2000 Experimental Brain Research

Exp Brain Res (2000) 133:303–311 Digital Object Identifier (DOI) 10.1007/s002210000382

R E S E A R C H A RT I C L E

Vlodek Siemionow · Guang H. Yue Vinoth K. Ranganathan · Jing Z. Liu · Vinod Sahgal

Relationship between motor activity-related cortical potential and voluntary muscle activation Received: 26 April 1999 / Accepted: 15 February 2000 / Published online: 31 March 2000 © Springer-Verlag 2000

Abstract The purpose of this study was to investigate the relationship between EEG-derived motor activityrelated cortical potential (MRCP) and voluntary muscle activation. Eight healthy volunteers participated in two experimental sessions. In one session, subjects performed isometric elbow-flexion contractions at four intensity levels [10%, 35%, 60%, and 85% maximal voluntary contraction (MVC)]. In another session, a given elbow-flexion force (35% MVC) was generated at three different rates (slow, intermediate, and fast). Thirty to 40 contractions were performed at each force level or rate. EEG signals were recorded from the scalp overlying the supplementary motor area (SMA) and contralateral sensorimotor cortex, and EMG signals were recorded from the skin surface overlying the belly of the biceps brachii and brachioradialis muscles during all contractions. In each trial, the force was used as the triggering signal for MRCP averaging. MRCP amplitude was measured from the beginning to the peak of the negative slope. The magnitude of MRCP from both EEG recording locations (sensorimotor cortex and SMA) was highly correlated with elbow-flexion force, rate of rising of force, and muscle EMG signals. These results suggest that MRCP represents cortical motor commands that scale the level of muscle activation. Key words Motor activity-related cortical potential · Electroencephalography · Electromyography · Voluntary contraction

V. Siemionow · G.H. Yue · V. Sahgal Department of Physical Medicine and Rehabilitation, The Cleveland Clinic Foundation, Cleveland, OH 44195, USA V. Siemionow · G.H. Yue (✉) · V.K. Ranganathan · J.Z. Liu Department of Biomedical Engineering, ND20, Lerner Research Institute, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, OH 44195, USA e-mail: [email protected] Tel: +1 216 445 9336, Fax: +1 216 444 9198

Introduction Motor activity-related cortical potential (MRCP) is described as EEG-derived brain potential associated with voluntary movements. Bates (1951) first recorded MRCP from human scalp during voluntary hand movements using a crude photographic superimposition technique. Subsequent successful recordings of human MRCP were reported by Kornhuber and Deecke (1965), who, based on the characteristics of the MRCP recording, described a slowly rising negative potential, known as Bereitschaftspotential or readiness potential, that precedes a more sharply rising negative potential, known as negative slope. The onset of both the readiness potential and negative slope occurs prior to the onset of the voluntary movement, and, hence, both are considered to indicate involvement of the underlying cortical fields in preparing for the desired movement (Kornhuber and Deecke 1965). The waveform of any given MRCP can be further defined into several more components (Gilden et al. 1966; Hazemann et al. 1989; Hallett 1994). Because of the advances in technology of recent years (e.g., multichannel EEG acquisition system, data processing and analysis software) and the noninvasive nature of surface EEG recordings, the number of studies of brain function (especially motor function) involving MRCP measurements is showing a rapid increase. To study central nervous system mechanisms underlying voluntary movements using MRCP, it is important to understand the relationship between the magnitude of MRCP and muscle activation, especially if one intends to draw conclusions that are critically dependent on a known relationship. A number of studies have investigated whether stronger muscle output accompanies a greater MRCP, and the results are mixed. For example, a higher amplitude of negative slope or readiness potential was associated with greater joint force (Becker and Kristeva 1980; Hink et al. 1983; Kutas and Donchin 1974; Nishihira et al. 1989; Shibata et al. 1993), but such an association was not found in other studies (Hazemann et al. 1989; Wilke and Lansing 1973). Furthermore, none

