Cutaneous reflexes evoked during human walking are reduced when self-induced
1 Radboud University Nijmegen Medical Center, Department of Rehabilitation Medicine, Nijmegen, The Netherlands
2 Sint Maartenskliniek, SMK-Research, Nijmegen, The Netherlands
3 Radboud University Nijmegen, Nijmegen Institute for Cognition and Information, Nijmegen, The Netherlands
4 University Hospital Balgrist, Spinal Cord Injury Center, Zurich, Switzerland
Abstract
, 百拇医药
Reflex responses are often less pronounced when they are self-induced, but this question has barely been investigated quantitatively. The issue is particularly relevant for locomotion since it has been shown that reflexes elicited during normal gait are important for the regulation of locomotion. The cortex is thought to be involved in the control of reflexes during gait, but it is unclear whether it plays a role in the modulation of these reflexes during the step cycle. During gait, weak electrical stimulation of the sural nerve elicits reflexes in various leg muscles. Are these reflexes different when subjects themselves trigger the stimuli instead of being randomly released by the computer Cutaneous reflexes were elicited by sural nerve stimulation in 16 phases of the gait cycle in healthy subjects. The stimuli were triggered either by computer or by the subjects themselves. In 6 out of 7 subjects it was observed that the facilitatory responses in leg muscles were smaller and the suppressive responses were more suppressive following self-generated stimuli. In some muscles such as tibialis anterior (TA) both effects were seen (reduced facilitation at end stance and exaggerated suppression at end swing). In all subjects the modulation of anticipatory influences was muscle specific. In the main group of six subjects, the mean reduction in reflex responses was strongest in the TA (max. 30.7%; mean over 16 phases was 12.5%) and weakest in peroneus longus (PL, max. 10.1%; mean over 16 phases was 2.6%). The observation that facilitation is reduced and suppression enhanced in several muscles is taken as evidence that anticipation of self-induced reflex responses reduces the excitatory drive to motoneurones, for example through presynaptic inhibition of facilitatory reflex pathways.
, 百拇医药
Both sensation and reflexes may differ depending on whether the stimuli are applied by oneself or someone else (Von Holst & Mittelstaedt, 1950; Blakemore et al. 2000). This has led to the notion of an ‘efference copy’, whereby a copy of central commands is forwarded to structures involved in processing incoming information. Do these principles also apply to gait During walking, there is modulation of sensation during the step cycle (Duysens et al. 1995). For example, when identical non-nociceptive electrical stimuli were applied to the sural nerve, detection of the stimuli was best near the end of swing. In contrast, there was a reduction in sensitivity of the foot just after touchdown. These findings indicate that some ‘expected’ input from the foot at touchdown is filtered out. Electrical stimulation of the sural nerve results not only in sensation, but also in reflex responses. Can subjects filter out these reflexes as well To investigate this, an experiment is needed in which subjects can anticipate these reflexes, e.g. by allowing them to voluntarily control stimuli themselves.
, 百拇医药
For locomotion, the cutaneous feedback from the foot is especially important (Duysens & Pearson, 1976; Zehr et al. 1997; Duysens et al. 1990; Bouyer & Rossignol, 2003). It is also thought that supraspinal loops, including the motor cortex, may be involved in the control of cutaneous reflexes evoked by electrical stimulation of the sural nerve (Delwaide et al. 1981; Nielsen et al. 1997; Duysens et al. 2004). Hence there is increasing evidence that cutaneous input to the cortex is not only important for fine motor control of upper limb muscles (Lemon, 1981) but also for the regulation of gait, especially when perturbations are presented (Christensen et al. 1999).
, 百拇医药
Most knowledge about cortical control comes from studies with transcranial magnetic stimulation (TMS). Application of TMS during gait was shown to facilitate some leg muscles directly (Schubert et al. 1997; Capaday et al. 1999; Bonnard et al. 2002) or the cutaneous reflexes to these muscles (Pijnappels et al. 1998; Christensen et al. 1999). At low intensity, however, it was shown that TMS suppressed corticospinal drive to motoneurones, probably through intracortical inhibitory circuits, i.e. removal of excitatory drive (Petersen et al. 2001). In addition, it was recently shown that such magnetically evoked responses during gait are under cognitive control (Camus et al. 2004). Hence, the hypothesis is advanced that humans can exert cognitive control over these reflexes during gait.
, http://www.100md.com
It remains difficult to predict what the effect is of adding corticospinal control over reflexes, e.g. through voluntary control of stimuli. Both facilitatory and suppressive effects might be expected. In the case of facilitatory reflex responses the cortical command would lead to enhanced or decreased reflex responses. Similarly, the suppressive responses may be exaggerated or diminished. These actions could be caused by several mechanisms. One likely candidate is presynaptic inhibition of sensory input at spinal level (Seki et al. 2003). An inhibitory effect on transcortical reflexes (Van Doornik et al. 2004), or selective premotoneuronal control of spinal interneurones are other possibilities.
, http://www.100md.com
A related question is whether these voluntary effects on reflexes are equally prominent throughout the step cycle. The amplitudes of the reflex responses are known to depend on the phase of stimulation (Yang & Stein, 1990; Duysens et al. 1990; Van Wezel et al. 1997; Zehr et al. 1997). In tibialis anterior (TA) for example, facilitatory responses are seen in early swing, while suppressive responses occur at end swing and early stance. If voluntary control involves primarily suppression over the whole step cycle then one would expect smaller TA facilitations at end stance and deeper TA suppressions at end swing. Alternatively it is possible that one of the features of the modulation (for TA the facilitation at end stance or the suppression at end swing) is more prominently under cortical control. In that case one would expect selective changes occurring in one of these features when the reflexes are induced voluntarily.
, 百拇医药
Methods
Similar methods as used in the present study have been described elsewhere in detail (Duysens et al. 1996; Van Wezel et al. 1997). Therefore, only the essentials and specific procedures are described below.
Experimental set-up
Seven healthy volunteers (4 males and 3 females; age range of 23–36 years; mean of 26 years) participated in the present study. None of them had a known history of neurological or motor disorder. The procedures conformed with the Declaration of Helsinki for experiments on humans and were approved by the local ethics committee. All subjects gave their written informed consent. They walked on a treadmill (Woodway, Weil am Rhein, Germany) with a constant velocity of 4 km h–1, wearing a safety harness fastened to an emergency brake at the ceiling.
, 百拇医药
The subjects had to wear special shoes with contact switches (designed in collaboration with Algra Fotometaal b.v., Wormerveer, The Netherlands) to detect foot contact. To elicit a reflex response, an electrical stimulation was given to the sural nerve at the lateral malleolus of the right leg, where the nerve is closest to the skin surface (approximately halfway between the lateral malleolus and the Achilles' tendon). Therefore, a home-made bipolar stimulation electrode was used. The exact position of the stimulation electrode was determined according to the optimal irradiation of the stimulus, corresponding to the innervation area of the sural nerve. To keep conditions stable throughout the experiment, the electrode was firmly attached to the skin with surgical tape. The stimulus consisted of a train of 5 rectangular pulses of 1 ms duration with a frequency of 200 Hz.
, http://www.100md.com
Bipolar electromyographic activity (EMG) with Ag–AgCl surface electrodes (Hellige, Freiburg, Germany) was recorded in the biceps femoris (BF), tibialis anterior (TA), gastrocnemius medialis (GM), gastrocnemius lateralis (GL), and the peroneus longus (PL) of the ipsilateral leg by using surface electrodes. The EMG signals were (pre-) amplified, high pass filtered (cut-off frequency 3 Hz), full wave rectified, and low pass filtered (cut-off frequency 300 Hz). The EMG signals were sampled along with the foot-switch signals of both feet and a digital code referring to the stimulus condition. Data were sampled at 1000 Hz and stored on a hard disk, where they were analysed using MATLAB (the MathWorks, Nantick, MI, USA).
, http://www.100md.com
Experimental protocol
Before each experiment, the perception threshold (PT) for the stimulation was determined by gradually increasing and decreasing the stimulus intensity, while standing quietly. During the experiment, all subjects were stimulated at 2 times the PT. This intensity gives a tactile, non-nociceptive sensation on the area of the foot innervated by the sural nerve. The PT had to be stable throughout the whole experiment. For this purpose, the subjects had to walk with the stimulus electrode for 15–30 min before the experiment started, until the PT was stable. The PT was determined again at the end of each experiment to be sure that it had not changed during the experiment. None of the subjects showed a change of more than 10% and therefore no subjects were excluded for further investigation.
, http://www.100md.com
The experiment consisted of three different conditions: a control condition without any stimulation, to measure the background activity during walking (the first condition in Fig. 1B); a condition with electrical stimulation triggered by the computer (second condition in Fig. 1B); and a condition with an electrical stimulation triggered by the subjects pushing a button which they held in their hands (last condition in Fig. 1B).
A, mean EMG (n= 10) of the biceps femoris (BF) during one step cycle in one subject. B, representation of the 3 conditions with response window settings. The top panel is the control condition, also shown between the dashed lines in A. The middle panel is the stimulation by the computer and the bottom panel represents the self-triggered stimulation after hearing an auditory cue. Time 0 is the time of stimulation. The time between the auditory cue and pushing the button in the self-triggered condition is variable per trial. For all conditions, one fixed response window was set around the time when responses were found (latency of 80 ms after stimulation and duration of 40 ms).
