Electrically evoked myoelectric signals.

Electrical stimulation of the nervous system is attracting increasing attention because of the possibilities it offers for physiological investigations, clinical diagnosis, muscle function assessment, noninvasive muscle characterization, and functional control of paralyzed extremities. Parameters of the myoelectric signal evoked by surface stimulation of a muscle motor point or by stimulation of a nerve trunk by means of implanted electrodes provide information about muscle performance and properties if the stimulation artifact is properly removed or suppressed. Comparison of these parameters with those obtained during voluntary contractions provides additional insight into muscle physiology. The relationships between myoelectric signal amplitude parameters, spectral parameters, and conduction velocity are discussed with special reference to muscle fatigue. This review focuses on a few methodological aspects concerning electrical stimulation of the peripheral nervous system, detection, and processing of the electrically evoked myoelectric signals in skeletal muscles. The state of the art of the following issues is discussed: (1) properties of voluntary and electrically evoked myoelectric signals; (2) techniques for evoking and detecting myoelectric signals; (3) techniques for suppression of stimulation artifacts; (4) effect of stimulation waveforms and electrode properties; (5) signal processing techniques for electrically evoked myoelectric signals; (6) physiological significance of myoelectric signal variables; (7) order of recruitment of motor units during electrical stimulation; (8) myoelectric manifestations of fatigue in electrically stimulated muscles; (9) assessment of crosstalk by electrical stimulation; and (10) applications in sport, rehabilitation, and geriatric medicine.     

Effects of fatigue of elbow extensor muscles voluntarily induced and induced by electromyostimulation on multi-joint movement organization

To investigate the capacity of the central nervous system to integrate and differentiate two different muscular fatigue states, the present study examines the changes on multi-joint movement organization following muscular fatigue of elbow extensor muscles (triceps brachii) induced by voluntary versus electrically induced contractions. Twenty right-handed male volunteers performed throws in the horizontal plane before and after two fatiguing procedures. First, success rate of throws was not affected by fatigue neither after voluntary contractions, nor after electrically induced contractions. Despite similar reductions of the maximal voluntary isometric force and the median frequency of the electromyographic signal following both fatiguing protocols, voluntary contractions induced greater changes in muscle activation, kinematics and kinetics during throws than electrically induced contractions. The changes observed following voluntary contractions are interpreted as a compensatory strategy involving a greater contribution of the wrist. In contrast, the greater activation of the triceps brachii could compensate the weakness of this muscle induced by fatigue without any modification of the initial multi-joint movement organization.     

Cross-correlation studies of movement-related cortical potentials during unilateral and bilateral muscle contractions in humans

A useful method of studying the degree of association between two signals of varying amplitude in the time domain is to use cross-correlation analysis. We applied this to the movement-related cortical potentials digitally filtered so as to eliminate the low frequency component before applying it during maximal unilateral left (UL L), unilateral right (UL R) and bilateral (BL) contractions in I I right-handed subjects. The recording electrode sites were over the right and left motor cortex areas (C3 and C4). The BL condition revealed higher cross-correlation levels of cortical activities between the two hemispheres than in UL L or UL R contraction [UL L, r = 0.68 (SEM 0.05); UL R, r = 0.73 (SEM 0.03); BL, r = 0.76 (SEM 0.02)]. The UL R revealed a positive phase difference [5 (SEM 2) ms] when the maximal cross-correlation coefficient was shown and UL L showed a negative phase difference [5 (SEM 3) ms]. However, BL revealed a smaller phase difference [2 (SEM 1) ms] than that for UL. It was concluded that during maximal BL contraction cortical cellular activities in both hemispheres was more synchronized in amplitude and time course compared with maximal UL contractions. Our data suggested that central common drive existed between the right and left motor areas during the maximal BL handgrip contractions and the amplitude of potentials of both hemispheres was modified by the interhemispheric inhibition mechanism as reported in other studies.     

Advances in processing of surface myoelectric signals: Part 1

During sustained voluntary or electrically elicted muscle contractions the surface myoelectric signal is nonstationary and it undergoes progressive changes reflecting the modifications of the motor unit action potentials and their propagation velocity. In particular, during sustained electrical stimulation, the evoked signals show progressive amplitude, time scaling and shape modification. The quantitative evaluation of these changes is important for non-invasive muscle characterisation and may be performed in either the time or frequency domain using parametric and nonparametric spectral analysis as well as alternative methodologies. The paper introduces the detection techniques, reviews and compares the methods of spectral estimation based on FFT and autoregressive models, and discusses their applications and limitations in extracting information from the surface myoelectric signal with particular regard to myoelectric manifestations of localised muscle fatigue during sustained contractions.     

