Neuroscience 298 (2015) 1–11
EVIDENCE FOR BETA CORTICOMUSCULAR COHERENCE DURING HUMAN STANDING BALANCE: EFFECTS OF STANCE WIDTH, VISION, AND SUPPORT SURFACE J. V. JACOBS, * G. WU AND K. M. KELLY
based on muscle or stance condition. Results demonstrate that significant beta CMC is evident during human standing balance, and that beta CMC is responsive to changes in mechanical, but not visual or surface, conditions. Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved.
Department of Rehabilitation and Movement Science, University of Vermont, 305 Rowell Building, 106 Carrigan Drive, Burlington, VT 05405, USA
Abstract—The role of the cerebral cortex in maintaining human standing balance remains unclear. Beta corticomuscular coherence (CMC) provides a measure of communication between the sensory-motor cortex and muscle, but past literature has not demonstrated significant beta CMC during human stance. This study evaluated the effects of stance width, vision, and surface compliance on beta CMC during human stance using methods to enhance sensitivity to CMC. Ten healthy, young adults stood for three 60-s trials in each of a wide or narrow stance width while on a firm surface and in narrow stance on a foam surface, each with eyes open or closed. Beta CMC was calculated between contralateral electroencephalographic and electromyographic recordings. Electromyography was recorded from bilateral tibialis anterior and gastrocnemius lateralis muscles. CMC magnitude was defined as the average integrated area of coherence spectrum above a significance threshold. Measures of center-of-pressure (COP) sway were derived from force plates under the subjects’ feet. Results of CMC from four muscles across six stance conditions (a total of 24 combinations) demonstrated significant average CMC magnitude from every subject in 20 combinations and significant average CMC magnitude in nine of 10 subjects in the remaining four combinations. The CMC magnitude was significantly larger in the wide-stance condition than in the narrow-stance condition with eyes open. No significant differences were detected when comparing eyes-open to eyes-closed conditions or when comparing firm- to foamsurface conditions. Correlations between CMC magnitude and COP sway elicited some significant relationships, but there was no consistent direction or pattern of correlation
Key words: corticomuscular coherence, standing balance, posture, cortex, EEG, EMG.
INTRODUCTION It is essential to understand the neural mechanisms of human standing balance in order to adequately interpret and treat balance disorders and to enable more accurate programing of bipedal robotics or neuroprosthetics. The role of the cerebral cortex in human standing balance remains unclear. Although decerebrate preparations of quadruped animals demonstrate a strong capacity to maintain standing balance without cortical resources (Mori, 1987; Musienko et al., 2008; Honeycutt et al., 2009), that capacity does not necessarily determine lack of cortical utilization in intact preparations, and such control may not generalize to the biomechanically more challenging condition of human bipedal stance (Skoyles, 2006). Indeed, multiple human-subject studies suggest a potential role for the cerebral cortex in the control of human standing balance. Specifically, standing balance is impaired by stroke to the cerebral cortex, particularly with regard to the integration of sensory input for the control of postural sway (Pe´rennou et al., 2000; Bonan et al., 2004). With regard to healthy subjects, studies of transcranial magnetic stimulation and neuroimaging demonstrate that standing balance associates with enhanced cortico-cortical and cortico-spinal excitability or with enhanced cerebral blood flow compared to nonstanding or supported conditions (Obata et al., 2009; Ouchi et al., 1999; Tokuno et al., 2009). Changes in corticospinal excitability, however, are not consistently identified (Daikuya et al., 2003). Studies of electroencephalography (EEG) have identified changes in EEG oscillation frequencies across many functional bands during quiet stance (Thibault et al., 2014), suggesting altered synchronized communication of cortical circuitry, and some studies have demonstrated changes in EEG rhythms in different visual conditions (Del Percio et al., 2007) or when
*Corresponding author. Tel: +1-802-656-8647; fax: +1-802-6566586. E-mail addresses: [email protected] (J. V. Jacobs), [email protected] (G. Wu), [email protected] (K. M. Kelly). Abbreviations: AP, anterior–posterior; CMC, corticomuscular coherence; COP, center of pressure; EEG, electromyography; EMG, electroencephalography; FEC, test condition of standing in narrow stance width on a foam surface with eyes closed; FEO, test condition of standing in narrow stance width on a foam surface with eyes open; GL, gastrocnemius lateralis; ML, medial–lateral; NEC, test condition of standing in narrow stance width on a firm surface with eyes closed; NEO, test condition of standing in narrow stance width on a firm surface with eyes open; TA, tibialis anterior; WEC, test condition of standing in wide stance width on a firm surface with eyes closed; WEO, test condition of standing in wide stance width on a firm surface with eyes open. http://dx.doi.org/10.1016/j.neuroscience.2015.04.009 0306-4522/Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved. 1
2
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
spontaneously transitioning from stable to unstable stance (Slobounov et al., 2009; Varghese et al., 2015). Thus, the cerebral cortex appears involved in human standing balance at least for the purpose of monitoring postural status and perhaps also in its subsequent control. Changes in cortical excitability, hemodynamic response, or EEG oscillations, however, do not necessarily signify direct control by the cerebral cortex on standing balance. In an effort to establish a direct functional connection between muscle and the cerebral cortex during human standing balance, some studies have examined corticomuscular coherence (CMC) (Masakado et al., 2008; Vecchio et al., 2008; Murnaghan et al., 2014). CMC is a measure whereby time-varying spectral power from signals of cortical and muscle function is correlated. CMC of the beta frequency band (roughly 13–30 Hz) is most studied, often during tasks of tonic, low-to-moderate levels of contraction, and found to represent both afferent and efferent coupling between sensory-motor regions of cerebral cortex and muscle (Mima and Hallett, 1999a,b, Mima et al., 2000; Witham et al., 2011; Campfens et al., 2013). Studies of CMC during human standing balance, however, did not demonstrate the existence of significant beta CMC (Masakado et al., 2008; Murnaghan et al., 2014), although one has identified the existence of alpha (8–12 Hz) CMC, which varied between groups of differing athletic training and visual conditions (Vecchio et al., 2008). In contrast to beta CMC, the alpha CMC is less specific to functions of the sensory-motor cortex and relates to perceptual processing over association cortex. Lack of significant beta CMC during stance may relate to either the methods of collection and processing or to the limited contexts of evaluated standing tasks. In specific, the studies that focused on evaluating beta CMC (Masakado et al., 2008; Murnaghan et al., 2014) (a) analyzed EEG data only over the vertex electrode and/or at one electrode just rostral to the vertex, (b) examined stance of unknown width or one-legged stance, (c) examined stance with a backboard support that either constrained sway or moved with the individual affixed to it, (d) evaluated only ankle plantarflexors, and (e) did not necessarily utilize processing techniques that enhance CMC (Mima and Hallett, 1999b). Because beta CMC is known to be strongly dependent on the individual, the task and type of muscle contraction elicited, the chosen muscle, as well as the location of EEG electrodes, more comprehensive study designs are needed in order to adequately assess beta CMC during human stance (Ushiyama et al., 2010, 2011, 2012; Gwin and Ferris, 2012; Campfens et al., 2013). Therefore, the objectives of this study were to (1) identify the existence of beta CMC during human standing balance through methods that enhance sensitivity and (2) identify the functional relevance of beta CMC to human standing balance through variations in sensory and biomechanical constraints. We sought to enhance sensitivity of identifying beta CMC by (a) conforming to suggested methods of processing (Mima and Hallett, 1999b), (b) evaluating both ankle plantar- and dorsi-flexor muscles, (c) accounting for
inter-individual variability by evaluating each individual’s electrode of maximal CMC across a broader set of electrodes overlying the sensory-motor cortex, (d) evaluating the integrated area of the CMC spectrum over a statistical threshold rather than peak CMC amplitudes, and (e) accounting for previously untested standing task conditions by varying stance width, vision, and surface compliance. We sought to identify the functional relevance of beta CMC to human standing balance by demonstrating its modulation to altered balance constraints between conditions of stance width, vision, and surface compliance, as well as through exploratory correlations with center-of-pressure (COP) measures of sway. We hypothesized that corticomuscular coupling does exist and that it is functionally relevant to standing balance. We, therefore, predicted non-zero CMC magnitude above the significance threshold across balance conditions as well as significant effects of stance conditions on beta CMC and significant correlations between beta CMC magnitudes and COP sway. Understanding the mechanisms by which the sensorimotor cortex communicates with muscle through beta CMC during human standing balance could offer important insight into causes of impaired balance with disease or injury and potentially provide a biomarker for such impairment or for improvement with clinical interventions.
