| | Effects of preparatory period on anticipatory postural control and contingent negative variation associated with rapid arm movement in standing postureReceived 17 May 2005; received in revised form 20 January 2006; accepted 20 January 2006. Abstract We investigated CNS motor preparation state and anticipatory postural muscle activation while subjects performed bilateral rapid arm movement at various intervals between warning and response stimulus (preparatory period) during standing. Motor preparation state was evaluated by integrated values of the late components of the contingent negative variation (late CNV), obtained by averaging electroencephalograms during the last 100 ms of the preparatory period. For quantifying anticipatory postural muscle activation, we measured the onset of burst activity in postural muscles (lumbar paraspinal, biceps femoris, and gastrocnemius) with respect to anterior deltoid activity and integrated values of preceding activation. Subjects performed the arm movement with minimal delay in the warning stimulus–response stimulus–motor response paradigm under preparatory periods of 2.0, 3.0 and 3.5s. Late CNV did not differ between the 2.0-s and 3.0-s period, but was significantly smaller in the 3.5-s period than in the 2.0-s period, suggesting difficulty in predicting response timing in the 3.5-s period. No change was found on integrated values of preceding activations of postural muscles. Burst onset of all postural muscles significantly preceded anterior deltoid activation in all periods. Burst activity for gastrocnemius only occurred earlier in the 3.5-s period than in the 2.0-s and 3.0-s periods. Weak correlations were observed between late CNV and onset time of gastrocnemius activity. It is suggested that earlier activation of gastrocnemius is a strategy adopted when response stimulus timing is relatively difficult to predict. 1. Introduction  In upper limb movement that would cause disequilibrium in a direction anterior to standing posture, postural muscle activation related to postural control of the trunk and legs precedes activation of agonists in the upper limb movement [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]. Preceding activation of postural muscles is reportedly performed based on predictions of postural perturbation caused by upper limb movement [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]. This is supported by studies demonstrating no preceding activation of postural muscles under conditions where postural equilibrium is undisturbed by upper limb movement, such as in a backward-leaning position [4], microgravity [8] or supine position [8], and by studies demonstrating earlier activation of muscles under conditions where center of foot pressure is shifted beyond the base of support by upper limb movement [1], [5]. Preceding activation of postural muscles has been observed in upper limb movement for self-paced conditions, but appears absent under reaction-time conditions [3], [7]. Under randomized conditions of presentation timing of response stimulus, preceding activation of postural muscles is not observed [3], [10]. Thus, when initiation of movement is easily predicted and/or postural control is in a state of readiness, preceding activation of postural muscles should also be observed under reaction-time conditions. However, few studies have systematically examined the effects of predictable presentation of response stimulus on anticipatory postural control associated with upper limb movement using various periods of movement preparation (preparation period) while standing. In addition, few studies have investigated relationships between anticipatory postural control and an index of motor preparation, such as electroencephalograms (EEGs). Gurfinkel and El’ner [6] reported that the supplementary motor area was closely related to anticipatory postural control associated with upper limb movement. They demonstrated that preceding activation of postural muscles was not observed in subjects with lesions of the supplementary motor area. Investigation of the relationship between activation state of the brain area related to anticipatory postural control and preceding activation of postural muscles associated with upper limb movement while standing should demonstrate that preceding activation of postural muscles is related to the motor preparation state. The effects of presentation timing of response stimulus on reaction time have previously been investigated using a paradigm of warning stimulus/response stimulus/motor response [11], [12], [13], [14], [15], [16], [17], [18]. When the preparation period is constant across trials and only a single response to the response stimulus is required, reaction time reportedly increases with preparation period [19]. Sato et al. [20] demonstrated that, in a task testing reaction time in response to periodic signals, when the period exceeded about 3s, timing prediction of signal presentation became difficult and style of response changed from anticipatory to reactive. These reports suggest that preceding activation of postural muscles is not observed when preparation period exceeds 3s, as difficulties in timing prediction of response stimulus prevent the state of motor preparation from reaching readiness. Brain activity during the preparation period in the above-mentioned paradigm has been investigated in subjects in a sitting position using contingent negative variation (CNV), which is obtained by averaging EEGs during the period [11], [12], [13], [14], [15], [16], [17], [18], [21], [22], [23]. Sources of CNV reportedly include the supplementary motor area [21], [22], [23], premotor area [21], [23] and prefrontal cortex [21], [23]. One factor reflected in the late component of CNV that appears at Cz from about 1s before response stimulus is the state of motor preparation. For example, the amplitude of late CNV is markedly larger under speed conditions than under accuracy conditions [11], [15], and displays positive correlations with speed of reaction movement [12], [13], [17]. McAdam et al. [16] demonstrated significantly decreased CNV amplitude and increased reaction time associated with elongation of the preparation period. This suggests that decreased CNV amplitude corresponds to increased difficulty in predicting response timing according to preparation period. Thus, amplitude of late CNV presumably decreases when preparation period exceeds around 3s. Amplitude of late CNV is presumably closely related to preceding time of postural muscles with respect to focal muscles. The present study investigated relationships between cerebral state of preparation for movement during the preparation period, qualitatively estimated using late CNV, and anticipatory postural control. Working hypotheses were as follows: (1) amplitude of late CNV will be smaller for preparatory periods exceeding 3s than for periods less than 3s; (2) preceding activation of postural muscles will become unobservable for preparatory periods exceeding 3s; (3) changing patterns of onset timing for postural muscles according to length of preparatory period will correlate with changing patterns of late CNV. 2. Methods 2.1. Subjects Subjects comprised 11 men and 7 women, with a mean age of 22.4±2.6 years (range, 19–29 years). No subjects displayed any history of neurological or orthopedic impairment. Informed consent was obtained from all subjects following an explanation of experimental protocols. On the day of experiment, from waking until end of the experiment, subjects were prohibited from consuming food or drink containing caffeine, such as coffee [14]. Mean height, weight and foot length were 165.6±6.6cm (range, 155.0–178.0cm), 59.5±7.0kg (range, 48–71kg) and 24.7±1.4cm (range, 22.2–27.1cm), respectively. 2.2. Equipment All measurements were taken while subjects were standing barefoot, with feet 10cm apart and parallel on an OR6-6 force platform (AMTI, USA). The force platform was used to measure position of the center of foot pressure in the anteroposterior direction (CFPy). Electronic momentum signal was sent to an analog CFPy calculator (Electro-design, Japan) and used to determine CFPy according to the following formula: where My is the moment in a sagittal plane and Fz is the vertical force component. CFPy data were then sent to an A/D converter (PIO9045; I/O-DATA, Japan) in a computer (PC9801BX2; NEC, Japan) with a 20 Hz sampling rate and 12-bit resolution. CFPy position was then calculated and shown as relative distance (%) from the heel compared to foot length. A buzz sound was generated by the computer when CFPy was located within ±1 cm of a mean position during quiet standing. Goshima [24] reported that standard deviation for CFPy fluctuations during quiet standing for 60 s was about 0.5 cm for normal subjects (mean age, 20 years). We therefore set the standard range of CFPy fluctuation during quiet standing at 1.0 cm, corresponding to two standard deviations. Hands of subjects were fixed to a wooden board (17 mm × 37 mm × 1.7 mm, weight: 0.7 kg), with both elbows flexed to 60°. Height of both hands was 10 cm below the shoulder joint ( Fig. 1). The board was suspended from a pole by metal wire, and a load was attached to the center of the board. Total weight of the board was set at 3% of body weight for men and 2% of body weight for women. Range of upper limb movement was from the initial height to shoulder height. In a preliminary experiment, cross-spectrum analysis for normalized EEG and electromyogram (EMG) of shoulder girdle muscles before response stimulus (a series of each data were transferred to obtain a mean of 0 and a variance of 1) confirmed that product of amplitudes corresponding to a frequency where the maximum square of EMG was obtained (51 Hz) was about 4% of the maximum square, indicated that EEG was hardly contaminated by EMG. Surface electrodes (M-00-S; Medicotest, Denmark) were arranged in bipolar configuration to record surface EMGs for the following muscles: anterior deltoid (AD); rectus abdominis (RA) at the level of the navel; lumbar paraspinal muscles (LP) at the level of the iliac crest; rectus femoris (RF); the long head of biceps femoris (BF); tibialis anterior (TA); the medial head of gastrocnemius (GcM). All recordings were made right-sided. Electrodes were aligned along the major axis of the muscle with an inter-electrode distance of about 3cm. Input impedance was reduced to <5kΩ. Signals from electrodes were amplified (×2000) and band-pass filtered (1.6–1.5kHz) using an EMG amplifier (BIOTOP-6R12; NEC-Sanei, Japan). Signals from the amplifier were sent to a computer (PC-9821V233; NEC) via an A/D converter (ADJ-98; CANOPUS, Japan) with sampling frequency of 1000Hz and 12-bit resolution. CNV was recorded using an ER1204 evoked potential system (NEC). Silver–silver chloride cup electrodes (diameter, 8mm) for CNV recording were affixed to the scalp at Fz, Cz and Pz in accordance with the international 10–20 system, and referred to linked ear lobes. A ground electrode was placed on the Fpz. Electrode input impedance was reduced to <5kΩ. Electro-oculograms (EOG) were recorded from a pair of electrodes placed above and below the left eye, to monitor artifacts accompanying blinking. To fix eye position, subjects were instructed to gaze at a fixation point presented on an Eye-trek face-mounted display (Olympus, Japan). Two tone-bursts were presented as warning stimulus (S1) and response stimulus (S2). The interval between stimuli was defined as the preparatory period, and was set at 2.0, 3.0 or 3.5s. The early component of CNV reportedly persists until 1000–1500ms after presentation of S1 [18]. We therefore isolated the late component of the CNV by applying a preparatory period of ≥2.0s. Frequency, duration and intensity of auditory stimuli were 2kHz, 100ms and 50dB above hearing threshold, respectively. Electroencephalographic data were band-pass filtered from 0.05 to 60Hz, and data were analyzed from 500ms before S1 to 1000ms after S2. CNV was amplified (×50,000) and sampled at 250Hz using the evoked potential system. 2.3. Procedure CFPy position was initially measured for 10s while subjects maintained quiet standing with upper limbs by their sides. A total of five trials were performed, with a 30-s period of seated rest between each trial. Mean values for the five trials were adopted as the representative CFPy position during quiet standing. Next, both hands were strapped to the wooden board with both elbows flexed at 60°, and subjects were instructed to keep muscles around the shoulder as relaxed as possible. CFPy position in this posture was also measured, and we ensured that no significant difference existed between mean values of CFPy position in each posture. After maintaining quiet standing for ≥3s, S1 was presented. S2 was then presented after the designated preparatory period. Subjects flexed the shoulders as fast as possible to minimize response time to onset of S2, stopping the hands at shoulder level. This position was then held for about 3s. Speed of movement was emphasized over accuracy. Lee et al. [9] noted that postural muscle action begins earlier than focal muscle action above a certain threshold velocity of arm movement, and that differences in onset timing display no significant changes up to a maximum velocity. To familiarize subjects with the task, 10 practice trials for each preparatory period were performed before measurement trials. A trial block in one preparation period comprised 30 trials with artifact-free EEGs. After completing the block, the next trial block for another preparation period was performed. Order of trial blocks was randomized for each subject. After obtaining measurements, subjects were asked in which preparatory period the timing of stimulus presentation was the most difficult to predict. 2.4. Data analysis Analyses of CFPy and EMG were performed using BIMUTAS II software (Kissei Comtec, Japan). Fig. 2 shows representative data for EMG during arm movement. The time course of EMG was analyzed for each trial. Trials with AD onset of ≤100ms in response to S2 were not considered performances of the reaction time task by the studies of middle latency stretch reflexes [25] and a number of reaction time studies [26], and were not used in the following analyses. 2.6. CNV analysis CNVs were obtained by averaging EEG separately for each preparatory period. Baseline for CNV was the 500ms segment just before S1 presentation. EEGs coinciding with artifacts >100μV before S2 were discarded from averaging. Only averaged waves were used in the following analyses of CNV. Integrated value of late CNV was calculated for the period from 100ms before S2 presentation (integrated late CNV). To investigate relationships between integrated late CNV and start of postural muscle activity, both values were normalized for each subject using the following formula: 2.7. CFP analysis Mean CFPy was calculated for the 150-ms period before S1 and 300–150ms before D0. 2.8. Statistical analysis The effects of preparatory period on AD reaction time, rate of premature AD reaction time, integrated late CNV, ʃEMG and the effect of postural muscles on start time of first EMG burst were assessed in the following process. First, Bartlett's test was used to assess differences in variance for each above-mentioned parameter of EMG and CNV. Next, depending on whether significant differences in variance were observed or not, one-way analysis of variance (ANOVA) or Friedman test was used to assess the effects, respectively. Post hoc multiple-comparison analysis was performed using Newman–Keuls procedures or the Tukey test to further examine differences suggested by ANOVA or Friedman test, respectively. Two-way ANOVA was used to study the effects of preparatory period and integration interval on mean amplitude of background EMG activity. When significant effects of integration intervals were shown, differences between BGS1 and BGbefore D0 were assessed using paired t-tests. A Student t-test was used to determine whether start time of first EMG burst differed from zero. A Pearson correlation was used to test for relationships between start time of first EMG burst and integrated late CNV. Alpha level was set at p<0.05. All statistical analyses were performed using Excel 2000 (Microsoft, Japan) with Stat Mate III (ATMS, Japan). 3. Results No significant effect of preparatory period was identified on reaction time of AD to S2 (2.0s: 168.1±46.1ms; 3.0s: 170.9±49.2ms; 3.5s: 169.7±47.5ms; F2,34=0.154, p=0.86). Trials with the premature AD onset (≤100ms) were excluded from the analysis of reaction time. Means and standard deviations of a percentage of the trials with premature AD onset across subjects were 4.8±8.3%, 2.4±3.9% and 3.4±6.4% in the preparatory period of 2.0, 3.0 and 3.5s, respectively. The effect of preparatory period on the rate of premature AD onset was not significant (F2,34=0.68, p=0.51). Integrated late CNV showed maximum amplitude at Cz in every preparatory period for all subjects when comparing electrode positions. Subsequent results for CNV were thus based on waveforms derived from Cz. Fig. 3 shows a representative CNV waveform for a single subject. A significant effect of preparatory period on integrated late CNV was observed (F2,34=4.80, p=0.015). Integrated late CNV was significantly smaller in the 3.5-s period than in the 2.0-s period (p<0.05) (Fig. 4). All subjects answered that predicting timing of response stimulus was most difficult for the 3.5-s preparatory period. The first EMG burst of postural muscles was observed in the rear of the body in all trials for all subjects. Burst activity of front muscles occurred subsequently. Burst onset of all rear muscles significantly preceded AD in all preparatory periods (2.0-s period: t17<−2.92, p<0.01; 3.0-s period: t17<−2.70, p<0.05; 3.5-s period: t17<−2.71, p<0.05). Most subjects showed the earliest burst in GcM for all preparatory periods (2.0-s period: 66.7%; 3.0-s period: 72.2%; 3.5-s period: 72.2%). Start times of first EMG burst for LP, BF and GcM are shown in Fig. 5. A significant effect of preparatory period was observed only on start time of GcM (F2,34=4.33, p=0.021). Start time of GcM was significantly earlier in the 3.5-s period than in the 2.0-s and 3.0-s periods (2.0-s: −41.9±36.0ms; 3.0-s: −39.8±39.7ms; 3.5-s: −47.8±39.3ms; p<0.05). A significant effect of preparatory period was observed on start time of postural muscle activity showing the earliest onset of burst in each period (F2,34=4.02; p=0.027). Start time of earliest muscle activity was significantly earlier in the 3.5-s period than in the 3.0-s period (p<0.05) (Fig. 6). The ʃEMG of GcM was calculated for 16, 17, and 17 subjects in the 2.0-s, 3.0-s and 3.5-s preparatory periods, respectively. Data for 16 subjects were thus used for ANOVA. No significant effect of preparatory period was found on ʃEMG of GcM (F2,30=0.61, p=0.55) (Fig. 7). A significant effect of integration intervals was found on mean amplitude of background EMG activity (integration interval: F2,51=18.3, p<0.0001; preparatory period: F2,51=0.22, p=0.80; integration intervals×preparatory period: F2,51=0.28, p=0.76). BGbefore D0 was significantly larger than BGS1 for 3.0-s and 3.5-s preparatory periods (2.0s: t17=1.90, p=0.075; 3.0s: t17=3.19, p=0.005; 3.5s: t17=2.52, p=0.022). Positions of CFPy in the two calculated intervals in all preparation periods were almost identical, with a maximum difference of 3mm. A weak significant correlation was found between start time of GcM burst and integrated value of late CNV (r=0.28, p=0.038) (Fig. 8a). A weak significant correlation was also found between start time of the earliest postural muscles and integrated late CNV (r=0.33, p=0.014) (Fig. 8b). 4. Discussion Reaction time of AD did not change according to preparation period. Generally, studies have reported that reaction time increases with increased preparation period [27], [28]. The result of reaction time in this study differed from the tendencies reported previously. One possible contributor to this difference was that a constant preparatory period was adopted in each trial block without inserting catch trials. The purpose of the present study was to investigate relationships between timing prediction of response stimulus and motor preparation based on this prediction. The experiment was performed under the hypothesis that predicting timing of response stimulus is difficult even without catch trials. To satisfy the present experimental condition for application of the reaction task, data with premature AD reaction time (≤100ms) were excluded from analyses. Data analysis based on the present experimental condition showed that the integrated value of late CNV was significantly smaller in the 3.5-s period than in the 2.0-s period. In addition, all subjects answered that the 3.5-s preparatory period was the most difficult to predict timing of response stimulus. In a task testing reaction time in response to periodic signals, prediction reportedly became difficult for subjects and style of response changed from anticipatory to reactive when period exceeded about 3s [20]. McAdam et al. [16] reported that timing prediction of response stimulus was more difficult with a 4.8-s preparatory period than with a 1.6-s preparatory period, corresponding to smaller amplitude of late CNV. From these reports, timing prediction of response stimulus would be difficult with a 3.5-s preparation period. We hypothesized above that preceding activation of postural muscles becomes difficult to observe when preparatory period exceeds 3s. However, the present results cannot support this hypothesis. Contrary to the hypothesis, start time of burst activity from GcM was earlier in the 3.5-s preparatory period than in the 3.0-s period. This suggests that adopted strategies for postural control differ between 3.5-s and 3.0-s preparatory periods. That is, even if presentation timing of response stimulus is difficult to predict, subjects should adopt the strategy of activating postural muscles earlier without abandoning preceding activation of muscles. Change in onset timing of preceding activation of postural muscles with respect to AD was observed most prominently in GcM. In the report by Bouisset and Zattara [2], in the task of bilateral upper limb movement of rapid shoulder flexion from 0° to 90° in the standing posture, preceding burst activities with respect to AD were not observed in the lower leg. Fujiwara et al. [5] also reported that preceding activation of postural muscles was less observable in the lower leg than in the trunk and thigh. Although no explanation was given for that phenomenon in those studies, one possible explanation is that effectiveness of postural control at the ankle joint differed between the present study and previous investigations. Comparing the upper limb movements applied, range of rapid arm movement was much smaller in the present study than in previous studies, in which shoulder flexion from 0° to 90° was applied. The larger range of shoulder flexion leads to longer periods of acceleration for upper limbs. Maximum acceleration of the limbs during arm movement must thus be larger in the previous studies than in the present study, causing larger forces in the direction opposing gravity. In this case, force acting on the floor via the sole should be decreased during postural control at the ankle joint, leading to reduced effectiveness of control at the ankle. Subjects presumably adopted strategies by which another part of the body was used for postural control. Horak and Nashner [29] reported that strategy of postural control changes according to the positional relationship between support surface and center of gravity during the maintenance of postural equilibrium. They demonstrated that when the support surface is narrow or unstable, muscles around the hip joint form the agonists of postural control, probably because postural control at the ankle is less effective. In the present study, postural control at the ankle joint was presumably effective because anti-gravitational forces generated by upper limb movement should be small, so preceding activation of GcM was observed. Background GcM activity was significantly larger at 300–150ms before AD burst than in the 150-ms period before warning stimulus, while no significant effect of preparatory period was found. This would represent a common strategy of postural muscle activity to react to postural disturbances caused by upper limb movement. This type of change in muscle activity has been categorized as preparatory postural adjustment that is differentiated from preceding burst activity associated with upper limb movement [9], [30]. In addition, in the 3.5-s period, contributions of prolonged preceding period compared with the 3.0-s period to magnitude of preceding activity would be very small. A significant but weak correlation was found between start time of GcM burst and integrated late CNV. Such weakness could be due to the fact that multiple factors other than motor preparation are reflected in late CNV, such as expectation, attention, motivation and cognition. Conversely, in postural muscle activities, the possibility exists that parameters such as length of preceding period, magnitude of tonus and magnitude of activity are controlled strategically. Individual differences in relationships between these above-mentioned factors are presumably the cause of weak correlations. 5. Conclusion Integrated value of late CNV was significantly smaller in the 3.5-s preparatory period than in the 2.0-s period, suggesting that difficulty of timing prediction was increased with the 3.5-s period. Conversely, postural muscles showed burst activity significantly earlier with the 3.5-s period than with the 2.0-s period, indicating that the applied strategy of postural control differed between these two periods. These results suggest that the postural control strategy adopted when timing prediction of response stimulus become difficult differs from the strategy used when prediction is relatively easy. References [1]. [1]Benvenuti F, Stanhope SJ, Thomas SL, Panzer VP, Hallett M. Flexibility of anticipatory postural adjustments revealed by self-paced and reaction-time arm movements. Brain Res. 1997;761:59–70. MEDLINE |
CrossRef
[2]. [2]Bouisset S, Zattara M. Anticipatory postural adjustments and dynamic asymmetry of voluntary movement. In: Gurfinkel VS, Ioffe ME, Massion J, Roll JP editor. Stance and motion: facts and concepts. New York: Plenum Press; 1988;p. 177–183. [3]. [3]De Wolf S, Slijper H, Latash ML. Anticipatory postural adjustments during self-paced and reaction-time movements. Exp Brain Res. 1998;121:7–19. MEDLINE |
CrossRef
[4]. [4]Fujiwara K, Toyama H, Asai H, Yamashina T. Effects of a sagittal position of the body gravity center and manual weight load on postural control during rapid arm-lifting. Jpn J Phys Fitness Sports Med. 1991;40:355–364. [5]. [5]Fujiwara K, Toyama H, Kunita K. Anticipatory activation of postural muscles associated with bilateral arm flexion in subjects with different quiet standing positions. Gait Posture. 2003;17:254–263. Abstract | Full Text |
Full-Text PDF (284 KB)
[6]. [6]Gurfinkel VS, El’ner AM. Participation of the secondary motor area of the frontal lobe of the brain in organizing postural components of human voluntary movement. Neirofiziologiia. 1988;20(1):7–15. MEDLINE [7]. [7]Horak FB, Esselman P, Anderson ME, Lynch MK. The effects of movement velocity, mass displaced, and task certainty on associated postural adjustments made by normal and hemiplegic individuals. J Neurol Neurosurg Psychiat. 1984;47:1020–1028. MEDLINE |
CrossRef
[8]. [8]Layne CS, Spooner BS. Effects of postural set on anticipatory muscle activation prior to rapid arm flexion. Res Quart Exerc Sport. 1992;63(2):196–199. [9]. [9]Lee WA, Michaels CF, Pai YC. The organization of torque and EMG activity during bilateral handle pulls by standing humans. Exp Brain Res. 1990;82:304–314. MEDLINE [10]. [10]Woollacott MH, Bonnet M, Yabe K. Preparatory process for anticipatory postural adjustments: modulation of leg muscles reflex pathways during preparation for arm movements in standing man. Exp Brain Res. 1984;55:263–271. MEDLINE [11]. [11]Gaillard AWK. The late CNV wave: preparation versus expectancy. Psycophysiology. 1977;14(6):563–568. [12]. [12]Hillyard SA, Galambos R. Effects of stimulus and response contingencies on a surface negative slow potential shift in man. Electroencephalogr Clin Neurophysiol. 1967;22:297–304. Abstract |
Full-Text PDF (651 KB)
|
CrossRef
[13]. [13]Hillyard SA. Relationships between the contingent negative variation (CNV) and reaction time. Physiol Behav. 1969;4:351–357.
CrossRef
[14]. [14]Kamijo K, Nishihira Y, Hatta A, Kaneda T, Kida T, Higashiura T, et al. Changes in arousal level by differential exercise intensity. Clin Neurophysiol. 2004;115:2693–2698. Abstract | Full Text |
Full-Text PDF (233 KB)
|
CrossRef
[15]. [15]Loveless NE, Sanford AJ. Slow potential correlates of preparatory set. Biol Psychol. 1974;1:303–314. MEDLINE |
CrossRef
[16]. [16]McAdam DW, Knott JR, Rebert CS. Cortical slow potential changes in man related to interstimulus interval. Psychophysiology. 1969;5:349–358. MEDLINE |
CrossRef
[17]. [17]Rebert CS, McAdam DW, Knott JR, Irwin DA. Slow potential change in human brain related to level of motivation. J Comp Physiol Psychol. 1967;63:20–23. MEDLINE |
CrossRef
[18]. [18]Rohrbaugh JW, Syndulko K, Lindsley DB. Brain wave components of the contingent negative variation in humans. Science. 1976;191:1055–1057. MEDLINE [19]. [19]Bevan W, Hardesty DL, Avant LL. Response latency with constant and variable interval schedules. Percept Motor Skills. 1965;20:969–972. MEDLINE [20]. [20]Sato M, Saito S, Tsukahara S. Analysis of human operator's reaction lag time by using a real time correlator. Fukushima J Med Sci. 1973;19:73–85. MEDLINE [21]. [21]Gemba H, Sasaki K, Tsujimoto T. Cortical field potentials associated with hand movements triggered by warning and imperative stimuli in the monkey. Neurosci Lett. 1990;113(3):275–280. MEDLINE |
CrossRef
[22]. [22]Ikeda A, Shibasaki H, Kaji R, Terada K, Nagamine T, Honda M, et al. Dissociation between contingent negative variation (CNV) and Bereitschafts potential (BP) in patients with Parkinsonism. Electroencephalogr Clin Neurophysiol. 1997;102:142–151. Abstract |
Full-Text PDF (1074 KB)
|
CrossRef
[23]. [23]Lamarche M, Louvel J, Buser P, Rektor I. Intracerebral recordings of slow potentials in a contingent negative variation paradigm: an exploration in epileptic patients. Electroencephalogr Clin Neurophysiol. 1995;95(4):268–276. Abstract |
Full-Text PDF (809 KB)
|
CrossRef
[24]. [24]Goshima K. Studies on items and normal ranges in stabilometry. Equilibr Res. 1986;45:368–387. [25]. [25]Diener HC, Dichgans J. Long loop reflexes and posture. In: Bles W, Brandt TH editor. Disorders of posture and gait. Amsterdam: Elsevier; 1986;p. 41–51. [26]. [26]Welford AT. Reaction times. New York: Academic press; 1980;. [27]. [27]Klemmer ET. Time uncertainty in simple reaction time. J Exp Psychol. 1956;51(3):179–184. MEDLINE |
CrossRef
[28]. [28]Mowrer OH. Preparatory set (expectancy): some methods of measurement. Psychol Monogr. 1940;52:233. [29]. [29]Horak FB, Nashner LM. Central programming of postural movements: adaptation to altered support-surface configurations. J Neurophysiol. 1986;55(6):1369–1381. MEDLINE [30]. [30]Gahery Y. Associated movements, postural adjustments and synergies: some comments about the history and significance of three motor concepts. Arch Ital Biol. 1987;125(4):345–360. MEDLINE Department of Human Movement and Health, Graduate School of Medical Science, Kanazawa University, Japan  Corresponding author at: Department of Human Movement and Health, Graduate School of Medical Science, Kanazawa University, 13-1 Takara-machi, Kanazawa 920-8640, Japan. Tel.: +81 76 265 2225; fax: +81 76 234 4219.
PII: S0966-6362(06)00013-0 doi:10.1016/j.gaitpost.2006.01.004 © 2006 Elsevier B.V. All rights reserved. | |
|
FULL TEXT
