- Email: [email protected]
Influences of illusionary position perception on anticipatory postural control associated with arm flexion
Journal of Electromyography and Kinesiology 13 (2003) 509–517 www.elsevier.com/locate/jelekin
Influences of illusionary position perception on anticipatory postural control associated with arm flexion Katsuo Fujiwara ∗, Kaoru Maeda, Hiroshi Toyama Department of Human Movement and Health, Graduate School of Medical Science, Kanazawa University, 13-1 Takara-machi, Kanazawa 920-8640, Japan Received 31 May 2002; received in revised form 23 April 2003; accepted 7 June 2003
Abstract We examined the effect of illusionary perception on anticipatory postural control associated with arm flexion with subjects in a standing position, using vibration stimulation of the Achilles’ tendon. Arm flexion was performed five times under each of the following conditions: (1) quiet standing, (2) vibration of the Achilles’ tendon at 100 Hz frequency and 1.5 mm amplitude with the trunk fixed by a stopper during quiet standing, and (3) a perceived standing position during vibration. The reproduced positions were located forward by about 20% of the foot length compared with the quiet standing position; these positions showed no significant differences among the five trials. In the first trial of arm flexion during vibration, the biceps femoris began activating approximately 40 ms before the anterior deltoid. The same time difference between activation of the two muscles was observed in the reproduced condition. As the vibration trials were repeated, this activation timing approached the value in the quiet standing condition. In both the biceps femoris and erector spinae, the mean amplitude of electromyogram for the first 50 ms after the start of activation did not differ significantly among the three conditions. 2003 Elsevier Ltd. All rights reserved. Keywords: Anticipatory postural control; Muscle activity; Vibration stimulation; Illusionary perception
1. Introduction Many previous studies have shown that the onset of activation of the postural muscles of the legs and trunk that control standing postures precedes that of the focal muscles that rapidly move the arm [3,7]. The preceding activation of the postural muscles is adjusted by a program that is selected in advance in order to moderate the effect of disturbances of posture and equilibrium caused by the arm movement [2,5,10,11,19,28,34]. When both arms are rapidly flexed to the front, body balance is disturbed by the increase in rotational momentum of the body in the anterior direction. It has been demonstrated that the postural muscles in the rear of the body begin activating before the focal muscles, and that this preceding activation leads to the enhancement of preparatory
Corresponding author. Tel.: +81-76-265-2225; fax: +81-76-2344219. E-mail address: [email protected] (K. Fujiwara). ∗
1050-6411/$ - see front matter 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1050-6411(03)00083-X
muscle tension in the postural muscles [10,34] and/or the postural movement to compensate for the forward transition of the center of gravity [13,20]. Many studies have found that the muscle action sequence of focal and postural muscles differs significantly according to behavioral condition, and particularly in self-paced arm movement, the postural muscles show a large preceding activation [4,8,11,19]. The activation timing of postural muscles is probably affected by the initial standing position just before arm movement [5]. Benvenuti et al. [4] found that muscle activation of the postural muscles begins earlier in a forward leaning posture with the center of foot pressure (CFP) at the metatarsal head than in a quiet standing posture. We previously reported that when subjects performed arm flexion movement with the initial CFP maintained at 30% of foot length from the heel, the CFP moved nearer to the quiet standing position, and preceding action of the postural muscles was not observed [12]. Therefore, we presume that activation of postural muscles changes according to the perception of initial standing position.
510
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
Goodwin et al. [14] reported that when the limb movement originated by a tonic vibratory reflex was obstructed, the illusion of reversal movement was produced. Eklund [9] observed that vibration of the Achilles’ tendon in a standing posture generated a reflex action in the gastrocnemius, and the body then leaned backward. It has also been found that when the leaning was obstructed, the subject perceived an illusory leaning forward of the body [23,30,31]. It has been reported that these illusions are caused by Ia impulses from muscle spindles being sent to the reference system in the neural center where they are collated with the body schema. In addition, it has been found that this illusionary perception of limb position influences the organization of motor action [17]. Gurfinkel et al. [18] reported that vibration stimulation of the neck muscles unilaterally with the head fixed relative to the trunk evoked an illusion of rotation of the head to the contralateral side, and a shift of the entire body laterally according to the illusion. This finding indicates that the illusionary perception of posture affects the organization of postural control. Kasai et al. [22] applied vibration to the tibialis anterior (TA) and soleus (Sol) of human subjects who were performing arm movement in a free-standing posture with their eyes closed. They observed that when vibration was applied to the tibialis anterior, the preceding action of the biceps femoris (BF) to the anterior deltoid (AD) began earlier than that in the free-standing posture, and that when applied to the soleus, it began later. In their study, body movement was not obstructed, and thus the subject leaned forward and backward when the vibration was applied to the tibialis anterior and soleus, respectively. These results were consistent with previous reports, in which the preceding action of biceps femoris began earlier in a forward leaning posture [4] and later in backward leaning [12]. Therefore, it is not clear whether the cause of changes in anticipatory postural control in response to vibration is an illusionary perception or an actual shift of standing position. If anticipatory postural control is affected by an illusionary perception of standing position, postural control in the illusionary perception should be similar to that with the standing position shifting actually to an illusionary position. Nashner [29] reported that abrupt inclination of a standing floor produces a stretch reflex in the triceps surae, and if the reflex does not work effectively for body balance, it is suppressed as the floor inclination is repeated. Furthermore, it has been reported that following a change in the condition of standing floor [20], the aspect of postural disturbance [32], or the gravity condition [6,24,30], the effectiveness of postural control is judged from control results and postural control is adjusted adaptively. In the present study, it is presumed that a similar change takes place. That is, in response to vibration of the Achilles’ tendon, an illusionary percep-
tion at which the body leans forward is produced and the preceding activation of postural muscles is then developed. This activation will lead to an unstable backward leaning of the body, and the preceding activation will be suppressed gradually as the unstable trial is repeated. In the present study, we investigated the effect of illusionary perception on anticipatory postural control associated with arm flexion with subjects in a standing position with their body fixed during vibration stimulation of the Achilles’ tendon. The working hypotheses were as follows: (1) activation of postural muscles will change based on the illusionary perception of standing position, (2) activation will adaptively change based on knowledge of the control result.
2. Methods 2.1. Subjects Large individual differences in the effect of vibration stimulation on positional perception have been observed [23]. We selected subjects in the preliminary test who clearly showed backward leaning of the body when mechanical vibration was applied to the Achilles’ tendon bilaterally, and who showed generation of an illusionary perception at which the body leaned forward in response to vibration when the trunk was fixed by a stopper. Subjects were nine men and three women, aged 19–42 years (mean 27.0 ± SD 0.7). All subjects appeared to be free of any neurological and orthopedic impairment. Informed consent was obtained from all subjects following an explanation of the experimental protocol. The mean values of height, weight, and foot length were 170.3 cm (SD = 5.2), 65.1 kg (SD = 7.7), and 25.4 cm (SD = 0.9), respectively. 2.2. Apparatus All measurements were taken with subjects standing on a force platform (WAMI, WA1001) composed of three load-cells. The pressure force signals (F1, F2, F3, Fig. 1A) detected on the load-cells were passed through an analog operation electric circuit, and the position (CFPy) of the CFP in the anteroposterior direction was calculated according to the formula shown below: CFPy ⫽ {(F2 ⫹ F3) / (F1 ⫹ F2 ⫹ F3)}·l (1: distance from an F1 load-cell to a straight line that connects between load-cells of F2 and F3, 400 mm). The electronic CFP signal was sent to a buzzer generator (HIRUTA, F-H6408) and an A/D converter (I/ODATA, PIO9045) in a computer (NEC, PC9801BX2).
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
511
Fig. 1. Experimental protocol. (A) Quiet standing condition. (B) Reproduced condition. (C) Vibration condition.