, 百拇医药
Every time the subjects had to stimulate themselves, a short auditory cue was given. This cue was above the level of the noise of the treadmill, but not loud enough to elicit a startle reflex. The subjects were instructed (to try) to push each time at the same interval after the auditory cue. They were told that it was not needed to push as fast as possible (as in a reaction test). They were also carefully instructed that they would not receive an electrical stimulation after the cue sound, but that they had to press the button first to receive the stimulus. A short training period preceded each experiment.
, 百拇医药
In all conditions, stimuli were delivered during 16 phases of the step cycle. Heel strike of the right leg was determined as phase 1. In the control and computer conditions, 10 trials were sampled in each phase (= 2 x 160 trials). To measure 10 trials per phase in the self-stimulation condition, the subjects had to push the button 10 times in each phase of the step cycle. This was difficult to achieve because of the variability in the response times and because of the very short duration of the different phases. Therefore, relatively more trials with auditory cues had to be given as compared to the computer-generated condition. A pilot study showed that 320 trials were enough to be sure that the subjects pushed the button at least 10 times in each phase (20 trials per phase). All 640 trials were used randomly. The trials were separated by a random interval in the range of 3.5–6.5 s. Hence, two stimuli were always separated by at least two step cycles without a stimulus. The whole experimental run lasted about 45 min.
, http://www.100md.com
Data analysis
To study the influence of self-stimulation on the phase-dependent modulation of the reflexes, the differences between stimulation by the computer and self-stimulation were analysed. To obtain pure reflex responses in these conditions, the background EMG activity of the control condition (mean of 10 trials) was subtracted point-by-point (bit-size 1 ms) from the reflex responses of both stimulation conditions for each separate phase of the step cycle.
, 百拇医药
Quantification of the reflex responses occurred by calculating the mean of the EMG data (with S.D.) over the period in which the responses occurred. Therefore, a single time window (see Fig. 1) was set around the reflexes with latencies of about 80 ms after the stimulation (Yang & Stein, 1990; Tax et al. 1995; Duysens et al. 1996; Van Wezel et al. 1997) for all 16 phases in each muscle. For all conditions, the same time window was used. In analogy with these responses in the cat (Duysens & Loeb, 1980) these medium latency responses in humans are also called P2 responses (Baken et al. 2005). When a muscle showed little or no response, no adequate window could be set. In that case, an average window was used, calculated from the time windows used to measure responses in (in order of priority) other nearby muscles or the same muscles in other subjects.
, 百拇医药
To make an intersubject comparison possible, the resulting data underwent both an amplitude normalization (to the maximum control value of each phase) and a time normalization (a subdivision of the step cycle into 16 phases).
For the statistical analysis of the differences between computer stimulation and self-stimulation, a Wilcoxon signed-rank test was used with a significance level of P < 0.05. The results of self-stimulation after hearing the cue sound were also compared with results where the subjects pushed the button randomly throughout the step cycle without any cue. The results of the computer-induced responses were also compared for conditions with and without preceding auditory cue.
, http://www.100md.com
Results
The electrical stimulation of the sural nerve resulted in reflex responses in all five leg muscles recorded. These reflexes had a latency of 80 ms after stimulation and lasted for 40 ms. However, the amplitude of these responses varied between the different phases of the step cycle (phase-dependent modulation of reflex responses).
An example of the mean EMG activity of a typical subject in BF after computer stimulation (grey) and self-stimulation (black) in the 16 phases of the step cycle. Dashed vertical bars represent the time window around the responses. In each phase, the stimulation was given at 0 ms.
, 百拇医药
The mean normalized reflex and background activity during the 16 phases of the step cycle in BF is plotted in Fig. 3 for the same subject as in Fig. 2. Results of experiments with and without an auditory cue preceding the subject pushing the button to apply the stimuli are displayed. It is obvious that in both cases the self-induced reflexes were smaller than the reflexes induced by the computer. Furthermore, it can be seen that the self-evoked reflex responses with and without an auditory cue showed a similar modulation. Therefore, and also because stimulation with the auditory cue results in a better distribution of the stimuli over the 16 phases (without a cue, there are often phases without stimulation), the remaining experiments were only performed with an auditory cue. Note that the use of this cue did not introduce a systematic delay between cue and response. In fact, the interval varied between 390 and 1180 ms and therefore there cannot be a systematic effect of the cue on the response. A similar control was performed for the computer-induced responses in 11 subjects, using conditioning intervals ranging from –50 to 180 ms, tested in at least 4 phases per subject. There were no significant differences in the responses between the conditions with and without a pre-cue.
, 百拇医药
Modulation of the reflex responses after computer- and self-stimulation and the level of background activity in BF during 16 phases of the step cycle in the same subject as in Fig. 2. A, the subject pushed the button to give a stimulation, after hearing an auditory cue. B, the subject pushed the button randomly throughout the step cycle, without an auditory cue. The horizontal line represents the stance phase. Significant differences (P < 0.05) between computer- and self-stimulation.
, 百拇医药
The mean normalized reflex activity and the background activity during the 16 phases of the step cycle in all five muscles are plotted in Fig. 4 (with an auditory cue for the subject condition). Data are from the same subject as in Figs 2 and 3. To obtain the pure reflex responses, the control data were subtracted from the reflex data, as explained in Methods. As can be observed in Fig. 4, all muscles showed a reduction in reflex amplitude after self-stimulation. However, this reduction was not the same for the various muscles. In the BF, there was a significant reduction in amplitude in the self-triggered condition as compared to the computer-triggered condition in phases 1, 3, 4, 6, 12, 13, 14 and 16 (Wilcoxon signed-rank, P < 0.05). The difference in reflex amplitude between self- and computer-stimulation was much smaller in TA than in BF (maximum decrease of 34%). GM showed larger significant differences than BF (decrease up to 76%). In addition, there was also a difference in phases of the step cycle in which the effects occurred. For BF this was at the beginning of stance and the end of swing, while in GM this was only in some phases during stance (phases 4, 6 and 7).
, http://www.100md.com
Modulation of the reflex responses after computer- and self-stimulation and the level of background activity in BF, TA, GM, GL and PL during 16 phases of the step cycle. Data are of the same subject as in Figs 2 and 3. The horizontal line represents the stance phase. Significant differences (P < 0.05) between computer- and self-stimulation.
The modulation seen in the subject of Figs 2, 3 and 4 was typical of the modulation in the whole population. All subjects, except one (subject 7), showed a reduction in reflex amplitude when they stimulated themselves as compared to computer stimulation. However, the level of reduction in reflex amplitude could differ slightly among the six subjects in the same muscle or in the same phases of the step cycle. Although there were some differences between the subjects, there was no reason not to group all subjects together, except for subject 7. The mean reflex modulation and background activity in all muscles for this group of six subjects can be seen in Fig. 5, while the data for subject 7 are shown separately.
, http://www.100md.com
Modulation of the reflex responses after computer- and self-stimulation and the level of background activity in all muscles during 16 phases of the step cycle. On the left-hand side, data are of the whole group, except for the subject (subject 7) who showed a different kind of modulation. Data of that subject are separately given on the right-hand site. Horizontal lines represent the mean stance phases of the group or subject 7. Significant differences (P < 0.05) between computer- and self-stimulation.
, http://www.100md.com
Subject 7
Subject 7 showed a different modulation in BF, TA, GL and PL as compared to the remaining population. In these muscles, the amplitude of the reflex responses after self-stimulation was irregular and sometimes larger than after computer stimulation. For BF, for example, this was significant in phases 2, 5, 7, 10 and 11. In contrast, only in phase 12 did the computer stimulation result in larger reflex responses as compared to self-stimulation. In GL, self-stimulation resulted in significantly larger responses in phases 1–7 and 13–16, whereas computer-stimulated responses were never larger. In GM, there was only a significant reduction in amplitude caused by the self-applied stimulation in phase 14.
, 百拇医药
To understand why this single person, who was familiar with this kind of stimulation, showed partly opposite results to the others, additional analyses were undertaken. First, repeating the experiment in the same subject resulted in almost exactly the same modulation for computer- as well as for self-stimulation. The increased responses for self-generated stimuli as compared to the computer-generated ones were observed. Second, a number of possible differences were investigated. Analyses of stimulus intensity used (in absolute values), mean background activity in the phases that the muscles were not active, step frequency, delay between the auditory cue and pushing the button, height of the subject, duration and latency of the reflexes all failed to yield significant deviations of this subject from the other subjects.
, 百拇医药
Subtraction
The difference in amplitude between subject stimulation and computer stimulation in the group can be seen in Fig. 6, in which the mean normalized amplitude of the reflex responses after computer stimulation was subtracted from the mean normalized amplitude after self-stimulation.
The mean activity after computer stimulation was subtracted from the activity after self-stimulation. The error bars represent the standard errors of the mean and the horizontal line represents the mean stance phase. Significant difference between the two conditions (P < 0.05).
