Evidence of a preprogrammed deactivation of the hamstring muscles for triggering rapid changes of posture in humans

Evidence of a preprogrammed deactivation of the hamstring muscles for triggering rapid changes of posture in humans

Electroencephalography and clinical Neurophysiology 105 (1997) 58–71 Evidence of a preprogrammed deactivation of the hamstring muscles for triggering...

924KB Sizes 1 Downloads 8 Views

Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

Evidence of a preprogrammed deactivation of the hamstring muscles for triggering rapid changes of posture in humans G. Cheron a ,*, A. Bengoetxea a, T. Pozzo b, M. Bourgeois a, J.P. Draye c a

Laboratory of Biomechanics, ISEPK, Universite´ Libre de Bruxelles, Avenue P. He´ger, CP168, 1050 Brussels, Belgium b Groupe d’Analyse du Mouvement, Universite´ de Bourgogne, Dijon, France c Parallel Information Processing Laboratory of the Faculte´ Polytechnique de Mons, Mons, Belgium Accepted for publication: 17 October 1996

Abstract Normal subjects were asked to make rapid flexions of the legs from a stationary initial standing posture in a self-paced mode. Because this movement implicates a rapid change in posture, questions were asked about the type of central command which must include the rupture of the erect posture and the accomplishment of the goal directed movement. Movements of the different segments of the body were recorded and analyzed using the optoelectronic ELITE system. Electromyographic (EMG) activities of 8 muscles of the lower limb on one side were recorded, rectified and integrated. The time relationships of the different EMG signals (activation or deactivation) were analyzed with respect to selected kinetic measures of the related segments of the body. In the majority of the subjects, before the movement onset, EMG events included a specific deactivation of the tonic EMG activity of the semimembranous (SM) and semitendinous (ST) muscles (time onset relative to the onset of the legs flexion: −196.9 ± 96.4 ms and −180.5 ± 89.7 ms, respectively). A second event was a phasic activation of the tibialis anterior (TA) muscle (time onset: −60.5 ± 117.6 ms). Conjugate cross-correlation analysis of these EMG signals demonstrated the existence of a common coordinated strategy between the deactivation of the hamstring and the TA activation. Even though a small horizontal displacement of the head was recorded prior to leg movement, it occurred too late to induce deactivation of the hamstring muscles. These results demonstrate that for rapid legs flexion, where the gravity forces are the main source of joint angle acceleration, the deactivation of the SM and ST muscles acts in conjunction with the phasic activation of the TA. The preprogrammed deactivation of the SM and ST muscles represents the early phase of the central command to switch from the standing to the squatting posture.  1997 Elsevier Science Ireland Ltd. All rights reserved Keywords: Posture; Whole-body movement; Electromyography

1. Introduction The majority of investigations that have focused on the coordination between movement and posture have confined experiments into 2 separate fields. The first one is represented by the classical ‘unexpected postural perturbation approach’ which has revealed the existence of a limited number of postural strategies (Nashner, 1977; Horak and Nashner, 1986). The second approach concerns the analysis of the anticipatory postural adjustments resulting from the voluntary displacement of a single body segment. Whatever the body segment and the type of movement used, for example a rapid upward arm movement * Corresponding author. Tel.: +32 2 6502187; fax: +32 2 6503745.

(Belen’kii et al., 1967; Lee, 1980; Bouisset and Zattara, 1981; Friedli et al., 1984; Horak et al., 1984; Brown and Frank, 1987; Lee et al., 1987; Zattara and Bouisset, 1988; Ramos and Stark, 1990a), a manipulative movement (Cordo and Nashner, 1982), a bimanual unloading task (Gahery and Massion, 1981; Hugon et al., 1982; Dufosse et al., 1985; Paulignan et al., 1989; Lum et al., 1992), a unilateral leg movement (Rogers and Pai, 1990; Rogers, 1992; Mouchino et al., 1992), or a fast forward bending movement (Crenna et al., 1987; Ramos and Stark, 1990b), electromyographic (EMG) studies have revealed the presence of postural responses that anticipate voluntary activity. These anticipatory activities seem to be an integral part of a particular program which prevents destabilization of the center of gravity due to limb or trunk movements

0924-980X/97/$17.00  1997 Elsevier Science Ireland Ltd. All rights reserved PII S0921-884X(96)9654 4-3

EEM 96544

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

(Massion, 1992). However, most of the natural movements of human necessitate rapid changes of whole body posture which are definitely different from the 2 main types of classical investigations relative to the reactive and anticipative motor process. Consequently, the purpose of the present study was to assess the motor strategies used to perform in a self-paced mode a rapid whole body movement inducing a complete change of posture. The central question is how does the nervous system organize the fast transition between 2 different positions of the body? Anticipatory postural adjustments have been demonstrated in whole body movements, such as rapid rising up onto tip-toe (Nardone and Schieppati, 1988; Yamashita et al., 1990; Kasai and Kawai, 1994). In this situation, the muscle actuators responsible for postural control were the same as those used to perform the destabilization task. Consequently, these latter anticipatory responses differ from the classical anticipatory postural adjustment because they were recorded from prime mover muscles of the goal directed task (standing on the toes). It has been demonstrated that the task of going up onto tip-toe is performed in 2 steps. The first step, initiated by a phasic burst of the tibialis anterior (TA) muscle, corresponds to a postural adjustment, while the second step, the movement itself, is performed mainly by the soleus (SOL) muscle (Nardone and Schieppati, 1988; Diener et al., 1993; Kasai and Kawai, 1994). Recently, Crenna and Frigo (1991) have proposed the existence of a common motor program for the initiation of various fast forward-oriented voluntary movements, such as initiation of gait, rising on tip-toe, fast forward bending of the trunk, standing up, and forward throwing with an upper limb. They demonstrated that the key sequence of the motor program is an early inhibition of the tonic activity of the SOL muscle followed by an activation of the TA muscle. These authors considered this sequence as a particular type of anticipatory postural synergy exerting a direct action on the position of the center of foot pressure allowing an appropriate contraction of the prime mover. However, for all of these forward movements the dissociation between the anticipative postural adjustment and the real prime mover action is difficult. The fact that the early sequence began with an inhibition of the tonic activity of the SOL starting around 90 ms before the TA burst could indicate the existence of a mechanism similar to the Hufschmidt phenomenon described for upper limb movements (Hufschmidt and Hufschmidt, 1954; Gottlieb et al., 1970; Hallet et al., 1975; Evarts, 1974; Flanders et al., 1994; Hoffman and Strick, 1995). This phenomenon is characterized by an antagonist inhibition as the earliest sign of a sensorymotor reaction task. The prompt initiation of movement depends not only on a brisk activation of the agonist but also on a previous interruption of posture controlled by the antagonist muscles. In order to analyze this phenomenon in a whole body movement configuration, a movement

