Locomotor sequence learning in visually guided walking

Reinforcement feedback impairs locomotor adaptation and retention

Introduction

Locomotor adaptation is a motor learning process used to alter spatiotemporal elements of walking that are driven by prediction errors, a discrepancy between the expected and actual outcomes of our actions. Sensory and reward prediction errors are two different types of prediction errors that can facilitate locomotor adaptation. Reward and punishment feedback generate reward prediction errors but have demonstrated mixed effects on upper extremity motor learning, with punishment enhancing adaptation, and reward supporting motor memory. However, an in-depth behavioral analysis of these distinct forms of feedback is sparse in locomotor tasks.

Methods

For this study, three groups of healthy young adults were divided into distinct feedback groups [Supervised, Reward, Punishment] and performed a novel locomotor adaptation task where each participant adapted their knee flexion to 30 degrees greater than baseline, guided by visual supervised or reinforcement feedback (Adaptation). Participants were then asked to recall the new walking pattern without feedback (Retention) and after a washout period with feedback restored (Savings).

Results

We found that all groups learned the adaptation task with external feedback. However, contrary to our initial hypothesis, enhancing sensory feedback with a visual representation of the knee angle (Supervised) accelerated the rate of learning and short-term retention in comparison to monetary reinforcement feedback. Reward and Punishment displayed similar rates of adaptation, short-term retention, and savings, suggesting both types of reinforcement feedback work similarly in locomotor adaptation. Moreover, all feedback enhanced the aftereffect of locomotor task indicating changes to implicit learning.

Discussion

These results demonstrate the multi-faceted nature of reinforcement feedback on locomotor adaptation and demonstrate the possible different neural substrates that underly reward and sensory prediction errors during different motor tasks.

"Walking selectivity" in the occipital place area in 8-year-olds, not 5-year-olds

A recent neuroimaging study in adults found that the occipital place area (OPA)-a cortical region involved in "visually guided navigation" (i.e. moving about the immediately visible environment, avoiding boundaries, and obstacles)-represents visual information about walking, not crawling, suggesting that OPA is late developing, emerging only when children are walking, not beforehand. But when precisely does this "walking selectivity" in OPA emerge-when children first begin to walk in early childhood, or perhaps counterintuitively, much later in childhood, around 8 years of age, when children are adult-like walking? To directly test these two hypotheses, using functional magnetic resonance imaging (fMRI) in two groups of children, 5- and 8-year-olds, we measured the responses in OPA to first-person perspective videos through scenes from a "walking" perspective, as well as three control perspectives ("crawling," "flying," and "scrambled"). We found that the OPA in 8-year-olds-like adults-exhibited walking selectivity (i.e. responding significantly more to the walking videos than to any of the others, and no significant differences across the crawling, flying, and scrambled videos), while the OPA in 5-year-olds exhibited no walking selectively. These findings reveal that OPA undergoes protracted development, with walking selectivity only emerging around 8 years of age.

Reduced corticospinal drive and inflexible temporal adaptation during visually guided walking in older adults

Corticospinal drive during walking is reduced in older adults compared with 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 with younger adults. Healthy young (n = 21; age 18-33 yr) and older adults (n = 20; age 68-80 yr) 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 a similar reduction in error asymmetry compared with younger adults; 2) however, older adults showed reduced adaptation in step time symmetry, despite demonstrating similar adaptation in step length asymmetry compared with younger adults; and 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.NEW & NOTEWORTHY Corticospinal input is essential for visually guided walking, especially when the walking pattern must be modified to accurately step on safe locations. Age-related changes in corticospinal drive are associated with inflexible step time, which necessitates different locomotor adaptation strategies in older adults.

Corticospinal Excitability Quantification During a Visually-Guided Precision Walking Task in Humans: Potential for Neurorehabilitation