, 百拇医药
All five muscles showed a different phase-dependent modulation of the reflex responses after computer and self-applied stimulation in the five muscles (note the differences in scale). In BF, there was a significant level of reduction in reflex responses in phases 3, 4, 5 and 6. In TA, the group had a significant reduction in amplitude in the self-applied condition as compared to the computer stimulation in phases 6–13.
Although there were differences among the various phases and muscles, the amplitude was generally smaller after self-stimulation than after computer stimulation. This was also true for the phases in which the reflex activity after computer stimulation was smaller than the background activity, the so-called reflex reversal (a shift from facilitatory (‘positive’) to suppressive (‘negative; below background activity’) responses; see especially in TA and also, for example, in phases 15 and 16 in BF in Fig. 5). In other words, in these phases, the suppressive responses were exaggerated (more suppressive).
, http://www.100md.com
To summarize all of the data the mean changes across all phases were calculated. Based on the group data presented in Figs 5 and 6, the mean reduction in reflex responses was found to be largest in the TA, namely 12.5% of the maximum background activity (mean of all phases and the 6 subjects, normalized with respect to the maximum background activity) with a maximum decrease of 30.7% in one phase. The mean reduction in BF, GM and GL was 9.7%, 7.1% and 8.5%, respectively (maximum decrease 23.2%, 23.2% and 23.0%, respectively). The smallest reduction, nevertheless present in all subjects, was observed in the PL with a mean reduction of 2.6% (maximum decrease of 10.1%). These data are plotted in Fig. 7 (grey columns).
, http://www.100md.com
Mean differences in EMG activity between the computer- and self-stimulation conditions in the five muscles. The error bars are standard errors of the mean. The differences in the time window of the reflex response are shown as grey columns. The EMG differences in the period 40 ms pre-stimulation are shown as black columns. The differences are averages of 16 phases and 6 subjects, normalized to the maximum background activity.
Pre-activation
, 百拇医药
It is known that the amplitude of reflexes depends on the level of background activity. The question arises as to whether the reduction in amplitude of the reflex responses after self-stimulation could have been related to a reduction in background activity due to the anticipation of the voluntary induction of reflexes. To examine this question, an extra condition was introduced with catch trials in a separate experiment on one subject. Occasionally, the subject pushed the button after hearing the auditory cue while no electrical stimulation followed. Such catch trials were added only in four phases (10, 11, 12, 13), to reduce the effect of not knowing whether or not stimulation would follow. These trials were then analysed in the same way as the other trials. No significant deviations from the background activity were observed in the period following the ‘dummy’ stimulation. For a further check on the total population to see whether the reduction in amplitude of the reflex responses after self-stimulation could have been related to a reduction in background, the differences in pre-stimulus EMG activity between computer- and self-stimulation was measured. A fixed time window was set for all 16 phases in all the muscles starting 40 ms before the stimulation until the moment of stimulation. A window of 40 ms was chosen because the reflexes also had a duration of about 40 ms. No significant change in EMG activity was found for this period 40 ms prior to the stimulation in the mean of six subjects (see Fig. 7). The same type of analysis was made for a 40 ms period just preceding the main response (30–70 ms following stimulation). Again no significant changes in EMG activity between the two conditions were observed.
, 百拇医药
Discussion
As compared to computer-triggered stimuli, self-induced sural nerve stimuli during gait elicited smaller responses if they were facilitatory and stronger suppressive responses if they were suppressive. How does the cortex achieve the reflex suppressions when the stimulation is self-produced A first possibility is that voluntary stimulations yielded direct suppression of the motoneurones involved. In seated subjects, for example, Gerilovsky et al. (2002) observed a reduction in EMG activity prior to electrical stimulation when the pulses were self-applied in contrast to computer-triggered. Similarly in principle, in the present experiments the subjects could have suppressed background activity, leading to an ‘anticipatory locomotor adjustment’ along the lines described by McFadyen & Winter (1991). According to the ‘automatic gain control’ (Matthews, 1986) it is to be expected that the amplitude of facilitatory reflexes is smaller when background activity is reduced. To examine this possibility, background activity during 40 ms prior to, or 40 ms after stimulation was examined. However, no change in activity was found in these periods related to anticipation of the stimulation. Furthermore, catch trials in which subjects did not receive stimulation after pushing the button failed to elicit significant changes in background activity. This is in contrast to the study of Gerilovsky et al. (2002), possibly because these authors used resting conditions, which allow subjects to focus more on the upcoming stimulation. Furthermore, perception of electrical pulses may differ between rest and gait (Duysens et al. 1995). At any rate, the present control experiments show that the observed reductions in amplitude cannot be ascribed to reduction in background activity.
, 百拇医药
A second more likely possibility is a premotoneuronal source for the suppression. Seki et al. (2003) recently demonstrated cortically induced presynaptic inhibition of cutaneous afferents to interneurones in the primate spinal cord just prior to and during voluntary movements. Similarly, it is possible that voluntary induction of reflexes results in decreased reflexes because cortical activation adds presynaptic inhibition to the reflex pathway. Did the cortex provide such inhibition in the present study The present experiments do not provide conclusive evidence for this but the literature would support this contention (Schubert et al. 1997, 1999; Petersen et al. 1998, 2001; Pijnappels et al. 1998). For example, TMS could possibly either activate or suppress excitatory neurones at cortical level or alternatively, TMS might activate spinal inhibitory circuits.
, 百拇医药
The present study shows that there is a similar phase-dependent modulation of cutaneous reflex responses irrespective of the type of stimulation applied. This result is most consistent with the view that the reflex modulation itself is not of cortical origin. Instead the modulation could rely on spinal locomotor centres (central pattern generators; see Burke, 1999). Suppressive responses in TA motoneurones are likely to arise from presynaptic inhibition of excitatory pathways (Van de Crommert et al. 2003). When reflexes are self-induced, further increases in such presynaptic inhibition would yield a deepening of suppression. Hence, exaggerated suppression could be due to reduced facilitation.
, http://www.100md.com
Other muscles
A secondary aim of the present study was to elucidate the role of the peroneal muscles in sural nerve-induced reflexes during walking. Under static conditions, strong facilitatory responses with short latencies of 44 ms can be elicited in the peroneus longus (Aniss et al. 1992). However, the responses were only seen when the muscle was pre-activated (except for one subject, with high stimulation intensities). Some suppressions with P2 latency (80 ms) were seen as well (Aniss et al. 1992). During gait, sural nerve responses in PL have not been studied so far. In the present study, background activity showed a burst during mid-stance (see also Fig. 5). This pattern of background activity corresponds to that in previous studies (Louwerens et al. 1995; Courtine & Schieppati, 2003). Sural nerve stimulation induces P2 suppressive responses irrespective of the origin of triggering (self-induced versus computer-generated) during mid stance. The suppression in PL was one of the most stable features in the present study and in fact, it was present also in the single subject showing substantial differences in other muscles.
, 百拇医药
Intersubject variability
How do individuals differ with respect to the modulation of responses due to anticipation It has to be emphasized that subjects were not instructed either to suppress or to enhance their reflexes. However, the majority of subjects showed reduced responses when the responses were elicited by their own action. Nevertheless, there was one subject (subject 7), in which computer stimulation often resulted in smaller responses than self-stimulation. Repeating the experiment confirmed this. It follows that individuals can differ with respect to their reflex behaviour. Such interindividual differences also show up in the phase-dependent modulation. It was observed in cats that there are often large individual differences in cutaneous responses during gait (Loeb, 1993). Also in humans, differences between subjects were observed. For example, the sural nerve-induced reflex reversal is seen in some subjects but not in others (Duysens et al. 1990; Yang & Stein, 1990; De Serres et al. 1995; Zehr et al. 1997). Similar variations are observed in H-reflex modulation (Simonsen et al. 2002).
, http://www.100md.com
These differences may be related to learning since subjects are able to generally decrease or increase the gain in certain reflex pathways over extended periods of time. The observation that walking humans can voluntarily modify their magnetic-evoked potentials (MEPs; Bonnard et al. 2002) suggests that it should also be possible to voluntarily change the reduction in reflex amplitudes induced by self-stimulation.
Functional significance
, http://www.100md.com
Use of an ‘internal forward model’ is suggested to be the basis of the differences in sensation between self-produced and externally produced stimuli (Wolpert, 1997). In such a model, an efference copy of the motor command is used to predict the sensory consequences of the ongoing motor act. This is then compared with the actual sensory feedback (re-afference) from the movement. Because self-produced sensations can be correctly predicted on the basis of motor commands, the resulting sensory input is irrelevant and can be suppressed.