59

which necessitates a clear rupture of the preceding posture has been chosen. Two particular points have been addressed: (i) Does the Hufschmidt phenomenon take place during an unloading task such as that elicited by a voluntary downward flexion of the leg in the standing posture? (ii) At which joint of the erect standing posture is the postural rupture programmed? 2. Methods 2.1. Subjects Informed consent was received from 11 healthy subjects (9 females, 2 males) aged between 21 and 39 years. All subjects were naive and did not receive any practice before the recordings were taken. 2.2. Description of the task Subjects stood quietly with their arms horizontally extended. Following a verbal instruction, the subject was asked to perform, as fast as possible, a complete flexion of the lower limb and to maintain the final flexed position. In order not to interfere with the initiation of the voluntary movement by some reaction time requirements, the subject was instructed to decide when he/she would produce the movement. Three different instructions were given: 1. To perform a complete flexion of the lower limbs, without requirements concerning the control of the final position, 2. To perform a complete flexion of the lower limbs producing a voluntary contraction in order to increase the rigidity of the flexed posture at the end of the movement, and 3. To perform a flexion of the legs to a mid-distance between full extension (initial position) and full flexion (final position of instructions 1 and 2). In each condition, 3 trials were performed and the fastest movement was later used for a quantitative analysis. 2.3. Movement recording and measurement equipment 2.3.1. Kinematics Movements were recorded and analyzed using the optoelectronic ELITE system (Ferrigno and Pedotti, 1985). This system consists of 2 CCD-cameras detecting retroreflective markers using a sampling rate of 100 Hz. The cameras were placed 4 m from the subject. Markers consisted of plastic spheres of 8 mm of diameter covered with reflecting material. Accuracy was 1/3000 of the working field. As the field used was 2 × 2 m, accuracy was 0.67 mm. Marker images recorded by the cameras were processed for real time shape recognition. Image coordinates of the marker centroids were reconstructed in 3 dimensions. The position in space of 11 passive markers, includ-

60

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

ing 9 links, were recorded. The markers were fastened onto the skin at the following points (Fig. 1): 1. The head: on the nose at the horizontal extent of the lower border of the orbit and on the meatus of the ear. The line of these 2 markers provide the ‘Frankfort plane’ (FP) (DeBeer, 1947), 2. The neck: at the level of C6 on the posterior border of the sterno-cleido mastoidus, 3. The upper limb: at the shoulder, on the acromial process and the lateral condyle of the elbow and wrist, 4. The trunk: on the tubercle of the antero-superior iliac crest and on the trochanter 5. The lower limb: on the lateral condyle of the knee, the lateral malleolus and on the 5th metatarsal. Numerical values of several parameters were calculated along the sagittal plane of the body before and during the flexion of the lower limb: (1) the hip angular position and

velocity, (2) the knee angular position and velocity, (3) the ankle angular position and velocity, (4) the angle of the FK with the earth vertical, and (5) the horizontal and vertical linear head accelerations. Velocity and acceleration signals were obtained by digitally differentiating position signals. Digital filtering was performed by means of a finite impulse response filter, in order to compute the velocity by a zero phase-shift differentiation (D’Amigo and Ferrigno, 1990). Supplementary experiments were performed in which the initial stance conditions were voluntarily changed. Extreme forward and backward leaning were subjectively determined by the subject and stabilized prior to the movement. 2.3.2. Electromyography Surface EMG patterns of 8 ipsilateral leg muscles were recorded using telemetry (Telemg, BTS). Muscle activity was recorded using pairs of silver-silver chloride surface electrodes spaced 2.5 cm apart on the following muscles: soleus (SOL), tibialis anterior (TA), biceps femoris (BF), semitendinous (ST), semimembranous (SM), vastus medialis (VM), vastus lateralis (VL) and rectus femoris (RF). Particular attention on cross-talk problem between those muscles showed it to be minimal in the current setting. In order to better describe the segmental distribution of the deactivation of tonic activity of the hip extensors and to test a possible contribution (by means of reciprocal inhibition) of an activation of the hip flexors or of trunk and neck muscle action, the EMG of the gluteus maximus (GM), the sartorius (SA), the tensor fasciae latae (TFL), the erector spinae at lumbar (ES), and the splenius capitis (SPL) muscles were recorded in 2 subjects. 2.4. Data analysis

Fig. 1. Experimental setup and initial posture of the subject. The locations of the 11 reflective markers recorded by the two cameras of the ELITE system are indicated by dots and associated numeration. Subjects were instructed to flex their legs as fast as possible and to reach and maintain a final squatting position. Electromyographic activity was recorded by surface electrodes. The approximate location of the 8 recorded muscles are indicated in the diagram. Abbreviations: BF, biceps femoris; ST, semitendinous; SM semimembranous; RF, rectus femoris; VM, vastus medialis; VL, vastus lateralis; SOL, soleus; TA, tibialis anterior. The directions of the three reference axes (X,Y,Z) are indicated in right part of the figure.

2.4.1. EMG signals Raw EMG signals (differential detection) were amplified 1000 times and bandpass filtered (10–2000 Hz). Thereafter, EMGs were digitized at 2 kHz, full-wave rectified and smoothed by means of a third order averaging filter with a time constant of 10 ms (Hof and Van Den Berg, 1981). Before the onset of movement, mean amplitude of tonic activity was measured on the rectified EMG signal for an interval of 500 ms before the onset of the silent period. The silent period onset was measured by visual inspection of the EMG records on the computer screen and was considered as the point from which the EMG amplitude definitely decreased. In order to test the temporal coordination between the different EMG patterns (activation and deactivation) preceding the onset of movement, conjugate cross-correlation functions (CCF) between different pairs of EMG signals (SM-ST, SM-TA, ST-TA, SM-SOL, ST-SOL and SOLTA) were calculated. The span of time lags or leads analyzed ranged from -300 ms to +300 ms (T = 600 ms) and

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

61

was centred on the onset of the silent period recorded on the hamstring muscles. The CCF between 2 functions f(t) and g(t) (both of which are measured during a total time T) is defined as: CCFfg (t) =

1 T

…T

f (t) g(t − t)dt

0

where t is the lag between both functions. The CCFs were computed using the time series analysis software Statistica. More details of the application of this statistical method to physiological data are available in Amblard et al. (1994) and Miller et al. (1993). 2.4.2. Analysis of movement dynamics For one representative movement dynamic torques of a four-link model including the upper limb, trunk, thigh and leg have been studied. This analysis was made in the sagittal plane and consisted of Lagrangian analysis (Onyshko and Winter, 1980; Ramos and Stark, 1990a; Ramos and Stark, 1990b). Briefly, in order to determine the variation in muscular torques applied to the different joints implicated in squatting, the kinematic data provided by the ELITE system (angles, angular velocities and angular acceleration) were introduced into the Lagrangian dynamic model. Assuming that at least one part of the body (the foot) is stationary in an inertial reference frame, and all joints are rotational and frictionless, dynamic equations of the multiple-link inverted pendulum can be written as:   dT dV d dT ÿ _ ‡ dv ˆ Q i dt dv2 dvi i where T and V are the kinetic energy and potential energy ¯ i is the generalized torque of the system, respectively. Q applied on each joint i. This generalized torque is dissociated into 2 components, one related to the total muscular torque (TMT) and one related to the gravitational torque (GT). In this model, TMT represents the resulting combination of agonist and antagonist muscular torque acting on the joint. The length of each of the four-links were given by experimental ELITE data. The mass-matrix was estimated using a method proposed by Clauser et al. (1969). The software of the present model is described in detail in Bourgeois (1991). 3. Results 3.1. General characteristics of the flexion movement The instructions given to the subjects were well respected and the ‘as fast as possible’ requirement produced very brief movements. For example, the mean duration of all the flexion movements of the knee was 727 ± 115 ms. The task of rapidly flexing the legs from the standing posture implicates the coordinated control of at least 3