The corticospinal tract has been shown to be involved in normal walking in humans. However, its contribution during more challenging locomotor tasks is still unclear. As the corticospinal tract can be a potential target to promote gait recovery after neurological injury, it is of primary importance to quantify its use during human walking. The aims of the current study were to: (1) quantify the effects of precision walking on corticospinal excitability as compared to normal walking; (2) assess if corticospinal modulation is related to task difficulty or participants' performance. Sixteen healthy participants walked on a treadmill during 2 tasks: regular walking (simple task) and stepping onto virtual targets (precision task). Virtual targets appeared randomly at 3 different step lengths: preferred, and ±20%. To assess corticospinal excitability, 25 motor evoked potentials (MEPs) were recorded from the tibialis anterior muscle in each task during walking. Performance for each participant (global success score; % of target hit) and task difficulty related to step length adjustments (success score for each step length) were also calculated. MEP size was larger during the precision task in all participants (mean increase of 93% ± 72%; P < .05) compared to the simple task. There was a correlation between MEP facilitation and individual performance (r = -.64; P < .05), but no difference in MEP size associated with task difficulty (P > .05). In conclusion, corticospinal excitability exhibits a large increase during the precision task. This effect needs to be confirmed in neurological populations to potentially provide a simple and non-invasive approach to increase corticospinal drive during gait rehabilitation.

Don't watch your step: gaze behavior adapts with practice of a target stepping task

Vision plays a vital role in locomotor learning, providing feedback information to correct movement errors, and feedforward information to inform learned movement plans. Gaze behavior, or the distribution of fixation locations, can quantify how visual information is used during the motor learning process. How gaze behavior adapts during motor learning and in response to changing motor performance is poorly understood. This study examines if and how an individual's gaze behavior adapts during a sequence learning, target stepping task. We monitored the gaze behavior of 12 healthy young adults while they walked on a treadmill and attempted to precisely step on moving targets that were separated by variable distances (80%, 100%, and 120% of preferred step length). Participants completed a total of 11 trial blocks of 102 steps each. We hypothesized that both mean fixation distance would increase (participants would look farther ahead), and step error would decrease with experience. Following practice, participants significantly increased their fixation distance (P < 0.001) by 0.27 ± 0.18 steps and decreased their step error (P < 0.001) by 4.0 ± 1.7 cm, supporting our hypothesis. Our results suggest that early in the learning process, participants gaze behavior emphasized gathering visual information necessary for feedback motor control. As motor performance improved with experience, participants shifted their gaze fixation farther ahead placing greater emphasis on the visual information used for feedforward motor control. These findings provide important information about how gaze behavior changes in parallel with improvements in walking performance.NEW & NOTEWORTHY People consistently vary how they use visual information to inform walking. However, what drives this variation and how sampled visual information changes with locomotor learning is not well understood. Here, we find that gaze fixation locations moved farther ahead while step error decreases as participants practice a target stepping task. The results suggest that participants increasingly used a feedforward locomotor control strategy with practice.

Brain activity and upper limb movement analysis in children with Down syndrome undergoing transcranial direct current stimulation combined with virtual reality training: study protocol for a randomized controlled trial

BACKGROUND

Children with Down syndrome have poorer functional and sensory skills compared to children with typical development. Virtual reality (VR) training could help improve these skills. Moreover, transcranial direct current stimulation (tDCS) has achieved promising results in terms of enhancing the effects of physical and sensory therapy by modulating cortical excitability.

METHODS/DESIGN

Two investigations are proposed: (1) an observational study with a convenience sample consisting of children with Down syndrome (group 1-cognitive age of 6 to 12 years according to the Wechsler Abbreviated Scale of Intelligence) and children with typical development 6 to 12 years of age (group 2). Both groups will undergo evaluations on a single day involving a three-dimensional analysis of upper limb movements, an analysis of muscle activity of the biceps and brachial triceps muscles and an analysis of visuospatial and cognitive-motor variables. (2) Analysis of clinical intervention: a pilot study and clinical trial will be conducted involving individuals with Down syndrome (cognitive age of 6 to 12 years according to the Wechsler Abbreviated Scale of Intelligence). The sample will be defined after conducting a pilot study with the same methodology as that to be used in the main study. The participants will be randomly allocated to two groups: An experimental group submitted to anodal tDCS combined with a VR game and a manual motor task and a control group submitted to sham tDCS combined with a VR game and a manual motor task. The training protocol will involve 10 sessions of active or sham tDCS during memory and motor task games. Three 20-min sessions will be held per week for a total of 10 sessions. Evaluations will be performed on three different occasions: pre-intervention, post-intervention (after 10 sessions) and follow-up (1 month after the intervention). Evaluations will consist of analyses of electroencephalographic signals, electromyographic signals of the biceps and triceps brachii, and the three-dimensional reconstruction of the reaching movement. The results will be analyzed statistically with the significance level set at 5% (p ≤ 0.05).