304

of these studies attempted to establish a relationship between the MRCP and muscle output by performing a statistical correlation analysis, perhaps because a majority of the studies investigated MRCP at only two levels of force. Similarly, the relationship between magnitude of MRCP and rate of force development has also not been well studied. In one report, a higher MRCP (readiness potential) was associated with slower movement (Becker et al. 1976); in another, no effect of rate of force development on the amplitude of MRCP was detected (Hazemann et al. 1989). A systematic study of these relationships (MRCP vs force, MRCP vs rate of force development) is needed. The signal of MRCP is determined by synchronous and asynchronous discharges of local and remote cortical neurons (Nunez 1981). Thus, the amplitude of each component of MRCP depends on the number of participating neurons, the temporal discharge pattern (degree of synchronization) and the rate of discharge of these neurons during the time period corresponding to the MRCP component. Because muscle force relates directly to the number of recruited spinal motoneurons and their discharge rate (Henneman and Mendell 1981; MilnerBrown et al. 1973) and because stronger neural input is needed for a faster contraction at a given force (forcevelocity relationship; Close 1972; Wilkie 1950), it is reasonable to hypothesize a positive relationship between the MRCP of cortical motor areas and muscle force and between the MRCP and the rate of force development. The purpose of our study was to determine the relationship between the amplitude of MRCP and isometric elbow-flexion force (four levels) and between the MRCP and rate of this force development (three levels). In general, many motor actions are accomplished without moving the body or a body part (e.g., isometric contractions). Specifically in this study, MRCP was recorded during isometric elbow-flexion contractions. Therefore, MRCP

Fig. 1 Schematic representation of experimental apparatus and data recording system. Subject’s forearm was fixed in a wrist cuff, which was connected to a force transducer. EEG, EMG, and force data were recorded during each isometric elbow-flexion contraction in an electrically shielded room

here represents motor activity-related cortical potential as opposed to movement-related cortical potential.

Materials and methods Subjects Eight right-handed volunteers (six males and two females, mean age ± SD 32.1±11.7 years, range 23–54 years) participated in the study. All individuals were healthy and had no known neuromuscular disorders. All experimental procedures were approved by the Institutional Review Board at The Cleveland Clinic Foundation. All subjects gave informed consent prior to participation in the study. Mechanical recording Subjects were seated comfortably in an experimental chair in an electrically shielded data-recording room. During the experiment, subjects performed isometric elbow-flexion contractions. The elbow-flexion force was recorded by a JR3 force-torque transducer (JR3 Universal Force-Movement Sensor System, Woodland, CA). The transducer was mounted on a custom-designed manipulandum that provided adjustment of the arm position and joint angle. The elbow rested on a padded support at hip height and the forearm was in an intermediate position between supination and pronation. The angle of the elbow joint was about 100° throughout the experiment. The shoulders and torso were securely clamped to the back of the chair (Fig. 1). The force signal was displayed on an oscilloscope and digitized (100 samples/s) by the Spike 2 data acquisition and analysis system (Cambridge Electronic Design Ltd., Cambridge, UK). The force data were recorded online on the hard drive of a personal computer. Electrical recording EEG recording Monopolar EEG data were recorded from the scalp using Ag-AgCl cup electrodes (10 mm diameter). One electrode was placed on the scalp overlying the supplementary motor area

305 (SMA, Cz). Another electrode was fixed over the contralateral sensorimotor cortex (C3). Positioning of the electrodes was based on the International 10-20 System (Jasper 1958). The active electrodes (Cz and C3) were referred to the common linked earlobes (A1 and A2). Impedance of each electrode was maintained below 5000 Ω. The EEG signal was amplified (×40,000) using EEG amplifiers (7P511, Grass Instrument Co., Quincy, MA). The time constant of the EEG recording was 2 s, with a low-pass cut-off frequency of 100 Hz. The output signal from the EEG amplifier was digitized (100 samples/s) using the Spike 2 system and recorded on the hard drive of the personal computer. EMG recording Surface EMG signals were simultaneously recorded from the biceps brachii and brachioradialis muscles. Bipolar electrodes (8 mm recording diameter) were attached to the skin overlying the belly of each muscle. The reference electrode was fixed on the skin overlying the lateral epicondyle near the elbow joint. The skin was cleaned with alcohol prior to electrode attachment. The EMG was amplified (×1000), filtered (10–3000 Hz), digitized (1000 samples/s), and recorded on the hard drive of the computer. Experimental procedures The study involved two experimental sessions. There were at least 3 days between the two sessions. In the first session, subjects performed elbow-flexion contractions at four submaximal force levels. In the second session, they performed elbow-flexion contractions at one force level (35% maximum) at three different rates (slow, intermediate, and fast). Thirty-five to 40 contractions were performed at each force level or rate. In each contraction, when the force reached 5% maximum level, a trigger signal was generated and stored on a separate channel of the data file containing EEG, EMG, and force signals. These trigger signals were later used for the triggered data averaging during data processing and analysis. In each session, the experiment began with two to three trials of maximal voluntary elbow-flexion contractions (MVC). The timing of the MVC task was based on a verbal count during which the subject increased the contraction force from zero to maximum in about 3 s and then maintained this force for another 3 s. The subject was verbally encouraged to maximize the force. A 2-min rest was provided between the MVC trials. The highest MVC force among the trials was used to calculate the submaximal forces. The four force levels performed during the first session were 10%, 35%, 60%, and 85% MVC. For each trial at each level, subjects increased the force to the target indicated by a horizontal cursor on an oscilloscope screen and then relaxed. They were trained to perform the four levels of contractions at the same rate (~200 N/s). A 5-min rest was provided between adjacent force levels. In the second session, subjects first practiced to increase the force to the target (35% MVC) at three distinct rates (slow, intermediate, and fast). (At the beginning, we attempted to perform the force at four distinct rates, but many subjects felt it was much easier to perform the force task at three rates.) Once they were able to perform the tasks, the slow-rate contractions were performed first, followed by the intermediate- and fast-rate contractions. When the fast-rate contractions were performed, the subjects sometimes overshot or undershot the target due to the difficulty of controlling the force at a high rate. However, the force fluctuation above or below the target was within 20% of the target force, and the mean force did not significantly deviate from the target. In both sessions, each trial was performed with subjects maintaining an open-eye, fixed-gaze condition to minimize the influence of eye movements on the EEG signal. Eye blinks were allowed during intertrial intervals during which the EEG signal was not used for the triggered averaging (see below). Each intertrial interval was about 5 s.

