Biology subjects

Daffertshofer, A.

Publications and source records attributed to Daffertshofer, A..

16 recordsLinked to original sources

bioRxiv2026

Cortico-muscular control of postural balance in older adults - effects of training on functional connectivity

BackgroundBalance training in older adults can lead to reduced centre of mass accelerations and reduced angular momenta after perturbations of unipedal stance, reflecting an enhanced ability to recover balance. It has been suggested that the co-occurring changes in muscle synergies indicated strategy-specific adaptations in feedback control. MethodsWe investigated the cortical involvement in such adaptations by focusing on the interaction between muscle synergies and cortical activity after perturbations. Twenty older adults (>65 years) underwent short-term and three-week long-term balance training, and we assessed their recovery from unpredictable mediolateral perturbations during unipedal stance. We measured high-density EEG and activation of leg and trunk muscles. The representations of the balance-related muscle synergies were localised in the cortex using coherence-based beamformers in the {beta}-frequency band. ResultsBalance performance was accompanied by task-specific {beta}-band activation in the somatotopic representation of the lower extremities in the primary motor cortex. The {beta}-power significantly dropped during the response to perturbations, while the coherence with the activation of muscle synergies significantly increased, especially for synergies active in the early stage of balance recovery. The task-related changes in cortico-synergy coherence, especially during the later phase of balance recovery, were significantly affected by short-term training. ConclusionRefinements of feedback control seem to underlie balance improvements in older adults. The significant changes in the cortico-synergy interaction after balance training suggest cortical involvement in these refinements.

neuroscience↗

bioRxiv2025

Age-related differences in the cortical motor network during uni-manual and bimanual coordination - an EEG study

Aging is accompanied by various neurophysiological changes that may affect cognitive and motor function. Compared to younger adults, motor performance in older adults engages more brain regions and exhibits elevated neural activity in the motor-related alpha (8-12 Hz) and beta (15-30 Hz) frequency bands of electro-encephalography (EEG) signals. To what extent these functional changes correlate with the quality of motor performance is underexplored. We recorded EEG in 19 young and 19 older adults during unimanual and bimanual visuomotor tasks with different degrees of difficulty. Older participants showed a lower quality of performance than younger adults, especially during asymmetric bimanual tasks. We analyzed source-localized activity in bilateral parietal and (pre-)motor areas to investigate, especially, the hypothesized pivotal role of left premotor cortex (PMCL) activity within the motor network during motor performance. In PMCL, beta activity was indeed significantly affected by age during bimanual performance, while alpha activity was altered in the bilateral (pre-)cuneus. When predicting error of performance via alpha and beta modulations, we found significant associations in bilateral M1 and (pre-)cuneusR during unimanual performance, while in bimanual performance, PMCL and (pre-)cuneusL were also included in the significant association. Our results confirm the importance of PMCL and (pre-)cuneusL for the performance of bimanual tasks, especially when the tasks are challenging. The age-related differences in alpha power in bilateral (pre-)cuneus and their associations with motor performance suggest that altered visuomotor integration has an important contribution to the reduction of the quality of motor performance in older adults.

neuroscience↗

bioRxiv2023

Machine learning approaches to predicting whether muscles can be elicited via TMS

BackgroundTranscranial magnetic stimulation (TMS) is a valuable technique for assessing the function of the motor cortex and cortico-muscular pathways. TMS activates the motoneurons in the cortex, and this activation is transmitted through the cortico-muscular pathway, after which it can be measured as a motor evoked potential (MEP) in the muscles. The position and orientation of the TMS coil and the intensity used to deliver a TMS pulse are considered central TMS setup parameters influencing the presence/absence of MEPs. New MethodWe sought to predict the presence of MEPs from TMS setup parameters using machine learning. We trained different machine learners using either within-subject or between-subject designs. ResultsWe obtained prediction accuracies of on average 77% and 65% with maxima up to up to 90% and 72% within and between subjects, respectively. Across the board, a bagging ensemble appeared to be the most suitable approach to predict the presence of MEPs, although a comparably simple logistic regression model also performed well. ConclusionsWhile the prediction between subjects clearly leaves room for improvement, the within-subject performance encourages to supplement TMS by machine learning to improve its diagnostic capacity with respect to motor impairment.

neuroscience↗

bioRxiv2023

Overlap in the cortical representation of hand and forearm muscles as assessed by navigated TMS

The representation of upper limb muscles in the motor cortex is complex. It contains areas of excitability that may overlap between muscles. We expected the cortical representations of synergistic muscle pairs to overlap more than those of non-synergistic muscles. To detail this, we used navigated transcranial magnetic stimulation in eight hand and forearm muscles of twenty healthy participants. We transformed the cortical representations of muscles to a template MRI to allow for group analysis. We found that the amount of overlap in cortical representations differed significantly between within-hand and within-forearm muscle combinations. Most synergistic muscle pairs, both within the hand, within the forearm and between them, had a larger overlap than non-synergistic muscle pairs. Our study supports the largely overlapping nature of cortical representations of upper limb muscles. We can particularly underscore that the overlap is elevated in muscles that usually act in a synergistic manner.

neuroscience↗

bioRxiv2022

Is phase-dependent stability related to phase-dependent gait robustness?

Predicting gait robustness is paramount for targeting interventions to prevent falls. Our analysis of a compass walker model revealed that phase-dependent stability measures have limited capacity to predict overall gait robustness. Interestingly, these measures vary substantially over the gait cycle which seemingly aligns with the robustness of both humans and walking models that likewise depends on the phase at which perturbations occur. To explore this in depth, we again used a compass walker model that can walk stably and periodically and applied forward and backward perturbations (instantaneous changes in angular velocities) to the stance or swing leg at multiple instants during the single-stance phase. We then estimated the degree to which phase-dependent stability measures correlate with phase-dependent gait robustness, quantified as the maximal perturbation-induced change in mechanical energy that the walker could tolerate without falling within 50 steps. We found that these phase-dependent stability measures did not adequately predict phase-dependent gait robustness in the compass walker model. We therefore conclude that these measures are unlikely to provide reliable predictions of gait robustness or fall risk in humans.

physiology↗

bioRxiv2022

The energetic effect of hip flexion and retraction in walking at different speeds: a modeling study

In human walking, power for propulsion is generated primarily via ankle and hip muscles. The addition of a passive hip spring to simple bipedal models appears more efficient than using only push-off impulse, at least, when hip spring associated energetic costs are not considered. Hip flexion and retraction torques, however, are not free, as they are produced by muscles demanding metabolic energy. Studies evaluating the inclusion of hip actuation costs, especially during the swing phase, and the hip actuations energetic benefits are few and far between. It is also unknown whether these possible benefits/effects may depend on speed. We simulated a planar flat-feet model walking stably over a range of speeds. We asked whether the addition of independent hip flexion and retraction remains energetically beneficial when considering work-based metabolic cost of transport (MCOT) with different efficiencies of doing positive and negative work. We found asymmetric hip actuation can reduce the estimated MCOT relative to ankle actuation by up to 6%, but only at medium speeds. The corresponding optimal strategy is zero hip flexion and some hip retraction actuation. The reason for this reduced MCOT is a reduction in collision loss being larger than the associated increase in hip negative work. Both terms require positive mechanical work to compensate, yielding a larger reduction in metabolic work per step than the reduction in step length. Our study shows how ankle actuation, hip flexion, and retraction actuation can be coordinated to reduce MCOT.

physiology↗

bioRxiv2022

Does ankle push-off correct for errors in anterior-posterior foot placement relative to center-of-mass state?

I.Understanding the mechanisms humans use to stabilize walking is vital for predicting falls in elderly. Modeling studies identified two potential mechanisms to stabilize gait in the anterior-posterior direction: foot placement control and ankle push-off control: Foot placement depends on position and velocity of the center-of-mass (CoM) and push-off modulates with deviations between actual and predicted CoM trajectories. While both control mechanisms have been reported in humans, it is unknown whether especially the latter one is employed in unperturbed steady-state walking. On the one hand, the covariance between CoM states and anterior-posterior foot placement served as a measure of foot placement control. On the other hand, we determined variations in ankle push-off as a function of differences between the actual foot placement and the one predicted from CoM states. We estimated the corresponding correlations and consider them indictors for push-off control based on foot placement errors. We found ankle push-off torque to be correlated to the foot placement error in 30 participants when walking at normal and slow speed, with mean correlation values of up to 0.45. Our study suggests that humans use a push-off strategy for correcting foot placement errors in steady-state walking.

