Skip to main content
NCBI home page
BMC Geriatrics logoLink to BMC Geriatrics
. 2025 Dec 17;25:1029. doi: 10.1186/s12877-025-06749-0

Network-based transcranial direct current stimulation may improve gait and cognitive function in older adults: a randomized controlled crossover study

Yufeng Zhang 1,2,#, Wei Pan 3,#, Suwang Zheng 1,2, Xiaofan Feng 1,2, Bowen Wang 1,2, Yajie Wang 1,2, Junhong Zhou 4,✉, Jiaojiao Lü 1,2,✉, Yu Liu 1,2
PMCID: PMC12709697  PMID: 41408177

Abstract

Objective

Gait abnormalities and cognition decline are significant challenges in aging, impacting the independence of older adults. Recent studies suggest that the dorsal attention network (DAN) and default network (DN) in the brain play crucial roles in modulating gait patterns and cognitive function. This study aimed to investigate whether concurrent transcranial direct current stimulation (tDCS) targeting DAN and DN could synergistically improve gait parameters and cognition among healthy older adults.

Methods

In this cross-over double-blind study, 28 healthy elderly aged 68.500 ± 4.647 years were randomized to receive tDCS or Sham stimulation, with electrode placement and current intensity optimised using the Stimweaver® technique. Immediately before and after stimulation, participants completed gait assessment under two conditions: walking at self-selected speed (single-task walking) and walking while counting backward (i.e., dual-task walking), as well as N-back (0-back and 1-back) tasks. Outcome measures included stride length, stride time, step width, gait speed in both single and dual-task walking, as well as dual-task cost to gait. Cognitive function outcomes encompassed accuracy, reaction time, and inverse efficiency scores (IES) in the 0-back and 1-back tasks.

Results

Compared to Sham stimulation, tDCS led to significant improvements in gait speed under both single-task and dual-task conditions, with average increases of 0.038 m/s and 0.053 m/s, respectively. Additionally, 1-back reaction time decreased by 57 ms and IES by 0.131 following tDCS, with no significant changes under Sham stimulation. However, no change in dual-task cost was observed. Moreover, improvements in working memory were associated with changes in gait speed in single-task walking. Blinding efficacy was excellent, and participants’ subjective belief in the type of stimulation received did not influence the observed functional benefits of tDCS.

Conclusion

In healthy older adults, a single session of tDCS designed to modulate DAN and DN excitability concurrently improved gait speed in both single and dual-task walking, as well as working memory. These preliminary findings suggest that gait and working memory may be modifiable through neuromodulation approaches involving DAN and DN. Further studies are warranted to explore the relationship between gait, working memory, and these two brain networks using neuroimaging means.

Trial registration

ChiCTR2300077186-11/01/2023.

Keywords: Older adults, Gait, Dorsal attention network, Default network, Working memory

Introduction

Walking is not a simple automated process, necessitating active cognitive involvement (e.g., executive function, attention, and working memory) that is regulated by a host of cortical networks within the brain [1–4]. Additionally, walking often involves concurrent activities such as talking, reading, or making decisions [5], a phenomenon known as “dual-tasking”. This simultaneous engagement of cognitive resources in multiple tasks can alter walking performance [6, 7]. Aging and age-related conditions may exacerbate this deterioration in gait performance, increasing the risk of falls [8, 9]. Therefore, it is important to identify and intervene the cortical elements pertaining to cognitive control of gait to improve gait performance.

Recent neuroimaging studies have suggested that the dorsal attention network (DAN) is associated with gait performance [10, 11]. Extensive research consistently characterizes DAN as facilitating top-down attentional allocation and commonly engaging in tasks requiring attention and mental control [12–14]. Increased excitability within the DAN correlates with enhanced gait performance [15–17]. Moreover, the interplay between cortical networks is pivotal for dynamically reconfiguring connectivity patterns in response to varied environmental demands [18, 19]. Unlike those task-positive networks (e.g., DAN), the default network (DN) consistently experiences suppression during tasks to optimize task completion [20, 21]. Notably, studies have demonstrated a negative association between DN activity and gait performance in older adults aged 65 and above [22]. As a cognitive function associated with gait [4], better working memory performance was significantly associated with greater load-dependent deactivation of DN [23]. Therefore, strategies aimed at simultaneously augmenting DAN activity while attenuating DN activity hold considerable promise for enhancing both gait and working memory.

Transcranial direct current stimulation (tDCS) is recognized as an effective modality for modulating the excitability of the cerebral cortex [24]. This technique involves applying a weak constant current to specific brain regions, resulting in enhancement or suppression of cortical activity and subsequent improvements in motor and cognitive performance [25, 26]. A recent pilot study from our research group implemented a novel high-definition transcranial direct current stimulation (HD-tDCS) protocol that simultaneously facilitated excitability in key nodes of the DAN and inhibited excitability in key nodes of the DN in healthy young adults [27]. This work provided preliminary evidence that this multifocal network-based tDCS protocol may induce changes of functionally cortical networks and reduce gait variability. Still, it is unclear if this type of tDCS intervention can help improve the diminished gait performance in the older adult population, and the relationship between the tDCS-induced changes in gait and that in cognitive performance remains unknown.