Data analysis Raw EEG data were inspected visually. Trials containing eye blinks or other signal artifacts were excluded. For each trial, the trigger signal triggered a 1.5-s window, 1 s before the trigger and 0.5 s after the trigger. The Spike 2 data analysis software performed signal averaging over the trials at each force level or rate. The amplitude of each averaged MRCP was measured from the beginning to the peak of a later component of the negative potential preceding onset of movement or muscle contraction. This component has been termed abrupt negative wave (Gilden et al. 1966) or negative slope by other investigators (Barrett et al. 1986; Deecke et al. 1976; Hallett 1994; Shibasaki et al. 1980). In many cases, it was difficult to clearly identify a single point indicating the beginning of the negative slope; however, the difficulty was eased by drawing a line with the negative slope and another line with the readiness potential and then taking the intercept of the two lines as the beginning of the negative slope (Fig. 2A). We chose to measure the negative slope because it is more directly related to planning and executing the voluntary movement (Gilden et al. 1966; Deecke et al. 1969). Our correlation data, presented in “Results,” support this assumption. In each trial, the EMG signal was rectified and averaged over 500 ms from the trigger point. This average EMG (on each trial) was then averaged again over all the trials performed at each force level or rate. The averaged EMG was expressed with absolute value (millivolts) and percentage of the average MVC EMG. The average force at each level or rate was also obtained by triggered averaging. Statistical analysis The amplitude of MRCP and average EMG at each force level or rate were statistically compared using one-way analysis of variance (ANOVA). The pairwise comparisons between adjacent force levels or rates were adjusted using Bonferroni’s correction. Thus P values less than 0.017 [α/no. of comparisons = (0.05/3) ≈ 0.017] are considered significant. Unless otherwise specified, the data are reported as means and standard deviations (means ± SD).

Results Relationship between MRCP and level of muscle activation Relationship between MRCP and force Subjects performed voluntary elbow-flexion contractions that resulted in 10%, 35%, 60%, and 85% MVC force. As the force increased, the amplitude of MRCP from the two recording locations (SMA and contralateral sensorimotor cortex) increased. (The amplitude of MRCP in this study was the measurement of negative slope indicated in “b” of Fig. 2A.) For the recording location overlying the SMA, MRCP values at the four force levels were 3.03±0.75 µV, 4.63±0.88 µV, 6.11±1.35 µV, and 8.75±2.17 µV, respectively. For the recording location overlying the sensorimotor cortex, MRCP values were 2.79±0.65 µV, 4.29±0.80 µV, 5.54±1.05 µV, and 7.56±1.76 µV, respectively. The difference in the MRCP value was significant (P<0.01) between every two adjacent force levels at both EEG recording locations. Figure 2B shows an example of MRCP data at four force levels along with the EMG and force measurements from one subject. Results of correlation analyses in-

306

Fig. 2 A Various components of MRCP (a readiness potential, b negative slope, c aftercontraction positivity). The shaded lines indicate how each component was defined. The vertical line indicates timing of the trigger, which occurred when the force reached 5% MVC. In this study, only b (negative slope) was measured. The MRCP recording shown in this figure was derived from EEG data recorded from the SMA site of a subject during contractions at 60% MVC. B Examples of MRCP (left), EMG (middle), and elbow-flexion force (right) at four levels of voluntary contraction (10%, 35%, 60%, and 85% MVC) of a subject. The MRCP was from the SMA recording site and the EMG was recorded from the biceps brachii muscle [the EMG was relatively low (8% MVC EMG) at the 10% force level but the EMG of the brachioradialis at the same force level was 17% of its MVC EMG (not shown)]. Each record depicts the average of about 35 trials. Vertical lines indicate timing of triggering (5% MVC force) and each arrow points to the beginning of the negative slope. C. Examples of MRCP (left), EMG (middle), and elbow-flexion force (right) at three rates (slow, intermediate, and fast) of force development of a subject. The MRCP was from the Cz electrode and the EMG was recorded from the biceps brachii muscle. Each record depicts the average of about 35 trials

dicated that there was a strong positive correlation between force and MRCP at both the SMA (r=0.95) and sensorimotor cortex (r=0.93) locations (Fig. 3). Relationship between MRCP and EMG The EMG signals of biceps brachii and brachioradialis muscles were compared with the MRCP values of both recording locations during the four levels of elbow-flexion contractions. As shown i0n Fig. 4, all four correlation analyses resulted in high correlation values. The values of r between MRCP at the SMA location and EMG of the biceps brachii (Fig. 4A) and brachioradialis (Fig. 4B) muscles were 0.85 and 0.84, respectively. The values of r between MRCP at the recording location of the sensorimotor

Fig. 3 Relationship between MRCP and force. A MRCP from the SMA location. B MRCP from the site of the sensorimotor cortex. Each symbol represents a subject (n=8) and shows the performance at each of four force levels. MRCP was highly correlated with elbow-flexion force at both recording locations

307 Fig. 4 Relationship between MRCP and muscle EMG across four levels of force. A MRCP from the SMA site compared with the EMG recorded from the biceps brachii muscle; B MRCP from the SMA site compared with the brachioradialis EMG; C MRCP from the motor cortex site compared with the EMG of the biceps brachii; D MRCP from the motor cortex site compared with the brachioradialis EMG. Data were recorded at the four force levels, but the EMG data at each force level are expressed as the actual percentage of MVC EMG. For example, the filled circle on the right in C shows an EMG value of about 45%, but the EMG was recorded during contractions of 85% MVC. Each symbol represents a subject. At each EEG or EMG location, the MRCP and EMG were highly correlated

cortex and EMG of biceps brachii (Fig. 4C) and brachioradialis (Fig. 4D) muscles were 0.84 and 0.81, respectively. Each symbol in each plot of Fig. 4 represents a subject, and a line connected the four values at the four force levels. For EMG, we normalized the average EMG at each force level to the average MVC EMG. In many subjects, the percentage EMG was substantially lower than the corresponding force. For example, the subject represented by the filled circle had an EMG of about 45% at 85% MVC force.

Duration of negative slope The time from beginning to the peak of the negative slope was measured at each level of force (shown in Fig. 2B, from black arrow to the peak). The duration at 10%, 35%, 60%, and 85% levels was 135±22 ms, 212±34 ms, 274±42 ms, and 502±62 ms, respectively. The magnitudes of the MRCP and the duration were highly correlated (r=0.91).

Relationship between MRCP and rate of force development Relationship between MRCP and rate of force increase Elbow-flexion force was increased to the target level of 35% MVC at three rates: slow (48.9±47.8 N/s), intermediate (133.3±97.9 N/s), and fast (403.3±231.5 N/s). At the SMA recording location, the MRCP values at the three rates (slow to fast) were 2.79±1.10 µV, 4.76±1.52 µV, and

6.50±1.86 µV, respectively (Fig. 5A). At the recording site of the sensorimotor cortex, the MRCP values were 2.42±0.92 µV, 4.33±1.53 µV, and 5.86±1.91 µV, respectively (Fig. 5B). The difference in the MRCP value was significant (P<0.01) between every two adjacent rates at both EEG recording locations. Figure 5 shows MRCP data points of individual subjects at each rate of force development. The correlation between MRCP and rate of force increase was r=0.84 at the SMA recording location and r=0.85 at the sensorimotor cortex recording site. Figure 2C is a single-subject example of MRCP (left), EMG (middle), and force (right) at three different rates. The correlation between EMG and rate of force increase was r=0.85 for the biceps brachii muscle and r=0.87 for the brachioradialis muscle. MRCP and EMG at three rates of force development The EMG signals of biceps brachii and brachioradialis muscles at three force rates were compared with the MRCP values at both recording locations. Both MRCP and EMG increased as the rate of force development increased (Fig. 6). The values of r between MRCP at the SMA recording location and EMG of biceps brachii (Fig. 6A) and brachioradialis (Fig. 6B) muscles were 0.85 and 0.86, respectively. The values of r between MRCP at the recording site of the motor cortex and EMG of biceps brachii (Fig. 6C) and brachioradialis (Fig. 6D) muscles were 0.71 and 0.75, respectively. Each symbol in each plot of Fig. 6 represents a subject and the three values at the three-force rate were connected by a line (left, low rate; middle, intermediate rate; right, high rate).

308

Duration of negative slope The duration from the beginning to the peak of the negative slope was measured at each rate of force development (Fig. 2C, from black arrow to the peak). The duration at slow, intermediate, and fast rates was 132±32 ms, 283±33 ms, and 380±34 ms, respectively. The correlation value between the measurements (MRCP and duration) was r=0.95.

Discussion The purpose of this study was to determine the relationship between MRCP and voluntary muscle activation. The major findings were that (1) MRCP (negative slope) was highly correlated with joint force and muscle EMG, and (2) MRCP was highly correlated with the rate of rise of force and associated muscle EMG. These data suggest that the MRCP measurement represents a cortical signal that controls voluntary muscle activation. Relationship between MRCP and level of muscle activation

Fig. 5 Relationship between MRCP and rate of force development. A MRCP from the SMA location. B MRCP from the site of the sensorimotor cortex. Each symbol represents a subject. MRCP was highly correlated with rate of force development at both EEG recording locations

Fig. 6 Relationship between MRCP and muscle EMG across three rates of force development. A MRCP from the SMA site compared with the EMG recorded from the biceps brachii muscle; B MRCP from the SMA site compared to the brachioradialis EMG; C MRCP from the motor cortex site was compared with the EMG of the biceps brachii; D MRCP from the motor cortex site compared with the brachioradialis EMG. Each symbol represents a subject. Data were recorded at the three rates of force development, but the EMG data at each rate are expressed as the actual percentage of MVC EMG. Thus, the three data points of each symbol (left, middle, right) represent EMG data recorded during slow-rate, intermediaterate, and fast-rate contractions, respectively. At each EEG or EMG location, the MRCP and EMG were highly correlated

The greater the force attempted, the higher the amplitude of MRCP from electrodes overlying two cortical areas: the SMA and the contralateral primary sensorimotor cortex. These cortical motor fields directly participate in planning and executing voluntary movements (for review see Ghez 1991a, 1991b). The results suggest that when stronger muscle output is planned, the motor cortex, SMA, and other cortical motor areas increase their

309

activity, and that is reflected in the MRCP signal. Although the underlying cortical field may not exclusively contribute to the MRCP signal from each electrode, it is reasonable to postulate that the field dominates the signal, given its known function in movement control and its proximity to the electrode. A more definite conclusion concerning the underlying signal source, however, cannot be drawn until multichannel, high-density EEG recordings and/or neuroimaging data acquired under the same experimental conditions are available. A stronger cortical signal for higher muscle force must result from participation of a larger number of cortical motoneurons and/or a higher discharge rate of the recruited neurons (Nunez 1981), assuming the degree of synchronization does not change significantly between the levels of muscle contraction. The discharge rate of cortical motoneurons has been reported to increase as static force increases (Cheney and Fetz 1980; Evarts 1968; Smith et al. 1975). Hepp-Reymond and co-workers (1989) observed a linear relationship between the discharge rate of motor cortex neurons and monkey finger pinch force. Data from neuroimaging studies have demonstrated that higher muscle force is associated with stronger brain activation. Dettmers et al. (1995) reported a high correlation between regional cerebral blood flow (rCBF) measured by positron emission tomography in a number of cortical motor areas and voluntary force. A recent functional MRI (fMRI) study (Dai et al. 1999) reported that in many motor cortical fields the fMRI signal and human joint force are positively correlated. Because increases in rCBF and fMRI signal reflect increases in synaptic activity (Jueptner and Weiller 1995; Raichle 1987), an elevated fMRI or rCBF signal at higher force levels reflects greater activity of motor cortical neurons, perhaps resulting from recruiting a larger number of output neurons and/or discharging at higher rates of the neurons being already active. Relationship between MRCP and rate of force development The higher the rate of rise of force, the greater was the MRCP amplitude from both recording locations (Fig. 5). Many neurons in monkey motor cortex discharge at a higher rate when the rate of force development is high (Hepp-Reymond et al. 1978; Smith et al. 1975). The higher discharge rate should contribute to a greater MRCP. In muscle, increases in the rate of force development or the speed of movement are associated with stronger EMG signals (Fig. 6). When muscles contract very rapidly, fast-twitch motor units or muscles can be selectively recruited (Smith et al. 1980) or recruited simultaneously with slow-twitch motor units (Burke 1981). Because the EMG signal reflects the signals of descending command and sensory feedback, greater EMG levels during fast contractions must at least be partially contributed by stronger cortical neuron activity.