physiology↗

bioRxiv2022

Predicting vertical ground reaction forces from 3D accelerometery using reservoir computers leads to accurate gait event detection

Accelerometers are low-cost measurement devices that can readily be used outside the lab. However, determining isolated gait events from accelerometer signals, especially foot-off events during running, is an open problem. We outline a two-step approach where machine learning serves to predict vertical ground reaction forces from accelerometer signals, followed by force-based event detection. We collected shank accelerometer signals and ground reaction forces from 21 adults during comfortable walking and running on an instrumented treadmill. We trained one common reservoir computer using segmented data using both walking and running data. Despite being trained on just a small number of strides, this reservoir computer predicted vertical ground reaction forces in continuous gait with high quality. The subsequent foot contact and foot off event detection proved highly accurate when compared to the golden standard based on co-registered ground reaction forces. Our proof-of-concept illustrates the capacity of combining accelerometry with machine learning for detecting isolated gait events irrespective of mode of locomotion.

neuroscience↗

bioRxiv2021

A new protocol for multiple muscle mapping using nTMS

The representation of muscles in the cortex can be mapped using navigated transcranial magnetic stimulation. The mapping can be quantified via measures like the centre of gravity or the size of the region of excitability. Determining these measures typically relies only on stimulation points that yield motor-evoked potentials (MEPs); stimulations that do not elicit an MEP, i.e., non-MEP points, are ignored entirely. In this study, we show how incorporating non-MEP points may affect the estimates of the size and centroid of the excitable area in eight hand and forearm muscles after mono-phasic single-pulse TMS. We performed test-retest assessments in twenty participants and estimated the reliability of centroids and sizes of the corresponding areas using inter-class correlation coefficients. For most muscles, the reliability turned out good, if not excellent. As expected, removing the non-MEP points significantly decreased area sizes and area weights, suggesting that conventional approaches that do not account for non-MEP points are likely to overestimate the regions of excitability.

neuroscience↗

bioRxiv2021

Ankle muscles activate: mediolateral center of pressure control ensures stable gait

During steady-state walking, mediolateral gait stability can be maintained by controlling the center of pressure (CoP). The CoP modulates the moment of the ground reaction force, which brakes and reverses movement of the center of mass (CoM) towards the lateral border of the base of support. In addition to foot placement, ankle moments serve to control the CoP. We hypothesized that, during steady-state walking, single stance ankle moments establish a CoP shift to correct for errors in foot placement. We expected ankle muscle activity to be associated with this complementary CoP shift. During treadmill walking, full-body kinematics, ground reaction forces and electromyography were recorded in thirty healthy participants. We found a negative relationship between preceding foot placement error and CoP displacement during single stance; steps that were too medial were compensated for by a lateral CoP shift and vice versa, steps that were too lateral were compensated for by a medial CoP shift. Peroneus longus, soleus and tibialis anterior activity correlated with these CoP shifts. As such, we identified an (active) ankle strategy during steady-state walking. As expected, absolute explained CoP variance by foot placement error decreased when walking with shoes constraining ankle moments. Yet, contrary to our expectations that ankle moment control would compensate for constrained foot placement, the absolute explained CoP variance by foot placement error did not increase when foot placement was constrained. We argue that this lack of compensation reflects the interdependent nature of ankle moment and foot placement control. We suggest that single stance ankle moments do not only compensate for preceding foot placement errors, but also assist control of the subsequent foot placement. Foot placement and ankle moment control are caught in a circular relationship, in which constraints imposed on one will also influence the other.

neuroscience↗

bioRxiv2021

Effect of aging-related network changes on unimanual sensorimotor learning - a simultaneous EEG-fMRI study