Voluntary Movement and Excitability of Cutaneous Eyeblink Reflexes

Reflex responses from eyelid muscles were evaluated when subjects prepared to or made brief or sustained voluntary contractions of eyelid or forearm muscles in a signaled reaction time task. The eyeblink reflex consists of an early and unilateral response, R1, and a longer latency bilateral response, R2. A similar pattern of reflex excitability was observed for all phasic movements. The R1 was transiently potentiated by the warning stimulus, increased again at the end of the preparatory period, increased more in the reaction interval and immediately after movement onset, and finally declined to control levels by 500 ms after onset of voluntary movement. The R2 responses were typically depressed at the same times that R1 responses were enhanced. When sustained contractions were performed and reflexes were elicited after movement eyelid onset, the patterns of early R1 enhancement and R2 depression occurred, both for about 500 ms, and gave way to continued R1 and also R2 enhancement as long as eyelid movement continued. The different behavioral responses of R1 and R2 reflect differences in the physiological properties of the separable neuroanatomical pathways for the two eyelid reflex responses.     

Estimation of single motor unit conduction velocity from surface electromyogram signals detected with linear electrode arrays

This work addresses the problem of estimating the conduction velocity (CV) of single motor unit (MU) action potentials from surface EMG signals detected with linear electrode arrays during voluntary muscle contractions. In ideal conditions, that is without shape or scale changes of the propagating signals and with additive white Gaussian noise, the maximum likelihood (ML) is the optimum estimator of delay. Nevertheless, other methods with computational advantages can be proposed; among them, a modified version of the beamforming algorithm is presented and compared with the ML estimator. In real cases, the resolution in delay estimation in the time domain is limited because of the sampling process. Transformation to the frequency domain allows a continuous estimation. A fast, high-resolution implementation of the presented multichannel techniques in the frequency domain is proposed. This approach is affected by a negligible decrease in performance with respect to ideal interpolation. Application of the ML estimator, based on two-channel information, to ten firings of each of three MUs provides a CV estimate affected by a standard deviation of 0.5 ms⁻¹; the modified beamforming and ML estimators based on five channels provide a CV standard deviation of less than 0.1 ms⁻¹ and allow the detection of statistically significant differences between the CVs of the three MUs. CV can therefore be used for MU classification.     

The electro-mechanical delay of the erector spinae muscle: influence of rate of force development,fatigue and electrode location

Summary Electro-mechanical delay (EMD) values of the erector spinae muscle were obtained using a technique based on the cross-correlation between the force and the electromyogram (EMG). Seven subjects performed a series of 20 submaximal dynamic isometric contractions in a seated position at two frequencies (0.5 Hz and 1 Hz) to study the influence of the rate of force development on EMD. Mean EMD values of 125.7 (SD 28.1) ms (1 Hz) and 136.8 (SD 28.6) ms (0.5 Hz) were shown to differ significantly (P=0.02). This finding supports the hypothesis that EMD is inversely related to the rate of force development and implies that the time to stretch the series elastic component is an important factor determining EMD. After performing a series of fatiguing contractions EMD did not differ significantly from the control value. Multiple regression analysis showed that maximal voluntary contraction force (MVC) and endurance time of the fatiguing exercise correlated significantly with EMD. The site from which the EMG signal was recorded had no significant influence on EMD. However, the coefficient of correlation between force and the EMG-signal differed significantly between electrode positions. The magnitude of the EMD values found emphasized the need to account for this delay when interpreting temporal patterns of activation of the muscles in, for example, lifting tasks.     

The pressor response to voluntary and electrically evoked isometric contractions in man

Summary Blood pressure and heart rate changes during sustained isometric exercise were studied in 11 healthy male volunteers. The responses were measured during voluntary and involuntary contractions of the biceps brachii at 30% of maximal voluntary contraction (MVC), and the triceps surae at 30% and 50% MVC. Involuntary contractions were evoked by percutaneous electrical stimulation of the muscle.Measurements of the time to peak tension of maximal twitch showed the biceps brachii (67.0±7.9 ms) muscle to be rapidly contracting, and the triceps surae (118.0±10.5 ms) to be slow contracting. The systolic and diastolic blood pressures increased linearly throughout the contractions, and systolic blood pressure increased more rapidly than diastolic. There was no significant difference in response to stimulated or voluntary contractions, nor was there any significant difference between the responses to contractions of the calf or arm muscles at the same relative tension.In contrast the heart rate rose to a higher level (P<0.01) in the biceps brachii than the triceps surae at given % MVC, and during voluntary compared with the electrically evoked contractions in the two muscle groups.It was concluded that the arterial blood pressure response to isometric contractions, unlike heart rate, is primarily due to a reflex arising within the active muscles (cf. Hultman and Sjöholm 1982) which is associated with relative tension but independent of contraction time and muscle mass.     