The buzzer generator was used to indicate the quiet standing position to the subject. In this apparatus, 1 cm displacement of CFP was calibrated to 0.1 V. The sound was generated when the CFP position was located within ±1 cm of a mean position during quiet standing. Goshima [15] reported that the SD of CFP fluctuation in an anteroposterior direction during quiet standing for 60 s was about 0.5 cm for normal subjects (aged 20 years). Therefore, we decided that the standard range of CFP fluctuation during quiet standing was 1.0 cm, which corresponded to 2 SD. The A/D converter was set at 200 Hz sampling rate with 12-bit resolution. Using the converted digital data, the mean position of CFP fluctuation was calculated and was shown by the relative distance (% FL) from the heel to the length of the foot. Mechanical vibration was bilaterally applied to the Achilles’ tendon through the skin using two vibrators. Each vibrator (Heiwa Electronic Industrial Co., TMT18) was independently strapped to the ankle region with a rubber belt. The vibration frequency was set at 100 Hz, as in a previous report [31], with an amplitude of 1.5 mm. Surface electrodes (Medicotest, M-00-S) were used in a bipolar derivation to record surface EMG activity of the following muscles: the anterior deltoid, rectus abdominis (RA) at the level of the navel, erector spinae (ES) at the level of the iliac crest, rectus femoris (RF), the long head of the biceps femoris, tibialis anterior, and soleus, on the left side. For each muscle, the electrodes were fixed over the muscle belly after shaving and cleaning the skin with alcohol. The electrodes were aligned along the great axis of the muscle with an inter-electrode
distance of about 2 cm. The electrode input impedance was reduced to below 5 k⍀. Signals from the electrodes were amplified (×2000) and band-pass filtered (1.6 Hz– 1.5 kHz) with an EMG amplifier (NEC-Sanei, BIOTOP-6R12). A stopper was set at the back of the subject’s body, with the position being level with the second to fourth thoracic vertebra, to prevent backward leaning of the body while vibrating the Achilles’ tendon. To confirm that the body would tend to incline backward in response to the vibration in the absence of a stopper, the pressure on the stopper was measured by a load-cell (KYOWA, LUR-A-50NSA1). The stopper was fixed by an electromagnet, and was released upon triggering an EMG burst onset of the anterior deltoid. The maximum backward displacement of the stopper was 14 cm behind exceeding the hindmost point of the subject’s heel, which prevented the subjects from falling. The sliding position of the stopper was measured from a sagittal view recorded using a high-speed camera (Photron, Fastcam rabbit mini 2) with a frame rate of 120 Hz. For subsequent analysis, signals from the force platform and EMG electrodes were recorded on a digital tape recorder (TEAC, RD-200E). The bandwidth of the recorder was 0–2.5 kHz. 2.3. Procedure Blindfolded subjects performed the following tasks while standing with their bare feet 10 cm apart and parallel with one another. The feet were aligned accurately with a reference board for the foot position. Subjects
512
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
grasped a grip bar (40 cm in length and 68 g in weight) with their hands and their arms were positioned close to the trochanter majors, and their elbows extended. Subjects started moving both arms at their own pace within 3 s after an oral cue from the examiner, and then stopped the arms at a frontal horizontal level position. This position was then held for 3 s (Fig. 1). Lee et al. [27] determined that postural muscle action begins earlier than focal muscle action above a certain threshold velocity of arm movement and that the difference in onset timing shows no significant change up to a maximum velocity. We had subjects perform arm movement at maximum speed in the present study to exclude the factor of arm movement dynamics in the start of muscle action. Before the start of data recording, the subjects were given 10 practice trials of arm movement. The measurement of arm movement was performed under three conditions: (1) the quiet standing condition, (2) the reproduced condition, and (3) the vibration condition. Measurement (1) was performed five times; thereafter, a set of measurements including (2) and (3) was performed five times. In the quiet standing condition (Fig. 1A), the mean position of CFP fluctuation was measured in subjects for 10 s. The trial was repeated five times with a 30-s period of seated rest between trials. The mean value for five trials was calculated and adopted as the representative CFP position during quiet standing. The subjects maintained the CFP within a range of ±1 cm of the mean position in quiet standing for 3 s and then started arm movement. Five trials of arm movement were carried out with a 30-s period of seated rest between trials. In the reproduced condition (Fig. 1B), the subjects maintained a quiet standing posture for 3 s with the trunk fixed by the stopper. Next, vibration stimulation was applied to the Achilles’ tendon at the same time when the buzzing sound indicating a quiet standing position to the subjects was stopped. The vibration was applied for 30 s, and the subjects were instructed to memorize the perceived standing position for the last 3 s. They sat on a chair immediately after the vibration stimulation was stopped. After sitting for 3 s, they reproduced the memorized standing position for 3 s and then started arm movement. In the vibration condition (Fig. 1C), the subjects maintained quiet standing for 3 s with the trunk fixed by the stopper. Next, the vibration stimulation was applied to the Achilles’ tendon at the same time when the buzzing sound was stopped. The vibration was applied for 30 s, and the subjects were instructed to memorize the perceived standing position for the last 3 s. Thereafter, they started arm movement.
computer (NEC, PC-9821V233) via an A/D converter (Canopus, ADJ-98) at 1000 Hz with 12-bit resolution. Subsequent analyses were performed using BIMUTAS II software (Kissei Comtec Co. Ltd.). Fig. 2 shows representative data during the vibration condition. The time course of muscle action in each trial was analyzed by visual inspection on a computer screen and the EMG amplitudes were analyzed as follows. Electromyograms of postural muscles were band-pass filtered (20–100 Hz) using the Butterworth method and then were full-wave rectified. The mean amplitude of EMG was computed for a period of ⫺2300 to ⫺300 ms with respect to the action onset of the anterior deltoid (D0) and that value was adopted as background activity. Muscle activity 2 SD above background activity lasting at least 50 ms in a period of ⫺150 to +100 ms with respect to D0 was identified as an increase in postural muscle activity (the first EMG burst), and the time difference between burst onset and D0 was measured as the start time of the first EMG burst. In the cases where the start time of the first EMG burst for postural muscles could not be measured due to large background activity, the start time was defined as when the rectified EMG deviated more than the mean + 2 SD of background activity within a period of ⫺150 to +100 ms with respect to D0. The mean amplitude of EMG of the postural muscle was computed during the first 50 ms after burst onset and that value was adopted as the activation magnitude of the postural muscle. The feedback information from the peripheral nerve was not used in this duration [1]. The anterior deltoid was then processed in the same manner as the postural muscle, with the duration and mean
2.4. Data analysis The data of force platform and muscle activity were reproduced from a data recorder and sent to a separate
Fig. 2. Representative data in the vibration condition. The arrow mark indicates the start time of the first EMG burst.
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
amplitude from D0 to peak amplitude calculated in order to determine the state of activation of the focal muscles that rapidly move the arm. The magnitude of illusionary perception of standing position was evaluated by the difference between the mean CFP positions that were maintained for 3 s before arm movement in the quiet standing and reproduced conditions. The distance of shift in CFP position in response to vibration stimulation was evaluated by the difference in the mean CFP positions for 3 s between just before the vibration and last (27–30 s) during the vibration. 2.5. Statistical analysis One-way repeated-measures analysis of variance was used to study the effect of the trial and experimental condition in each parameter of EMG activity and CFP position. Post-hoc multiple-comparison analysis using the Newman–Keuls procedures was performed to examine the differences suggested by the analysis of variance. When no significant effect of trial was found in each parameter, the mean value across five trials was used as a representative value for the experimental condition; otherwise, the value in each trial was used as the representative value. A paired t-test was used to compare the CFP positions evaluated in various conditions. The alpha level was set at p ⬍ 0.05. All statistics were calculated using Excel 2000 (Microsoft Corp.) with StatMate III (ATMS Co. Ltd.).
3. Results The CFP position during maintaining quiet standing was 42.9% FL (SD = 5.7). In the reproduced condition, the reproduced CFP position was not significantly affected by the trial, and thus the mean value across five trials was adopted as the representative value for each subject. The mean CFP position for 12 subjects was 63.6% FL (SD = 6.9), and it was located further forward by 20.6% FL (SD = 7.7) compared with that during the quiet standing condition (p ⬍ 0.001). For the period of the vibration, the CFP position shifted slightly but significantly forward (p ⬍ 0.05). The shift distance was 1.7%
513
FL (SD = 0.7) in the reproduced condition and 2.9% FL (SD = 1.3) in the vibration condition. When the arm movement was performed during vibration, some subjects leaned their body backward and the position of the stopper was shifted to the backmost limit. Five subjects showed this phenomenon in the first trial and this number decreased as the trials were repeated (three in the second and third trials, two in the fourth trial, and one in the fifth trial). The duration and mean amplitude from D0 to peak amplitude of the anterior deltoid did not show a significant effect of trial or experimental condition (Table 1). The analysis of the effect of experimental condition for each subject showed that in very few subjects (one or three subjects), the duration and mean amplitude in the reproduced and vibration conditions differed significantly from that in the quiet standing condition. The first EMG burst of the postural muscles was observed in the rear of the body in all trials. Muscle activation in the front of the body followed that in the rear of the body. Fig. 3 shows the start time of the biceps femoris and erector spinae. The biceps femoris in the vibration condition began activating earliest in the first trial, and the start time was significantly later as the trials were repeated (F 4,44 = 5.33, p ⬍ 0.001). When the start time of the biceps femoris in the first trial under the vibration condition was adopted as a representative value for that condition, a significant effect of experimental condition on start time was found (F 2,22 = 22.4, p ⬍ 0.001). Post-hoc analyses indicated that the start time in the vibration condition (⫺71.1 ± 31.7 ms in the first trial) did not differ significantly from that in the reproduced condition (⫺70.9 ± 23.5 ms) but was significantly earlier than that in the quiet standing condition (⫺33.7 ± 21.7 ms, p ⬍ 0.001). When the start time of the biceps femoris in the fifth trial under the vibration condition (⫺29.9 ± 23.8 ms) was adopted as a representative value, a significant effect of experimental condition on start time was found (F 2,22 = 19.8, p ⬍ 0.001). Post-hoc analyses indicated that it did not differ significantly from the representative value in the quiet standing condition, but it did show a significant difference from that in the reproduced condition (p ⬍ 0.001).
Table 1 Duration and mean amplitude from D0 to peak amplitude in EMG of the anterior deltoid in each experimental condition n = 12
Duration (ms) Mean amplitude (µ V/ms)
Mean SD Mean SD
Experimental condition Quiet standing
Reproduction
Vibration
132 28 290 154
136 45 231 121
149 35 238 152
514
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
Fig. 3. Start time in the biceps femoris and erector spinae. Values in the quiet standing condition and reproduction condition are averaged across five trials. Values in the vibration condition are for each of the five trials. Asterisks indicate significant differences (p ⬍ 0.001) relative to the quiet standing condition.
Fig. 4. Means and standard deviations of background activity in the biceps femoris and erector spinae. Asterisks indicate significant differences (p ⬍ 0.05) relative to the quiet standing condition.
Fig. 5. Means and standard deviations of activation magnitude in the biceps femoris and erector spinae.
In the erector spinae, no significant effect of experimental condition on start time was found (⫺45.4 ± 11.8 ms in the quiet standing condition, ⫺51.1 ± 25.9 ms in the vibration condition, and ⫺57.9 ± 18.0 ms in the reproduced condition). Background activity in the biceps femoris and erector spinae is shown in Fig. 4. No significant effect of trial was found in either postural muscle. The effect of experimental condition was significant in both muscles (biceps femoris: F 2,22 = 4.91, p ⬍ 0.05; erector spinae: F 2,22 = 3.98, p ⬍ 0.05), and the background activity in the reproduced condition was significantly greater than that in the quiet standing condition (p ⬍ 0.05). Activation magnitude in the biceps femoris and erector spinae is shown in Fig. 5. There were no significant effects of trial and experimental condition in either postural muscle.
4. Discussion 4.1. Illusionary perception of standing position in response to vibration The CFP position shifted forward by 2% or 3% FL during vibration of the Achilles’ tendon, and the background activity of the biceps femoris and erector spinae showed no significant difference between the conditions of quiet standing and vibration. These results suggest that the momentum around the ankle joint changed only a little in response to vibration, and the muscle sense information from the postural muscles of the trunk and thigh did not change. Therefore, the illusionary perception of standing position may be produced mainly by muscle sense information from the triceps surae. Additionally, it is possible that the information from
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
mechanoreceptors that relate to the momentum change around the foot joint causes the illusionary perception. The reproduced positions were located forward by about 20% FL of the foot length compared with the quiet standing position. This means that the illusion that the body is leaning forward by about 20% FL is generated during vibration. Many researchers have reported that when the triceps surae is vibrated with the trunk fixed by a stopper, the illusion that the body is leaning forward is generated [23,30,31]. The illusionary perception of standing position showed no significant differences among the five trials. In contrast, the modality of postural control changed noticeably as the trial of arm flexion during the vibration was repeated. It has been found that the body schema in the perceptual reference system exists for a while in a changeless state under the weightless condition and after its experience [6,30]. The present result supports this finding. However, Roll et al. [30] clarified that the reference frame was not fixed but recomposed, because on the 21st space flight day, the illusion created by vibration stimulation of the leg muscles during trunk fixation disappeared. We will examine in future studies the transformation of the illusionary perception of standing position over the long term. 4.2. Activity of postural muscle The earliest onset of muscle activation was observed in the biceps femoris and erector spinae in most trials. Many studies have reported that during arm flexion movement, the postural muscles in the rear begin activating before the focal muscle [1,5,11,25]. In this study, therefore, the focus of analysis was on the biceps femoris and erector spinae. 4.2.1. EMG amplitude The mean amplitudes of EMG in both muscles during arm flexion movement showed no significant difference among the three experimental conditions, suggesting that the magnitude control of postural muscle activity was similarly performed in the three conditions. Many researchers have reported that the magnitude of postural muscle activation is a function of the magnitude in postural demand, which changes according to the size of the support base surface or the magnitude of postural disturbance by arm movement [1,5,7,26,27,33]. In the present study, the size of support base surface was constant in every experimental condition. In addition, the time duration and mean amplitude from D0 to peak activity in the deltoid showed no significant differences among three conditions. Thus, it is possible that the degree of physical disturbance caused by arm movement does not differ significantly among the experimental conditions. Even if the magnitude of physical disturbance is held constant, the disturbance effect increases
515
in parallel with leaning the body forward, because the forward movable range of CFP decreases in the forward leaning posture. It is suggested that the inclination degree in this study was not the factor influencing the relative magnitude of disturbance. 4.2.2. Timing of anticipatory postural control The biceps femoris started to activate earlier in the reproduced condition and the first trial in the vibration condition than in the quiet standing condition. The start time showed no significant difference between the former two conditions. In the first trial in the vibration condition, the subjects encounter a novel situation that is not previously experienced. In the first vibration trial, the subjects undoubtedly use the body schema that has been used until then, without changing it. As a result, it is possible that the postural control is similar to that of the forward leaning posture, based on the sensory information of standing position during vibration. The disturbance of posture and/or balance associated with arm movement may be anticipated, and a postural control program should be prepared. The preceding activation of the postural muscles is not observed when a subject’s body is attached with restrainers to a wall [3,7,11]. One explanation given for this phenomenon is that no disturbance of equilibrium associated with arm movement was anticipated. Horak et al. [19] clarified that equilibrium is never disturbed during slow speed arm movements, and in such cases, the necessity for preceding activation of the postural muscles decreased. Aruin and Latash [1] suggested that the preceding activation of the biceps femoris and erector spinae for postural control functioned to resist the forward transition of the center of gravity and to maintain it within the support base in the foot. Kasai and Taga [21] reported that the force exerted with arm movement influenced the EMG onset of postural muscle and postural stability. Judging from these reports, it may be appropriate to presume that movement of the CFP on the support due to arm movement was anticipated and the activation timing of the postural muscle was adjusted. For the erector spinae, the activation timing showed no significant difference among experimental conditions, in contrast to the case of the biceps femoris. This suggests that a control principle different from the biceps femoris acts in the erector spinae. Muscle sensory information from the trunk, in every subject, is thought to correlate with sensory information from the neck muscles and the vestibular organs, and perhaps to play a role in the perception of the direction of gravity [16]. Massion [28] divided postural adjustment into balance control over the entire body and the positional control of specific body parts in order to provide a reference axis for action. It is possible that the muscle activity of the biceps femoris strongly affected the first type of postural adjust-
516
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
ment, and that of the erector spinae activity affected the second. The present results likely reflect the same kind of functional difference in the sensory–motor system between the trunk and leg.
[8]
[9]
4.2.3. Adaptive change in the timing of muscle activity As mentioned in the Introduction, it has been observed that the postural control under the non-daily condition changes adaptively in a short time to a new behavioral requirement, based on the knowledge of the control result [20,24,29,30,32]. Under the vibration condition in the present study, the preceding time of the first EMG burst in the biceps femoris significantly decreased with advances in the trial, and this value became almost equal to that in the quiet standing condition until the fifth trial. This change is possibly generated due to a recognition of the need to alternate the inappropriate postural control in the first trial of the vibration condition. In this case, the appropriate postural control will be that in the real standing position, which is the quiet standing position. In fact, in the first trial of vibration, arm movement caused a marked disturbance of body balance, and many subjects shifted up to the movable limit for the position of the stopper. The number of subjects showing this disturbance decreased as the trials advanced, indicating that the modality of muscle activity in the postural muscle is adjusted adaptively in order to maintain equilibrium based on the knowledge of the control result. In support of the preceding argument, it has been definitely shown that the activation timing of the biceps femoris is mainly regulated based on the perception of initial standing position, and that postural control is adjusted adaptively in order to maintain equilibrium based on the knowledge of the control result.
[10]
[11]
[12]
[13]
[14]
[15] [16]
[17]
[18]
[19]
[20]
References [1] A.S. Aruin, M.L. Latash, Directional specificity of postural muscles in feed-forward postural reactions during fast voluntary arm movement, Exp. Brain Res. 103 (1995) 323–332. [2] A.S. Aruin, M.L. Latash, The role of motor action in anticipatory postural adjustments studied with self-induced and externally triggered perturbations, Exp. Brain Res. 106 (1995) 291–300. [3] V.E. Belen’kii, V.S. Gurfinkel, R.L. Pal’tsev, On elements of control of voluntary movements, Biofizika 12 (1967) 135–141. [4] F. Benvenuti, S.J. Stanhope, S.L. Thomas, V.P. Panzer, M. Hallett, Flexibility of anticipatory postural adjustments revealed by self-paced and reaction-time arm movements, Brain Res. 761 (1997) 59–70. [5] S. Bouisset, M. Zattara, Anticipatory postural adjustments and dynamic asymmetry of voluntary movement, in: V.S. Gurfinkel, M.E. Ioffe, J. Massion, J.P. Roll (Eds.), Stance and Motion: Facts and Concepts, Plenum Press, New York, 1988, pp. 177–183. [6] G. Cle´ ment, V.S. Gurfinkel, F. Lestienne, M.I. Lipshits, K.E. Popov, Adaptation of postural control to weightlessness, Exp. Brain Res. 57 (1984) 61–72. [7] P.J. Cordo, L.M. Nashner, Properties of postural adjustments
[21]
[22]
[23]
[24]
[25] [26]
[27]
associated with rapid arm movements, J. Neurophysiol. 47 (1982) 287–302. S. De Wolf, H. Slijper, M.L. Latash, Anticipatory postural adjustments during self-paced and reaction-time movements, Exp. Brain Res. 121 (1998) 7–19. G. Eklund, General features of vibration-induced effects on balance, Upsala J. Med. Sci. 77 (1972) 112–124. W.G. Friedli, L. Cohen, M. Hallett, S. Stanhope, S.R. Simon, Postural adjustments associated with rapid voluntary arm movements. II. Biomechanical analysis, J. Neurol. Neurosurg. Psychiatry 51 (1988) 232–243. W.G. Friedli, M. Hallett, S.R. Simon, Postural adjustments associated with rapid voluntary arm movements. 1. Electromyographic data, J. Neurol. Neurosurg. Psychiatry 47 (1984) 611– 622. K. Fujiwara, H. Toyama, H. Asai, T. Yamashina, Effects of a sagittal position of the body gravity center and manual weightload on postural control during rapid arm-lifting, Jpn. J. Phys. Fitness Sports Med. 40 (1991) 355–364. K. Fujiwara, H. Toyama, K. Kunita, H. Asai, A. Miyaguchi, Modality of postural movement in men and women with both arms flexed during standing, Percept. Mot. Skills 93 (2001) 611–625. G.M. Goodwin, D.I. Mccloskey, P.B.C. Matthews, The contribution of muscle afferents to kinaesthesia shown by vibration induced illusions of movement and by the effects of paralyzing joint afferents, Brain 95 (1972) 705–748. K. Goshima, Studies on items and normal ranges in stabilometry, Equilibrium Res. 45 (1986) 368–387. V.S. Gurfinkel, Y.S. Levik, K.E. Popov, B.N. Smetanin, V.Y. Shlikov, Body scheme in the control of postural activity, in: V.S. Gurfinkel, M.E. Loffe, J. Massion, J.P. Roll (Eds.), Stance and Motion, Plenum Press, New York, 1988, pp. 185–193. V.S. Gurfinkel, Y.S. Levick, Perceptual and automatic aspects of the postural body scheme, in: J. Paillard (Ed.), Brain and Space, Oxford University Press, New York, 1991, pp. 147–162. V.S. Gurfinkel, Y.P. Ivanenko, Y.S. Levick, The influence of head rotation on human upright posture during balanced bilateral vibration, Neuroreport 7 (1995) 137–140. F.B. Horak, P. Esselman, M.E. Anderson, M.K. Lynch, The effects of movement velocity, mass displaced, and task certainty on associated postural adjustments made by normal and hemiplegic individuals, J. Neurol. Neurosurg. Psychiatry 47 (1984) 1020–1028. F.B. Horak, L.M. Nashner, Central programming of postural movements: adaptation to altered support-surface configurations, J. Neurophysiol. 55 (1986) 1369–1381. T. Kasai, T. Taga, Effects of varying load conditions on the organization of postural adjustments during voluntary arm flexion, J. Mot. Behav. 24 (1992) 359–365. T. Kasai, S. Yahagi, K. Shimura, Effects of vibration-induced postural illusion on anticipatory postural adjustment of voluntary arm movement in standing humans, Gait Posture 15 (2002) 94–100. J.R. Lackner, M.S. Levine, Changes in apparent body orientation and sensory localization induced by vibration of postural muscles, Aviat. Space Environ. Med. 50 (1979) 346–354. C.S. Layne, B.S. Spooner, Effects of postural set on anticipatory muscle activation prior to rapid arm flexion, Res. Q. Exerc. Sport 63 (1992) 196–199. W.A. Lee, Anticipatory control of postural and task muscles during rapid arm flexion, J. Mot. Behav. 12 (1980) 185–196. W.A. Lee, C.F. Michaels, Y.C. Pai, The organization of torque and EMG activity during bilateral handle pulls by standing humans, Exp. Brain Res. 82 (1990) 304–314. W.A. Lee, T.S. Buchanan, M.W. Rogers, Effects of arm acceleration and behavioral conditions on the organization of postural
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
[28] [29] [30]
[31]
[32]
[33]
[34]
adjustments during arm flexion, Exp. Brain Res. 66 (1987) 257–270. J. Massion, Movement, posture and equilibrium: interaction and coordination, Prog. Neurobiol. 38 (1992) 35–56. L.M. Nashner, Adapting reflexes controlling the human posture, Exp. Brain Res. 26 (1976) 59–72. J.P. Roll, K. Popov, V. Gurfinkel, M. Lipshits, C. Andre´ -Deshays, J.C. Gilhodes, C. Quoniam, Sensorimotor and perceptual function of muscle proprioception in microgravity, J. Vestibular Res. 3 (1993) 259–273. J.P. Roll, J.P. Vedel, R. Roll, Eye, head and skeletal muscle spindle feedback in the elaboration of body references, in: J.H.J. Allum, M. Hulliger (Eds.), Progress in Brain Research, vol. 80, Elsevier Science Publishers, Amsterdam, 1989, pp. 113–123. A. Struppler, L. Gerilovsky, C. Jakob, Self-generated rapid taps directed to the opposite forearm in man: anticipatory reduction in the muscle activity of the target arm, Neurosci. Lett. 159 (1993) 115–118. S. Vernazza, M. Cincera, A. Pedotti, J. Massion, Balance control during lateral arm raising in humans, Neuroreport 7 (1996) 1543–1548. M.H. Woollacott, M. Bonnet, K. Yabe, Preparatory process for anticipatory postural adjustments: modulation of leg muscles reflex pathways during preparation for arm movements in standing man, Exp. Brain Res. 55 (1984) 263–271.
Katsuo Fujiwara received his Ph.D. from Tsukuba University in 1984. He is professor at the Department of Human Movement and Health, Graduate School of Medical Science, Kanazawa University (2001 to present). He is a member of the International Society of Electrophysiological Kinesiology, the International Society of Posture and Gait Research, and the Society for Neuroscience. He serves as permanent director of the Japanese Society of Health and Behavior Science. He is now analyzing the anticipatory processes of postural control in the brain by using electromyograms, evoked potentials, and event-related potentials.
517
Kaoru Maeda received his master’s degree in health science from Kanazawa University in 2002. Since 2002, he has been a doctoral student at Kanazawa University. He is currently performing research on the relationship between postural control and contingent negative variation.
Hiroshi Toyama received his Ph.D. from Kananazawa University in 1997. He is associate professor at the Department of Human Movement and Health, the Graduate School of Medicine, Kanazawa University (2001 to present). His current research interest is the automatization of human motor control. He is a member of the International Society of Postural and Gait Research, International Society of Electrophysiological Kinesiology, the International Brain Research Organization, etc.
Influences of illusionary position perception on anticipatory postural control associated with arm flexion Katsuo Fujiwara ∗, Kaoru Maeda, Hiroshi Toyama Department of Human Movement and Health, Graduate School of Medical Science, Kanazawa University, 13-1 Takara-machi, Kanazawa 920-8640, Japan Received 31 May 2002; received in revised form 23 April 2003; accepted 7 June 2003
Abstract We examined the effect of illusionary perception on anticipatory postural control associated with arm flexion with subjects in a standing position, using vibration stimulation of the Achilles’ tendon. Arm flexion was performed five times under each of the following conditions: (1) quiet standing, (2) vibration of the Achilles’ tendon at 100 Hz frequency and 1.5 mm amplitude with the trunk fixed by a stopper during quiet standing, and (3) a perceived standing position during vibration. The reproduced positions were located forward by about 20% of the foot length compared with the quiet standing position; these positions showed no significant differences among the five trials. In the first trial of arm flexion during vibration, the biceps femoris began activating approximately 40 ms before the anterior deltoid. The same time difference between activation of the two muscles was observed in the reproduced condition. As the vibration trials were repeated, this activation timing approached the value in the quiet standing condition. In both the biceps femoris and erector spinae, the mean amplitude of electromyogram for the first 50 ms after the start of activation did not differ significantly among the three conditions. 2003 Elsevier Ltd. All rights reserved. Keywords: Anticipatory postural control; Muscle activity; Vibration stimulation; Illusionary perception
1. Introduction Many previous studies have shown that the onset of activation of the postural muscles of the legs and trunk that control standing postures precedes that of the focal muscles that rapidly move the arm [3,7]. The preceding activation of the postural muscles is adjusted by a program that is selected in advance in order to moderate the effect of disturbances of posture and equilibrium caused by the arm movement [2,5,10,11,19,28,34]. When both arms are rapidly flexed to the front, body balance is disturbed by the increase in rotational momentum of the body in the anterior direction. It has been demonstrated that the postural muscles in the rear of the body begin activating before the focal muscles, and that this preceding activation leads to the enhancement of preparatory
Corresponding author. Tel.: +81-76-265-2225; fax: +81-76-2344219. E-mail address: [email protected] (K. Fujiwara). ∗
1050-6411/$ - see front matter 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1050-6411(03)00083-X
muscle tension in the postural muscles [10,34] and/or the postural movement to compensate for the forward transition of the center of gravity [13,20]. Many studies have found that the muscle action sequence of focal and postural muscles differs significantly according to behavioral condition, and particularly in self-paced arm movement, the postural muscles show a large preceding activation [4,8,11,19]. The activation timing of postural muscles is probably affected by the initial standing position just before arm movement [5]. Benvenuti et al. [4] found that muscle activation of the postural muscles begins earlier in a forward leaning posture with the center of foot pressure (CFP) at the metatarsal head than in a quiet standing posture. We previously reported that when subjects performed arm flexion movement with the initial CFP maintained at 30% of foot length from the heel, the CFP moved nearer to the quiet standing position, and preceding action of the postural muscles was not observed [12]. Therefore, we presume that activation of postural muscles changes according to the perception of initial standing position.
510
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
Goodwin et al. [14] reported that when the limb movement originated by a tonic vibratory reflex was obstructed, the illusion of reversal movement was produced. Eklund [9] observed that vibration of the Achilles’ tendon in a standing posture generated a reflex action in the gastrocnemius, and the body then leaned backward. It has also been found that when the leaning was obstructed, the subject perceived an illusory leaning forward of the body [23,30,31]. It has been reported that these illusions are caused by Ia impulses from muscle spindles being sent to the reference system in the neural center where they are collated with the body schema. In addition, it has been found that this illusionary perception of limb position influences the organization of motor action [17]. Gurfinkel et al. [18] reported that vibration stimulation of the neck muscles unilaterally with the head fixed relative to the trunk evoked an illusion of rotation of the head to the contralateral side, and a shift of the entire body laterally according to the illusion. This finding indicates that the illusionary perception of posture affects the organization of postural control. Kasai et al. [22] applied vibration to the tibialis anterior (TA) and soleus (Sol) of human subjects who were performing arm movement in a free-standing posture with their eyes closed. They observed that when vibration was applied to the tibialis anterior, the preceding action of the biceps femoris (BF) to the anterior deltoid (AD) began earlier than that in the free-standing posture, and that when applied to the soleus, it began later. In their study, body movement was not obstructed, and thus the subject leaned forward and backward when the vibration was applied to the tibialis anterior and soleus, respectively. These results were consistent with previous reports, in which the preceding action of biceps femoris began earlier in a forward leaning posture [4] and later in backward leaning [12]. Therefore, it is not clear whether the cause of changes in anticipatory postural control in response to vibration is an illusionary perception or an actual shift of standing position. If anticipatory postural control is affected by an illusionary perception of standing position, postural control in the illusionary perception should be similar to that with the standing position shifting actually to an illusionary position. Nashner [29] reported that abrupt inclination of a standing floor produces a stretch reflex in the triceps surae, and if the reflex does not work effectively for body balance, it is suppressed as the floor inclination is repeated. Furthermore, it has been reported that following a change in the condition of standing floor [20], the aspect of postural disturbance [32], or the gravity condition [6,24,30], the effectiveness of postural control is judged from control results and postural control is adjusted adaptively. In the present study, it is presumed that a similar change takes place. That is, in response to vibration of the Achilles’ tendon, an illusionary percep-
tion at which the body leans forward is produced and the preceding activation of postural muscles is then developed. This activation will lead to an unstable backward leaning of the body, and the preceding activation will be suppressed gradually as the unstable trial is repeated. In the present study, we investigated the effect of illusionary perception on anticipatory postural control associated with arm flexion with subjects in a standing position with their body fixed during vibration stimulation of the Achilles’ tendon. The working hypotheses were as follows: (1) activation of postural muscles will change based on the illusionary perception of standing position, (2) activation will adaptively change based on knowledge of the control result.
2. Methods 2.1. Subjects Large individual differences in the effect of vibration stimulation on positional perception have been observed [23]. We selected subjects in the preliminary test who clearly showed backward leaning of the body when mechanical vibration was applied to the Achilles’ tendon bilaterally, and who showed generation of an illusionary perception at which the body leaned forward in response to vibration when the trunk was fixed by a stopper. Subjects were nine men and three women, aged 19–42 years (mean 27.0 ± SD 0.7). All subjects appeared to be free of any neurological and orthopedic impairment. Informed consent was obtained from all subjects following an explanation of the experimental protocol. The mean values of height, weight, and foot length were 170.3 cm (SD = 5.2), 65.1 kg (SD = 7.7), and 25.4 cm (SD = 0.9), respectively. 2.2. Apparatus All measurements were taken with subjects standing on a force platform (WAMI, WA1001) composed of three load-cells. The pressure force signals (F1, F2, F3, Fig. 1A) detected on the load-cells were passed through an analog operation electric circuit, and the position (CFPy) of the CFP in the anteroposterior direction was calculated according to the formula shown below: CFPy ⫽ {(F2 ⫹ F3) / (F1 ⫹ F2 ⫹ F3)}·l (1: distance from an F1 load-cell to a straight line that connects between load-cells of F2 and F3, 400 mm). The electronic CFP signal was sent to a buzzer generator (HIRUTA, F-H6408) and an A/D converter (I/ODATA, PIO9045) in a computer (NEC, PC9801BX2).
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
511
Fig. 1. Experimental protocol. (A) Quiet standing condition. (B) Reproduced condition. (C) Vibration condition.
The buzzer generator was used to indicate the quiet standing position to the subject. In this apparatus, 1 cm displacement of CFP was calibrated to 0.1 V. The sound was generated when the CFP position was located within ±1 cm of a mean position during quiet standing. Goshima [15] reported that the SD of CFP fluctuation in an anteroposterior direction during quiet standing for 60 s was about 0.5 cm for normal subjects (aged 20 years). Therefore, we decided that the standard range of CFP fluctuation during quiet standing was 1.0 cm, which corresponded to 2 SD. The A/D converter was set at 200 Hz sampling rate with 12-bit resolution. Using the converted digital data, the mean position of CFP fluctuation was calculated and was shown by the relative distance (% FL) from the heel to the length of the foot. Mechanical vibration was bilaterally applied to the Achilles’ tendon through the skin using two vibrators. Each vibrator (Heiwa Electronic Industrial Co., TMT18) was independently strapped to the ankle region with a rubber belt. The vibration frequency was set at 100 Hz, as in a previous report [31], with an amplitude of 1.5 mm. Surface electrodes (Medicotest, M-00-S) were used in a bipolar derivation to record surface EMG activity of the following muscles: the anterior deltoid, rectus abdominis (RA) at the level of the navel, erector spinae (ES) at the level of the iliac crest, rectus femoris (RF), the long head of the biceps femoris, tibialis anterior, and soleus, on the left side. For each muscle, the electrodes were fixed over the muscle belly after shaving and cleaning the skin with alcohol. The electrodes were aligned along the great axis of the muscle with an inter-electrode
distance of about 2 cm. The electrode input impedance was reduced to below 5 k⍀. Signals from the electrodes were amplified (×2000) and band-pass filtered (1.6 Hz– 1.5 kHz) with an EMG amplifier (NEC-Sanei, BIOTOP-6R12). A stopper was set at the back of the subject’s body, with the position being level with the second to fourth thoracic vertebra, to prevent backward leaning of the body while vibrating the Achilles’ tendon. To confirm that the body would tend to incline backward in response to the vibration in the absence of a stopper, the pressure on the stopper was measured by a load-cell (KYOWA, LUR-A-50NSA1). The stopper was fixed by an electromagnet, and was released upon triggering an EMG burst onset of the anterior deltoid. The maximum backward displacement of the stopper was 14 cm behind exceeding the hindmost point of the subject’s heel, which prevented the subjects from falling. The sliding position of the stopper was measured from a sagittal view recorded using a high-speed camera (Photron, Fastcam rabbit mini 2) with a frame rate of 120 Hz. For subsequent analysis, signals from the force platform and EMG electrodes were recorded on a digital tape recorder (TEAC, RD-200E). The bandwidth of the recorder was 0–2.5 kHz. 2.3. Procedure Blindfolded subjects performed the following tasks while standing with their bare feet 10 cm apart and parallel with one another. The feet were aligned accurately with a reference board for the foot position. Subjects
512
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
grasped a grip bar (40 cm in length and 68 g in weight) with their hands and their arms were positioned close to the trochanter majors, and their elbows extended. Subjects started moving both arms at their own pace within 3 s after an oral cue from the examiner, and then stopped the arms at a frontal horizontal level position. This position was then held for 3 s (Fig. 1). Lee et al. [27] determined that postural muscle action begins earlier than focal muscle action above a certain threshold velocity of arm movement and that the difference in onset timing shows no significant change up to a maximum velocity. We had subjects perform arm movement at maximum speed in the present study to exclude the factor of arm movement dynamics in the start of muscle action. Before the start of data recording, the subjects were given 10 practice trials of arm movement. The measurement of arm movement was performed under three conditions: (1) the quiet standing condition, (2) the reproduced condition, and (3) the vibration condition. Measurement (1) was performed five times; thereafter, a set of measurements including (2) and (3) was performed five times. In the quiet standing condition (Fig. 1A), the mean position of CFP fluctuation was measured in subjects for 10 s. The trial was repeated five times with a 30-s period of seated rest between trials. The mean value for five trials was calculated and adopted as the representative CFP position during quiet standing. The subjects maintained the CFP within a range of ±1 cm of the mean position in quiet standing for 3 s and then started arm movement. Five trials of arm movement were carried out with a 30-s period of seated rest between trials. In the reproduced condition (Fig. 1B), the subjects maintained a quiet standing posture for 3 s with the trunk fixed by the stopper. Next, vibration stimulation was applied to the Achilles’ tendon at the same time when the buzzing sound indicating a quiet standing position to the subjects was stopped. The vibration was applied for 30 s, and the subjects were instructed to memorize the perceived standing position for the last 3 s. They sat on a chair immediately after the vibration stimulation was stopped. After sitting for 3 s, they reproduced the memorized standing position for 3 s and then started arm movement. In the vibration condition (Fig. 1C), the subjects maintained quiet standing for 3 s with the trunk fixed by the stopper. Next, the vibration stimulation was applied to the Achilles’ tendon at the same time when the buzzing sound was stopped. The vibration was applied for 30 s, and the subjects were instructed to memorize the perceived standing position for the last 3 s. Thereafter, they started arm movement.
computer (NEC, PC-9821V233) via an A/D converter (Canopus, ADJ-98) at 1000 Hz with 12-bit resolution. Subsequent analyses were performed using BIMUTAS II software (Kissei Comtec Co. Ltd.). Fig. 2 shows representative data during the vibration condition. The time course of muscle action in each trial was analyzed by visual inspection on a computer screen and the EMG amplitudes were analyzed as follows. Electromyograms of postural muscles were band-pass filtered (20–100 Hz) using the Butterworth method and then were full-wave rectified. The mean amplitude of EMG was computed for a period of ⫺2300 to ⫺300 ms with respect to the action onset of the anterior deltoid (D0) and that value was adopted as background activity. Muscle activity 2 SD above background activity lasting at least 50 ms in a period of ⫺150 to +100 ms with respect to D0 was identified as an increase in postural muscle activity (the first EMG burst), and the time difference between burst onset and D0 was measured as the start time of the first EMG burst. In the cases where the start time of the first EMG burst for postural muscles could not be measured due to large background activity, the start time was defined as when the rectified EMG deviated more than the mean + 2 SD of background activity within a period of ⫺150 to +100 ms with respect to D0. The mean amplitude of EMG of the postural muscle was computed during the first 50 ms after burst onset and that value was adopted as the activation magnitude of the postural muscle. The feedback information from the peripheral nerve was not used in this duration [1]. The anterior deltoid was then processed in the same manner as the postural muscle, with the duration and mean
2.4. Data analysis The data of force platform and muscle activity were reproduced from a data recorder and sent to a separate
Fig. 2. Representative data in the vibration condition. The arrow mark indicates the start time of the first EMG burst.
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
amplitude from D0 to peak amplitude calculated in order to determine the state of activation of the focal muscles that rapidly move the arm. The magnitude of illusionary perception of standing position was evaluated by the difference between the mean CFP positions that were maintained for 3 s before arm movement in the quiet standing and reproduced conditions. The distance of shift in CFP position in response to vibration stimulation was evaluated by the difference in the mean CFP positions for 3 s between just before the vibration and last (27–30 s) during the vibration. 2.5. Statistical analysis One-way repeated-measures analysis of variance was used to study the effect of the trial and experimental condition in each parameter of EMG activity and CFP position. Post-hoc multiple-comparison analysis using the Newman–Keuls procedures was performed to examine the differences suggested by the analysis of variance. When no significant effect of trial was found in each parameter, the mean value across five trials was used as a representative value for the experimental condition; otherwise, the value in each trial was used as the representative value. A paired t-test was used to compare the CFP positions evaluated in various conditions. The alpha level was set at p ⬍ 0.05. All statistics were calculated using Excel 2000 (Microsoft Corp.) with StatMate III (ATMS Co. Ltd.).
3. Results The CFP position during maintaining quiet standing was 42.9% FL (SD = 5.7). In the reproduced condition, the reproduced CFP position was not significantly affected by the trial, and thus the mean value across five trials was adopted as the representative value for each subject. The mean CFP position for 12 subjects was 63.6% FL (SD = 6.9), and it was located further forward by 20.6% FL (SD = 7.7) compared with that during the quiet standing condition (p ⬍ 0.001). For the period of the vibration, the CFP position shifted slightly but significantly forward (p ⬍ 0.05). The shift distance was 1.7%
513
FL (SD = 0.7) in the reproduced condition and 2.9% FL (SD = 1.3) in the vibration condition. When the arm movement was performed during vibration, some subjects leaned their body backward and the position of the stopper was shifted to the backmost limit. Five subjects showed this phenomenon in the first trial and this number decreased as the trials were repeated (three in the second and third trials, two in the fourth trial, and one in the fifth trial). The duration and mean amplitude from D0 to peak amplitude of the anterior deltoid did not show a significant effect of trial or experimental condition (Table 1). The analysis of the effect of experimental condition for each subject showed that in very few subjects (one or three subjects), the duration and mean amplitude in the reproduced and vibration conditions differed significantly from that in the quiet standing condition. The first EMG burst of the postural muscles was observed in the rear of the body in all trials. Muscle activation in the front of the body followed that in the rear of the body. Fig. 3 shows the start time of the biceps femoris and erector spinae. The biceps femoris in the vibration condition began activating earliest in the first trial, and the start time was significantly later as the trials were repeated (F 4,44 = 5.33, p ⬍ 0.001). When the start time of the biceps femoris in the first trial under the vibration condition was adopted as a representative value for that condition, a significant effect of experimental condition on start time was found (F 2,22 = 22.4, p ⬍ 0.001). Post-hoc analyses indicated that the start time in the vibration condition (⫺71.1 ± 31.7 ms in the first trial) did not differ significantly from that in the reproduced condition (⫺70.9 ± 23.5 ms) but was significantly earlier than that in the quiet standing condition (⫺33.7 ± 21.7 ms, p ⬍ 0.001). When the start time of the biceps femoris in the fifth trial under the vibration condition (⫺29.9 ± 23.8 ms) was adopted as a representative value, a significant effect of experimental condition on start time was found (F 2,22 = 19.8, p ⬍ 0.001). Post-hoc analyses indicated that it did not differ significantly from the representative value in the quiet standing condition, but it did show a significant difference from that in the reproduced condition (p ⬍ 0.001).
Table 1 Duration and mean amplitude from D0 to peak amplitude in EMG of the anterior deltoid in each experimental condition n = 12
Duration (ms) Mean amplitude (µ V/ms)
Mean SD Mean SD
Experimental condition Quiet standing
Reproduction
Vibration
132 28 290 154
136 45 231 121
149 35 238 152
514
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
Fig. 3. Start time in the biceps femoris and erector spinae. Values in the quiet standing condition and reproduction condition are averaged across five trials. Values in the vibration condition are for each of the five trials. Asterisks indicate significant differences (p ⬍ 0.001) relative to the quiet standing condition.
Fig. 4. Means and standard deviations of background activity in the biceps femoris and erector spinae. Asterisks indicate significant differences (p ⬍ 0.05) relative to the quiet standing condition.
Fig. 5. Means and standard deviations of activation magnitude in the biceps femoris and erector spinae.
In the erector spinae, no significant effect of experimental condition on start time was found (⫺45.4 ± 11.8 ms in the quiet standing condition, ⫺51.1 ± 25.9 ms in the vibration condition, and ⫺57.9 ± 18.0 ms in the reproduced condition). Background activity in the biceps femoris and erector spinae is shown in Fig. 4. No significant effect of trial was found in either postural muscle. The effect of experimental condition was significant in both muscles (biceps femoris: F 2,22 = 4.91, p ⬍ 0.05; erector spinae: F 2,22 = 3.98, p ⬍ 0.05), and the background activity in the reproduced condition was significantly greater than that in the quiet standing condition (p ⬍ 0.05). Activation magnitude in the biceps femoris and erector spinae is shown in Fig. 5. There were no significant effects of trial and experimental condition in either postural muscle.
4. Discussion 4.1. Illusionary perception of standing position in response to vibration The CFP position shifted forward by 2% or 3% FL during vibration of the Achilles’ tendon, and the background activity of the biceps femoris and erector spinae showed no significant difference between the conditions of quiet standing and vibration. These results suggest that the momentum around the ankle joint changed only a little in response to vibration, and the muscle sense information from the postural muscles of the trunk and thigh did not change. Therefore, the illusionary perception of standing position may be produced mainly by muscle sense information from the triceps surae. Additionally, it is possible that the information from
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
mechanoreceptors that relate to the momentum change around the foot joint causes the illusionary perception. The reproduced positions were located forward by about 20% FL of the foot length compared with the quiet standing position. This means that the illusion that the body is leaning forward by about 20% FL is generated during vibration. Many researchers have reported that when the triceps surae is vibrated with the trunk fixed by a stopper, the illusion that the body is leaning forward is generated [23,30,31]. The illusionary perception of standing position showed no significant differences among the five trials. In contrast, the modality of postural control changed noticeably as the trial of arm flexion during the vibration was repeated. It has been found that the body schema in the perceptual reference system exists for a while in a changeless state under the weightless condition and after its experience [6,30]. The present result supports this finding. However, Roll et al. [30] clarified that the reference frame was not fixed but recomposed, because on the 21st space flight day, the illusion created by vibration stimulation of the leg muscles during trunk fixation disappeared. We will examine in future studies the transformation of the illusionary perception of standing position over the long term. 4.2. Activity of postural muscle The earliest onset of muscle activation was observed in the biceps femoris and erector spinae in most trials. Many studies have reported that during arm flexion movement, the postural muscles in the rear begin activating before the focal muscle [1,5,11,25]. In this study, therefore, the focus of analysis was on the biceps femoris and erector spinae. 4.2.1. EMG amplitude The mean amplitudes of EMG in both muscles during arm flexion movement showed no significant difference among the three experimental conditions, suggesting that the magnitude control of postural muscle activity was similarly performed in the three conditions. Many researchers have reported that the magnitude of postural muscle activation is a function of the magnitude in postural demand, which changes according to the size of the support base surface or the magnitude of postural disturbance by arm movement [1,5,7,26,27,33]. In the present study, the size of support base surface was constant in every experimental condition. In addition, the time duration and mean amplitude from D0 to peak activity in the deltoid showed no significant differences among three conditions. Thus, it is possible that the degree of physical disturbance caused by arm movement does not differ significantly among the experimental conditions. Even if the magnitude of physical disturbance is held constant, the disturbance effect increases
515
in parallel with leaning the body forward, because the forward movable range of CFP decreases in the forward leaning posture. It is suggested that the inclination degree in this study was not the factor influencing the relative magnitude of disturbance. 4.2.2. Timing of anticipatory postural control The biceps femoris started to activate earlier in the reproduced condition and the first trial in the vibration condition than in the quiet standing condition. The start time showed no significant difference between the former two conditions. In the first trial in the vibration condition, the subjects encounter a novel situation that is not previously experienced. In the first vibration trial, the subjects undoubtedly use the body schema that has been used until then, without changing it. As a result, it is possible that the postural control is similar to that of the forward leaning posture, based on the sensory information of standing position during vibration. The disturbance of posture and/or balance associated with arm movement may be anticipated, and a postural control program should be prepared. The preceding activation of the postural muscles is not observed when a subject’s body is attached with restrainers to a wall [3,7,11]. One explanation given for this phenomenon is that no disturbance of equilibrium associated with arm movement was anticipated. Horak et al. [19] clarified that equilibrium is never disturbed during slow speed arm movements, and in such cases, the necessity for preceding activation of the postural muscles decreased. Aruin and Latash [1] suggested that the preceding activation of the biceps femoris and erector spinae for postural control functioned to resist the forward transition of the center of gravity and to maintain it within the support base in the foot. Kasai and Taga [21] reported that the force exerted with arm movement influenced the EMG onset of postural muscle and postural stability. Judging from these reports, it may be appropriate to presume that movement of the CFP on the support due to arm movement was anticipated and the activation timing of the postural muscle was adjusted. For the erector spinae, the activation timing showed no significant difference among experimental conditions, in contrast to the case of the biceps femoris. This suggests that a control principle different from the biceps femoris acts in the erector spinae. Muscle sensory information from the trunk, in every subject, is thought to correlate with sensory information from the neck muscles and the vestibular organs, and perhaps to play a role in the perception of the direction of gravity [16]. Massion [28] divided postural adjustment into balance control over the entire body and the positional control of specific body parts in order to provide a reference axis for action. It is possible that the muscle activity of the biceps femoris strongly affected the first type of postural adjust-
516
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
ment, and that of the erector spinae activity affected the second. The present results likely reflect the same kind of functional difference in the sensory–motor system between the trunk and leg.
[8]
[9]
4.2.3. Adaptive change in the timing of muscle activity As mentioned in the Introduction, it has been observed that the postural control under the non-daily condition changes adaptively in a short time to a new behavioral requirement, based on the knowledge of the control result [20,24,29,30,32]. Under the vibration condition in the present study, the preceding time of the first EMG burst in the biceps femoris significantly decreased with advances in the trial, and this value became almost equal to that in the quiet standing condition until the fifth trial. This change is possibly generated due to a recognition of the need to alternate the inappropriate postural control in the first trial of the vibration condition. In this case, the appropriate postural control will be that in the real standing position, which is the quiet standing position. In fact, in the first trial of vibration, arm movement caused a marked disturbance of body balance, and many subjects shifted up to the movable limit for the position of the stopper. The number of subjects showing this disturbance decreased as the trials advanced, indicating that the modality of muscle activity in the postural muscle is adjusted adaptively in order to maintain equilibrium based on the knowledge of the control result. In support of the preceding argument, it has been definitely shown that the activation timing of the biceps femoris is mainly regulated based on the perception of initial standing position, and that postural control is adjusted adaptively in order to maintain equilibrium based on the knowledge of the control result.
[10]
[11]
[12]
[13]
[14]
[15] [16]
[17]
[18]
[19]
[20]
References [1] A.S. Aruin, M.L. Latash, Directional specificity of postural muscles in feed-forward postural reactions during fast voluntary arm movement, Exp. Brain Res. 103 (1995) 323–332. [2] A.S. Aruin, M.L. Latash, The role of motor action in anticipatory postural adjustments studied with self-induced and externally triggered perturbations, Exp. Brain Res. 106 (1995) 291–300. [3] V.E. Belen’kii, V.S. Gurfinkel, R.L. Pal’tsev, On elements of control of voluntary movements, Biofizika 12 (1967) 135–141. [4] F. Benvenuti, S.J. Stanhope, S.L. Thomas, V.P. Panzer, M. Hallett, Flexibility of anticipatory postural adjustments revealed by self-paced and reaction-time arm movements, Brain Res. 761 (1997) 59–70. [5] S. Bouisset, M. Zattara, Anticipatory postural adjustments and dynamic asymmetry of voluntary movement, in: V.S. Gurfinkel, M.E. Ioffe, J. Massion, J.P. Roll (Eds.), Stance and Motion: Facts and Concepts, Plenum Press, New York, 1988, pp. 177–183. [6] G. Cle´ ment, V.S. Gurfinkel, F. Lestienne, M.I. Lipshits, K.E. Popov, Adaptation of postural control to weightlessness, Exp. Brain Res. 57 (1984) 61–72. [7] P.J. Cordo, L.M. Nashner, Properties of postural adjustments
[21]
[22]
[23]
[24]
[25] [26]
[27]
associated with rapid arm movements, J. Neurophysiol. 47 (1982) 287–302. S. De Wolf, H. Slijper, M.L. Latash, Anticipatory postural adjustments during self-paced and reaction-time movements, Exp. Brain Res. 121 (1998) 7–19. G. Eklund, General features of vibration-induced effects on balance, Upsala J. Med. Sci. 77 (1972) 112–124. W.G. Friedli, L. Cohen, M. Hallett, S. Stanhope, S.R. Simon, Postural adjustments associated with rapid voluntary arm movements. II. Biomechanical analysis, J. Neurol. Neurosurg. Psychiatry 51 (1988) 232–243. W.G. Friedli, M. Hallett, S.R. Simon, Postural adjustments associated with rapid voluntary arm movements. 1. Electromyographic data, J. Neurol. Neurosurg. Psychiatry 47 (1984) 611– 622. K. Fujiwara, H. Toyama, H. Asai, T. Yamashina, Effects of a sagittal position of the body gravity center and manual weightload on postural control during rapid arm-lifting, Jpn. J. Phys. Fitness Sports Med. 40 (1991) 355–364. K. Fujiwara, H. Toyama, K. Kunita, H. Asai, A. Miyaguchi, Modality of postural movement in men and women with both arms flexed during standing, Percept. Mot. Skills 93 (2001) 611–625. G.M. Goodwin, D.I. Mccloskey, P.B.C. Matthews, The contribution of muscle afferents to kinaesthesia shown by vibration induced illusions of movement and by the effects of paralyzing joint afferents, Brain 95 (1972) 705–748. K. Goshima, Studies on items and normal ranges in stabilometry, Equilibrium Res. 45 (1986) 368–387. V.S. Gurfinkel, Y.S. Levik, K.E. Popov, B.N. Smetanin, V.Y. Shlikov, Body scheme in the control of postural activity, in: V.S. Gurfinkel, M.E. Loffe, J. Massion, J.P. Roll (Eds.), Stance and Motion, Plenum Press, New York, 1988, pp. 185–193. V.S. Gurfinkel, Y.S. Levick, Perceptual and automatic aspects of the postural body scheme, in: J. Paillard (Ed.), Brain and Space, Oxford University Press, New York, 1991, pp. 147–162. V.S. Gurfinkel, Y.P. Ivanenko, Y.S. Levick, The influence of head rotation on human upright posture during balanced bilateral vibration, Neuroreport 7 (1995) 137–140. F.B. Horak, P. Esselman, M.E. Anderson, M.K. Lynch, The effects of movement velocity, mass displaced, and task certainty on associated postural adjustments made by normal and hemiplegic individuals, J. Neurol. Neurosurg. Psychiatry 47 (1984) 1020–1028. F.B. Horak, L.M. Nashner, Central programming of postural movements: adaptation to altered support-surface configurations, J. Neurophysiol. 55 (1986) 1369–1381. T. Kasai, T. Taga, Effects of varying load conditions on the organization of postural adjustments during voluntary arm flexion, J. Mot. Behav. 24 (1992) 359–365. T. Kasai, S. Yahagi, K. Shimura, Effects of vibration-induced postural illusion on anticipatory postural adjustment of voluntary arm movement in standing humans, Gait Posture 15 (2002) 94–100. J.R. Lackner, M.S. Levine, Changes in apparent body orientation and sensory localization induced by vibration of postural muscles, Aviat. Space Environ. Med. 50 (1979) 346–354. C.S. Layne, B.S. Spooner, Effects of postural set on anticipatory muscle activation prior to rapid arm flexion, Res. Q. Exerc. Sport 63 (1992) 196–199. W.A. Lee, Anticipatory control of postural and task muscles during rapid arm flexion, J. Mot. Behav. 12 (1980) 185–196. W.A. Lee, C.F. Michaels, Y.C. Pai, The organization of torque and EMG activity during bilateral handle pulls by standing humans, Exp. Brain Res. 82 (1990) 304–314. W.A. Lee, T.S. Buchanan, M.W. Rogers, Effects of arm acceleration and behavioral conditions on the organization of postural
K. Fujiwara et al. / Journal of Electromyography and Kinesiology 13 (2003) 509–517
[28] [29] [30]
[31]
[32]
[33]
[34]
adjustments during arm flexion, Exp. Brain Res. 66 (1987) 257–270. J. Massion, Movement, posture and equilibrium: interaction and coordination, Prog. Neurobiol. 38 (1992) 35–56. L.M. Nashner, Adapting reflexes controlling the human posture, Exp. Brain Res. 26 (1976) 59–72. J.P. Roll, K. Popov, V. Gurfinkel, M. Lipshits, C. Andre´ -Deshays, J.C. Gilhodes, C. Quoniam, Sensorimotor and perceptual function of muscle proprioception in microgravity, J. Vestibular Res. 3 (1993) 259–273. J.P. Roll, J.P. Vedel, R. Roll, Eye, head and skeletal muscle spindle feedback in the elaboration of body references, in: J.H.J. Allum, M. Hulliger (Eds.), Progress in Brain Research, vol. 80, Elsevier Science Publishers, Amsterdam, 1989, pp. 113–123. A. Struppler, L. Gerilovsky, C. Jakob, Self-generated rapid taps directed to the opposite forearm in man: anticipatory reduction in the muscle activity of the target arm, Neurosci. Lett. 159 (1993) 115–118. S. Vernazza, M. Cincera, A. Pedotti, J. Massion, Balance control during lateral arm raising in humans, Neuroreport 7 (1996) 1543–1548. M.H. Woollacott, M. Bonnet, K. Yabe, Preparatory process for anticipatory postural adjustments: modulation of leg muscles reflex pathways during preparation for arm movements in standing man, Exp. Brain Res. 55 (1984) 263–271.
Katsuo Fujiwara received his Ph.D. from Tsukuba University in 1984. He is professor at the Department of Human Movement and Health, Graduate School of Medical Science, Kanazawa University (2001 to present). He is a member of the International Society of Electrophysiological Kinesiology, the International Society of Posture and Gait Research, and the Society for Neuroscience. He serves as permanent director of the Japanese Society of Health and Behavior Science. He is now analyzing the anticipatory processes of postural control in the brain by using electromyograms, evoked potentials, and event-related potentials.
517
Kaoru Maeda received his master’s degree in health science from Kanazawa University in 2002. Since 2002, he has been a doctoral student at Kanazawa University. He is currently performing research on the relationship between postural control and contingent negative variation.
Hiroshi Toyama received his Ph.D. from Kananazawa University in 1997. He is associate professor at the Department of Human Movement and Health, the Graduate School of Medicine, Kanazawa University (2001 to present). His current research interest is the automatization of human motor control. He is a member of the International Society of Postural and Gait Research, International Society of Electrophysiological Kinesiology, the International Brain Research Organization, etc.