Sensorimotor coordination requires orchestrated network activity mediated by inter- and intra-hemispheric, excitatory and inhibitory neuronal interactions. Aging-related structural changes may alter these interactions. Disbalancing strength and timing of excitation and inhibition may limit motor performance. This is particularly true during motor coordination tasks that have to be learned through practice. To investigate this, we simultaneously acquired electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) in two groups of healthy adults (young N=13: 20-25y and elderly N=14: 59-70y), while they were practicing a unimanual motor task. Both groups learned the task during brain scanning, which was confirmed by a 24h follow-up retention test. On average, quality of performance of older participants stayed significantly below that of the younger ones. Accompanying decreases in motor-event-related EEG-source beta band power ({beta}, 15-30 Hz) were lateralized in both groups towards the contralateral side, albeit more so in younger participants. In the latter, the mean {beta}-power during motor learning in bilateral pre-motor cortex (PM1) was significantly higher than in the older group. Combined EEG/fMRI analysis revealed positive correlations between fMRI signals and source-reconstructed {beta}-amplitude time courses in contralateral and ipsilateral M1, and negative correlations in bilateral PM1 for both groups. The {beta}-positive fMRI response in bilateral M1 might be explained by an increased cross-talk between hemispheres during periods of pronounced {beta}-activity. During learning, the Rolandic {beta}-power relative to rest was higher in bilateral PM1 in younger participants, suggesting less task-related beta band desynchronization in this (better performing) group. We also found positive correlations between Rolandic {beta}-amplitude and fMRI-BOLD in bilateral M1 and negative correlations bilateral in PM1. This indicates that increased {beta}-amplitudes are associated with increased M1 "activity" (positive BOLD response) and decreased PM1 "activity" (negative BOLD response). Our results point at decreased pre-motor inhibitory inputs to M1 as possible source for increased interhemispheric crosstalk and an aging-related decline in motor performance. HighlightsO_LISensorimotor coordination performance decreases with increasing age. C_LIO_LIDuring motor learning the {beta}-power in pre-motor areas is reduced with age. C_LIO_LIEEG/fMRI points at less effective inhibitory inputs from PM1 to ipsilateral M1 in older adults. C_LI

neuroscience↗

bioRxiv2020

The validation of new phase-dependent gait stability measures: a modeling approach

Identification of individuals at risk of falling is important when designing fall prevention methods. Current measures that estimate gait stability and robustness appear limited in predicting falls in older adults. Inspired by recent findings on changes in phase-dependent local stability within a gait cycle, we devised several phase-dependent stability measures and tested for their usefulness to predict gait robustness in compass walker models. These measures are closely related to the often-employed maximum finite-time Lyapunov exponent and maximum Floquet multiplier that both assess a systems response to infinitesimal perturbations. As such, they entail linearizing the system but this is realized in a rotating hypersurface orthogonal to the period-one solution followed by estimating the trajectory-normal divergence rate of the swing phases and the foot strikes. We correlated the measures with gait robustness, i.e. the largest perturbation a walker can handle, in two compass walker models with either point or circular feet to estimate their prediction accuracy. To also test for the dependence of the measures under state space transform, we represented the point-feet walker in both Euler-Lagrange and Hamiltonian canonical form. Our simulations revealed that for most of the measures their correlation with gait robustness differs between models and between different state space forms. In particular the latter may jeopardize many stability measures predictive capacity for gait robustness. The only exception that consistently displayed strong correlations is the divergence of foot strike. Our results admit challenges of using phase-dependent stability measures as objective means to estimate the risk of falling.

bioengineering↗

bioRxiv2020

Step width and frequency to modulate: Active foot placement control ensures stable gait

Step-by-step foot placement control, relative to the center of mass (CoM) kinematic state, is generally considered a dominant mechanism for maintenance of gait stability. By adequate (mediolateral) positioning of the center of pressure with respect to the CoM, the ground reaction force generates a moment that prevents falling. In healthy individuals, foot placement is complemented mainly by ankle moment control ensuring stability. To evaluate possible compensatory relationships between step-by-step foot placement and complementary ankle moments, we investigated the degree of (active) foot placement control during steady-state walking, and under either foot placement-, or ankle moment constraints. Thirty healthy participants walked on a treadmill, while full-body kinematics, ground reaction forces and EMG activities were recorded. As a replication of earlier findings, we first showed step-by-step foot placement is associated with preceding CoM state and hip ab-/adductor activity during steady-state walking. Tight control of foot placement appears to be important at normal walking speed because there was a limited change in the degree of foot placement control despite the presence of a foot placement constraint. At slow speed, the degree of foot placement control decreased substantially, suggesting that tight control of foot placement is less essential when walking slowly. Step-by-step foot placement control was not tightened to compensate for constrained ankle moments. Instead compensation was achieved through increases in step width and stride frequency.