In this experimental study, we investigated the therapeutic potential of tDCS on change in gait and working memory in a sample of healthy community-dwelling older adults. Specifically, our primary aim was to examine whether a concurrent, single-session tDCS protocol targeting both DAN and DN brain networks could synergistically induce acute change in the outcomes of gait speed and working memory among healthy older adults. A secondary aim of the study was to explore whether tDCS-induced changes in gait speed (in both single-task and dual-task conditions) were related to changes in working memory - specifically, accuracy, reaction time, and inverse efficiency score (IES; calculated as reaction time divided by accuracy). Additionally, we also explored the potential influences of the subjective speculations of participants on the intervention type on their performance.

Methods

Participants

The healthy older adults in the community were recruited through online advertisements, and flyers. The inclusion criteria were as follows: (1) Age 60 years or above; (2) Completion of primary school education or higher; (3) Capable of standing or ambulating without assistance; (4) Cognitive health assessed by the Montreal Cognitive Assessment - Beijing Version (MoCA-B) scores: >19 for primary school education, >22 for secondary school, and >24 for college education [28]; (5) Normal or corrected vision and hearing. Exclusion criteria were as follows: (1) Presence of severe mental disorders or other neurodegenerative diseases (e.g., Parkinson’s, stroke, vascular dementia); (2) Use of psychotropic medications (e.g., sedatives, hypnotics); (3) Any contraindications related to the use of tDCS, such as metal-implanted devices in the brain; (4) Self-reported lower-extremity pain or chronic issues significantly affecting gait; (5) Severe depression, anxiety, or sleep disorders. The study received approval from the Ethics Committee of Shanghai University of Sport (No: 102772023RT119) and was registered on the Chinese Clinical Trial Registry Platform (ChiCTR2300077186). All participants signed a statement of informed consent in the study.

Study design

This within-subject, randomized, double-blinded, sham-controlled study consisted of three study visits. On Visit 1, participants completed screening and familiarization protocol and signed written informed consent. The evaluations included MoCA-B [28], anxiety assessments (the validated Chinese version of Geriatric Anxiety Inventory-GAI) [29], depression evaluations (the validated Chinese version of the Geriatric Depression Scale - GDS) [30], as well as the recording of demographic features (e.g. age, height, weight). The familiarization protocol included N-back tasks and two gait conditions: walking at a self-selected speed (i.e., single-task walking) and walking while counting backward (i.e., dual-task walking). On visits 2 and 3, participants underwent gait and working memory assessments both before and immediately after a 20-minute session of either tDCS or Sham (i.e., control) stimulation. The sequence of tDCS conditions was randomized, and there was a minimum 7-day washout period. Following each experimental visit, participants completed a questionnaire of blinding efficacy and side effects [31]. The study was reported in accordance with the 2010 CONSORT guidelines: extension for randomized crossover trials.

Stimulation protocol

The Stimweaver® optimization technique was used to compute the optimal stimulation parameters, including electrode placement and current intensity [32, 33]. This algorithm enabled the modelling of an electric field tailored to concurrently enhance excitability in regions of the DAN while suppressing activity in the DN, based on predefined target coordinates derived from fMRI-based volumetric brain maps. The targeted electric field strength was constrained within physiologically safe limits, with efforts to minimize spread to non-target cortical areas. Based on the modeling results, seven electrodes were positioned at AF3, CP1, CP2, CP5, F7, FPZ, and FZ [27] (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Electrical current flow model (A) (B), electrical placements of the tDCS montage (C) and current intensities by locations of the electrodes (D). A The targets of tDCS by montages targeting the DAN (depicted in red) and DN (depicted in blue). B Electrical current flow model: The heat map reflects the strength and polarity of the electrical current flow modal. Red and blue representpositive and negative electrical currents. Darker and lighter colors mean stronger and weaker electrical currents, respectively. C The electrical placements of anodal electrodes (red circles: F7, CP1, and CP2) and cathodal electrodes (blue circles: FPz, AF3, Fz, and CP5). D The injected current intensities by each electrode

The tDCS was administered with the participant seated and at rest for 20 min, incorporating a 30-second fade-in and a 30-second fade-out. In the Sham stimulation, the same montage was applied, but stimulation was delivered for only 1 min (with a 30-second fade-in and a 30-second fade-out). Participants were instructed to notify study personnel if they experienced discomfort during stimulation. After the experimental visit, all participants completed a questionnaire assessing the efficacy of blinding (i.e., guessing the type of stimulation received - real, sham, or uncertain) and reported any side effects experienced during the stimulation period. Side effects, such as pain, itching, burning, and skin redness, were evaluated on a severity scale ranging from none (0) to mild (1), moderate (2), or severe (3). Furthermore, an open-ended section in the questionnaire was included to inquire about any additional discomfort experienced by participants [31].

