Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jul 27.
Published in final edited form as: Neurosci Lett. 2020 May 24;732:135090. doi: 10.1016/j.neulet.2020.135090

Gamma Somatosensory Cortical Oscillations are attenuated during the Stance Phase of Human Walking

Sarah Baker 1,2, Mike Trevarrow 1,2, James Gehringer 1, Hannah Bergwell 1,2, David Arpin 1, Elizabeth Heinrichs-Graham 2,3, Tony W Wilson 2,3, Max J Kurz 1,2
PMCID: PMC7336339  NIHMSID: NIHMS1601597  PMID: 32461106

Abstract

It is well appreciated that processing of peripheral feedback by the somatosensory cortices plays a prominent role in the control of human motor actions like walking. However, very few studies have actually quantified the somatosensory cortical activity during walking. In this investigation, we used electroencephalography (EEG) and beamforming source reconstruction methods to quantify the frequency specific neural oscillations that are induced by an electrical stimulation that is applied to the right tibial nerve under the following experimental conditions: 1) sitting, 2) standing in place, and 3) treadmill walking. Our experimental results revealed that the peripheral stimulation induced a transient increase in theta-alpha (4–12 Hz; 50–350 ms) and gamma (40–80 Hz; 40–100 ms) activity in the leg region of the contralateral somatosensory cortices. The strength of the gamma oscillations were similar while sitting and standing, but were markedly attenuated while walking. Conversely, the strength of the theta-alpha oscillations were not different across the respective experimental conditions. Prior research suggests the afferent feedback from the Ia sensory fibers are likely attenuated during walking, while afferent feedback from the Aβ polysynaptic sensory fibers are not. We suggest that the attenuated gamma oscillations seen during walking reflect the gating of the Ia afferents, while the similarity of theta-alpha oscillations across the experimental conditions is associated with the afferent information from the type II (Aα and β polysynaptic sensory fibers.

Keywords: Electroencephalography, Gait, Sensory, Cortical Oscillations

Introduction

The seminal investigation by Penfield and Boldrey (1937) identified that the post-central gyrus is involved in the processing of afferent sensory information, and has a topological organization that is representative of the different areas of the body (Penfield and Boldrey, 1937). Numerous fMRI, electroencephalographic (EEG) and magnetoencephalographic (MEG) follow-up studies have verified these initial findings through paradigms that employ tactile stimulation of the skin or electrical stimulation of the peripheral nerves (Del Gratta et al., 2002; Hamalainen et al., 1990; Kakigi et al., 2000; Nakamura et al., 1998). Although these neuroimaging experiments have been pivotal for refining our understanding of the topological organization of the somatosensory cortices, more recent experiments have been focused on identifying how spectrally-specific changes in somatosensory cortical activity reflect different neural computations. It is now well recognized that peripheral stimulation applied to the foot and hand produces an immediate and transient increase in the somatosensory cortical oscillations across the 10–75 Hz frequency bands (Dockstader et al., 2008; Gaetz and Cheyne, 2003; Kurz et al., 2018a; Spooner et al., 2018; Wiesman et al., 2017), and that these neural oscillations are often followed by a decrease in strength of the oscillations seen across the alpha (8–16 Hz) and beta (18–26 Hz) frequency bands at later latencies (e.g., 150–400 ms). Furthermore, it has been shown that the strength of these spectrally-specific neural oscillations are attenuated if an identical peripheral stimulation is applied within a specific time window relative to the first stimulation (Kurz et al., 2018a; Spooner et al., 2018; Wiesman et al., 2017).

Several EEG and MEG studies have extended these initial findings by showing that the somatosensory cortical activity is attenuated during voluntary movements (Arpin et al., 2018; Avikainen et al., 2002; Kristeva-Feige et al., 1996; Kurz et al., 2018b; Macerollo et al., 2016; Staines et al., 2000). However, very few efforts have been made to evaluate the somatosensory cortical activity during walking. The few studies that have been conducted have evaluated differences in the electrode-level evoked-potentials as participants stand or walk on a treadmill (Altenmuller et al., 1995; Dietz et al., 1985). The outcomes from these investigations have shown that compared with standing, the strength of the early (N80) and late (P220) somatosensory evoked-potentials are markedly attenuated during walking. These seminal studies have also identified that the somatosensory evoked-potentials are more attenuated during the stance phase when the foot is in contact with the ground compared with the evoked-potentials seen during the swing phase when the foot is not in contact with the ground. Together, these results imply that the polysynaptic spinal pathways that involve the integration of afferent sensory information from the Aα and Aβ foot mechanoreceptors alter the intensity of the evoked somatosensory cortical responses seen during walking. Despite these insights, evoked-potentials only capture the phase-locked component of the neural signal and thus cortical oscillations seen during walking remain poorly understood. Information about the underlying oscillatory responses may provide new and pivotal data on the somatosensory neural computations that underlie the motor control of walking. Furthermore, such new insights might provide a spring-board for advancing our understanding of how the somatosensory neural computations impact the standing postural control strategies and walking dynamics seen in various patient populations.

In the present study, we aimed to further advance our understanding of the somatosensory cortical oscillations associated with walking. We used EEG and beamformer source reconstruction methods to identify how the sensorimotor cortical oscillations are altered under the following experimental conditions: 1) sitting, 2) standing in place, and 3) during the stance phase of walking. Based on the prior sensor level experimental work by Altenmuller and colleagues (1995), we hypothesized that the somatosensory cortical oscillations would be the strongest for the sitting condition, followed by standing and walking.

Methods and Materials

Twenty-five healthy adults (Age = 25.9 ± 4.4 yrs.; Females = 12) with no neurological or musculoskeletal impairments were recruited to participate in this experimental investigation. The Institutional Review Board (IRB) at the University of Nebraska Medical Center reviewed and approved this research study, and informed consent was acquired from all participants.

A portable EEG system (eego sport, ANT Neuro, Netherlands) with 64 Ag/AgCl electrodes arranged in a 10/20 configuration (Wavegaurd, ANT Neuro, Netherlands) was used to collect the electrocortical signals at 1024 Hz. Electrode gel was used to maintain the conductivity throughout the experiment, and the impedance levels were set below 20 kΩ. The participants completed the following experimental conditions: 1) sitting in a chair with the feet off the floor with the distal portion of the legs resting on a bench, 2) standing in place, and 3) walking on a treadmill. Across all three conditions, unilateral electrical stimulation was applied to the tibial nerve posterior to the medial malleolus of the right foot (DS7AH, Digitimer, Fort Lauderdale, FL). The stimulator consisted of two electrodes that were spaced 50.8 mm apart. We only stimulated the right leg to avoid any potential hemispheric differences since prior research has noted that handedness can affect the strength of the cortical activity (Jung et al., 2003). The electrical stimulation consisted of a 0.2 ms constant-current current square wave that was set to elicit a subtle but visible flexor twitch in the hallux. The average amplitude of the electrical stimulation across participants was 25.8 ± 3 μA and was held constant across the three conditions.

During the sitting condition, 150 electrical stimulations were applied with a 2000 ms inter-stimulus interval (ISI). The participants were instructed to fix their vision on a red cross that was displayed on monitor that was positioned ~1 m away at eye-level. During the standing condition, the participants had their feet positioned shoulder width apart, and the same simulation paradigm that was used for the sitting condition was repeated (i.e., 150 electrical stimulations at an ISI of 2000 ms) as the participant viewed a cross that was displayed on the monitor. For the walking condition, the participants walked at 0.45 m/s (1.0 mph) on a treadmill (Woodway, Waukesha, WI, USA) for 10 minutes as 150 electrical stimulations were once again applied. A footswitch system that was placed underneath the right heel was used to trigger a single-pulse electrical stimulation that occurred 200 ms after heel-contact and was triggered to occur every other heel-contact. A 200 ms delay was used to avoid the mechanical induced noise that is known to be created by the foot colliding with the ground at heel-contact (Snyder et al., 2015). The inter-stimulus interval was approximately 3250 ms since the onset of the trigger was dependent upon the timing of the participant’s spatiotemporal walking biomechanics. Similar to the other experimental conditions, the participants were also instructed to look at a red cross on a monitor that was positioned ~1 m in front of them as they walked on the treadmill. The epochs for all of the conditions were defined offline and were of 1600 ms duration, including a pre-stimulation baseline of 250 ms.

Prior to the experiment, three fiducial points and the EEG electrodes were digitized to determine their three-dimensional position (Patriot, Polhemus Colchester, VT, USA). With this coordinate system, each participant’s EEG data were coregistered with a standard adult MRI based on the three external landmarks (i.e., fiducials), and the digitized EEG electrodes locations prior to source space analyses.

Artifact rejection was based on a fixed threshold method and was supplemented with visual inspection. The continuous time series were divided into epochs of 1600 ms duration, with the baseline being defined as −550 ms to −300 ms before the stimulus onset. The artifact-free epochs were transformed into the time-frequency domain using complex demodulation (Kovach and Gander, 2016), and the resulting spectral power estimations per EEG electrode were averaged over the respective trials to generate time-frequency plots of the mean spectral density. For the lower frequency range (5–35 Hz), a resolution of 2 Hz by 25 ms was used for the time-frequency transform, while for the higher frequencies (>35 Hz) we used 1 Hz by 50 ms. These electrode-level data were then normalized by dividing the power value of each time-frequency bin by the respective bin’s baseline power. The specific time-frequency windows used for imaging were determined by statistical analysis of the electrode-level spectrograms across the entire electrode net. Each data point in the spectrogram was initially evaluated using a mass univariate approach based on the general linear model. To reduce the risk of false positive results while maintaining reasonable sensitivity, a two-stage procedure was followed to control for Type 1 error. In the first stage, one-sample t-tests were conducted on each data point and the output spectrogram of t-values was thresholded at p < 0.05 to define time-frequency bins containing potentially significant oscillatory deviations across all conditions. In stage two, time-frequency bins that survived the threshold were clustered with temporally and/or spectrally neighboring bins that were also above the (p < 0.05) threshold and a cluster value was derived by summing all of the t-values of all data points in the cluster. Nonparametric permutation testing was then used to derive a distribution of cluster-values and the significance level of the observed clusters (from stage one) were tested directly using this distribution (Ernst, 2004; Maris and Oostenveld, 2007). 1,000 permutations were computed to build a distribution of cluster values.

The time-frequency windows that contained significant oscillatory events across all participants were subjected to the dynamic imaging of coherent sources (DICS) beamformer with a four-shell ellipsoidal head model to calculate the 3D images of the local power of neuronal current over the time-frequency ranges of interest (Gross et al., 2001). The single images were derived from the cross spectral densities of all combinations of EEG electrodes, and the solution of the forward problem for each location on a grid specified by input voxel space. Following convention, the source power in these images was normalized per subject using a separately averaged pre-stimulus noise period of equal duration and bandwidth (Hillebrand and Barnes, 2005; Hillebrand et al., 2005; Van Veen et al., 1997). Thus, the normalized power per voxel was computed over the entire brain volume per participant at 7 mm3 resolution. The resulting functional images, which were co-registered with a high-resolution adult MRI prior to beamforming, were transformed into standardized Talairach space. To evaluate the conditional differences, we extracted the neural time course (i.e., virtual sensor) of the peak somatosensory voxel across all three conditions, per participant, and these data were transformed into the time-frequency domain. Lastly, a one-way repeated measures ANOVA with Bonferroni corrected post-hoc t-tests was used to evaluate conditional differences in the average activity seen in the same time windows. All imaging procedures were done with the Brain Electrical Source Analysis (BESA) software (BESA v7.0; Grafelfing, Germany) and the statistical comparisons were performed with SPSS (IBM, Armonk, NY, USA).

Results

Time frequency analyses were conducted by collapsing the data across all conditions in order to statistically identify the time-frequency windows of interest that were common across all experimental conditions. This analysis revealed a prominent increase in the strength of the cortical oscillations that stretched across the 4–80 Hz frequency range that was initiated immediately after the stimulation. Permutation testing identified a significant increase in gamma range (40–80 Hz) oscillations within the 40–100 ms time window (Figure 1A), as well as an increase in the theta-alpha (4–12 Hz) range that occurred shortly after and was sustained within the 50–350 ms time window (Figure 1B). The topologic maps in Figure 1 show that the power increase seen in the gamma range primarily resided in electrodes located over the medial sensorimotor cortices, while the power increase seen in theta-alpha bad appeared in electrodes that spanned the premotor, sensorimotor and parietal areas. For illustrative purposes, we also show the individual time frequency components for the sitting, standing, and walking conditions (Figure 2). Visual inspection of these figures suggests that the gamma oscillations were notably attenuated during the walking condition. Yet, the theta-alpha oscillations appeared to be sustained across the respective conditions.

Figure 1.

Figure 1.

Open in a new tab

Time-frequency spectrograms from an electrode located near the sensorimotor strip (e.g., Cz). For visualization, the time-frequency spectrograms were averaged across participants and respective conditions. Time (in ms) is denoted on the x-axis, with 0 ms defined as the onset of the tibial nerve electrical stimulation, and the boxes represent the significant time-frequency windows identified via permutation testing. Spectral power is expressed as the percent difference from the baseline period, and the color scale bar is shown on the bottom. The topologic maps shown in the right panels have the same color scale as the time-frequency spectrograms. As shown in Figure A, there was a strong transient increase in the neuronal activity at the gamma (40–80 Hz) frequency from the 40–100 ms. The topologic map shown to the right displays the area where the increase in gamma power during the 40–100 ms time window was detected at the electrode level. As show in Figure B, the stimulation also induced an increase in the theta-alpha (4–12 Hz) frequency range during the 50–350 ms time window. The topologic figure shown to the right displays the area where the increase in the theta-alpha power during the 50–350 ms time window was detected at the electrode level.

Figure 2.

Figure 2.

Open in a new tab

Time-frequency spectrograms from the same electrode located near the sensorimotor cortical area (e.g., CZ). The displayed time-frequency spectrograms were averaged across participants for the sitting (A), standing (B) and walking (C) conditions, respectively. Time (in ms) is denoted on the x-axis, with 0 ms defined as the onset of the tibial nerve electrical stimulation. Spectral power is expressed as the percent difference from the baseline period, and the color scale bar is shown on the bottom. The top panels are the time frequency spectrogram seen for the gamma oscillations, while the bottom panels are the time frequency spectrograms seen for the alpha-theta (4–12 Hz) oscillations. Inspection of the respective figures shows that the increase in the theta-alpha oscillations was present across all of the respective conditions. However, the transient increase in the gamma (40–80 Hz) oscillations was prominent for the sitting and standing conditions, but markedly attenuated during walking.

Beamforming of the theta-alpha (4–12 Hz) range was performed using the time period and bandwidth identified in the electrode-level spectrogram analysis (50–350 ms) using a baseline period of equal duration and bandwidth (−550 to −250 ms). The images revealed that the theta-alpha oscillations were centered on the contralateral leg region of postcentral gyrus (Figure 3A). The peak voxel seen in this region was used to subsequently extract a neural time series for each participant under the respective conditions. Figure 4A displays the respective theta-alpha neural time courses where 0 ms represents the onset of the stimulus and the greyed area represents the beamformed time window (50–350 ms). Qualitative inspection of the neural time courses revealed that the theta-alpha oscillations were similar across the respective experimental conditions. To confirm this observation, the average activity across the 50 to 350 ms time window of the neural time course was then calculated, and a one-way ANOVA showed that there were no significant differences in the strength of the theta-alpha oscillations between the respective conditions (P=0.593; Figure 4A). Overall, these results indicate that the strength of these lower frequency oscillations were similar while sitting, standing and walking.

Figure 3.

Figure 3.

Open in a new tab

Grand averaged beamformer images collapsed across conditions for (A) theta-alpha (4–12 Hz) activity and (B) gamma (40–80 Hz) activity. Scale bar represent pseudo-t values. The location of neural generators for the theta-alpha and gamma oscillations resided in the contralateral leg region of the post-central gyrus.

Figure 4.

Figure 4.

Open in a new tab

Neural time courses for the theta-alpha (A; 4–12 Hz) and gamma (B; 40–80 Hz) responses within the peak voxel of the leg region of the somatosensory cortices for the respective conditions. Time (in ms) is denoted on the x-axis, with 0 ms defined as the onset of the tibial nerve electrical stimulation. The percent change in the power relative to the baseline period is shown on the y-axis. The sitting condition is shown in blue, standing condition is shown in red, and the walking condition is shown in black. The greyed area in the respective neural time courses represents the time window that was imaged and was subsequently used to calculate the average power differences between the respective conditions. The bar graphs in the right panels display the average relative power calculated for the corresponding time windows (i.e., greyed area in time course). The results show that the stimulation induced changes at the gamma frequency were considerably attenuated for the walking condition when compared with the sitting and standing conditions. Conversely, the stimulus induced changes at the theta-alpha frequency were not significantly different across the respective conditions. *P<0.05

To examine the source of the gamma oscillations, beamforming was also performed using the time period and bandwidth identified in the electrode-level spectrogram analysis (40–100 ms, 40–80 Hz) using a baseline period of equal duration and bandwidth (−550 to −490 ms). The images revealed that the gamma oscillations were also centered on the contralateral leg region of the postcentral gyrus (Figure 3B). The peak voxel in this region across all three conditions was subsequently used as a seed to extract voxel time series data for each participant under the respective conditions. Figure 4B displays the respective gamma neural time courses where 0 ms represents the onset of the stimulus and the greyed area represents the beamformed time window (40–100 ms). Qualitative inspection of the neural time courses revealed that the gamma oscillations appeared to be similar between the sitting and standing conditions, but were notably attenuated during the walking condition. To follow-up on this observation, the average activity across the 40 to 100 ms time window was subsequently calculated. A one-way ANOVA showed that there was a significant difference in the strength of the gamma oscillations between the respective conditions (P=0.0001; Figure 4B), and post-hoc analyses indicated that the average strength of the gamma oscillations was not different between the sitting and standing conditions (P=0.76), but was significantly weaker during the walking condition compared to both the sitting (P=0.001) and standing conditions (P=0.0001). Overall, these results show that the strength of the gamma oscillations were diminished while walking, but were similar while sitting and standing.

Discussion

This investigation used EEG and beamforming source reconstruction methods to quantify the changes in the somatosensory cortical oscillatory activity induced by an electrical stimulation of the tibial nerve while sitting, standing and walking. The data-driven approach employed in this investigation revealed that for all conditions the peripheral stimulation induced a transient increase in the theta-alpha (4–12Hz) and gamma (40–80 Hz) oscillations in the leg region of the contralateral somatosensory cortices. Furthermore, we identified that the amount of change in the gamma oscillations were similar while sitting and standing, but were markedly attenuated while walking. Conversely, the induced change the theta-alpha oscillations were similar across the respective experimental conditions. Further interpretation of our experimental results are discussed in the proceeding sections.

Our sitting and standing results are aligned with prior research that has shown a peripheral stimulation prompts an increase in the somatosensory cortical oscillations across the 10–75 Hz frequency bands (Dockstader et al., 2008; Gaetz and Cheyne, 2003; Kurz et al., 2018a; Spooner et al., 2018; Wiesman et al., 2017). However, our results reveal that the somatosensory gamma oscillations are attenuated during the stance phase of walking. Our prior MEG research found similar attenuation of the higher frequency somatosensory cortical oscillations while participants performed a hand motor task (Kurz et al., 2018b). Together these results support the notion that high frequency somatosensory cortical oscillations are significantly diminished during a motor action. Prior animal research has indicated that the suppressed cortical responses during movement might be the result of gating of the afferent feedback at several levels of the ascending lemniscal pathway (e.g., dorsal column nuclei, medial lemniscus, thalamus), and through cortico-cortical connections (Seki and Fetz, 2012; Seki et al., 2003). It has been postulated that this inhibitory effect is necessary for gaining greater control of the spinal circuits that govern the desired motor action (Seki et al., 2003), and has been referred to in the literature as centrifugal gating or central gating (Arpin et al., 2018; Jones et al., 1989; Saradjian, 2015; Wasaka et al., 2005). It is plausible that central gating may have played a role in our experimental outcomes since sitting and standing has fewer motor demands compared to walking. Along similar lines, the attenuation of the gamma oscillations seen at the cortical level may be a direct result of the neural computations performed at the spinal cord level. Prior studies have shown that the monosynaptic stretch reflex (i.e., Hoffman reflex) generated via stimulation of the type Ia sensory fibers plays a prominent role in the control of standing posture, but is inhibited when there is an isotonic contraction of the gastrocnemius-soleus musculature while walking (Cohen and Starr, 1985; Dietz et al., 1984, 1985; Trimble et al., 2000). Hence, we speculate that the attenuation of the gamma oscillations seen during the walking condition may reflect an inhibition of the afferent information provided by the Ia sensory fibers.

The intensity of the induced change in the theta-alpha oscillations were similar while sitting, standing and walking. This implies that the neural generators oscillating at these lower frequencies represents the processing of similar afferent sensory information that was present during the respective conditions. The time frequency components showed that, compared with the gamma oscillations (40–100 ms), the theta-alpha oscillations occurred at a slightly later time point and were sustained over a longer duration (50–350 ms). We suggest that these oscillations might be related to the processing of the afferent feedback from the type II polysynaptic spinal reflexes (e.g., Aβ fibers). The polysynaptic spinal pathways have longer latencies since they involve additional computations performed by the spinal interneurons and are influenced by the supraspinal motor circuitry (Dietz et al., 1985). Such pathways would likely not be suppressed during walking since they are known to play a prominent role in adapting the leg kinematics while walking and for the maintenance of postural balance.

The seminal studies on somatosensory cortical activity during standing versus walking evaluated the evoked potentials seen in the electrodes located on the scalp (Altenmuller et al., 1995; Dietz et al., 1985). Although these studies have provided unique insights, identifying the neural source generating the activity seen at the electrode level is critical as it allows more firm conclusions regarding the cortical sources involved. To this end, we used beamforming in the current study to identify the cortical source of the respective neural oscillations. This algorithm has certain advantages over the traditional equivalent-current dipole fitting techniques that have been predominately used in the literature for quantifying the sensorimotor cortical activity during walking (Snyder et al., 2015), since they do not require prior assumptions on the number of sources to model nor solving the inverse problem (Van Veen et al., 1997). We suggest that further employment of the beamforming methods used in this investigation will likely enhance the field’s understanding of the cortical neural generators that are involved in the control of human standing balance and walking.

Conclusions

Over the last decade, there has been a growing number of investigations that have used EEG to evaluate the sensorimotor cortical activity underlying the control of walking and standing postural balance (Hamacher et al., 2015; Wittenberg et al., 2017). Yet, very few of these investigations have specifically targeted the somatosensory cortical activity. Furthermore, limited attention has been paid to the neural oscillations seen at the source level. Our results are the first to show that the high frequency gamma somatosensory cortical oscillations are attenuated during the stance phase of walking, while the theta-alpha oscillations do not appear to appreciably change. We propose that linking biomechanical measures of the spatiotemporal kinematics with the observed changes in the somatosensory oscillatory cortical dynamics will provide unique insight on the sensory processing that occurs while walking. Furthermore, application of the neuroscience methods presented here could also be used to derive how different disease states (i.e., multiple sclerosis) and developmental disabilities (i.e., cerebral palsy) influence the cortical processing of afferent sensory information during different standing postural conditions and while walking.

Highlights.

  • EEG is used to quantify the strength of somatosensory cortical oscillations

  • Theta-alpha oscillations did not differ between sitting, standing, or walking

  • Gamma oscillations were similar while sitting and standing

  • Gamma oscillations were markedly attenuated while walking

Acknowledgements

This work was partially supported by the National Institutes of Health (1R01HD086245, R01HD101833, P20GM130447).

Footnotes

Publisher's Disclaimer: This is a PDF fi le of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its fi nal form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declarations of interest

The authors have nothing to declare.

References

  1. Altenmuller E, Berger W, Prokop T, Trippel M, Dietz V, 1995. Modulation of sural nerve somatosensory evoked potentials during stance and different phases of the step-cycle. Electroencephalogr Clin Neurophysiol 96, 516–525. [DOI] [PubMed] [Google Scholar]
  2. Arpin DJ, Gehringer JE, Wilson TW, Kurz MJ, 2018. Movement-Related Somatosensory Activity Is Altered in Patients with Multiple Sclerosis. Brain Topogr 31, 700–707. [DOI] [PubMed] [Google Scholar]
  3. Avikainen S, Forss N, Hari R, 2002. Modulated activation of the human SI and SII cortices during observation of hand actions. Neuroimage 15, 640–646. [DOI] [PubMed] [Google Scholar]
  4. Cohen LG, Starr A, 1985. Vibration and muscle contraction affect somatosensory evoked potentials. Neurology 35, 691–698. [DOI] [PubMed] [Google Scholar]
  5. Del Gratta C, Della Penna S, Ferretti A, Franciotti R, Pizzella V, Tartaro A, Torquati K, Bonomo L, Romani GL, Rossini PM, 2002. Topographic organization of the human primary and secondary somatosensory cortices: comparison of fMRI and MEG findings. Neuroimage 17, 1373–1383. [DOI] [PubMed] [Google Scholar]
  6. Dietz V, Quintern J, Berger W, 1984. Cerebral evoked potentials associated with the compensatory reactions following stance and gait perturbation. Neurosci Lett 50, 181–186. [DOI] [PubMed] [Google Scholar]
  7. Dietz V, Quintern J, Berger W, 1985. Afferent control of human stance and gait: evidence for blocking of group I afferents during gait. Exp Brain Res 61, 153–163. [DOI] [PubMed] [Google Scholar]
  8. Dockstader C, Gaetz W, Cheyne D, Wang F, Castellanos FX, Tannock R, 2008. MEG event-related desynchronization and synchronization deficits during basic somatosensory processing in individuals with ADHD. Behav Brain Funct 4, 8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ernst MD, 2004. Permutation methods: a basis for exact inference. Statistical Science 19, 676–685. [Google Scholar]
  10. Gaetz WC, Cheyne DO, 2003. Localization of human somatosensory cortex using spatially filtered magnetoencephalography. Neurosci Lett 340, 161–164. [DOI] [PubMed] [Google Scholar]
  11. Gross J, Kujala J, Hamalainen M, Timmermann L, Schnitzler A, Salmelin R, 2001. Dynamic imaging of coherent sources: Studying neural interactions in the human brain. Proc Natl Acad Sci U S A 98, 694–699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hamacher D, Herold F, Wiegel P, Hamacher D, Schega L, 2015. Brain activity during walking: A systematic review. Neurosci Biobehav Rev 57, 310–327. [DOI] [PubMed] [Google Scholar]
  13. Hamalainen H, Kekoni J, Sams M, Reinikainen K, Naatanen R, 1990. Human somatosensory evoked potentials to mechanical pulses and vibration: contributions of SI and SII somatosensory cortices to P50 and P100 components. Electroencephalogr Clin Neurophysiol 75, 13–21. [DOI] [PubMed] [Google Scholar]
  14. Hillebrand A, Barnes GR, 2005. Beamformer analysis of MEG data. Int Rev Neurobiol 68, 149–171. [DOI] [PubMed] [Google Scholar]
  15. Hillebrand A, Singh KD, Holliday IE, Furlong PL, Barnes GR, 2005. A new approach to neuroimaging with magnetoencephalography. Hum Brain Mapp 25, 199–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jones SJ, Halonen JP, Shawkat F, 1989. Centrifugal and centripetal mechanisms involved in the ‘gating’ of cortical SEPs during movement. Electroencephalogr Clin Neurophysiol 74, 36–45. [DOI] [PubMed] [Google Scholar]
  17. Jung P, Baumgartner U, Bauermann T, Magerl W, Gawehn J, Stoeter P, Treede RD, 2003. Asymmetry in the human primary somatosensory cortex and handedness. Neuroimage 19, 913–923. [DOI] [PubMed] [Google Scholar]
  18. Kakigi R, Hoshiyama M, Shimojo M, Naka D, Yamasaki H, Watanabe S, Xiang J, Maeda K, Lam K, Itomi K, Nakamura A, 2000. The somatosensory evoked magnetic fields. Prog Neurobiol 61, 495–523. [DOI] [PubMed] [Google Scholar]
  19. Kovach CK, Gander PE, 2016. The demodulated band transform. J Neurosci Methods 261, 135–154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kristeva-Feige R, Rossi S, Pizzella V, Lopez L, Erne S, Edrich J, Rossini PM, 1996. Neuromagnetic study of movement-related somatosensory gating. Electroencephalogr Clin Neurophysiol Suppl 46, 337–342. [PubMed] [Google Scholar]
  21. Kurz MJ, Wiesman AI, Coolidge NM, Wilson TW, 2018a. Children with Cerebral Palsy Hyper-Gate Somatosensory Stimulations of the Foot. Cereb Cortex 28, 2431–2438. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kurz MJ, Wiesman AI, Coolidge NM, Wilson TW, 2018b. Haptic exploration attenuates and alters somatosensory cortical oscillations. J Physiol 596, 5051–5061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Macerollo A, Chen JC, Parees I, Sadnicka A, Kassavetis P, Bhatia KP, Kilner JM, Rothwell JC, Edwards MJ, 2016. Abnormal movement-related suppression of sensory evoked potentials in upper limb dystonia. Eur J Neurol 23, 562–568. [DOI] [PubMed] [Google Scholar]
  24. Maris E, Oostenveld R, 2007. Nonparametric statistical testing of EEG-and MEG-data. J Neurosci Methods 164, 177–190. [DOI] [PubMed] [Google Scholar]
  25. Nakamura A, Yamada T, Goto A, Kato T, Ito K, Abe Y, Kachi T, Kakigi R, 1998. Somatosensory homunculus as drawn by MEG. Neuroimage 7, 377–386. [DOI] [PubMed] [Google Scholar]
  26. Penfield W, Boldrey E, 1937. Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain 60, 389–443. [Google Scholar]
  27. Saradjian AH, 2015. Sensory modulation of movement, posture and locomotion. Neurophysiol Clin 45, 255–267. [DOI] [PubMed] [Google Scholar]
  28. Seki K, Fetz EE, 2012. Gating of sensory input at spinal and cortical levels during preparation and execution of voluntary movement. J Neurosci 32, 890–902. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Seki K, Perlmutter SI, Fetz EE, 2003. Sensory input to primate spinal cord is presynaptically inhibited during voluntary movement. Nat Neurosci 6, 1309–1316. [DOI] [PubMed] [Google Scholar]
  30. Snyder KL, Kline JE, Huang HJ, Ferris DP, 2015. Independent Component Analysis of Gait-Related Movement Artifact Recorded using EEG Electrodes during Treadmill Walking. Front Hum Neurosci 9, 639. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Spooner RK, Wiesman AI, Mills MS, O’Neill J, Robertson KR, Fox HS, Swindells S, Wilson TW, 2018. Aberrant oscillatory dynamics during somatosensory processing in HIV-infected adults. Neuroimage Clin 20, 85–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Staines WR, Brooke JD, McIlroy WE, 2000. Task-relevant selective modulation of somatosensory afferent paths from the lower limb. Neuroreport 11, 1713–1719. [DOI] [PubMed] [Google Scholar]
  33. Trimble MH, Du P, Brunt D, Thompson FJ, 2000. Modulation of triceps surae H-reflexes as a function of the reflex activation history during standing and stepping. Brain Res 858, 274–283. [DOI] [PubMed] [Google Scholar]
  34. Van Veen BD, van Drongelen W, Yuchtman M, Suzuki A, 1997. Localization of brain electrical activity via linearly constrained minimum variance spatial filtering. IEEE Trans Biomed Eng 44, 867–880. [DOI] [PubMed] [Google Scholar]
  35. Wasaka T, Nakata H, Kida T, Kakigi R, 2005. Changes in the centrifugal gating effect on somatosensory evoked potentials depending on the level of contractile force. Exp Brain Res 166, 118–125. [DOI] [PubMed] [Google Scholar]
  36. Wiesman AI, Heinrichs-Graham E, Coolidge NM, Gehringer JE, Kurz MJ, Wilson TW, 2017. Oscillatory dynamics and functional connectivity during gating of primary somatosensory responses. J Physiol 595, 1365–1375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wittenberg E, Thompson J, Nam CS, Franz JR, 2017. Neuroimaging of Human Balance Control: A Systematic Review. Front Hum Neurosci 11, 170. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES