Improvement in gait stability in older adults after ten sessions of standing balance training

Leila Alizadehsaravi; Sjoerd M Bruijn; Wouter Muijres; Ruud A J Koster; Jaap H van Dieën

doi:10.1371/journal.pone.0242115

. 2022 Jul 27;17(7):e0242115. doi: 10.1371/journal.pone.0242115

Improvement in gait stability in older adults after ten sessions of standing balance training

Leila Alizadehsaravi ^1,^¤a, Sjoerd M Bruijn ¹, Wouter Muijres ^1,^¤b, Ruud A J Koster ¹, Jaap H van Dieën ^1,^*

Editor: Jeremy P Loenneke²

PMCID: PMC9328559 PMID: 35895709

Abstract

Balance training aims to improve balance and transfer acquired skills to real-life tasks. How older adults adapt gait to different conditions, and whether these adaptations are altered by balance training, remains unclear. We hypothesized that reorganization of modular control of muscle activity is a mechanism underlying adaptation of gait to training and environmental constraints. We investigated the transfer of standing balance training, shown to enhance unipedal balance control, to gait and adaptations in neuromuscular control of gait between normal and narrow-base walking in twenty-two older adults (72.6 ± 4.2 years). At baseline, after one, and after ten training sessions, kinematics and EMG of normal and narrow-base treadmill walking were measured. Gait parameters and temporal activation profiles of five muscle synergies were compared between time-points and gait conditions. Effects of balance training and an interaction between training and gait condition on step width were found, but not on synergies. After ten training sessions step width decreased in narrow-base walking, while step width variability decreased in both conditions. Trunk center of mass displacement and velocity, and the local divergence exponent, were lower in narrow-base compared to normal walking. Activation duration in narrow-base compared to normal walking was shorter for synergies associated with dominant leg weight acceptance and non-dominant leg stance, and longer for the synergy associated with non-dominant heel-strike. Time of peak activation associated with dominant leg stance occurred earlier in narrow-base compared to normal walking, while it was delayed in synergies associated with heel-strikes and non-dominant leg stance. The adaptations of synergies to narrow-base walking may be interpreted as related to more cautious weight transfer to the new stance leg and enhanced control over center of mass movement in the stance phase. The improvement of gait stability due to standing balance training is promising for less mobile older adults.

Introduction

Falls in older adults mostly occur during walking [1,2]. Thus, if effects of standing balance training programs do not transfer to improvements in gait stability, they are unlikely to decrease the number of falls. On the one hand, effects of balance training have been described as task specific, i.e., leading to performance improvement of the task that has been trained but not to improvements in other tasks [3]. On the other hand, transfer to gait stability from solely standing balance training [4–6] is suggested by improved clinical balance scores, gait parameters, and performance on the timed up and go, and other tests. In addition, studies that trained standing balance found reduced falls [7]. Consequently, the existence of transfer from standing balance training to gait, as well as the mechanisms underlying such a transfer, if present, are insufficiently clear. In addition, while improved balance has been reported even after one training session [8], the required time course for transfer is unknown.

Fall prevention training programs aim to improve balance control employing plastsicity of the neuromuscular system. To prevent falls, one needs to be able to adapt gait when facing environmental challenges, such as when forced to walk with a narrow step width. Older adults show more pronounced adaptations to narrow-base walking than young adults [11], possibly because they are more cautious in the presence of balance threats [12]. An interaction between training and stabilizing demands may be expected. On the one hand, increased confidence after training may result in less adaptation to a challenging condition. On the other hand, balance training may enhance the ability to adapt to challenging conditions. Therefore, if transfer of standing balance training to gait occurs, an altered modulation of gait between normal and narrow-base walking might be expected after training, but the direction of change is unpredictable.

Transfer of balance training to gait should become apparent in biomechanical gait parameters related to balance control. Relevant gait parameters would be variability and local dynamic stability of gait, as these were shown to be associated with a history of falls in older adults [9]. In addition, larger trunk mediolateral center of mass (CoM) excursions and velocities are expected to cause an increased fall risk [10] and both these parameters as well as their variability are larger in older than young adults [11]. Moreover, in narrow-base walking, which challenges mediolateral balance control, CoM displacement and CoM velocity are decreased and local dynamic stability is increased indicating enhanced control [11,12]. Step width and its variability reflect active balance control during gait [13]. Mediolateral gait stability is thought to be actively controlled by adjusting foot placement [14], as centre of mass kinematics during the preceding swing phase strongly correlate with foot position in the next stance phase [15,16]. The strong coupling between centre of mass kinematics and foot placement is found to decrease in conditions in which gait is stabilized, such as in external lateral stabilization, increasing confidence that lateral gait stability is indeed controlled by foot placement adjustment [17]. This decoupling coincides with a decrease in step width and step width variability [18]. Moreover, increased step width and step width variability have been found in older compared to young adults and in fallers compared to non-fallers [19–21]. Consequently, step width and step width variability are considered to be proxies of the quality of active mediolateral control of gait stability using foot placement adjustments and may therefore be useful to evaluate the effect of balance training. Transfer of balance training effects to gait could thus be reflected in increased local dynamic stability, decreased trunk CoM displacement and velocity, decreased trunk CoM displacement variability, and decreased step width and step width variability.

Motor tasks are thought to be performed through modular control of muscles, reflected in so-called muscle synergies [22]. A potential mechanism for standing balance training to affect gait is through shared muscle synergies between standing balance and gait. Human gait has been reported to be controlled by four to eight of such muscle synergies [23–25], the combination of which shapes the overall motor output [26,27]. Standing balance is also reported to be controlled through a limited number of muscle synergies [25], and it has shared synergies with gait [25]. These shared synergies were identified as subserving postural ability. Hence, standing balance training could change synergies involved in standing balance and, by extension, the shared synergies involved in gait. The overlap in modular control provides a mechanism through which standing balance training can influence gait stability.

Muscle synergies consist of time-dependent patterns (activation profiles) and time-independent components (muscle weightings). Motor adaptation is assumed to result from altering either of these in response to task and environmental demands [28,29]. Furthermore, due to aging and related changes in sensory and motor organs, adapted synergies are likely required to maintain motor performance [30,31]. Synergy analysis of gait and gait related-tasks revealed a less efficient modular control in older compared to young adults [30–32]. To adapt to environmental challenges, similarly widened activation profiles appear to be used [29,32], indicating a more robust control and suggesting that the age effects described reflect a more cautious control. Balance training might alter synergies in gait and the adaptation of these synergies to task demands, as has been shown in a comparison between expert and novice dancers [33,34].

Previously, we have shown that training of standing balance control improved balance robustness. This was defined as the time to balance loss in unipedal standing on a platform with decreasing rotational stiffness around a sagittal axis. Balance robustness increased by 33% already after one training session, with no further improvement after ten sessions. Balance performance, defined as absolute mediolateral center of mass velocity, was improved by 19% in perturbed unipedal standing after one session and by 18% in unperturbed unipedal standing after ten sessions [35]. In the current study, we aimed to investigate the transfer of effects of standing balance training to normal and narrow-base walking in older adults, as well as the adaptation of older adults to narrow-base walking. The modulation of balance control between two conditions aimed to test adaptability of balance control to environmental constraints and effects of training were studied to analyse the plasticity of balance control. To this end, we evaluated normal walking and narrow-base walking on a virtual beam, both on a treadmill, before training and after one and ten training sessions. We focused on mediolateral balance control, because larger mediolateral instability has been shown to be associated with falls in older adults [36,37], narrow-base walking challenges mediolateral stability and also standing balance was evaluated in the mediolateral direction. To assess performance in narrow-base walking, we calculated foot placement errors, the percentage of steps in the beam, step width, and the step width variability [38]. For both gait conditions, we calculated trunk CoM displacement and trunk CoM displacement variability, trunk CoM velocity, and the local divergence exponent of trunk movement as measures of gait stability.

We extracted muscle synergies to characterize effects of training on the neuromuscular control of gait and on adaptations to narrow-base walking. We focused on changes in timing of muscle activation by assessing the synergy activation profiles’ full width at half maximum (FWHM), reflecting activation duration, and Center of Activation (CoA), reflecting time of peak activation. We hypothesized that adaptations to narrow-base walking would be reflected in enhanced gait stability [39] and in more prolonged muscle activation. Furthermore, we hypothesized that training effects would transfer to gait as reflected in improved gait stability and narrow-base walking performance and activation profiles with less prolonged muscle activation. These effects were expected to be more pronounced in narrow-base walking compared to normal walking and more pronounced after ten training sessions compared to a single session.

We would like to emphasize that the outcome variables step width and step width variability were added to the analysis based on reviewer feedback. In contrast to other variables, these showed a significant effect of training. Even though step width and step width variability seem to be sensitive to quality of balance control in gait, step width measures are not necessarily affected by postural balance training. Postural balance is controlled using torques around the ankle and hip of the standing leg [40,41], but does not include stepping (i.e. foot placement). However, as in postural balance, the stance leg is used to regulate stabilizing torques in the stance phase gait [42]. This stance leg control co-determines the control of foot placement [43] and therefore reduced step width and step width variability are likely results of improved control by the stance leg. These predictions on step width and step width variability were conceived, based on reviewer comments, after the data were processed and analyzed. These predictions are thus explorative and aim to drive future research [44,45] and we ask the reader to consider the limitations of the evidence provided by these variables. While this result is in line with our primary hypothesis and indicates transfer of effects of standing balance training to gait, in a strict sense it cannot be considered a planned analysis.

Methods

The methods described here partially overlap with those described in a previous paper [35], as data were obtained in the same cohort.

Participants

Twenty-two older (72.6 ± 4.2 years old; mean ± SD, 11 females) healthy volunteers, without a history of falls in the preceding year, participated in this study. The required sample size was estimated at twenty-two based on power analysis for an F test of a repeated measures ANOVA, assuming a Cohen’s f of 0.44 (based on meta-analysis of the effect of standing balance training on steady-state balance [46]) and a correlation among repeated measures of 0.6 (β = 0.8, G * power 3.1.9.2, Düsseldorf, Germany), comparable to similar studies [47,48]. Participants were recruited through a radio announcement, by contacting older adults who had previously participated in our research, and via flyers and information meetings. Individuals with obesity (BMI > 30), cognitive impairment (MMSE < 24), peripheral neuropathy, a history of neurological or orthopaedic impairment, use of medication that may negatively affect balance, inability to walk for 3 minutes without aid, and performing sports with balance training as an explicit component (e.g., Yoga or Pilates) were excluded. All participants provided written informed consent before participation and the procedures were approved by the ethical review board of the Faculty of Behavioural & Movement Sciences, VU Amsterdam (VCWE-2018-171).

Experimental procedures

Participants completed an initial measurement to determine baseline values (Pre), and a single-session individual balance training (30-minutes), a second measurement (Post1) to compare to baseline to assess single-session training effects. The program was continued with a 3-week balance training (9 sessions x 45 minutes training), and a third measurement (Post2) to compare to baseline to assess 10-session training effects (Fig 1).

The measurements consisted of one experimental condition on a robot-controlled platform (standing balance) and two experimental conditions performed on a treadmill: virtual-beam walking (Fig 2) and normal walking.

The training sessions consisted of exercises solely focused on unipedal balancing with blocks of 40–60 seconds exercises in which balance was challenged by different surface conditions, static vs dynamic conditions, self-perturbations and external perturbations while catching a ball in a dual tasking exercise (e.g. catching, throwing and passing a ball) [49]. Participants performed the exercises in a group (except for the first, individual session) and were always under supervision of the physiotherapist in our research team (for details see S1 File).

Instrumentation and data acquisition

Balance robustness and performance were evaluated using a custom-made balance platform controlled by a robot arm (HapticMaster, Motek, Amsterdam, the Netherlands) and results were reported previously [35]. To quantify transfer to gait, participants were instructed to walk for 4.5 minutes on a treadmill with an embedded force plate. For estimating the local divergence exponent, a minimum of 150 steps is recommended [50]. We expected that a total duration of 4.5 minutes was needed to reach that number. To avoid effects of gait speed on outcome measures, this was kept constant at 3.5 km/h for all participants [51].

For safety reasons, handrails were installed on either side of the treadmill, and an emergency stop button was placed within easy reach (MotekForcelink, Amsterdam, the Netherlands). We assessed walking in two conditions, normal walking and narrow-base walking, in a randomized order, with a minimum of two minutes seated rest in between conditions. A narrow-base walking paradigm was chosen because narrow-base walking has been shown to challenge mediolateral stability in older adults [12,34]. In this condition, participants were instructed to place their entire foot inside a green light-beam path (12 cm width) projected in the middle of the treadmill (Bonte Technology/ForceLink, Culemborg, The Netherlands) as accurately as possible [38]. Participants were acquainted with the experimental setup to minimize habituation effects. For familiarization, participants performed 30 seconds of narrow-base walking before the measurement.

Kinematic data were obtained by two Optotrak 3020 camera arrays sampling at 50 Hz (Northern Digital, Waterloo, Canada). Ten active marker clusters (3 markers each) were placed on the posterior surface of the thorax (1), pelvis (1), arms (2), calves (4), and feet (2) (Fig 2). Positions of anatomical landmarks were digitized by a 4-marker probe and a full-body 3D-kinematics model of the participant was formed relating clusters to the anatomical landmarks [52]. The position of the foot segments was obtained through cluster markers on both feet, digitizing the medial and lateral aspects of the calcaneus, and the heads of metatarsals one and five [38]. Additionally, to allow assessment of performance in narrow-base walking, position and orientation of the projected beam were determined by digitizing the four outer bounds of the beam on the treadmill.

Surface electromyography (EMG) data were recorded from 11 muscles; 5 unilateral muscles of the dominant leg: tibialis anterior (TAD), vastus lateralis (VLD), lateral gastrocnemius (GLD), soleus (SOD), peroneus longus (PLD), and 6 bilateral muscles: rectus femoris (RFD, RFN), biceps femoris (BFD, BFN), and gluteus medius (GMD, GMN). These muscles were selected based on a previous study that showed changes in walking synergies due to long-term training [34]. The ankle muscles were chosen for their key role in postural stability [53–55], and the gluteus medius for its role in mediolateral stability [56].

Bipolar electrodes were placed in accordance with SENIAM recommendations [57]. EMG data were sampled at a rate of 2000 Hz and amplified using a 16-channel TMSi Porti system (TMSi, Twente, The Netherlands). The dominant leg was the preferred stance leg for unipedal stance. The preferred stance leg was reported by the participant prior to the experiment and confirmed by the experimenter by asking the participant to kick an imaginary soccer ball. The supporting leg was considered the preferred stance leg [35]. Focus was on this leg, because we extensively assessed unipedal balance control on this leg as reported earlier [35].

Data analysis

Gait events

The first 30 seconds of all gait trials were removed, to avoid habituation effects. Heel-strikes were detected through a peak detection algorithm based on the center of pressure [58]. This algorithm proved to be precise when the center of pressure moved in a butterfly pattern. However, for narrow-base walking, the feet share a common area in the middle of the treadmill. Therefore, identifying which leg touched the surface was problematic, as the butterfly pattern was not formed in narrow-base walking. Hence, heel-strikes were detected based on the center of pressure peak detection, but the associated leg was identified based on kinematic data of the foot marker. 160 strides per participant per condition were used to calculate all gait variables (i.e., stability variables and muscle synergies).

Gait stability

To quantify gait performance, we evaluated foot placement, step width, and CoM behaviour. Evaluation of foot placement was only performed for narrow-base walking. We assessed foot placement error, determined as the mean mediolateral distance of the furthest edge of the foot from the edge of the beam. If the entire foot was within the beam the error was set to zero. The foot placement error was used because it is by design a direct quantification of task performance. Hence, we expected it to be the most sensitive measure to changes in task performance. In addition, the percentage of steps inside the beam was defined as the number of the steps in which the whole foot was placed within the beam. To assess foot placement in both walking conditions, step width and its variability were determined. These measures were computed as the mean and standard deviation of the distance between the mediolateral position of the left and right foot over 160 strides. Additionally, the trajectory of the CoM of the trunk was estimated from the mediolateral movement of the trunk markers [15,59]. From this, we calculated mean and standard deviation of the peak-to-peak mediolateral trunk CoM displacement and mean of CoM velocity per stride. Trunk CoM displacement, displacement variability, and velocity are commonly used measures to express postural stability. However, particularly for use during gait these measures are not entirely undisputed as they are not necessarily minimized for task execution. Hence, local dynamic stability was also evaluated using the LDE, as evidence suggests its validity in the context of gait stability and falling [50]. Calculation of the LDE was based on Rosenstein’s algorithm [60,61]. We used the time normalized time-series (i.e., 160 strides of data were time normalized to 16000 samples, preserving between stride variability) of trunk CoM velocity to reconstruct a state space with 5 embedding dimensions at 10 samples time delay [59]. The divergence for each point and its nearest neighbour was calculated and the LDE was determined by a linear fit over half a stride to the averaged log transformed divergence.

Muscle synergies

EMG data were high-pass (50 Hz, bidirectional, 4th order Butterworth) [29] and notch filtered (50 Hz and its harmonics up to the Nyquist frequency, 1 Hz bandwidth, bidirectional, 1st order Butterworth). The filtered data were Hilbert transformed, rectified and low-pass filtered (20 Hz, bidirectional, 2nd order Butterworth). Each channel was normalized to the maximum activation obtained for an individual per measurement point per trial. Synergies were extracted from 11 muscles using non-negative matrix factorization based on Lee and Seung’s multiplicative update rule [62] with 50 repetitions with a maximum of 1000 iterations to update the components and at a tolerance of 10⁻⁶. Five synergies were extracted from the whole dataset to account for a minimum of 85% of the variance in the EMG data (Fig 7). It has been shown that perturbations during walking change the temporal activation profiles as compared to normal walking, while muscle weightings are preserved [63]. Therefore, in the current study we fixed muscle weightings between conditions and time-points to be able to identify changes in the temporal activation. These muscle weightings were extracted from the concatenated EMG data of both conditions at all time-points. This allowed for objective comparison of synergy activation profiles between normal and narrow-base walking and between time-points. The time-normalized EMG data of the muscles E_{11 x (3 x 2 x 100 x 160),} was factorized to two matrices: time-invariant muscle weightings, W_{11 x 5}, and temporal activation profiles of the factorization, A_{5 x (3 x 2 x 100 x 160)}, where 11 was the number of muscles, 3 the number of time-points, 2 the number of conditions, 100 the number of samples in each stride,160 the number of strides, and 5 the number of synergies.