Gait assessment

A 16-foot GAITRite pressure mat (ProtoKinetics Zeno Walkway, ZenoMetrics, LLC, Peekskill, NY, United States, 120 Hz sampling frequency) was used to measure gait. Each participant completed two trials of gait assessment under two conditions: single-task walking and dual-task walking. During each of the four assessment periods (i.e., pre- and post-tDCS and Sham stimulation), the order of these trials was randomized with a 60-second intermission in between. Participants initiated each trial by standing 1 m away from the gait mat, walking straightforwardly to traverse the mat, performing a 180-degree turn off the mat, walking away, and then returning to the starting position. This ‘lap’ was completed three times per trial. In the dual-task condition, participants were instructed to walk at their self-selected speed while verbally subtracting 3 from a randomly provided three-digit number just before the trial commenced. No explicit instructions on task priority were given in the dual-task walking. Gait characteristics for each trial were recorded by the pressure mat.

Working memory test

The performance of gait is associated with several aspects of executive function [4]. Here, we used the Digital N-back task to evaluate participants’ working memory. The assessment was conducted using the Psychtoolbox in MATLAB (MathWorks, Natick, MA, United States). Participants were assigned to perform two N-back tasks: 0-back and 1-back. In the 0-back task, participants identified whether the stimulus matched a randomly presented number (0–9) at the beginning. For the 1-back task, participants determined if the stimulus matched the one presented immediately before. Each block comprised 30 trials both in 0-back task and 1-back task. Participants completed two blocks for each task. Target trials occurred in 1/3 of the blocks. The instruction for participants was to complete N-back tasks with both speed and accuracy. For the N-back task, three outcome measures were analyzed: accuracy (percentage of correct responses), reaction time (mean response time of correct responses only), and IES.

Study outcomes

Primary outcomes for gait were gait speed in both single-task and dual-task walking conditions and those for working memory were the accuracy, reaction time and IES of 1-back.

Secondary outcomes for gait comprised stride length, stride time, and step width in both single- and dual-task walking conditions, as well as the dual-task cost (i.e., percent decrement in performance between normal walking and dual-task walking conditions). The dual-task cost was calculated for four parameters: gait speed, stride length, stride width, and stride time, using the following formula:

graphic file with name d33e502.gif

Secondary outcomes for working memory included the accuracy, reaction time and IES of 0-back.

Blinding and randomization

Participants were randomly assigned to one of two sequences (A and B) using the SPSS (version 29.0; IBM, Armonk, NY, USA). In Sequence A, participants first received the tDCS followed by the Sham stimulation, while in Sequence B, the order was reversed. The allocation sequence was concealed from both participants and investigators using sequentially numbered, opaque, sealed envelopes. Additionally, the tDCS and Sham stimulation were identical in appearance and administered in the same manner to maintain blinding. Blinding was assessed at the end of the study to ensure its effectiveness.

Statistical analysis

An a priori sample size calculation was performed using G*Power 3.1.9.7 software for an analysis of covariance (ANCOVA) model estimating fixed effects, main effects, and interactions. The desired statistical power was set at 0.80, with a significance level of 0.05 and an effect size of 0.40. This calculation indicated that a minimum total sample size of 26 participants was required. Considering a potential dropout rate of 15%, a total of 30 participants were included in the study.

Continuous variables are represented by the mean ± standard deviation. The normality of primary and secondary outcomes was assessed using the Shapiro-Wilk normality test. For continuous outcome variables that violated normality assumptions, including stride length and stride time under dual-task conditions, as well as all working memory-related measures (accuracy, reaction time and IES), natural logarithmic transformation was applied to enhance the normality of their distribution.

The ANCOVA was used for data analysis, to examine the effects of tDCS on primary and secondary outcomes, adjusting for its own MoCA scores, age and pre-stimulation [27]. The dependent variable of each model was the post-stimulation outcomes of gait (e.g., gait speed) or N-back tasks (e.g., accuracy) and the independent variable was stimulation (i.e., tDCS, Sham stimulation). To explore the relationship between changes in cognitive performance and gait speed, simple linear regression analyses were conducted using changes in 1-back performance (accuracy, reaction time, and IES) as predictors, and changes in gait speed under single-task and dual-task conditions as outcomes. Given the exploratory nature of the analysis, covariates were not included.

The Mann-Whitney U test was employed to investigate tDCS-induced side effects between two stimulations. Blinding efficacy and its impact on functional outcomes were evaluated in two steps. Initially, the Fisher’s Exact Test was employed to examine discrepancies in participant guesses regarding tDCS conditions (i.e., real, sham, or uncertain) across the different tDCS conditions. Subsequently, the ANOVA model was utilized to explore the potential influences of one’s belief in the type of stimulation received on the effects of those stimulation conditions that led to improved gait and N-back tasks. The model factor was the guess of stimulation type (i.e., real, sham, or uncertain), and the dependent variables were the improved outcomes from pre- to post-stimulation. Statistical analysis was performed using SPSS 29.0, with significance defined as p values below 0.05. Partial eta square (η2p) values were reported as an index of effect size.

Result

A total of 93 participants were initially screened for eligibility in the study, of whom 30 met the predefined inclusion criteria. Subsequently, two participants voluntarily withdrew from the study due to scheduling conflicts, leaving 28 participants who successfully completed all study tests (Fig. 2). Table 1 shows the demographics and the cognitive, mood status of the participants.

Fig. 2.

Fig. 2

Open in a new tab

The consolidated standards of reporting trials (CONSORT) diagram

Table 1.