Fig. 2. Characteristics of the angular position and velocity of the ‘Frankfort plane’ (FK) (A and E), the hip (B and F), the knee (C and G) and the ankle (D and H) before and during the rapid flexion of the legs in two representative subjects. The movement represented in A–D and in E–H were executed following instructions 1 and 3, respectively. Zero time indicates onset of flexion movement.

main joints: the hip, knee and ankle. Fig. 2 illustrates in 2 representative subjects the main sequence of the movement performed following instruction 1 (A–D) and 3 (E– H) and represented by the angular variation and angular velocity of the hip (B,F), knee (C,G), and ankle (D,H). The top of this figure (A and E) illustrates the small deviation of the FK during this ample body movement. For the 11 subjects analyzed in this study, the mean angular displacement of the angle formed by this plane with the earth vertical was 9.8 ± 3.8 deg. The mean angular velocity of this joint was 29.0 ± 14.4 deg/s (Fig. 2A). The mean angular displacement of the hip, regarded as the angle between the line joining the iliac crest and the trochanter, and the line joining the trochanter and the knee, was 54.9 ± 12.5 deg. The mean angular velocity of this joint was 180.0 ± 44.1 deg/s (Fig. 2B). The mean angular displacement of the knee was 105.7 ± 11.0 deg, with the mean angular velocity reaching 340.3 ± 44.8 deg/s (Fig. 2C). The mean angular displacement of the ankle was

62

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

Table 1 Main characteristics of different joint motions (mean and SD) during the rapid flexion of the legs in 11 subjects following instructions 1–3 Instruction Instruction 1 Mean SD Instruction 2 Mean SD Instruction 3 Mean SD

Am. FK (deg)

V. FK (deg/s)

Am. H. (deg)

V. H. (deg/s)

Am. K. (deg)

V. K. (deg/s)

Duration K (ms)

Am. A. (deg)

V. A. (deg/s)

12.62 4.39

34.31 12.67

76.06 11.65

220.64 34.01

140.03 6.64

347.33 43.36

723.00 83.43

19.78 6.77

77.94 37.01

11.23 2.99

33.94 11.06

75.58 14.99

231.25 39.99

132.84 7.40

348.98 53.37

783.00 187.03

19.33 6.23

75.68 24.06

9.79 3.77

29.00 14.36

54.90 12.54

180.02 44.10

105.70 11.04

340.31 44.84

660.00 140.43

20.71 8.09

81.12 37.40

Abbreviations: Am, angular amplitude; V, angular velocity; FK, Frankfort plane; H, hip; K, knee; A, ankle.

20.7 ± 8.1 deg, and the mean of the maximal angular velocity of this joint was 81.1 ± 37.4 deg/s (Fig. 2D) This small extent of ankle motion may have been due to the fact that leg flexion was accompanied by an elevation of the heel reducing the dorsi-flexion of the ankle. Fig. 2 also illustrates a nearly perfect synchronization between the angular velocity peaks of the hip (Fig. 2B) and knee (Fig. 2C). However, the velocity profiles of the ankle (Fig. 2D) were more complex than that of the knee. The former were subserved by different components. The comparison of the mean movement parameters recorded with the different instrustions is given in Table 1. 3.2. EMG signals analysis During quiet stance preceding the rapid flexion of the legs, EMG activities were clearly present (tonic activity) only from the ST and the SM (Fig. 3A, black area). The tonic EMG patterns of the ST and SM were recorded in 64% and 79% of the studied movements (n = 33), respectively. The tonic activity in the SM and ST muscles was usually found in subjects standing at ease and was not a hallmark of the task at hand. The first EMG event related to the triggering of the rapid flexion of the leg was expressed by a clear deactivation of the tonic activity of the ST and SM muscles manifested as inhibition and silent periods. In considering all trials (3 trials in each condition) performed by each subject, this early silent period in the ST and SM activities was recorded in 90% of the subjects, at least in one of the 3 conditions (see Instructions 1–3). Fig. 3A shows that the inhibition of the ST and SM muscles coincides with a burst action of the TA. However, the synchrony of the inhibition of the ST and the SM and the excitation of the TA muscle was not always observed as in Fig. 3. The distribution of the inhibition latency of the SM and ST muscles and the burst latency of the TA muscle with respect to the onset of leg movement are shown in Fig. 4A–C, respectively. The mean onset latencies of the inhibition of the ST and SM with respect to the onset of the angular velocity of knee flexion were −180.5 ± 89.7 ms and −196.9 ± 96.4 ms, respectively. The corresponding

Fig. 3. Typical EMG responses prior to, during and after the flexion of the legs in a representative subject. In A, the first trace corresponds to the ˙ ) in one single movement. The black area angular velocity of the knee (K in the EMG signal of the ST and SM muscles indicates the tonic activity present before the initiation of leg movement. The arrows indicate the onset of deactivation of the tonic activity of the ST and SM muscles (silent period, SP). Zero time indicates the onset of flexion movement. In B, a stick diagram of the flexion movement in the sagittal plane corresponds to the EMGs illustrated in A. The sampling of the stick diagram was made at 100 Hz. The large arrow indicates the direction of movement.

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

63

tions in the onset of the early events were recorded during the 3 different conditions concerning the amplitude and the voluntary brake of the flexion movement, there was no tendency to find an earlier anticipative event in one particular situation. For example, Fig. 5 shows a typical experiment for which we superimpose the kinematic (angular velocity of knee flexion) and the EMG signals of 4 leg muscles (BF, ST, SM and TA) corresponding to the instruction 3 (Fig. 5B) and number 2 (Fig. 5C). While the amplitude of the 2 movements are clearly different (for comparison see the stick diagram of Fig. 5B,C) the first phases of velocity profiles of knee flexion are identical in the 2 situations. Only the late phases of these velocity curves are significantly different. Moreover, the time relationship between the onset of the movement and the onset of the ST and SM inhibition and the TA activation are also largely identical. This figure also demonstrates the fact that

Fig. 4. Histograms of the distribution of inhibition latency of the tonic activity of the SM (A) and ST (B) muscles and of activation latency of the TA muscle (C). The arrows indicate the onset of the angular velocity of the knee during the flexion of the legs.

range of these values were −39.0 to −402.0 ms for the ST and −11.0 to −402.0 ms for the SM. From the 33 studied movements the early action of the TA was recorded in 27 movements but was associated with the presence of the silent period (deactivation) on the SM and ST in only 25 movements. The mean onset latency of the activation of the TA was −60.5 ± 117 ms (range of −294.0 to +280.0 ms). In most situations the onset of the inhibition in the SM and ST muscles occurred in advance with respect to the onset of TA activation. Indeed, when the inhibition occurred early, the TA activation followed the same trend. Furthermore, although these time varia-

Fig. 5. In A, the superposition of two single trials executed following instruction 2 and 3 (see text) in one representative subject. The first two traces represent the angular velocity profile of these two different movements. The black arrow and the dotted line indicate the onset of flexion movement. The small downward and upward arrows indicate the onset of the inhibition in the ST and SM muscle and the phasic activation of the TA muscle, respectively. In B and C, stick diagrams representing the flexion movement performed following instructions 3 and 2, respectively.

64

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

by a clear burst of activation of the hip flexor muscles represented here by the rectus femoris (RF), the tensor faciae latae (TFL) and the sartorius (SAR). Like the uniarticular knee extensor muscles represented in this figure by the vastus lateralis (VL), all the hip flexor muscles recorded in the present task discharged only at or after the onset of the movement and not before. This figure also shows the absence of tonic activation of the gluteus maximus (GM) prior to the movement onset and its very small contribution to the control of the flexion movement. 3.3. Effect of changing stance conditions on the tonic activity of the SM and ST and their deactivation For this experiment, subjects were instructed to modify their initial posture prior to performing a fast flexion of the leg. The movement was recorded firstly in the natural standing position adopted by the subject (Fig. 8A). The

Fig. 6. Muscle activation pattern of the rectus femoris (RF), the vastus lateralis (VL), the vastus medialis (VM) and the tibialis anterior (TA) muscles during a movement illustrated by the stick diagram. The first trace corresponds to the angular velocity of the knee (K). The open arrows indicate the late onset of the RF, VL and VM EMG with respect to the onset of the flexion of the legs (zero time indicated by the dotted line). The black arrow indicates the anticipative onset of the TA muscle.

the inhibition of the tonic activity of the ST and SM is initiated before the onset of the TA activation. In this particular case the time difference between these 2 events was 87 ms. Fig. 6 shows the EMG pattern of the different principal muscles of the quadriceps (RF, VL and VM) in relation to the onset of the flexion of the leg and the anticipative activity of the TA. This characteristic pattern of the quadriceps activation was recorded in all subjects. Their onsets were always clearly located after the onset of leg flexion. Their peak of activity occurred during the deceleration phase of the velocity profile of the knee angular motion. Fig. 7 illustrates the absence of an anticipative EMG activation of neck and trunk muscles. The splenius capitis (SPL) and erectus spinae (lumbar) (ES) were inactive before the onset of the knee flexion. Moreover, the deactivation of the tonic activity of the SM muscle (black area) occurring in this case at −178.5 ms was not accompanied

Fig. 7. Muscle activation pattern of the splenius capitis (SPL), the erector spinae at lumbar level (ES), the gluteus maximus (GM), the semimembranous (SM), the rectus femoris (RF), the tensor faciae latae (TFL), the sartorius (SAR) and the vastus lateralis (VL) muscles during a flexion movement performed following instruction 1. The zero time and the associated line indicate the onset of movement. Movement kinetics are ˙ ). The dotted line indicates the represented by knee angular velocity (K onset of the deactivation of the tonic activity (black area) of the SM muscle.

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

65

Fig. 8. The influence of stance conditions upon the tonic activity of the SM and ST muscles and their deactivation for a representative subject. In A, the flexion of the legs is performed from the natural upright posture indexed by the vertical line joining the markers number 3 and 10 in the stick diagram. In B, the flexion of the legs is performed from a forward position of the body (see the direction of the arrow with respect to the vertical line in the stick diagram). In this situation, the tonic activity (shaded area) of the ST and SM muscles prior to the movement is increased. In C, flexion of the legs is performed from a backward position of the body (see the direction of the arrow with respect to the vertical line in the stick diagram). In this situation the tonic activity of the ST and SM is absent. For each of the three blocks the first trace corresponds to the angular velocity of the knee (K), the open triangle and the associated dotted line indicate the onset of the movement.

subject subsequently adopted a forward bending posture (Fig. 8B), from which the same movement was initiated. Finally, flexion of the leg was initiated from a backward bending posture (Fig. 8C). Fig. 8 shows that leg movements were practically the same regardless of the initial posture. Small differences in the maximal angular velocity of the knee were recorded in these situations. A higher value was obtained from the forward sway position (429 deg/s, Fig. 8B), whereas a lower value was obtained from the backward sway position (336 deg/s, Fig. 8C). A change in the initial posture induced a great difference in the tonic activity of the SM and ST muscles. During the forward standing posture (Fig. 8B), an increase in the tonic activities of the SM and ST compared with those recorded during the natural posture (Fig. 8A) was observed. Conversely, during the backward posture (Fig. 8C) these tonic activities were absent. In this

latter situation an important activity in the TA which participates in the maintenance of the body equilibrium was recorded before the movement. Whatever the initial posture for which the tonic activities of the SM and ST were present (Fig. 8A,B), the onset of the inhibition of these 2 muscles constantly occurred before the onset of knee flexion (−175 ms in A and −109 ms in B). Moreover, the synchronization between the onset of the silent period of the SM and ST muscles and the phasic activation of the TA muscle was better in the forward posture (Fig. 8B) than in the natural one (Fig. 8A). These 2 events occurred at the same time in B, while the onset of the silent period began 122 ms before TA activation in A. 3.4. Cross-correlation analysis Fig. 9 shows cross-correlation functions (CCFs) calcu-

66

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

nals vary in opposite directions. This latter situation was encountered for the conjugate CCF between the deactivation of the ST or the SM and the activation of the TA muscle (Fig. 9A,B). In these cases, the largest CCFmax were negative with a mean of −0.39 ± 0.14 and −0.43 ± 0.14 for the ST-TA and SM-TA muscle combinations, respectively. These cross-correlation peaks occurred around a mean time lag of −77.3 ± 58.6 ms and of −82.8 ± 65.9 ms, respectively. The histograms of Fig. 9 show the distribution of the CCFmax (open bar) and their respective time lags (filled bar) for the different combinations tested and for all the movements realized by all the subjects. For the ST-TA and SM-TA combinations all the CCFmax values were negative (Fig. 9D,E) and always occurred at a negative time lag (Fig. 9G,H). The CCF between the 2 hamstring ST and SM muscles (Fig. 9C,F,I) displays a positive peak at a time lag of 0 ms which means that the deactivation of these muscles were synchronously coordinated. For comparison, the CCF obtained for the other combinations between the hamstring and the SOL muscles and between the SOL and the TA muscles (Fig. 9J–O) were not so regular throughout the subjects and movements. While the CCFmax may reach significant levels, the sign of the correlation was frequently opposite and may appear at positive or negative time lags depending of the subject and trial. 3.5. Analysis of movement dynamics Fig. 9. Conjugate cross-correlation analysis of the EMG patterns between the ST and TA muscles (A, D, G), the SM and TA muscles (B, E, H) and the ST and SM muscles (C,F,I) during the time period preceding the onset of movement. Each of the 3 curves (A, B, C) corresponds to the CCF of one movement executed by one representative subject. In D, E and F distribution of the CCFmax measured in the 11 subjects (60 movements out of 99) for the ST-TA, SM-TA and ST-SM muscle combinations, respectively. In G, H and I distribution of the time lags corresponding to the CCFmax of the preceding correlations, respectively. In J, K and L distribution of the CCFmax for the ST-SOL, SM-SOL and SOL-TA muscle combination, respectively. In M, N and O, distribution of the time lags corresponding to the CCFmax of the preceding correlations, respectively. (No of obs, number of observations).

lated during the time period preceding the onset of movement for the EMG signals of the ST and TA (A), the SM and TA (B) and the ST and SM muscles (C). This statistical analysis was performed on all the trials for which a deactivation was recorded in the ST and SM muscles and associated with an activation in the TA muscle. This situation was encountered in 60 movements out of the 99 recorded. The peak value, designated CCFmax, provides a measure of the correlation degree between both chosen EMG signals. A reasonable estimate of the 5% level of significance is CCFmax = ± 0.15 (Miller et al., 1993). A positive CCFmax with a given time lag means that with the appropriate lag both EMG signals vary in the same direction. Conversely, a negative CCFmax means that the 2 sig-

The biomechanical analysis is focused upon the evolution of the total muscular torque (TMT) acting on the hip, knee and ankle prior to the movement onset and during the early phase of squatting. The main purpose of this analysis was to determine the muscular torque change that is responsible for the triggering of movement. The torque is considered positive when it produces a rotation (around the center of the proximal joint with respect to the ground level) of the segment in a counter-clockwise direction. For example, the muscular torque exerted by the hamstring muscles on the upper segments of the body (head, arm and trunk) produces a positive torque (extension) at the hip. Conversely, the torque induced by gravity (not illustrated) produces a negative torque (flexion) at the same joint. During erect posture a constant muscular torque with positive values oscillating around 15 Nm was measured at the hip and ankle joint. The muscular torque at the hip decreased 67 ms before the onset of the flexion movement and reached 0 Nm. Thereafter, a negative peak of 20 Nm was recorded in phase with the onset of movement. This dynamical pattern was consistent with a total inhibition of the extensor drive acting on the hip. Moreover, this analysis demonstrated the absence of a clear and consistent hip flexion torque as those produced by a burst action of hip flexor muscles. In fact, a burst action of this type would be expected to produce a negative torque peak of a similar amplitude as that measured later on during the movement

67

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

and largely superior to 20 Nm. The suppression of the constant positive torque was also present at the ankle joint. A similar pattern was obtained at the knee but in the opposite direction. 3.6. Influence of early head movements on anticipatory EMG responses Because the vestibular system may be postulated as a candidate for triggering automatic postural responses in the leg muscles (Inglis et al., 1995), small early variations of the upper part of the body could produce or influence the deactivation of the ST and SM tonic activities recorded in the present situation. In order to test this possibility the horizontal and vertical components of the linear motion of the head and the angular variation of the FK with respect to the earth vertical were analysed. Fig. 10 shows (for one representative subject) the horizontal linear displacements of the head markers number 1 and 2, their horizontal and vertical linear accelerations and the angular position and velocity of the FK plane for 2 different trials superimposed and in succession. All these curves are compared with the time of occurrence of the silent period of the SM muscle. Before the onset of the flexion movement (marked by the stripped line and the black arrow) around −380 ms, but just after the onset of the deactivation of the SM muscle (see the short dotted line), a small horizontal displacement of the head was clearly present. Fig. 10 shows that the flexion movement was accompanied by an extension of the head. The maximal angular velocity of the head occurred 100 ms before the onset of the leg movement. Although the head forward motion occurred before flexion of the legs, Fig. 10 definitely shows, from the timing point of view, that the vestibular signals generated by this horizontal head acceleration (represented here with a greater magnification factor than the vertical head acceleration) do not serve as a trigger for the deactivation of the SM muscles. The mean values concerning the amplitude of the linear head acceleration and the time relationship between the head movement and the deactivation of the SM (DT) are given in Table 2. Fig. 10. Analysis of the time relationships between the different components of the head motion and the anticipative deactivation of the SM muscle during rapid flexion of the legs. Two different trials (A,B) are illustrated for a chosen subject. The first two traces (Hh D(1) and Hh D(2)) correspond to the horizontal displacement of the head markers number 1 and 2, respectively. The shaded area represents the displacement of the markers induced by the forward sway of the body in the absence of the extension movement of the head. From top to bottom: the ¨ (1,2)) of the superposition of the horizontal linear acceleration (Hh D head markers number 1 and 2; the superposition of the vertical linear ¨ (1,2)) of the head markers number 1 and 2; the anguacceleration (Hv D lar variation of the ‘Frankfort-plane’ (FK); the angular velocity of the ˙ , and the EMG of the SM muscle. The black arrow indicates the onset FK of the movement. The open arrow indicates the onset of the horizontal forward displacement of the head. Abbreviations: F, forwards; B, backwards; U, upwards; D, downwards

Table 2 Mean value of the time relationship between the onset of the silent period of the SM muscle and the peak of the horizontal acceleration of the head (DT)

Mean SD

DT (ms)

Hor. Acc (mm/s2)

Vert. Acc (mm/s2)

FK FK angular amplitude velocity (deg) (deg/s)

+312.5 104.6

2015.9 726.8

−10810.8 2610.9

9.2 7.2

25.9 21.7

Mean values of the horizontal and vertical acceleration of the head and the angular amplitude and velocity of the ‘Frankfort Plane’ (FK) are also given for all the subjects presenting anticipatory deactivation of the SM.

68

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

4. Discussion This study has attempted to assess the motor strategy used by normal human subjects in performing a rapid flexion of the legs during the standing posture. In the majority of subjects conjugate cross-correlation analysis have demonstrated the existence of a coordination strategy represented by 2 main synergistic actions: (1) the early suppression of the tonic activity of the SM and ST muscles, and (2) the burst activity of the TA muscle. 4.1. Anticipatory postural adjustments versus the triggering action of the movement The comparison of the present EMG events recorded prior to the movement onset with the anticipatory responses of the experiment of Bouisset and Zattara (1987) and Zattara and Bouisset (1988) during both unilateral and bilateral arm raising is interesting. Their anticipatory activities exhibited a distal to proximal trend whilst from the present paradigm, a reverse proximo-distal trend has been recorded. During their experiments, the soleus became silent about 60 ms prior to voluntary movement, whereas the SM became active 30 ms prior to the movement. In the present experiment, although the inhibition remained the first event, it occurred in the SM at about 197 ms prior to the flexion of the legs, whereas the TA became active 106 ms prior to voluntary movement. In this context, the simulation experiment performed by Ramos and Stark (1990a) demonstrates that an anticipatory activation of the SM greatly increases the backwards motion of the body. We can therefore deduce from this study that a clearcut silencing of the SM like those reported in the present study must facilitate the forward motion of the body. Does the silent period of the SM and ST really represent an anticipatory postural adjustment or merely a part of the synergy (triggering action) leading to leg flexion? The fact that in the majority of subjects the deactivation of the biarticular SM and ST muscles were significantly correlated with the early activation of the TA muscle (Fig. 9A,B,D,E) and that the onset of this deactivation occurred earlier than the phasic activity of the TA, which is also demonstrated by the negative time lags of the cross-correlation, suggest the existence of a triggering action played by the silent period of the ST and SM. In general, during any classical experimental approach of postural control, the anticipatory component of postural activity has to readdress the center of gravity into a stable position which accounts for the limb or body movement (Horstmann and Dietz, 1990). In the present experiment, the movement itself implicated a profound change between the initial erect posture and the final flexed posture. In this particular case, movement and postural control may be viewed as a single process subserved by a global strategy. The aim of this motor strategy is to perform a rapid change of posture and it necessitates 2 prepro-

grammed actions: (1) the ‘unlocking’ of the previous erect posture (subserved by the inhibition of the tonic hip extensor activity) and (2) the forward tilt of the body (subserved by the burst activity of the TA). After these preprogrammed events, the principal force used during the unloading movement is provided by gravity and controlled by an intense co-contraction of all the leg muscles. It is also interesting to compare the motor program used for leg flexion movement with those described by Crenna and Frigo (1991) for forward-oriented movements and characterized by an inhibition of the SOL followed by a burst activation of the TA. Only the TA burst is recorded in the present movement and an anticipative inhibition is recorded at the level of the hamstring muscle but not at the level of the SOL. Moreover, the cross-correlation analysis between the EMG of the SOL and TA and between the 2 hamstring muscles and the SOL in the present movement demonstrated the absence of a common coordination sequence between these muscles (Fig. 9J–O). Conversely, the same statistical analysis demonstrated the existence of a common coordinated pattern between the hamstring deactivation and the TA activation. Kinematics analysis of the leg movement in all of the forward-oriented movement studied by Crenna and Frigo (1991) and in the present study reveals the presence of an initial dorsal flexion of the ankle for which the TA would be considered as the prime mover. Therefore, we may consider this initial flexion of the ankle as a proper element of the movement itself and not as a separate anticipatory postural adjustment. In fact, during whole body movement the application of the concept of anticipatory postural adjustment is not easy because the postural requirement became an integral part of the movement task. The suggestion that the silent period is a preprogrammed central command may not be completely upheld as EMG activity of deep hip flexor muscles was not recorded. For example, the suppression of activity in hip extensor could be related with a sort of reciprocal inhibition mechanism, to some activity in the illiopsoas muscle which could per se initiate (flexing the hip) the unlocking process of standing posture. In other words the deactivation of the ST and SM could be only a counterpart of the excitation of hip flexor muscles at the onset of flexion movement. Illiospoas activity was not recorded using a needle because this type of recording of a deep and almost inaccessible muscle is not ethically possible during such a type of rapid whole body movement. Moreover, indirect arguments are in favor of an absence of illiopsoas activity at the latency of hamstring inhibition. (1) We have showed that the EMG activity of the rectus femoris, the sartorius and the tensor of faciae latae, a 2 hip flexor and synergist of the illiopsoas were not active before the flexion movement (Fig. 7). (2) The significance of this absence of early activation of the sartorius is reinforced by the fact that it was reported that, in combined hip and knee movements, the sartorius is known to produce its greatest

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

activity when hip flexion is accompanied by maximum knee flexion (Basmajian, 1967) (the condition encountered during squatting). (3) Biomechanical analysis performed on the basis of the present kinematic data demonstrates that a prime mover activation of the hip flexor (initial active muscular torque at the hip) is not necessary for producing the whole flexion of the body. The present biomechanical analysis reinforces the idea that the deactivation of the tonic activity represented the first event of the ‘unlocking’ of the upright posture facilitating the rapid flexion of the leg with the action of gravity. 4.2. The Hufschmidt phenomenon and the interruption of the standing posture The anticipative inhibition of the tonic activity of the hamstring muscle resembles the Hufschmidt phenomenon (Hufschmidt and Hufschmidt, 1954). These authors have demonstrated that in a simple sensory-motor reaction task in which the subject must maintain a tonic contraction of the biceps muscle and voluntarily react to an electrical stimulus by a contraction of their triceps muscle, the agonist burst (triceps) was usually preceded some 50 ms earlier by an inhibition of the tonic biceps activity. This early inhibition was also reported in other movement tasks, both in man (Gottlieb et al., 1970; Hallet et al., 1975; Hoffman and Strick, 1990) and in monkeys (Evarts, 1974; Hoffman and Strick, 1995). Hallet et al. (1975) have reported that this inhibition is present in patients with pan-sensory neuropathy suggesting a central preprogrammed nature of this command. Moreover, a recent study of Hoffman and Strick (1995) have demonstrated in a monkey, after lesions of the contralateral arm area of the primary motor cortex (M1), the abolition of the early inhibition. They have suggested that this absence of antagonist inhibition was compatible with a prior proposal of Preston et al. (1967) that an important function of M1 is to interrupt posture at the onset of movement. The deactivation of the SM and ST muscles before a rapid squatting movement could also be considered to result from the same type of central command. Therefore, the present results could be in favor of a generalization of the Hufschmidt phenomenon inducing the interruption of different types of posture. The fact that the first event of this descending command is a silencing could be relevant to the analysis of the postural inflexibility (Horak et al., 1992) and the motor block phenomenon (Giladi et al., 1992) frequently observed in Parkinson’s disease (PD). The present approach would be useful in order to test the ability of PD patients to demonstrate the silencing command prior to switch from one posture to another. 4.3. The influence of head movement on the anticipatory responses It is well known that the vestibular system acts as

69

an internal reference system which controls static upright position (Allum and Pfaltz, 1985; Bussel et al., 1980; Allum and Keshner, 1986) and dynamic equilibrium (Pozzo et al., 1991). In particular, it has been seen in the cat, that a change in the position of the head alters the tone of the extensor muscles of the 4 limbs (Magnus and DeKleijn, 1912). The deafferentation of the otolithic system induces a relative suppression of the extensor muscles (Magnus, 1926). Moreover, the vestibular system is sensitive to the range of head accelerations present during quiet stance (Benson et al., 1986). In the present paradigm, although the main vertical component of the head movement was concomitant to leg flexion, the horizontal component was initiated just before the principal movement of the body. The amplitude of this earlier horizontal head acceleration was about 0.5 m/s2 which is of an order of magnitude higher than the corresponding perceptual threshold of the vestibular system (0.04 m/s2, Melvill-Jones and Young, 1978). Can this horizontal head acceleration be responsible for the silent period of the hip extensor? The effect of a sudden head tilt during stance has been evaluated by Dietz et al. (1988). They demonstrated that a forward directed head tilt during stance was followed by slight tonic activity of the gastrocnemius muscle at a latency of 55 ms and later on the TA muscle (latency of 95 ms). The extent of head tilt induced during the latter study by a torque motor was comparable to the extent of the voluntary forward movement of the head recorded in the present study. Although there is no data concerning the effect of head tilt on hamstring muscles, it seems unlikely, on the basis of the latency data, that horizontal head movement initiates early anticipatory deactivation of the hamstring muscles. Moreover, the fact that our results have demonstrated a functional link between the silent period of the hamstring and the burst activity of the TA, is not compatible with the idea that horizontal head movement initiates these anticipative responses, because an early activation of the TA was induced by a backward and not a forward tilt of the head (Dietz et al., 1988). More recently, Inglis et al. (1995) demonstrated that the vestibular system does not play a critical role in triggering the earliest postural responses following galvanic vestibular stimulation. Acknowledgements We thank Bernard Dan and Paul Stapley for useful discussion and Michelle Plasch for secretarial assistance. This work was supported by the Belgian National Fund for Scientific Research (F.N.R.S.), by the Research Fund of the University of Brussels (U.L.B.) and by the Banque Nationale de Belgique. T. Pozzo was supported by a grant from the Institut National de la Recherche Me´dicale, France(INSERM/CRE No. 92107).

70

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71

References Allum, J.H.J. and Keshner, E.A. Vestibular and proprioceptive control of sway stabilization. In: W. Bles and T. Brandt (Eds.), Disorders of Posture and Gait, Elsevier, Amsterdam, 1986, pp. 19–39. Allum, J. and Pfaltz, C.R. Visual and vestibular contributions to pitch sway stabilization in the ankle muscles of normals and patients with bilateral vestibular deficits. Exp. Brain Res., 1985, 58: 82–94. Amblard, B., Assiante, C., Hamid, L. and Marchand, A. A statistical approach to sensorimotor strategies: conjugate cross-correlations. J. Motor Behav., 1994, 26: 103–112. Basmajian, J. Muscles Alive: Their Functions Revealed by Electromyography, 2nd edn. Williams and Wilkins, Baltimore, 1967. Belen’kii, V.Y., Gurfinkel, V.S. and Paltsev, Y.I. Elements of control of voluntary movements. Biophysics, 1967, 12: 154–160. Benson, A.J., Spencer, M.B. and Stott, J.R.R. Thresholds for the detection of the direction of whole-body, linear movement in the horizontal plane. Aviat. Space Environ. Med., 1986, 57: 1088–1096. Bouisset, S. and Zattara, M.A. A sequence of postural movements precedes voluntary movement. Neurosci. Lett., 1981, 22: 263–270. Bouisset, S. and Zattara, M. Biomechanical study of the programming of anticipatory postural adjustments associated with voluntary movements. J. Biomech., 1987, 20: 735–742. Bourgeois, M. Contribution a` l’Analyse Biome´canique de l’Organisation du Mouvement Volontaire Complexe par Mode`le Mathe´matique et Approche Myo-Cyberne´tique. PhD Thesis, Universite´ Libre de Bruxelles, 1991. Brown, J.F. and Frank, J.S. Influence of event anticipation on postural actions accompanying voluntary movement. Exp. Brain Res., 1987, 67: 645–650. Bussel, B., Katz, R., Pierrot-Deseilligny, E., Bergego, C.M, and Hayat, A. Vestibular and proprioceptive influences on the postural reactions to a sudden displacement in man. In: J.E. Desmedt (Ed.), Spinal and Supraspinal Mechanisms of Voluntary Motor Control and Locomotion. Progress in Clinical Neurophysiology, Vol. 8, Karger, Basel, 1980, pp. 310–322. Clauser, C.E., Mc Conville, J.T. and Young, J.W. Weight, volume and center of mass of segments of the human body. Technical documentary report, AMRL-TR 69–70 Wrigth-Patterson, 1969. Cordo, P. and Nashner, L. Properties of postural adjustments associated with rapid arm movements. J. Neurophysiol., 1982, 47: 287–302. Crenna, P. and Frigo, C. A motor program for the initiation of forwardoriented movements in humans. J. Physiol., 1991, 437: 635–653. Crenna, P., Frigo, C., Massion, J. and Pedotti, A. Forward and backward axial synergies in man. Exp. Brain Res., 1987, 65: 538–548. D’Amigo, M. and Ferrigno, G. Technique for the evaluation of derivatives from noisy biomechanical displacement data using a modelbased bandwidth-selection procedure. Med. Biol. Eng. Comput., 1990, 28: 407–415. DeBeer, G.R. How animals hold their heads. Proc. Linn. Soc. Lond., 1947, 159: 125–139. Dietz, V., Horstmann, G.A. and Berger, W. Fast head tilt has only a minor effect on quick compensatory reactions during the regulation of stance and gait. Exp. Brain Res., 1988, 73: 470–476. Diener, H.C., Bacher, M., Guschlbauer, B., Thomas, C. and Dichgans, J. The coordination of posture and voluntary movement in patients with hemiparesis. J. Neurol., 1993, 240, 161–167. Dufosse, M., Hugon, M. and Massion, J. Postural forearm changes induced by predictable in time or voluntary triggered unloading in man. Exp. Brain Res., 1985, 60: 330–334. Evarts, E.V. Sensorimotor cortex associated with movements triggered by visual as compared to somesthetic inputs. In: F.O. Schmitt and F.G. Worden (Eds.), The Neurosciences, Third Study Program. MIT Press, Cambridge, MA, 1974, pp. 327–337. Ferrigno, G. and Pedotti, A. Elite: a digital dedicated hardware system for movement analysis via real time TV-signal processing. IEEE Trans. Biomed. Eng., 1985, 32: 943–950.

Flanders, M., Pellegrini, J.J. and Soechting, J.F. Spatial/temporal characteristics of a motor pattern for reaching. J. Neurophysiol., 1994, 171: 811–813. Friedli, W.G., Hallet, M. and Simon, S.R. Postural adjustments associated with rapid voluntary arm movements. 1. Electromyographic data. J. Neurol. Neurosurg. Psychiatry, 1984, 47: 611–622. Gahery, Y. and Massion, J. Coordination between posture and movement. Trends Neurosci., 1981, 4: 199–202. Giladi, N., Mcmahon, D., Przedborski, S., Flaster, E. Guillory, S., Kostic, V. and Fahn, S. Motor blocks in Parkinson’s disease. Neurology, 1992, 42: 333–339. Gottlieb, G.L., Agarwal, G.C. and Stark, L. Interactions between voluntary and postural mechanisms of the human motor system. J. Neurophysiol., 1970, 33: 365–381. Hallet, M., Shahani, B.T. and Young, R.R. EMG analysis of stereotyped voluntary movements in man. J. Neurol. Neurosurg. Psychiatry, 1975, 38: 1154–1162. Hof, A.L. and Van Den Berg, J.W. EMG to force processing. I. An electrical analogue of the Hill muscle model. J. Biomech., 1981, 11: 747–758. Hoffman, D.S. and Strick, P.L. Step-tracking movements of the wrist in humans. II. EMG analysis. J. Neurosci., 1990, 10: 142–152. Hoffman, D.S. and Strick, P.L. Effects of a primary motor cortex lesion on step-tracking movements of the wrist. J. Neurophysiol., 1995, 73: 891–895. Horak, F.B. and Nashner, L.M. Central programming of postural movements: adaptation to altered support-surface configurations. J. Neurophysiol., 1986, 55: 1369–1381. Horak, F.B., Nutt, J.G. and Nashner, L.M. Postural inflexibility in parkinsonian subjects. J. Neurol. Sci., 1992, 111, 46–58. Horak, F.B., Esselman, P., Anderson, M.E. and Lynch, M.K. The effects of movement velocity, mass displaced, and task certainty on associated postural adjustments made by normal and hemiplegic individuals. J. Neurol. Neurosurg. Psychiatry, 1984, 47: 1020–1028. Horstmann, G.A. and Dietz, V. A basic posture control mechanism: The stabilization of the centre of gravity. Electroenceph. clin. Neurophysiol., 1990, 76: 165–176. Hufschmidt, H.J. and Hufschmidt, T. Antagonist inhibition as the earliest sign of a sensory-motor reaction. Nature, 1954, 25: 605. Hugon, M., Massion, J. and Wiesendanger, M. Anticipatory postural changes induced by active unloading and comparison with passive unloading in man. Pflu¨gers Arch., 1982, 393: 292–296. Inglis, J.T., Shupert, C.L., Hlavacka, F. and Horak, F.B. Effect of galvanic vestibular stimulation on human postural responses during support surface translations. J. Neurophysiol., 1995, 73: 896–901. Kasai, T. and Kawai, K. Quantitative EMG analysis of anticipatory postural adjustments of voluntary contraction of leg muscles in standing man. Electroenceph. clin. Neurophysiol., 1994, 93: 184–187. Lee, W.A. Anticipatory control of postural and task muscles during rapid arm flexion. J. Motor Behav., 1980, 12: 185–196. Lee, W.A., Buchanan, T.S. and Rogers, M.W. Effects of arm acceleration and behavioural conditions on the organization of postural adjustments during arm flexion. Exp. Brain Res., 1987, 66: 257–270. Lum, P.S., Reinkensmeyer, D.J., Lehman, S.L., Li, P.Y. and Stark, L.W. Feedforward stabilization in a bimanual unloading task. Exp. Brain Res., 1992, 89: 172–180. Magnus, R. The physiology of posture: Cameron Lectures. Lancet, 1926, i: 531–536, 585–588. Magnus, R. and DeKleijn, A. Die Abh :a¨ngigkeit des Tonus der extremita¨tenmuskeln von der Kopfstellung. Pflu¨gers Arch., 1912, 145: 455–548. Massion, J. Movement, posture and equilibrium: interaction and coordination. Prog. Neurobiol., 1992, 38: 35–56. Melvill-Jones, G. and Young, L.R. Subjective detection of vertical acceleration: a velocity dependent response? Acta Otolaryngol., 1978, 85: 45–53. Miller, L.E., van Kan, P.L.E., Sinkjaer, T., Andersen, T., Harris, G.D.

G. Cheron et al. / Electroencephalography and clinical Neurophysiology 105 (1997) 58–71 and Houk, J.C. Correlation of primate red nucleus discharge with muscle activity during free-form arm movements. J. Physiol. (Lond.), 1993, 469: 213–243. Mouchino, L., Aurenty, R., Massion, J., Pedotti, A. Coordination between equilibrium and head-trunk orientation during leg movement: a new strategy built up by training. J Neurophysiol., 1992, 67: 1587– 1598. Nashner, L.M. Fixed patterns of rapid postural responses among leg muscles during stance. Exp. Brain Res., 1977, 30: 13–24 :. Nardone, A. and Schieppati, M. Postural adjustments associated with voluntary contraction of leg muscles in standing man. Exp. Brain Res., 1988, 69: 469–480. Onyshko, S. and Winter, D.A. A mathematical model for the dynamics of human locomotion. J. Biomech., 1980, 13: 361–368. Paulignan, Y., Dufosse, M., Hugon, M. and Massion, J. Acquisition of co-ordination between posture and movement in a bimanual task. Exp. Brain Res., 1989, 77: 337–348. Pozzo, T., Berthoz, A. and Lefort, L. Head stabilization during various locomotor tasks in humans. Exp. Brain Res., 1991, 82: 97–106. Preston, J.B., Shende, M.C. and Uemura, K. The motor cortex-pyramidal system: patterns of facilitation and inhibition on motoneurons innervating limb musculature of cat and baboon and their possible adapta-

71

tive significance. In: M.D. Yahr and D.P. Purpura. (Eds.), Neurophysiological Basis of Normal and Abnormal Motor Activities, Raven, New York, 1967, pp. 61–72. Ramos, C.F. and Stark, L.W. Postural maintenance during movement: Simulations of a two joint model. Biol. Cybern., 1990a, 63: 363–375. Ramos, C.F. and Stark, L.W. Postural maintenance during fast forward bending: a model simulation experiment determines the ‘reduced trajectory’. Exp. Brain Res., 1990b, 82: 651–657. Rogers, M.W. Influence of task dynamics on the organization of interlimb responses accompanying standing human leg flexion movements. Brain Res., 1992, 579: 353–356. Rogers, M.W. and Pai, Y. Dynamic transitions in stance support accompanying leg flexion movements in man. Exp. Brain Res., 1990, 81: 398–402. Yamashita, N., Nakabayashi, T. and Moritani, T. Inter-relationships among anticipatory EMG activity, Hoffmann reflex amplitude and EMG reaction time during voluntary standing movement. Eur. J. Appl. Physiol., 1990, 60: 98–103. Zattara, M. and Bouisset, S. Posturo-kinetic organization during the early phase of voluntary upper limb movement. 1. Normal subjects. J. Neurol. Neurosurg. Psychiatry, 1988, 51: 956–965.