A convenient method to reduce crosstalk in surface EMG. Cobb Award-winning article, 2001.

OBJECTIVES: Crosstalk in surface EMG can be reduced by the use of spatial filters. We compared a variety of spatial filters to establish the most effective and the least complex method to reduce crosstalk. METHODS: Six different spatial filters described in the literature were tested in 8 healthy volunteers. Electrode arrays were placed over the anterior tibial and triceps surae muscles. Selective muscle activation was achieved both by supramaximal nerve stimulation and by maximal voluntary contraction. Selectivity of activation was guaranteed by using intramuscular wire electrodes. Crosstalk was quantified by dividing the amount of EMG activity recorded during pure agonist activation (i.e. the muscle directly under the electrode array) by the EMG activity recorded during pure antagonist activation. This was done for both compound muscle action potentials and voluntary muscle activation. The amount of crosstalk recorded with the different spatial filters was compared with that recorded with a standard bipolar lead. RESULTS: Crosstalk was most reduced by the "double-differential" (DD) filter, yielding an up to 6-fold improvement of EMG selectivity. We then compared signals recorded with this DD filter with those recorded with the less complex "branched electrode". As expected on theoretical grounds, signals from both filter types were identical. CONCLUSIONS: Crosstalk is best reduced using a "double-differentiating" recording technique, which can be achieved easily using a branched electrode instead of a standard bipolar lead. This technique can be used with all conventional EMG equipment.     

High‐pass filtering surface EMG in an attempt to better represent the signals detected at the intramuscular level

Surface electromyography (EMG) is often used to represent activation profiles of the underlying musculature. The purpose of this study was to assess the potential of high‐pass (HP) filtering to improve the matching of surface EMG signals to those signals recorded intramuscularly. EMG was recorded at the skin surface over the infraspinatus and supraspinatus muscles as well as from fine‐wire electrodes placed in the infraspinatus, supraspinatus, and teres minor muscles. The surface EMG signals were HP‐filtered at 18 cutoff frequencies (0–510 HZ in 30 HZ increments), and the time‐histories were correlated with the signals from the wire electrodes. HP filtering did not significantly alter the correlated muscle activation waveform relationship between the surface and wire signals until cutoffs reached 240 HZ . HP filtering of the surface signals did not improve the representation of the muscle fiber‐level activation profile, but the results suggest that enough information resides in the high‐frequency components of the signal to reproduce the activation time–history profile of the muscle. Muscle Nerve, 2010     

A method for positioning electrodes during surface EMG recordings in lower limb muscles

PURPOSE: The aim of this work is to provide information about the degree of inter-subject uniformity of location of innervation zone (IZ) in 13 superficial muscles of the lower limb. The availability of such information will allow researchers to standardize and optimize their electrode positioning procedure and to obtain accurate and repeatable estimates of surface electromyography (sEMG) signal amplitude, spectral variables and muscle fiber conduction velocity. METHODS: Surface EMG signals from gluteus maximus, gluteus medius, tensor faciae latae, biceps femoris, semitendinosus, vastus medialis obliquus, vastus lateralis, rectus femoris, tibialis anterior, peroneus longus, soleus, gastrocnemius medialis and lateralis muscles of ten healthy male subjects aged between 25 and 34 years (average = 29.2 years, S.D. = 2.5 years) were recorded to assess individual IZ location and signal quality. RESULTS: Tensor faciae latae, biceps femoris, semitendinosus, vastus lateralis, gastrocnemius medialis and lateralis showed a high level of both signal quality and IZ location uniformity. In contrast, rectus femoris, gluteus medius and peroneus longus were found to show poor results for both indexes. Gluteus maximus, vastus medialis obliquus and tibialis anterior were found to show high signal quality but low IZ location uniformity. Finally, soleus muscle was found to show low signal quality but high IZ location uniformity. CONCLUSIONS: This study identifies optimal electrode sites for muscles in the lower extremity by providing a standard landmarking technique for the localization of the IZ of each muscle so that surface EMG electrodes can be properly positioned between the IZ and a tendon.     

A surface EMG multi-electrode technique for characterizing muscle activation patterns in mice during treadmill locomotion

In mice a new method for 2x4-channel surface electromyography (EMG) recordings of the vastus lateralis and biceps femoris muscles during locomotion on a treadmill with varying speed is presented. The approach involves high-speed-videography (sampling interval 2.5 ms) in concert with the application of chronically implanted surface EMG multi-electrodes (EMG sampling rate 4000 Hz, frequency range 10-700 Hz). The recordings are started 2 days after surgery and finished 2 weeks after surgery. During the whole investigation period EMG recordings of both muscles have been possible. The monopolar EMG activities recorded by the electrode-arrays and the bipolar EMG signals derived from the monopolar activities permit an evaluation of the extent of myo-electrical activation in larger regions of both muscles and co-ordination between the flexor and extensor muscles. Bipolar EMG signals indicate propagation of activities along the muscle fibers and a slight effect of non-propagating signal components. Thus, the cross talk between these muscles is small and does not influence the evaluation of the EMG results. The resolution of the simultaneously recorded synchronized data allows a precise temporal correlation of kinematic and EMG parameters.     

Electrophysiological evidence of doubly innervated branched muscle fibers in the human brachioradialis muscle

OBJECTIVE: Motor-unit action potentials (MUAPs) with unstable satellite (late-latency) components are found in EMG signals from the brachioradialis muscles of normal subjects. We analyzed the morphology and blocking behavior of these MUAPs to determine their anatomical origin. METHODS: EMG signals were recorded from the brachioradialis muscles of 5 normal subjects during moderate-level isometric contractions. MUAP waveforms, discharge patterns, and blocking were determined using computer-aided EMG decomposition. RESULTS: Twelve MUAPs with unstable satellite potentials were detected, always two together in the same signal. Each MUAP also had a second unstable component associated with its main spike. The blocking behavior of the unstable components depended on how close together the two MUAPs were when they discharged. CONCLUSIONS: The latencies and blocking behavior indicate that the unstable components came from branched muscle fibers innervated by two different motoneurons. The satellite potentials were due to action potentials that traveled to the branching point along one branch and back along the other. The blockings were due to action-potential collisions when both motoneurons discharged close together in time. SIGNIFICANCE: Animal studies suggest that branched muscle fibers may be a normal characteristic of series-fibered muscles. This study adds to our understanding of these muscles in humans.     

Comparison of electromyographic responses for the superficial quadriceps muscles: Cycle versus knee‐extensor ergometry

The purpose of this study was to compare the electromyographic (EMG) amplitude and mean power frequency (MPF) versus power output relationships for the three superficial quadriceps muscles during incremental cycle (CE) and knee‐extensor (KE) ergometry in the same subjects. Eight men performed incremental CE and KE tests to exhaustion. Surface EMG signals were recorded simultaneously from the vastus lateralis (VL), rectus femoris (RF), and vastus medialis (VM). Polynomial regression analyses on a subject‐by‐subject basis indicated that the relationship between EMG amplitude versus power output was best‐fit with either a linear, quadratic, or cubic model for CE, whereas the relationship was best‐fit with a linear model for all subjects for the KE test. No consistent relationship was found for EMG MPF within subjects and between muscle groups for CE or KE. Compared with CE, however, the EMG amplitude during KE exercise was, on average, approximately 87% and 30% higher for the RF and VM muscles, respectively. These results suggest that KE exercise may be a better mode of examining EMG amplitude in the quadriceps muscle during incremental exercise than traditional CE. Muscle Nerve, 2009     

The influence of electrode placement over the innervation zone on electromyographic amplitude and mean power frequency versus isokinetic torque relationships

The purpose of this investigation was to examine the influence of electrode placement over the estimated innervation zone (IZ) for the vastus lateralis, as well as proximal and distal to the estimated IZ, on the torque-related patterns for electromyographic (EMG) amplitude and mean power frequency (MPF) during concentric and eccentric isokinetic muscle actions of the leg extensors. Eleven men performed randomly ordered, submaximal to maximal concentric and eccentric isokinetic muscle actions of the dominant leg extensors in 10% increments from 10 to 90% peak torque (PT). Surface EMG signals were recorded simultaneously from the vastus lateralis muscle with bipolar electrode arrangements placed over the estimated IZ, as well as proximal and distal to the estimated IZ. The results indicated that there were no consistent differences among the proximal, IZ, and distal electrode placement sites for the patterns of responses for absolute and normalized EMG amplitude and MPF versus torque, or the mean absolute and normalized EMG amplitude and MPF values. Thus, these findings suggested that during concentric and eccentric isokinetic muscle actions of the leg extensors, electrode placement over the estimated IZ for the vastus lateralis had no effect on the patterns of responses or mean values for absolute and normalized EMG amplitude and MPF versus torque.     

The influence of electrode shift over the innervation zone and normalization on the electromyographic amplitude and mean power frequency versus isometric torque relationships for the vastus medialis muscle

The purpose of the present study was to: (a) examine the influence of the innervation zone (IZ) for the vastus medialis on EMG signals from bipolar electrode arrangements that have their center point directly over the IZ, 10mm distal to the IZ, 10mm proximal to the IZ, and 20mm distal to the IZ, and (b) investigate the effects of normalization on EMG amplitude and center frequency values over the IZ. Ten men (mean+/-S.D. age=23.6+/-3.0 years) performed submaximal to maximal isometric muscle actions of the dominant leg extensors, and four separate bipolar surface EMG signals were detected simultaneously from the vastus medialis. One bipolar electrode arrangement had its center point located directly over the IZ, while the other electrode arrangements had their center points 10mm proximal, 10mm distal, and 20mm distal to the IZ. The results showed that there were no consistent patterns among the four electrode arrangements for the absolute and normalized EMG amplitude and mean power frequency (MPF) versus isometric torque relationships. Generally speaking, the IZ had the greatest effect on the EMG signal when the center point of the bipolar electrode arrangement was directly over it or 10mm proximal to it. In addition, normalization reduced the influence of the IZ on the absolute EMG amplitude and MPF values. Thus, these findings supported the practice of normalization, and indicated that it is a useful technique for reducing the influence of electrode location on EMG amplitude and MPF data. Future studies should examine the potential for movement of the IZ during isometric muscle actions.     

Comparison of conventional and decomposition-enhanced spike triggered averaging techniques.

OBJECTIVE: Spike triggered averaging (STA) is a technique to extract an estimate of a recurring motor unit potential from a complex electromyographic (EMG) signal. In conventional STA (C-STA), potentials related to the discharges of single intramuscular motor units are isolated and used to trigger an averager to obtain an individual surface-detected motor unit potential (S-MUP) from an EMG signal. In decomposition-enhanced STA (DE-STA), EMG signal decomposition algorithms determine discharges of a number of different motor units (4 to 10) that can be used to trigger an averager to obtain their corresponding S-MUPs. We tested the accuracy of extracting and averaging S-MUPs using DE-STA compared to C-STA for the same EMG signals. METHODS: We compared the intramuscular potentials used for triggering and the resultant averaged S-MUPs that were common in both techniques. RESULTS: We found no statistically significant differences in the metrics used to describe the triggering potentials and S-MUPs. CONCLUSIONS: We conclude that DE-STA is an accurate and efficient method to obtain a large number of intramuscular motor unit potentials and their corresponding S-MUP in proximal and distal muscles.     

Electrophysiologic mapping of the segmental anatomy of the muscles of the lower extremity.

Our knowledge of the specific root innervation of skeletal muscles is derived from accumulated clinical experience. While performing selective posterior rhizotomy for treatment of spasticity in children with cerebral palsy, we made direct electrophysiologic measurement of the root innervation of the lower extremity. We stimulated ventral roots from L2 to S2 while recording from all muscles simultaneously. The size of the evoked compound muscle action potential was used as an indication of the amount of innervation derived from stimulation of a given spinal root. We found the major root innervation for the 8 muscles studied to be: adductor longus, vastus medialis, and vastus lateralis, L3; tibialis anterior; L4; peroneus longus, L5; and medial gastrocnemius, lateral gastrocnemius, and gluteus maximus, S1. In general, each muscle received innervation from 3 or more roots. Prefixed or postfixed innervation patterns were found in 27.9% of legs examined, and there was asymmetry of innervation in 29.8%. We conclude that the segmental innervation of lower extremity muscles is broader than previously thought. Anomalous innervation occurs so frequently that caution should be used in attributing any pattern of clinical or EMG findings to a specific spinal level.     

Nerve stimulation boosts botulinum toxin action in spasticity.

Spasticity leads to functional and structural changes in nerves and muscles, which alter skeletal muscle function. To evaluate whether short-term electrical nerve stimulation (NS) improves the effect of botulinum toxin in spastic skeletal muscle, we studied changes in the amplitude of the compound muscle action potential (CMAP) recorded from the extensor digitorum brevis (EDB) muscle in response to peroneal nerve stimulation at the ankle after injection of botulinum toxin type A (BTXA) alone or combined with short-term NS. In paraparetic patients, both EDB muscles were injected with BTXA; and NS was applied to one EDB muscle alone. All patients received a 30-minute session of electrical NS once a day for 5 consecutive days after BTXA injection. We used two different stimulation frequencies (low-frequency, 4 Hz; and high-frequency, 25 Hz). EDB-CMAP amplitudes were evaluated before BTXA injection (day 0) and changes in CMAP amplitude, expressed as a percentage (CMAP%), were measured at various time points over a 30-day period after BTXA injection. We compared changes in the CMAP% amplitude on the stimulated and contralateral nonstimulated sides. We also studied the electromyographic activity recorded from EDB muscles over a 30-day period. CMAP% amplitudes measured at all time points after BTXA injections were significantly reduced in both EDB muscles. On days 4, 10, and 15, the CMAP% amplitude reduction was significantly greater for the low-frequency stimulated EDB than for the contralateral nonstimulated EDB. No significant differences in CMAP% were observed for the high-frequency stimulated and nonstimulated EDB. After BTXA injection, spontaneous activity appeared in both EDB muscles; but it appeared earlier and involved larger areas in the stimulated than in the nonstimulated EDB. In conclusion, short-term NS accelerates the effectiveness of intramuscular BTXA injections on the neuromuscular blockade in patients with spastic paraparesis and could induce a rapid and persistent improvement in spasticity. Its action probably arises mainly from low-frequency NS.     

Muscle-fiber conduction velocity during concentric and eccentric actions on a flywheel exercise device

A gravity-independent flywheel exercise device (FWED) has been proven effective as a countermeasure to loss of strength and muscle atrophy induced by simulated microgravity. This study assessed muscle-fiber conduction velocity (CV) and surface EMG instantaneous mean power spectral frequency (iMNF) during brief bouts of fatiguing concentric (CON) and eccentric (ECC) exercise on a FWED in order to identify electromyographic (EMG) variables that can be used to provide objective indications of muscle status when exercising with a FWED. Multichannel surface EMG signals were recorded from vastus lateralis and medialis muscles of nine men during: (1) isometric, 60-s action at 50% of maximum voluntary action (MVC); (2) two isometric, linearly increasing force ramps (0-100% MVC); and (3) dynamic CON/ECC coupled actions on the FWED. Muscle-fiber CV and iMNF were computed over time during the three tasks. During ramps, CV, but not iMNF, increased with force (P < 0.001). Conduction velocity and iMNF decreased with the same normalized rate of change in constant-force actions. During CON/ECC actions, the normalized rate of change over time was larger for CV than iMNF (P < 0.05). These results suggest that, during fatiguing, dynamic, variable-force tasks, changes in CV cannot be indirectly inferred by EMG spectral analysis. This underlines the importance of measuring both CV and spectral variables for muscle assessment in dynamic tasks.     

Activation linearity and parallelism of the superficial quadriceps across the isometric intensity spectrum

The purpose of this study was to assess neuromuscular activation of the three superficial portions of the quadriceps femoris muscles during linearly increasing isometric contraction intensities. Thirty healthy volunteers were assessed for isometric electromyographic (EMG) activity of the vastus medialis (VM), vastus lateralis (VL), and rectus femoris (RF) muscles with the knee at 60 degrees of flexion. For 5 s, subjects performed isometric contractions equivalent to 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90% of the average of three maximal voluntary contractions (MVC), in random order. Full-wave rectified and integrated EMG signals over the middle 3 s of each contraction were expressed as a percentage of the activity recorded during the three averaged MVCs. One sample t-tests and 95% confidence intervals were calculated at each relative torque level. A two-factor analysis of variance (muscle by intensity) with repeated measures was performed to evaluate parallel activation across the intensity levels. Activation linearity was assessed via regression analysis for each muscle. VM activation was shown to be significantly lower than expected at 20-70% MVC. VL and RF activations were significantly higher than expected at 10% MVC, and RF EMG was less than expected at 40-70% MVC. EMG of VM was shown to increase significantly more than VL and RF from 80% to 90% MVC. Significant linear and quadratic relations were also demonstrated for all three muscles. Parallel activation of the superficial quadriceps muscles occurred from low to moderate intensities, whereas convergence was noted at near maximal intensities.     

The electromyographic signal as a presymptomatic indicator of organophosphates in the body

Organophosphate (OP) compounds are present in household and agricultural pesticides as well as in nerve agents. The toxic effects of these chemicals result from their anticholinesterase activity, which disrupts nerve junctions and parasympathetic effector sites, leading to a variety of symptoms and possible death. When the anticholinesterase agents in OP compounds reach the neuromuscular junction, they cause a disruption in the firing of muscle fiber action potentials. This effect has the potential of altering the time course of the electromyographic (EMG) signal detected by surface electrodes. We investigated the association between OP compound dose, surface EMG changes, and overt signs of OP toxicity. Daily doses of 10-15 microg/kg of diisopropylfluorophosphate (DFP) were injected into the calf muscle of four rhesus monkeys while surface EMG signals were recorded from two thigh muscles bilaterally. With increasing number of doses, the EMG signal presented an increasing number of time gaps. The presence of the gaps was evident prior to any overt symptoms of cholinesterase toxicity. These findings can lead to the development of noninvasive technology for indicating the presence of OP compounds in muscle tissue prior to clinical abnormalities.     

Time-frequency analysis and estimation of muscle fiber conduction velocity from surface EMG signals during explosive dynamic contractions

Time-frequency analysis of the surface electromyographic (EMG) signal is used to assess muscle fiber membrane properties during dynamic contractions. The aim of this study was to compare the direct estimation of average muscle fiber conduction velocity (CV) with instantaneous mean frequency (iMNF) of surface EMG signals in isometric and explosive dynamic contractions. The muscles investigated were the vastus lateralis and medialis of both thighs in 12 male subjects. The isometric contractions were at linearly increasing force (0-100% of the maximal voluntary contraction in 10s). The explosive contractions were performed on a multipurpose ergometer-dynamometer (MED). The subject, sitting on the MED, performed six explosive contractions, separated by 2min rest, by pushing against two force platforms and thrusting himself backwards with the maximum possible speed, while completely extending his legs. The estimated CV significantly increased with force in both the isometric (mean+/-S.D., from 3.24+/-0.34 to 5.12+/-0.31m/s for vastus lateralis and from 3.17+/-0.26 to 5.11+/-0.34m/s for vastus medialis, with force in the range 10-100% of the maximal voluntary contraction level) and explosive contractions (from 4.36+/-0.49 to 5.00+/-0.47m/s for vastus lateralis, and from 4.32+/-0.46 to 4.94+/-0.44m/s for vastus medialis, with force in the range 17.5-100% of maximal thrusting force). Moreover, estimated CV was not significantly different at the maximal force in the two exercises. On the contrary, iMNF, computed from the Choi-Williams time-frequency transform, was significantly lower in the explosive (57.7+/-8.2 and 66.5+/-10.3Hz for vastus laterialis and medialis, respectively) than in the isometric exercises (73.7+/-9.2 and 75.0+/-8.5Hz for vastus laterialis and medialis, respectively) and did not change with force in any of the conditions. It was concluded that EMG spectral features provide different information with respect to average muscle fiber CV in dynamic contractions. Thus, in general, they cannot be used to infer CV changes during the exertion of a dynamic task. A joint analysis of CV and EMG spectral features is necessary in this type of contractions.     

An examination of the Runs Test, Reverse Arrangements Test, and modified Reverse Arrangements Test for assessing surface EMG signal stationarity

The purpose of this study was to examine the accuracy of the Runs Test, Reverse Arrangements Test, and modified Reverse Arrangements Test for assessing stationarity of surface electromyographic (EMG) signals. Five stationary signals were generated by custom programs written with LabVIEW programming software. These signals consisted of sine waves, sums of sine waves, and sums of sine waves and random noise. The sixth signal was a stationary computer generated surface EMG signal downloaded from the surface EMG for the non-invasive assessment of muscles (SENIAM) project database. There were no changes in the amplitude or frequency contents of the stationary signals over time. Several nonstationary signals were also created, including a nonstationary chirp signal generated with LabVIEW programming software, a nonstationary computer generated surface EMG signal downloaded from the SENIAM project database, and a real surface EMG signal recorded from the biceps brachii during a concentric isokinetic muscle action of the forearm flexors at a velocity of 30 degrees s(-1). Both the stationary and nonstationary signals were tested for stationarity using the Runs Test, Reverse Arrangements Test, and modified Reverse Arrangements Test. The results indicated that each of the three stationarity tests demonstrated at least one form of inaccuracy (i.e. false positive and/or false negative results) in examining the stationarity of the test signals. These findings may reflect the fact that these tests were designed to determine whether or not a signal is random, rather than examine signal stationarity exclusively. Thus, the Runs Test, Reverse Arrangements Test, and modified Reverse Arrangements Test may not be appropriate for assessing stationarity in surface EMG signals.     

Evaluation of an automated method for analysing the electromyogram.

An automated system, incorporating the ANOPS-101 mini-computer, has been used to analyse the EMG. The vastus medialis (VM) and biceps brachii (BB) muscles were studied in 28 controls, 16 patients with myopathies, and in 26 patients with denervating disorders. For each muscle mean values were computed for the durations and numbers of phases of muscle action potentials; the mean density and amplitude of the deflections in the interference pattern were also measured. A higher incidence of abnormalities could be detected in myopathic than in neuropathic disorders; for both conditions the incidence was significantly greater in BB than in VM. For the diagnosis of denervation the most useful measurement was that of potential duration; for the detection of myopathies amplitude determinations were also very useful. The present results have been compared with those of other published studies in which automatic EMG analysis has been employed.     

EMG spectral differences among the quadriceps femoris during the stretch reflex

Introduction: The purpose of this study was to examine the electromyographic (EMG) spectral characteristics of the quadriceps femoris muscles during tendon tap stretch reflexes. Methods: Sixteen healthy subjects (mean ± SD age = 21.2 ± 2.8 years) performed tendon tap reflexes of the leg extensors as surface EMG signals were detected from the vastus lateralis (VL), rectus femoris (RF), and vastus medialis (VM) muscles of the dominant thigh. All EMG signals were processed with a wavelet analysis, and the resulting spectra were decomposed with nonparametric spectral decomposition. Results: The results showed that the spectra for the VL had significantly more high‐frequency power than those for the RF and VM, with similar spectral shapes for the RF and VM. Conclusions: These findings could be due to differences in the width of the innervation zone, or the fiber type composition of the muscles, although the latter seems to be more likely. Muscle Nerve 52 : 826–831, 2015     

Congenital focal muscle dysplasia in the lower extremities from probable abnormal innervation: a case report.

A two-year-seven-month-old girl with pes equinovarus congenita, muscle hypotonia and weakness limited to the lower extremities is presented. Upon admission to our hospital, she could stand with support but could not walk alone. Serum creatine kinase level was normal and the electromyogram was nondiagnostic. The muscle CT disclosed an almost total absence of bilateral vastus lateralis and medialis, rectus femoris and gastrocnemius muscles. The biopsied vastus lateralis muscle was almost completely replaced by fat tissue, and a small amount of muscle tissue showed uniform type 1 fiber and an aggregate of atrophic fibers in one fascicle. Because of an absence of progressive muscle weakness and neurogenic EMG findings, the authors conclude that the muscle pathology was due to the congenital anomalous condition of probable abnormal innervation to developing muscles.     

Surface EMG profiles during different walking cadences in humans

The ensemble electromyogram (EMG) patterns associated with different walking cadences were examined in 11 normal subjects. Five muscle groups were studied: the rectus femoris, vastus lateralis, lateral hamstring, tibialis anterior and soleus muscles of the right lower extremity. The myoelectric signals were telemetered, full-wave rectified and smoothed. Subjects walked at cadences of 115, 95 and 75 steps/min. Footswitches indicated the different phases of the stride. Six or more strides per subject were averaged for each cadence. Cadence-related changes in (1) mean EMG amplitude during stance, and during swing, and (2) the shape of the EMG patterns, were analyzed. One-way repeated-measures analyses of variance on the mean EMG amplitude in stance and in swing revealed significant changes with cadence (P less than 0.05) in all muscles examined. The magnitude of these changes could be related to the mechanical function of the muscles involved. The shape of the EMG patterns generally remained similar at the different cadences. The timing of EMG activity was closely related to the normalized stride time and remained invariant at different cadences.