The baseline characteristics of participants

Age (years) Sex
[n(%) = females]		 Height (m) Weight (kg) BMI (kg/m2) Education (years) MoCA score GDS score GAI score
68.500 ± 4.647 13 (46.429%) 1.632 ± 0.059 65.911 ± 7.635 24.724 ± 2.408 10.643 ± 2.778 25.714 ± 1.697 7.654 ± 4.783 2.615 ± 3.940
Open in a new tab

BMI Body Mass Index, GAI the validated Chinese version of Geriatric Anxiety Inventory, GDS the validated Chinese version of the Geriatric Depression Scale

The effects of tDCS on gait characteristics

The primary analyses demonstrated a significant stimulation effect for gait speed of single-task walking (F = 4.068, p = 0.049, η2p = 0.073) and dual-task walking (F = 5.117, p = 0.028, η2p = 0.091) (Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

The effects of tDCS and Sham stimulation on gait speed in single-task walking (A) and in dual-task walking. Following tDCS administration, participants exhibited significant improvements of gait speed in both single-task (A) and dual-task (B) walking compared to received Sham stimulation. Error bars represent standard deviations, # indicates the significant differences between two stimulation conditions

The secondary analyses showed a trend toward significant stimulation effect for stride length of single-task walking (F = 3.422, p = 0.070, η2p = 0.063) and dual-task walking (F = 2.914, p = 0.094, η2p = 0.054). No stimulation effect was evident in stride width and stride time (all F < 2.654, p > 0.109, η2p < 0.049). In addition, no significant difference was found in dual-task cost (all F < 2.064, p > 0.157, η2p < 0.039). The detailed values are shown in Table 2.

Table 2.

Effects of tDCS and Sham stimulation on mean values-based gait parameters and working memory

Conditions Outcomes tDCS Sham Stimulation effect
Pre Post Pre Post F p η p 2
Single-task walking gait speed (cm/sec) 117.582 ± 16.944 121.422 ± 17.683 118.857 ± 15.276 118.627 ± 16.674 4.068 0.049* 0.073
stride length (cm) 123.866 ± 11.252 126.218 ± 12.243 124.713 ± 10.966 124.895 ± 12.307 3.422 0.070 0.063
stride width (cm) 7.130 ± 2.408 7.553 ± 2.405 7.207 ± 2.934 7.421 ± 2.638 1.027 0.316 0.020
stride time (sec) 1.075 ± 0.086 1.060 ± 0.091 1.059 ± 0.075 1.061 ± 0.072 2.654 0.109 0.049
Dual-task walking gait speed (cm/sec) 104.523 ± 17.179 109.778 ± 16.308 104.524 ± 17.156 106.166 ± 17.571 5.117 0.028* 0.091
stride length (cm) 117.660 ± 14.217 120.457 ± 15.197 118.398 ± 12.272 119.053 ± 13.323 2.914 0.094 0.054
stride width (cm) 7.764 ± 2.998 7.776 ± 3.181 7.654 ± 3.370 7.861 ± 2.779 0.444 0.508 0.009
stride time (sec) 1.159 ± 0.149 1.138 ± 0.130 1.159 ± 0.149 1.134 ± 0.124 0.088 0.768 0.002
Dual-task cost gait speed (%) −10.549 ± 13.081 −8.665 ± 13.746 −11.796 ± 10.922 −10.423 ± 8.757 0.235 0.630 0.005
stride length (%) −4.911 ± 8.966 −4.391 ± 9.845 −4.773 ± 9.402 −4.448 ± 8.764 0.040 0.842 0.001
stride width (%) 7.571 ± 18.309 0.522 ± 25.258 6.563 ± 30.448 7.778 ± 23.546 1.780 0.188 0.034
stride time (%) 7.577 ± 7.332 7.262 ± 5.809 9.379 ± 10.796 6.752 ± 7.722 2.064 0.157 0.039
0-back accuracy 0.900 ± 0.040 0.913 ± 0.028 0.932 ± 0.028 0.937 ± 0.030 0.904 0.346 0.017
reaction time (sec) 0.586 ± 0.096 0.558 ± 0.072 0.573 ± 0.075 0.559 ± 0.062 1.123 0.294 0.022
IES 0.655 ± 0.118 0.612 ± 0.082 0.616 ± 0.083 0.599 ± 0.070 0.696 0.408 0.013
1-back accuracy 0.847 ± 0.074 0.901 ± 0.060 0.880 ± 0.069 0.886 ± 0.054 2.703 0.106 0.050
reaction time (sec) 0.670 ± 0.137 0.613 ± 0.101 0.636 ± 0.103 0.628 ± 0.146 4.974 0.030* 0.089
IES 0.809 ± 0.218 0.678 ± 0.141 0.732 ± 0.152 0.717 ± 0.199 9.540 0.003* 0.158
Open in a new tab

* significant difference, IES inverse efficiency score, all models adjusted for baseline value, MoCA, age

Effect of tDCS on working memory

The primary analyses showed that there was a significant stimulation effect for 1-back in reaction time (F = 4.974, p = 0.030, η2p = 0.089) and IES (F = 9.540, p = 0.003, η2p = 0.158) (Fig. 4).

Fig. 4.

Fig. 4

Open in a new tab

The effects of tDCS and Sham stimulation on working memory. Significant reductions in reaction time (A) and inverse efficiency score (B) in 1-back after receiving tDCS. Error bars represent standard deviations. # indicates the significant differences between two stimulation conditions

The secondary analyses indicated that no significant effect was observed in 0-back (all F < 1.123, p > 0.294, η2p < 0.022). The detailed values are shown in Table 2.

Linear regression analyses examined the associations between changes in 1-back performance and gait speed from pre- to post-stimulation. During single-task walking, a significant negative association was observed between the change in 1-back reaction time and the change in gait speed (β = −0.541, p = 0.003). Additionally, changes in IES were also significantly associated with gait speed (β = −0.430, p = 0.022). These findings suggest that improvements in cognitive performance - reflected by faster reaction times and greater processing efficiency - were linked to greater improvements in gait speed under single-task conditions. No significant associations were found in the dual-task condition (Table 3).

Table 3.

Linear regression analysis of the association between change in 1-back and change in gait speed

Δ Gait speed in single-task walking Δ Gait speed in dual-task walking
R 2 Standardized β coefficients
(95% confidence interval)		 p R 2 Standardized β coefficients
(95% confidence interval)		 p
Δ accuracy 0.057

0.238

(−14.719, 60.371)

 0.223 0.011

0.107

(−20.231, 34.943)

 0.588
Δ reaction time 0.293

−0.541

(−101.828, −23.420)

 0.003* 0.048

−0.220

(−50.931, 14.363)

 0.260
Δ IES 0.185

−0.430

(−41.488, −3.480)

 0.022* 0.079

−0.281

(−25.034, 3.974)

 0.148
Open in a new tab

* significant difference; IES, inverse efficiency score

Blinding efficacy and side effects

For side effects, none of the participants reported serious events associated with the intervention. Only mild-to-moderate side effects were reported (Table 4). The Mann-Whitney U test revealed that significant difference in tingling (p = 0.013) and redness (p = 0.045). In addition, the results showed no other types of discomfort were reported by the participants.

Table 4.

Effects of tDCS and Sham stimulation on side effects

tDCS Sham p
none mild moderate severe none mild moderate severe
tingling 4 21 3 0 11 17 0 0 0.013*
itching 22 6 0 0 25 3 0 0 0.279
burning 20 7 1 0 24 4 0 0 0.182
pain 20 8 0 0 24 4 0 0 0.197
redness 22 6 0 0 27 1 0 0 0.045*
fatigue 28 0 0 0 26 2 0 0 0.154
phosphene 28 0 0 0 28 0 0 0 1
Open in a new tab

* significant difference

For blinding efficacy, the Fisher’s Exact Test showed that the blinding was successful (p = 0.114), that is, 11 participants after tDCS and 14 participants after Sham stimulation guessed the stimulation condition correctly. Additionally, the ANOVA model revealed no effect of subjective guess on the improvements in the gait speed both in normal and dual-task walking within tDCS condition, as well as the 1-back tasks (all F < 3.237, p > 0.084, η2p < 0.111).

Discussion

This double-blinded and sham-controlled study provided that a single 20-minute session of tDCS designed to concurrently facilitate the excitability of the DAN and suppress the excitability of the DN appears to alter the regulation of gait parameters and cognitive function. Specifically, older adults who underwent this network-based tDCS demonstrated improvements in gait speed during both single- and dual-task walking, with a trend towards enhanced stride length. Additionally, their working memory performance, specifically in the 1-back task, showed improvement after receiving this targeted tDCS. Furthermore, faster reaction time in cognitive tasks was associated with higher gait speed during single-task walking. Importantly, no severe side effects were noted, with only mild-to-moderate side effects reported. These results suggest a potential role of one or both of these spatially-distinct brain networks in regulating both gait parameters and cognitive function.

Gait and working memory

In our study, we found that gait speed was improved both in single- and dual-task walking, alongside a discernible trend towards improved stride length across similar conditions. In addition to statistical significance, the observed changes in gait speed may have clinical relevance. A small but meaningful improvement is typically defined as 0.05 m/s [34]. In our study, gait speed increased by 0.038 m/s during single-task and 0.053 m/s during dual-task walking, with the latter exceeding the clinical threshold - suggesting potential real-world benefits for older adults. Walking, being a continuous task, evidently necessitates sustained attention for effective filtering of stimuli and mitigation of distractors [35]. Impairments in attentional mechanisms manifest as a diminished ability to focus and sustain task performance, particularly evident in tasks demanding heightened motor coordination, such as dual-task walking [36]. Our tDCS intervention aimed to enhance attentional processes by upregulating key DAN nodes, as supported by the significant improvements observed in 1-back task performance. Additionally, our tDCS protocol strategically aimed to suppress activity within the DN, given its propensity to interfere with task execution, particularly in contexts necessitating external attentional focus [13, 21]. By effectively modulating the DN, we sought to optimize cognitive resources for the successful completion of walking tasks.

Moreover, the typical inverse correlation between the DAN and DN, believed to support sustained attention [37], suggested that tDCS may regulate this relationship and subsequently improve gait performance. Studies have documented a diminished anticorrelation between these networks in the context of abnormal gait, indicating potential disruptions in attentional processes associated with gait dysfunction [10]. Therefore, the observation that gait was altered following tDCS targeting the DAN and DN provides unique causal evidence implicating a likely role of one or both of these networks in the regulation of gait in older adults. However, neurophysiological assessments were lacking to assess the effects of the tested tDCS montage on cortical function. Without such an assessment, our study could not conclusively determine whether tDCS targeting DAN and DN effectively altered their cortical excitability and functional connectivity. Future investigations should consider incorporating concurrent functional Magnetic Resonance Imaging (fMRI) during tDCS sessions to elucidate the functional connectivity between brain networks.

Furthermore, our findings reveal a significant correlation between gait performance and working memory. Impaired working memory capacity not only diminishes gait speed but also compromises the execution of complex movements and impairs the ability to respond effectively to environmental changes [38, 39]. Prior studies have demonstrated that a six-week regimen of working memory training can notably enhance the single-task gait speed among older adults, which aligns with the outcomes of our investigation [40]. Nonetheless, we did not identify a discernible relationship between working memory and performance during dual-task walking. We postulate that this inconsistency may stem from the heightened demand for additional cognitive resources inherent in dual-task locomotion [36]. Among these, the ventral attention network, as another component of the attention network, may play a role in detecting unattended or unexpected stimuli and triggering shifts of attention to the cognitive task in the dual-task walking condition [12, 22]. Further research is warranted to explore the intricacies of the relationship between brain networks and dual-task walking performance, including the exploration of tDCS interventions targeting the excitation of the ventral attention network.

Side effects

While our study did not uncover any severe adverse effects, it is noteworthy that participants reported experiencing heightened sensations of tingling and redness following tDCS administration compared to Sham stimulation. These commonly reported adverse reactions are consistent with the literature [24, 27, 41], suggesting their typicality in tDCS interventions. Notably, our adherence to established safety guidelines regarding current intensity and stimulation duration underscores the safety profile of our stimulation protocol [24]. Additionally, despite the increased incidence of side effects observed during tDCS sessions, it is imperative to highlight that participants’ subjective perceptions did not exert any discernible influence on the objective outcomes of the study. This resilience of study outcomes to participants’ reported experiences further bolsters the robustness and reliability of our findings.

Limitations and future direction

Several limitations should be noted. First, although the dual-task gait assessment utilized a commonly employed cognitive task (i.e., backward counting), it represented only one type of cognitive demand and may not have imposed sufficient cognitive load to elicit measurable interference. This may partly explain the absence of a significant dual-task cost. Future studies should incorporate a broader range of cognitive tasks with varying complexity to better characterize cognitive-motor interactions. Second, although the gait protocol was designed to capture steady-state walking, the exact number of steps per participant was not recorded, which may limit the precision of gait-related outcomes. Future research should quantify and report step count to enhance methodological rigour. Third, as this was an exploratory study, only the acute effects of a single tDCS session were examined. Multi-session and dose - response designs are needed to evaluate longer-term effects and optimize stimulation protocols. Finally, the network-based tDCS montage was not directly compared with traditional stimulation sites such as the dorsolateral prefrontal cortex (dlPFC). Comparative studies are warranted to determine the relative efficacy of network-targeted versus region-specific stimulation strategies in modulating cognitive and motor functions in older adults.

Conclusion

tDCS targeting DAN and DN shows promise as an intervention to improve gait and working memory in older adults. Despite the absence of imaging evidence, our results offer novel insights into the potential unique modulation of gait and cognitive function by these spatially distinct yet functionally linked cortical networks. Further investigations are warranted to elucidate the neural mechanisms underlying these effects and bolster the evidence base by employing larger sample sizes, and rigorous control conditions.

Acknowledgements

We would like to thank Dr. On-Yee Lo from the Hinda and Arthur Marcus Institute for Aging Research, Hebrew SeniorLife, Boston, MA, United States, and Harvard Medical School, Harvard University, Boston, MA, United States, for her invaluable assistance with the electrical stimulation protocol used in our experiments.

Abbreviations

ANCOVA

Analysis of covariance

BMI

Body mass index

DAN

Dorsal attention network

dlPFC

Dorsolateral prefrontal cortex

DN

Dorsal attention network

fMRI

Functional magnetic resonance imaging

GAI

the Validated Chinese version of geriatric anxiety inventory

GDS

the Validated Chinese version of the geriatric depression scale

MoCA-B

Montreal cognitive assessment - Beijing version

η2p

Partial eta square

tDCS

transcranial direct current stimulation

Authors’ contributions

YZ, WP, JZ, JL, and YL designed the study. YZ, WP, SZ, XF, BW, and YW collected the data. YZ, JZ, and JL analyzed the data. YZ drafted the manuscript. JZ, JL, and YL revised the manuscript. All authors read and approved the final version of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (12472328 and 12302418), the Sports Science & Technology Project of Shanghai (24J013) and the Student Innovation and Entrepreneurship Training Program (STYK20250402).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Shanghai University of Sport (No. 102772023RT119) and registered on the Chinese Clinical Trial Registry Platform (ChiCTR2300077186). All participants gave written informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Yufeng Zhang and Wei Pan contributed equally to this work and share first authorship. 

Contributor Information

Junhong Zhou, Email: JunhongZhou@hsl.harvard.edu.

Jiaojiao Lü, Email: ljj27@163.com.