EXPERIMENTAL PROCEDURES Subjects Ten healthy young subjects (five males, five females; mean (range) age = 23 (20–29) year, height = 170 (152–183) cm, weight = 69 (48–88) kg; nine right leg dominant) provided written informed consent to participate in the protocol, which was approved by the local institutional review board. Subjects were recruited via postings on internet forums and community bulletin boards. Subjects were included based on a self-report if they were non-smokers with no known history of neurological, musculoskeletal or psychiatric disorders, no cancer or cancer treatment, and were not taking any centrally active medications. Protocol Subjects were first prepared for testing. Following shaving and cleaning with a conductive gel to obtain impedances below 10 kX, bipolar surface electromyography (EMG) electrodes (1-cm silver/silver-chloride disk electrodes with fixed 2-cm inter-electrode distance; Myotronics, Kent, WA, USA) were applied according to the SENIAM guidelines (http://www.seniam.org/) along the length of the contracted muscle bellies of the left and right tibialis anterior (TA) and gastrocnemius lateralis (GL). These muscles were chosen for the study because (1) these muscles provide complimentary plantarflexor and dorsiflexor activation for modulating multidirectional postural sway, (2) each muscle is known to generate CMC during other tasks, and (3) these distal muscles do not require filtering against cardiac artifact. Specifically, the GL was chosen for recording because it
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
exhibits relatively strong beta CMC during seated contraction (Murnaghan et al., 2014) and because the GL correlates with sway displacement in an anticipatory/ feedforward manner (Gatev et al., 1999; Masani et al., 2003), which we believe strongly suggests cortical influence and thus a good target for examining CMC during stance. The TA muscle was chosen for recording because it exhibits beta CMC during varied tasks of contraction (Gwin and Ferris, 2012; Ushiyama et al., 2012) and because it acts as a dorsiflexor for the correction of backward sway (Basmajian and De Luca, 1985) as well as provides an important source for afferent monitoring of postural sway (Day et al., 2013). In addition, the TA muscle was not evaluated in previous studies of beta CMC during human stance (Masakado et al., 2008; Murnaghan et al., 2014). We placed a Waveguard 128-channel EEG headcap (sintered silver/silver-chloride electrodes; standard 10/5 system placement (Oostenveld and Praamstra, 2001); Advanced Neuro Technology, Enschede, the Netherlands) on the subjects and used a conductive electrode gel (Electro-gel; Electro-Cap International; Eaton, OH, USA) to obtain impedances below 5 kX. Recordings were generated from the classic 10/20 system (Oostenveld and Praamstra, 2001) augmented by midline and peri-midline electrodes along the parietal to frontal zones (Fig. 1A). The task was to maintain quiet stance for three 60-s trials in each of six conditions: wide stance on a firm surface with eyes open (WEO) or closed (WEC), narrow stance on a firm surface with eyes open (NEO) or closed (NEC), and narrow stance on a foam surface (4inches, medium-density, T-Foam, Alimed Inc., Dedham, MA, USA) with eyes open (FEO) or closed (FEC). Standing on foam was in narrow stance with the intention of providing a progressively difficult standing condition from the conditions on a firm surface. Subjects were instructed to stand with their bare feet as close together as possible without any contact occurring between them for narrow-stance conditions and with the lateral edges of their feet placed along marked lines of a measured shoulder width apart for the wide-stance conditions. In all conditions, subjects were instructed to stand with arms relaxed at the sides, face relaxed, and head oriented straight ahead. When standing with eyes open, subjects were requested to fix their gaze at an eye-level target and to minimize blinking. In all conditions, subjects stood with each foot on separate force plates (AMTI OR6-6; Watertown, MA, USA) that were used to generate COP measures. The order of the six stance conditions was randomized using a random number generator for each of the three repetition blocks. Subjects wore a gait belt and were guarded by an experimenter located just behind the subject in order to prevent falls to the ground. No contact was made between the subjects and the assistants unless the center of mass appeared to sway far enough relative to the base of support that the subject would have fallen without assistance. No assistance was required for this cohort. Subjects were monitored for self-reported fatigue
3
and were allowed breaks throughout the study to prevent fatigue. Data collection and processing EMG and force-plate signals were collected at 2000 Hz and were synchronized through Nexus recording software (VICON, Centennial, CO, USA). The EMG signals were pre-amplified at the skin’s surface by 1000. The EMG signals were then filtered from 1 to 100 Hz, and the COP signals were low-pass filtered to 100 Hz using a 4th-order Butterworth filter. EEG data were collected at 1024 Hz using a DC amplifier and preprocessed using ASA software version 4.7.3 (Advanced Neuro Technology, Enschede, the Netherlands). An external trigger signal was used to synchronize the collection of EMG, force-plate, and EEG signals. Following collection, blinking artifacts were removed using the principal components analysis artifact correction parameter, selecting the two components that most represented the artifacts’ characteristics. Next, event markers were identified to splice continuously recorded EEG data into epochs of 70 s, defined from 5 s before the external trigger signal that represented the start of the trial to 5 s after the 60-s trial. Finally, the signals were band-pass filtered from 1 to 60 Hz. The remainder of the data processing was completed using customized Matlab code (The Mathworks Inc., Natick, MA, USA). EMG and COP data were resampled to 1024 Hz and aligned with the EEG data at the onset of the external trigger signal. In preparation for calculating CMC, the EEG data from each electrode were re-referenced to the common average value of all electrodes in order to improve the signal-to-noise ratio of coherence measures (Mima and Hallett, 1999b). The Matlab coherence function was used to compute CMC between the filtered EMG of each muscle and the reference-adjusted EEG of electrode Cz as well as contralateral electrodes CCP1h, CP1, C1, FC1, and FCC1h for the right TA and GL muscles, or C2, CCP2h, CP2, FC2, and FCC2h for the left TA and GL muscles (Fig. 1E, F) with a Hanning window of one second and zero overlap. This grouping of electrodes was chosen because peri-midline electrodes that overlie the sensorimotor cortex have been shown to exhibit significant beta CMC with distal leg muscles, particularly when the contraction is isotonic rather than isometric (Gwin and Ferris, 2012). CMC values are unitless between zero and one, with a value of one indicating a perfect relationship in time varying spectral power of EEG and EMG signals, and a value of zero indicating independence. A threshold of significant CMC was determined by the 95% confidence limit = 1 (0.05)1/n 1, where n is the number of windows used for spectral estimation (Halliday et al., 1995). Once calculated, the CMC was plotted and the maximum peak in the beta (13–30 Hz) frequency band was manually selected with an interactive graphing function in Matlab. The beta frequencies were selected because this frequency band is known to (1) arise from the sensory-motor cortex (2) be
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
4
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
Fig. 1. (A) Electroencephalographic (EEG) recording montage and electrodes examined for CMC; electrodes with gray fill represent those that were examined for CMC. An example trial from one subject in the NEO condition illustrates (B) electromyographic (EMG) traces from each muscle; (C) center of pressure (COP) traces under each foot in the anterior-posterior (black traces) and medial–lateral (gray traces) directions, with positive values indicating forward or rightward displacements; (D) EEG traces from all the electrodes examined for cortico-muscular coherence (CMC); (E) CMC spectrum lines from all 6 contralateral or midline EEG electrodes with each muscle across all frequency bands (gray shaded region represents the beta band, gray dashed line represents the significance threshold); (F) a magnified CMC line from (E), identifying the area above the threshold for significant CMC (referred to as CMC magnitude), the beta frequency at which peak CMC is located, and the peak amplitude of CMC.
relevant to sub-maximal muscle contraction, and (3) be appropriate for analysis using surface EEG signals (Grosse et al., 2002). Once the peak was selected, the true peak was automatically identified within the close vicinity of the user-selected peak by the Matlab software. The magnitude of significant CMC was then determined as the integrated area of the CMC spectrum around the selected peak above the significance threshold (Fig. 1F).
Trials without significant CMC were valued at zero. The electrode that exhibited the maximum CMC magnitude was then identified for each individual and trial, and each trial’s CMC magnitude from the electrode of maximum magnitude was averaged by condition for each subject for analysis. The location of maximum CMC magnitude was selected to enhance sensitivity to CMC as well as to account for inter-individual and task-related differences
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
in functional representation of the sensory-motor cortex (Kurtzer et al., 2005; Scott, 2008; Koessler et al., 2009). The root-mean-square, path length, range, and mean velocity of COP displacements in the anterior–posterior (AP) and medial–lateral (ML) directions were also averaged by condition for each subject for analysis. Statistical analysis Descriptive statistics of frequency histograms determined the incidence of subjects with significant CMC per condition and muscle in order to first establish that beta CMC exists; for comparison, these frequencies were determined as the number of subjects with an average area of CMC spectrum above the significance threshold as well as with an average peak value of CMC above the significance threshold. Frequency histograms were also generated to identify the frequency at which each electrode location represented the location of maximum CMC magnitude for each condition and muscle. The functional relevance of beta CMC was then evaluated by identifying whether it changes significantly due to stance width, vision, or support surface. Due to the sample size and a common failure to exhibit normal distributions, as assessed by Shapiro–Wilk tests, Wilcoxon Signed-Ranks Tests were used to generate a priori comparisons of the WEO and NEO conditions in order to assess effects of stance width on CMC, the WEO and WEC conditions in order to assess effects of vision on CMC, and the NEO and FEO conditions in order to assess the effects of support surface on CMC. Other combinations of conditions reflecting effects of stance width, vision or surface (including NEO and NEC, FEO and FEC, WEC and NEC, as well as NEC and FEC) were also examined as a post hoc exploratory analysis. Friedman’s ANOVAs were not used across all conditions because we did not have an a priori interest in all potential combinations of condition comparisons. In order to further evaluate the functional relevance of beta CMC during stance, we generated an exploratory analysis of Spearman’s rho correlation coefficients between CMC magnitudes and four measures of whole-body COP sway (maximum range, mean velocity, total displacement, and root-mean-square amplitude). Lastly, in order to determine whether the right and left limbs operated synchronously or with some independence, a post hoc evaluation of cross-correlation coefficients between left and right COP displacements was computed with a Matlab function. Significance was assumed at an alpha level of 0.05, and all analyses were performed in SPSS version 21 software (IBM, Armonk, NY, USA).
RESULTS Fig. 1B–D illustrates time trajectories of EMG from four leg muscles, COP under each foot, and EEG from 11 electrodes. It was evident that both anterior and posterior leg muscles are activated during the 60 s of stance, and that these muscle activations are synchronized with the COP sway (forward COP displacement coincided with activation of the GL and/or inhibition of the TA, and backward COP displacement coincided with activation of
5
the TA and/or inhibition of the GL). In addition, COP displacements under the left and right foot were not completely synchronized, suggesting an asymmetric control of standing posture from each side of the body. Further, the CMC peaks were heterogeneous across electrodes and observed in alpha, beta, and gamma ranges (Fig. 1E). Again, however, this study focused on the beta band because it is known to (1) arise from the sensory-motor cortex (2) be relevant to sub-maximal muscle contraction, and (3) be appropriate for analysis using surface EEG signals (Grosse et al., 2002). Existence of significant beta CMC during human standing balance When evaluating the existence of significant beta CMC using the area of CMC spectrum above the significance threshold (Fig. 2), all 10 subjects exhibited significant beta CMC magnitudes in 20 of the 24 muscle-condition combinations. Significant CMC magnitudes were evident in nine of 10 subjects for the remaining four musclecondition combinations. Three separate subjects were responsible for these four instances of dropout. Although the incidence of significant beta CMC was consistent using the average area above the significance threshold, peak amplitudes of beta CMC were sometimes low and the incidence of significant peak amplitudes of beta CMC was lower than that of the area of CMC spectrum above significance (Fig. 2). The difference in incidence reflects inter-trial variability within subjects and conditions, such that some trials did not reach the threshold of significant beta CMC. Therefore, when averaging peak amplitudes, even when one or two trials exhibit amplitudes above the significance threshold, if at least one trial exhibited a peak amplitude far enough below significance to decrease the average value below the threshold for significance, then no significant beta CMC would be identified in that average despite some trials exhibiting significant CMC. In contrast, when averaging the area above the significance threshold, any trials without CMC values that cross the threshold for significance would be counted as a value of zero; thus, if any one trial exhibited beta CMC above the significance threshold, the average value across trials would also be non-zero, thereby indicating evidence of significant beta CMC. The group’s mean peak frequency of CMC ranged between 20.5 and 23.9 Hz across all muscle-condition combinations. Wilcoxon signed ranks tests revealed no significant differences due to condition except for an effect of support surface, in which the peak frequency was lower in the FEO condition than in the NEO condition: mean (95% confidence interval) was 20.5 (19.1–21.9) Hz in the FEO condition and 23.8 (22.2–25.5) Hz in the NEO condition (Wilcoxon Z = 2.19; P = 0.027). Location of EEG electrodes with maximum CMC magnitudes The EEG electrode location of maximum CMC magnitudes was consistent across trials within subjects and within each condition, but the location of maximum
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
6
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
Fig. 2. Frequency histograms of the number of subjects exhibiting significant beta CMC for each condition and muscle as assessed by either a nonzero average of the area of the CMC spectrum above the significance threshold (filled bars) or an average peak amplitude of CMC above the significance threshold (unfilled bars). WEO, wide, eyes open; WEC, wide, eyes closed; NEO, narrow, eyes open; NEC, narrow, eyes closed; FEO, foam, eyes open; FEC, foam, eyes closed.
CMC magnitude was not consistent across conditions or among different subjects (Fig. 3). The vertex (Cz) electrode that was the focus of past studies of beta CMC during human stance (Masakado et al., 2008; Murnaghan et al., 2014) was the most common (or tied as the most common) location of maximum CMC in seven of the 24 muscle-condition combinations. The FC1/FC2 electrodes were most common in nine muscle-condition combinations; FCC1h/FCC2h, six combinations; C1/C2, four combinations; CCP1h/CCP2h, four combinations; and CP1/CP2, two combinations. Functional relevance of beta CMC during human stance: condition effects Fig. 4 illustrates the individual means and group means of beta CMC magnitudes (as assessed by the area of CMC spectrum above the significance threshold), contrasting the a priori condition comparisons of WEO to NEO, WEO to WEC, and NEO to FEO for the left and right TA and GL muscles. Beta CMC magnitudes were significantly larger in the WEO condition than in the NEO condition for the right GL muscle. It is worthy to note that this effect remained significant even after a very conservative Bonferroni correction of 12 musclecondition combinations. No other significant effects of condition were evident in the a priori comparisons or in the exploratory comparisons of other condition pairs. Functional relevance of beta CMC during human stance: correlations with sway and sway measures With four COP variables each assessed in the ML and AP directions, and CMC from four muscles each in six
conditions, a total of 192 correlations were generated in this exploratory analysis. Seven significant correlations were evident from these 192 tests (Table 1), such that only the root-mean-square and range of COP sway elicited significant correlations with beta CMC. The significant correlations varied with regard to the direction of correlation, the condition in which the correlation was evident, and the muscle of CMC with which the correlation was evident. Table 2 presents the mean values of the root-meansquare and range of COP sway for the WEO, NEO, NEC, and FEO conditions. Both the AP root-mean square and range of COP sway were significantly different between the WEO and NEO conditions (Wilcoxon Z = 2.50, 2.29; P = 0.010, 0.020) and between the NEO and FEO conditions (Wilcoxon Z = 2.40, 2.80; P = 0.014, 0.002), but not between the NEO and NEC conditions (Wilcoxon Z = 1.19, 0.97; P = 0.27, 0.37). The ML range of COP sway was significantly different between the WEO and NEO conditions (Wilcoxon Z = 2.80; P = 0.002), the NEO and FEO conditions (Wilcoxon Z = 2.80; P = 0.002), as well as the NEO and NEC conditions (Wilcoxon Z = 2.80; P = 0.002). The ML root-meansquare of COP sway was significantly different between the WEO and NEO conditions (Wilcoxon Z = 2.80; P = 0.002), but not the NEO and FEO conditions (Wilcoxon Z = 1.58; P = 0.13) or the NEO and NEC conditions (Wilcoxon Z = 0.05; P = 1.0). Because CMC magnitudes exhibited different values between muscles, and because CMC was responsive to a change in stance width at only the right limb, we performed a post hoc analysis of cross-correlation between the COP displacements of the right vs left limb
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
7
Fig. 3. Frequency histograms of the number of subjects for whom each electrode represented the electrode of maximum CMC for each muscle and condition.
(Table 3). The results exception of standing on did not approach a value unique behavior of the left
demonstrate that, with the foam, correlation coefficients of 1.0, thus demonstrating a vs right limb.
DISCUSSION The results supported our hypothesis that beta CMC exists during human standing balance and partially
supported the hypothesis that beta CMC is functionally relevant to changes in stance condition and to postural sway during stance. In specific, our methods of processing elicited significant CMC from all subjects and all recorded muscles in most of the tested conditions, at least demonstrating significant CMC from nine of 10 subjects in all conditions and muscles. Beta CMC was consistently sensitive to changes in biomechanical orientation, such that beta CMC was higher when
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
8
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
Fig. 4. Individual (gray lines) and group (black circles) mean CMC magnitudes for each muscle. CMC magnitudes represent the integrated area of the CMC spectrum above the significance threshold. Rows specify muscles. Columns specify condition comparisons for effects of stance width, vision, and support surface. Wilcoxon signed rank test statistics are provided for each comparison. Error bars on the group means represent 95% confidence intervals.
standing in a wide stance width compared to a narrow stance width for the right GL muscle. No consistent effect on beta CMC magnitude was evident, however, when transitioning from eyes-open to eyes-closed conditions, or when transitioning from a firm to foam surface. Significant correlations with measures of COP
sway were evident, although the direction of correlations differed among muscle-condition combinations. Our ability to demonstrate significant beta CMC in contrast to previous studies (Masakado et al., 2008; Murnaghan et al., 2014) likely reflects differences in processing methods. Our data demonstrate that it may be
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
9
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11 Table 1. Significant correlations between beta CMC magnitudes and measures of COP sway Stance condition
Muscle of CMC
Measure of COP sway
Spearman’s Rho (P-value)
Wide, Eyes open Narrow, Eyes open Narrow, Eyes open Narrow, Eyes closed Narrow, Eyes closed Foam, Eyes closed Foam, Eyes closed
Right Tibialis anterior Left Gastrocnemius lateralis Left Gastrocnemius lateralis Right Gastrocnemius lateralis Left Tibialis anterior Right Tibialis anterior Right Tibialis anterior
Medial–lateral Root-mean-square Anterior–posterior Root-mean-square Medial–lateral Root-mean-square Medial–lateral Root-mean-square Medial–lateral Range of displacement Anterior–posterior Range of displacement Medial–lateral Range of displacement
0.709 (0.022) 0.661 (0.038) +0.636 (0.048) 0.891 (0.00054) 0.685 (0.029) +0.685 (0.029) +0.709 (0.022)
Table 2. Anterior–posterior (AP) and medial–lateral (ML) COP sway by condition COP measure
Condition
Mean (95% CI) value
AP root-mean-square COP (% foot length)
WEO NEO NEC FEO WEO NEO NEC FEO WEO NEO
14.9 (8.2–21.7) 19.8 (14.7–24.9) 18.2 (15.0–21.4) 15.4 (11.5–19.3) 11.5 (9.2–13.8) 13.4 (11.1–15.7) 14.3 (12.0–16.6) 22.7 (18.9–26.6) 62.9 (58.1–67.6) 119.0 (109.9– 128.1) 119.1 (109.8– 128.5) 118.2 (108.7– 127.8) 3.0 (2.6–3.4) 12.9 (10.7–15.1) 16.0 (13.7–18.3) 17.8 (14.4–21.2)
AP range of COP displacement (% foot length) ML root-mean-square COP (% stance width)
NEC FEO ML range of COP displacement (% stance width)
WEO NEO NEC FEO
important to account for inter-individual Koessler et al., 2009) and inter-condition (Kurtzer et al., 2005; Scott, 2008) differences by analyzing the electrode of maximal CMC for each subject and by evaluating multiple electrodes along and adjacent to the midline over sensorymotor cortex. In addition, it was clear from our analysis that inter-trial variability in beta CMC renders it important to evaluate multiple trials. Sensitivity for identifying significant beta CMC is also enhanced by evaluating the area above the threshold of significant CMC rather than the peak amplitude of CMC. We also re-referenced our data to a common average reference in order to enhance signal-to-noise ratio, as previously suggested (Mima and Hallett, 1999b). In our attempt to understand the functional significance of beta CMC, we did identify that beta CMC modulates with stance width at the right GL muscle. The decrease in beta CMC from wide stance to narrow
Table 3. Coefficients from Cross-correlations of COP under left vs right limb Condition
WEO NEO NEC FEO
Mean (95% CI) correlation coefficient Anterior–posterior COP
Medial–lateral COP
0.77 0.63 0.55 0.92
0.52 (0.32–0.72) 0.20 ( 0.44–0.03) 0.17 ( 0.34–0.00) 0.70 (0.48–0.92)
(0.73–0.82) (0.50–0.77) (0.42–0.67) (0.89–0.95)
stance at the GL muscle may reflect a transition to a sway-modulated isotonic contraction from an isometric contraction afforded by the mechanical stability of a wide stance. Indeed, more sway was evident in the narrow-stance conditions than in the wide-stance condition. Studies have demonstrated that beta CMC is most evident in isometric conditions (Gwin and Ferris, 2012; Ushiyama et al., 2012). The unilateral effect on only the GL muscle could reflect a combination of the plantarflexors’ role in standing postural sway and limb asymmetries during standing balance. Our data demonstrate that COP displacements under each limb were not completely symmetric. Given these limb asymmetries, it would therefore be expected that afferent feedback about limb status or descending control of a limb’s function would also differ, which would subsequently generate different CMC responses. The specificity of effect of stance width on the GL muscle and not the TA muscle could reflect that the GL muscle is more active to modify postural sway than the TA muscle. Whereas the GL muscle has been found to associate with postural sway with a leading phase shift relative to the sway signal (Gatev et al., 1999; Masani et al., 2003), the TA muscle’s influence on sway may be more transient to correct larger backward sway displacements and otherwise acts as a more passive source of afferent feedback regarding joint position during postural sway (Basmajian and De Luca, 1985; Day et al., 2013). The GL muscle’s responsiveness to changes in stance width, thus, reflects the GL muscle’s influence on postural sway and suggests that either cortical monitoring of
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
10
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
afferent input from the GL muscle’s activation or control by the cortex on the GL becomes up-regulated in the beta frequency band when in a wide stance. Lack of effect of changes in vision or surface on beta CMC was not expected. Previous studies have identified that cortical lesions affect postural sway under changing sensory conditions (Bonan et al., 2004), and changes in the availability of vision alter activity of the parietal and prefrontal cortex in healthy subjects (Ouchi et al., 1999; Del Percio et al., 2007). Thus, we anticipated that this cortical role in sensory integration during standing balance would elicit altered communication between the sensory-motor cortex and the muscle in order to modulate postural sway. Lack of effect of vision or surface on CMC in this study, however, is consistent with another study using transcranial magnetic stimulation that demonstrated a lack of change in corticospinal excitability between standing conditions with or without vision, or when standing on a firm vs foam surface (Baudry et al., 2014). Thus, the parieto-prefrontal changes that occur when standing with vs without vision must mediate changes in postural sway via extra-pyramidal pathways not involving the sensory-motor cortex. Although this study provides the initial evidence that beta CMC exists during human standing balance and provides insight into the responsiveness of corticomuscular communication to changes in mechanics and sensory conditions, certain methodological considerations warrant further study. Our study evaluated only the TA and GL muscles, but other muscles, such as the soleus and more proximal muscles, are known to modulate postural sway as well. The relative influence of proximal musculature might be even more relevant to ML sway or to hip strategies adopted during more difficult standing conditions such as when on a foam surface. Communication of these other muscles with the sensory-motor cortex during these difficult standing conditions is worth further investigation in future studies. In addition, although we identified significant correlations between postural sway and beta CMC magnitudes, these correlations did not demonstrate a clear pattern with regard to muscle, condition, direction of sway, or direction of correlation. Due to the large number of correlations generated by our analysis, it remains unclear whether these relationships are true and reflective of complex interactions, or if they are spurious findings amid multiple comparisons derived from an exploratory analysis. Lastly, functional associations between CMC and sway dynamics may be more apparent in other frequency bands; previous studies have demonstrated the alpha and gamma bands are sensitive to detecting unstable stance and can be sensitive to changes in sensory conditions or be discriminative between groups with different athletic backgrounds (Vecchio et al., 2008; Slobounov et al., 2009). Additional studies with larger sample sizes are, thus, needed in order to evaluate interactions among many potential contributing factors as well as to better understand whether corticomuscular communication with other muscles and at other frequencies functionally modulates postural sway.
CONCLUSIONS This study demonstrates that, contrary to past literature, significant beta CMC can be consistently detected across subjects after accounting for important methodological considerations of EEG referencing, selected electrodes used for analysis, and choice in outcome measure. For the healthy young adults of this study, corticomuscular communication with the sensorymotor cortex appears responsive to changes in mechanical orientation during standing balance, whereas pathways that do not directly involve the sensory-motor cortex appear to mediate changes in muscle activation due to changes in visual or surface conditions. Although this study focuses on the functional relevance of beta CMC during standing balance for healthy young adults, the functional relevance of corticomuscular coupling may differ for older populations or for those with health conditions (Grosse et al., 2002). Thus, while we have confirmed the existence of beta CMC during healthy human stance with a novel insight into the functional responsiveness of beta CMC to changes in stance conditions, its relevance to impaired standing balance should still be examined. Acknowledgments—A Research Incentives Grant from the University of Vermont’s College of Nursing and Health Sciences funded the study. We thank Sarah Aghjayan, Juvena Hitt, Elena Isaacson, Anne Patti, Julia Parks, Claire Purcell, and Xi Wen for assistance with data collection and processing.
REFERENCES Basmajian JV, De Luca CJ (1985) Muscles alive. 5th ed. Baltimore, MD: Williams and Wilkins. Baudry S, Penzer F, Duchateau J (2014) Vision and proprioception do not influence the excitability of the corticomotoneuronal pathway during upright standing in young and elderly adults. Neuroscience 268:247–254. Bonan IV, Colle FM, Guichard JP, Vicaut E, Eisenfisz M, Tran Ba Huy P, Yelnik AP (2004) Reliance on visual information after stroke. Part I: Balance on dynamic posturography. Arch Phys Med Rehabil 85:268–273. Campfens SF, Schouten AC, van Putten MJ, van der Kooij H (2013) Quantifying connectivity via efferent and afferent pathways in motor control using coherence measures and joint position perturbations. Exp Brain Res 228:141–153. Daikuya S, Tanino Y, Nishimori T, Takasaki K, Suzuki T (2003) The silent period from soleus and gastrocnemius muscles in relation to conditions of standing. Electromyogr Clin Neurophysiol 43:217–222. Day JT, Lichtwark GA, Cresswell AG (2013) Tibialis anterior muscle fascicle dynamics adequately represent postural sway during standing balance. J Appl Physiol 115:1742–1750. Del Percio C, Brancucci A, Bergami F, Marzano N, Fiore A, Di Ciolo E, Aschieri P, Lino A, Vecchio F, Iacoboni M, Gallamini M, Babiloni C, Eusebi F (2007) Cortical alpha rhythms are correlated with body sway during quiet open-eyes standing in athletes: a high-resolution EEG study. NeuroImage 36:822–829. Gatev P, Thomas S, Kepple T, Hallett M (1999) Feedforward ankle strategy of balance during quiet stance in adults. J Physiol 514:915–928. Grosse P, Cassidy MJ, Brown P (2002) EEG–EMG, MEG–EMG and EMG–EMG frequency analysis: physiological principles and clinical applications. Clin Neurophysiol 113:1523–1531.
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11 Gwin JT, Ferris DP (2012) Beta- and gamma-range human lower limb corticomuscular coherence. Front Hum Neurosci 6:258. Halliday DM, Rosenberg JR, Amjad AM, Breeze P, Conway BA, Farmer SF (1995) A framework for the analysis of mixed time series/point process data–theory and application to the study of physiological tremor, single motor unit discharges and electromyograms. Prog Biophys Mol Biol 64:237–278. Honeycutt CF, Gottschall JS, Nichols TR (2009) Electromyographic responses from the hindlimb muscles of the decerebrate cat to horizontal support surface perturbations. J Neurophysiol 101:2751–2761. Koessler L, Maillard L, Benhadid A, Vignal JP, Felblinger J, Vespignani H, Braun M (2009) Automated cortical projection of EEG sensors: anatomical correlation via the international 10–10 system. NeuroImage 46:64–72. Kurtzer I, Herter TM, Scott SH (2005) Random change in cortical load representation suggests distinct control of posture and movement. Nat Neurosci 8:498–504. Masakado Y, Ushiba J, Tsutsumi N, Takahashi Y, Tomita Y, Kimura A, Liu M (2008) EEG–EMG coherence changes in postural tasks. Electromyogr Clin Neurophysiol 48:27–33. Masani K, Popovic MR, Nakazawa K, Kouzaki M, Nozaki D (2003) Importance of body sway velocity information in controlling ankle extensor activities during quiet stance. J Neurophysiol 90:3774–3782. Mima T, Hallett M (1999a) Corticomuscular coherence: a review. J Clin Neurophysiol 16:501–511. Mima T, Hallett M (1999b) Electroencephalographic analysis of cortico-muscular coherence: reference effect, volume conduction and generator mechanism. Clin Neurophysiol 110:1892–1899. Mima T, Matsuoka T, Hallett M (2000) Functional coupling of human right and left cortical motor areas demonstrated with partial coherence analysis. Neurosci Lett 287:93–96. Mori S (1987) Integration of posture and locomotion in acute decerebrate cats and in awake, freely moving cats. Prog Neurobiol 28:161–195. Murnaghan CD, Squair JW, Chua R, Inglis JT, Carpenter MG (2014) Cortical contributions to control of posture during unrestricted and restricted stance. J Neurophysiol 111:1920–1926. Musienko PE, Zelenin PV, Lyalka VF, Orlovsky GN, Deliagina TG (2008) Postural performance in decerebrated rabbit. Behav Brain Res 190:124–134. Obata H, Sekiguchi H, Nakazawa K, Ohtsuki T (2009) Enhanced excitability of the corticospinal pathway of the ankle extensor and flexor muscles during standing in humans. Exp Brain Res 197:207–213.
11
Oostenveld R, Praamstra P (2001) The five percent electrode system for high-resolution EEG and ERP measurements. Clin Neurophysiol 112:713–719. Ouchi Y, Okada H, Yoshikawa E, Nobezawa S, Futatsubashi M (1999) Brain activation during maintenance of standing postures in humans. Brain 122(Pt. 2):329–338. Pe´rennou DA, Leblond C, Amblard B, Micallef JP, Rouget E, Pe´lissier J (2000) The polymodal sensory cortex is crucial for controlling lateral postural stability: evidence from stroke patients. Brain Res Bull 53:359–365. Scott SH (2008) Inconvenient truths about neural processing in primary motor cortex. J Physiol 586:1217–1224. Skoyles (2006) Human balance, the evolution of bipedalism and disequilibrium syndrome. Med Hypotheses 66:1060–1068. Slobounov S, Cao C, Jaiswal N, Newell KM (2009) Neural basis of postural instability identified by VTC and EEG. Exp Brain Res 199:1–16. Thibault RT, Lifshitz M, Jones JM, Raz A (2014) Posture alters human resting-state. Cortex 58:199–205. Tokuno CD, Taube W, Cresswell AG (2009) An enhanced level of motor cortical excitability during the control of human standing. Acta Physiol (Oxf) 195:385–395. Ushiyama J, Takahashi Y, Ushiba J (2010) Muscle dependency of corticomuscular coherence in upper and lower limb muscles and training-related alterations in ballet dancers and weightlifters. J Appl Physiol (1985) 109:1086–1095. Ushiyama J, Suzuki T, Masakado Y, Hase K, Kimura A, Liu M, Ushiba J (2011) Between-subject variance in the magnitude of corticomuscular coherence during tonic isometric contraction of the tibialis anterior muscle in healthy young adults. J Neurophysiol 106:1379–1388. Ushiyama J, Masakado Y, Fujiwara T, Tsuji T, Hase K, Kimura A, Liu M, Ushiba J (2012) Contraction level-related modulation of corticomuscular coherence differs between the tibialis anterior and soleus muscles in humans. J Appl Physiol (1985) 112:1258–1267. Varghese JP, Beyer KB, Williams L, Miyasike-daSilva V, McIlroy WE (2015) Standing still: Is there a role for the cortex? Neurosci Lett 590:18–23. Vecchio F, Del Percio C, Marzano N, Fiore A, Toran G, Aschieri P, Gallamini M, Cabras J, Rossini PM, Babiloni C, Eusebi F (2008) Functional cortico-muscular coupling during upright standing in athletes and nonathletes: a coherence electroencephalographic– electromyographic study. Behav Neurosci 122:917–927. Witham CL, Riddle CN, Baker MR, Baker SN (2011) Contributions of descending and ascending pathways to corticomuscular coherence in humans. J Physiol 589:3789–3800.
(Accepted 6 April 2015) (Available online 11 April 2015)
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
EVIDENCE FOR BETA CORTICOMUSCULAR COHERENCE DURING HUMAN STANDING BALANCE: EFFECTS OF STANCE WIDTH, VISION, AND SUPPORT SURFACE J. V. JACOBS, * G. WU AND K. M. KELLY
based on muscle or stance condition. Results demonstrate that significant beta CMC is evident during human standing balance, and that beta CMC is responsive to changes in mechanical, but not visual or surface, conditions. Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved.
Department of Rehabilitation and Movement Science, University of Vermont, 305 Rowell Building, 106 Carrigan Drive, Burlington, VT 05405, USA
Abstract—The role of the cerebral cortex in maintaining human standing balance remains unclear. Beta corticomuscular coherence (CMC) provides a measure of communication between the sensory-motor cortex and muscle, but past literature has not demonstrated significant beta CMC during human stance. This study evaluated the effects of stance width, vision, and surface compliance on beta CMC during human stance using methods to enhance sensitivity to CMC. Ten healthy, young adults stood for three 60-s trials in each of a wide or narrow stance width while on a firm surface and in narrow stance on a foam surface, each with eyes open or closed. Beta CMC was calculated between contralateral electroencephalographic and electromyographic recordings. Electromyography was recorded from bilateral tibialis anterior and gastrocnemius lateralis muscles. CMC magnitude was defined as the average integrated area of coherence spectrum above a significance threshold. Measures of center-of-pressure (COP) sway were derived from force plates under the subjects’ feet. Results of CMC from four muscles across six stance conditions (a total of 24 combinations) demonstrated significant average CMC magnitude from every subject in 20 combinations and significant average CMC magnitude in nine of 10 subjects in the remaining four combinations. The CMC magnitude was significantly larger in the wide-stance condition than in the narrow-stance condition with eyes open. No significant differences were detected when comparing eyes-open to eyes-closed conditions or when comparing firm- to foamsurface conditions. Correlations between CMC magnitude and COP sway elicited some significant relationships, but there was no consistent direction or pattern of correlation
Key words: corticomuscular coherence, standing balance, posture, cortex, EEG, EMG.
INTRODUCTION It is essential to understand the neural mechanisms of human standing balance in order to adequately interpret and treat balance disorders and to enable more accurate programing of bipedal robotics or neuroprosthetics. The role of the cerebral cortex in human standing balance remains unclear. Although decerebrate preparations of quadruped animals demonstrate a strong capacity to maintain standing balance without cortical resources (Mori, 1987; Musienko et al., 2008; Honeycutt et al., 2009), that capacity does not necessarily determine lack of cortical utilization in intact preparations, and such control may not generalize to the biomechanically more challenging condition of human bipedal stance (Skoyles, 2006). Indeed, multiple human-subject studies suggest a potential role for the cerebral cortex in the control of human standing balance. Specifically, standing balance is impaired by stroke to the cerebral cortex, particularly with regard to the integration of sensory input for the control of postural sway (Pe´rennou et al., 2000; Bonan et al., 2004). With regard to healthy subjects, studies of transcranial magnetic stimulation and neuroimaging demonstrate that standing balance associates with enhanced cortico-cortical and cortico-spinal excitability or with enhanced cerebral blood flow compared to nonstanding or supported conditions (Obata et al., 2009; Ouchi et al., 1999; Tokuno et al., 2009). Changes in corticospinal excitability, however, are not consistently identified (Daikuya et al., 2003). Studies of electroencephalography (EEG) have identified changes in EEG oscillation frequencies across many functional bands during quiet stance (Thibault et al., 2014), suggesting altered synchronized communication of cortical circuitry, and some studies have demonstrated changes in EEG rhythms in different visual conditions (Del Percio et al., 2007) or when
*Corresponding author. Tel: +1-802-656-8647; fax: +1-802-6566586. E-mail addresses: [email protected] (J. V. Jacobs), [email protected] (G. Wu), [email protected] (K. M. Kelly). Abbreviations: AP, anterior–posterior; CMC, corticomuscular coherence; COP, center of pressure; EEG, electromyography; EMG, electroencephalography; FEC, test condition of standing in narrow stance width on a foam surface with eyes closed; FEO, test condition of standing in narrow stance width on a foam surface with eyes open; GL, gastrocnemius lateralis; ML, medial–lateral; NEC, test condition of standing in narrow stance width on a firm surface with eyes closed; NEO, test condition of standing in narrow stance width on a firm surface with eyes open; TA, tibialis anterior; WEC, test condition of standing in wide stance width on a firm surface with eyes closed; WEO, test condition of standing in wide stance width on a firm surface with eyes open. http://dx.doi.org/10.1016/j.neuroscience.2015.04.009 0306-4522/Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved. 1
2
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
spontaneously transitioning from stable to unstable stance (Slobounov et al., 2009; Varghese et al., 2015). Thus, the cerebral cortex appears involved in human standing balance at least for the purpose of monitoring postural status and perhaps also in its subsequent control. Changes in cortical excitability, hemodynamic response, or EEG oscillations, however, do not necessarily signify direct control by the cerebral cortex on standing balance. In an effort to establish a direct functional connection between muscle and the cerebral cortex during human standing balance, some studies have examined corticomuscular coherence (CMC) (Masakado et al., 2008; Vecchio et al., 2008; Murnaghan et al., 2014). CMC is a measure whereby time-varying spectral power from signals of cortical and muscle function is correlated. CMC of the beta frequency band (roughly 13–30 Hz) is most studied, often during tasks of tonic, low-to-moderate levels of contraction, and found to represent both afferent and efferent coupling between sensory-motor regions of cerebral cortex and muscle (Mima and Hallett, 1999a,b, Mima et al., 2000; Witham et al., 2011; Campfens et al., 2013). Studies of CMC during human standing balance, however, did not demonstrate the existence of significant beta CMC (Masakado et al., 2008; Murnaghan et al., 2014), although one has identified the existence of alpha (8–12 Hz) CMC, which varied between groups of differing athletic training and visual conditions (Vecchio et al., 2008). In contrast to beta CMC, the alpha CMC is less specific to functions of the sensory-motor cortex and relates to perceptual processing over association cortex. Lack of significant beta CMC during stance may relate to either the methods of collection and processing or to the limited contexts of evaluated standing tasks. In specific, the studies that focused on evaluating beta CMC (Masakado et al., 2008; Murnaghan et al., 2014) (a) analyzed EEG data only over the vertex electrode and/or at one electrode just rostral to the vertex, (b) examined stance of unknown width or one-legged stance, (c) examined stance with a backboard support that either constrained sway or moved with the individual affixed to it, (d) evaluated only ankle plantarflexors, and (e) did not necessarily utilize processing techniques that enhance CMC (Mima and Hallett, 1999b). Because beta CMC is known to be strongly dependent on the individual, the task and type of muscle contraction elicited, the chosen muscle, as well as the location of EEG electrodes, more comprehensive study designs are needed in order to adequately assess beta CMC during human stance (Ushiyama et al., 2010, 2011, 2012; Gwin and Ferris, 2012; Campfens et al., 2013). Therefore, the objectives of this study were to (1) identify the existence of beta CMC during human standing balance through methods that enhance sensitivity and (2) identify the functional relevance of beta CMC to human standing balance through variations in sensory and biomechanical constraints. We sought to enhance sensitivity of identifying beta CMC by (a) conforming to suggested methods of processing (Mima and Hallett, 1999b), (b) evaluating both ankle plantar- and dorsi-flexor muscles, (c) accounting for
inter-individual variability by evaluating each individual’s electrode of maximal CMC across a broader set of electrodes overlying the sensory-motor cortex, (d) evaluating the integrated area of the CMC spectrum over a statistical threshold rather than peak CMC amplitudes, and (e) accounting for previously untested standing task conditions by varying stance width, vision, and surface compliance. We sought to identify the functional relevance of beta CMC to human standing balance by demonstrating its modulation to altered balance constraints between conditions of stance width, vision, and surface compliance, as well as through exploratory correlations with center-of-pressure (COP) measures of sway. We hypothesized that corticomuscular coupling does exist and that it is functionally relevant to standing balance. We, therefore, predicted non-zero CMC magnitude above the significance threshold across balance conditions as well as significant effects of stance conditions on beta CMC and significant correlations between beta CMC magnitudes and COP sway. Understanding the mechanisms by which the sensorimotor cortex communicates with muscle through beta CMC during human standing balance could offer important insight into causes of impaired balance with disease or injury and potentially provide a biomarker for such impairment or for improvement with clinical interventions.
EXPERIMENTAL PROCEDURES Subjects Ten healthy young subjects (five males, five females; mean (range) age = 23 (20–29) year, height = 170 (152–183) cm, weight = 69 (48–88) kg; nine right leg dominant) provided written informed consent to participate in the protocol, which was approved by the local institutional review board. Subjects were recruited via postings on internet forums and community bulletin boards. Subjects were included based on a self-report if they were non-smokers with no known history of neurological, musculoskeletal or psychiatric disorders, no cancer or cancer treatment, and were not taking any centrally active medications. Protocol Subjects were first prepared for testing. Following shaving and cleaning with a conductive gel to obtain impedances below 10 kX, bipolar surface electromyography (EMG) electrodes (1-cm silver/silver-chloride disk electrodes with fixed 2-cm inter-electrode distance; Myotronics, Kent, WA, USA) were applied according to the SENIAM guidelines (http://www.seniam.org/) along the length of the contracted muscle bellies of the left and right tibialis anterior (TA) and gastrocnemius lateralis (GL). These muscles were chosen for the study because (1) these muscles provide complimentary plantarflexor and dorsiflexor activation for modulating multidirectional postural sway, (2) each muscle is known to generate CMC during other tasks, and (3) these distal muscles do not require filtering against cardiac artifact. Specifically, the GL was chosen for recording because it
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
exhibits relatively strong beta CMC during seated contraction (Murnaghan et al., 2014) and because the GL correlates with sway displacement in an anticipatory/ feedforward manner (Gatev et al., 1999; Masani et al., 2003), which we believe strongly suggests cortical influence and thus a good target for examining CMC during stance. The TA muscle was chosen for recording because it exhibits beta CMC during varied tasks of contraction (Gwin and Ferris, 2012; Ushiyama et al., 2012) and because it acts as a dorsiflexor for the correction of backward sway (Basmajian and De Luca, 1985) as well as provides an important source for afferent monitoring of postural sway (Day et al., 2013). In addition, the TA muscle was not evaluated in previous studies of beta CMC during human stance (Masakado et al., 2008; Murnaghan et al., 2014). We placed a Waveguard 128-channel EEG headcap (sintered silver/silver-chloride electrodes; standard 10/5 system placement (Oostenveld and Praamstra, 2001); Advanced Neuro Technology, Enschede, the Netherlands) on the subjects and used a conductive electrode gel (Electro-gel; Electro-Cap International; Eaton, OH, USA) to obtain impedances below 5 kX. Recordings were generated from the classic 10/20 system (Oostenveld and Praamstra, 2001) augmented by midline and peri-midline electrodes along the parietal to frontal zones (Fig. 1A). The task was to maintain quiet stance for three 60-s trials in each of six conditions: wide stance on a firm surface with eyes open (WEO) or closed (WEC), narrow stance on a firm surface with eyes open (NEO) or closed (NEC), and narrow stance on a foam surface (4inches, medium-density, T-Foam, Alimed Inc., Dedham, MA, USA) with eyes open (FEO) or closed (FEC). Standing on foam was in narrow stance with the intention of providing a progressively difficult standing condition from the conditions on a firm surface. Subjects were instructed to stand with their bare feet as close together as possible without any contact occurring between them for narrow-stance conditions and with the lateral edges of their feet placed along marked lines of a measured shoulder width apart for the wide-stance conditions. In all conditions, subjects were instructed to stand with arms relaxed at the sides, face relaxed, and head oriented straight ahead. When standing with eyes open, subjects were requested to fix their gaze at an eye-level target and to minimize blinking. In all conditions, subjects stood with each foot on separate force plates (AMTI OR6-6; Watertown, MA, USA) that were used to generate COP measures. The order of the six stance conditions was randomized using a random number generator for each of the three repetition blocks. Subjects wore a gait belt and were guarded by an experimenter located just behind the subject in order to prevent falls to the ground. No contact was made between the subjects and the assistants unless the center of mass appeared to sway far enough relative to the base of support that the subject would have fallen without assistance. No assistance was required for this cohort. Subjects were monitored for self-reported fatigue
3
and were allowed breaks throughout the study to prevent fatigue. Data collection and processing EMG and force-plate signals were collected at 2000 Hz and were synchronized through Nexus recording software (VICON, Centennial, CO, USA). The EMG signals were pre-amplified at the skin’s surface by 1000. The EMG signals were then filtered from 1 to 100 Hz, and the COP signals were low-pass filtered to 100 Hz using a 4th-order Butterworth filter. EEG data were collected at 1024 Hz using a DC amplifier and preprocessed using ASA software version 4.7.3 (Advanced Neuro Technology, Enschede, the Netherlands). An external trigger signal was used to synchronize the collection of EMG, force-plate, and EEG signals. Following collection, blinking artifacts were removed using the principal components analysis artifact correction parameter, selecting the two components that most represented the artifacts’ characteristics. Next, event markers were identified to splice continuously recorded EEG data into epochs of 70 s, defined from 5 s before the external trigger signal that represented the start of the trial to 5 s after the 60-s trial. Finally, the signals were band-pass filtered from 1 to 60 Hz. The remainder of the data processing was completed using customized Matlab code (The Mathworks Inc., Natick, MA, USA). EMG and COP data were resampled to 1024 Hz and aligned with the EEG data at the onset of the external trigger signal. In preparation for calculating CMC, the EEG data from each electrode were re-referenced to the common average value of all electrodes in order to improve the signal-to-noise ratio of coherence measures (Mima and Hallett, 1999b). The Matlab coherence function was used to compute CMC between the filtered EMG of each muscle and the reference-adjusted EEG of electrode Cz as well as contralateral electrodes CCP1h, CP1, C1, FC1, and FCC1h for the right TA and GL muscles, or C2, CCP2h, CP2, FC2, and FCC2h for the left TA and GL muscles (Fig. 1E, F) with a Hanning window of one second and zero overlap. This grouping of electrodes was chosen because peri-midline electrodes that overlie the sensorimotor cortex have been shown to exhibit significant beta CMC with distal leg muscles, particularly when the contraction is isotonic rather than isometric (Gwin and Ferris, 2012). CMC values are unitless between zero and one, with a value of one indicating a perfect relationship in time varying spectral power of EEG and EMG signals, and a value of zero indicating independence. A threshold of significant CMC was determined by the 95% confidence limit = 1 (0.05)1/n 1, where n is the number of windows used for spectral estimation (Halliday et al., 1995). Once calculated, the CMC was plotted and the maximum peak in the beta (13–30 Hz) frequency band was manually selected with an interactive graphing function in Matlab. The beta frequencies were selected because this frequency band is known to (1) arise from the sensory-motor cortex (2) be
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
4
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
Fig. 1. (A) Electroencephalographic (EEG) recording montage and electrodes examined for CMC; electrodes with gray fill represent those that were examined for CMC. An example trial from one subject in the NEO condition illustrates (B) electromyographic (EMG) traces from each muscle; (C) center of pressure (COP) traces under each foot in the anterior-posterior (black traces) and medial–lateral (gray traces) directions, with positive values indicating forward or rightward displacements; (D) EEG traces from all the electrodes examined for cortico-muscular coherence (CMC); (E) CMC spectrum lines from all 6 contralateral or midline EEG electrodes with each muscle across all frequency bands (gray shaded region represents the beta band, gray dashed line represents the significance threshold); (F) a magnified CMC line from (E), identifying the area above the threshold for significant CMC (referred to as CMC magnitude), the beta frequency at which peak CMC is located, and the peak amplitude of CMC.
relevant to sub-maximal muscle contraction, and (3) be appropriate for analysis using surface EEG signals (Grosse et al., 2002). Once the peak was selected, the true peak was automatically identified within the close vicinity of the user-selected peak by the Matlab software. The magnitude of significant CMC was then determined as the integrated area of the CMC spectrum around the selected peak above the significance threshold (Fig. 1F).
Trials without significant CMC were valued at zero. The electrode that exhibited the maximum CMC magnitude was then identified for each individual and trial, and each trial’s CMC magnitude from the electrode of maximum magnitude was averaged by condition for each subject for analysis. The location of maximum CMC magnitude was selected to enhance sensitivity to CMC as well as to account for inter-individual and task-related differences
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
in functional representation of the sensory-motor cortex (Kurtzer et al., 2005; Scott, 2008; Koessler et al., 2009). The root-mean-square, path length, range, and mean velocity of COP displacements in the anterior–posterior (AP) and medial–lateral (ML) directions were also averaged by condition for each subject for analysis. Statistical analysis Descriptive statistics of frequency histograms determined the incidence of subjects with significant CMC per condition and muscle in order to first establish that beta CMC exists; for comparison, these frequencies were determined as the number of subjects with an average area of CMC spectrum above the significance threshold as well as with an average peak value of CMC above the significance threshold. Frequency histograms were also generated to identify the frequency at which each electrode location represented the location of maximum CMC magnitude for each condition and muscle. The functional relevance of beta CMC was then evaluated by identifying whether it changes significantly due to stance width, vision, or support surface. Due to the sample size and a common failure to exhibit normal distributions, as assessed by Shapiro–Wilk tests, Wilcoxon Signed-Ranks Tests were used to generate a priori comparisons of the WEO and NEO conditions in order to assess effects of stance width on CMC, the WEO and WEC conditions in order to assess effects of vision on CMC, and the NEO and FEO conditions in order to assess the effects of support surface on CMC. Other combinations of conditions reflecting effects of stance width, vision or surface (including NEO and NEC, FEO and FEC, WEC and NEC, as well as NEC and FEC) were also examined as a post hoc exploratory analysis. Friedman’s ANOVAs were not used across all conditions because we did not have an a priori interest in all potential combinations of condition comparisons. In order to further evaluate the functional relevance of beta CMC during stance, we generated an exploratory analysis of Spearman’s rho correlation coefficients between CMC magnitudes and four measures of whole-body COP sway (maximum range, mean velocity, total displacement, and root-mean-square amplitude). Lastly, in order to determine whether the right and left limbs operated synchronously or with some independence, a post hoc evaluation of cross-correlation coefficients between left and right COP displacements was computed with a Matlab function. Significance was assumed at an alpha level of 0.05, and all analyses were performed in SPSS version 21 software (IBM, Armonk, NY, USA).
RESULTS Fig. 1B–D illustrates time trajectories of EMG from four leg muscles, COP under each foot, and EEG from 11 electrodes. It was evident that both anterior and posterior leg muscles are activated during the 60 s of stance, and that these muscle activations are synchronized with the COP sway (forward COP displacement coincided with activation of the GL and/or inhibition of the TA, and backward COP displacement coincided with activation of
5
the TA and/or inhibition of the GL). In addition, COP displacements under the left and right foot were not completely synchronized, suggesting an asymmetric control of standing posture from each side of the body. Further, the CMC peaks were heterogeneous across electrodes and observed in alpha, beta, and gamma ranges (Fig. 1E). Again, however, this study focused on the beta band because it is known to (1) arise from the sensory-motor cortex (2) be relevant to sub-maximal muscle contraction, and (3) be appropriate for analysis using surface EEG signals (Grosse et al., 2002). Existence of significant beta CMC during human standing balance When evaluating the existence of significant beta CMC using the area of CMC spectrum above the significance threshold (Fig. 2), all 10 subjects exhibited significant beta CMC magnitudes in 20 of the 24 muscle-condition combinations. Significant CMC magnitudes were evident in nine of 10 subjects for the remaining four musclecondition combinations. Three separate subjects were responsible for these four instances of dropout. Although the incidence of significant beta CMC was consistent using the average area above the significance threshold, peak amplitudes of beta CMC were sometimes low and the incidence of significant peak amplitudes of beta CMC was lower than that of the area of CMC spectrum above significance (Fig. 2). The difference in incidence reflects inter-trial variability within subjects and conditions, such that some trials did not reach the threshold of significant beta CMC. Therefore, when averaging peak amplitudes, even when one or two trials exhibit amplitudes above the significance threshold, if at least one trial exhibited a peak amplitude far enough below significance to decrease the average value below the threshold for significance, then no significant beta CMC would be identified in that average despite some trials exhibiting significant CMC. In contrast, when averaging the area above the significance threshold, any trials without CMC values that cross the threshold for significance would be counted as a value of zero; thus, if any one trial exhibited beta CMC above the significance threshold, the average value across trials would also be non-zero, thereby indicating evidence of significant beta CMC. The group’s mean peak frequency of CMC ranged between 20.5 and 23.9 Hz across all muscle-condition combinations. Wilcoxon signed ranks tests revealed no significant differences due to condition except for an effect of support surface, in which the peak frequency was lower in the FEO condition than in the NEO condition: mean (95% confidence interval) was 20.5 (19.1–21.9) Hz in the FEO condition and 23.8 (22.2–25.5) Hz in the NEO condition (Wilcoxon Z = 2.19; P = 0.027). Location of EEG electrodes with maximum CMC magnitudes The EEG electrode location of maximum CMC magnitudes was consistent across trials within subjects and within each condition, but the location of maximum
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
6
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
Fig. 2. Frequency histograms of the number of subjects exhibiting significant beta CMC for each condition and muscle as assessed by either a nonzero average of the area of the CMC spectrum above the significance threshold (filled bars) or an average peak amplitude of CMC above the significance threshold (unfilled bars). WEO, wide, eyes open; WEC, wide, eyes closed; NEO, narrow, eyes open; NEC, narrow, eyes closed; FEO, foam, eyes open; FEC, foam, eyes closed.
CMC magnitude was not consistent across conditions or among different subjects (Fig. 3). The vertex (Cz) electrode that was the focus of past studies of beta CMC during human stance (Masakado et al., 2008; Murnaghan et al., 2014) was the most common (or tied as the most common) location of maximum CMC in seven of the 24 muscle-condition combinations. The FC1/FC2 electrodes were most common in nine muscle-condition combinations; FCC1h/FCC2h, six combinations; C1/C2, four combinations; CCP1h/CCP2h, four combinations; and CP1/CP2, two combinations. Functional relevance of beta CMC during human stance: condition effects Fig. 4 illustrates the individual means and group means of beta CMC magnitudes (as assessed by the area of CMC spectrum above the significance threshold), contrasting the a priori condition comparisons of WEO to NEO, WEO to WEC, and NEO to FEO for the left and right TA and GL muscles. Beta CMC magnitudes were significantly larger in the WEO condition than in the NEO condition for the right GL muscle. It is worthy to note that this effect remained significant even after a very conservative Bonferroni correction of 12 musclecondition combinations. No other significant effects of condition were evident in the a priori comparisons or in the exploratory comparisons of other condition pairs. Functional relevance of beta CMC during human stance: correlations with sway and sway measures With four COP variables each assessed in the ML and AP directions, and CMC from four muscles each in six
conditions, a total of 192 correlations were generated in this exploratory analysis. Seven significant correlations were evident from these 192 tests (Table 1), such that only the root-mean-square and range of COP sway elicited significant correlations with beta CMC. The significant correlations varied with regard to the direction of correlation, the condition in which the correlation was evident, and the muscle of CMC with which the correlation was evident. Table 2 presents the mean values of the root-meansquare and range of COP sway for the WEO, NEO, NEC, and FEO conditions. Both the AP root-mean square and range of COP sway were significantly different between the WEO and NEO conditions (Wilcoxon Z = 2.50, 2.29; P = 0.010, 0.020) and between the NEO and FEO conditions (Wilcoxon Z = 2.40, 2.80; P = 0.014, 0.002), but not between the NEO and NEC conditions (Wilcoxon Z = 1.19, 0.97; P = 0.27, 0.37). The ML range of COP sway was significantly different between the WEO and NEO conditions (Wilcoxon Z = 2.80; P = 0.002), the NEO and FEO conditions (Wilcoxon Z = 2.80; P = 0.002), as well as the NEO and NEC conditions (Wilcoxon Z = 2.80; P = 0.002). The ML root-meansquare of COP sway was significantly different between the WEO and NEO conditions (Wilcoxon Z = 2.80; P = 0.002), but not the NEO and FEO conditions (Wilcoxon Z = 1.58; P = 0.13) or the NEO and NEC conditions (Wilcoxon Z = 0.05; P = 1.0). Because CMC magnitudes exhibited different values between muscles, and because CMC was responsive to a change in stance width at only the right limb, we performed a post hoc analysis of cross-correlation between the COP displacements of the right vs left limb
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
7
Fig. 3. Frequency histograms of the number of subjects for whom each electrode represented the electrode of maximum CMC for each muscle and condition.
(Table 3). The results exception of standing on did not approach a value unique behavior of the left
demonstrate that, with the foam, correlation coefficients of 1.0, thus demonstrating a vs right limb.
DISCUSSION The results supported our hypothesis that beta CMC exists during human standing balance and partially
supported the hypothesis that beta CMC is functionally relevant to changes in stance condition and to postural sway during stance. In specific, our methods of processing elicited significant CMC from all subjects and all recorded muscles in most of the tested conditions, at least demonstrating significant CMC from nine of 10 subjects in all conditions and muscles. Beta CMC was consistently sensitive to changes in biomechanical orientation, such that beta CMC was higher when
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
8
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
Fig. 4. Individual (gray lines) and group (black circles) mean CMC magnitudes for each muscle. CMC magnitudes represent the integrated area of the CMC spectrum above the significance threshold. Rows specify muscles. Columns specify condition comparisons for effects of stance width, vision, and support surface. Wilcoxon signed rank test statistics are provided for each comparison. Error bars on the group means represent 95% confidence intervals.
standing in a wide stance width compared to a narrow stance width for the right GL muscle. No consistent effect on beta CMC magnitude was evident, however, when transitioning from eyes-open to eyes-closed conditions, or when transitioning from a firm to foam surface. Significant correlations with measures of COP
sway were evident, although the direction of correlations differed among muscle-condition combinations. Our ability to demonstrate significant beta CMC in contrast to previous studies (Masakado et al., 2008; Murnaghan et al., 2014) likely reflects differences in processing methods. Our data demonstrate that it may be
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
9
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11 Table 1. Significant correlations between beta CMC magnitudes and measures of COP sway Stance condition
Muscle of CMC
Measure of COP sway
Spearman’s Rho (P-value)
Wide, Eyes open Narrow, Eyes open Narrow, Eyes open Narrow, Eyes closed Narrow, Eyes closed Foam, Eyes closed Foam, Eyes closed
Right Tibialis anterior Left Gastrocnemius lateralis Left Gastrocnemius lateralis Right Gastrocnemius lateralis Left Tibialis anterior Right Tibialis anterior Right Tibialis anterior
Medial–lateral Root-mean-square Anterior–posterior Root-mean-square Medial–lateral Root-mean-square Medial–lateral Root-mean-square Medial–lateral Range of displacement Anterior–posterior Range of displacement Medial–lateral Range of displacement
0.709 (0.022) 0.661 (0.038) +0.636 (0.048) 0.891 (0.00054) 0.685 (0.029) +0.685 (0.029) +0.709 (0.022)
Table 2. Anterior–posterior (AP) and medial–lateral (ML) COP sway by condition COP measure
Condition
Mean (95% CI) value
AP root-mean-square COP (% foot length)
WEO NEO NEC FEO WEO NEO NEC FEO WEO NEO
14.9 (8.2–21.7) 19.8 (14.7–24.9) 18.2 (15.0–21.4) 15.4 (11.5–19.3) 11.5 (9.2–13.8) 13.4 (11.1–15.7) 14.3 (12.0–16.6) 22.7 (18.9–26.6) 62.9 (58.1–67.6) 119.0 (109.9– 128.1) 119.1 (109.8– 128.5) 118.2 (108.7– 127.8) 3.0 (2.6–3.4) 12.9 (10.7–15.1) 16.0 (13.7–18.3) 17.8 (14.4–21.2)
AP range of COP displacement (% foot length) ML root-mean-square COP (% stance width)
NEC FEO ML range of COP displacement (% stance width)
WEO NEO NEC FEO
important to account for inter-individual Koessler et al., 2009) and inter-condition (Kurtzer et al., 2005; Scott, 2008) differences by analyzing the electrode of maximal CMC for each subject and by evaluating multiple electrodes along and adjacent to the midline over sensorymotor cortex. In addition, it was clear from our analysis that inter-trial variability in beta CMC renders it important to evaluate multiple trials. Sensitivity for identifying significant beta CMC is also enhanced by evaluating the area above the threshold of significant CMC rather than the peak amplitude of CMC. We also re-referenced our data to a common average reference in order to enhance signal-to-noise ratio, as previously suggested (Mima and Hallett, 1999b). In our attempt to understand the functional significance of beta CMC, we did identify that beta CMC modulates with stance width at the right GL muscle. The decrease in beta CMC from wide stance to narrow
Table 3. Coefficients from Cross-correlations of COP under left vs right limb Condition
WEO NEO NEC FEO
Mean (95% CI) correlation coefficient Anterior–posterior COP
Medial–lateral COP
0.77 0.63 0.55 0.92
0.52 (0.32–0.72) 0.20 ( 0.44–0.03) 0.17 ( 0.34–0.00) 0.70 (0.48–0.92)
(0.73–0.82) (0.50–0.77) (0.42–0.67) (0.89–0.95)
stance at the GL muscle may reflect a transition to a sway-modulated isotonic contraction from an isometric contraction afforded by the mechanical stability of a wide stance. Indeed, more sway was evident in the narrow-stance conditions than in the wide-stance condition. Studies have demonstrated that beta CMC is most evident in isometric conditions (Gwin and Ferris, 2012; Ushiyama et al., 2012). The unilateral effect on only the GL muscle could reflect a combination of the plantarflexors’ role in standing postural sway and limb asymmetries during standing balance. Our data demonstrate that COP displacements under each limb were not completely symmetric. Given these limb asymmetries, it would therefore be expected that afferent feedback about limb status or descending control of a limb’s function would also differ, which would subsequently generate different CMC responses. The specificity of effect of stance width on the GL muscle and not the TA muscle could reflect that the GL muscle is more active to modify postural sway than the TA muscle. Whereas the GL muscle has been found to associate with postural sway with a leading phase shift relative to the sway signal (Gatev et al., 1999; Masani et al., 2003), the TA muscle’s influence on sway may be more transient to correct larger backward sway displacements and otherwise acts as a more passive source of afferent feedback regarding joint position during postural sway (Basmajian and De Luca, 1985; Day et al., 2013). The GL muscle’s responsiveness to changes in stance width, thus, reflects the GL muscle’s influence on postural sway and suggests that either cortical monitoring of
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
10
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11
afferent input from the GL muscle’s activation or control by the cortex on the GL becomes up-regulated in the beta frequency band when in a wide stance. Lack of effect of changes in vision or surface on beta CMC was not expected. Previous studies have identified that cortical lesions affect postural sway under changing sensory conditions (Bonan et al., 2004), and changes in the availability of vision alter activity of the parietal and prefrontal cortex in healthy subjects (Ouchi et al., 1999; Del Percio et al., 2007). Thus, we anticipated that this cortical role in sensory integration during standing balance would elicit altered communication between the sensory-motor cortex and the muscle in order to modulate postural sway. Lack of effect of vision or surface on CMC in this study, however, is consistent with another study using transcranial magnetic stimulation that demonstrated a lack of change in corticospinal excitability between standing conditions with or without vision, or when standing on a firm vs foam surface (Baudry et al., 2014). Thus, the parieto-prefrontal changes that occur when standing with vs without vision must mediate changes in postural sway via extra-pyramidal pathways not involving the sensory-motor cortex. Although this study provides the initial evidence that beta CMC exists during human standing balance and provides insight into the responsiveness of corticomuscular communication to changes in mechanics and sensory conditions, certain methodological considerations warrant further study. Our study evaluated only the TA and GL muscles, but other muscles, such as the soleus and more proximal muscles, are known to modulate postural sway as well. The relative influence of proximal musculature might be even more relevant to ML sway or to hip strategies adopted during more difficult standing conditions such as when on a foam surface. Communication of these other muscles with the sensory-motor cortex during these difficult standing conditions is worth further investigation in future studies. In addition, although we identified significant correlations between postural sway and beta CMC magnitudes, these correlations did not demonstrate a clear pattern with regard to muscle, condition, direction of sway, or direction of correlation. Due to the large number of correlations generated by our analysis, it remains unclear whether these relationships are true and reflective of complex interactions, or if they are spurious findings amid multiple comparisons derived from an exploratory analysis. Lastly, functional associations between CMC and sway dynamics may be more apparent in other frequency bands; previous studies have demonstrated the alpha and gamma bands are sensitive to detecting unstable stance and can be sensitive to changes in sensory conditions or be discriminative between groups with different athletic backgrounds (Vecchio et al., 2008; Slobounov et al., 2009). Additional studies with larger sample sizes are, thus, needed in order to evaluate interactions among many potential contributing factors as well as to better understand whether corticomuscular communication with other muscles and at other frequencies functionally modulates postural sway.
CONCLUSIONS This study demonstrates that, contrary to past literature, significant beta CMC can be consistently detected across subjects after accounting for important methodological considerations of EEG referencing, selected electrodes used for analysis, and choice in outcome measure. For the healthy young adults of this study, corticomuscular communication with the sensorymotor cortex appears responsive to changes in mechanical orientation during standing balance, whereas pathways that do not directly involve the sensory-motor cortex appear to mediate changes in muscle activation due to changes in visual or surface conditions. Although this study focuses on the functional relevance of beta CMC during standing balance for healthy young adults, the functional relevance of corticomuscular coupling may differ for older populations or for those with health conditions (Grosse et al., 2002). Thus, while we have confirmed the existence of beta CMC during healthy human stance with a novel insight into the functional responsiveness of beta CMC to changes in stance conditions, its relevance to impaired standing balance should still be examined. Acknowledgments—A Research Incentives Grant from the University of Vermont’s College of Nursing and Health Sciences funded the study. We thank Sarah Aghjayan, Juvena Hitt, Elena Isaacson, Anne Patti, Julia Parks, Claire Purcell, and Xi Wen for assistance with data collection and processing.
REFERENCES Basmajian JV, De Luca CJ (1985) Muscles alive. 5th ed. Baltimore, MD: Williams and Wilkins. Baudry S, Penzer F, Duchateau J (2014) Vision and proprioception do not influence the excitability of the corticomotoneuronal pathway during upright standing in young and elderly adults. Neuroscience 268:247–254. Bonan IV, Colle FM, Guichard JP, Vicaut E, Eisenfisz M, Tran Ba Huy P, Yelnik AP (2004) Reliance on visual information after stroke. Part I: Balance on dynamic posturography. Arch Phys Med Rehabil 85:268–273. Campfens SF, Schouten AC, van Putten MJ, van der Kooij H (2013) Quantifying connectivity via efferent and afferent pathways in motor control using coherence measures and joint position perturbations. Exp Brain Res 228:141–153. Daikuya S, Tanino Y, Nishimori T, Takasaki K, Suzuki T (2003) The silent period from soleus and gastrocnemius muscles in relation to conditions of standing. Electromyogr Clin Neurophysiol 43:217–222. Day JT, Lichtwark GA, Cresswell AG (2013) Tibialis anterior muscle fascicle dynamics adequately represent postural sway during standing balance. J Appl Physiol 115:1742–1750. Del Percio C, Brancucci A, Bergami F, Marzano N, Fiore A, Di Ciolo E, Aschieri P, Lino A, Vecchio F, Iacoboni M, Gallamini M, Babiloni C, Eusebi F (2007) Cortical alpha rhythms are correlated with body sway during quiet open-eyes standing in athletes: a high-resolution EEG study. NeuroImage 36:822–829. Gatev P, Thomas S, Kepple T, Hallett M (1999) Feedforward ankle strategy of balance during quiet stance in adults. J Physiol 514:915–928. Grosse P, Cassidy MJ, Brown P (2002) EEG–EMG, MEG–EMG and EMG–EMG frequency analysis: physiological principles and clinical applications. Clin Neurophysiol 113:1523–1531.
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
J. V. Jacobs et al. / Neuroscience 298 (2015) 1–11 Gwin JT, Ferris DP (2012) Beta- and gamma-range human lower limb corticomuscular coherence. Front Hum Neurosci 6:258. Halliday DM, Rosenberg JR, Amjad AM, Breeze P, Conway BA, Farmer SF (1995) A framework for the analysis of mixed time series/point process data–theory and application to the study of physiological tremor, single motor unit discharges and electromyograms. Prog Biophys Mol Biol 64:237–278. Honeycutt CF, Gottschall JS, Nichols TR (2009) Electromyographic responses from the hindlimb muscles of the decerebrate cat to horizontal support surface perturbations. J Neurophysiol 101:2751–2761. Koessler L, Maillard L, Benhadid A, Vignal JP, Felblinger J, Vespignani H, Braun M (2009) Automated cortical projection of EEG sensors: anatomical correlation via the international 10–10 system. NeuroImage 46:64–72. Kurtzer I, Herter TM, Scott SH (2005) Random change in cortical load representation suggests distinct control of posture and movement. Nat Neurosci 8:498–504. Masakado Y, Ushiba J, Tsutsumi N, Takahashi Y, Tomita Y, Kimura A, Liu M (2008) EEG–EMG coherence changes in postural tasks. Electromyogr Clin Neurophysiol 48:27–33. Masani K, Popovic MR, Nakazawa K, Kouzaki M, Nozaki D (2003) Importance of body sway velocity information in controlling ankle extensor activities during quiet stance. J Neurophysiol 90:3774–3782. Mima T, Hallett M (1999a) Corticomuscular coherence: a review. J Clin Neurophysiol 16:501–511. Mima T, Hallett M (1999b) Electroencephalographic analysis of cortico-muscular coherence: reference effect, volume conduction and generator mechanism. Clin Neurophysiol 110:1892–1899. Mima T, Matsuoka T, Hallett M (2000) Functional coupling of human right and left cortical motor areas demonstrated with partial coherence analysis. Neurosci Lett 287:93–96. Mori S (1987) Integration of posture and locomotion in acute decerebrate cats and in awake, freely moving cats. Prog Neurobiol 28:161–195. Murnaghan CD, Squair JW, Chua R, Inglis JT, Carpenter MG (2014) Cortical contributions to control of posture during unrestricted and restricted stance. J Neurophysiol 111:1920–1926. Musienko PE, Zelenin PV, Lyalka VF, Orlovsky GN, Deliagina TG (2008) Postural performance in decerebrated rabbit. Behav Brain Res 190:124–134. Obata H, Sekiguchi H, Nakazawa K, Ohtsuki T (2009) Enhanced excitability of the corticospinal pathway of the ankle extensor and flexor muscles during standing in humans. Exp Brain Res 197:207–213.
11
Oostenveld R, Praamstra P (2001) The five percent electrode system for high-resolution EEG and ERP measurements. Clin Neurophysiol 112:713–719. Ouchi Y, Okada H, Yoshikawa E, Nobezawa S, Futatsubashi M (1999) Brain activation during maintenance of standing postures in humans. Brain 122(Pt. 2):329–338. Pe´rennou DA, Leblond C, Amblard B, Micallef JP, Rouget E, Pe´lissier J (2000) The polymodal sensory cortex is crucial for controlling lateral postural stability: evidence from stroke patients. Brain Res Bull 53:359–365. Scott SH (2008) Inconvenient truths about neural processing in primary motor cortex. J Physiol 586:1217–1224. Skoyles (2006) Human balance, the evolution of bipedalism and disequilibrium syndrome. Med Hypotheses 66:1060–1068. Slobounov S, Cao C, Jaiswal N, Newell KM (2009) Neural basis of postural instability identified by VTC and EEG. Exp Brain Res 199:1–16. Thibault RT, Lifshitz M, Jones JM, Raz A (2014) Posture alters human resting-state. Cortex 58:199–205. Tokuno CD, Taube W, Cresswell AG (2009) An enhanced level of motor cortical excitability during the control of human standing. Acta Physiol (Oxf) 195:385–395. Ushiyama J, Takahashi Y, Ushiba J (2010) Muscle dependency of corticomuscular coherence in upper and lower limb muscles and training-related alterations in ballet dancers and weightlifters. J Appl Physiol (1985) 109:1086–1095. Ushiyama J, Suzuki T, Masakado Y, Hase K, Kimura A, Liu M, Ushiba J (2011) Between-subject variance in the magnitude of corticomuscular coherence during tonic isometric contraction of the tibialis anterior muscle in healthy young adults. J Neurophysiol 106:1379–1388. Ushiyama J, Masakado Y, Fujiwara T, Tsuji T, Hase K, Kimura A, Liu M, Ushiba J (2012) Contraction level-related modulation of corticomuscular coherence differs between the tibialis anterior and soleus muscles in humans. J Appl Physiol (1985) 112:1258–1267. Varghese JP, Beyer KB, Williams L, Miyasike-daSilva V, McIlroy WE (2015) Standing still: Is there a role for the cortex? Neurosci Lett 590:18–23. Vecchio F, Del Percio C, Marzano N, Fiore A, Toran G, Aschieri P, Gallamini M, Cabras J, Rossini PM, Babiloni C, Eusebi F (2008) Functional cortico-muscular coupling during upright standing in athletes and nonathletes: a coherence electroencephalographic– electromyographic study. Behav Neurosci 122:917–927. Witham CL, Riddle CN, Baker MR, Baker SN (2011) Contributions of descending and ascending pathways to corticomuscular coherence in humans. J Physiol 589:3789–3800.
(Accepted 6 April 2015) (Available online 11 April 2015)
Please cite this article in press as: Jacobs JV et al. Evidence for beta corticomuscular coherence during human standing balance: Effects of stance width, vision, and support surface. Neuroscience (2015), http://dx.doi.org/10.1016/j.neuroscience.2015.04.009