Fig 7 — Muscles monitored unilaterally on the dominant side (D): tibialis anterior (TA), vastus lateralis (VL), lateral gastrocnemius (GLD, soleus (SO), and peroneus longus (PLD). Muscle collected on the dominant (D) and non-dominant side (N): rectus femoris (RFD, RFN), biceps femoris (BFD, BFN), and gluteus medius (GMD, GMN).

To compare activation profiles, we evaluated the FWHM per stride for each activation profile. The FWHM is defined as the number of data points above half of the maximum of the activation profile, after subtracting the minimum activation [64]. In addition, we evaluated the CoA (indicating the center of the distribution of activation timing within a gait cycle) per stride, defined as the angle of the vector that points to the center of mass in the activation profile transformed to polar coordinates [65]. The FWHM metric reflects the duration of activation but is naïve of timing of activation. The CoA metric reflects the timing of activation but is naïve of duration of activation. FWHM and CoA were averaged over 160 strides per participant per condition. For CoA data, circular averaging was used.

Statistics

Shapiro-Wilk’s test was performed on all measures. In case of non-normally distributed data, a log transformation was performed. One-way repeated measures ANOVA were performed to investigate the effects of Training (Pre, Post1, Post2) on foot placement errors, and % of steps within the beam. Post hoc comparisons (paired sample t-tests), with Holm’s correction for multiple comparisons were performed to investigate the effect of one and ten training sessions (Pre vs Post1 and Pre vs Post2, respectively).

Two-way repeated-measures ANOVAs were used to identify effects of Training (Pre, Post1, Post2) and Condition (normal and narrow-base walking), as well as their interaction on step width; step width variability; FWHM; and on trunk kinematics: CoM displacement, CoM displacement variability, CoM velocity, and LDE. When the assumption of sphericity was violated, the Greenhouse-Geisser correction was used. In case of a significant effect of Training, or an interaction of Training x Condition, post hoc tests with Holm’s correction for multiple comparisons were performed. To identify effects of Training and Condition (normal and narrow-base walking), as well as their interaction, on CoA a parametric two-way ANOVA for circular data was used using the Circular Statistic MATLAB toolbox [66]. In all statistical analyses α = 0.05 was used.

Results

One participant was not able to perform the treadmill walking trials for the full duration and data for this participant were excluded.

Gait performance

Performance in narrow-base walking, as reflected in foot placement errors, did not change significantly with Training (F_{2,40 =} 1.479, p ₌ 0.242; Fig 3a). Also, the percentage of the steps within the beam did not change significantly with Training (F_{2,40 =} 2.934, p ₌ 0.065; Fig 3b). However, Training had a significant effect on step width (F_{1.57,31.36 =} 7.121, p ₌ 0.005; Fig 4a). Moreover, the interaction of Training and Condition significantly affected step width (F_{1,20 =} 261.075, p < 0.001). Post hoc analyses showed that step width decreased after ten sessions and only in narrow-base walking (t = 4.062, p < 0.001), and step width was smaller in narrow-base compared to normal walking at all time-points (t = -12.302, p < 0.001; t = -11.763, p < 0.001; t = -14.386, p < 0.001; for Pre, Post1, and Post2, respectively).Training also had a significant effect on step width variability (F_{2,40 =} 8.724, p < 0.001; Fig 4b). Post hoc testing showed that step width variability had not significantly changed after one session (t = 0.898, p = 0.375), but was decreased after ten sessions (t = 3.982, p < 0.001). There was no significant effect of Condition or interaction of Training and Condition on step width variability (F_{1,20 =} 0.576, p = 0.457; F_{2,40 =} 1.994, p = 0.149).

Fig 4 — (a) Step width (b) Step width variability in narrow-base and normal walking at time-points Pre, Post1, and Post2. Thin lines represent individual subject data. Thick horizontal lines indicate means over subjects. Black, normal walking; red filled-dot, narrow-base walking.

Training did not significantly affect trunk CoM displacement, displacement variability, and velocity (F_{2,40 =} 2.729, p = 0.082; F_{2,40 =} 0.469, p = 0.628; F_{2,40 =} 2.024, p = 0.145). Condition significantly affected all three variables, with lower displacement and velocity (F_{1,20 =} 96.007, p < 0.001; F_{1,20 =} 168.26, p < 0.001; respectively, Fig 5a & 5b), but larger CoM displacement variability (F_{1,20 =} 4.678, p = 0.042, Fig 5c), in narrow-base compared to normal walking. No significant interactions of Training x Condition were found (p > 0.05). Training did not significantly affect LDE (F_{2,40 =} 0.205, p = 0.814), but Condition did, with lower values (indicating improved stability) in narrow-base compared to normal walking (F_{1,20 =} 26.223, p < 0.001; Fig 6). No significant interaction of Training x Condition was found for the LDE (F_{1.3,24.699 =} 3.112, p = 0.078).

Fig 5 — (a) Mediolateral center of mass displacement and (b) variability, and (c) center of mass velocity in narrow-base and normal walking at time-points Pre, Post1, and Post2. Thin lines represent individual subject data. Thick horizontal lines indicate means over subjects. Black, normal walking; red filled-dot, narrow-base walking.

Fig 6 — Thin lines represent individual subject data. Thick horizontal lines indicate means over subjects. Black, normal walking; red filled-dot, narrow-base walking.

Muscle synergies

Five muscle synergies were extracted with a fixed muscle weighting matrix W (Fig 7) and activation profiles per individual per condition and time-point (Fig 8). This accounted for 87 ± 2% of the variance in the EMG data. Based on muscle weightings and activation profiles, the first synergy was predominantly active in the stance phase of the dominant leg, with major involvement of soleus and gastrocnemius lateralis. The second synergy was active during the weight acceptance phase of the dominant leg, where the quadriceps (vastus lateralis, rectus femoris) muscles were most engaged. The third synergy resembled partial mirror images of synergies 1 and 2 for the non-dominant leg but differed, because only a subset of muscles was measured. It was mainly active in the stance phase of the non-dominant leg, with major involvement of the gluteus medius and rectus femoris. It lacks muscle activation during the push-off (represented in synergy 1), because lower leg muscles were not measured and represented thigh muscle activity related to weight acceptance (represented in synergy 2). The fourth synergy was activated prior to dominant leg heel-strike with engagement mostly of the biceps femoris of the dominant leg. Finally, the fifth synergy appeared to be the mirror image of the fourth synergy, with pronounced engagement of the biceps femoris of the non-dominant leg.

Fig 8 — The x-axis in the Cartesian coordinates represents one gait cycle. One gait cycle in polar coordinate is [0, 2π]. Black, normal walking; red, narrow-base walking.

FWHM

None of the FWHMs were significantly affected by Training. FWHMs were found to be smaller in narrow-base compared to normal walking in the synergies associated with weight acceptance of the dominant leg (synergy 2, F_{1,20 =} 92.86, p < 0.001) and the stance phase of the non-dominant leg (synergy 3, F_{1,20 =} 17.06, p < 0.001; Fig 9). In contrast, FWHM of synergies associated with heel-strike in narrow-base compared to normal walking was only greater for the non-dominant leg (synergy 5, F_{1,20 =} 8.603, p = 0.008) and not the dominant leg (synergy 4, F_{1,20 =} 2.198, p = 0.153; Fig 9). In none of the synergies, FWHM was significantly affected by the interaction of Training x Condition (p > 0.05).

Fig 9 — Thin lines represent individual subject data. Thick horizontal lines indicate means over subjects. Black, normal walking; red, narrow-base walking.

CoA

None of the CoAs were significantly affected by Training (p > 0.05). CoA of synergy 1, associated with dominant leg stance, occurred significantly earlier in narrow-base compared to normal walking (F_{1,20 =} 6.005, p = 0.015; Fig 9). CoAs of synergy 3 (F_{1,20 =} 9.832, p = 0.002), associated with non-dominant stance leg, and synergies 4 (F_{1,20 =} 22.109, p < 0.001) and 5 (F_{1,20 =} 18.308, p < 0.001), associated with heel-strike, were delayed in narrow-base compared to normal walking (Fig 9).

Discussion

We studied whether effects of standing balance training transferred to gait in older adults. Additionally, we investigated adaptations in neuromuscular control of gait in older adults between normal and narrow-base walking, and the effect of one and ten sessions of standing balance training on this. We expected the neural mechanisms underlying balance in standing and walking to be adaptable (modulated between conditions) and plastic (modified by training). We also expected transfer of training effects to be most pronounced in the narrow-base condition, given its challenging nature, and higher resemblance to unipedal standing balance.

We have previously found improvements in robustness of standing balance already after one session and improvements in performance of standing balance after the first session and after ten sessions of standing balance training [35]. Here, we found a decreased step width in narrow-base walking and decreased step width variability in narrow-base and normal walking after ten sessions of training. However, we found no changes in gait synergies after training. In line with our expectations, we found increased activation duration in synergies during narrow-base walking, which suggests enhanced stability in narrow-base walking compared to normal walking.

Transfer of training effects

As previously reported, we found improved balance robustness and performance after standing balance training [35]. In this study, after ten sessions we found transfer from standing balance training to gait. This transfer effect was manifested in a decreased step width in narrow-base walking and a decreased step width variability in both gait conditions.

Our results showed that although the participants did not better comply with the task instruction (to step within the beam), they did use narrower steps in line with those instructions. Foot placement errors did not show improvement, despite an ample scope for improvement. When interpreting the 1.5 cm mean step error, one should consider that the width of the beam was 12 cm, while the average foot width is about 10 cm [67]. This means that participants, on average, had a 2 cm margin to achieve a zero error in foot placement. Additionally, the percentages of steps inside the beam were 27.6% (SD 17.8%) pre-training and 33.3% (SD 20.1%) post-training. Although these differences were not significant (p = 0.11), the numbers show an increasing trend and suggest that there is no ceiling effect. Also, the best-recorded performance was 67% of steps within the beam, so scores much higher than the mean were achievable. The significant effect of training on step width during narrow-base walking indicates that performance improved in line with the instructions, but the improvement was not large enough to be detected by foot placement error and steps outside the beam, probably as both variables discard information on foot placement within the beam.

A reduced step width was observed only in narrow-base and not in normal walking, which might suggest that the transfer occurs more readily to a task similar to the trained task. The challenge in narrow-base walking is in nature similar to unipedal standing, since foot placement control for gait stability [14] is constrained and the CoM needs to be controlled relative to a narrow base of support during the single support phase. Nevertheless, normal walking also involves some degree of control of center of mass movement during single support [43], and a decreased step width variability in both conditions indicates transfer to this task as well. Foot placement variability has been associated with fall risk [68–70]. So, this decrease can be interpreted as a positive training outcome.

Other kinematic gait parameters were not affected by training. This might suggest that these parameters are insensitive to training. However, they were sensitive to differences between narrow-base and normal gait conditions. Previously, improved gait parameters were reported after 5 to 12 weeks of balance training [4,71]. Therefore, a longer duration of training might have led to more improvements in mediolateral gait stability, but more studies are needed to confirm this speculation.

Even though we found evidence of transfer, gait synergies were not affected by training. Synergy metrics may be insensitive, because subtle variations in the activation of some muscles, for example through improved feedback control, underlie improved balance. Alternatively, changes may have occurred in muscles that we did not include in our measurements. Young adults with years of experience in balance training in ballet showed different motor modules during narrow-base walking compared to novice ballet dancers [25], indicating that long-term effects of training may be apparent in muscle synergies. Note, however, that these ballet dancers’ training experience most certainly included gait-related tasks [25]. In addition, the dancers had trained for at least ten years.

The lack of changes in synergies, despite modulated synergies between narrow-base and normal walking, may suggest that FWHM and COA are not sensitive to differences that occur with training. This could be due to several factors. One potential factor could be the decreased robustness of the synergies analysis when EMG data collection is performed on different days. This could hamper comparison between time-points, especially if EMG is not normalized to maximum voluntary activation. We chose not to measure the maximum voluntary activation of 11 muscles, to avoid that muscle fatigue would cause inadequate balance control during the main experiment. Given that participants were healthy older adults and the training comprised of only ten sessions spread over three weeks, we did not expect any changes in the number of synergies pre- and post-training. Changes in the number of synergies are observed typically only in individuals with severe movement disorders [72]. Further investigation is required to identify the underlying mechanisms of inter-task transfer.

The training program used excluded all exercises that directly targeted gait stability, solely focusing on the transfer of balance skill to gait as a result of standing balance training. It used unstable surfaces to make sure that the training was challenging and challenges were incremented according to guidelines in literature [73]. The current indication of transfer effects offers a valuable clue for clinicians and future studies, as exercises of standing balance form a substantial component of many training programs used in practice and showed potential to be used in training of older adults with limited mobility (to start with stationary exercises and transfer the training effects to walking). Our results highlight the necessity to optimize training methods and duration. More significant improvements in balance skills may be required to transfer acquired skills to daily-life tasks.

Adaptation to narrow-base walking

We expected that neuromuscular control in older adults would be sufficiently plastic to adapt to narrow-base walking. In line with literature [11], our participants appeared to control CoM movements more tightly during narrow-base walking than during normal walking, as reflected in a lower CoM displacement and velocity, lower step width, and higher local dynamic stability (i.e. lower LDE). Furthermore, again in line with literature [11], variability of CoM displacement was larger in narrow-base walking. This larger variability might reflect on-line corrections of the CoM trajectory to match it to the constrained foot placement. Confronted with a narrower base, older adults reduced mediolateral CoM displacement and velocity more than young adults [11]. The LDE was also lower in narrow-base walking compared to normal, implying higher local dynamic stability, or in other words a faster attenuation of perturbation effects. This effect has previously been reported for young adults [74] but apparently older adults manage to achieve a qualitatively similar adaptation. Overall, the decreased center of mass displacement and velocity and the faster attenuation of perturbation effects, reflected in the higher LDE, would facilitate dealing with the challenge of this condition.

The mechanical changes observed in narrow-base walking were accompanied by changes in the neuromuscular control of gait. An increase of the FWHM has been suggested to increase the ability to deal with mechanical perturbations of the gait pattern [29]. We did find differences in the FWHM of the activation profiles between narrow-base walking and normal walking. However, during narrow-base walking our participants only increased the FWHM of the activation profile associated with the non-dominant leg heel-strike (synergy 5), although a similar tendency could be observed for the dominant leg (synergy 4). These adaptations of the activation profiles may reflect enhanced control over foot placement or preparation for weight acceptance on the new stance leg in the narrow-base condition. In contrast, participants shortened the FWHM of the activation profiles associated with the stance phase of the non-dominant leg and weight acceptance of the dominant leg. These synergies share muscle activation related to weight acceptance and the change in the activation profiles is mainly visible in a slower build-up of muscle activity (Fig 8). This may reflect a slower weight acceptance by the new support leg, possibly related to the lower activation peak during push-off observable in synergy 1.

Also, the CoA of the activation profiles was different between normal and narrow-base walking. Narrow-base walking coincided with an earlier CoA of the activation profile associated with dominant leg stance (synergy 1) and delayed CoAs of the activation profile associated with dominant and non-dominant leg heel-strikes (synergies 4 and 5). Earlier CoA in the dominant leg stance phase appears to be a consequence of the reduction in activation during the second peak of the activation profile (Fig 8). This reduction in activation would reflect a decrease in muscle activity related to push-off and possibly reflects a more cautious gait. The earlier CoA of the activation profile associated with heel-strike reflects a more sustained activation following a slower build-up (Fig 8). Again, this may be related to active control over CoM movement during the stance phase or a more cautious weight acceptance. In the supplementary material we reported the Falls Efficacy Scale International (FES-I) results measured at Pre, Post2, and retention (2 weeks after the last training session) time-points. FES-I decreased between Pre and retention time-points, and between Pre and Post2, however insignificant, but the trend suggested increasing confidence, which may have contributed to a less cautious behaviour (S1 File). The active control of CoM movement is supported by the fact that muscles that would contribute to mediolateral control, specifically tibialis anterior, peroneus longus, and gluteus medius are part of these synergies. To check that changes in CoA and FWHM of the activation profiles were not due to changes in the duration of gait phases, we assessed single support and double support times as percentages of the stride times, and no effects of Condition were found.

Limitations

There are several limitations in our study that need to be addressed. First, we did not include a young group to investigate the influence of aging on the transfer effect of standing balance training to gait. Second, we did not include a control group to identify the normal variation over time, independent of the training program. Third, it was hypothesized that transfer would occur through shared synergies. In the current study muscle weightings of the synergy analysis were extracted based solely on gait data rather than a combination with standing balance data. Therefore, it is less likely that any one of these synergies is optimally defined as being a shared synergy between the tasks. Moreover, we included only 11 muscles in the synergy analysis not representative of whole-body activity. It could be that the mechanical changes observed would be more closely associated with changes in control of muscles in the upper body [75]. We extracted five synergies to describe leg muscle activity across both narrow-base and normal walking, together accounting for 87% of the variation in muscle activity. In spite of differences in muscles measured, participant age, and walking conditions between studies, (the number and the general grouping and activation of) these synergies resembled results reported previously [26,76–80]. We kept the muscle weightings in these synergies constant over conditions and time-points to investigate variations in the activation profile. We repeated the analysis for six synergies to achieve a variance accounted for over 90%. However, the results were not any different.