DISCUSSION

The optimization of the results obtained with virtual reality training is believed to be related to the interactive experience with a wide range of activities and scenarios involving multiple sensory channels and the creation of exercises, the intensity of which can be adjusted to the needs of children. Therefore, the proposed study aims to complement the literature with further information on tDCS and VR training considering different variables to provide the scientific community with clinical data on this combination of interventions.

TRIAL REGISTRATION

Brazilian Clinical Trials Registry (REBEC) protocol number RBR-43pk59 registered on 2019 March 27 https://ensaiosclinicos.gov.br/rg/RBR-43pk59 and Human Research Ethics Committee number 3.608.521 approved on 2019 September 30. Protocol version 2021 October 20. Any changes to the protocol will be reported to the committees and approved. Informed consent will be obtained from all participants by the clinical research coordinator and principal investigator.

Visuomotor errors drive step length and step time adaptation during 'virtual' split-belt walking: the effects of reinforcement feedback

Precise foot placement is dependent on changes in spatial and temporal coordination between two legs in response to a perturbation during walking. Here, we used a 'virtual' split-belt adaptation task to examine the effects of reinforcement (reward and punishment) feedback about foot placement on the changes in error, step length and step time asymmetry. Twenty-seven healthy adults (20 ± 2.5 years) walked on a treadmill with continuous feedback of the foot position and stepping targets projected on a screen, defined by a visuomotor gain for each leg. The paradigm consisted of a baseline period (same gain on both legs), visuomotor adaptation period (split: one high = 'fast', one low = 'slow' gain) and post-adaptation period (same gain). Participants were divided into 3 groups: control group received no score, reward group received increasing score for each target hit, and punishment group received decreasing score for each target missed. Re-adaptation was assessed 24 ± 2 h later. During early adaptation, the slow foot undershot and fast foot overshot the stepping target. Foot placement errors were gradually reduced by late adaptation, accompanied by increasing step length asymmetry (fast < slow step length) and step time asymmetry (fast > slow step time). Only the punishment group showed greater error reduction and step length re-adaptation on the next day. The results show that (1) explicit feedback of foot placement alone drives adaptation of both step length and step time asymmetry during virtual split-belt walking, and (2) specifically, step length re-adaptation driven by visuomotor errors may be enhanced by punishment feedback.

Increased central common drive to ankle plantar flexor and dorsiflexor muscles during visually guided gait

When we walk in a challenging environment, we use visual information to modify our gait and place our feet carefully on the ground. Here, we explored how central common drive to ankle muscles changes in relation to visually guided foot placement. Sixteen healthy adults aged 23 ± 5 years participated in the study. Electromyography (EMG) from the Soleus (Sol), medial Gastrocnemius (MG), and the distal and proximal ends of the Tibialis anterior (TA) muscles and electroencephalography (EEG) from Cz were recorded while subjects walked on a motorized treadmill. A visually guided walking task, where subjects received visual feedback of their foot placement on a screen in real-time and were required to place their feet within narrow preset target areas, was compared to normal walking. There was a significant increase in the central common drive estimated by TA-TA and Sol-MG EMG-EMG coherence in beta and gamma frequencies during the visually guided walking compared to normal walking. EEG-TA EMG coherence also increased, but the group average did not reach statistical significance. The results indicate that the corticospinal tract is involved in modifying gait when visually guided placement of the foot is required. These findings are important for our basic understanding of the central control of human bipedal gait and for the design of rehabilitation interventions for gait function following central motor lesions.

Different neural substrates for precision stepping and fast online step adjustments in youth

Humans can navigate through challenging environments (e.g., cluttered or uneven terrains) by modifying their preferred gait pattern (e.g., step length, step width, or speed). Growing behavioral and neuroimaging evidence suggests that the ability to modify preferred step patterns requires the recruitment of cognitive resources. In children, it is argued that prolonged development of complex gait is related to the ongoing development of involved brain regions, but this has not been directly investigated yet. Here, we aimed to elucidate the relationship between structural brain properties and complex gait in youth aged 9–18 years. We used volumetric analyses of cortical grey matter (GM) and whole-brain voxelwise statistical analyses of white matter (WM), and utilized a treadmill-based precision stepping task to investigate complex gait. Moreover, precision stepping was performed on step targets which were either unperturbed or perturbed (i.e., unexpectedly shifting to a new location). Our main findings revealed that larger unperturbed precision step error was associated with decreased WM microstructural organization of tracts that are particularly associated with attentional and visual processing functions. These results strengthen the hypothesis that precision stepping on unperturbed step targets is driven by cortical processes. In contrast, no significant correlations were found between perturbed precision stepping and cortical structures, indicating that other (neural) mechanisms may be more important for this type of stepping.

Performance of a visuomotor walking task in an augmented reality training setting

Visual cues can be used to train walking patterns. Here, we studied the performance and learning capacities of healthy subjects executing a high-precision visuomotor walking task, in an augmented reality training set-up. A beamer was used to project visual stepping targets on the walking surface of an instrumented treadmill. Two speeds were used to manipulate task difficulty. All participants (n = 20) had to change their step length to hit visual stepping targets with a specific part of their foot, while walking on a treadmill over seven consecutive training blocks, each block composed of 100 stepping targets. Distance between stepping targets was varied between short, medium and long steps. Training blocks could either be composed of random stepping targets (no fixed sequence was present in the distance between the stepping targets) or sequenced stepping targets (repeating fixed sequence was present). Random training blocks were used to measure non-specific learning and sequenced training blocks were used to measure sequence-specific learning. Primary outcome measures were performance (% of correct hits), and learning effects (increase in performance over the training blocks: both sequence-specific and non-specific). Secondary outcome measures were the performance and stepping-error in relation to the step length (distance between stepping target). Subjects were able to score 76% and 54% at first try for lower speed (2.3 km/h) and higher speed (3.3 km/h) trials, respectively. Performance scores did not increase over the course of the trials, nor did the subjects show the ability to learn a sequenced walking task. Subjects were better able to hit targets while increasing their step length, compared to shortening it. In conclusion, augmented reality training by use of the current set-up was intuitive for the user. Suboptimal feedback presentation might have limited the learning effects of the subjects.

Visual targeting one step before force plates has no effect on gait parameters in orthopaedic patients during level walking

In clinical gait analysis, it is challenging to acquire usable force plate data for a patient in a limited amount of time. The aim of this study was to compare three measurement protocols, to investigate if any one of them was more time-efficient than the others at collecting kinetic data. Three conditions were compared for 15 orthopaedic patients: 1) approaching the force plate with four steps, 2) approaching the force plate with six steps, and 3) approaching the force plate with four steps while stepping on a target one step before the first force plate. Then, the following characteristics were analysed: the rate of usable force plate steps, the spatio-temporal parameters, the full-body gait kinematics, and the lower body kinetics. For the condition with four steps and targeting, the rate of usable force plate steps was highest: 84% (6.8 usable trials out of 8.1 trials on average per patient). Left hip adduction and rotation, right shoulder flexion, and total left hip power were the gait parameters with statistically significant differences between the four and six step approach. Left cadence, right step time, left thorax lateroflexion, left shoulder abduction, total right knee power, hip rotation, thorax tilt, and head tilt on both sides were statistically different between the four step approach with targeting and without targeting. None of the differences in gait parameters (except for head tilt) were of clinical relevance. Therefore, approaching the force plate with four steps and stepping on a foot-sized target one step prior to stepping on the force plate increases the rate of usable kinetic data.

Seeing the Errors You Feel Enhances Locomotor Performance but Not Learning

In human motor learning, it is thought that the more information we have about our errors, the faster we learn. Here, we show that additional error information can lead to improved motor performance without any concomitant improvement in learning. We studied split-belt treadmill walking that drives people to learn a new gait pattern using sensory prediction errors detected by proprioceptive feedback. When we also provided visual error feedback, participants acquired the new walking pattern far more rapidly and showed accelerated restoration of the normal walking pattern during washout. However, when the visual error feedback was removed during either learning or washout, errors reappeared with performance immediately returning to the level expected based on proprioceptive learning alone. These findings support a model with two mechanisms: a dual-rate adaptation process that learns invariantly from sensory prediction error detected by proprioception and a visual-feedback-dependent process that monitors learning and corrects residual errors but shows no learning itself. We show that our voluntary correction model accurately predicted behavior in multiple situations where visual feedback was used to change acquisition of new walking patterns while the underlying learning was unaffected. The computational and behavioral framework proposed here suggests that parallel learning and error correction systems allow us to rapidly satisfy task demands without necessarily committing to learning, as the relative permanence of learning may be inappropriate or inefficient when facing environments that are liable to change.