References

  • 1.Hausdorff JM, Yogev G, Springer S, Simon ES, Giladi N. Walking is more like catching than tapping: gait in the elderly as a complex cognitive task. Exp Brain Res. 2005;164(4):541–8. [DOI] [PubMed] [Google Scholar]
  • 2.Yogev-Seligmann G, Hausdorff JM, Giladi N. The role of executive function and attention in gait. Mov Disord. 2008;23(3):329–42. quiz 472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Montero-Odasso M, Almeida QJ, Bherer L, Burhan AM, Camicioli R, Doyon J, et al. Consensus on shared measures of mobility and cognition: from the Canadian consortium on neurodegeneration in aging (CCNA). J Gerontol A Biol Sci Med Sci. 2019;74(6):897–909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Doi T, Shimada H, Makizako H, Tsutsumimoto K, Uemura K, Anan Y, et al. Cognitive function and gait speed under normal and dual-task walking among older adults with mild cognitive impairment. BMC Neurol. 2014;14:67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Hillel I, Gazit E, Nieuwboer A, Avanzino L, Rochester L, Cereatti A, et al. Is every-day walking in older adults more analogous to dual-task walking or to usual walking? Elucidating the gaps between gait performance in the lab and during 24/7 monitoring. Eur Rev Aging Phys Act. 2019;16:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Orcioli-Silva D, Islam A, Baker MR, Gobbi LTB, Rochester L, Pantall A. Bi-anodal transcranial direct current stimulation combined with treadmill walking decreases motor cortical activity in young and older adults. Front Aging Neurosci. 2021;13:739998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Al-Yahya E, Dawes H, Smith L, Dennis A, Howells K, Cockburn J. Cognitive motor interference while walking: a systematic review and meta-analysis. Neurosci Biobehav Rev. 2011;35(3):715–28. [DOI] [PubMed] [Google Scholar]
  • 8.Hassan SA, Bonetti LV, Kasawara KT, Beal DS, Rozenberg D, Reid WD. Decreased automaticity contributes to dual task decrements in older compared to younger adults. Eur J Appl Physiol. 2022;122(4):965–74. [DOI] [PubMed] [Google Scholar]
  • 9.Belur P, Hsiao D, Myers PS, Earhart GM, Rawson KS. Dual-task costs of texting while walking forward and backward are greater for older adults than younger adults. Hum Mov Sci. 2020;71:102619. [DOI] [PubMed] [Google Scholar]
  • 10.Lo OY, Halko MA, Zhou J, Harrison R, Lipsitz LA, Manor B. Gait speed and gait variability are associated with different functional brain networks. Front Aging Neurosci. 2017;9:390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Yu Q, Li Q, Fang W, Wang Y, Zhu Y, Wang J, et al. Disorganized resting-state functional connectivity between the dorsal attention network and intrinsic networks in Parkinson’s disease with freezing of gait. Eur J Neurosci. 2021;54(7):6633–45. [DOI] [PubMed] [Google Scholar]
  • 12.Vossel S, Geng JJ, Fink GR. Dorsal and ventral attention systems: distinct neural circuits but collaborative roles. Neuroscientist. 2014;20(2):150–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Miller EK, Buschman TJ. Cortical circuits for the control of attention. Curr Opin Neurobiol. 2013;23(2):216–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Corbetta M, Shulman GL. Control of goal-directed and stimulus-driven attention in the brain. Nat Rev Neurosci. 2002;3(3):201–15. [DOI] [PubMed] [Google Scholar]
  • 15.Potvin-Desrochers A, Martinez-Moreno A, Clouette J, Parent-L’Ecuyer F, Lajeunesse H, Paquette C. Upregulation of the parietal cortex improves freezing of gait in Parkinson’s disease. J Neurol Sci. 2023;452:120770. [DOI] [PubMed] [Google Scholar]
  • 16.Young DR, Parikh PJ, Layne CS. The posterior parietal cortex is involved in gait adaptation: a bilateral transcranial direct current stimulation study. Front Hum Neurosci. 2020;14:581026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Maidan I, Droby A, Jacob Y, Giladi N, Hausdorff JM, Mirelman A. The neural correlates of falls: alterations in large-scale resting-state networks in elderly fallers. Gait Posture. 2020;80:56–61. [DOI] [PubMed] [Google Scholar]
  • 18.van den Heuvel MP, Sporns O. Network hubs in the human brain. Trends Cogn Sci. 2013;17(12):683–96. [DOI] [PubMed] [Google Scholar]
  • 19.Bertolero MA, Yeo BTT, Bassett DS, D’Esposito M. A mechanistic model of connector hubs, modularity and cognition. Nat Hum Behav. 2018;2(10):765–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Christoff K, Irving ZC, Fox KC, Spreng RN, Andrews-Hanna JR. Mind-wandering as spontaneous thought: a dynamic framework. Nat Rev Neurosci. 2016;17(11):718–31. [DOI] [PubMed] [Google Scholar]
  • 21.Greicius MD, Krasnow B, Reiss AL, Menon V. Functional connectivity in the resting brain: a network analysis of the default mode hypothesis. Proc Natl Acad Sci U S A. 2003;100(1):253–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Droby A, Varangis E, Habeck C, Hausdorff JM, Stern Y, Mirelman A, et al. Effects of aging on cognitive and brain inter-network integration patterns underlying usual and dual-task gait performance. Front Aging Neurosci. 2022;14:956744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Liu S, Poh JH, Koh HL, Ng KK, Loke YM, Lim JKW, et al. Carrying the past to the future: distinct brain networks underlie individual differences in human spatial working memory capacity. Neuroimage. 2018;176:1–10. [DOI] [PubMed] [Google Scholar]
  • 24.Antal A, Alekseichuk I, Bikson M, Brockmöller J, Brunoni AR, Chen R, et al. Low intensity transcranial electric stimulation: safety, ethical, legal regulatory and application guidelines. Clin Neurophysiol. 2017;128(9):1774–809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Nitsche MA, Paulus W. Excitability changes induced in the human motor cortex by weak transcranial direct current stimulation. J Physiol. 2000;527(Pt 3):633–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Narmashiri A, Akbari F. The effects of transcranial direct current stimulation (tDCS) on the cognitive functions: a systematic review and meta-analysis. Neuropsychol Rev. 2023. 10.1016/j.arr.2022.101738. [DOI] [PubMed] [Google Scholar]
  • 27.Zhou R, Zhou J, Xiao Y, Bi J, Biagi MC, Ruffini G, et al. Network-based transcranial direct current stimulation may modulate gait variability in young healthy adults. Front Hum Neurosci. 2022;16:877241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Chen KL, Xu Y, Chu AQ, Ding D, Liang XN, Nasreddine ZS, et al. Validation of the Chinese version of Montreal cognitive assessment basic for screening mild cognitive impairment. J Am Geriatr Soc. 2016;64(12):e285-90. [DOI] [PubMed] [Google Scholar]
  • 29.Yan Y, Xin T, Wang D, Tang D. Application of the geriatric anxiety inventory-Chinese version (GAI-CV) to older people in Beijing communities. Int Psychogeriatr. 2014;26(3):517–23. [DOI] [PubMed] [Google Scholar]
  • 30.Liu J, Wang Y, Wang XH, Song RH, Yi XH. Reliability and validity of the Chinese version of geriatric depression scale among Chinese urban community-dwelling elderly population. Chin J Clin Psychol. 2013;21(1):39–41. [Google Scholar]
  • 31.Zhang Y, Zhou Z, Zhou J, Qian Z, Lü J, Li L, et al. Temporal interference stimulation targeting right frontoparietal areas enhances working memory in healthy individuals. Front Hum Neurosci. 2022;16:918470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Ruffini G, Fox MD, Ripolles O, Miranda PC, Pascual-Leone A. Optimization of multifocal transcranial current stimulation for weighted cortical pattern targeting from realistic modeling of electric fields. Neuroimage. 2014;89:216–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Fischer DB, Fried PJ, Ruffini G, Ripolles O, Salvador R, Banus J, et al. Multifocal tDCS targeting the resting state motor network increases cortical excitability beyond traditional tDCS targeting unilateral motor cortex. Neuroimage. 2017;157:34–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Perera S, Mody SH, Woodman RC, Studenski SA. Meaningful change and responsiveness in common physical performance measures in older adults. J Am Geriatr Soc. 2006;54(5):743–9. [DOI] [PubMed] [Google Scholar]
  • 35.Wang J, Liu J, Wang Z, Sun P, Li K, Liang P. Dysfunctional interactions between the default mode network and the dorsal attention network in subtypes of amnestic mild cognitive impairment. Aging. 2019;11(20):9147–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Killane I, Donoghue OA, Savva GM, Cronin H, Kenny RA, Reilly RB. Relative association of processing speed, short-term memory and sustained attention with task on gait speed: a study of community-dwelling people 50 years and older. J Gerontol A Biol Sci Med Sci. 2014;69(11):1407–14. [DOI] [PubMed] [Google Scholar]
  • 37.Greenwood PM, Blumberg EJ, Scheldrup MR. Hypothesis for cognitive effects of transcranial direct current stimulation: externally- and internally-directed cognition. Neurosci Biobehav Rev. 2018;86:226–38. [DOI] [PubMed] [Google Scholar]
  • 38.Montero-Odasso M, Bergman H, Phillips NA, Wong CH, Sourial N, Chertkow H. Dual-tasking and gait in people with mild cognitive impairment. The effect of working memory. BMC Geriatr. 2009;9:41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Montero-Odasso M, Verghese J, Beauchet O, Hausdorff JM. Gait and cognition: a complementary approach to understanding brain function and the risk of falling. J Am Geriatr Soc. 2012;60(11):2127–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Azadian E, Majlesi M, Jafarnezhadgero AA. The effect of working memory intervention on the gait patterns of the elderly. J Bodyw Mov Ther. 2018;22(4):881–7. [DOI] [PubMed] [Google Scholar]
  • 41.Brunoni AR, Amadera J, Berbel B, Volz MS, Rizzerio BG, Fregni F. A systematic review on reporting and assessment of adverse effects associated with transcranial direct current stimulation. Int J Neuropsychopharmacol. 2011;14(8):1133–45. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Data Availability Statement

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Articles from BMC Geriatrics are provided here courtesy of BMC

RESOURCES