, 百拇医药
The reduction of self-induced reflexes, i.e. as seen in the present study, has to be considered in the same context. For example, there is a reduction in firing rate of vestibular neurones during active head movements as compared to passive head movements (McCrea et al. 1999). Von Holst & Mittelstaedt (1950) found that the reflexive eye, neck and limb movements produced by passive rotation of the head in space were usually absent during active head movements. In addition, reduction in amplitude of the middle latency blink reflex was observed when this reflex was self-elicited as compared to stimuli delivered by the experimenter (Meincke et al. 2003). In all these studies, the primary result is a reduction in reflex effects during voluntary self-stimulation. Like self-produced sensations, the self-produced stimuli are presumably seen as less relevant and can be partly blocked. The resulting reflex responses are therefore suppressed and a transition is made from a reactive strategy (dependence on reflexes) to a proactive control (feedforward strategy). More generally one observes a similar shift to suppression of reflex responses in situations when such reflexes are likely to be perceived as disturbing. For example, when a cat is forced to perform beam walking there is a suppression of H-reflexes as compared to the situation of normal walking overground (Llewellyn et al. 1990). Reflex responses are crucial in reactions to unexpected perturbations. In contrast, when such perturbations are predictable (as when the stimuli are self-induced), the responses lose most of their potential functional benefit and therefore can be suppressed.
, 百拇医药
References
Aniss AM, Gandevia SC & Burke D (1992). Reflex responses in active muscles elicited by stimulation of low-threshold afferents from the human foot. J Neurophysiol 67, 1375–1384.
Baken BCM, Dietz V & Duysens J (2005). Phase-dependent modulation of short latency cutaneous reflexes during walking in man. Brain Res 1031, 268–275.
Blakemore S-J, Wolpert DM & Frith CD (2000). Why can't you tickle yourselfNeuroreport 11, 11–16.
, http://www.100md.com
Bonnard M, Camus M, Coyle T & Pailhous J (2002). Task-induced modulation of motor evoked potentials in upper-leg muscles during human gait; a TMS study. Eur J Neurosci 16, 2225–2230.
Bouyer LJ & Rossignol S (2003). Contribution of cutaneous inputs from the hindpaw to the control of locomotion. II. Spinal cats. J Neurophysiol 90, 3640–3653.
Burke RE (1999). The use of state-dependent modulation of spinal reflexes as a tool to investigate the organization of spinal interneurons. Exp Brain Res 128, 263–277.
, 百拇医药
Camus M, Pailhous J & Bonnard M (2004). Cognitive tuning of corticospinal excitability during human gait: adaptation to the phase. Eur J Neurosci 20, 1101–1107.
Capaday C, Lavoie BA, Barbeau H, Schneider C & Bonnard M (1999). Studies on the corticospinal control of human walking. I. Responses to focal transcranial magnetic stimulation of the motor cortex. J Neurophysiol 81, 129–139.
Christensen LOD, Morita H, Petersen N & Nielsen J (1999). Evidence suggesting that a transcortical reflex pathway contributes to cutaneous reflexes in the tibialis anterior muscle during walking in man. Exp Brain Res 124, 59–68.
, http://www.100md.com
Courtine G & Schieppati M (2003). Human walking along a curved path. II. Gait features and EMG patterns. Eur J Neurosci 18, 191–205.
De Serres SJ, Yang JF & Patrick SK (1995). Mechanism for reflex reversal during walking in human tibialis anterior muscle revealed by single motor unit recording. J Physiol 488, 249–258.
Delwaide PJ, Toulouse P & Crenna P (1981). Hypothetical role of long-loop reflex pathways. Appl Neurophysiol 44, 171–176.
, 百拇医药
Duysens J, Baken BCM, Burgers L, Plat FM, den Otter AR & Kremer HPH (2004). Cutaneous reflexes from the foot during gait in hereditary spastic paraparesis. Clin Neurophysiol 115, 1057–1062.
Duysens J & Loeb GE (1980). Modulation of ipsi- and contralateral reflex responses in unrestrained walking cats. J Neurophysiol 44, 1024–1037.
Duysens J & Pearson KG (1976). The role of cutaneous afferents from the distal hindlimb in the regulation of the step cycle of thalamic cats. Exp Brain Res 24, 245–255.
, http://www.100md.com
Duysens J, Tax AAM, Murrer L & Dietz V (1996). Backward and forward walking use different patterns of phase-dependent modulation of cutaneous reflexes in humans. J Neurophysiol 76, 301–310.
Duysens J, Tax AAM, Nawijn S, Berger W, Prokop T & Altenmuller E (1995). Gating of sensation and evoked potentials following foot stimulation during human gait. Exp Brain Res 105, 423–431.
Duysens J, Trippel M, Horstmann GA & Dietz V (1990). Gating and reversal of reflexes in ankle muscles during human walking. Exp Brain Res 82, 351–358.
, http://www.100md.com
Gerilovsky L, Philipova D & Georgieva S (2002). Reduction of the muscle activity associated with self-imposed electrical stimulation of mixed nerves supplying lower limb muscles in man. J Electromyogr Kinesiol 12, 247–258.
Lemon RN (1981). Functional properties of monkey motor cortex neurons receiving afferent input from the hand and fingers. J Physiol 311, 497–519.
Llewellyn M, Yang JF & Prochazka A (1990). Human H-reflexes are smaller in difficult beam walking than in normal treadmill walking. Exp Brain Res 83, 22–28.
, http://www.100md.com
Loeb GE (1993). The distal hindlimb musculature of the cat: interanimal variability of locomotor activity and cutaneous reflexes. Exp Brain Res 96, 125–140.
Louwerens JW, Van Linge B, De Klerk LW, Mulder PG & Snijders CJ (1995). Peroneus longus and tibialis anterior muscle activity in the stance phase. A quantified electromyographic study of 10 controls and 25 patients with chronic ankle instability. Acta Orthop Scand 66, 517–523.
, http://www.100md.com
McCrea RA, Gdowski GT, Boyle R & Belton T (1999). Firing behavior of vestibular neurons during active and passive head movements: vestibulo-spinal and other non-eye-movement related neurons. J Neurophysiol 82, 416–428.
McFadyen BJ & Winter DA (1991). Anticipatory locomotor adjustments during obstructed human walking. Neurosci Res Comm 9, 37–44.
Matthews PB (1986). Observations on the autonomic compensation of reflex gain on varying the pre-existing level of motor discharge in man. J Physiol 374, 73–90.
, 百拇医药
Meincke U, Tpper R & Gouzoulis-Mayfrank E (2003). The electrically elicited R2 blink reflex consists of distinct subcomponents. Electromyogr Clin Neurophysiol 43, 91–95.
Nielsen J, Petersen N & Fedirchuk B (1997). Evidence suggesting a transcortical pathway from cutaneous foot afferents to tibialis anterior motoneurones in man. J Physiol 501, 473–484.
Petersen NT, Butler JE, Marchand-Pauvert V, Fisher R, Ledebt A, Pyndt HS et al. (2001). Suppression of EMG activity by transcranial magnetic stimulation in human subjects during walking. J Physiol 537, 651–656.
, 百拇医药
Petersen N, Christensen LOD & Nielsen J (1998). The effect of transcranial magnetic stimulation on the soleus H reflex during human walking. J Physiol 513, 599–610.
Pijnappels M, Van Wezel BMH, Colombo G, Dietz V & Duysens J (1998). Cortical facilitation of cutaneous reflexes in leg muscles during human gait. Brain Res 787, 149–153.
Schubert M, Curt A, Colombo G, Berger W & Dietz V (1999). Voluntary control of human gait: conditioning of magnetically evoked motor responses in a precision stepping task. Exp Brain Res 126, 583–588.
, 百拇医药
Schubert M, Curt A, Jensen L & Dietz V (1997). Corticospinal input in human gait: modulation of magnetically evoked motor responses. Exp Brain Res 115, 234–246.
Seki K, Perlmutter SI & Fetz EE (2003). Sensory input to the primate spinal cord is presynaptically inhibited during voluntary movement. Nat Neurosci 6, 1309–1316.
Simonsen EB, Dyhre-Poulsen P, Alkjaer T, Aagaard P & Magnusson SP (2002). Interindividual differences in H-reflex modulation during normal walking. Exp Brain Res 142, 108–115.
, 百拇医药
Tax AAM, Van Wezel BMH & Dietz V (1995). Bipedal reflex coordination to tactile stimulation of the sural nerve during human running. J Neurophysiol 73, 1947–1964.
Van de Crommert HW, Steijvers PJ, Mulder T & Duysens J (2003). Suppressive musculocutaneous reflexes in tibialis anterior following upper leg stimulation at the end of the swing phase. Exp Brain Res 149, 405–412.
Van Doornik J, Masakado Y, Sinkjaer T & Nielsen JB (2004). The suppression of the long-latency stretch reflex in the human tibialis anterior muscle by transcranial magnetic stimulation. Exp Brain Res 157, 403–406.
, 百拇医药
Van Wezel BMH, Ottenhoff FAM & Duysens J (1997). Dynamic control of location-specific information in tactile cutaneous reflexes from the foot during human walking. J Neurosci 17, 3804–3814.
Von Holst E & Mittelstaedt H (1950). Das Reafferenzprinzip [in German]. Naturwiss 37, 464–476.
Wolpert DM (1997). Computational approaches to motor control. Trends Cogn Sci 1, 209–216.
Yang JF & Stein RB (1990). Phase-dependent reflex reversal in human leg muscles during walking. J Neurophysiol 63, 1109–1117.
Zehr EP, Komiyama T & Stein RB (1997). Cutaneous reflexes during human gait: electromyographic and kinematic responses to electrical stimulation. J Neurophysiol 77, 3311–3325., http://www.100md.com(B. C. M. Baken, P. H. J. A. Nieuwenhuijz)
2 Sint Maartenskliniek, SMK-Research, Nijmegen, The Netherlands
3 Radboud University Nijmegen, Nijmegen Institute for Cognition and Information, Nijmegen, The Netherlands
4 University Hospital Balgrist, Spinal Cord Injury Center, Zurich, Switzerland
Abstract
, 百拇医药
Reflex responses are often less pronounced when they are self-induced, but this question has barely been investigated quantitatively. The issue is particularly relevant for locomotion since it has been shown that reflexes elicited during normal gait are important for the regulation of locomotion. The cortex is thought to be involved in the control of reflexes during gait, but it is unclear whether it plays a role in the modulation of these reflexes during the step cycle. During gait, weak electrical stimulation of the sural nerve elicits reflexes in various leg muscles. Are these reflexes different when subjects themselves trigger the stimuli instead of being randomly released by the computer Cutaneous reflexes were elicited by sural nerve stimulation in 16 phases of the gait cycle in healthy subjects. The stimuli were triggered either by computer or by the subjects themselves. In 6 out of 7 subjects it was observed that the facilitatory responses in leg muscles were smaller and the suppressive responses were more suppressive following self-generated stimuli. In some muscles such as tibialis anterior (TA) both effects were seen (reduced facilitation at end stance and exaggerated suppression at end swing). In all subjects the modulation of anticipatory influences was muscle specific. In the main group of six subjects, the mean reduction in reflex responses was strongest in the TA (max. 30.7%; mean over 16 phases was 12.5%) and weakest in peroneus longus (PL, max. 10.1%; mean over 16 phases was 2.6%). The observation that facilitation is reduced and suppression enhanced in several muscles is taken as evidence that anticipation of self-induced reflex responses reduces the excitatory drive to motoneurones, for example through presynaptic inhibition of facilitatory reflex pathways.
, 百拇医药
Both sensation and reflexes may differ depending on whether the stimuli are applied by oneself or someone else (Von Holst & Mittelstaedt, 1950; Blakemore et al. 2000). This has led to the notion of an ‘efference copy’, whereby a copy of central commands is forwarded to structures involved in processing incoming information. Do these principles also apply to gait During walking, there is modulation of sensation during the step cycle (Duysens et al. 1995). For example, when identical non-nociceptive electrical stimuli were applied to the sural nerve, detection of the stimuli was best near the end of swing. In contrast, there was a reduction in sensitivity of the foot just after touchdown. These findings indicate that some ‘expected’ input from the foot at touchdown is filtered out. Electrical stimulation of the sural nerve results not only in sensation, but also in reflex responses. Can subjects filter out these reflexes as well To investigate this, an experiment is needed in which subjects can anticipate these reflexes, e.g. by allowing them to voluntarily control stimuli themselves.
, 百拇医药
For locomotion, the cutaneous feedback from the foot is especially important (Duysens & Pearson, 1976; Zehr et al. 1997; Duysens et al. 1990; Bouyer & Rossignol, 2003). It is also thought that supraspinal loops, including the motor cortex, may be involved in the control of cutaneous reflexes evoked by electrical stimulation of the sural nerve (Delwaide et al. 1981; Nielsen et al. 1997; Duysens et al. 2004). Hence there is increasing evidence that cutaneous input to the cortex is not only important for fine motor control of upper limb muscles (Lemon, 1981) but also for the regulation of gait, especially when perturbations are presented (Christensen et al. 1999).
, 百拇医药
Most knowledge about cortical control comes from studies with transcranial magnetic stimulation (TMS). Application of TMS during gait was shown to facilitate some leg muscles directly (Schubert et al. 1997; Capaday et al. 1999; Bonnard et al. 2002) or the cutaneous reflexes to these muscles (Pijnappels et al. 1998; Christensen et al. 1999). At low intensity, however, it was shown that TMS suppressed corticospinal drive to motoneurones, probably through intracortical inhibitory circuits, i.e. removal of excitatory drive (Petersen et al. 2001). In addition, it was recently shown that such magnetically evoked responses during gait are under cognitive control (Camus et al. 2004). Hence, the hypothesis is advanced that humans can exert cognitive control over these reflexes during gait.
, http://www.100md.com
It remains difficult to predict what the effect is of adding corticospinal control over reflexes, e.g. through voluntary control of stimuli. Both facilitatory and suppressive effects might be expected. In the case of facilitatory reflex responses the cortical command would lead to enhanced or decreased reflex responses. Similarly, the suppressive responses may be exaggerated or diminished. These actions could be caused by several mechanisms. One likely candidate is presynaptic inhibition of sensory input at spinal level (Seki et al. 2003). An inhibitory effect on transcortical reflexes (Van Doornik et al. 2004), or selective premotoneuronal control of spinal interneurones are other possibilities.
, http://www.100md.com
A related question is whether these voluntary effects on reflexes are equally prominent throughout the step cycle. The amplitudes of the reflex responses are known to depend on the phase of stimulation (Yang & Stein, 1990; Duysens et al. 1990; Van Wezel et al. 1997; Zehr et al. 1997). In tibialis anterior (TA) for example, facilitatory responses are seen in early swing, while suppressive responses occur at end swing and early stance. If voluntary control involves primarily suppression over the whole step cycle then one would expect smaller TA facilitations at end stance and deeper TA suppressions at end swing. Alternatively it is possible that one of the features of the modulation (for TA the facilitation at end stance or the suppression at end swing) is more prominently under cortical control. In that case one would expect selective changes occurring in one of these features when the reflexes are induced voluntarily.
, 百拇医药
Methods
Similar methods as used in the present study have been described elsewhere in detail (Duysens et al. 1996; Van Wezel et al. 1997). Therefore, only the essentials and specific procedures are described below.
Experimental set-up
Seven healthy volunteers (4 males and 3 females; age range of 23–36 years; mean of 26 years) participated in the present study. None of them had a known history of neurological or motor disorder. The procedures conformed with the Declaration of Helsinki for experiments on humans and were approved by the local ethics committee. All subjects gave their written informed consent. They walked on a treadmill (Woodway, Weil am Rhein, Germany) with a constant velocity of 4 km h–1, wearing a safety harness fastened to an emergency brake at the ceiling.
, 百拇医药
The subjects had to wear special shoes with contact switches (designed in collaboration with Algra Fotometaal b.v., Wormerveer, The Netherlands) to detect foot contact. To elicit a reflex response, an electrical stimulation was given to the sural nerve at the lateral malleolus of the right leg, where the nerve is closest to the skin surface (approximately halfway between the lateral malleolus and the Achilles' tendon). Therefore, a home-made bipolar stimulation electrode was used. The exact position of the stimulation electrode was determined according to the optimal irradiation of the stimulus, corresponding to the innervation area of the sural nerve. To keep conditions stable throughout the experiment, the electrode was firmly attached to the skin with surgical tape. The stimulus consisted of a train of 5 rectangular pulses of 1 ms duration with a frequency of 200 Hz.
, http://www.100md.com
Bipolar electromyographic activity (EMG) with Ag–AgCl surface electrodes (Hellige, Freiburg, Germany) was recorded in the biceps femoris (BF), tibialis anterior (TA), gastrocnemius medialis (GM), gastrocnemius lateralis (GL), and the peroneus longus (PL) of the ipsilateral leg by using surface electrodes. The EMG signals were (pre-) amplified, high pass filtered (cut-off frequency 3 Hz), full wave rectified, and low pass filtered (cut-off frequency 300 Hz). The EMG signals were sampled along with the foot-switch signals of both feet and a digital code referring to the stimulus condition. Data were sampled at 1000 Hz and stored on a hard disk, where they were analysed using MATLAB (the MathWorks, Nantick, MI, USA).
, http://www.100md.com
Experimental protocol
Before each experiment, the perception threshold (PT) for the stimulation was determined by gradually increasing and decreasing the stimulus intensity, while standing quietly. During the experiment, all subjects were stimulated at 2 times the PT. This intensity gives a tactile, non-nociceptive sensation on the area of the foot innervated by the sural nerve. The PT had to be stable throughout the whole experiment. For this purpose, the subjects had to walk with the stimulus electrode for 15–30 min before the experiment started, until the PT was stable. The PT was determined again at the end of each experiment to be sure that it had not changed during the experiment. None of the subjects showed a change of more than 10% and therefore no subjects were excluded for further investigation.
, http://www.100md.com
The experiment consisted of three different conditions: a control condition without any stimulation, to measure the background activity during walking (the first condition in Fig. 1B); a condition with electrical stimulation triggered by the computer (second condition in Fig. 1B); and a condition with an electrical stimulation triggered by the subjects pushing a button which they held in their hands (last condition in Fig. 1B).
A, mean EMG (n= 10) of the biceps femoris (BF) during one step cycle in one subject. B, representation of the 3 conditions with response window settings. The top panel is the control condition, also shown between the dashed lines in A. The middle panel is the stimulation by the computer and the bottom panel represents the self-triggered stimulation after hearing an auditory cue. Time 0 is the time of stimulation. The time between the auditory cue and pushing the button in the self-triggered condition is variable per trial. For all conditions, one fixed response window was set around the time when responses were found (latency of 80 ms after stimulation and duration of 40 ms).
, 百拇医药
Every time the subjects had to stimulate themselves, a short auditory cue was given. This cue was above the level of the noise of the treadmill, but not loud enough to elicit a startle reflex. The subjects were instructed (to try) to push each time at the same interval after the auditory cue. They were told that it was not needed to push as fast as possible (as in a reaction test). They were also carefully instructed that they would not receive an electrical stimulation after the cue sound, but that they had to press the button first to receive the stimulus. A short training period preceded each experiment.
, 百拇医药
In all conditions, stimuli were delivered during 16 phases of the step cycle. Heel strike of the right leg was determined as phase 1. In the control and computer conditions, 10 trials were sampled in each phase (= 2 x 160 trials). To measure 10 trials per phase in the self-stimulation condition, the subjects had to push the button 10 times in each phase of the step cycle. This was difficult to achieve because of the variability in the response times and because of the very short duration of the different phases. Therefore, relatively more trials with auditory cues had to be given as compared to the computer-generated condition. A pilot study showed that 320 trials were enough to be sure that the subjects pushed the button at least 10 times in each phase (20 trials per phase). All 640 trials were used randomly. The trials were separated by a random interval in the range of 3.5–6.5 s. Hence, two stimuli were always separated by at least two step cycles without a stimulus. The whole experimental run lasted about 45 min.
, http://www.100md.com
Data analysis
To study the influence of self-stimulation on the phase-dependent modulation of the reflexes, the differences between stimulation by the computer and self-stimulation were analysed. To obtain pure reflex responses in these conditions, the background EMG activity of the control condition (mean of 10 trials) was subtracted point-by-point (bit-size 1 ms) from the reflex responses of both stimulation conditions for each separate phase of the step cycle.
, 百拇医药
Quantification of the reflex responses occurred by calculating the mean of the EMG data (with S.D.) over the period in which the responses occurred. Therefore, a single time window (see Fig. 1) was set around the reflexes with latencies of about 80 ms after the stimulation (Yang & Stein, 1990; Tax et al. 1995; Duysens et al. 1996; Van Wezel et al. 1997) for all 16 phases in each muscle. For all conditions, the same time window was used. In analogy with these responses in the cat (Duysens & Loeb, 1980) these medium latency responses in humans are also called P2 responses (Baken et al. 2005). When a muscle showed little or no response, no adequate window could be set. In that case, an average window was used, calculated from the time windows used to measure responses in (in order of priority) other nearby muscles or the same muscles in other subjects.
, 百拇医药
To make an intersubject comparison possible, the resulting data underwent both an amplitude normalization (to the maximum control value of each phase) and a time normalization (a subdivision of the step cycle into 16 phases).
For the statistical analysis of the differences between computer stimulation and self-stimulation, a Wilcoxon signed-rank test was used with a significance level of P < 0.05. The results of self-stimulation after hearing the cue sound were also compared with results where the subjects pushed the button randomly throughout the step cycle without any cue. The results of the computer-induced responses were also compared for conditions with and without preceding auditory cue.
, http://www.100md.com
Results
The electrical stimulation of the sural nerve resulted in reflex responses in all five leg muscles recorded. These reflexes had a latency of 80 ms after stimulation and lasted for 40 ms. However, the amplitude of these responses varied between the different phases of the step cycle (phase-dependent modulation of reflex responses).
An example of the mean EMG activity of a typical subject in BF after computer stimulation (grey) and self-stimulation (black) in the 16 phases of the step cycle. Dashed vertical bars represent the time window around the responses. In each phase, the stimulation was given at 0 ms.
, 百拇医药
The mean normalized reflex and background activity during the 16 phases of the step cycle in BF is plotted in Fig. 3 for the same subject as in Fig. 2. Results of experiments with and without an auditory cue preceding the subject pushing the button to apply the stimuli are displayed. It is obvious that in both cases the self-induced reflexes were smaller than the reflexes induced by the computer. Furthermore, it can be seen that the self-evoked reflex responses with and without an auditory cue showed a similar modulation. Therefore, and also because stimulation with the auditory cue results in a better distribution of the stimuli over the 16 phases (without a cue, there are often phases without stimulation), the remaining experiments were only performed with an auditory cue. Note that the use of this cue did not introduce a systematic delay between cue and response. In fact, the interval varied between 390 and 1180 ms and therefore there cannot be a systematic effect of the cue on the response. A similar control was performed for the computer-induced responses in 11 subjects, using conditioning intervals ranging from –50 to 180 ms, tested in at least 4 phases per subject. There were no significant differences in the responses between the conditions with and without a pre-cue.
, 百拇医药
Modulation of the reflex responses after computer- and self-stimulation and the level of background activity in BF during 16 phases of the step cycle in the same subject as in Fig. 2. A, the subject pushed the button to give a stimulation, after hearing an auditory cue. B, the subject pushed the button randomly throughout the step cycle, without an auditory cue. The horizontal line represents the stance phase. Significant differences (P < 0.05) between computer- and self-stimulation.
, 百拇医药
The mean normalized reflex activity and the background activity during the 16 phases of the step cycle in all five muscles are plotted in Fig. 4 (with an auditory cue for the subject condition). Data are from the same subject as in Figs 2 and 3. To obtain the pure reflex responses, the control data were subtracted from the reflex data, as explained in Methods. As can be observed in Fig. 4, all muscles showed a reduction in reflex amplitude after self-stimulation. However, this reduction was not the same for the various muscles. In the BF, there was a significant reduction in amplitude in the self-triggered condition as compared to the computer-triggered condition in phases 1, 3, 4, 6, 12, 13, 14 and 16 (Wilcoxon signed-rank, P < 0.05). The difference in reflex amplitude between self- and computer-stimulation was much smaller in TA than in BF (maximum decrease of 34%). GM showed larger significant differences than BF (decrease up to 76%). In addition, there was also a difference in phases of the step cycle in which the effects occurred. For BF this was at the beginning of stance and the end of swing, while in GM this was only in some phases during stance (phases 4, 6 and 7).
, http://www.100md.com
Modulation of the reflex responses after computer- and self-stimulation and the level of background activity in BF, TA, GM, GL and PL during 16 phases of the step cycle. Data are of the same subject as in Figs 2 and 3. The horizontal line represents the stance phase. Significant differences (P < 0.05) between computer- and self-stimulation.
The modulation seen in the subject of Figs 2, 3 and 4 was typical of the modulation in the whole population. All subjects, except one (subject 7), showed a reduction in reflex amplitude when they stimulated themselves as compared to computer stimulation. However, the level of reduction in reflex amplitude could differ slightly among the six subjects in the same muscle or in the same phases of the step cycle. Although there were some differences between the subjects, there was no reason not to group all subjects together, except for subject 7. The mean reflex modulation and background activity in all muscles for this group of six subjects can be seen in Fig. 5, while the data for subject 7 are shown separately.
, http://www.100md.com
Modulation of the reflex responses after computer- and self-stimulation and the level of background activity in all muscles during 16 phases of the step cycle. On the left-hand side, data are of the whole group, except for the subject (subject 7) who showed a different kind of modulation. Data of that subject are separately given on the right-hand site. Horizontal lines represent the mean stance phases of the group or subject 7. Significant differences (P < 0.05) between computer- and self-stimulation.
, http://www.100md.com
Subject 7
Subject 7 showed a different modulation in BF, TA, GL and PL as compared to the remaining population. In these muscles, the amplitude of the reflex responses after self-stimulation was irregular and sometimes larger than after computer stimulation. For BF, for example, this was significant in phases 2, 5, 7, 10 and 11. In contrast, only in phase 12 did the computer stimulation result in larger reflex responses as compared to self-stimulation. In GL, self-stimulation resulted in significantly larger responses in phases 1–7 and 13–16, whereas computer-stimulated responses were never larger. In GM, there was only a significant reduction in amplitude caused by the self-applied stimulation in phase 14.
, 百拇医药
To understand why this single person, who was familiar with this kind of stimulation, showed partly opposite results to the others, additional analyses were undertaken. First, repeating the experiment in the same subject resulted in almost exactly the same modulation for computer- as well as for self-stimulation. The increased responses for self-generated stimuli as compared to the computer-generated ones were observed. Second, a number of possible differences were investigated. Analyses of stimulus intensity used (in absolute values), mean background activity in the phases that the muscles were not active, step frequency, delay between the auditory cue and pushing the button, height of the subject, duration and latency of the reflexes all failed to yield significant deviations of this subject from the other subjects.
, 百拇医药
Subtraction
The difference in amplitude between subject stimulation and computer stimulation in the group can be seen in Fig. 6, in which the mean normalized amplitude of the reflex responses after computer stimulation was subtracted from the mean normalized amplitude after self-stimulation.
The mean activity after computer stimulation was subtracted from the activity after self-stimulation. The error bars represent the standard errors of the mean and the horizontal line represents the mean stance phase. Significant difference between the two conditions (P < 0.05).
, 百拇医药
All five muscles showed a different phase-dependent modulation of the reflex responses after computer and self-applied stimulation in the five muscles (note the differences in scale). In BF, there was a significant level of reduction in reflex responses in phases 3, 4, 5 and 6. In TA, the group had a significant reduction in amplitude in the self-applied condition as compared to the computer stimulation in phases 6–13.
Although there were differences among the various phases and muscles, the amplitude was generally smaller after self-stimulation than after computer stimulation. This was also true for the phases in which the reflex activity after computer stimulation was smaller than the background activity, the so-called reflex reversal (a shift from facilitatory (‘positive’) to suppressive (‘negative; below background activity’) responses; see especially in TA and also, for example, in phases 15 and 16 in BF in Fig. 5). In other words, in these phases, the suppressive responses were exaggerated (more suppressive).
, http://www.100md.com
To summarize all of the data the mean changes across all phases were calculated. Based on the group data presented in Figs 5 and 6, the mean reduction in reflex responses was found to be largest in the TA, namely 12.5% of the maximum background activity (mean of all phases and the 6 subjects, normalized with respect to the maximum background activity) with a maximum decrease of 30.7% in one phase. The mean reduction in BF, GM and GL was 9.7%, 7.1% and 8.5%, respectively (maximum decrease 23.2%, 23.2% and 23.0%, respectively). The smallest reduction, nevertheless present in all subjects, was observed in the PL with a mean reduction of 2.6% (maximum decrease of 10.1%). These data are plotted in Fig. 7 (grey columns).
, http://www.100md.com
Mean differences in EMG activity between the computer- and self-stimulation conditions in the five muscles. The error bars are standard errors of the mean. The differences in the time window of the reflex response are shown as grey columns. The EMG differences in the period 40 ms pre-stimulation are shown as black columns. The differences are averages of 16 phases and 6 subjects, normalized to the maximum background activity.
Pre-activation
, 百拇医药
It is known that the amplitude of reflexes depends on the level of background activity. The question arises as to whether the reduction in amplitude of the reflex responses after self-stimulation could have been related to a reduction in background activity due to the anticipation of the voluntary induction of reflexes. To examine this question, an extra condition was introduced with catch trials in a separate experiment on one subject. Occasionally, the subject pushed the button after hearing the auditory cue while no electrical stimulation followed. Such catch trials were added only in four phases (10, 11, 12, 13), to reduce the effect of not knowing whether or not stimulation would follow. These trials were then analysed in the same way as the other trials. No significant deviations from the background activity were observed in the period following the ‘dummy’ stimulation. For a further check on the total population to see whether the reduction in amplitude of the reflex responses after self-stimulation could have been related to a reduction in background, the differences in pre-stimulus EMG activity between computer- and self-stimulation was measured. A fixed time window was set for all 16 phases in all the muscles starting 40 ms before the stimulation until the moment of stimulation. A window of 40 ms was chosen because the reflexes also had a duration of about 40 ms. No significant change in EMG activity was found for this period 40 ms prior to the stimulation in the mean of six subjects (see Fig. 7). The same type of analysis was made for a 40 ms period just preceding the main response (30–70 ms following stimulation). Again no significant changes in EMG activity between the two conditions were observed.
, 百拇医药
Discussion
As compared to computer-triggered stimuli, self-induced sural nerve stimuli during gait elicited smaller responses if they were facilitatory and stronger suppressive responses if they were suppressive. How does the cortex achieve the reflex suppressions when the stimulation is self-produced A first possibility is that voluntary stimulations yielded direct suppression of the motoneurones involved. In seated subjects, for example, Gerilovsky et al. (2002) observed a reduction in EMG activity prior to electrical stimulation when the pulses were self-applied in contrast to computer-triggered. Similarly in principle, in the present experiments the subjects could have suppressed background activity, leading to an ‘anticipatory locomotor adjustment’ along the lines described by McFadyen & Winter (1991). According to the ‘automatic gain control’ (Matthews, 1986) it is to be expected that the amplitude of facilitatory reflexes is smaller when background activity is reduced. To examine this possibility, background activity during 40 ms prior to, or 40 ms after stimulation was examined. However, no change in activity was found in these periods related to anticipation of the stimulation. Furthermore, catch trials in which subjects did not receive stimulation after pushing the button failed to elicit significant changes in background activity. This is in contrast to the study of Gerilovsky et al. (2002), possibly because these authors used resting conditions, which allow subjects to focus more on the upcoming stimulation. Furthermore, perception of electrical pulses may differ between rest and gait (Duysens et al. 1995). At any rate, the present control experiments show that the observed reductions in amplitude cannot be ascribed to reduction in background activity.
, 百拇医药
A second more likely possibility is a premotoneuronal source for the suppression. Seki et al. (2003) recently demonstrated cortically induced presynaptic inhibition of cutaneous afferents to interneurones in the primate spinal cord just prior to and during voluntary movements. Similarly, it is possible that voluntary induction of reflexes results in decreased reflexes because cortical activation adds presynaptic inhibition to the reflex pathway. Did the cortex provide such inhibition in the present study The present experiments do not provide conclusive evidence for this but the literature would support this contention (Schubert et al. 1997, 1999; Petersen et al. 1998, 2001; Pijnappels et al. 1998). For example, TMS could possibly either activate or suppress excitatory neurones at cortical level or alternatively, TMS might activate spinal inhibitory circuits.
, 百拇医药
The present study shows that there is a similar phase-dependent modulation of cutaneous reflex responses irrespective of the type of stimulation applied. This result is most consistent with the view that the reflex modulation itself is not of cortical origin. Instead the modulation could rely on spinal locomotor centres (central pattern generators; see Burke, 1999). Suppressive responses in TA motoneurones are likely to arise from presynaptic inhibition of excitatory pathways (Van de Crommert et al. 2003). When reflexes are self-induced, further increases in such presynaptic inhibition would yield a deepening of suppression. Hence, exaggerated suppression could be due to reduced facilitation.
, http://www.100md.com
Other muscles
A secondary aim of the present study was to elucidate the role of the peroneal muscles in sural nerve-induced reflexes during walking. Under static conditions, strong facilitatory responses with short latencies of 44 ms can be elicited in the peroneus longus (Aniss et al. 1992). However, the responses were only seen when the muscle was pre-activated (except for one subject, with high stimulation intensities). Some suppressions with P2 latency (80 ms) were seen as well (Aniss et al. 1992). During gait, sural nerve responses in PL have not been studied so far. In the present study, background activity showed a burst during mid-stance (see also Fig. 5). This pattern of background activity corresponds to that in previous studies (Louwerens et al. 1995; Courtine & Schieppati, 2003). Sural nerve stimulation induces P2 suppressive responses irrespective of the origin of triggering (self-induced versus computer-generated) during mid stance. The suppression in PL was one of the most stable features in the present study and in fact, it was present also in the single subject showing substantial differences in other muscles.
, 百拇医药
Intersubject variability
How do individuals differ with respect to the modulation of responses due to anticipation It has to be emphasized that subjects were not instructed either to suppress or to enhance their reflexes. However, the majority of subjects showed reduced responses when the responses were elicited by their own action. Nevertheless, there was one subject (subject 7), in which computer stimulation often resulted in smaller responses than self-stimulation. Repeating the experiment confirmed this. It follows that individuals can differ with respect to their reflex behaviour. Such interindividual differences also show up in the phase-dependent modulation. It was observed in cats that there are often large individual differences in cutaneous responses during gait (Loeb, 1993). Also in humans, differences between subjects were observed. For example, the sural nerve-induced reflex reversal is seen in some subjects but not in others (Duysens et al. 1990; Yang & Stein, 1990; De Serres et al. 1995; Zehr et al. 1997). Similar variations are observed in H-reflex modulation (Simonsen et al. 2002).
, http://www.100md.com
These differences may be related to learning since subjects are able to generally decrease or increase the gain in certain reflex pathways over extended periods of time. The observation that walking humans can voluntarily modify their magnetic-evoked potentials (MEPs; Bonnard et al. 2002) suggests that it should also be possible to voluntarily change the reduction in reflex amplitudes induced by self-stimulation.
Functional significance
, http://www.100md.com
Use of an ‘internal forward model’ is suggested to be the basis of the differences in sensation between self-produced and externally produced stimuli (Wolpert, 1997). In such a model, an efference copy of the motor command is used to predict the sensory consequences of the ongoing motor act. This is then compared with the actual sensory feedback (re-afference) from the movement. Because self-produced sensations can be correctly predicted on the basis of motor commands, the resulting sensory input is irrelevant and can be suppressed.
, 百拇医药
The reduction of self-induced reflexes, i.e. as seen in the present study, has to be considered in the same context. For example, there is a reduction in firing rate of vestibular neurones during active head movements as compared to passive head movements (McCrea et al. 1999). Von Holst & Mittelstaedt (1950) found that the reflexive eye, neck and limb movements produced by passive rotation of the head in space were usually absent during active head movements. In addition, reduction in amplitude of the middle latency blink reflex was observed when this reflex was self-elicited as compared to stimuli delivered by the experimenter (Meincke et al. 2003). In all these studies, the primary result is a reduction in reflex effects during voluntary self-stimulation. Like self-produced sensations, the self-produced stimuli are presumably seen as less relevant and can be partly blocked. The resulting reflex responses are therefore suppressed and a transition is made from a reactive strategy (dependence on reflexes) to a proactive control (feedforward strategy). More generally one observes a similar shift to suppression of reflex responses in situations when such reflexes are likely to be perceived as disturbing. For example, when a cat is forced to perform beam walking there is a suppression of H-reflexes as compared to the situation of normal walking overground (Llewellyn et al. 1990). Reflex responses are crucial in reactions to unexpected perturbations. In contrast, when such perturbations are predictable (as when the stimuli are self-induced), the responses lose most of their potential functional benefit and therefore can be suppressed.
, 百拇医药
References
Aniss AM, Gandevia SC & Burke D (1992). Reflex responses in active muscles elicited by stimulation of low-threshold afferents from the human foot. J Neurophysiol 67, 1375–1384.
Baken BCM, Dietz V & Duysens J (2005). Phase-dependent modulation of short latency cutaneous reflexes during walking in man. Brain Res 1031, 268–275.
Blakemore S-J, Wolpert DM & Frith CD (2000). Why can't you tickle yourselfNeuroreport 11, 11–16.
, http://www.100md.com
Bonnard M, Camus M, Coyle T & Pailhous J (2002). Task-induced modulation of motor evoked potentials in upper-leg muscles during human gait; a TMS study. Eur J Neurosci 16, 2225–2230.
Bouyer LJ & Rossignol S (2003). Contribution of cutaneous inputs from the hindpaw to the control of locomotion. II. Spinal cats. J Neurophysiol 90, 3640–3653.
Burke RE (1999). The use of state-dependent modulation of spinal reflexes as a tool to investigate the organization of spinal interneurons. Exp Brain Res 128, 263–277.
, 百拇医药
Camus M, Pailhous J & Bonnard M (2004). Cognitive tuning of corticospinal excitability during human gait: adaptation to the phase. Eur J Neurosci 20, 1101–1107.
Capaday C, Lavoie BA, Barbeau H, Schneider C & Bonnard M (1999). Studies on the corticospinal control of human walking. I. Responses to focal transcranial magnetic stimulation of the motor cortex. J Neurophysiol 81, 129–139.
Christensen LOD, Morita H, Petersen N & Nielsen J (1999). Evidence suggesting that a transcortical reflex pathway contributes to cutaneous reflexes in the tibialis anterior muscle during walking in man. Exp Brain Res 124, 59–68.
, http://www.100md.com
Courtine G & Schieppati M (2003). Human walking along a curved path. II. Gait features and EMG patterns. Eur J Neurosci 18, 191–205.
De Serres SJ, Yang JF & Patrick SK (1995). Mechanism for reflex reversal during walking in human tibialis anterior muscle revealed by single motor unit recording. J Physiol 488, 249–258.
Delwaide PJ, Toulouse P & Crenna P (1981). Hypothetical role of long-loop reflex pathways. Appl Neurophysiol 44, 171–176.
, 百拇医药
Duysens J, Baken BCM, Burgers L, Plat FM, den Otter AR & Kremer HPH (2004). Cutaneous reflexes from the foot during gait in hereditary spastic paraparesis. Clin Neurophysiol 115, 1057–1062.
Duysens J & Loeb GE (1980). Modulation of ipsi- and contralateral reflex responses in unrestrained walking cats. J Neurophysiol 44, 1024–1037.
Duysens J & Pearson KG (1976). The role of cutaneous afferents from the distal hindlimb in the regulation of the step cycle of thalamic cats. Exp Brain Res 24, 245–255.
, http://www.100md.com
Duysens J, Tax AAM, Murrer L & Dietz V (1996). Backward and forward walking use different patterns of phase-dependent modulation of cutaneous reflexes in humans. J Neurophysiol 76, 301–310.
Duysens J, Tax AAM, Nawijn S, Berger W, Prokop T & Altenmuller E (1995). Gating of sensation and evoked potentials following foot stimulation during human gait. Exp Brain Res 105, 423–431.
Duysens J, Trippel M, Horstmann GA & Dietz V (1990). Gating and reversal of reflexes in ankle muscles during human walking. Exp Brain Res 82, 351–358.
, http://www.100md.com
Gerilovsky L, Philipova D & Georgieva S (2002). Reduction of the muscle activity associated with self-imposed electrical stimulation of mixed nerves supplying lower limb muscles in man. J Electromyogr Kinesiol 12, 247–258.
Lemon RN (1981). Functional properties of monkey motor cortex neurons receiving afferent input from the hand and fingers. J Physiol 311, 497–519.
Llewellyn M, Yang JF & Prochazka A (1990). Human H-reflexes are smaller in difficult beam walking than in normal treadmill walking. Exp Brain Res 83, 22–28.
, http://www.100md.com
Loeb GE (1993). The distal hindlimb musculature of the cat: interanimal variability of locomotor activity and cutaneous reflexes. Exp Brain Res 96, 125–140.
Louwerens JW, Van Linge B, De Klerk LW, Mulder PG & Snijders CJ (1995). Peroneus longus and tibialis anterior muscle activity in the stance phase. A quantified electromyographic study of 10 controls and 25 patients with chronic ankle instability. Acta Orthop Scand 66, 517–523.
, http://www.100md.com
McCrea RA, Gdowski GT, Boyle R & Belton T (1999). Firing behavior of vestibular neurons during active and passive head movements: vestibulo-spinal and other non-eye-movement related neurons. J Neurophysiol 82, 416–428.
McFadyen BJ & Winter DA (1991). Anticipatory locomotor adjustments during obstructed human walking. Neurosci Res Comm 9, 37–44.
Matthews PB (1986). Observations on the autonomic compensation of reflex gain on varying the pre-existing level of motor discharge in man. J Physiol 374, 73–90.
, 百拇医药
Meincke U, Tpper R & Gouzoulis-Mayfrank E (2003). The electrically elicited R2 blink reflex consists of distinct subcomponents. Electromyogr Clin Neurophysiol 43, 91–95.
Nielsen J, Petersen N & Fedirchuk B (1997). Evidence suggesting a transcortical pathway from cutaneous foot afferents to tibialis anterior motoneurones in man. J Physiol 501, 473–484.
Petersen NT, Butler JE, Marchand-Pauvert V, Fisher R, Ledebt A, Pyndt HS et al. (2001). Suppression of EMG activity by transcranial magnetic stimulation in human subjects during walking. J Physiol 537, 651–656.
, 百拇医药
Petersen N, Christensen LOD & Nielsen J (1998). The effect of transcranial magnetic stimulation on the soleus H reflex during human walking. J Physiol 513, 599–610.
Pijnappels M, Van Wezel BMH, Colombo G, Dietz V & Duysens J (1998). Cortical facilitation of cutaneous reflexes in leg muscles during human gait. Brain Res 787, 149–153.
Schubert M, Curt A, Colombo G, Berger W & Dietz V (1999). Voluntary control of human gait: conditioning of magnetically evoked motor responses in a precision stepping task. Exp Brain Res 126, 583–588.
, 百拇医药
Schubert M, Curt A, Jensen L & Dietz V (1997). Corticospinal input in human gait: modulation of magnetically evoked motor responses. Exp Brain Res 115, 234–246.
Seki K, Perlmutter SI & Fetz EE (2003). Sensory input to the primate spinal cord is presynaptically inhibited during voluntary movement. Nat Neurosci 6, 1309–1316.
Simonsen EB, Dyhre-Poulsen P, Alkjaer T, Aagaard P & Magnusson SP (2002). Interindividual differences in H-reflex modulation during normal walking. Exp Brain Res 142, 108–115.
, 百拇医药
Tax AAM, Van Wezel BMH & Dietz V (1995). Bipedal reflex coordination to tactile stimulation of the sural nerve during human running. J Neurophysiol 73, 1947–1964.
Van de Crommert HW, Steijvers PJ, Mulder T & Duysens J (2003). Suppressive musculocutaneous reflexes in tibialis anterior following upper leg stimulation at the end of the swing phase. Exp Brain Res 149, 405–412.
Van Doornik J, Masakado Y, Sinkjaer T & Nielsen JB (2004). The suppression of the long-latency stretch reflex in the human tibialis anterior muscle by transcranial magnetic stimulation. Exp Brain Res 157, 403–406.
, 百拇医药
Van Wezel BMH, Ottenhoff FAM & Duysens J (1997). Dynamic control of location-specific information in tactile cutaneous reflexes from the foot during human walking. J Neurosci 17, 3804–3814.
Von Holst E & Mittelstaedt H (1950). Das Reafferenzprinzip [in German]. Naturwiss 37, 464–476.
Wolpert DM (1997). Computational approaches to motor control. Trends Cogn Sci 1, 209–216.
Yang JF & Stein RB (1990). Phase-dependent reflex reversal in human leg muscles during walking. J Neurophysiol 63, 1109–1117.
Zehr EP, Komiyama T & Stein RB (1997). Cutaneous reflexes during human gait: electromyographic and kinematic responses to electrical stimulation. J Neurophysiol 77, 3311–3325., http://www.100md.com(B. C. M. Baken, P. H. J. A. Nieuwenhuijz)