Figure - available from: Experimental Brain Research
This content is subject to copyright. Terms and conditions apply.
Representative examples of rectified electromyographic (EMG) signals during the adaptation periods on the perturbed (right) side. The paired TA muscle activities during the swing phase before the foot strike on the same side and the MG and LG muscle activities during the mid-to-terminal stance phase before the foot strike on the opposite side were segmented into 400 ms data segments. The vertical dashed lines with the filled and open triangles indicate foot strike events on the perturbed and unperturbed (left) sides, respectively. TA tibialis anterior, MG medial gastrocnemius, LG lateral gastrocnemius, EA early adaptation, LA late adaptation
Source publication
Different neural contributions to motor learning might be involved when different error sizes of perturbation are introduced. Although the corticospinal drive contributes to abrupt gait adaptation processes, no studies have investigated whether cortical involvement during gait differs between perturbations applied abruptly and gradually. This study...
Citations
... The copyright holder for this preprint (which this version posted July 25, 2023. ; https://doi.org/10.1101/2023.07.21.550018 doi: bioRxiv preprint Depending on the gait cycle phase of interest, previous work segmented EMG signals using predetermined fixed duration (Kitatani et al. 2016;Jensen et al. 2019) and number (e.g., 100) (Kitatani et al. 2022;Nielsen et al. 2008) of epochs. Such an approach ensures that the length and number of all epochs remains the same across trials and subjects, which is convenient for subsequent computational analyses. ...
Plantarflexors provide propulsion during walking (late stance) and receive input from both corticospinal tract (CST) and corticoreticulospinal tract (CReST). Both descending motor tracts exhibit some frequency-specificity, which allows potential differentiation of neural drive from each tract using intermuscular coherence (IMC). Stroke may differentially affect each tract, thus impair the function of plantarflexors. However, the evidence concerning this frequency-specificity and its relation to plantarflexors’ neuromechanics post-stroke remains very limited. Here, we investigated the intermuscular coherences of alpha, beta, and low-gamma bands between the Soleus (SOL), Lateral Gastrocnemius (LG), and Medial Gastrocnemius (MG) muscles and their relationships with walking-specific measures (propulsive impulse; speed). Fourteen individuals with chronic stroke walked on a treadmill at self-selected and fast walking speed (SSWS and FWS, respectively). Inter-limb IMC comparisons revealed that beta LG-MG (SSWS) and low-gamma SOL-LG (FWS) IMCs were degraded on the paretic side. At the same time, within each limb, the IMCs, which were significantly different to a surrogate dataset denoting random coherence, were in the alpha band (both speeds). Further, alpha LG-MG IMC was positively correlated with propulsive impulse in the paretic limb (SSWS). Findings suggest differential functional role of alpha and beta/low-gamma, which may be related to the frequency-specificity of the underlying descending drives. The persistence of alpha in plantarflexors and its strong positive relationship with propulsive impulse suggests relative preservation and/or upregulation of CReST. Future research should address whether entraining motor system at alpha frequencies via neuromodulation can improve the neuromechanical function of paretic plantarflexors and subsequently promote post-stroke walking recovery.
Key Points Summary
Cortical and subcortical motor drives may be frequency-specific, have a role in walking, and be degraded after stroke.
Whether this frequency-specificity exists and how it is related to neuromechanical function of ankle plantarflexors post-stroke remains to be determined.
Here, we investigated bilaterally the intermuscular coherences of alpha, beta, and low-gamma bands for the Soleus (SOL), Lateral Gastrocnemius (LG), and Medial Gastrocnemius (MG) muscles and their relationships with walking-specific measures (propulsive impulse; self-selected and fast speed) during treadmill walking in individuals post-stroke.
The beta LG-MG (self-selected speed) and low-gamma SOL-LG (fast speed) were degraded on the paretic side.
Alpha coherence was significantly present across plantarflexors mainly on the non-paretic side (both speeds).
Paretic alpha LG-MG was positively correlated with paretic propulsive impulse (self-selected speed).
Given that paretic propulsive impulse is impaired post-stroke, entraining the motor system at alpha frequency via neuromodulation may improve propulsive impulse and subsequently promote post-stroke walking recovery.
... Electromyography coherence analysis has demonstrated a common neural drive at 15-45 Hz to the tibialis anterior that is modulated during walking adaptation (Sato and Choi, 2019;Oshima et al., 2021;Kitatani et al., 2022). During normal walking, a significant amount of coherence can be found between EMG recorded from the proximal and distal ends of the tibialis anterior in the alpha (8-15 Hz), beta (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30), and gamma (30-45 Hz) frequencies during the swing phase of gait. ...
... However, we did not observe consistent beta-gamma coherence differences between pre-slow and post-slow that would be explained by fatigue. Alternatively, reduced modulation in beta-gamma coherence during split-belt adaptation may reflect less cortical involvement, due to greater reliance on implicit processes (Kitatani et al., 2022). ...
Healthy aging is associated with reduced corticospinal drive to leg muscles during walking. Older adults also exhibit slower or reduced gait adaptation compared to young adults. The objective of this study was to determine age-related changes in the contribution of corticospinal drive to ankle muscles during walking adaptation. Electromyography (EMG) from the tibialis anterior (TA), soleus (SOL), medial, and lateral gastrocnemius (MGAS, LGAS) were recorded from 20 healthy young adults and 19 healthy older adults while they adapted walking on a split-belt treadmill. We quantified EMG-EMG coherence in the beta-gamma (15–45 Hz) and alpha-band (8–15 Hz) frequencies. Young adults demonstrated higher coherence in both the beta-gamma band coherence and alpha band coherence, although effect sizes were greater in the beta-gamma frequency. The results showed that slow leg TA-TA coherence in the beta-gamma band was the strongest predictor of early adaptation in double support time. In contrast, early adaptation in step length symmetry was predicted by age group alone. These findings suggest an important role of corticospinal drive in adapting interlimb timing during walking in both young and older adults.
Locomotor adaptation to abrupt and gradual perturbations are likely driven by fundamentally different neural processes. The aim of this study was to quantify brain dynamics associated with gait adaptation to a gradually introduced gait perturbation, which typically results in smaller behavioral errors relative to an abrupt perturbation. Loss of balance during standing and walking elicits transient increases in midfrontal theta oscillations that have been shown to scale with perturbation intensity. We hypothesized there would be no significant change in anterior cingulate theta power (4–7 Hz) with respect to pre-adaptation when a gait perturbation is introduced gradually because the gradual perturbation acceleration and stepping kinematic errors are small relative to an abrupt perturbation. Using mobile electroencephalography (EEG), we measured gait-related spectral changes near the anterior cingulate, posterior cingulate, sensorimotor, and posterior parietal cortices as young, neurotypical adults ( n = 30) adapted their gait to an incremental split-belt treadmill perturbation. Most cortical clusters we examined (>70%) did not exhibit changes in electrocortical activity between 2–50 Hz. However, we did observe gait-related theta synchronization near the left anterior cingulate cortex during strides with the largest errors, as measured by step length asymmetry. These results suggest gradual adaptation with small gait asymmetry and perturbation magnitude may not require significant cortical resources beyond normal treadmill walking. Nevertheless, the anterior cingulate may remain actively engaged in error monitoring, transmitting sensory prediction error information via theta oscillations.
Cortical resources are typically engaged for balance and mobility in older adults, but these resources are impaired post-stroke. Although slowed balance and mobility after stroke have been well-characterized, the effects of unilateral cortical lesions due to stroke on neuromechanical control of balance is poorly understood. Our central hypothesis is that stroke impairs the ability to rapidly and effectively engage the cerebral cortex during balance and mobility behaviors, resulting in asymmetrical contributions of each limb to balance control. Using electroencephalography (EEG), we assessed cortical N1 responses evoked over fronto-midline regions (Cz) during balance recovery in response to backward support-surface perturbations loading both legs, as well as posterior-lateral directions that preferentially load the paretic or nonparetic leg. Cortical N1 responses were smaller and delayed in the stroke group. While older adults exhibited weak or absent relationships between cortical responses and clinical function, stroke survivors exhibited strong associations between slower N1 latencies and slower walking, lower clinical mobility, and lower balance function. We further assessed kinetics of balance recovery during perturbations using center of pressure rate of rise. During backward support-surface perturbations that loaded the legs bilaterally, balance recovery kinetics were not different between stroke and control groups and were not associated with cortical response latency. However, lateralized perturbations revealed slower kinetic reactions during paretic loading compared to controls, and to non-paretic loading within stroke participants. Individuals post stroke had similar nonparetic-loaded kinetic reactions to controls implicating that they effectively compensate for impaired paretic leg kinetics when relying on the non-paretic leg. In contrast, paretic-loaded balance recovery revealed time-synchronized associations between slower cortical responses and slower kinetic reactions only in the stroke group, potentially reflecting the limits of cortical engagement for balance recovery revealed within the behavioral context of paretic motor capacity. Overall, our results implicate individuals after stroke may be uniquely limited in their balance ability by the slowed speed of their cortical engagement, particularly under challenging balance conditions that rely on the paretic leg. We expect this neuromechanical insight will enable progress toward an individualized framework for the assessment and treatment of balance impairments based on the interaction between neuropathology and behavioral context.
Corticospinal drive during walking is reduced in older adults compared to young adults, but it is not clear how this decrease might compromise one's ability to adjust stepping, particularly during visuomotor adaptation. We hypothesize that age-related changes in corticospinal drive could predict differences in older adults' step length and step time adjustments in response to visual perturbations compared to younger adults. Healthy young (n = 21; age 18-33 years) and older adults (n = 20; age 68-80 years) were tested with a treadmill task, incorporating visual feedback of the foot position and stepping targets in real-time. During adaptation, the visuomotor gain was reduced on one side, causing the foot cursor and step targets to move slower on that side of the screen (i.e., split-visuomotor adaptation). Corticospinal drive was quantified by coherence between electromyographic signals in the beta-gamma frequency band (15-45 Hz). The results showed that 1) older adults adapted to visuomotor perturbations during walking, with similar reduction in error asymmetry compared to younger adults; 2) however, older adults showed reduced adaptation in step time symmetry, despite demonstrating similar adaptation in step length asymmetry compared to younger adults; 3) smaller overall changes in step time asymmetry was associated with reduced corticospinal drive to the tibialis anterior in the slow leg during split-visuomotor adaptation. These findings suggest that changes in corticospinal drive may affect older adults' control of step timing in response to visual challenges. This could be important for safe navigation when walking in different environments or dealing with unexpected circumstances.
Rhythmic auditory stimulation (RAS) improves gait symmetry in neurological patients with asymmetric gait patterns. However, whether RAS can accelerate gait adaptation remains unclear. This study aimed to investigate whether RAS during gait adaptation can enhance learning after-effects and savings of gait symmetries. Furthermore, we investigated the differences in coherence of paired surface electromyographic (EMG) recordings during gait adaptation between with and without RAS. Nineteen healthy young adults were subjected to continuous treadmill gait with swing phase perturbation (adaptation period) with or without RAS (RAS or no-RAS condition) for 5 or 10 min (short- or long-time condition), without the perturbation for 5 min (de-adaptation period), and with the perturbation for another 5 min (re-adaptation period). Swing phase and step length symmetries were significantly greater in the RAS conditions than in the no-RAS conditions during the adaptation period. Learning after-effects and savings of gait symmetries were significantly greater in the RAS conditions than in the no-RAS conditions in the early de-adaptation and re-adaptation periods, respectively. There were no significant differences in savings in the early re-adaptation period between the short- and long-time conditions in the RAS condition. EMG-EMG coherence in the rectus femoris muscle in the beta band (15-35 Hz) on the perturbed side was significantly lower during the early adaptation period in the RAS than in the no-RAS conditions. Therefore, RAS may enhance learning efficiency by reducing common neural drives from cortical structure during gait adaptation, which could induce high savings of a learned gait pattern, even within short-time periods.