Conclusions

In conclusion, after ten sessions of standing balance training, older adults decreased their step width in narrow-base walking and decreased the step width variability in narrow-base and normal. This suggests a transfer of balance skills from standing to walking after ten training sessions. However, there was no evidence in adaptations of neuromuscular control due to balance training associated with changes in control of mediolateral gait stability. In addition, older adults adapted mediolateral CoM kinematics and the step width between normal and narrow-base walking, and this was associated with changes in synergies governing the activation of leg muscles. Our results suggest that in older population the neural mechanisms are still adaptable and acquired skills can be transferred from standing to walking.

Supporting information

S1 File

(DOCX)

Click here for additional data file.^{(298.8KB, docx)}

S2 File

(MAT)

Click here for additional data file.^{(243.8KB, mat)}

S3 File

(CSV)

Click here for additional data file.^{(601B, csv)}

S4 File

(CSV)

Click here for additional data file.^{(7.6KB, csv)}

S5 File

(MAT)

Click here for additional data file.^{(242.7KB, mat)}

Acknowledgments

The research team would like to thank the individuals who participated in the experiment.

Data Availability

All relevant data are within the article and its Supporting information files.

Funding Statement

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 721577. SMB was funded by a VIDI grant (016.Vidi.178.014) from the Dutch Organization for Scientific Research (NWO) and RAJK was supported by a grant of the European Research Council (grant No. 715945). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Rubenstein LZ, Josephson KR, Robbins AS. Falls in the nursing home. Ann Intern Med. 1994;121: 442–451. doi: 10.7326/0003-4819-121-6-199409150-00009 [DOI] [PubMed] [Google Scholar]
2.Berg WP, Alessio HM, Mills EM, Tong C. Circumstances and consequences of falls in independent community-dwelling older adults. Age Ageing. 1997;26: 261–268. doi: 10.1093/ageing/26.4.261 [DOI] [PubMed] [Google Scholar]
3.Giboin L- S, Gruber M, Kramer A. Task-specificity of balance training. Hum Mov Sci. 2015;44: 22–31. doi: 10.1016/j.humov.2015.08.012 [DOI] [PubMed] [Google Scholar]
4.Hirase T, Inokuchi S, Matsusaka N, Okita M. Effects of a Balance Training Program Using a Foam Rubber Pad in Community-Based Older Adults: A Randomized Controlled Trial. J Geriatr Phys Ther. 2015;38: 62–70. doi: 10.1519/JPT.0000000000000023 [DOI] [PubMed] [Google Scholar]
5.Kümmel J, Kramer A, Giboin LS, Gruber M. Specificity of Balance Training in Healthy Individuals: A Systematic Review and Meta-Analysis. Sport Med. 2016;46: 1261–1271. doi: 10.1007/s40279-016-0515-z [DOI] [PubMed] [Google Scholar]
6.Nam HC, Cha HG, Kim MK. The effects of exercising on an unstable surface on the gait and balance ability of normal adults. J Phys Ther Sci. 2016;28: 2102–2104. doi: 10.1589/jpts.28.2102 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Nyman SR. Tai Chi for the Prevention of Falls Among Older Adults: A Critical Analysis of the Evidence. J Aging Phys Act. 2021;29: 343–352. doi: 10.1123/japa.2020-0155 [DOI] [PubMed] [Google Scholar]
8.van Dieën JH, van Leeuwen M, Faber GS. Learning to balance on one leg: motor strategy and sensory weighting. Am J Physiol Circ Physiol. 2015. doi: 10.1152/jn.00434.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Toebes MJP, Hoozemans MJM, Furrer R, Dekker J, Van Dieën JH. Local dynamic stability and variability of gait are associated with fall history in elderly subjects. Gait Posture. 2012;36: 527–531. doi: 10.1016/j.gaitpost.2012.05.016 [DOI] [PubMed] [Google Scholar]
10.Hof AL, Gazendam MGJ, Sinke WE. The condition for dynamic stability. J Biomech. 2005;38: 1–8. doi: 10.1016/j.jbiomech.2004.03.025 [DOI] [PubMed] [Google Scholar]
11.Arvin M, Mazaheri M, Hoozemans MJM, Pijnappels M, Burger BJ, Verschueren SMP, et al. Effects of narrow base gait on mediolateral balance control in young and older adults. J Biomech. 2016;49: 1264–1267. doi: 10.1016/j.jbiomech.2016.03.011 [DOI] [PubMed] [Google Scholar]
12.Manuscript A. NIH Public Access. 2013;35: 472–477. [Google Scholar]
13.Reimold NK, Knapp HA, Henderson RE, Wilson L, Chesnutt AN, Dean JC. Altered active control of step width in response to mediolateral leg perturbations while walking. Sci Rep. 2020;10: 1–15. doi: 10.1038/s41598-020-69052-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Bruijn SM, Van Dieën JH. Control of human gait stability through foot placement. J R Soc Interface. 2018;15. doi: 10.1098/rsif.2017.0816 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Hurt CP, Rosenblatt N, Crenshaw JR, Grabiner MD. Variation in trunk kinematics influences variation in step width during treadmill walking by older and younger adults. Gait Posture. 2010;31: 461–464. doi: 10.1016/j.gaitpost.2010.02.001 [DOI] [PubMed] [Google Scholar]
16.Wang Y, Srinivasan M. Stepping in the direction of the fall: The next foot placement can be predicted from current upper body state in steady-state walking. Biol Lett. 2014;10. doi: 10.1098/rsbl.2014.0405 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Mahaki M, Bruijn SM, Van Dieën JH. The effect of external lateral stabilization on the use of foot placement to control mediolateral stability in walking and running. PeerJ. 2019;2019. doi: 10.7717/peerj.7939 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Donelan JM, Shipman DW, Kram R, Kuo AD. Mechanical and metabolic requirements for active lateral stabilization in human walking. J Biomech. 2004;37: 827–835. doi: 10.1016/j.jbiomech.2003.06.002 [DOI] [PubMed] [Google Scholar]
19.Nordin E, Moe-Nilssen R, Ramnemark A, Lundin-Olsson L. Changes in step-width during dual-task walking predicts falls. Gait Posture. 2010;32: 92–97. doi: 10.1016/j.gaitpost.2010.03.012 [DOI] [PubMed] [Google Scholar]
20.Callisaya ML, Blizzard L, Schmidt MD, McGinley JL, Srikanth VK. Ageing and gait variability-a population-based study of older people. Age Ageing. 2010;39: 191–197. doi: 10.1093/ageing/afp250 [DOI] [PubMed] [Google Scholar]
21.Skiadopoulos A, Moore EE, Sayles HR, Schmid KK, Stergiou N. Step width variability as a discriminator of age-related gait changes. J Neuroeng Rehabil. 2020;17: 1–13. doi: 10.1186/s12984-020-00671-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Ivanenko YP, Cappellini G, Dominici N, Poppele RE, Lacquaniti F. Modular control of limb movements during human locomotion. J Neurosci. 2007;27: 11149–11161. doi: 10.1523/JNEUROSCI.2644-07.2007 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Ivanenko YP, Cappellini G, Poppele RE, Lacquaniti F. Spatiotemporal organization of α‐motoneuron activity in the human spinal cord during different gaits and gait transitions. Eur J Neurosci. 2008;27: 3351–3368. doi: 10.1111/j.1460-9568.2008.06289.x [DOI] [PubMed] [Google Scholar]
24.Janshen L, Santuz A, Ekizos A, Arampatzis A. Fuzziness of muscle synergies in patients with multiple sclerosis indicates increased robustness of motor control during walking. Sci Rep. 2020;10: 1–14. doi: 10.1038/s41598-020-63788-w [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Chvatal SA, Ting LH. Common muscle synergies for balance and walking. Front Comput Neurosci. 2013;7: 1–14. doi: 10.3389/fncom.2013.00001 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Bizzi E, Cheung VCK, d’Avella A, Saltiel P, Tresch M. Combining modules for movement. Brain Res Rev. 2008;57: 125–133. doi: 10.1016/j.brainresrev.2007.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.d’Avella A, Saltiel P, Bizzi E. Combinations of muscle synergies in the construction of a natural motor behavior. Nat Neurosci. 2003;6: 300–308. doi: 10.1038/nn1010 [DOI] [PubMed] [Google Scholar]
28.d’Avella A. Modularity for motor control and motor learning. Advances in Experimental Medicine and Biology. Springer, Cham; 2016. pp. 3–19. doi: 10.1007/978-3-319-47313-0_1 [DOI] [PubMed] [Google Scholar]
29.Santuz A, Ekizos A, Eckardt N, Kibele A, Arampatzis A. Challenging human locomotion: Stability and modular organisation in unsteady conditions. Sci Rep. 2018;8: 1–13. doi: 10.1038/s41598-018-21018-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Baggen RJ, va Dieën JH., Van Roie E, Verschueren SM, Giarmatzis G, Delecluse C, et al. Age-related differences in muscle synergy organization during step ascent at different heights and directions. Appl Sci. 2020;10. doi: 10.3390/app10061987 [DOI] [Google Scholar]
31.da Silva Costa AA, Moraes R, Hortobágyi T, Sawers A. Older adults reduce the complexity and efficiency of neuromuscular control to preserve walking balance. Exp Gerontol. 2020; 111050. doi: 10.1016/j.exger.2020.111050 [DOI] [PubMed] [Google Scholar]
32.Martino G, Ivanenko YP, d’Avella A, Serrao M, Ranavolo A, Draicchio F, et al. Neuromuscular adjustments of gait associated with unstable conditions. J Neurophysiol. 2015;114: 2867–2882. doi: 10.1152/jn.00029.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Wang Y, Watanabe K, Asaka T. Effect of dance on multi-muscle synergies in older adults: a cross-sectional study. BMC Geriatr. 2019;19: 340. doi: 10.1186/s12877-019-1365-y [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Sawers A, Allen JL, Ting LH. Long-term training modifies the modular structure and organization of walking balance control. J Neurophysiol. 2015;114: 3359–3373. doi: 10.1152/jn.00758.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Alizadehsaravi L, Koster RAJ, Muijres W, Maas H, Bruijn SM, van Dieën JH. The underlying mechanisms of improved balance after one and ten sessions of balance training in older adults. Hum Mov Sci. 2022;81: 1–14. doi: 10.1016/j.humov.2021.102910 [DOI] [PubMed] [Google Scholar]
36.Hilliard MJ, Martinez KM, Janssen I, Edwards B, Mille ML, Zhang Y, et al. Lateral Balance Factors Predict Future Falls in Community-Living Older Adults. Arch Phys Med Rehabil. 2008;89: 1708–1713. doi: 10.1016/j.apmr.2008.01.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Maki BE, Holliday PJ, Topper AK. A prospective study of postural balance and risk of falling in an ambulatory and independent elderly population. Journals Gerontol. 1994;49: M72–M84. Available: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=emed3&NEWS=N&AN=1994105124. [DOI] [PubMed] [Google Scholar]
38.Kluft N, van Dieën JH, Pijnappels M. The degree of misjudgment between perceived and actual gait ability in older adults. Gait Posture. 2017;51: 275–280. doi: 10.1016/j.gaitpost.2016.10.019 [DOI] [PubMed] [Google Scholar]
39.Magnani RM, Bruijn SM, van Dieën JH, Forbes PA. Stabilization demands of walking modulate the vestibular contributions to gait. Sci Rep. 2021;11: 1–13. doi: 10.1038/s41598-021-93037-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Nashner LM. The organization of human postural movements: A formal basis and experimental synthesis. 1985;8: 135–172. doi: 10.1017/S0140525X00020008 [DOI] [Google Scholar]
41.Runge CF, Shupert CL, Horak FB, Zajac FE. Ankle and hip postural strategies defined by joint torques. Gait Posture. 1999;10: 161–170. doi: 10.1016/s0966-6362(99)00032-6 [DOI] [PubMed] [Google Scholar]
42.Reimann H, Fettrow T, Jeka JJ. Strategies for the Control of Balance During Locomotion. Kinesiol Rev. 2018;7: 18–25. doi: 10.1123/kr.2017-0053 [DOI] [Google Scholar]
43.van Leeuwen AM, van Dieën JH, Daffertshofer A, Bruijn SM. Ankle muscles drive mediolateral center of pressure control to ensure stable steady state gait. Sci Rep. 2021;11: 1–14. doi: 10.1038/s41598-021-00463-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Kerr NL. HARKing: Hypothesizing after the results are known. Personal Soc Psychol Rev. 1998;2: 196–217. doi: 10.1207/s15327957pspr0203_4 [DOI] [PubMed] [Google Scholar]
45.Rowbottom DP, Alexander RMN. The role of hypotheses in biomechanical research. Sci Context. 2012;25: 247–262. doi: 10.1017/s0269889712000051 [DOI] [PubMed] [Google Scholar]
46.Muehlbauer T, Gollhofer A, Lesinski M, Hortoba T, Granacher U. Effects of Balance Training on Balance Performance in Healthy Older Adults: A Systematic Review and Meta-analysis. Sport Med. 2015;45: 1721–1738. doi: 10.1007/s40279-015-0375-y [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Bisson E, Contant B, Sveistrup H, Lajoie Y. Functional balance and dual-task reaction times in older adults are improved by virtual reality and biofeedback training. Cyberpsychology Behav. 2007;10: 16–23. doi: 10.1089/cpb.2006.9997 [DOI] [PubMed] [Google Scholar]
48.Nagy E, Feher-Kiss A, Barnai M, Domján-Preszner A, Angyan L, Horvath G. Postural control in elderly subjects participating in balance training. Eur J Appl Physiol. 2007;100: 97–104. doi: 10.1007/s00421-007-0407-x [DOI] [PubMed] [Google Scholar]
49.Sherrington C, Tiedemann A, Fairhall N, Close JCT, Lord SR. Exercise to prevent falls in older adults: an updated meta-analysis and best practice recommendations. N S W Public Health Bull. 2011;22: 78–83. doi: 10.1071/NB10056 [DOI] [PubMed] [Google Scholar]
50.Bruijn SM, Meijer OG, Beek PJ, Van Dieen JH. Assessing the stability of human locomotion: A review of current measures. J R Soc Interface. 2013;10. doi: 10.1098/rsif.2012.0999 [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Bruijn SM, van Dieën JH, Meijer OG, Beek PJ. Statistical precision and sensitivity of measures of dynamic gait stability. J Neurosci Methods. 2009;178: 327–333. doi: 10.1016/j.jneumeth.2008.12.015 [DOI] [PubMed] [Google Scholar]
52.Cappozzo A, Catani F, Della Croce U, Leardini A. Position and orientation in space of bones during movement: anatomical frame definition and determination. Clin Biomech. 1995;10: 171–178. doi: 10.1016/0268-0033(95)91394-t [DOI] [PubMed] [Google Scholar]
53.Neptune RR, McGowan CP. Muscle contributions to frontal plane angular momentum during walking. J Biomech. 2016;49: 2975–2981. doi: 10.1016/j.jbiomech.2016.07.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Vieira TMM, Minetto MA, Hodson-Tole EF, Botter A. How much does the human medial gastrocnemius muscle contribute to ankle torques outside the sagittal plane? Hum Mov Sci. 2013;32: 753–767. doi: 10.1016/j.humov.2013.03.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Sozzi S, Honeine JL, Do MC, Schieppati M. Leg muscle activity during tandem stance and the control of body balance in the frontal plane. Clin Neurophysiol. 2013;124: 1175–1186. doi: 10.1016/j.clinph.2012.12.001 [DOI] [PubMed] [Google Scholar]
56.Afschrift M, Pitto L, Aerts W, van Deursen R, Jonkers I, De Groote F. Modulation of gluteus medius activity reflects the potential of the muscle to meet the mechanical demands during perturbed walking. Sci Rep. 2018;8: 1–11. doi: 10.1038/s41598-017-17765-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol. 2000;10: 361–374. doi: 10.1016/s1050-6411(00)00027-4 [DOI] [PubMed] [Google Scholar]
58.Roerdink M, Coolen BH, Clairbois BHE, Lamoth CJC, Beek PJ. Online gait event detection using a large force platform embedded in a treadmill. J Biomech. 2008;41: 2628–2632. doi: 10.1016/j.jbiomech.2008.06.023 [DOI] [PubMed] [Google Scholar]
59.Bruijn SM, Van Dieën JH, Daffertshofer A. Beta activity in the premotor cortex is increased during stabilized as compared to normal walking. Front Hum Neurosci. 2015;9: 1–13. doi: 10.3389/fnhum.2015.00001 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Rosenstein MT, Collins JJ, De Luca CJ. A practical method for calculating largest Lyapunov exponents from small data sets. Phys D Nonlinear Phenom. 1993;65: 117–134. doi: 10.1016/0167-2789(93)90009-P [DOI] [Google Scholar]
61.Bruijn S. SjoerdBruijn/LocalDynamicStability: First release. 2017 [cited 2 Sep 2020]. 10.5281/ZENODO.573285.
62.Lee DD, Seung HS. Learning the parts of objects by non-negative matrix factorization. Nature. 1999;401: 788–791. doi: 10.1038/44565 [DOI] [PubMed] [Google Scholar]
63.Oliveira ASC, Gizzi L, Kersting UG, Farina D. Modular organization of balance control following perturbations during walking. J Neurophysiol. 2012;108: 1895–1906. doi: 10.1152/jn.00217.2012 [DOI] [PubMed] [Google Scholar]
64.Patla AE. Some characteristics of EMG patterns during locomotion: implications for the locomotor control process. J Mot Behav. 1985;17: 443–461. doi: 10.1080/00222895.1985.10735360 [DOI] [PubMed] [Google Scholar]
65.Ivanenko YP, Poppele RE, Lacquaniti F. Motor control programs and walking. Neurosci a Rev J bringing Neurobiol Neurol psychiatry. 2006;12: 339–348. doi: 10.1177/1073858406287987 [DOI] [PubMed] [Google Scholar]
66.Harrison D, Kanji GK. The development of analysis of variance for circular data. J Appl Stat. 1988;15: 197–223. doi: 10.1080/02664768800000026 [DOI] [Google Scholar]
67.Jurca A, Žabkar J, Džeroski S. Analysis of 1.2 million foot scans from North America, Europe and Asia. Sci Rep. 2019;9: 1–11. doi: 10.1038/s41598-018-37186-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Socie M, Sandroff B, Pula J, Hsiao-Wecksler E, Motl R, Sosnoff J. Footfall Placement Variability and Falls in Multiple Sclerosis. Ann Biomed Eng. 2012;41. doi: 10.1007/s10439-012-0636-y [DOI] [PubMed] [Google Scholar]
69.Dean JC, Kautz SA. Foot placement control and gait instability among people with stroke Jesse. J Rehabil Res Dev. 2016;52: 577–590. doi: 10.1682/JRRD.2014.09.0207.Foot [DOI] [PMC free article] [PubMed] [Google Scholar]
70.Maki BE. Gait changes in older adults: predictors of falls or indicators of fear. J Am Geriatr Soc. 1997;45: 313–320. doi: 10.1111/j.1532-5415.1997.tb00946.x [DOI] [PubMed] [Google Scholar]
71.Martínez-Amat A; Hita-Contreras F; Lomas-Vega R; Caballero-Martínez I; Alvarez P.J.; Martínez-López E. Effects of 12-week proprioception training program on postural stability, gait, and balance in older adults: A controlled clinical trial. J Strength Cond. 2013;27: 2180–2188. [DOI] [PubMed] [Google Scholar]
72.Ambrosini E, Parati M, Peri E, De Marchis C, Nava C, Pedrocchi A, et al. Changes in leg cycling muscle synergies after training augmented by functional electrical stimulation in subacute stroke survivors: a pilot study. J Neuroeng Rehabil. 2020;17: 1–14. doi: 10.1186/s12984-019-0634-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
73.Sherrington C, Michaleff ZA, Fairhall N, Paul SS, Tiedemann A, Whitney J, et al. Exercise to prevent falls in older adults: An updated systematic review and meta-analysis. Br J Sports Med. 2017;51: 1749–1757. doi: 10.1136/bjsports-2016-096547 [DOI] [PubMed] [Google Scholar]
74.McAndrew Young PM, Dingwell JB. Voluntary changes in step width and step length during human walking affect dynamic margins of stability. Gait Posture. 2012/04/01. 2012;36: 219–224. doi: 10.1016/j.gaitpost.2012.02.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Shih H- JS, Gordon J, Kulig K. Trunk control during gait: Walking with wide and narrow step widths present distinct challenges. J Biomech. 2021;114: 110135. doi: 10.1016/j.jbiomech.2020.110135 [DOI] [PubMed] [Google Scholar]
76.Monaco V, Ghionzoli A, Micera S. Age-related modifications of muscle synergies and spinal cord activity during locomotion. J Neurophysiol. 2010;104: 2092–2102. doi: 10.1152/jn.00525.2009 [DOI] [PubMed] [Google Scholar]
77.Ivanenko YP, Poppele RE, Lacquaniti F. Five basic muscle activation patterns account for muscle activity during human locomotion. J Physiol. 2004;556: 267–282. doi: 10.1113/jphysiol.2003.057174 [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Oliveira AS, Gizzi L, Farina D, Kersting UG. Motor modules of human locomotion: Influence of EMG averaging, concatenation, and number of step cycles. Front Hum Neurosci. 2014;8. doi: 10.3389/fnhum.2014.00335 [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Cappellini G, Ivanenko YP, Poppele RE, Lacquaniti F. Motor patterns in human walking and running. J Neurophysiol. 2006;95: 3426–3437. doi: 10.1152/jn.00081.2006 [DOI] [PubMed] [Google Scholar]
80.van den Hoorn W, Hodges PW, van Dieen JH, Hug F. Effect of acute noxious stimulation to the leg or back on muscle synergies during walking. J Neurophysiol. 2015;113: 244–254. doi: 10.1152/jn.00557.2014 [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0242115.r001

Decision Letter 0

Fabio Augusto Barbieri

10 Feb 2021

PONE-D-20-30116

Neuromuscular control of gait stability in older adults is adapted to environmental demands but not improved after standing balance training

PLOS ONE

Dear Dr. van Dieën,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Firstly, sorry for the slow review process, but I had to submit for a third reviewer. One reviewer suggested to reject the paper. The other two reviewers were favorable with the publication of the paper, but many points need to be improved. Please address all suggestions of the three reviewers. Please try to address the suggestions of the reviewer that indicated to reject the paper too (Reviewer 2). There are some relevant aspects in the review. For example, the rationale that training standing balance would improve walking balance needs to be improved. Also, I ask special attention in the discussion to explain the absence of effect of the balance training - both reviewers have indicated that this aspect should be better explained.

Please submit your revised manuscript by Mar 27 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

We look forward to receiving your revised manuscript.

Kind regards,

Fabio A. Barbieri, PhD

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The study investigated the skill transfer of standing balance training to walking balance among older adults. Walking balance was tested in normal and narrow-base walking conditions. The findings quantified a series of gait measures including muscle synergies, COM displacement, and gait stability. The study had a null finding on the training effect on gait despite positive effects of training on stance balance reported in an earlier study using the same cohort of participants. The study is generally well-done, and the results are relevant to both researchers and clinicians. Addressing the issues noted below will increase the clarity of the manuscript.

Major issues:

1. Interpretation of null effect: The authors speculate that the duration of training was the cause of the null effect. I disagree with this cause for two reasons. First, if the duration was too short, I would have expected to see some improvement between post 1 and post 2. Second, there is a significant effect of training on standing balance from your previous paper (ref#24, Alizadehsaravi et al. 2020). If the duration was sufficient for standing, what is the rationale that it was not sufficient for walking?

2. Please justify the selection of muscles. Specifically, muscles that were examined were mostly sagittal, but the task is more medial-lateral than sagittal, so perhaps it is not surprising that sagittal muscles synergies were not affected by the training. Gluteus medius is the only muscle that acts mainly in the ML plane (hip abductor).

3. Recommendation for the structure of the discussion. The following text: “The gait measures were different between conditions but not affected by training” was emphasized several times in the discussion, and demonstrates that these two issues were conflated throughout. It would increase clarity to separate the discussion for the gait adaptation from the discussion for training. The adaptation is shown in several different gait measures and, in my opinion, is the most interesting finding of this study. I suggest the authors have stand-alone discussion paragraph(s) focusing on changes in gait control for narrow base walking and what they would mean in real life locomotion and falls. More in-depth discussion should address the questions such as, are all the changes in gait control adaptation to the postural challenge? Could they be maladaptive or mechanical consequences of a narrower base of support?

4. This is not necessary, but it could be of interest: Would it be possible to quantify the synergies for standing balance? In lines 58-59 the authors note that four synergies are shared between walking and reactive balance control. Adding the standing balance synergies to the current paper on gait synergies would provide a more comprehensive study.

5. Please specify the type of synergy analysis that was conducted, for example, principal component analysis, UCM analysis, or other approach.

Minor comments:

1. Line 34 “Falls in older adults mostly occur during walking. Therefore, skills acquired during stance balance should transfer to gait and improve gait stability”. The logic and/or the wording is not clear across the first two sentences, it is explained more fully in later sentences, but the second sentence should not start with “Therefore,…”.

2. Line 51 and line 276 - Confidence is mentioned in a couple of locations in the manuscript, but confidence was not measured in the current study. Perhaps it should be noted as a need for future study.

3. Line 105-108 – Please explicitly describe the exercise plan, including progressions and progression criteria, so that the reader can determine if the training was sufficiently challenging and was appropriately progressed. Note that the progressions were not available in previous publication either. The null effects can possibly be explained by these training details.

4. Since the journal does not have a word limit, I recommend that more details of the methods are included in the current paper, rather than simply directing readers to the previous paper for the method section (Alizadehsaravi et al., 2020). For example, familiarization trials mentioned in figure 1 can also be explained in the text. Rationales could be provided for the treadmill settings (4.5 minutes at a constant speed of 3.5 km/h).

5. Foot placement error was reported to indicate the narrow base walking performance. The mean error is less than 2 cm, which is presumably quite small. Could the training be subject to a ceiling effect? It could be helpful to also report step width and step width variability, to aid the interpretation of narrow base walking performance in comparison with normal walking.

Editorial comments:

Line 90 - Exclusion criteria 4 minutes without an aid – in previous paper it is three minutes.

Line 211 typo “did not”

Figures - Be consistent with the scales, number of decimal points – use different markers for normal and narrow gait, to accommodate people without color printers.

Figure 6 – Authors should provide the x-axis label. Error bars description?

Line 227 – “87+2%”. Consistent with spacing.

Reviewer #2: The current manuscript reports the results of a study that sought to examine whether standing balance training altered or improved balance performance and/or its control in older adults. Measures derived from the center of mass (i.e., displacement, velocity, LDE) and foot placement were used to characterize performance, while a host of muscle recruitment and coordination metrics are employed to characterize the control of balance. While further efforts to understand the neuromuscular basis for walking and balance control, as well as the effect of training or aging, are always welcome, my enthusiasm for the current manuscript is dampened by issues regarding study motivation and scientific rationale, technical/methodological choices, and the lack of a discussion that explores, explains, or puts the current results in content with previous research.

Major

Study motivation and scientific rationale

1. It is unclear to me why training in standing balance would have an effect on the control or performance of walking balance. This would seem to go against the principle of specificity, which has been frequently described and cited in the balance literature (Grabiner; Oddsson). Additionally, multiple studies have shown standing balance does not equate to walking balance (Mackey, 2005; Owings, 2000). It is therefore unclear what evidence would support the idea that training standing balance would improve walking balance or alter its control.

References

Oddsson et al., 2007. How to improve gait and balance function in elderly individuals—compliance with principles of train.

Grabiner et al., 2012. Task-specific training reduces trip-related fall risk in women.

Mackey et al., 2005. Postural steadiness during quiet stance does not associate with ability to recover balance in older women.

Owings et al., 2000. Measures of postural stability are not predictors of recovery from large postural disturbances in healthy older

2. The lack of scientific rationale is reflected in the lack of structure and organization in the introduction. As currently written, each paragraph in the introduction does not build on the previous paragraph in a way that creates a logical argument and directs the reader to an obvious gap/need (i.e., study objective), and an accompanying hypothesis. In fact, no specific statement of purpose and/or testable hypotheses are provided. Similarly, the contents within each introduction paragraph do not support or back up the topic sentence at the start of each paragraph. Greater attention to "building your argument" based on the published literature is required in the introduction.

Technical issues

3. It is not clear why the extraction of muscle synergies was limited to five. A number of studies have shown the number of muscle synergies to change with skill or training. This methodological choice would therefore seem at odds with the goal of the study; to examine whether training alters neuromuscular control of walking balance. It would seem more suitable to identify the number of muscle synergies for each subject in each condition that is required to explain the original EMG.

4. The extracted synergies are reported to account for 85% of the variance in the EMG signals. Is this sufficient? Typically, studies that extract muscle synergies from EMG signals use between 90 or 95% variance accounted for. Similarly, was the variance accounted in a specific muscle, across all conditions? Additional explanation and justification for how and why muscle synergies were extracted in the manner they were is required.

5. Parametric inferential statistics are used to compare conditions and examine the effect of training. However, no assessments of normality were performed or reported. Please provide the results of Shapiro-Wilk's tests if parametric statistics are to be used.

6. Little information is provided about study participants. Did the study participants have a history of falling? High or low balance confidence or falls self-efficacy scores? In the absence of additional information describing the balance and falls characteristics of the study participants, it is difficult to interpret the study results and put them in context.

Discussion

7. As currently written, the discussion largely reiterates the study results. At present there is no explanation for why training failed to have an effect on walking balance performance or control. This seems critical as it appears to be the primary motivation for the study. Also, there is no discussion of how the current study results regarding muscle synergies and balance control compare to or differ from studies previously conducted (e.g., da Silva Costa, 2020; Allen, 2020; Monaco, 2010). As a result, it is not clear what additional insight is provided by the current study to advance our understanding of balance control and aging. A major revision to the discussion is likely required.

Minor

1. Please review the manuscript for grammatical and spelling errors.

2. It is unclear why gait speed was kept constant between walking conditions (regular and narrow), between participants, and between time points (i.e., pre and post). Would walking speed not be one of the variables expected to improve with training? Please provide justification for both the selection of this specific gait speed, as well as its constraint across and within study conditions and time points.

3. Results: the section describing the spatial components of muscle synergies reads more like a discussion or interpretation of the spatial patterns. Consider revising this section to simply report rather than interpret the results.

4. Figure 1 is not required. The section in the methods provides sufficient information on the study protocol.

5. Figures 3-5, and 8 are very difficult to interpret. I would advise removing the lines connecting each of the phases of the study.

6. Justification for the selection of the wide range of balance performance and control metrics are required. Upon what basis did the authors expect these metrics to change as a function of training?

7. Additional details regarding the standing balance training and subsequent results are required. Perhaps the training procedures could be included in an Appendix. The results of the standing training should be briefly summarized in the results to provide the reader with context.

Reviewer #3: This manuscript describes a study investigating the effect of balance training on neuromuscular and stability control in older adults during normal and narrow base walking. The findings indicated no training effects on these parameters. As expected, the base of support constraint resulted in walking adaptations in both neuromuscular and stability parameters. Although the manuscript is well organized, some aspects could be improved to facilitate readability to those not familiarized with some parameters analyzed in this study. The discussion could also be improved to explain the absence of effect of the balance training.

1) The abstract would benefit from the use of plain language, particularly when describing the results. The use of jargon (particularly FWHM and CoA) makes it difficult to follow and understand the main findings.

2) The first two sentences of the introduction suggest a logical interconnection that is not the case. The fact that people often fall during walking does not imply that a “balance training should transfer to gait and improve gait stability”.

3) How was the sample size determined for this study? Was there any sample size calculation?

4) The description of the balance training protocol (page 6, lines 104-109) is very brief. It is essential to provide additional information regarding the duration of each session, the progression of the exercises throughout the sessions, and the organization of each session (warm-up, main activities, etc.).

5) When describing the FWHM and CoA (page 9, lines 186-192), include the meaning of these variables for those not familiarized with these variables.

6) For the beam walking task, did the participants step off the beam at any moment? What was the maximum foot placement error accepted when the participant eventually did not step with the entire foot over the beam?

7) Since the beam width (12 cm) is remarkably close to the average foot width, how foot placement error could be affected by balance training? Wouldn’t you have a floor effect for this variable?

8) Page 13 (lines 286-287): what do you mean by “robustness of gait”?

9) The discussion about the absence of effect of the balance training in all investigated variables was incredibly brief (page 14, lines 313-317). Most of the discussion is focused on the effects of the base of support manipulation. The discussion regarding the balance training needs to improve to explain the absence of an effect. Only arguing that the duration of the training protocol was short is not enough. Why did you choose a 3-week intervention if studies are showing that more time is needed to observe gait improvements? Were the variables not sensitive enough to capture gains due to balance training?

10) I recommend including a paragraph with the limitations of this study.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2022 Jul 27;17(7):e0242115. doi: 10.1371/journal.pone.0242115.r002

Author response to Decision Letter 0

8 Mar 2022

Reply to reviewers

We would like to thank the reviewers for their comments and suggestions, which have helped improve this paper. The introduction and discussion sections of the manuscript have been modified extensively. Below we have copied the reviewers' comments and have provided a point-by-point reply.

Reviewer #1: The study investigated the skill transfer of standing balance training to walking balance among older adults. Walking balance was tested in normal and narrow-base walking conditions. The findings quantified a series of gait measures, including muscle synergies, COM displacement, and gait stability. The study had a null finding on the training effect on gait despite positive effects of training on stance balance reported in an earlier study using the same cohort of participants. The study is generally well-done, and the results are relevant to both researchers and clinicians. Addressing the issues noted below will increase the clarity of the manuscript.

Major issues:

1- Interpretation of null effect: The authors speculate that the duration of training was the cause of the null effect. I disagree with this cause for two reasons. First, if the duration was too short, I would have expected to see some improvement between post 1 and post 2. Second, there is a significant effect of training on standing balance from your previous paper (ref#24, Alizadehsaravi et al. 2020). If the duration was sufficient for standing, what is the rationale that it was not sufficient for walking?

We thank the reviewer for this comment. Following the reviewer’s suggestions in minor comment #5, step width and step width variability have been added to the gait kinematics analysis. Based on these new analyses we did find a positive effect of the balance training on gait. The transfer effect was evident in decreased step width and decreased step width variability after three weeks of standing balance training. Note that the primary measures, defined as the foot placement error and the percentage of the steps within the beam, did not change after the training. Therefore, we conclude that transfer did happen, but some caution is needed as this was not demonstrated by our primary outcome variable. The introduction and methods have been updated to include step width and step width variability and the discussion has been modified extensively.

2- Please justify the selection of muscles. Specifically, muscles that were examined were mostly sagittal, but the task is more medial-lateral than sagittal, so perhaps it is not surprising that sagittal muscles synergies were not affected by the training. Gluteus medius is the only muscle that acts mainly in the ML plane (hip abductor).

We thank the reviewer for this comment. We agree with the reviewer that most of the measured muscles were not acting solely in the frontal plane. Although we focused on balance in the frontal plane, motions in all three planes are important for a stable gait pattern. Moreover, muscles do not have effects in one plane, but produce effects in multiple planes. For example, sagittal plane push-off, which is mainly generated by the calf muscles, acts at some distance to the mid-sagittal plane through which push-off modulation can contribute to frontal plane balance control. We measured almost all possible ankle muscles as they play a key role in postural stability 4–6. A simulation study also showed that in the late stance of a gait cycle, the gluteus medius rotates the body towards the ipsilateral leg, while the soleus and gastrocnemius rotate the body towards the contralateral leg4.

Biceps femoris, rectus femoris, and vastus lateralis were measured as they contribute to stabilizing the knee around heel strike and in single leg stance.

Moreover, we chose similar muscles to a previous study that showed changes in synergies due to long-term training during walking3. We also measured the adductor longus, but we omitted it from the analysis in view of the signals being very noisy, especially in female subjects.

We now provide this motivation in the manuscript Lines 194-197.

3- Recommendation for the structure of the discussion. The following text: "The gait measures were different between conditions but not affected by training" was emphasized several times in the discussion, and demonstrates that these two issues were conflated throughout. It would increase clarity to separate the discussion for the gait adaptation from the discussion for training. The adaptation is shown in several different gait measures and, in my opinion, is the most interesting finding of this study. I suggest the authors have stand-alone discussion paragraph(s) focusing on changes in gait control for narrow base walking and what they would mean in real life locomotion and falls. More in-depth discussion should address the questions such as, are all the changes in gait control adaptations to the postural challenge? Could they be maladaptive or mechanical consequences of a narrower base of support?

We thank the reviewer for the suggestions. The discussion has been updated and restructured to general findings, the effect of narrow base, transfer effect, limitations, and conclusions, to address this suggestion. Additionally, lines 465-471 have been added to the discussion (effects of narrow base) to focus on changes in gait control for narrow base walking and what it means in real life.

About the last part of the reviewer's comment, we do not expect the modulation of the gait parameters between the narrow base and normal gait to be maladaptive because the decreased center of mass displacement and velocity are beneficial for balance control. Note that in a challenging condition the efficiency-safety trade-off moves more to the side of safety. However, the adaptation to the mechanical constraints that we imposed will have mechanical consequences. It was our aim to understand the neural responses needed to successfully deal with these mechanical constraints.

Since in our new analysis we found the effect of training to be transferred to gait, we have rewritten the discussion according to the new findings. There are separate paragraphs discussing the adaptation and the training effect as the reviewer suggested.

4- This is not necessary, but it could be of interest: Would it be possible to quantify the synergies for standing balance? In lines 58-59 the authors note that four synergies are shared between walking and reactive balance control. Adding the standing balance synergies to the current paper on gait synergies would provide a more comprehensive study.

We thank the reviewer for the comment. The paper we referred to in the introduction showed that most muscle synergies used in perturbation responses during standing were also used in perturbation responses during walking, suggesting common neural mechanisms for reactive balance across different contexts. (DOI: 10.3389/fncom.2013.00048)

The reviewer's suggestion is indeed interesting. However, to make a fair comparison of the synergies we would need to use the same set of muscles in standing and in gait. Our balance study was unipedal (unlike the study we referred to), which will make a significant difference in the grouping of muscles into synergies. Therefore, we decided not to include the synergies of standing balance in this study. We have analyzed the synergies of stance leg and trunk muscles and have observed training effects on perturbed standing balance, which is reported elsewhere (DOI: https://doi.org/10.1101/2021.03.31.437824).

5- Please specify the type of synergy analysis that was conducted, for example, principal component analysis, UCM analysis, or other approach.

We thank the reviewer for the comment. We have now more clearly written that we use NNMF to extract synergies;

“Synergies were extracted from 11 muscles using non-negative matrix factorization based on Lee and Seung's multiplicative update rule7 with 50 repetitions with a maximum of 1000 iterations to update the components and at a tolerance of 10-6.”Lines 245-247.

Minor comments:

1- Line 34 "Falls in older adults mostly occur during walking. Therefore, skills acquired during stance balance should transfer to gait and improve gait stability". The logic and/or the wording is not clear across the first two sentences, it is explained more fully in later sentences, but the second sentence should not start with "Therefore,…".

We thank the reviewer for this comment. We noted the suggestion, and the manuscript has been updated as follows:

“Falls in older adults mostly occur during walking [1,2]. Thus, if skills acquired by balance training programs do not transfer to improvements in gait stability, they are unlikely to decrease the number of falls.” Lines 37-39.

2- Line 51 and line 276 - Confidence is mentioned in a couple of locations in the manuscript, but confidence was not measured in the current study. Perhaps it should be noted as a need for future study.

We thank the reviewer for the comment. We know that older adults take wider step in order to increase their base of support and to be robust against the perturbations, which can be regarded as cautious behavior. Based on our new results, the decreased step width and decreased step width variability after the longer duration of training, we concluded that the improved gait biomechanics may reflect an increased confidence. This has been further discussed in the discussion section.

We have measured the concern of falls using the FESI questionnaire, but so far did not report it. We have updated the manuscript's supplementary materials to include the FESI scores. Between Pre and Post2, although there was a trend, there was no significant decrease in FESI score. However, at retention, the FESI score had decreased significantly, implying that weeks after training, participants felt more confident about their balance ability. The results are shown below:

"A repeated measures ANOVA indicated that concern of falling was affected by balance training. Post-hoc analysis showed that concern of falling was not significantly changed immediately after the training program but was decreased at the retention measurement 2 weeks after the end of training (t = 2.16, p = 0.072; t = 2.82, p = 0.022, respectively."

Figure 1. FES-I scores at different time points. Each of the lines between timepoints represents the score of a single participant.

3- Line 105-108 – Please explicitly describe the exercise plan, including progressions and progression criteria, so that the reader can determine if the training was sufficiently challenging and was appropriately progressed. Note that the progressions were not available in previous publications either. The null effects can possibly be explained by these training details.

We thank the reviewer for the comment. We have provided more information on the training program and the training materials as supplementary material. Progression criteria were based on the physical therapist's observation during the training sessions; if participants were able to perform the task for 60 seconds, the difficulty level was increased. The progression was hard to measure objectively as the training was group-based. But the general progression plan was as follows:

Table 1. Guideline for training progression

Number Exercise Duration/Frequency

Warm-up

1) Head rotations Rotate head to either side 5 x

3 repetitions

2) Back stretching Stretch 3 x

3 repetitions

3) Trunk rotations 5 rotations to both sides

3 repetitions

Exercises

4) Balancing

- One leg stance (when possible)

- Switch the legs

- Unstable surfaces 3 x 60 seconds

2 repetitions

5) Balancing eyes-closed

- One leg stance (when possible)

- Switch the legs

- Unstable surfaces 3 x 60 seconds

2 repetitions

6) Displacement of weight

- One leg stance

- Switch the legs

- Unstable surfaces 3 x 60 seconds

2 repetitions

7) Passing/throwing around a ball in groups of 4

Fitness ball

- One leg

- Unstable surface

2 kg ball

- One leg

- Unstable surface

Alternative approaches:

- Make the circle bigger.

- With back towards each other in order to induce more trunk rotations. 5 rounds both directions

3 repetitions

Pass the big ball around while stopping it on foot and role it to the other person.

Fitness ball

- One leg

- Unstable surface

2 kg ball

- One leg

- Unstable surface 5 rounds both directions

3 repetitions

Figure. 2. Balance training materials

Note that exercises that specifically target gait were excluded from the training to investigate the transfer effect.

4- Since the journal does not have a word limit, I recommend that more details of the methods are included in the current paper, rather than simply directing readers to the previous paper for the method section (Alizadehsaravi et al., 2020). For example, familiarization trials mentioned in figure 1 can also be explained in the text. Rationales could be provided for the treadmill settings (4.5 minutes at a constant speed of 3.5 km/h).

We thank the reviewer for the comment. We have included the following additional information in the methods section:

“To quantify transfer to gait, participants were instructed to walk for 4.5 minutes on a treadmill with an embedded force plate. For estimating the local divergence exponent, a minimum of 150 steps is recommended [41]. We expected that a total duration of 4.5 minutes was needed to reach that number. To avoid effects of gait speed on outcome measures, this was kept constant at 3.5 km/h for all participants [42].” Lines 165-169.

“A narrow-base walking paradigm was chosen because narrow-base walking has been shown to challenge mediolateral stability in older adults [12,34].” Lines 174-175.

“Participants were acquainted with the experimental setup to minimize habituation effects. For familiarization, participants performed 30 seconds of narrow-base walking before the measurement.” Lines 178-180.

5- Foot placement error was reported to indicate the narrow base walking performance. The mean error is less than 2 cm, which is presumably quite small. Could the training be subject to a ceiling effect? It could be helpful to also report step width and step width variability, to aid the interpretation of narrow base walking performance in comparison with normal walking.

We thank the reviewer for this valuable comment. The suggestion has made us do further analysis and we found that step width and step width variability were decreased after training. These new findings do indicate a transfer effect of balance training to gait. The new results have been added to the manuscript. Regarding a possible ceiling in the mean step error, one should consider that the width of the beam is 12 cm, whilst the average foot width is about 10 cm (Jurca, Zabkar, Dzeroski, 2019). This means that participants, on average, had a 2 cm margin in foot placement that was not seen as error. So, the task did not require perfect control. Additionally, the percentages of steps inside the beam were 27.6% (SE 3.9%) pre-training and 33.3% (SE 4.4%) post-training. We do believe this suggests that there is ample room for improvement. The best recorded performance was 67% of steps within the beam, which demonstrated to us that scores much higher than the mean could be obtained, and that even these best performances were not close to the end of the scale.

Editorial comments:

1- Line 90 - Exclusion criteria 4 minutes without an aid – in the previous paper it is three minutes.

Thank you for the comment. Indeed, it was three minutes. The manuscript has been corrected. Line 140.

2- Line 211 typo "did not"

The repetition of "did not" has been removed. Thank you for noticing.

3- Figures - Be consistent with the scales, number of decimal points – use different markers for normal and narrow gait, to accommodate people without color printers.

Thank you for the comment. We double-checked the manuscript and have corrected the figures as suggested.

4- Figure 6 – Authors should provide the x-axis label. Error bars description?

Thank you for the comment. The error bars were not needed and have been removed, thank you.

The X-axis label was an arbitrary unit and has been added to the revised version.

5- Line 227 – "87+2%". Consistent with spacing.

Thank you for the comment. We have corrected the manuscript and put spaces in between as follows: 87 ± 2%

Major

Study motivation and scientific rationale

1- It is unclear to me why training in standing balance would have an effect on the control or performance of walking balance. This would seem to go against the principle of specificity, which has been frequently described and cited in the balance literature (Grabiner; Oddsson). Additionally, multiple studies have shown standing balance does not equate to walking balance (Mackey, 2005; Owings, 2000). It is therefore unclear what evidence would support the idea that training standing balance would improve walking balance or alter its control.

References

Oddsson et al., 2007. How to improve gait and balance function in elderly individuals—compliance with principles of train.

Grabiner et al., 2012. Task-specific training reduces trip-related fall risk in women.

Mackey et al., 2005. Postural steadiness during quiet stance does not associate with ability to recover balance in older women.

Owings et al., 2000. Measures of postural stability are not predictors of recovery from large postural disturbances in healthy older

We thank the reviewer for the references. Based on the first reviewer’s suggestion we have added two new measures to our gait evaluation. We found decreased step width in narrow-base and decreased step width variability in both conditions after training. Therefore, we argue that there was a transfer effect from standing balance to narrow-base walking, although it was not visible in the primary outcome measures, perhaps due to the lower sensitivity of those measures.

The articles suggested by the reviewer were reviewed carefully.

Indeed, the reviewer is right, that in general, training is very task specific. However, as outlined in our introduction, studies that trained standing balance found reduced falls, and falls in daily life often happen during walking. We have now further clarified why transfer effects may be expected in lines 41-44 and lines 71-80.

2- The lack of scientific rationale is reflected in the lack of structure and organization in the introduction. As currently written, each paragraph in the introduction does not build on the previous paragraph in a way that creates a logical argument and directs the reader to an obvious gap/need (i.e., study objective), and an accompanying hypothesis. In fact, no specific statement of purpose and/or testable hypotheses are provided. Similarly, the contents within each introduction paragraph do not support or back up the topic sentence at the start of each paragraph. Greater attention to "building your argument" based on the published literature is required in the introduction.

Thank you for the comment. We have completely reworked the introduction, and made sure to link the paragraphs, and provide a better rationale for our study.

Technical issues

3- It is not clear why the extraction of muscle synergies was limited to five. A number of studies have shown the number of muscle synergies to change with skill or training. This methodological choice would therefore seem at odds with the goal of the study; to examine whether training alters neuromuscular control of walking balance. It would seem more suitable to identify the number of muscle synergies for each subject in each condition that is required to explain the original EMG.

Thank you for the comment. Several authors have reported the number of synergies during gait, and most have reported 4 to 6 synergies. We determined the number of synergies from the concatenated data of all three-measurement timepoints. So, this constrains the number of synergies as well as the muscle weighting factors to be constant across timepoints, while differences in activation profiles between pre- and post-training can be observed. Additionally, having a different number of synergies (i.e., not constraining muscle weighting factors over time points) and different activation profiles will not allow a conclusive result regarding changes due to training. The problem then is that any difference in muscle weightings might as well be explained by different activation profiles (and vice versa), rather than training.

4- The extracted synergies are reported to account for 85% of the variance in the EMG signals. Is this sufficient? Typically, studies that extract muscle synergies from EMG signals use between 90 or 95% variance accounted for. Similarly, was the variance accounted in a specific muscle, across all conditions? Additional explanation and justification for how and why muscle synergies were extracted in the manner they were is required.

Thank you for the comment. Several other studies have also used an 85% threshold [9][10] Considering the number of muscles measured, 5 synergies is an acceptable number, and adding more synergies wouldn't add much value to the current reconstructed EMG. To check for this, we have added one more synergy (see figure below) while VAF improved to >90%, it was only 4% higher with 6 than with 5 synergies.

Unfortunately, there are only a few studies on synergies in older adults that we could refer to, but considering the lower signal to noise ratio of EMG data in older adults, lowering the threshold to 85% for reconstructing the data is not unreasonable (as shown by Baggen et al.[9]), and using 90% might cause overfitting. There are no standards for choosing the threshold or the method, but our choices were based on our research questions and our target group.

Figure 2. decomposed synergies to temporal and spatial components, with 6 (top) and 5 (bottom) synergies

5- Parametric inferential statistics are used to compare conditions and examine the effect of training. However, no assessments of normality were performed or reported. Please provide the results of Shapiro-Wilk's tests if parametric statistics are to be used.

Thank you for the suggestion. Shapiro-Wilk’s test has now been performed and in two of our measures (CoM variability and vCoM) data were not normally distributed. Therefore, log transformation was performed before applying the repeated measure ANOVA. The results were not changed. The Shapiro-Wilk has also been applied to the two new measures (step width and step width variability).

6- Little information is provided about study participants. Did the study participants have a history of falling? High or low balance confidence or falls self-efficacy scores? In the absence of additional information describing the balance and falls characteristics of the study participants, it is difficult to interpret the study results and put them in context.

The participants in our study had no history of falls in the year prior to the measurement. The methods section for participants has been expanded as below:

“Twenty-two older (72.6 ± 4.2 years old; mean ± SD, 11 females) healthy volunteers, without a history of falls in the preceding year, participated in this study. The required sample size was estimated at twenty-two based on power analysis for an F test of a repeated measure, assuming an effect size of 0.44 and correlation among repeated measures of 0.6 (β = 0.8, G * power 3.1.9.2, Düsseldorf, Germany), comparable to similar studies [38,39].” Lines 132-135.

Balance performance of these participants was a primary outcome. This was extensively reported in a previous paper, but results are briefly reported as well as in this paper. FESI scores are provided in the supplementary material.

Discussion

7- As currently written, the discussion largely reiterates the study results. At present there is no explanation for why training failed to have an effect on walking balance performance or control. This seems critical as it appears to be the primary motivation for the study. Also, there is no discussion of how the current study results regarding muscle synergies and balance control compare to or differ from studies previously conducted (e.g., da Silva Costa, 2020; Allen, 2020; Monaco, 2010). As a result, it is not clear what additional insight is provided by the current study to advance our understanding of balance control and aging. A major revision to the discussion is likely required.

We thank the reviewer for this critical comment. The new findings of training effects for step width and step width variability forced us to rewrite the whole discussion.

We tried to make it clear in the introduction that the novel aspects of our study are that: 1) we addressed balance training and transfer to gait in older adults, 2) we quantified the effect of short and longer duration of training, 3) we solely focused on standing balance training. We used unstable balance training and excluded exercises that targeted gait stability directly, to focus on transfer of balance skill to gait as a result of standing balance training. This suggests the effectiveness of the standing balance training in older adults, providing a clue for clinicians and direction for future studies, that standing balance training may be valuable for older adults with declined mobility. We have rewritten the discussion to reflect these ideas more clearly. The idea behind it is if the standing balance training can be transferred to gait, there might be a potential clue for clinicians to use standing balance training to improve gait postural control in less mobile older adults.

Minor

1- Please review the manuscript for grammatical and spelling errors.

Thank you for the comments. The manuscript has been checked and corrected.

2- It is unclear why gait speed was kept constant between walking conditions (regular and narrow), between participants, and between time points (i.e., pre and post). Would walking speed not be one of the variables expected to improve with training? Please provide justification for both the selection of this specific gait speed, as well as its constraint across and within study conditions and time points.

Thank you for the comment. Our rationale to use a constant gait speed is that we were primarily interested in the effect of balance training on gait stability, and not specifically on gait speed, while speed is known to affect stability. Our main focus was on effects on the quality of balance control, which would be reflected in changes in stability measures (LDE, vCoM) and task performance measures (foot placement error, % steps in beam). Had we not constrained gait speed, but taken it as an additional outcome measure, it would have become a confounder in the analysis of all other variables. We have more explicitly stated this in the manuscript: “To avoid effects of gait speed on outcome measures, this was kept constant at 3.5 km/h for all participants.” Lines 168-169.

3- Results: the section describing the spatial components of muscle synergies reads more like a discussion or interpretation of the spatial patterns. Consider revising this section to simply report rather than interpret the results.

Thank you for the comment. We have updated the manuscript as suggested.

4- Figure 1 is not required. The section in the methods provides sufficient information on the study protocol.

Thank you for the suggestion. We prefer to keep the figure to help grasping the experimental procedure in one glance.

5- Figures 3-5, and 8 are very difficult to interpret. I would advise removing the lines connecting each of the phases of the study.

Thank you for the comment. We have updated the figures.

6- Justification for the selection of the wide range of balance performance and control metrics are required. Upon what basis did the authors expect these metrics to change as a function of training?

Thank you for the comment.

Performance measure: Mean foot placement error

This measure was primarily used, rather than comparable measures of medio-lateral foot placement (e.g. step width), because it is by design the most direct quantification of task performance. Hence, we expected it to be the most sensitive measure to changes in task performance.

Kinematics derived: Trunk CoM displacement, trunk CoM velocity, and Local divergence exponent (LDE)

The first two are the standard and most widespread used measures to express gait stability. However, particularly for use during gait these measures are not entirely undisputed as they are not necessarily minimized for task execution. On one hand, relatively minor CoM movement might indicate good motor control and, therefore, good balance ability. On the other hand, relatively large CoM movement might indicate that it is not minimized because of good balance ability.

Because aforementioned metrics leave room for interpretation the choice was made to add another type of measure with proven validity for gait. The choice for the Lyapunov exponent as an outcome measure was based on the existence of substantial evidence suggesting its validity in the context of gait stability and falling [4].

Improved performance and stability of gait would directly express transfer of training at the level of task performance.

As described above we expected changes in gait performance to result from improved control, which we assessed in terms of muscle synergies that could be expected to be common to control of stance and gait. To quantify changes in these synergies we used the full width at half maximum (FWHM) and Centre of activation (CoA). The FWHM metric addresses the duration of activation but is naïve of timing of activation. The CoA metric addresses the timing of activation but is naïve of duration of activation. By including both we get the entire picture. These measures have been shown to be sensitive to task challenge and thus we expected them to be sensitive also to changes in capacity of our subjects to control their balance in a given task.

These justifications for the selection of measures have been added to their corresponding part in the method section.

7- Additional details regarding the standing balance training and subsequent results are required. Perhaps the training procedures could be included in an Appendix. The results of the standing training should be briefly summarized in the results to provide the reader with context.

Thank you for your suggestions. The result about standing balance training have been summarized in the introduction section of manuscript as suggested. In addition, the training protocol has been described in more detail in the supplementary material.

1- The abstract would benefit from the use of plain language, particularly when describing the results. The use of jargon (particularly FWHM and CoA) makes it difficult to follow and understand the main findings.

We thank the reviewer for this comment. We have updated the manuscript to incorporate the definition in plain language.

2- The first two sentences of the introduction suggest a logical interconnection that is not the case. The fact that people often fall during walking does not imply that a "balance training should transfer to gait and improve gait stability".

Thank you for the comment. The introduction has been modified as follows:

3- How was the sample size determined for this study? Was there any sample size calculation?

Thank you for the comment. More details have been added to the manuscript regarding the sample size as follows:

“The required sample size was estimated at twenty-two based on power analysis for an F test of a repeated measure, assuming an effect size of 0.44 and correlation among repeated measures of 0.6 (β = 0.8, G * power 3.1.9.2, Düsseldorf, Germany), comparable to similar studies [38,39].” Lines 132-135.

4- The description of the balance training protocol (page 6, lines 104-109) is very brief. It is essential to provide additional information regarding the duration of each session, the progression of the exercises throughout the sessions, and the organization of each session (warm-up, main activities, etc.).

We thank the reviewer for the suggestion. Additional information has been added as supplementary material. It now reads:

“Participants first trained for 30 minutes individually between pre and post1 measurement timepoint on a single day. Then, for three weeks with a frequency of three times per week, for 45 minutes per session, they trained in a group of 4 to 8 people. A training session consisted of blocks of 40-60 second exercises in which balance was challenged by different surface conditions, static conditions, perturbations, and dual tasks. Over the course of the training period, the difficulty level was increased by using more challenging exercises, challenging surface conditions (foam, balance boards), and perturbations. Progression criteria were based on the researcher’s observation during the training sessions; if participants were able to perform the task for 60 seconds, the difficulty level would be increased using different balance boards or by limiting the sensory inputs (Figure. S1.1.). Participants were encouraged to train at their individual balance ability level. To test the transfer of acquired skill to gait, none of the exercises included stepping, jumping, or locomotion. To maintain safety, exercises were carried out in groups of two under supervision of the researchers.”

5- When describing the FWHM and CoA (page 9, lines 186-192), include the meaning of these variables for those not familiarized with these variables.

Thank you for the comment. The following text has been added to the manuscript:

"The FWHM is defined as the number of data points above half of the maximum of the activation profile, after subtracting the minimum activation [56]. In addition, we evaluated the CoA (indicating the center of the distribution of activation timing within a gait cycle) per stride, defined as the angle of the vector that points to the center of mass in the activation profile transformed to polar coordinates [57]. The FWHM metric reflects the duration of activation but is naïve of timing of activation. The CoA metric reflects the timing of activation but is naïve of duration of activation.” Lines 263-269.

6- For the beam walking task, did the participants step off the beam at any moment? What was the maximum foot placement error accepted when the participant eventually did not step with the entire foot over the beam?

Thank you for the comment. We did not dichotomize success or failure in the beam walking trial. Participants stepped outside the beam sometimes, but this did not lead to task failure, as the beam was a virtual beam. Instead, we measured where participants placed their feet relative to the target (beam) pre and post training. No participant had a near-fall, nor did anyone have to stop the trial. The foot placement error was determined as the distance between the outer edge of the foot and the edge of the beam. We measured this difference for every step, with a foot placement error indicating that a part or the whole foot was placed outside the beam. When the foot was completely inside the beam, the error was zero. Based on a comment of the first reviewer, we have now added two new measures to our gait evaluation: step width and step width variability. After training we found decreased step width in narrow-base walking and decreased step width variability in both normal and narrow-base walking.

7- Since the beam width (12 cm) is remarkably close to the average foot width, how foot placement error could be affected by balance training? Wouldn't you have a floor effect for this variable?

Thank you for the comment. It should be noted, when interpreting the 1.5 cm mean step error, that the width of the beam is 12 cm whilst average foot width is about 10 cm (Jurca, Zabkar, Dzeroski, 2019). This means that participants, on average, had a 2 cm margin in foot placement that was not seen as error. So, the task did not require perfect control. Additionally, the percentages of steps inside the beam were 27.6% (SE 3.9%) pre-training and 33.3% (SE 4.4%) post-training. Although these differences were not significant (p=0.11), we do believe this suggests that there is ample room for improvement. The best recorded performance was 67% of steps within the beam, which demonstrated to us that scores much higher than the mean could be obtained, and that even these best performances were not close to the end of the scale.

The % of the steps within the beam has been added to the manuscript as below:

“Also, the percentage of the steps within the beam did not change significantly with Training (F2,40= 2.934, p = 0.065; Fig 3.b).” Lines 293-294.

However, as we mentioned in previous response, we have added two new measures to our gait evaluation which shows different result. We found decreased step width in narrow-base and decreased step width variability in both conditions after training

8- Page 13 (lines 286-287): what do you mean by "robustness of gait"?

Thank you for the comment. Robustness of gait refers to the ability to prevent falls in the presence of perturbations and managing to regain balance quickly. However, this part is removed from the discussion.

9- The discussion about the absence of effect of the balance training in all investigated variables was incredibly brief (page 14, lines 313-317). Most of the discussion is focused on the effects of the base of support manipulation. The discussion regarding the balance training needs to improve to explain the absence of an effect. Only arguing that the duration of the training protocol was short is not enough. Why did you choose a 3-week intervention if studies are showing that more time is needed to observe gait improvements? Were the variables not sensitive enough to capture gains due to balance training?

Thank you for your suggestions. Addition of the step width outcome measures, suggested by the reviewer, have changed interpretation of the effect of balance training. The discussion has been updated and restructured into the following sections general findings, effect of training, effect of narrow base, limitations, and conclusions. The section ‘Transfer of training effects’ in the discussion now has a more extensive discussion on the effect of training on the outcome measures.

The measures were sensitive enough to show the modulation between narrow-base and normal walking, therefore, the lack of sensitivity of the measures is unlikely.

10- I recommend including a paragraph with the limitations of this study.

Thank you for your suggestion. The discussion has been updated and the limitations of the study are addressed in lines 508-525.

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

Yes

1. Hu MH, Woollacott MH. Multisensory training of standing balance in older adults: II. Kinematic and electromyographic postural responses. Journals Gerontol 1994;49:.

2. Bohm S, Mandla-Liebsch M, Mersmann F, Arampatzis A. Exercise of Dynamic Stability in the Presence of Perturbations Elicit Fast Improvements of Simulated Fall Recovery and Strength in Older Adults: A Randomized Controlled Trial. Front Sport Act Living 2020;2:1–10.

3. Sawers A, Allen JL, Ting LH. Long-term training modifies the modular structure and organization of walking balance control. J Neurophysiol 2015;114:3359–3373.

4. Neptune RR, McGowan CP. Muscle contributions to frontal plane angular momentum during walking. J Biomech 2016;49:2975–2981.

5. Vieira TMM, Minetto MA, Hodson-Tole EF, Botter A. How much does the human medial gastrocnemius muscle contribute to ankle torques outside the sagittal plane? Hum Mov Sci 2013;32:753–767.

6. Sozzi S, Honeine JL, Do MC, Schieppati M. Leg muscle activity during tandem stance and the control of body balance in the frontal plane. Clin Neurophysiol 2013;124:1175–1186.

7. Hien TD, Tuan D Van, At P Van, Son LH. Novel algorithm for non-negative matrix factorization. New Math Nat Comput 2015;11:121–133.

8. Rubenstein LZ, Josephson KR, Robbins AS. Falls in the nursing home. Ann Intern Med 1994;121:442–451.

9. Berg WP, Alessio HM, Mills EM, Tong C. Circumstances and consequences of falls in independent community-dwelling older adults. Age Ageing 1997;26:261–268.

10. Sawers A, Allen JL, Ting LH. Long-term training modifies the modular structure and organization of walking balance control. J Neurophysiol 2015;114:3359–3373.

11. Chvatal SA, Ting LH. Common muscle synergies for balance and walking. Front Comput Neurosci 2013;7:1–14.

12. Bekius A, Bach MM, van der Krogt MM, de Vries R, Buizer AI, Dominici N. Muscle Synergies During Walking in Children With Cerebral Palsy: A Systematic Review. Front Physiol 2020;11:.

13. Chvatal SA, Torres-Oviedo G, Safavynia SA, Ting LH. Common muscle synergies for control of center of mass and force in nonstepping and stepping postural behaviors. J Neurophysiol 2011;106:999–1015.

Attachment

Submitted filename: Answer_to_Reviewers_PlosOne_Final.docx

Click here for additional data file.^{(599.5KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0242115.r003

Decision Letter 1

Jeremy P Loenneke

29 Mar 2022

PONE-D-20-30116R1Improvement in gait stability in older adults after ten sessions of standing balance trainingPLOS ONE

Dear Dr. van Dieën,

All three reviewers have found merit with your manuscript, however, there are still some suggested changes. Please pay particular attention to Reviewer 1 who has additional comments that need to be addressed within the manuscript (e.g. great justification in places).

Please submit your revised manuscript by May 13 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Jeremy P Loenneke

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #1: The changes based on the previous review have strengthened the manuscript. However, there are a few issues to be addressed to increase clarity and readability.

Major issues:

1. Justification for step width and step width variability based on stability/control of gait. While we (review team) suggested the authors examine this measure, the authors need to provide strong justifications for why these variables were included and add appropriate hypotheses. Line 107 and lines 122-126 are redundant (“The two latter variables were added to the analysis upon a reviewer’s suggestion” line 107). I appreciate that the authors clearly and explicitly state that SW and SWvar were not initially included. I recommend removing the statement from line 107 and keeping lines 122-126. However, you should not rely only on the suggestion of a reviewer, there must be a rationale beyond suggestion of a third party. Increase the justification for these measures based on stability/empirical data. You state earlier that SW and SWvar are associated with age and with older fallers (lines 65-67), expand on that.

2. Skill transfer. Justify the use of the terminology ‘skill transfer’. Step width and step width variability are, of course, not observed during standing, so explicitly state the skill that is transferred? What you observed is a change in gait parameters after standing balance training. This may or may not be ‘skill transfer’, so clarify what skill was transferred and the rationale/evidence that support this interpretation.

3. Adaptable and plastic. Line 376 - Authors used the set of terms – adaptable (modulated between conditions) and plastic (modified by training) – for the first time in the discussion. These terms increase clarity when describing/interpreting the different statistical outcomes. Thus, it would be helpful to include and define these terms in the introduction and perhaps build them into the purpose/aim/hypotheses. Similarly, line 384-386 seems to be related to the condition comparison effects, but not training effects. Be explicit. Please also explain if/how these terms relate to motor modules? We also suggest you use the terms in your headings.

Minor issues:

Line 37-38 – for clarity change “…skills acquired by balance training programs…” to “…skills acquired by standing balance training programs…” (clarification added as many balance training programs including gait activities)

Line 43 – similar comment as above – were the four references [4-8] only examining standing balance training, or balance training that went beyond standing tasks?

Line 48-56- This paragraph is not clear. What is an ‘altered modulation’ in the last sentence – this seems to be a change in gait that has changed (this may be related to the terms adaptable and plastic, and thus may increase clarity by adding the terms to the introduction)? Also, I don’t follow the logic – if skill transfer (from standing to gait) occurs, then you expect gait to be different between normal and narrow-base walking (NW and NBW, respectively)? Or do you expect a *greater* difference between NW and NBW after training than before training? If the latter is an accurate description of the expectation, why is that expected? Further, the following statement is unclear - “increased confidence after training may result in less adaptation to a challenging condition”. I would assume that increased confidence in a task would allow more adaptations. Also, confidence is included in this paragraph, but is only superficially addressed in one sentence, and is not included in the concluding statement of the paragraph.

Line 87 – ‘similarly widened activation profiles’ – similar to what?

Line 134 - Why did authors assume effect size of 0.44. Is it a partial eta squared value?

Lien 158 - specify perturbations (self-perturbation or externally applied, etc.)

Line 200- How did authors quantify leg dominance?

Line 428 - Authors compared motor modules of novice vs expert ballet dancers and specify that their training period was ten years. It would be helpful to relate how that fits in with the current results.

Line 451 - redundant text with the text in methods section

Editorial comments:

Line 246 - Underscores before the reference

Line 291 onwards - At multiple locations, “T” is capitalized for training in the middle of the sentence

Line 398- Be consistent in format for reporting mean +/- SE 27.6% (SE 3.9%) pre-training and 33.3% (SE 4.4%)

Reviewer #2: (No Response)

Reviewer #3: I want to thank the authors for their thoughtful revisions. The new results and the authors’ modifications improved the manuscript considerably. I have just a minor comment. In three parts of the manuscript (two in the Introduction and one in the Discussion), the authors commented on one of the reviewer’s suggestions of adding step width and step width variability variables to argue that these analyses were not pre-planned. You could mention this only in the Discussion (p. 19, lines 451-453). I do not see the reason to keep repeating this information in different parts of the manuscript.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

PLoS One. 2022 Jul 27;17(7):e0242115. doi: 10.1371/journal.pone.0242115.r004

Author response to Decision Letter 1

3 May 2022

We would like to thank the reviewers for their valuable comments. Without their comments the article would have not reached this level of quality. Below there are point-by-point answers to reviewers’ comments.

Reviewer #1: The changes based on the previous review have strengthened the manuscript. However, there are a few issues to be addressed to increase clarity and readability.

Major issues:

Line 107 and lines 122-126 are redundant (“The two latter variables were added to the analysis upon a reviewer’s suggestion” line 107). I appreciate that the authors clearly and explicitly state that SW and SWvar were not initially included. I recommend removing the statement from line 107 and keeping lines 122-126.

However, you should not rely only on the suggestion of a reviewer, there must be a rationale beyond suggestion of a third party. Increase the justification for these measures based on stability/empirical data. You state earlier that SW and SWvar are associated with age and with older fallers (lines 65-67), expand on that.

Thank you for the suggestion. Line 107 has been removed. The texts below have been added to the introduction:

(Lines 67-74): Mediolateral gait stability is thought to be actively controlled by adjusting foot placement (Bruijn & Van Dieën, 2018), as centre of mass kinematics during the preceding swing phase strongly correlate with foot position in the next stance phase (Hurt et al., 2010; Wang & Srinivasan, 2014). The strong coupling between centre of mass kinematics and foot placement is found to decrease in conditions in which gait is stabilized, such as in external lateral stabilization, increasing confidence that lateral gait stability is indeed controlled by foot placement adjustment (Mahaki et al., 2019). This decoupling coincides with a decrease in step width and step width variability (Donelan et al., 2004). Moreover, increased step width and step width variability have been found in older compared to young adults and in fallers compared to non-fallers (Callisaya et al., 2010; Nordin et al., 2010; Skiadopoulos et al., 2020).

(Lines 76-78): Consequently, step width and step width variability are considered to be proxies of the quality of active mediolateral control of gait stability using foot placement adjustments and may therefore be useful to evaluate the effect of balance training.

(Lines 135-146): Even though step width and step width variability seem to be sensitive to quality of balance control in gait, step width measures are not necessarily affected by postural balance training. Postural balance is controlled using torques around the ankle and hip of the standing leg (Nashner, 1985; Runge et al., 1999), but does not include stepping (i.e. foot placement). However, as in postural balance, the stance leg is used to regulate stabilizing torques in the stance phase gait (Reimann et al., 2018). This stance leg control co-determines the control of foot placement (van Leeuwen et al., 2021) and therefore reduced step width and step width variability are likely results of improved control by the stance leg. These predictions on step width and step width variability were conceived, based on reviewer comments, after the data were processed and analyzed. These predictions are thus explorative and aim to drive future research (Kerr, 1998; Rowbottom & Alexander, 2012) and we ask the reader to consider the limitations of the evidence provided by these variables.

We thank the reviewer for the comment. We agree that strictly speaking we do not provide evidence for transfer of acquired skill, while we do provide evidence for transfer of an effect of training. We now use the more neutral term transfer, omitting skill. There are changes throughout the manuscript in Key words, introduction, and discussion.

Lines 37-39: Thus, if effects of standing balance training programs do not transfer to improvements in gait stability, they are unlikely to decrease the number of falls.

Lines 44-46: Consequently, the existence of transfer from standing balance training to gait, as well as the mechanisms underlying such a transfer, if present, are insufficiently clear.

Lines 109-113: In the current study, we aimed to investigate the transfer of effects of standing balance training to normal and narrow-base walking in older adults, as well as the adaptation of older adults to narrow-base walking.

Lines 146-148: While this result is in line with our primary hypothesis and indicates transfer of effects of standing balance training to gait, in a strict sense it cannot be considered a planned analysis.

Line 398: We studied whether effects of standing balance training transferred to gait in older adults.

Lines 403-404: We also expected transfer of training effects to be most pronounced in the narrow-base condition.

Lines 415-416: In this study, after ten sessions we found transfer from standing balance training to gait.

Lines 474-475: (to start with stationary exercises and transfer the training effects to walking)

Thank you for the comment. We agree these two terms should have been introduced earlier in the introduction. We now introduce adaptable and plastic in the introduction.

Lines 111- 113: The modulation of balance control between two conditions aimed to test adaptability of balance control to environmental constraints and effects of training were studied to analyse the plasticity of balance control.

Lines 48-50: Fall prevention training programs aim to improve balance control employing plastsicity of the neuromuscular system . To prevent falls, one needs to be able to adapt gait when facing environmental challenges, such as when forced to walk with a narrow step width.

Minor issues:

Thank you for the comment. Skills acquired by balance training programs is now changed to effects of standing balance training programs as suggested.

Line 43 – similar comment as above – were the four references [4-8] only examining standing balance training, or balance training that went beyond standing tasks?

Studies 4, 5, and 8 refer to standing balance training. Study 6 used gait and 7 a combination of gait and standing and sitting, lying on the floor, etc. The latter two references have thus been removed.

The text now reads as (Lines 41-43): On the other hand, transfer to gait stability from solely standing balance training [4-6] is suggested by improved clinical balance scores, gait parameters, and performance on the timed up and go, and other tests.

Thank you for the comments. Indeed, we expect a change in the difference between NW and NBW after training, that is an interaction between Training and Condition. However, we were unsure about the direction of this effect. We have rephrased this section to (Lines: 49-58):

“To prevent falls, one needs to be able to adapt gait when facing environmental challenges, such as when forced to walk with a narrow step width. Older adults show more pronounced adaptations to narrow-base walking than young adults [11], possibly because they are more cautious in the presence of balance threats [12]. Therefore, an interaction between training and stabilizing demands may be expected. On the one hand, increased confidence after training may result in less adaptation to a challenging condition. On the other hand, balance training may enhance the ability to adapt to challenging conditions. Therefore, if transfer of standing balance training to gait occurs, an altered modulation of gait between normal and narrow-base walking might be expected after training, but the direction of change is unpredictable.”

We expected that training might increase confidence. We did not elaborate on the confidence in the main manuscript, since it was not the focus of the research. However, we provided as short introduction to the topic and the results in the supplementary material for curious readers to test our speculation (copied below).

Several studies showed a strong correlation between concern of falling and balance performance (Thiamwong & Suwanno, 2014; Young & Mark Williams, 2015). It has been shown that poor balance performance is mediated by changes in the allocation of attention in the presence of concern of falling [(Young & Mark Williams, 2015)]. Concern of falling is reduced after training in older adults, which is associated with improved balance performance (Kumar et al., 2016; Thiamwong & Suwanno, 2014). To assess concern of falling, we used the Falls Efficacy Scale International (FES-I) questionnaire at pre, post2, and retention time-points (Kempen et al., 2007). FES-I outcomes are on a scale of 16 to 64, with 16 indicating minimum concern of falling and 64 severe concern of falling.

A repeated measures ANOVA indicated that concern of falling was affected by balance training (F2,42= 4.37, P = 0.039; Figure S2.1). Post-hoc analysis showed that concern of falling was not significantly changed immediately after the training program, but was decreased at retention (t = 2.16, p = 0.072; t = 2.82, p = 0.022, respectively), implying that weeks after training participants felt more confident about their balance ability.

Line 87 – ‘similarly widened activation profiles’ – similar to what?

Thank you for the comment. The word “similarly” has been removed.

Line 134 - Why did authors assume effect size of 0.44. Is it a partial eta squared value?

We chose this effect size from the meta-analysis article, DOI 10.1007/s40279-015-0375-y, which expressed effects size using Cohen’s f. This has now been mentioned in the text as:

Lines 154-158: The required sample size was estimated at twenty-two based on power analysis for an F test of a repeated measures ANOVA, assuming a Cohen’s f of 0.44 (based on meta-analysis of the effect of standing balance training on steady-state balance (Muehlbauer et al., 2015)) and a correlation among repeated measures of 0.6 (β = 0.8, G * power 3.1.9.2, Düsseldorf, Germany), comparable to similar studies (Bisson et al., 2007; Nagy et al., 2007).

Line 158 - specify perturbations (self-perturbation or externally applied, etc.)

We have further specified the perturbations, and the text now reads (Lines 181-182): self-perturbations and external perturbations while catching a ball in a dual tasking exercise.

Line 200- How did authors quantify leg dominance?

We have added this to the methods section as (Lines 225-227): The preferred stance leg was reported by the participant prior to the experiment and confirmed by the experimenter by asking the participant to kick an imaginary soccer ball. The supporting leg was considered the preferred stance leg (Alizadehsaravi et al., 2022).

The mentioned study was a cross-sectional study, in which the experts had a minimum of 10 years training, including all sorts of training in ballet, which could vary from standing balance to gait and more, but with balance control as a core element. The aim of mentioning this study was to show that a long-term training may cause reorganization and permanent changes.

The text (Lines 447-455) now reads as: Even though we found evidence of transfer, gait synergies were not affected by training. Synergy metrics may be insensitive, because subtle variations in the activation of some muscles, for example through improved feedback control, underlie improved balance. Alternatively, changes may have occurred in muscles that we did not include in our measurements. Young adults with years of experience in balance training in ballet showed different motor modules during narrow-base walking compared to novice ballet dancers (Chvatal & Ting, 2013), indicating that long-term effects of training may be apparent in muscle synergies. Note, however, that these ballet dancers’ training experience most certainly included gait-related tasks (Chvatal & Ting, 2013). In addition, the dancers had trained for at least ten years.

Line 451- redundant text with the text in methods section

We removed the redundant text and kept the one in the introduction.

Editorial comments: Line 246 - Underscores before the reference

The underscores have been removed, thank you for noticing that.

Line 291 onwards - At multiple locations, “T” is capitalized for training in the middle of the sentence

Thank you for the comment. We capitalized the T intentionally where we refer to Training as a factor in the ANOVA.

Line 398- Be consistent in format for reporting mean +/- SE 27.6% (SE 3.9%) pre-training and 33.3% (SE 4.4%)

Thanks. We reported the SD now.

Reviewer #2: (No Response)

Thank you for the compliment and the time you devoted to review our paper. We have now removed the first statement and acknowledged the changes at the end of the introduction as reviewer #1 suggested and the discussion as you suggested.

References:

Alizadehsaravi, L., Koster, R. A. J., Muijres, W., Maas, H., Bruijn, S. M., & van Dieën, J. H. (2022). The underlying mechanisms of improved balance after one and ten sessions of balance training in older adults. Human Movement Science, 81(December 2021), 1–14. https://doi.org/10.1016/j.humov.2021.102910

Bisson, E., Contant, B., Sveistrup, H., & Lajoie, Y. (2007). Functional balance and dual-task reaction times in older adults are improved by virtual reality and biofeedback training. Cyberpsychology and Behavior, 10(1), 16–23. https://doi.org/10.1089/cpb.2006.9997

Bruijn, S. M., & Van Dieën, J. H. (2018). Control of human gait stability through foot placement. Journal of the Royal Society Interface, 15(143). https://doi.org/10.1098/rsif.2017.0816

Callisaya, M. L., Blizzard, L., Schmidt, M. D., McGinley, J. L., & Srikanth, V. K. (2010). Ageing and gait variability-a population-based study of older people. Age and Ageing, 39(2), 191–197. https://doi.org/10.1093/ageing/afp250

Chvatal, S. A., & Ting, L. H. (2013). Common muscle synergies for balance and walking. Frontiers in Computational Neuroscience, 7(May), 1–14. https://doi.org/10.3389/fncom.2013.00048

Donelan, J. M., Shipman, D. W., Kram, R., & Kuo, A. D. (2004). Mechanical and metabolic requirements for active lateral stabilization in human walking. Journal of Biomechanics, 37(6), 827–835. https://doi.org/10.1016/j.jbiomech.2003.06.002

Hurt, C. P., Rosenblatt, N., Crenshaw, J. R., & Grabiner, M. D. (2010). Variation in trunk kinematics influences variation in step width during treadmill walking by older and younger adults. Gait and Posture, 31(4), 461–464. https://doi.org/10.1016/j.gaitpost.2010.02.001

Kempen, G. I. J. M., Zijlstra, G. A. R., & van Haastregt, J. C. M. (2007). [The assessment of fear of falling with the Falls Efficacy Scale-International (FES-I). Development and psychometric properties in Dutch elderly]. Tijdschrift voor gerontologie en geriatrie, 38(4), 204–212.

Kerr, N. L. (1998). HARKing: Hypothesizing after the results are known. Personality and Social Psychology Review, 2(3), 196–217. https://doi.org/10.1207/s15327957pspr0203_4

Kumar, A., Delbaere, K., Zijlstra, G. A. R., Carpenter, H., Iliffe, S., Masud, T., Skelton, D., Morris, R., & Kendrick, D. (2016). Exercise for reducing fear of falling in older people living in the community: Cochrane systematic review and Meta-Analysis. Age and Ageing, 45(3), 345–352. https://doi.org/10.1093/ageing/afw036

Mahaki, M., Bruijn, S. M., & Van Dieën, J. H. (2019). The effect of external lateral stabilization on the use of foot placement to control mediolateral stability in walking and running. PeerJ, 2019(10). https://doi.org/10.7717/peerj.7939

Muehlbauer, T., Gollhofer, A., Lesinski, M., Hortoba, T., & Granacher, U. (2015). Effects of Balance Training on Balance Performance in Healthy Older Adults : A Systematic Review and Meta-analysis. Sports Med, 45(45), 1721–1738. https://doi.org/10.1007/s40279-015-0375-y

Nagy, E., Feher-Kiss, A., Barnai, M., Domján-Preszner, A., Angyan, L., & Horvath, G. (2007). Postural control in elderly subjects participating in balance training. European Journal of Applied Physiology, 100(1), 97–104. https://doi.org/10.1007/s00421-007-0407-x

Nashner, L. M. (1985). The organization of human postural movements : A formal basis and experimental synthesis. 8, 135–172. https://doi.org/10.1017/S0140525X00020008

Nordin, E., Moe-Nilssen, R., Ramnemark, A., & Lundin-Olsson, L. (2010). Changes in step-width during dual-task walking predicts falls. Gait & Posture, 32(1), 92–97. https://doi.org/https://doi.org/10.1016/j.gaitpost.2010.03.012

Reimann, H., Fettrow, T., & Jeka, J. J. (2018). Strategies for the Control of Balance During Locomotion. Kinesiology Review, 7(1), 18–25. https://doi.org/10.1123/kr.2017-0053

Rowbottom, D. P., & Alexander, R. M. N. (2012). The role of hypotheses in biomechanical research. Science in Context, 25(2), 247–262. https://doi.org/10.1017/S0269889712000051

Runge, C. F., Shupert, C. L., Horak, F. B., & Zajac, F. E. (1999). Ankle and hip postural strategies defined by joint torques. Gait and Posture, 10(2), 161–170. https://doi.org/10.1016/S0966-6362(99)00032-6

Skiadopoulos, A., Moore, E. E., Sayles, H. R., Schmid, K. K., & Stergiou, N. (2020). Step width variability as a discriminator of age-related gait changes. Journal of NeuroEngineering and Rehabilitation, 17(1), 1–13. https://doi.org/10.1186/s12984-020-00671-9

Thiamwong, L., & Suwanno, J. (2014). Effects of simple balance training on balance performance and fear of falling in rural older adults. International Journal of Gerontology, 8(3), 143–146. https://doi.org/10.1016/j.ijge.2013.08.011

van Leeuwen, A. M., van Dieën, J. H., Daffertshofer, A., & Bruijn, S. M. (2021). Ankle muscles drive mediolateral center of pressure control to ensure stable steady state gait. Scientific Reports, 11(1), 1–14. https://doi.org/10.1038/s41598-021-00463-8

Wang, Y., & Srinivasan, M. (2014). Stepping in the direction of the fall: The next foot placement can be predicted from current upper body state in steady-state walking. Biology Letters, 10(9). https://doi.org/10.1098/rsbl.2014.0405

Young, W. R., & Mark Williams, A. (2015). How fear of falling can increase fall-risk in older adults: Applying psychological theory to practical observations. Gait and Posture, 41(1), 7–12. https://doi.org/10.1016/j.gaitpost.2014.09.006

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(49.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0242115.r005

Decision Letter 2

Jeremy P Loenneke

29 Jun 2022

Improvement in gait stability in older adults after ten sessions of standing balance training

PONE-D-20-30116R2

Dear Dr. van Dieën,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jeremy P Loenneke

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #1: The authors have addressed all of my concerns. This is nice piece of work, and will stimulate other research.

Reviewer #2: (No Response)

Reviewer #3: I have no further comments. The authors addressed all concerns raised in the previous rounds. Congratulations on the manuscript.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

PLoS One. doi: 10.1371/journal.pone.0242115.r006

Acceptance letter

Jeremy P Loenneke

5 Jul 2022

PONE-D-20-30116R2

Improvement in gait stability in older adults after ten sessions of standing balance training.

Dear Dr. van Dieën:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jeremy P Loenneke

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 File

(DOCX)

Click here for additional data file.^{(298.8KB, docx)}

S2 File

(MAT)

Click here for additional data file.^{(243.8KB, mat)}

S3 File

(CSV)

Click here for additional data file.^{(601B, csv)}

S4 File

(CSV)

Click here for additional data file.^{(7.6KB, csv)}

S5 File

(MAT)

Click here for additional data file.^{(242.7KB, mat)}

Attachment

Submitted filename: Answer_to_Reviewers_PlosOne_Final.docx

Click here for additional data file.^{(599.5KB, docx)}

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(49.8KB, docx)}

Data Availability Statement

All relevant data are within the article and its Supporting information files.

[pone.0242115.ref001] 1.Rubenstein LZ, Josephson KR, Robbins AS. Falls in the nursing home. Ann Intern Med. 1994;121: 442–451. doi: 10.7326/0003-4819-121-6-199409150-00009 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref002] 2.Berg WP, Alessio HM, Mills EM, Tong C. Circumstances and consequences of falls in independent community-dwelling older adults. Age Ageing. 1997;26: 261–268. doi: 10.1093/ageing/26.4.261 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref003] 3.Giboin L- S, Gruber M, Kramer A. Task-specificity of balance training. Hum Mov Sci. 2015;44: 22–31. doi: 10.1016/j.humov.2015.08.012 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref004] 4.Hirase T, Inokuchi S, Matsusaka N, Okita M. Effects of a Balance Training Program Using a Foam Rubber Pad in Community-Based Older Adults: A Randomized Controlled Trial. J Geriatr Phys Ther. 2015;38: 62–70. doi: 10.1519/JPT.0000000000000023 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref005] 5.Kümmel J, Kramer A, Giboin LS, Gruber M. Specificity of Balance Training in Healthy Individuals: A Systematic Review and Meta-Analysis. Sport Med. 2016;46: 1261–1271. doi: 10.1007/s40279-016-0515-z [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref006] 6.Nam HC, Cha HG, Kim MK. The effects of exercising on an unstable surface on the gait and balance ability of normal adults. J Phys Ther Sci. 2016;28: 2102–2104. doi: 10.1589/jpts.28.2102 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref007] 7.Nyman SR. Tai Chi for the Prevention of Falls Among Older Adults: A Critical Analysis of the Evidence. J Aging Phys Act. 2021;29: 343–352. doi: 10.1123/japa.2020-0155 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref008] 8.van Dieën JH, van Leeuwen M, Faber GS. Learning to balance on one leg: motor strategy and sensory weighting. Am J Physiol Circ Physiol. 2015. doi: 10.1152/jn.00434.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref009] 9.Toebes MJP, Hoozemans MJM, Furrer R, Dekker J, Van Dieën JH. Local dynamic stability and variability of gait are associated with fall history in elderly subjects. Gait Posture. 2012;36: 527–531. doi: 10.1016/j.gaitpost.2012.05.016 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref010] 10.Hof AL, Gazendam MGJ, Sinke WE. The condition for dynamic stability. J Biomech. 2005;38: 1–8. doi: 10.1016/j.jbiomech.2004.03.025 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref011] 11.Arvin M, Mazaheri M, Hoozemans MJM, Pijnappels M, Burger BJ, Verschueren SMP, et al. Effects of narrow base gait on mediolateral balance control in young and older adults. J Biomech. 2016;49: 1264–1267. doi: 10.1016/j.jbiomech.2016.03.011 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref012] 12.Manuscript A. NIH Public Access. 2013;35: 472–477. [Google Scholar]

[pone.0242115.ref013] 13.Reimold NK, Knapp HA, Henderson RE, Wilson L, Chesnutt AN, Dean JC. Altered active control of step width in response to mediolateral leg perturbations while walking. Sci Rep. 2020;10: 1–15. doi: 10.1038/s41598-020-69052-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref014] 14.Bruijn SM, Van Dieën JH. Control of human gait stability through foot placement. J R Soc Interface. 2018;15. doi: 10.1098/rsif.2017.0816 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref015] 15.Hurt CP, Rosenblatt N, Crenshaw JR, Grabiner MD. Variation in trunk kinematics influences variation in step width during treadmill walking by older and younger adults. Gait Posture. 2010;31: 461–464. doi: 10.1016/j.gaitpost.2010.02.001 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref016] 16.Wang Y, Srinivasan M. Stepping in the direction of the fall: The next foot placement can be predicted from current upper body state in steady-state walking. Biol Lett. 2014;10. doi: 10.1098/rsbl.2014.0405 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref017] 17.Mahaki M, Bruijn SM, Van Dieën JH. The effect of external lateral stabilization on the use of foot placement to control mediolateral stability in walking and running. PeerJ. 2019;2019. doi: 10.7717/peerj.7939 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref018] 18.Donelan JM, Shipman DW, Kram R, Kuo AD. Mechanical and metabolic requirements for active lateral stabilization in human walking. J Biomech. 2004;37: 827–835. doi: 10.1016/j.jbiomech.2003.06.002 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref019] 19.Nordin E, Moe-Nilssen R, Ramnemark A, Lundin-Olsson L. Changes in step-width during dual-task walking predicts falls. Gait Posture. 2010;32: 92–97. doi: 10.1016/j.gaitpost.2010.03.012 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref020] 20.Callisaya ML, Blizzard L, Schmidt MD, McGinley JL, Srikanth VK. Ageing and gait variability-a population-based study of older people. Age Ageing. 2010;39: 191–197. doi: 10.1093/ageing/afp250 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref021] 21.Skiadopoulos A, Moore EE, Sayles HR, Schmid KK, Stergiou N. Step width variability as a discriminator of age-related gait changes. J Neuroeng Rehabil. 2020;17: 1–13. doi: 10.1186/s12984-020-00671-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref022] 22.Ivanenko YP, Cappellini G, Dominici N, Poppele RE, Lacquaniti F. Modular control of limb movements during human locomotion. J Neurosci. 2007;27: 11149–11161. doi: 10.1523/JNEUROSCI.2644-07.2007 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref023] 23.Ivanenko YP, Cappellini G, Poppele RE, Lacquaniti F. Spatiotemporal organization of α‐motoneuron activity in the human spinal cord during different gaits and gait transitions. Eur J Neurosci. 2008;27: 3351–3368. doi: 10.1111/j.1460-9568.2008.06289.x [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref024] 24.Janshen L, Santuz A, Ekizos A, Arampatzis A. Fuzziness of muscle synergies in patients with multiple sclerosis indicates increased robustness of motor control during walking. Sci Rep. 2020;10: 1–14. doi: 10.1038/s41598-020-63788-w [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref025] 25.Chvatal SA, Ting LH. Common muscle synergies for balance and walking. Front Comput Neurosci. 2013;7: 1–14. doi: 10.3389/fncom.2013.00001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref026] 26.Bizzi E, Cheung VCK, d’Avella A, Saltiel P, Tresch M. Combining modules for movement. Brain Res Rev. 2008;57: 125–133. doi: 10.1016/j.brainresrev.2007.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref027] 27.d’Avella A, Saltiel P, Bizzi E. Combinations of muscle synergies in the construction of a natural motor behavior. Nat Neurosci. 2003;6: 300–308. doi: 10.1038/nn1010 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref028] 28.d’Avella A. Modularity for motor control and motor learning. Advances in Experimental Medicine and Biology. Springer, Cham; 2016. pp. 3–19. doi: 10.1007/978-3-319-47313-0_1 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref029] 29.Santuz A, Ekizos A, Eckardt N, Kibele A, Arampatzis A. Challenging human locomotion: Stability and modular organisation in unsteady conditions. Sci Rep. 2018;8: 1–13. doi: 10.1038/s41598-018-21018-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref030] 30.Baggen RJ, va Dieën JH., Van Roie E, Verschueren SM, Giarmatzis G, Delecluse C, et al. Age-related differences in muscle synergy organization during step ascent at different heights and directions. Appl Sci. 2020;10. doi: 10.3390/app10061987 [DOI] [Google Scholar]

[pone.0242115.ref031] 31.da Silva Costa AA, Moraes R, Hortobágyi T, Sawers A. Older adults reduce the complexity and efficiency of neuromuscular control to preserve walking balance. Exp Gerontol. 2020; 111050. doi: 10.1016/j.exger.2020.111050 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref032] 32.Martino G, Ivanenko YP, d’Avella A, Serrao M, Ranavolo A, Draicchio F, et al. Neuromuscular adjustments of gait associated with unstable conditions. J Neurophysiol. 2015;114: 2867–2882. doi: 10.1152/jn.00029.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref033] 33.Wang Y, Watanabe K, Asaka T. Effect of dance on multi-muscle synergies in older adults: a cross-sectional study. BMC Geriatr. 2019;19: 340. doi: 10.1186/s12877-019-1365-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref034] 34.Sawers A, Allen JL, Ting LH. Long-term training modifies the modular structure and organization of walking balance control. J Neurophysiol. 2015;114: 3359–3373. doi: 10.1152/jn.00758.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref035] 35.Alizadehsaravi L, Koster RAJ, Muijres W, Maas H, Bruijn SM, van Dieën JH. The underlying mechanisms of improved balance after one and ten sessions of balance training in older adults. Hum Mov Sci. 2022;81: 1–14. doi: 10.1016/j.humov.2021.102910 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref036] 36.Hilliard MJ, Martinez KM, Janssen I, Edwards B, Mille ML, Zhang Y, et al. Lateral Balance Factors Predict Future Falls in Community-Living Older Adults. Arch Phys Med Rehabil. 2008;89: 1708–1713. doi: 10.1016/j.apmr.2008.01.023 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref037] 37.Maki BE, Holliday PJ, Topper AK. A prospective study of postural balance and risk of falling in an ambulatory and independent elderly population. Journals Gerontol. 1994;49: M72–M84. Available: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=emed3&NEWS=N&AN=1994105124. [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref038] 38.Kluft N, van Dieën JH, Pijnappels M. The degree of misjudgment between perceived and actual gait ability in older adults. Gait Posture. 2017;51: 275–280. doi: 10.1016/j.gaitpost.2016.10.019 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref039] 39.Magnani RM, Bruijn SM, van Dieën JH, Forbes PA. Stabilization demands of walking modulate the vestibular contributions to gait. Sci Rep. 2021;11: 1–13. doi: 10.1038/s41598-021-93037-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref040] 40.Nashner LM. The organization of human postural movements: A formal basis and experimental synthesis. 1985;8: 135–172. doi: 10.1017/S0140525X00020008 [DOI] [Google Scholar]

[pone.0242115.ref041] 41.Runge CF, Shupert CL, Horak FB, Zajac FE. Ankle and hip postural strategies defined by joint torques. Gait Posture. 1999;10: 161–170. doi: 10.1016/s0966-6362(99)00032-6 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref042] 42.Reimann H, Fettrow T, Jeka JJ. Strategies for the Control of Balance During Locomotion. Kinesiol Rev. 2018;7: 18–25. doi: 10.1123/kr.2017-0053 [DOI] [Google Scholar]

[pone.0242115.ref043] 43.van Leeuwen AM, van Dieën JH, Daffertshofer A, Bruijn SM. Ankle muscles drive mediolateral center of pressure control to ensure stable steady state gait. Sci Rep. 2021;11: 1–14. doi: 10.1038/s41598-021-00463-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref044] 44.Kerr NL. HARKing: Hypothesizing after the results are known. Personal Soc Psychol Rev. 1998;2: 196–217. doi: 10.1207/s15327957pspr0203_4 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref045] 45.Rowbottom DP, Alexander RMN. The role of hypotheses in biomechanical research. Sci Context. 2012;25: 247–262. doi: 10.1017/s0269889712000051 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref046] 46.Muehlbauer T, Gollhofer A, Lesinski M, Hortoba T, Granacher U. Effects of Balance Training on Balance Performance in Healthy Older Adults: A Systematic Review and Meta-analysis. Sport Med. 2015;45: 1721–1738. doi: 10.1007/s40279-015-0375-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref047] 47.Bisson E, Contant B, Sveistrup H, Lajoie Y. Functional balance and dual-task reaction times in older adults are improved by virtual reality and biofeedback training. Cyberpsychology Behav. 2007;10: 16–23. doi: 10.1089/cpb.2006.9997 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref048] 48.Nagy E, Feher-Kiss A, Barnai M, Domján-Preszner A, Angyan L, Horvath G. Postural control in elderly subjects participating in balance training. Eur J Appl Physiol. 2007;100: 97–104. doi: 10.1007/s00421-007-0407-x [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref049] 49.Sherrington C, Tiedemann A, Fairhall N, Close JCT, Lord SR. Exercise to prevent falls in older adults: an updated meta-analysis and best practice recommendations. N S W Public Health Bull. 2011;22: 78–83. doi: 10.1071/NB10056 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref050] 50.Bruijn SM, Meijer OG, Beek PJ, Van Dieen JH. Assessing the stability of human locomotion: A review of current measures. J R Soc Interface. 2013;10. doi: 10.1098/rsif.2012.0999 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref051] 51.Bruijn SM, van Dieën JH, Meijer OG, Beek PJ. Statistical precision and sensitivity of measures of dynamic gait stability. J Neurosci Methods. 2009;178: 327–333. doi: 10.1016/j.jneumeth.2008.12.015 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref052] 52.Cappozzo A, Catani F, Della Croce U, Leardini A. Position and orientation in space of bones during movement: anatomical frame definition and determination. Clin Biomech. 1995;10: 171–178. doi: 10.1016/0268-0033(95)91394-t [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref053] 53.Neptune RR, McGowan CP. Muscle contributions to frontal plane angular momentum during walking. J Biomech. 2016;49: 2975–2981. doi: 10.1016/j.jbiomech.2016.07.016 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref054] 54.Vieira TMM, Minetto MA, Hodson-Tole EF, Botter A. How much does the human medial gastrocnemius muscle contribute to ankle torques outside the sagittal plane? Hum Mov Sci. 2013;32: 753–767. doi: 10.1016/j.humov.2013.03.003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref055] 55.Sozzi S, Honeine JL, Do MC, Schieppati M. Leg muscle activity during tandem stance and the control of body balance in the frontal plane. Clin Neurophysiol. 2013;124: 1175–1186. doi: 10.1016/j.clinph.2012.12.001 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref056] 56.Afschrift M, Pitto L, Aerts W, van Deursen R, Jonkers I, De Groote F. Modulation of gluteus medius activity reflects the potential of the muscle to meet the mechanical demands during perturbed walking. Sci Rep. 2018;8: 1–11. doi: 10.1038/s41598-017-17765-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref057] 57.Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol. 2000;10: 361–374. doi: 10.1016/s1050-6411(00)00027-4 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref058] 58.Roerdink M, Coolen BH, Clairbois BHE, Lamoth CJC, Beek PJ. Online gait event detection using a large force platform embedded in a treadmill. J Biomech. 2008;41: 2628–2632. doi: 10.1016/j.jbiomech.2008.06.023 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref059] 59.Bruijn SM, Van Dieën JH, Daffertshofer A. Beta activity in the premotor cortex is increased during stabilized as compared to normal walking. Front Hum Neurosci. 2015;9: 1–13. doi: 10.3389/fnhum.2015.00001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref060] 60.Rosenstein MT, Collins JJ, De Luca CJ. A practical method for calculating largest Lyapunov exponents from small data sets. Phys D Nonlinear Phenom. 1993;65: 117–134. doi: 10.1016/0167-2789(93)90009-P [DOI] [Google Scholar]

[pone.0242115.ref061] 61.Bruijn S. SjoerdBruijn/LocalDynamicStability: First release. 2017 [cited 2 Sep 2020]. 10.5281/ZENODO.573285.

[pone.0242115.ref062] 62.Lee DD, Seung HS. Learning the parts of objects by non-negative matrix factorization. Nature. 1999;401: 788–791. doi: 10.1038/44565 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref063] 63.Oliveira ASC, Gizzi L, Kersting UG, Farina D. Modular organization of balance control following perturbations during walking. J Neurophysiol. 2012;108: 1895–1906. doi: 10.1152/jn.00217.2012 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref064] 64.Patla AE. Some characteristics of EMG patterns during locomotion: implications for the locomotor control process. J Mot Behav. 1985;17: 443–461. doi: 10.1080/00222895.1985.10735360 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref065] 65.Ivanenko YP, Poppele RE, Lacquaniti F. Motor control programs and walking. Neurosci a Rev J bringing Neurobiol Neurol psychiatry. 2006;12: 339–348. doi: 10.1177/1073858406287987 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref066] 66.Harrison D, Kanji GK. The development of analysis of variance for circular data. J Appl Stat. 1988;15: 197–223. doi: 10.1080/02664768800000026 [DOI] [Google Scholar]

[pone.0242115.ref067] 67.Jurca A, Žabkar J, Džeroski S. Analysis of 1.2 million foot scans from North America, Europe and Asia. Sci Rep. 2019;9: 1–11. doi: 10.1038/s41598-018-37186-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref068] 68.Socie M, Sandroff B, Pula J, Hsiao-Wecksler E, Motl R, Sosnoff J. Footfall Placement Variability and Falls in Multiple Sclerosis. Ann Biomed Eng. 2012;41. doi: 10.1007/s10439-012-0636-y [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref069] 69.Dean JC, Kautz SA. Foot placement control and gait instability among people with stroke Jesse. J Rehabil Res Dev. 2016;52: 577–590. doi: 10.1682/JRRD.2014.09.0207.Foot [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref070] 70.Maki BE. Gait changes in older adults: predictors of falls or indicators of fear. J Am Geriatr Soc. 1997;45: 313–320. doi: 10.1111/j.1532-5415.1997.tb00946.x [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref071] 71.Martínez-Amat A; Hita-Contreras F; Lomas-Vega R; Caballero-Martínez I; Alvarez P.J.; Martínez-López E. Effects of 12-week proprioception training program on postural stability, gait, and balance in older adults: A controlled clinical trial. J Strength Cond. 2013;27: 2180–2188. [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref072] 72.Ambrosini E, Parati M, Peri E, De Marchis C, Nava C, Pedrocchi A, et al. Changes in leg cycling muscle synergies after training augmented by functional electrical stimulation in subacute stroke survivors: a pilot study. J Neuroeng Rehabil. 2020;17: 1–14. doi: 10.1186/s12984-019-0634-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref073] 73.Sherrington C, Michaleff ZA, Fairhall N, Paul SS, Tiedemann A, Whitney J, et al. Exercise to prevent falls in older adults: An updated systematic review and meta-analysis. Br J Sports Med. 2017;51: 1749–1757. doi: 10.1136/bjsports-2016-096547 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref074] 74.McAndrew Young PM, Dingwell JB. Voluntary changes in step width and step length during human walking affect dynamic margins of stability. Gait Posture. 2012/04/01. 2012;36: 219–224. doi: 10.1016/j.gaitpost.2012.02.020 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref075] 75.Shih H- JS, Gordon J, Kulig K. Trunk control during gait: Walking with wide and narrow step widths present distinct challenges. J Biomech. 2021;114: 110135. doi: 10.1016/j.jbiomech.2020.110135 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref076] 76.Monaco V, Ghionzoli A, Micera S. Age-related modifications of muscle synergies and spinal cord activity during locomotion. J Neurophysiol. 2010;104: 2092–2102. doi: 10.1152/jn.00525.2009 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref077] 77.Ivanenko YP, Poppele RE, Lacquaniti F. Five basic muscle activation patterns account for muscle activity during human locomotion. J Physiol. 2004;556: 267–282. doi: 10.1113/jphysiol.2003.057174 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref078] 78.Oliveira AS, Gizzi L, Farina D, Kersting UG. Motor modules of human locomotion: Influence of EMG averaging, concatenation, and number of step cycles. Front Hum Neurosci. 2014;8. doi: 10.3389/fnhum.2014.00335 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0242115.ref079] 79.Cappellini G, Ivanenko YP, Poppele RE, Lacquaniti F. Motor patterns in human walking and running. J Neurophysiol. 2006;95: 3426–3437. doi: 10.1152/jn.00081.2006 [DOI] [PubMed] [Google Scholar]

[pone.0242115.ref080] 80.van den Hoorn W, Hodges PW, van Dieen JH, Hug F. Effect of acute noxious stimulation to the leg or back on muscle synergies during walking. J Neurophysiol. 2015;113: 244–254. doi: 10.1152/jn.00557.2014 [DOI] [PubMed] [Google Scholar]

PERMALINK

Improvement in gait stability in older adults after ten sessions of standing balance training

Leila Alizadehsaravi

Sjoerd M Bruijn

Wouter Muijres

Ruud A J Koster

Jaap H van Dieën

Roles

Abstract

Introduction

Methods

Participants

Experimental procedures

Fig 1. Block diagram of the study; training and gait assessment.

Fig 2. Narrow-base walking on a treadmill.

Instrumentation and data acquisition

Data analysis

Gait events

Gait stability

Muscle synergies

Fig 7. Time-invariant muscle weightings of synergies extracted from concatenated data, over all individuals, conditions, and time-points.

Statistics

Results

Gait performance

Fig 3.

Fig 4.

Fig 5.

Fig 6. Local divergence exponents in narrow-base and normal walking at time-points Pre, Post1, and Post2.

Muscle synergies

Fig 8. Activation profiles of the extracted synergies as time series and in polar coordinates in narrow-base and normal walking at time-points Pre (solid), Post1 (dash-dot), and Post2 (dotted).

FWHM

Fig 9. FWHM and CoA of five synergies, in narrow-base and normal walking at time-points Pre, Post1, and Post2.

CoA

Discussion

Transfer of training effects

Adaptation to narrow-base walking

Limitations

Conclusions

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Fabio Augusto Barbieri

Roles

Author response to Decision Letter 0

Decision Letter 1

Jeremy P Loenneke

Roles

Author response to Decision Letter 1

Decision Letter 2

Jeremy P Loenneke

Roles

Acceptance letter

Jeremy P Loenneke

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases