at or distal to the neuromuscular junction, and central fatigue is a reduction in the maximal capacity of the central nervous system to optimally recruit motor units to produce force (12). It is well established that both neuromuscular and peripheral fatigue increase with increasing intensity during exercise bouts of similar duration (4). Although the causes of fatigue after different types of exercise (i.e., task dependency) are now better understood, the time course of fatigue are largely unexplored during self-paced exercise and have been addressed by only a couple of studies. For instance, substantial reductions of MVC levels and evoked peak force response (as an index of peripheral fatigue) occur after approximately 20% of the total duration of a self-paced TT with further increases as the exercise continues (11). In whole-body exercises, central fatigue is believed to occur after peripheral fatigue has already developed (8) and to be dependent on exercise duration (21).

Some attempts have been made to determine the extent to which different aspects of fatigue develop during self-paced exercise of different durations. Recently, Thomas et al. (24) found greater levels of central fatigue after 40-km cycling TT and 20-km cycling TT compared with 4-km cycling TT, but a higher level of peripheral fatigue after the 4-km TT compared with the 20-km TT and 40-km TT (24). Wuthrich et al. (26) found that peripheral fatigue was reduced to a similar extent in TT lasting 15 or 30 min, but to different levels after cycling and running. Importantly, however, the measurements in the study by Wuthrich et al. (26) were performed 10 min after exercise termination, allowing substantial recovery to occur. In fact, even when, as in Thomas et al. (24), fatigue was measured within 2.5 min after exercise cessation, this time delay is probably long enough to substantially affect the results because we have shown that substantial recovery of peripheral fatigue occurs within this period after intense exercises (10,11). Robust data on the development of the different components of fatigue and their relationship to the pacing strategies adopted during TT of different durations is yet to be provided.

Accordingly, this study was designed to evaluate the time course of fatigue during and immediately after termination of self-paced exercise of three different durations. Specifically, the first aim of this study was to determine if there is a difference in the time course and the extent to which peripheral and central fatigue develops during TT of different durations. The second aim was to determine if a “critical level” of peripheral fatigue develops during fatiguing exercise, according to the model proposed by Amann (2,3). The third aim was to investigate the time course of skeletal muscle recruitment and peripheral fatigue during and between the trials. We hypothesized that: (i) a similar “critical level” of peripheral fatigue will not develop across trials, (ii) more central fatigue will occur during longer compared with shorter trials, and (iii) there will be differences between trials in the extent and time course of muscle recruitment.



Twelve subjects (11 men, one woman) trained in both endurance and strength (training greater than seven times a week) participated in the study. Subjects with previous or present injuries of the right leg were excluded. Average (± SD) age, body mass, and height were 24.0 ± 6.3 yr, 75.0 ± 11.4 kg, and 179 ± 10 cm, respectively. The study was approved by the Regional Ethics Committee Vest in Norway and The Research and Ethics Committee of the Faculty of Health Science of the University of Cape Town. The subjects gave their written informed consent to participate in the study. Subjects were given a full explanation of the details and rationale of the study and were informed that they were free to withdraw at any time. The possibility that electrical stimulation might cause discomfort was fully explained as was the nature of the risks involved.

Experimental Design

Each subject made nine visits to the laboratory, that is, six familiarization visits and three experimental visits separated by 2–3 d. During all familiarization visits, the subjects were accustomed to neuromuscular function assessment and performing a TT with repetitive self-paced concentric knee extension movements of the right leg on the KinCom dynamometer (Kinematic Communicator, Chattecx Corp., Chattanooga, TE). Subjects performed two familiarization trials for each of the three experimental trials during visits 1–6. The experimental TT during visits 7 to 9 were performed in randomized order with TT of 3-, 10-, and 40-min duration. The subjects were not allowed to consume alcohol or other stimulants for 24 h preceding the exercise tests. Neuromuscular function was assessed before, during, and after the TT.


Settings and warm-up

On arrival at the laboratory, subjects were secured to the dynamometer by chest and hip strapping to avoid excessive lateral and frontal plane movements. The seating was adjusted for each subject, with the right knee femoral epicondyle aligned with the axis of the rotation arm of the dynamometer. Settings were saved for the next sessions. The right lower leg was attached to the lever arm just above the lateral malleolus. The left leg was not active at any time and was secured to the dynamometer by strapping around the upper leg. Subjects kept their hands crossed in front of their upper body and in the same position during all experiments.

The subjects warmed up by performing four sets of 15 concentric contractions at 30% of MVC force with 30-s breaks between sets. After 1-min rest, 4-s isometric contractions of 2% × 50% and 2% × 70% of MVC force were performed. The rest period between each contraction was 30 s. A target line on a computer screen was used for visual feedback of the force recordings.

Preexercise and postexercise measurements

Pre-TT neuromuscular function (Fig. 1A) was assessed after the warm-up. Five concentric MVC (MVCCON) were performed and followed by electrical stimulation (see below) within 2 s. After a 2-min break, three isometric MVC (MVCISO), each lasting 4 s were performed with a 1-min break between MVC. Electrical stimulation on relaxed muscles consisted in single stimulus (SS), paired stimuli at 100 Hz (PS100), and paired stimuli at 10 Hz (PS10) and was performed immediately (<2 s) after the first two MVCISO. PS100 (superimposed doublet) was applied when the subject produced maximal force during the third MVCISO. The third MVCISO was repeated if superimposed doublet was not applied when voluntary force was above a force threshold of less than 5% below voluntary force measured during the larger of the first two MVCISO. Neuromuscular function assessment (4-s MVCISO + electrical stimulation) was also performed immediately (<2 s) after completion of the TT. MVCISO with superimposed doublet was assessed immediately (<5 s) after the first neuromuscular function assessment after TT termination.


Experimental TT

The TT started 5 min after the end of pre-TT neuromuscular function assessment. Each TT consisted of sets of 15 concentric contractions, and between sets there was a 5- or 10-s period with neuromuscular function assessment or rest with the knee angle at 90° (Fig. 1B). Numbers of sets were 4 for 3-min TT, 16 for 10-min TT, and 64 for 40-min TT (Fig. 1C). Range of motion during the concentric contractions was from 90° to 15° of knee flexion (0° is full extension). Each contraction consisted of a knee extension phase at 60°·s−1 during which subjects were encouraged to try to accelerate the lever arm of the dynamometer and a passive knee flexion phase at 120°·s−1 in which no muscle activity or effort was required. One cycle lasted 2.2 s including a 1.4-s extension phase and a 0.8-s passive flexion phase.

The 3-min TT had MVCCON during the 15th contraction followed by a 10-s neuromuscular function assessment period (Fig. 1B-1) for all sets. The 10-min TT had MVCCON and 10-s neuromuscular function assessment after sets 1–4 and sets 14–16, and MVCCON and 5-s neuromuscular function assessment (Fig. 1B-2) after sets 6, 8, 10, and 12. The 40-min TT had MVCCON and 10-s neuromuscular function assessment after sets 1–4 and 62–64, and MVCCON and 5-s neuromuscular function assessment after sets 6, 8, 12, 16, 24, 32, 40, 48, 56, and 60. Remaining sets involving 15 self-paced contractions (without any MVC) was followed by 5-s rest (Fig. 1B-3). To limit the neuromuscular function assessment period during the TT to only 5 or 10 s, PS10 was not assessed during the TT. The subjects were asked to report their RPE (5) for each set of contraction containing neuromuscular function assessment. During each TT, the participant was encouraged to complete the TT exerting as much force as possible. The subjects received constant visual feedback of their force production and how much of the TT they had completed.

Data Collection

Electrical stimulation

A high-voltage (maximal voltage, 400 V) constant current stimulator (DS7AH; Digitimer, Hertfordshire, UK) triggered by a Power Lab and Lab Chart software (ADInstruments, Bella Vista, Australia) was used to deliver square-wave stimuli of 1-ms duration. The femoral nerve was stimulated percutaneously via a 10-mm diameter self-adhesive cathode electrode (Skintact, Austria) pressed manually by the investigator onto the skin at the femoral triangle. The anode, a 130 × 80 mm self-adhesive electrode (Cefar-Compex Scandinavia AB, Sweden) was applied to the gluteal fold. The optimal stimulation intensity for one SS was determined by increasing the current gradually from 5 mA until a plateau in force (15–35 mA) was reached. The current was then increased to 130% to ensure that the intensity was supramaximal throughout the experiment. The current was kept constant for the same subject for all types of electrical stimulation. The subjects were instructed to relax fully while the electrical stimulation was applied.

Neuromuscular function assessment

Neuromuscular function assessment consisted of an MVC (MVCCON or MVCISO or both) and a sequence of electrical stimulation. The sequences of SS, PS100, and PS10 were used for neuromuscular function assessment before and after the TT. In breaks during the TT, neuromuscular function assessment consisted of either a 10-s neuromuscular function assessment period (including SS + PS100, a 4-s isometric MVCISO and SS + PS100) or a 5-s neuromuscular function assessment period (including SS + PS100) (Fig. 1B). Breaks between sequences of SS, PS100, and PS10 were 1.5 s, and breaks between MVC and SS were 1–2 s. Because it was expected that an MVCCON would not fully potentiate the muscle in the early stages of the TT, MVCISO before electrical stimulation were performed in the first four breaks during all TT. To ensure similarities between TT, the last three neuromuscular function assessments also included both MVCCON and MVCISO. During MVCISO, subjects were instructed to reach maximum force within 1 s and then to maintain this level for 3 s while they received strong verbal encouragement. MVCCON was performed with maximal effort at the same velocity used during the self-paced contractions, that is, 60°·s−1.

EMG acquisition

The electromyography (EMG) signals from the vastus lateralis (VL) and vastus medialis (VM) were recorded using DE-2.1 single differential surface sensors with 10 mm interelectrode distance (Delsys Inc., Boston, MA). Delsys’ recommendations were used for the placement of the sensors on the skin. The skin was shaved and wiped with isopropyl alcohol before the sensors were applied. The reference electrode was applied to the patella. The EMG signals were sampled at 2000 Hz and amplified using Bagnoli-8 (Delsys Inc.). The EMG signals were transferred together with simultaneous force and electrical stimulation recordings into Power Lab (ADInstruments) and filtered using a band pass filter with a bandwidth of 15–500 Hz in Lab Chart Pro software (ADInstruments).

Experimental Variables and Data Analysis


Concentric force for the whole extension phase was averaged for each contraction. Average concentric force for each set was used for further analysis. MVCISO force was measured for 1 s during the period of peak force development, that is, 500 ms before and after peak force. The largest MVCISO force before the trials was taken as the pre-TT MVCISO. The peak force responses to electrical stimulation are reported as evoked peak force. PS10·PS100−1 (evoked peak force for PS10·PS100−1) was calculated as an index of low-frequency fatigue (25).

Voluntary activation (VA) (1) was calculated as the ratio of the amplitude of the superimposed PS100 evoked peak force over the size of the control PS100-evoked peak force: VA = (1 − superimposed PS100 evoked peak force / control PS100 evoked peak force) × 100.


The root mean square (RMS) of the EMG data of VL and VM was measured during the whole extension phase for each concentric contraction and for 1 s around peak force for MVCISO, that is, 500 ms before and after peak force. M-wave peak to peak amplitude in response to SS was also measured. RMS for the concentric contractions during and after the trials was normalized to RMS of the preexercise MVCCON for each TT. In addition, RMS measured during concentric contractions was normalized by dividing it with the M-wave peak to peak amplitude of the following SS to calculate RMS·M−1 (skeletal muscle recruitment) (19). Because of the similarities, results for M-wave and RMS·M−1 are presented as an average of VL and VM data. Separate data for VL and VM are presented as supplemental material (see Table, Supplemental Digital Content 1, Central-, and peripheral fatigue responses for 3, 10, and 40 min TT,, and Table, Supplemental Digital Content 2, Responses during 3, 10, and 40 min TT in % of pre-TT,

Statistical Analyses

The data were analyzed with Statistica 12.0 (Stat Soft Inc., Tulsa, OK). Descriptive statistics are presented as means ± SD. Two-way repeated-measures ANOVA were used to detect differences over time (pre-TT vs post-TT, or each 25% of TT duration), at one point of time between distances (3, 10, and 40 min TT), and to detect differences in evoked peak force and MVC after all TT (time × condition). When F values were significant, a Tukey post hoc test was used to localize the differences. Pearson correlation coefficients were calculated between selected pairs of variables. The statistical significance was defined at P < 0.05.


Pre-TT Responses

Before the TT, all responses with the exception of MVCCON (Table 1) and MVCCON RMS·M−1 for VL (see Table, Supplemental Digital Content 1, central, and peripheral fatigue responses for 3, 10, and 40 min TT, were similar (P > 0.05) between 3-, 10- and 40-min TT.

Neuromuscular, central, and peripheral fatigue responses for 3-, 10-, and 40-min TT.

Responses During the TT


On average, self-paced force for the entire bout was 64% ± 8%, 55% ± 12% and 46% ± 8% of pre-TT MVCCON for 3-, 10-, and 40-min TT, respectively. These values were different (P < 0.01) among all TT (Fig. 2A). Self-paced force was also different (P < 0.05) between all TT for the first but not for the last set of self-paced contractions. In addition, self-paced force increased during the final 25% of the 10-min (P < 0.05) and 40-min TT (P < 0.001) but not during the final 25% of the 3-min TT (Fig. 2B).

Responses in % of pre-TT for 3-, 10-, and 40-min duration TT. Measurements during all TT; self-paced force every set (A), self-paced force for relative duration (B), RMS·M−1 for every set (C), RMS·M−1 for relative duration (D), MVCCON RMS·M−1 for every set (E), MVCCON RMS·M−1 forrelative duration (F), PS100 evoked peak force for every set (G), PS100-evoked peak force for relative duration (H). Data are shown as mean ± SD, n = 12, n = 11 for panels C, D, E, and F. PS100, paired stimuli at 100 Hz. Significant difference between 3-min TT and 10-min TT: &P < 0.05, &&P < 0.01, &&&P < 0.001; significant difference between 10-min TT and 40-min TT: $P < 0.05, $$P < 0.01, $$$P < 0.001; significant difference between 3-min TT and 40-min TT: £P < 0.05, ££P < 0.01, £££P < 0.001; significant difference between all TT: ¤P < 0.05, ¤¤P < 0.01, ¤¤¤P < 0.001. Significant difference from previous 25% of the TT-duration for 3-min TT: #P < 0.05, ##P < 0.01, ###P < 0.001; 10-min TT: @P < 0.05, @@P < 0.01, @@P < 0.001; and 40-min TT: βP < 0.05, ββP < 0.01, βββP < 0.001.

Central command

RMS·M−1 during the self-paced exercise for the entire trial was 75% ± 10%, 66% ± 10%, and 55% ± 9% of pre-TT MVCCON for 3-, 10-, and 40-min TT, respectively. These values were different (P < 0.05) among all TT (Fig. 2C). RMS·M−1 increased during all TT, in particular during the final 25% for 10-min and 40-min TT (Fig. 2D, P < 0.01). RMS·M−1 for each set of self-paced contractions was lower (P < 0.01) than RMS·M−1 during MVCCON during all TT. RMS·M−1 during MVCCON remained unchanged (P > 0.05) from pre-TT during all TT, indicating no reduction in the maximal skeletal muscle recruitment detected in concentric contractions during the TT (Fig. 2E and F).

Peripheral responses

PS100 (Figs. 2G and H) and SS (see Table, Supplemental Digital Content 2, responses during 3, 10, and 40 min TT in % of pre-TT, declined during TT, including a decrease (P < 0.05) during the final 25% of the duration for all TT.


RPE increased progressively during all trials peaking at 18.0 ± 1.4, 19.1 ± 1.2, and 19.0 ± 0.8 at exercise termination for 3-, 10-, and 40-min TT, respectively. Final RPE were not significantly different between TT (Fig. 3).

RPE during all TT for relative duration (A) and absolute duration (B). Data are shown as means ± SD (n = 12). Significant difference between all TT for the first four contractions: ¤P < 0.05.

Post-TT Responses

All responses related to MVC force and evoked peak force were reduced (P < 0.05) immediately after all TT, with the exception of M-waves peak-to-peak amplitude for the 3-min TT (Table 1). PS100, in contrast to SS, was reduced more (P < 0.05) after 40-min TT compared with 3-min TT. Central fatigue was not detected for the 3-min TT, but VA decreased significantly for the 10-min TT (P < 0.05) and 40-min TT (P < 0.001) (Table 1 and Fig. 4). Central fatigue was also detected for the 40-min TT when using the EMG parameters because RMS·M−1 during MVCISO was significantly decreased (P < 0.05) (Table 1).

VA measured as percentage of pre-TT responses for all TT. Data are shown as means ± SD, n = 12. Significant difference from pre-TT: *P < 0.05, ***P < 0.001. Significant difference between 3-min TT and 40-min TT: £P < 0.05.


The novel and significant findings of this study were that: (i) peripheral fatigue did not reach a common critical threshold before exercise termination across the TT of different duration; (ii) in contrast to the 3-min trial, during the 10-min and 40-min TT, peripheral fatigue tended to stabilize at ∼80% of baseline values during the middle section of the trials before increasing again in the final stages; (iii) central fatigue occurred during the 40-min TT and 10-min TT but not during the TT of 3 min; and (iv) both skeletal muscle recruitment and peripheral fatigue increased during the final stages of the TT. Together, these findings indicate that the mechanisms of neuromuscular fatigue of knee extensor muscles, as well as the time course and magnitude of peripheral fatigue, depend on the duration and intensity of self-paced TT.

Peripheral fatigue at the end of trials

To the best of our knowledge, no other studies have measured peripheral fatigue after single-joint self-paced exercise of more than one distance (11), although peripheral fatigue has been compared after cycling or running TT of two (26) or three (24) distances. Peripheral fatigue after single-joint isometric (6,14,18) and dynamic (7) exercise with different intensity or duration between trials has also been reported; however, this is the first study to examine these parameters during self-paced TT of multiple durations.

During whole-body dynamic exercises, Thomas et al. (24) reported more peripheral fatigue for the shortest compared with the two other cycling trials, and Wuthrich et al. (26) reported no differences in peripheral fatigue between distances after either running or cycling. In the present study, we found that the level of peripheral fatigue was dependent on the index used to assess this subset of fatigue. In fact, similar changes in evoked peak force were found for SS, but greater reductions were found in the 40-min trial compared with the 3-min trial for PS100 and M-wave peak to peak amplitude (Table 1). To our knowledge, significant greater reductions in either high-frequency doublet or M-wave after long TT compared with short TT has not been previously reported (24,26). Thomas et al. (24) argued that the most likely explanation for the greater degree of peripheral fatigue for the shortest trial was the higher intensity of the shortest trial, and that the longer trials may have negatively affected motivation causing reduced RPE and exercise intensity. In the present study, the end spurt in skeletal muscle recruitment (Fig. 2D) and self-paced force production (Fig. 2B) contributed to the significant decrease in MVC and evoked peak force at the end of all trials. Therefore, the level of peripheral fatigue is dependent of the combination of intensity and duration of the trials (24).

Peripheral fatigue during trials

To our knowledge, comparison of the time course of peripheral fatigue between TT of different durations has not been reported in previous studies. Our data demonstrate a progressive increase in peripheral fatigue during the whole 3-min TT (Fig. 2H). On the contrary, and similar to our previous study (11), the level of peripheral fatigue stabilized during the middle part of the 10- and 40-min TT.

Although the end values of evoked peak force for all TT at first impression might seem to support the existence of a critical threshold in peripheral fatigue, a closer look at the development of skeletal muscle recruitment, force production, and evoked peak force responses suggests otherwise. Because skeletal muscle recruitment increased in the last part of all trials, a possible critical threshold of peripheral fatigue could not, at least directly, inhibit force production during the trials.

It is possible that evoked peak force is reduced as the result of a regulated process within each muscle fiber, the so-called peripheral governor (15). Support for a peripheral governor comes from the significant reduction in rate of force development and rate of relaxation between TT (see Table, Supplemental Digital Content 1, central, and peripheral fatigue responses for 3-, 10-, and 40-min TT, and from the relatively small reduction in M-wave peak to peak amplitude compared with evoked peak force (Table 1). Therefore, it is probable that the decrease in evoked peak force is caused mainly by processes within the muscle fibers, presumably involving Ca2+ handling (15). The peripheral mechanism involves regulation of actin and myosin by Ca2+ and myosin light chain phosphorylation, regulation of membrane excitability through Cl and K+ channels, and the regulation of Ca2+ availability and the sarcoplasmic reticulum Ca2+ release channel (15). The peripheral governor would in fact explain the initial drop in contractility for all TT, but somewhat stable values (∼80% of pre-TT values) during most of the 10- and 40-min TT despite the different intensities and amount of contractions.

Central fatigue after trials of 10- and 40-min duration

It has been suggested (12,21) and recently shown (24) that central fatigue was greater after exercise of longer duration compared with shorter duration. The findings of the present study support that hypothesis, although the changes in “central fatigue” for the 40-min TT were smaller than the 7%–11% reduction in VA after cycling TT of 4–40 km (24). The present study therefore for the second time shows that central fatigue is greater for a long TT compared with a short TT, and for the first time shows this phenomenon for single-joint TT. Further studies should investigate if there is a difference in central fatigue between single-joint and whole-body locomotor exercise and try to understand the reasons for higher central fatigue in longer exercises.

Neuromuscular fatigue and pacing

Reduction in MVC force reached similar levels at cessation of all TT (Table 1) and is in agreement with previous studies (24,26). For the 3-min TT, reduction in MVC force can be explained by peripheral fatigue and not by central fatigue. For the 10- and 40-min TT, reduction in MVC force can mainly be explained by peripheral fatigue, but also to a small extent by central fatigue (Table 1, Figs. 2 and 4).

It has been suggested that feedback from fatiguing muscles reduces central motor drive and hence exercise intensity once a critical level of peripheral fatigue has been reached (2,3). In the present study, peripheral fatigue, measured as either SS or PS100, increased constantly during the 3-min TT, whereas during self-paced contractions, EMG increased and force remained mostly constant, even though there was an apparent “end-spurt” in the EMG signal (Figs. 2A–D, Table, Supplemental Digital Content 2, Responses during 3-, 10-, and 40-min TT in % of pre-TT, Therefore, for the 3-min TT, our data do not support the existence of a negative feedback system whereby peripheral fatigue limits the extent of force production.

The 10- and 40-min TT, on the other hand, showed a somewhat different picture. In these trials, force production during self-paced contractions increased in the final 50% of the trials, which was accompanied by an increase in EMG signal (Table, Supplemental Digital Content 2, Responses during 3-, 10-, and 40-min TT in % of pre-TT,, whereas peripheral fatigue was essentially stable during the mid-part of the trials (first 75% of the 10-min TT and between 25% and 75% of the 40-min TT), increasing only after the end-spurt. It is therefore possible that in these more “endurance”-like tasks, the level of peripheral fatigue is one of the factors affecting pacing, as it does not increase progressively as during the 3-min TT. Such “break” is then released very close to the end of the task, resulting in increase in muscle recruitment, force production, and consequentially an increase in peripheral fatigue. There does not seem to be, however, any evidence for a direct inhibitory action of peripheral fatigue on muscle recruitment.

Central regulation

Increasing skeletal muscle recruitment to compensate for decreasing muscle contractility is expected in trials to task failure (6,8). Such compensation may be expected also for self-paced exercise, because force production tends to be more or less stable for the most part of exercise (Fig. 2). It has been suggested that RPE is an important factor regulating exercise performance (23); however, some also argue that RPE is unrelated to feedback from the muscles (16). RPE and skeletal muscle recruitment are related during constant load cycling (9) but in the middle part of the 10- and 40-min TT in this study RPE continued to increase, whereas skeletal muscle recruitment and the level of peripheral fatigue remained unchanged. Therefore, we suggest that RPE and exercise performance are not only related to feed-forward processes within the brain (16) or to feedback from the working muscles (2,3), but also likely reflect a response to anticipation of exercise duration and intensity (20).

Limitations and perspectives

This study was performed in a laboratory on a dynamometer, which allowed us to control for potentiation, and to measure parameters, such as evoked peak force and M-wave, during the TT and within 2–5 s after exercise cessation to limit any recovery of muscle function that would reduce the degree of peripheral fatigue that could be detected. Peripheral fatigue and normalized RMS could therefore be compared with self-paced force during exercise. We studied single-joint exercise which may cause more peripheral fatigue than does two-legged exercise (e.g., (17)). Therefore, the results cannot be generalized to activities, such as running or cycling. With this proviso, future studies should investigate real endurance TT involving running and cycling with measurement of fatigue during and immediately after exercise termination to better understand the complex interaction between the central and peripheral regulators of exercise performance. Neuromuscular function has indeed been measured during cycling at constant intensity (22) and in short breaks between repeated running sprints (13), and a future challenge is to design studies to better understand the regulators for exercise performance during TT in whole-body exercises.


These data demonstrate that the extent of peripheral and central fatigue that contribute to reductions in force of single-limb dynamic contractions depend on the duration and intensity of self-paced exercise. Despite similar levels of MVC force, evoked peak force for SS, and sensation of fatigue, less evoked peak force for PS100, M-wave peak to peak amplitude and VA occurred during the 40-min TT compared with the 3-min TT. Peripheral fatigue did not reach a common critical threshold before exercise termination across the TT of different duration, and the increase in peripheral fatigue during the final stages of all TT was compensated by increasing skeletal muscle recruitment to maintain or increase force production.

This research was funded by the University of Cape Town Staff Research Fund, the Medical Research Council of South Africa, Discovery Health and the National Research Foundation.

The authors have no conflict of interest to disclose. The results of the present study do not constitute endorsement by ACSM.

We are grateful to the subjects who volunteered to participate in the study.

The experiments were performed at Sogn og Fjordane University College, Norway. C. F. and T. D. N. conceptualized and designed the study. C. F. collected the data. C. F. analyzed the data, while all authors interpreted the data. C. F. drafted the article. All authors contributed to the manuscript and approved the final version of the article.


1. Allen GM, Gandevia SC, McKenzie DK. Reliability of measurements of muscle strength and voluntary activation using twitch interpolation. Muscle Nerve. 1995; 18(6): 593–600.
2. Amann M. Central and peripheral fatigue: interaction during cycling exercise in humans. Med Sci Sports Exerc. 2011; 43(11): 2039–45.
3. Amann M. Significance of group III and IV muscle afferents for the endurance exercising human. Clin Exp Pharmacol Physiol. 2012; 39(9): 831–5.
4. Amann M, Dempsey JA. Locomotor muscle fatigue modifies central motor drive in healthy humans and imposes a limitation to exercise performance. J Physiol. 2008; 586(Pt 1): 161–73.
5. Borg GA. Perceived exertion. Exerc Sport Sci Rev. 1974; 2: 131–53.
6. Burnley M, Vanhatalo A, Jones AM. Distinct profiles of neuromuscular fatigue during muscle contractions below and above the critical torque in humans. J Appl Physiol. 2012; 113(2): 215–23.
7. Christian RJ, Bishop DJ, Billaut F, Girard O. Peripheral fatigue is not critically regulated during maximal, intermittent, dynamic leg extensions. J Appl Physiol. 2014; 117(9): 1063–73.
8. Decorte N, Lafaix PA, Millet GY, et al. Central and peripheral fatigue kinetics during exhaustive constant-load cycling. Scand J Med Sci Sports. 2012; 22(3): 381–91.
9. Fontes EB, Smirmaul BP, Nakamura FY, et al. The relationship between rating of perceived exertion and muscle activity during exhaustive constant-load cycling. Int J Sports Med. 2010; 31(10): 683–8.
10. Froyd C, Beltrami FG, Jensen J, et al. Potentiation and electrical stimulus frequency during self-paced exercise and recovery. J Hum Kinet. 2014; 42: 91–101.
11. Froyd C, Millet GY, Noakes TD. The development of peripheral fatigue and short-term recovery during self-paced high-intensity exercise. J Physiol. 2013; 591(Pt 5): 1339–46.
12. Gandevia SC. Spinal and supraspinal factors in human muscle fatigue. Physiol Rev. 2001; 81(4): 1725–89.
13. Goodall S, Charlton K, Howatson G, Thomas K. Neuromuscular fatigability during repeated-sprint exercise in male athletes. Med Sci Sports Exerc. 2015; 47(3): 528–36.
14. Katayama K, Amann M, Pegelow DF, et al. Effect of arterial oxygenation on quadriceps fatigability during isolated muscle exercise. Am J Physiol Regul Integr Comp Physiol. 2007; 292(3): R1279–86.
15. Macintosh BR, Holash RJ, Renaud JM. Skeletal muscle fatigue - regulation of excitation-contraction coupling to avoid metabolic catastrophe. J Cell Sci. 2012; 125(Pt9): 2105–14.
16. 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.
17. Matkowski B, Place N, Martin A, et al. Neuromuscular fatigue differs following unilateral vs bilateral sustained submaximal contractions. Scand J Med Sci Sports. 2011; 21(2): 268–76.
18. Millet GY, Aubert D, Favier FB, et al. Effect of acute hypoxia on central fatigue during repeated isometric leg contractions. Scand J Med Sci Sports. 2009; 19(5): 695–702.
19. Millet GY, Martin V, Maffiuletti NA, et al. Neuromuscular fatigue after a ski skating marathon. Can J Appl Physiol. 2003; 28(3): 434–45.
20. Noakes TD. Time to move beyond a brainless exercise physiology: the evidence for complex regulation of human exercise performance. Appl Physiol Nutr Metab. 2011; 36(1): 23–35.
21. Place N, Yamada T, Bruton JD, et al. Muscle fatigue: from observations in humans to underlying mechanisms studied in intact single muscle fibres. Eur J Appl Physiol. 2010; 110(1): 1–15.
22. Sidhu SK, Cresswell AG, Carroll TJ. Motor cortex excitability does not increase during sustained cycling exercise to volitional exhaustion. J Appl Physiol. 2012; 113(3): 401–9.
23. Smirmaul BP, Dantas JL, Nakamura FY, et al. The psychobiological model: a new explanation to intensity regulation and (in)tolerance in endurance exercise. Rev Bras Educ Fís Esporte. 2013; 27(2): 333–40.
24. Thomas K, Goodall S, Stone M, Howatson G, St Clair Gibson A, Ansley L. Central and peripheral fatigue in male cyclists after 4-, 20-, and 40-km time trials. Med Sci Sports Exerc. 2015; 47(3): 537–46.
25. Verges S, Maffiuletti NA, Kerherve H, Decorte N, Wuyam B, Millet GY. Comparison of electrical and magnetic stimulations to assess quadriceps muscle function. J Appl Physiol. 2009; 106(2): 701–10.
26. Wuthrich TU, Eberle EC, Spengler CM. Locomotor and diaphragm muscle fatigue in endurance athletes performing time-trials of different durations. Eur J Appl Physiol. 2014; 114(8): 1619–33.


Supplemental Digital Content

© 2016 American College of Sports Medicine