neuroscience↗

bioRxiv2017

Comparing average network signals and neural mass signals in systems with low-synchrony

1Neural mass models are accepted as efficient modelling techniques to model empirical observations such as disturbed oscillations or neuronal synchronization. Neural mass models are based on the mean-field assumption, i.e. they capture the mean-activity of a neuronal population. However, it is unclear if neural mass models still describe the mean activity of a neuronal population when the underlying neural network topology is not homogenous. Here, we test whether the mean activity of a neuronal population can be described by neural mass models when there is neuronal loss and when the connections in the network become sparse. To this end, we derive two neural mass models from a conductance based leaky integrate-and-firing (LIF) model. We then compared the power spectral densities of the mean activity of a network of inhibitory and excitatory LIF neurons with that of neural mass models by computing the Kolmogorov-Smirnov test statistic. Firstly, we found that when the number of neurons in a fully connected LIF-network is larger than 300, the neural mass model is a good description of the mean activity. Secondly, if the connection density in the LIF-network does not exceed a crtical value, this leads to desynchronization of neurons within the LIF-network and to failure of neural mass description. Therefore we conclude that neural mass models can be used for analysing empirical observations if the neuronal network of interest is large enough and when neurons in this system synchronize.

neuroscience↗

bioRxiv2017

Network structure of the human musculoskeletal system shapes neural interactions on multiple timescales

Human motor control requires the coordination of muscle activity under the anatomical constraints imposed by the musculoskeletal system. Interactions within the central nervous system are fundamental to motor coordination, but the principles governing functional integration remain poorly understood. We used network analysis to investigate the relationship between anatomical and functional connectivity amongst 36 muscles. Anatomical networks were defined by the physical connections between muscles and functional networks were based on intermuscular coherence assessed during postural tasks. We found a modular structure of functional networks that was strongly shaped by the anatomical constraints of the musculoskeletal system. Changes in postural tasks were associated with a frequency-dependent reconfiguration of the coupling between functional modules. These findings reveal distinct patterns of functional interactions between muscles involved in flexibly organising muscle activity during postural control. Our network approach to the motor system offers a unique window into the neural circuitry driving the musculoskeletal system.

neuroscience↗

bioRxiv2017

Functional Connectivity Dynamics of the Resting State across the Human Adult Lifespan

Functional Connectivity (FC) during resting-state or task conditions is not fixed but inherently dynamic. Yet, there is no consensus on whether fluctuations in FC may resemble isolated transitions between discrete FC states rather than continuous changes. This quarrel hampers advancing the study of dynamic FC. This is unfortunate as the structure of fluctuations in FC can certainly provide more information about developmental changes, aging, and progression of pathologies. We merge the two perspectives and consider dynamic FC as an ongoing network reconfiguration, including a stochastic exploration of the space of possible steady FC states. The statistical properties of this random walk deviate both from a purely "order-driven" dynamics, in which the mean FC is preserved, and from a purely "randomness-driven" scenario, in which fluctuations of FC remain uncorrelated over time. Instead, dynamic FC has a complex structure endowed with long-range sequential correlations that give rise to transient slowing and acceleration epochs in the continuous flow of reconfiguration. Our analysis for fMRI data in healthy elderly revealed that dynamic FC tends to slow down and becomes less complex as well as more random with increasing age. These effects appear to be strongly associated with age-related changes in behavioural and cognitive performance. HighlightsO_LIDynamic Functional Connectivity (dFC) at rest and during cognitive task performs a "complex" (anomalous) random walk. C_LIO_LISpeed of dFC slows down with aging. C_LIO_LIResting dFC replaces complexity by randomness with aging. C_LIO_LITask performance correlates with the speed and complexity of dFC. C_LI

neuroscience↗