Single motor unit and spectral surface EMG analysis during low-force, sustained contractions of the upper trapezius muscle

Intramuscular and surface electromyographic (EMG) activities were recorded from the left and right upper trapezius muscle of eight healthy male subjects during 5-min long static contractions at 2% and 5% of the maximal voluntary contraction (MVC) force. Intramuscular signals were detected by wire electrodes while surface EMG signals were recorded with linear adhesive electrode arrays. The surface EMG signals were averaged using the potentials extracted from the intramuscular EMG decomposition as triggers. The conduction velocity of single motor units (MUs) was estimated over time from the averaged surface potentials while average rectified value and mean power spectral frequency were computed over time from 0.5 s epochs of surface EMG signal. It was found that (1) MUs were progressively recruited after the beginning of sustained contractions of the upper trapezius muscle at 2% and 5% MVC, (2) the conduction velocity of the MUs active since the beginning of the contraction significantly decreased over time, and (3) although the CV of single MUs significantly decreased, the mean power spectral frequency of the surface EMG did not show a consistent trend over time. It was concluded that spectral surface EMG analysis, being affected by many physiological mechanisms, may show limitations for the objective assessment of localized muscle fatigue during low force, sustained contractions. On the contrary, single motor unit conduction velocity may provide an early indication of changes in muscle fiber membrane properties with sustained activity.     

Motor unit recruitment order during voluntary and electrically induced contractions in the tibialis anterior

 The recruitment order of motor units (MU) was compared during voluntary and electrically induced contractions. With the use of spike-triggered averaging, a total of 302 MUs with recruitment thresholds ranging from 1% to 88% of maximal voluntary contraction were recorded in the human tibialis anterior muscle in five subjects. The mean (±SD) MU force was 98.3±93.3 mN (mean torque 16.8±15.9 mNm) and the mean contraction time (CT) 46.2±12.7 ms. The correlation coefficients (r) between MU twitch force and CT versus the recruitment threshold in voluntary contractions were +0.68 and –0.38 (P<0.001), respectively. In voluntary contractions, MUs were recruited in order of increasing size except for only 6% of the cases; whereas, during transcutaneous electrical stimulation (ES) at the muscle motor point, MU pairs showed a reversal of recruitment order in 28% and 35% of the observations, respectively, when the pulse durations were 1.0 ms or 0.1 ms. This recruitment reversal during ES was not related to the magnitude of the difference in voluntary recruitment thresholds between MUs. It is concluded that if the reversal of MU recruitment observed during ES is biophysically controlled by differences in their nerve axon input impedance, in percutaneous stimulation at the motor point, other factors such as the size and the morphological organisation of the axonal branches can also influence the order of activation. Received: 24 May 1996 / Accepted: 30 September 1996     

Myo-electric signals to augment speech recognition

It is proposed that myo-electric signals can be used to augment conventional speech-recognition systems to improve their performance under acoustically noisy conditions (e.g. in an aircraft cockpit). A preliminary study is performed to ascertain the presence of speech information within myo-electric signals from facial muscles. Five surface myo-electric signals are recorded during speech, using Ag-AgCl button electrodes embedded in a pilot oxygen mask. An acoustic channel is also recorded to enable segmentation of the recorded myo-electric signal. These segments are processed off-line, using a wavelet transform feature set, and classified with linear discriminant analysis. Two experiments are performed, using a ten-word vocabulary consisting of the numbers 'zero' to 'nine'. Five subjects are tested in the first experiment, where the vocabulary is not randomised. Subjects repeat each word continuously for 1 min; classification errors range from 0.0% to 6.1%. Two of the subjects perform the second experiment, saying words from the vocabulary randomly; classification errors are 2.7% and 10.4%. The results demonstrate that there is excellent potential for using surface myo-electric signals to enhance the performance of a conventional speech-recognition system.     

Blind separation of multichannel electrogastrograms using independent component analysis based on a neural network

The electrogastrogram (EGG) is an abdominal surface measurement of gastric myo-electrical activity which regulates gastric contractions. It is of great clinical importance to record and analyse multichannel EGGs, which provide more information on the propagation and co-ordination of gastric contractions. EGGs are, however, contaminated by myo-electric interference from other organs and artefacts such as motion and respiration. The aim of the study is to separate the gastric signal from noisy multichannel EGGs without any information on the interference, using independent component analysis. A neural-network model is proposed, and corresponding unsupervised learning algorithms are developed to achieve the separation. The performance of the proposed method is investigated using artificial data simulating real EGG signals. Experimental EGG data are obtained from humans and dogs. The processed results of both simulated and real EGG data show the following: first, the proposed method is able to separate normal gastric slow waves from respiratory artefacts and random noises. It is also able to extract gastric slow waves, even when the EGG is contaminated by severe respiratory and ECG artefacts. Secondly, when the stomach contains various gastric electric signals with different frequencies, the proposed method is able to separate these different signals, as illustrated by simulations. These data suggest that the proposed method can be used to separate gastric slow waves, respiratory and motion artefacts, and intestinal myo-electric interference that are mixed in the EGG. It can also be used to detect gastric slow-wave uncoupling, during which the stomach has multiple gastric signals with different frequencies. It is believed that the proposed method may also be applicable to other biomedical signals.     

Myo-electric signals from two extrinsic hand muscles and force tremor during isometric handgrip

Summary The objective of the present study was to investigate the myo-electric signs of muscle fatigue and the isometric force tremor of two extrinsic hand muscles, extensor digitorum communis and flexor digitorum superficialis, during isometric power grip. In addition, the synergy between flexor and extensor muscles and hand differences in a right-handed population have been studied. During isometric hand-dynamometry the myo-electric signal was recorded using surface electrodes and isometric force tremor was recorded using a special load cell. Eight subjects participated in this study and contractions were performed at 20%, 40%, 60% and 80% of maximal voluntary contraction (MVC) with left and right hands. The decrease of mean power frequency (MPF) with duration of contraction was greater in the left extensor as compared to the ipsilateral flexor muscle. No differences in the decrease in MPF with the duration of the contraction were found between the right extensor and flexor muscles. Isometric force tremor root mean square did not change during contractions at a given contraction level. Isometric tremor amplitude increased from 20% to 60% MVC and decreased at higher contraction levels. Tremor amplitude was higher in the left hand at all contraction levels but 60% MVC. These data would suggest differences in fatiguability and muscle fibre composition between the dominant and nondominant hand, which may be due to preferred use. The significance of force tremor for the evaluation of recruitment order and muscle fatigue is discussed.     

On the evaluation of muscle fiber conduction velocity considering waveform properties of an electromyogram in M. biceps brachii during voluntary isometric contraction

The surface myoelectric signal during 20% maximum voluntary contraction was measured in m. biceps brachii using array electrodes for ten subjects in order to evaluate the distribution of muscle fiber conduction velocity (MFCV) in a whole muscle. MFCV was estimated by two calculating methods of the peak maximum method and the cross-correlation method from the myoelectric signals which were processed by techniques of the averaging and the non-averaging. It was found that the values of MFCV depended on the location irrespective of the kind of calculating method used and the kind of processing technique of myoelectric signal. In both the motor end-plate zone and the tendon zone, the values of MFCV showed more than 7.0 m/s. In the regions other than the motor end-plate zone and the tendon zone, the values of MFCV showed about 3.90 m/s which were almost constant. The statistical differences of the values of MFCV were in the same locations measured not recognized between the two calculating methods nor between the two processing techniques. In the cross-correlation method, the relation between MFCV and the electrode location was evaluated by both the maximum correlation coefficient and the amplitude ratio between the different neighboring channels to evaluate the conductive waveform properties of action potentials. The changes of parameters (i.e., MFCV, maximum correlation coefficient and amplitude ratio) depended on the electrode location. The values of MFCV significantly increased in the regions of the motor end-plate zone and the tendon zone, where the maximum correlation coefficient and amplitude ratio significantly decreased. The values of the coefficient of variance (CV) of three parameters in those regions were larger than those in other regions, i.e., the regions other than the motor end-plate zone and the tendon zone. A high maximum correlation coefficient and a high amplitude ratio were necessary for a reliable measurement of the MFCV.     

Muscle fiber conduction velocity is more affected after eccentric than concentric exercise

It has been shown that mean muscle fiber conduction velocity (CV) can be acutely impaired after eccentric exercise. However, it is not known whether this applies to other exercise modes. Therefore, the purpose of this experiment was to compare the effects of eccentric and concentric exercises on CV, and amplitude and frequency content of surface electromyography (sEMG) signals up to 24?h post-exercise. Multichannel sEMG signals were recorded from biceps brachii muscle of the exercised arm during isometric maximal voluntary contraction (MVC) and electrically evoked contractions induced by motor-point stimulation before, immediately after and 2?h after maximal eccentric (ECC group, N?=?12) and concentric (CON group, N?=?12) elbow flexor exercises. Isometric MVC decreased in CON by 21.7?±?12.0% (±SD, p?<?0.01) and by 30.0?±?17.7% (p?<?0.001) in ECC immediately post-exercise when compared to baseline. At 2?h post-exercise, ECC showed a reduction in isometric MVC by 24.7?±?13.7% (p?<?0.01) when compared to baseline, while no significant reduction (by 8.0?±?17.0%, ns) was observed in CON. Similarly, reduction in CV was observed only in ECC both during the isometric MVC (from baseline of 4.16?±?0.3 to 3.43?±?0.4?m/s, p?<?0.001) and the electrically evoked contractions (from baseline of 4.33?±?0.4 to 3.82?±?0.3?m/s, p?<?0.001). In conclusion, eccentric exercise can induce a greater and more prolonged reduction in muscle force production capability and CV than concentric exercise.     

Myo-electric fatigue manifestations revisited: power spectrum,conduction velocity,and amplitude of human elbow flexor muscles during isolated and repetitive endurance contractions at 30% maximal voluntary contraction

Summary A brief survey of the literature on manifestations of myo-electric fatigue has disclosed a surprisingly sharp conflict between early studies, focusing on neuromotor regulatory mechanisms, and more recent studies which stress the determinant influence of local metabolism and skewed homeostasis. Favoured explanations concerning changes in the electromyographic (EMG) spectrum were synchronization/grouping of motor unit (MU) firing and conduction velocity (CV) decreases of the action potential propagation. The notion of mutual exclusivity interwoven with these theories prompted us to reinvestigate the EMG of moderate level, static endurance contraction. Ten men in their twenties performed isometric elbow flexion (elbow angle 135°) at 30%6 maximal voluntary contraction (MVC), and the surface EMG of the brachioradialis (BR) and biceps brachii (BB) muscles was recorded. Initially the CV — determined by cross-correlation — was 4.3 m · s^–1 (BR) and 4.6 m · s^–1 (BB). At exhaustion the CV of the BR muscle had declined by 33%, roughly twice the decrease of the BB CV. Substantially larger relative median frequency (f _m) reductions of 50% (BR) and 43% (BB) were found. Simultaneously, the root-mean-square amplitudes grew by 150% (BR) and 120% (BB). All changes during contraction reached the same level of significance (P<0.001, both muscles). From the largely uniform relative increases inf _m and CV during the last 4 min of a 5-min recovery period, variations in CV were suggested to produce equivalent shifts inf _m. The gradually increasing discrepancies between relative decreases inf _m and CV during contraction presumably reflected centrally mediated regulation of MU firing patterns (notably synchronization). After the 5-min recovery another 11 endurance contractions at 30% MVC were executed, separated by 5-min intervals. The series of contractions reduced the endurance time to one-third of the 153 s initially sustained, while the terminal CV recordings increased by 1.0 (BR) and 0.6 (BB) m · s^–1, and the terminalf _m increased by 24 (BR) and 14 (BB) Hz. The relative CV decreased in direct proportion to the endurance time and thef _m decreases varied with the CV; the findings did not support a causal link between CV decrease (signifying impaired fibre excitability) and the force failure of exhaustion.     

The origin of activity in the biceps brachii muscle during voluntary contractions of the contralateral elbow flexor muscles

During strong voluntary contractions, activity is not restricted to the target muscles. Other muscles, including contralateral muscles, often contract. We used transcranial magnetic stimulation (TMS) to analyse the origin of these unintended contralateral contractions (termed “associated” contractions). Subjects (n = 9) performed maximal voluntary contractions (MVCs) with their right elbow-flexor muscles followed by submaximal contractions with their left elbow flexors. Electromyographic activity (EMG) during the submaximal contractions was matched to the associated EMG in the left biceps brachii during the right MVC. During contractions, TMS was delivered to the motor cortex of the right or left hemisphere and excitatory motor evoked potentials (MEPs) and inhibitory (silent period) responses recorded from left biceps. Changes at a spinal level were investigated using cervicomedullary stimulation to activate corticospinal paths (n = 5). Stimulation of the right hemisphere produced silent periods of comparable duration in associated and voluntary contractions (218 vs 217 ms, respectively), whereas left hemisphere stimulation caused a depression of EMG but no EMG silence in either contraction. Despite matched EMG, MEPs elicited by right hemisphere stimulation were ∼1.5–2.5 times larger during associated compared to voluntary contractions (P < 0.005). Similar inhibition of the associated and matched voluntary activity during the silent period suggests that associated activity comes from the contralateral hemisphere and that motor areas in this (right) hemisphere are activated concomitantly with the motor areas in the left hemisphere. Comparison of the MEPs and subcortically evoked potentials implies that cortical excitability was greater in associated contractions than in the matched voluntary efforts.     

The influence of contraction amplitude and firing history on spike-triggered averaged trapezius motor unit potentials

The spike-triggered averaged (STA) technique was used to examine trapezius motor unit potentials and their dependence on contraction amplitude and firing history. Individual motor unit firings were identified by a fine-wire intramuscular electrode, while STA-derived potentials were extracted from the simultaneously recorded surface electromyographic (SEMG) signal. Amplitude-controlled contractions and contractions with typing tasks and mental stress were carried out. STA potentials were mostly derived from 20 s intervals of firing. Motor unit synchrony was estimated by peristimulus time histograms (PSTHs). An association between SEMG amplitude and STA-derived motor unit potentials was found: motor unit area showed a four-fold increase when SEMG amplitude increased from 1.5 to 10.5% of the root mean square-detected SEMG signal at maximal voluntary contraction (%EMG_max). Low- and higher threshold motor unit potentials, all with recruitment thresholds <10% EMG_max, had similar area at the same contraction amplitude. A significant increase in the STA-derived potentials was observed after 3 min of constant-amplitude contractions; however, this difference was reduced after 10 min and no longer present after 30 min of contraction. Motor unit synchrony accounted for, on average, 2.8% additional firings within 2 ms of the triggering motor unit. We conclude that the increase in STA-derived potentials with contraction amplitude is, to a major extent, due to motor unit synchrony, limiting the applicability of this method in postural muscles presenting wide motor unit potentials. The similar area of motor units at same SEMG amplitude may indicate that trapezius motor units recruited below 10% EMG_max are of similar size and thus not organized according to the Henneman size principle.     

Pattern of pulses that maximize force output from single human thenar motor units

We assessed the sequence of nerve impulses that maximize force output from individual human thenar motor units. When these motor units were stimulated intraneurally by a variable sequence of seven pulses, the pattern of pulses that elicited maximum force always started with a short (5-15 ms) interpulse interval termed a "doublet. " The twitch force summation caused by this "doublet" elicited, on average, 48 +/- 13% (SD) of the maximum tetanic force. The peak amplitude of "doublet" forces was 3.5 times that of the initial twitches, and twitch potentiation appeared to have little influence on twitch force summation elicited by the "doublets." For some units, the second optimal interpulse interval was also short. Peak forces elicited by the third to sixth interpulse intervals did not change substantially when the last interpulse interval was varied between 5 to 55 ms, so maximum force could not be attributed to any unique interpulse interval. Each successive pulse contributed a smaller force increment. When five to seven pulses were delivered in an optimal sequence, the evoked force was close to that recorded during maximal tetanic stimulation. In contrast, maximal force-time integral was evoked with one short interpulse interval (5-15 ms) then substantially longer interpulse intervals (>100 ms). Maximum force and force-time integrals were therefore elicited by different patterns of stimuli. We conclude that a brief initial interpulse interval (5-15 ms) is required to elicit maximum "doublet" force from human thenar motor units and that near-maximal tetanic forces can be elicited by only five or six additional post-"doublet" pulses if appropriately spaced in time. However, the rate at which these post-"doublet" stimuli must be provided is fairly uncritical. In contrast, maximum post-"doublet" force-time integrals were obtained at intervals corresponding to motoneuronal firing rates of approximately 7 Hz, rates close to that typically used to recruit motor units and to maintain weak voluntary contractions.