Magnetic cortical stimulation during dynamic (fast) contractions evokes a greater muscle response than during steady (slow) contractions with similar background EMG activity (Aranyi et al. 1988). Motor unit recruitment threshold was significantly lowered and motor unit discharge rate was substantially higher during fast isometric contractions (Desmedt and Godaux 1977; Masakado et al. 1995). These findings indicate that during fast muscle contractions the recruitment threshold of spinal motoneurons is lowered resulting in the recruitment of a larger number of motor units and a higher discharge rate that contribute to a greater EMG. It is also possible that the cortical motoneuron excitability is higher (recruitment threshold is lower) during fast contractions so that a larger number of cortical output neurons are recruited and discharge at higher rates, leading to greater MRCP and EMG. The high correlation value between the MRCP and EMG at different rates of force development (Fig. 6) indicates that this may be the case. Comparisons of results of the present study with data in the literature MRCP and force Few studies have investigated the question of whether higher voluntary force is accompanied by greater MRCP, and the results are mixed. Kutas and Donchin (1974) recorded MRCP during three levels of hand-squeezing exercise (~25%, 50%, and 75% maximal grip force) from scalp electrodes overlying several cortical locations (CZ, C3, and C4). The component of MRCP measured was readiness potential (termed N1). It was found that, at both C3 and C4, a higher amplitude of MRCP was associated with the higher grip force of the contralateral hand. However, no correlation studies were performed, and results from the Cz electrode were not reported. Kristeva et al. (1990) compared MRCP signals among index finger flexion movements against three levels of inertial loads: 0 g, 250 g, and 400 g. A significant difference in these measurements was observed only between conditions of 0 g load and 250 g load, or between 0 g load and 400 g load. No difference in MRCP was seen between conditions of 250 g load and 400 g load. Perhaps a 150-g difference between the two load levels was too small to have an effect on the MRCP. Average index finger flexion strength is well beyond 8000 g (Ejeskar and Ortengren 1981) on average; 150 g was less than 2% of the maximal load. (In our study, the difference between the two adjacent force levels was 25%.) Shibata et al. (1993) reported a significant increase in motor potential (MP), one component of the MRCP that consists of part of the negative slope (Hallett 1994) at the Cz location when subjects increased isometric elbow flexion force from 10% to 50% MVC. Nishihira et al. (1989) found that the magnitude of MRCP (MP) at Cz and C3 or C4 was greater when the subjects applied about 22 kg handgrip force than when the applied force

310

was 12 kg. Becker and Kristeva (1980) had their subjects perform two levels of finger-squeezing exercise, one with higher force and the other with lower force. MRCP was greater at Cz and C3 when the subjects’ finger force was high than when the force was low. Similarly, the readiness potential was greater for a high-force index finger flexion task than for the low-force task (Hink et al. 1983). All four studies mentioned in this paragraph examined MRCP at only two force levels. Two studies did not find an association between voluntary force and MRCP. Wilke and Lansing (1973) reported that the pre-movement negative potential from the C3 or C4 electrode was the same between two elbowextension movements against different loads. Perhaps the difference in load between the two tasks was too small (1 lb vs 3 lb) compared to the total weight that the elbow flexor muscles of college-age subjects can remove (usually >50 lb); thus, the difference in weight was less than 5% of the maximum weight that subjects could move. Hazemann et al. (1989) compared MRCP from the C3 or C4 electrode between two force levels of finger extension (about 200 g vs about 500 g). There was no significant difference in the amplitude of MRCP between the two force conditions. Again, the reason for this may be that the difference in force was too small (assuming that, on average, the maximal strength of the subjects was 5000 g, a difference of 300 g is only 6% of the maximal force). MRCP and rate of force development To our knowledge, only two studies have addressed this issue. In one report (Becker et al. 1976), higher MRCP (readiness potential) was associated with slower arm movement (10 cm/s vs 110 cm/s for fast movement) from an electrode (C3’ or C4’) positioned 10 mm anterior to C3 or C4. In another (Hazemann et al. 1989), subjects generated 500-g index finger extension force in 160 ms (high rate) and in 660 ms (low rate). MRCP was measured from the C3 or C4 electrode, depending on the hand used. No effect of rate of force development on the amplitude of MRCP was detected. One possible explanation is that the component of MRCP measured in these studies did not reflect cortical neural activity related directly to the planning and execution of the muscle contraction. Both studies measured the amplitude from EEG baseline to the peak or before peak of the negative potential. If the two studies measured the same component, as we did in the present study (negative slope, component “b” in Fig. 2), the results may have been different. Conclusions This study systematically examined the relationship between MRCP and voluntary muscle activation. The negative slope of the MRCP was highly correlated with the level of voluntary joint force, the rate of force develop-

ment, and the associated EMG. These results suggest that the negative slope, a major component of MRCP, represents descending signals from cortical motor centers that directly control voluntary muscle activation. The majority of studies in the literature that have addressed these issues tested MRCP at only two force levels, making it impossible to perform any meaningful correlation analysis. For the studies that found no effect of voluntary force on the MRCP, the difference in force or resistance between the tested levels was seemingly too small. The two studies examining MRCP and rate of force development failed to show a dependency between the two variables, possibly because they did not measure the most appropriate component of the MRCP. Acknowledgements This study was supported by NIH grants (NS 35130, NS 37400) to G.H.Y. and by the departmental research funds of Physical Medicine and Rehabilitation at the Cleveland Clinic Foundation. We thank the anonymous reviewers, whose comments improved the manuscript.

References Aranyi Z, Mathis J, Hess CW, Rosler KM (1988) Task-dependent facilitation of motor evoked potentials during dynamic and steady muscle contractions. Muscle Nerve 21:1309–1316 Barrett G, Shibasaki H, Neshige R (1986) Cortical potentials preceding voluntary movement: evidence for three periods of preparation in man. Electroencephalogr Clin Neurophysiol 63:327–339 Bates AV (1951) Electrical activity of the cortex accompanying movement. J Physiol (Lond) 113:240–254 Becker W, Kristeva R (1980) Cerebral potentials prior to various force deployments. Prog Brain Res 54:189–194 Becker W, Iwase K, Jürgens R, Kornhuber HH (1976) Bereitschaftspotential preceding voluntary slow and rapid hand movement. In: McCallum WC, Knott JR (eds) The responsive brain. Wright and Son, Bristol, UK, pp 99–102 Burke RE (1981) Motor units: anatomy, physiology, and functional organization. In: Brooks VB (ed) Handbook of physiology, sect 1. The nervous system, vol II. American Physiological Society, Bethesda, MD, pp 390–405 Cheney PD, Fetz EE (1980) Functional classes of primate corticomotoneuronal cells and their relation to active force. J Neurophysiol 44:773–791 Close RI (1972) Dynamic properties of mammalian skeletal muscles. Physiol Rev 52:129–197 Dai TH, Liu JZ, Siemionow V, Ng TC, Yue GH (1999) Simultaneous measurements of human functional MRI, muscle force and EMG. Soc Neurosci Abstr 25:1915 Deecke L, Scheid P, Kornhuber HH (1969) Distribution of readiness potential, pre-motion positivity, and motor potential of the human cerebral cortex preceding voluntary finger movements. Exp Brain Res 7:158–168 Deecke L, Grozinger B, Kornhuber HH (1976) Voluntary finger movement in man: cerebral potentials and theory. Biol Cybern 23:99–119 Desmedt JE, Godaux E (1977) Ballistic contractions in man: characteristic recruitment pattern of single motor units of the tibialis anterior muscle. J Physiol (Lond) 264:673–693 Dettmers C, Fink GR, Lemon RN, Stephan KM, Passingham RE, Silbersweig D, Holmes A, Ridding MC, Brooks DJ, Frackowiak RS (1995) Relation between cerebral activity and force in the motor areas of the human brain. J Neurophysiol 74:802–815 Ejeskar A, Ortengren R (1981) Isolated finger flexion force – a methodological study. Hand 13:223–230

311 Evarts EV (1968) Relation of pyramidal tract activity to force exerted during voluntary movement. J Neurophysiol 31:14–27 Ghez G (1991a) The control of movement. In: Kandel ER, Schwartz JH, Jessell TM (eds) Principles of neuroscience. Elsevier, New York, pp 533–547 Ghez G (1991b) Voluntary movement. In: Kandel ER, Schwartz JH, Jessell TM (eds) Principles of neuroscience. Elsevier, New York, pp 609–625 Gilden L, Vaughan HG Jr, Costa LD (1966) Summated human EEG potentials with voluntary movement. Electroencephalogr Clin Neurophysiol 20:433–438 Hallett M (1994) Movement-related cortical potentials. Electromyogr Clin Neurophysiol 34:5–13 Hazemann P, Metral S, Lille F (1989) Influence of force, speed and duration of isometric contraction upon slow cortical potentials in man. In: Otto DA (ed) Multidisciplinary perspectives in event-related brain potential research. U.S. Government Printing Office, Washington, DC, pp 107–111 Henneman E, Mendell LM (1981) Functional organization of motoneuron pool and its input. In: Brookhart JM, Mountcastle VB (eds) Handbook of physiology, sect 1. The nervous system, vol II. Motor control, part 1. American Physiological Society, Bethesda, MD, pp 423–507 Hepp-Reymond MC, Wyss UR, Anner R (1978) Neuronal coding of static force in primate motor cortex. J Physiol (Paris) 74: 287–291 Hepp-Reymond MC, Wannier TMJ, Maier MA, Rufener EA (1989) Sensorimotor cortical control of isometric force in the monkey. In: Allum JHJ, Hulliger M (eds) Progress in brain research, vol 80. Elsevier, New York, pp 451–463 Hink RF, Deecke L, Kornhuber HH (1983) Force uncertainty of voluntary movement and human movement-related potentials. Biol Psychol 16:197–210 Jasper HH (1958) Report of the Committee on Methods of Clinical Examination in Electroencephalography. Appendix: the ten twenty electrode system of the International Federation. Electroencephalogr Clin Neurophysiol 10:370–375 Jueptner M, Weiller C (1995) Review: does measurement of regional cerebral blood flow reflect synaptic activity? Implications for PET and fMRI. Neuroimage 2:148–156 Kornhuber HH, Deecke L (1965) Hirnpotential Änderungen bei Willkürbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pflugers Arch Ges Physiol 284:1–17 Kristeva R, Cheyne D, Lang W, Lindinger G, Deecke L (1990) Movement-related potentials accompanying unilateral and bilateral finger movements with different inertial loads. Electroencephalogr Clin Neurophysiol 75:410–418

Kutas M, Donchin E (1974) Studies of squeezing: handedness, responding hand, response force, and asymmetry of readiness potential. Science 186:545–548 Masakado Y, Akaboshi K, Nagata M, Kimura A, Chino N (1995) Motor unit firing behavior in slow and fast contractions of the first dorsal interosseous muscle of healthy men. Electroencephalogr Clin Neurophysiol 97:290–295 Milner-Brown HS, Stein RB, Yemm R (1973) The orderly recruitment of human motor units during voluntary isometric contractions. J Physiol (Lond) 230:359–370 Nishihira Y, Fujita T, Araki H (1978) Movement-associated cortical potential preceding voluntary movement. Jpn J Phys Fitness Sports Med 27:140–148 Nishihira Y, Araki H, Ishihara A (1989) Cerebral motor potential preceding grip strength movement. J Sports Med Phys Fitness 29:297–303 Nunez PL (1981) Electric fields of the brain: the neurophysics of EEG. Oxford University Press, New York, pp 101–105 Raichle ME (1987) Circulatory and metabolic correlates of brain function in normal humans. In: Plum F (ed) Handbook of physiology, sect 1. The nervous system. Higher function of the brain, vol I. American Physiological Society, Washington, DC, pp 643–674 Shibasaki H, Barrett G, Halliday E, Halliday AM (1980) Components of the movement-related cortical potential and their scalp topography. Electroencephalogr Clin Neurophysiol 49: 213–226 Shibata M, Moritani T, Kubota K (1993) The relationships between movement-related cortical potentials and the force level in man. Med Sci Sports Exerc [Suppl] 25:197 Shibata M, Oda S, Moritani T (1997) The relationships between movement-related cortical potentials and motor unit activity during muscle contraction. J Electromyogr Kinesiol 7:79–85 Smith AM, Hepp-Reymond MC, Wyss UR (1975) Relation of activity in precentral cortical neurons to force and rate of force change during isometric contractions of finger muscles. Exp Brain Res 23:315–332 Smith JL, Betts B, Edgerton VR, Zernicke RF (1980) Rapid ankle extension during paw shakes: selective recruitment of fast ankle extensors. J Neurophysiol 43:612–620 Wilkie DR (1950) The relation between force and velocity in human muscle. J Physiol (Lond) 110:249–280 Wilke JT, Lansing RW (1973) Variations in the motor potential with force exerted during voluntary arm movements in man. Electroencephalogr Clin Neurophysiol 35:259–265