A new surface electromyography analysis method to determine spread of muscle fiber conduction velocities.

Muscle fiber conduction velocity (MFCV) estimation from surface signals is widely used to study muscle function, e.g., in neuromuscular disease and in fatigue studies. However, most analysis methods do not yield information about the velocity distribution of the various motor unit action potentials. We have developed a new method-the interpeak latency method (IPL)-to calculate both the mean MFCV and the spread of conduction velocities in vivo, from bipolar surface electromyogram (sEMG) during isometric contractions. sEMG was analyzed in the biceps brachii muscle in 15 young male volunteers. The motor unit action potential peaks are automatically detected with a computer program. Associated peaks are used to calculate a mean MFCV and the SD. The SD is taken as a measure of the MFCV spread. The main finding is that the IPL method can derive a measure of MFCV spread at different contraction levels. In conclusion, the IPL method provides accurate values for the MFCV and additionally gives information about the scatter of conduction velocities.     

Effect of power, pedal rate, and force on average muscle fiber conduction velocity during cycling.

Muscle fiber conduction velocity (MFCV) provides indications on motor unit recruitment strategies due to the relation between conduction velocity and fiber diameter. The aim of this study was to investigate MFCV of thigh muscles during cycling at varying power outputs, pedal rates, and external forces. Twelve healthy male participants aged between 19 and 30 yr cycled on an electronically braked ergometer at 45, 60, 90, and 120 rpm. For each pedal rate, subjects performed two exercise intensities, one at an external power output corresponding to the previously determined lactate threshold (100% LT) and the other at half of this power output (50% LT). Surface electromyogram signals were detected during cycling from vastus lateralis and medialis muscles with linear adhesive arrays of eight electrodes. In both muscles, MFCV was higher at 100% LT compared with 50% LT for all average pedal rates except 120 rpm (mean +/- SE, 4.98 +/- 0.19 vs. 4.49 +/- 0.18 m/s; P < 0.001). In all conditions, MFVC increased with increasing instantaneous knee angular speed (from 4.14 +/- 0.16 to 5.08 +/- 0.13 m/s in the range of instantaneous angular speeds investigated; P < 0.001). When MFCV was compared at the same external force production (i.e., 90 rpm/100% LT vs. 45 rpm/50% LT, and 120 rpm/100% LT vs. 60 rpm/50% LT), MFCV was higher at the faster pedal rate (5.02 +/- 0.17 vs. 4.64 +/- 0.12 m/s, and 4.92 +/- 0.19 vs. 4.49 +/- 0.11 m/s, respectively; P < 0.05) due to the increase in inertial power required to accelerate the limbs. It was concluded that, during repetitive dynamic movements, MFCV increases with the external force developed, instantaneous knee angular speed, and average pedal rate, indicating progressive recruitment of large, high conduction velocity motor units with increasing muscle force.     

Task-related differences in peroneus longus muscle fiber conduction velocity

It has been identified that the peroneus longus presents a regional activity. Specifically, a greater activation of the anterior and posterior compartments has been observed during eversion, whereas a lower activation of the posterior compartment has been reported during plantarflexion. In addition to myoelectrical amplitude, motor unit recruitment can be inferred indirectly from muscle fiber conduction velocity (MFCV). However, there are few reports of MFCV of the regions that make up a muscle, and even less, MFCV of the peroneus longus compartments. This study aimed to analyze the MFCV of peroneus longus compartments during eversion and plantarflexion. Twenty-one healthy individuals were assessed. High-density surface electromyography was recorded from the peroneus longus during eversion and plantarflexion at 10%, 30%, 50%, and 70% of maximal voluntary isometric contraction. The posterior compartment presented a lower MFCV than the anterior compartment during plantarflexion, and both compartments did not show differences in MFCV during eversion; however, the posterior compartment showed an increase in MFCV during eversion compared to plantarflexion. Differences observed in the MFCV of the peroneus longus compartments could support a regional activation strategy and, to some extent, explain different motor unit recruitment strategies of the peroneus longus during ankle movements.     

Spike-triggered average torque and muscle fiber conduction velocity of low-threshold motor units following submaximal endurance contractions.

The motor unit twitch torque is modified by sustained contraction, but the association to changes in muscle fiber electrophysiological properties is not fully known. Thus twitch torque, muscle fiber conduction velocity, and action potential properties of single motor units were assessed in 11 subjects following an isometric submaximal contraction of the tibialis anterior muscle until endurance. The volunteers activated a target motor unit at the minimum discharge rate in eight 3-min-long contractions, three before and five after an isometric contraction at 40% of the maximal torque, sustained until endurance. Multichannel surface electromyogram signals and joint torque were averaged with the target motor unit potential as trigger. Discharge rate (mean +/- SE, 6.6 +/- 0.2 pulses/s) and interpulse interval variability (33.3 +/- 7.0%) were not different in the eight contractions. Peak twitch torque and recruitment threshold increased significantly (93 +/- 29 and 12 +/- 5%, P <0.05) in the contraction immediately after the endurance task with respect to the preendurance values (0.94 +/- 0.26 mN.m and 3.7 +/- 0.5% of the maximal torque), whereas time to peak of the twitch torque did not change (74.4 +/- 10.1 ms). Muscle fiber conduction velocity decreased and action potential duration increased in the contraction after the endurance (6.3 +/- 1.8 and 9.8 +/- 1.8%, respectively, P <0.05; preendurance values, 3.9 +/- 0.2 m/s and 11.1 +/- 0.8 ms), whereas the surface potential peak-to-peak amplitude did not change (27.1 +/- 3.1 microV). There was no significant correlation between the relative changes in muscle fiber conduction velocity or surface potential duration and in peak twitch torque (R2= 0.04 and 0.10, respectively). In conclusion, modifications in peak twitch torque of low-threshold motor units with sustained contraction are mainly determined by mechanisms not related to changes in action potential shape and in its propagation velocity.     

Electromyogram power spectra frequencies associated with motor unit recruitment strategies

The isolated contributions of motor unit recruitment and firing rate variations to the median frequency of the electromyogram's power density spectrum were determined. Orderly stimulation of the cat gastrocnemius motor units via nerve electrodes gave rise to linearly increasing median frequency regardless of the action potential firing rate of the active motor units. Increase in the discharge rate of all the motor units resulted in nearly constant median frequency. It was concluded that the increasing average conduction velocity during motor unit recruitment is the major contributor to variations in the electromyogram median frequency. The possibility of using the median frequency as the index to identify the recruitment control strategies employed by various muscles during increasing force contraction is suggested.     

Inference of motor unit recruitment order in voluntary and electrically elicited contractions

The relationship between surface myoelectric signal parameters and the level of voluntary or electrically elicited contractions was studied in 32 experiments on the tibialis anterior muscle of 22 healthy human subjects. Contractions were performed at 20 and 80% of the maximum voluntary contraction torque. Two levels of stimulation current were used, yielding, respectively, a maximum M wave and an M wave approximately 30% of the maximum. A four-bar electrode probe was used to detect single- and double-differential signals from which mean and median frequency of the power spectrum and average muscle fiber conduction velocity were estimated. Measurements obtained from voluntary contractions showed a positive correlation between contraction levels and both conduction velocity and spectral parameters. Conduction velocity increased by 21.2 +/- 10.9% when voluntary contraction level increased from 20 to 80% of the maximal value. Spectral parameters increased by similar amounts. Tetanic electrical stimulation was applied to a muscle motor point for 20 s via surface electrodes. Rectangular current pulses with 0.1-ms width and frequencies of 20, 25, 30, 35, and 40 Hz were used. Four types of behavior were observed with increasing stimulation level: 1) the two spectral parameters and conduction velocity both increased with stimulation in 15 experiments, 2) the two spectral parameters decreased and conduction velocity increased in 8 experiments, 3) the two spectral parameters and conduction velocity both decreased in 6 experiments, and 4) the two spectral parameters increased and conduction velocity decreased in 3 experiments. Conduction velocity increased with increasing stimulation current in 72% of the experiments, indicating a recruitment order similar to that of voluntary contractions, although it decreased in the other 28% of the cases, indicating a reverse order of recruitment. Contrary to what is observed in direct stimulation of nerves, motor units are not in general recruited in reverse order of size during electrical stimulation of a muscle motor point. This discrepancy may be the result of geometric factors or a lack of correlation between axonal branch diameter and the diameter of the parent motoneuron axon. Changes of conduction velocity and spectral parameters in opposite directions may be the result of the combined effect of the motor unit recruitment order and of the different tissue filtering function associated with the geometric location of the recruited motor units within the muscle.     

High-density surface EMG study on the time course of central nervous and peripheral neuromuscular changes during 8 weeks of bed rest with or without resistive vibration exercise

The aim of the present study was to assess the time course and the origin of adaptations in neuromuscular function as a consequence of prolonged bed rest with or without countermeasure. Twenty healthy males volunteered to participate in the present study and were randomly assigned to either an inactive control group (Ctrl) or to a resistive vibration exercise (RVE) group. Prior to, and seven times during bed rest, we recorded high-density surface electromyogram (sEMG) signals from the vastus lateralis muscle during isometric knee extension exercise at a range of contraction intensities (5–100% of maximal voluntary isometric torque). The high-density sEMG signals were analyzed for amplitude (root mean square, RMS), frequency content (median frequency, F_med) and muscle fiber conduction velocity (MFCV) in an attempt to describe bed rest-induced changes in neural activation properties at the levels of the motor control and muscle fibers. Without countermeasures, bed rest resulted in a significant progressive decline in maximal isometric knee extension strength, whereas RMS remained unaltered throughout the bed rest period. In line with observed muscle atrophy, both F_med and MFCV declined during bed rest. RVE training during bed rest resulted in maintained maximal isometric knee extension strength, and a strong increase (～30%) in maximal EMG amplitude, from 10 days of bed rest on. Exclusion of other factors led to the conclusion that the RVE training increased motor unit firing rates as a consequence of an increased excitability of motor neurons. An increased firing rate might have been essential under training sessions, but it did not affect isometric voluntary torque capacity.     

Myoelectric signal conduction velocity and spectral parameters: influence of force and time

Reports on measurement of muscle fiber conduction velocity in humans are scarce. Inferences on the behavior of conduction velocity have been drawn from the behavior of myoelectric spectral parameters. The present report contains information on conduction velocity and spectral parameters studied at various muscle contraction levels and during and after sustained contractions. The following results have been obtained from measurements on the tibialis anterior muscle. Conduction velocity demonstrated a positive correlation with limb circumference and with muscle force output. Thus we conclude that the diameters of the muscle fibers of high-threshold motor units are, on an average, larger than those of low-threshold motor units. The study of a sustained contraction and on the recovery after such a contraction revealed that conduction velocity consistently decreased during a strong contraction as did various myoelectric spectral parameters. However, the spectral parameters decreased approximately twice as much as did the conduction velocity, and we conclude that factors other than the conduction velocity along the muscle fibers affect the myoelectric signal during a high-level contraction. These other factors appertain to changes in the firing statistics of individual motor units as well as the correlation between the firings of different motor units.     

Electromyogram median frequency, spectral compression and muscle fibre conduction velocity during sustained sub-maximal contraction of the brachioradialis muscle.

Changes in the median frequency of the power spectrum of the surface electromyogram (EMG) are commonly used to detect muscle fatigue. Previous research has indicated that changes in the median frequency are related to decreases in muscle fibre conduction velocity (MFCV) during sustained fatiguing contractions. However, in experimental studies the median frequency has been consistently observed to decrease by a relatively greater amount than MFCV. In this paper, a new estimate of EMG frequency compression, the Spectral Compression Estimate (SCE), is compared with the median frequency of the EMG power spectrum, the median frequency of the EMG amplitude spectrum and MFCV measured during sustained, isometric, fatiguing contractions of the brachioradialis muscle at 30, 50 and 80% maximum voluntary contraction (MVC). The SCE is found to provide a better estimate of the observed changes in MFCV than the median frequency of either the EMG power spectrum or EMG amplitude spectrum.     

Orderly recruitment among motoneurons supplying different muscles.

Virtually all movements involve the recruitment of motor units from multiple muscles. Given the functional diversity of motor units (motoneurons and the muscle fibers they supply), the effective production of specific movements undoubtedly depends upon some principle(s) to organize the ensemble of active motor units. The principle acting to organize the recruitment of motor units within muscles is the size principle, whereby the first motor units to be recruited have the smallest values for axonal conduction velocity and contractile force, and are the slowest to contract and fatigue. Here we consider the possibility that the size principle applies in the recruitment of motor units across muscles, i.e., that regardless of their muscles of origin, active motor units are recruited in rank order, for example, from low to high conduction velocity. The benefits of orderly recruitment across muscles could be similar to the acknowledged advantages of orderly recruitment within muscles. One benefit is that the neural process involved in organizing active motor units would be simplified. In a muscle-based scheme, the size principle would organize only those motor units within individual muscles, leaving the nervous system with the additional task of coordinating the relative activities of motor units from different muscles. By contrast, in an ensemble-based scheme, orderly recruitment of all motor units according to the size principle would automatically coordinate motor units both within and across motor nuclei. Another potential benefit is the provision for movements with smooth trajectory, the result of interleaving the divergent torque contributions made by motor units from muscles that differ in their orientations about joints. Otherwise, if order were restricted within muscles, the torque trajectory of a joint would change unevenly as participating muscles begin contracting at different times and grade activity at different rates. These considerations support speculation that motor units recruited from co-contracting muscles are collectively recruited according to the size principle.     

The influence of force and circulation on average muscle fibre conduction velocity during local muscle fatigue

Two series of experiments were performed to examine the relationship between force and change in average muscle fibre conduction velocity (MFCV) during local muscle fatigue. The average MFCV was estimated using the cross-correlation method. In the first experiment this relationship was studied with surface EMG of vastus lateralis at force levels from 10 to 100% of maximal voluntary contraction (MVC) with and without occluded circulation. The product of relative force and time was held constant. At 10-20% MVC, MFCV increased slightly under the 2 conditions. Between 30-40% MVC, MFCV decreased, this decline in conduction velocity being significantly greater with occluded circulation. Above 40% MVC the decline in MFCV was larger at higher forces, but without any differences between the ischaemic and non-ischaemic conditions. In the second experiment the relationship between change in force and MFCV was studied during sustained maximal voluntary contractions of biceps brachii. MFCV declined during the first 26-39 s of the contraction, followed by an increase. Since this increase occurred when the force had dropped to 30-50% of the initial maximal force, a partial restoration of blood flow is thought to be responsible for this phenomenon. Because an increase in MFCV was noted, despite a further decline in force, this implies that at high force levels the change in MFCV during fatigue could partly be caused by mechanisms different from those accounting for the force loss. It is concluded that above 40% MVC intramuscular pressure is sufficiently high to cause ischaemia, and MFCV is found to be very sensitive to changes in intramuscular blood flow.     

Muscle fiber conduction velocity estimation by using normalized peak-averaging technique

In order to compute the muscle fiber conduction velocity (MFCV) and to clarify how action potentials are conducted, the normalized peak-averaging technique (NPAT) was newly employed together with computer softwares. Twelve pairs of surface electromyograms were selected from biceps brachii muscles during contraction at a level of 50% of the maximum voluntary isometric contraction in seven healthy volunteers. The techniques to compute MFCV from the time delay of the peaks (P-NPAT) and from the cross correlation (CC-NPAT) of averaged pulses were compared to the cross-correlation technique (CCT). The spread rate of averaged pulses was computed to estimate the spread of MFCVs in different motor units. Tri-phasic averaged pulses were obtained clearly by averaging more than 500 detected pulses. The P-NPAT and CC-NPAT highly correlated with the CCT in the computed MFCVs. The MFCVs obtained by P-NPAT were generally larger than those obtained by CCT, and the spread rates had in the definite values. These results suggest that the MFCV could be computed and the spread of MFCVs would be estimated from averaged pulses. The MFCV of a patient with myotonic dystrophy was also studied, and it was suggested that the NPAT would be clinically useful.     

Axonal conduction velocity of motor units of rat's medial gastrocnemius muscle

Axonal conduction velocity and its relations to different contractile properties of motor units of medial gastrocnemius muscle were investigated in nine Wistar rats anaesthetized with pentobarbitone. Functionally isolated motor units were identified as slow (S), fast resistant (FR) and fast fatigable (FF). Axons of S motor units conducted significantly more slowly than of fast units, while there was considerable overlap between conduction velocities measured for FR and FF types. The mean values of conduction velocity were 50.9 m/s for S, 68.9 m/s for FR and 71.3 m/s for FF type motor units. Strong and significant negative correlation between conduction velocity and contraction time as well as half-relaxation time was demonstrated. However, only a weak correlation between conduction velocity and twitch tension, tetanic tension or fatigue index was found. The multiple regression analysis revealed that the major factor to determine conduction velocity was contraction time.     

Relationship between innervation zone width and mean muscle fiber conduction velocity during a sustained isometric contraction

Objectives:

To examine the relationship between the biceps brachii muscle innervation zone (IZ) width and the mean muscle fiber conduction velocity (MFCV) during a sustained isometric contraction.

Methods:

Fifteen healthy men performed a sustained isometric elbow flexion exercise at their 60% maximal voluntary contraction (MVC) until they could not maintain the target force. Mean MFCV was estimated through multichannel surface electromyographic recordings from a linear electrode array. Before exercise, IZ width was quantified. Separate non-parametric one-way analyses of variance (ANOVAs) were used to examine whether there was a difference in each mean MFCV variable among groups with different IZ width. In addition, separate bivariate correlations were also performed to examine the relationships between the IZ width and the mean MFCV variables during the fatiguing exercise.

Results:

There was a significant difference in the percent decline of mean MFCV (%ΔMFCV) among groups with different IZ width (χ² (3)=11.571, p=0.009). In addition, there was also a significant positive relationship between the IZ width and the %ΔMFCV (Kendall’s tau= 0.807; p<0.001).

Conclusions:

We believe that such relationship is likely influenced by both muscle fiber size and the muscle fiber type composition.     

Motor unit activation patterns during concentric wrist flexion in humans with different muscle fibre composition

Muscle activity was recorded from the flexor carpi radialis muscle during static and dynamic-concentric wrist flexion in six subjects, who had exhibited large differences in histochemically identified muscle fibre composition. Motor unit recruitment patterns were identified by sampling 310 motor units and counting firing rates in pulses per second (pps). During concentric wrist flexion at 30% of maximal exercise intensity the mean firing rate was 27 (SD 13) pps. This was around twice the value of 12 (SD 5) pps recorded during sustained static contraction at 30% of maximal voluntary contraction, despite a larger absolute force level during the static contraction. A similar pattern of higher firing rates during dynamic exercise was seen when concentric wrist flexion at 60% of maximal exercise intensity [30 (SD 14) pps] was compared with sustained static contraction at 60% of maximal voluntary contraction [19 (SD 8) pps]. The increase in dynamic exercise intensity was accomplished by recruitment of additional motor units rather than by increasing the firing rate as during static contractions. No difference in mean firing rates was found among subjects with different muscle fibre composition, who had previously exhibited marked differences in metabolic response during corresponding dynamic contractions. It was concluded that during submaximal dynamic contractions motor unit firing rate cannot be deduced from observations during static contractions and that muscle fibre composition may play a minor role. Accepted: 5 May 1998     

Amplitude cancellation reduces the size of motor unit potentials averaged from the surface EMG.

The purpose of the study was to evaluate the influence of selected physiological parameters on amplitude cancellation in the simulated surface electromyogram (EMG) and the consequences for spike-triggered averages of motor unit potentials derived from the interference and rectified EMG signals. The surface EMG was simulated from prescribed recruitment and rate coding characteristics of a motor unit population. The potentials of the motor units were detected on the skin over a hand muscle with a bipolar electrode configuration. Averages derived from the EMG signal were generated using the discharge times for each of the 24 motor units with lowest recruitment thresholds from a population of 120 across three conditions: 1) excitation level; 2) motor unit conduction velocity; and 3) motor unit synchronization. The area of the surface-detected potential was compared with potentials averaged from the interference, rectified, and no-cancellation EMGs. The no-cancellation EMG comprised motor unit potentials that were rectified before they were summed, thereby preventing cancellation between the opposite phases of the potentials. The percent decrease in area of potentials extracted from the rectified EMG was linearly related to the amount of amplitude cancellation in the interference EMG signal, with the amount of cancellation influenced by variation in excitation level and motor unit conduction velocity. Motor unit synchronization increased potentials derived from both the rectified and interference EMG signals, although cancellation limited the increase in area for both potentials. These findings document the influence of amplitude cancellation on motor unit potentials averaged from the surface EMG and the consequences for using the procedure to characterize motor unit properties.     

Reliability of muscle fibre conduction velocity distribution estimation from surface EMG

Most of the neuromuscular diseases induce changes in muscle fibre characteristics. For example, Duchenne dystrophy is characterized by a specific loss of fast fibres, and an increase in small diameter fibres. These morphological changes may lead to large modifications in the distribution of fibre diameters, possibly producing bimodal distributions. It has already been shown that it is possible to reveal these morphological modifications through the distribution of muscle fibre conduction velocity (MFCV) as estimated from needle electromyography (EMG) recordings. In this paper, we investigate whether such changes can be extracted from surface EMG signals.

Simulation allows generation of surface EMG signals in which features are well described especially at a morphological level. Therefore, we generated a database of simulated signals both in voluntary and electrically elicited contraction conditions using a bimodal distribution of muscle fibre diameters. MFCV distributions were computed using two short-term methods based on cross-correlation and peak-to-peak techniques for voluntary contraction signals, and using a deconvolution method in time domain for electrically elicited signals. MFCV distributions were compared with true ones, as generated from modelling.

This work reveals that estimating MFCV distribution through these methods does not appear yet as precise and robust enough to accurately characterize changes in redistribution of various muscle fibre diameters. However, it opens to new experimental protocols that can be explored in order to improve the robustness of MFCV distribution estimation for the follow-up of patients suffering from neuromuscular disorders.     

Relationship between average muscle fibre conduction velocity and EMG power spectra during isometric contraction, recovery and applied ischemia

The relationship between muscle fibre conduction velocity (MFCV) and the power spectrum of surface EMGs in 3 human volunteers was studied during isometric contractions at 40% maximum voluntary contraction. In addition, the recovery of these two parameters was measured during short lasting contractions at the same force level every 30 s. The recovery phase was also studied during ischaemia, thereby preventing the recovery of MFCV. The mean MFCV was calculated by the cross-correlation method. The measurements were facilitated by a real-time estimation of the cross-correlation and the MFCV and by a graphic display of the digitised signal. During contraction a nearly linear relation was found between MFCV and the median frequency of the power spectrum (MPF). During recovery this relationship was lost in one subject: MPF restored much faster then MFCV. During recovery under ischemia MFCV did not recover, but MPF recovered partially in all subjects. It is concluded that the shift of the power spectrum to lower frequencies during fatigue cannot be explained by changes in MFCV alone. Central mechanisms also influence the power spectrum and studying the recovery of local muscle fatigue during ischemia may separate these influences from that of MFCV on the power spectrum during fatigue.     

Motor unit recruitment strategies investigated by surface EMG variables.

During isometric contractions of increasing strength, motor units (MUs) are recruited by the central nervous system in an orderly manner starting with the smallest, with muscle fibers that usually show the lowest conduction velocity (CV). Theory predicts that the higher the velocity of propagation of the action potential, the higher the power at high frequencies of the detected surface signal. These considerations suggest that the power spectral density of the surface detected electromyogram (EMG) signal may give indications about the MU recruitment process. The purpose of this paper is to investigate the potential and limitations of spectral analysis of the surface EMG signal as a technique for the investigation of muscle force control. The study is based on a simulation approach and on an experimental investigation of the properties of surface EMG signals detected from the biceps brachii during isometric linearly increasing torque contractions. Both simulation and experimental data indicate that volume conductor properties play an important role as confounding factors that may mask any relation between EMG spectral variables and estimated CV as a size principle parameter during ramp contractions. The correlation between spectral variables and CV is thus significantly lower when the MU pool is not stable than during constant-torque isometric contractions. Our results do not support the establishment of a general relationship between spectral EMG variables and torque or recruitment strategy.     

Acute changes in motor unit discharge property after concentric versus eccentric contraction exercise in knee extensor

This study aimed to investigate the motor unit firing property immediately after concentric or eccentric contraction exercise. Eighteen healthy men performed repetitive maximal isokinetic knee extension exercises with only concentric or eccentric contraction until they exerted less than 80% of the baseline strength. Before and after the fatiguing exercise, high-density surface electromyography of the vastus lateralis was recorded during submaximal ramp-up isometric contraction and individual motor units were identified. Only motor units that could be tracked before and after exercise were analyzed. Muscle cross-sectional area of the vastus lateralis was measured using ultrasound, and electrically evoked torque was recorded before and after the exercise. Sixty-five and fifty-three motor units were analyzed before and after the concentric and eccentric contractions, respectively. The results showed that motor units with moderate to high recruitment thresholds significantly decreased recruitment thresholds under both conditions, and the motor unit discharge rates significantly increased after concentric contraction compared to eccentric contraction. A greater muscle cross-sectional area was observed with concentric contraction. The evoked torque was significantly decreased under both conditions, but no difference between the conditions. These results suggest that fatiguing exercise with concentric contraction contributes to greater neural input to muscles and metabolic responses than eccentric contraction.