https://orcid.org/0009-0009-7820-1527
Figures
Abstract
Background
Post-stroke cognitive impairment is a common and severe complication of stroke, significantly affecting patients’ daily life and rehabilitation. Although transcranial direct current stimulation combined with computerized cognitive training has shown potential for improvement, current studies still have key limitations. Most have used only simultaneous intervention protocols, without systematically comparing sequential combination strategies, making it difficult to determine the optimal clinical approach.
Objectives
The core objective of this study will be to address the existing research gap by comparing sequential intervention strategies and to resolve the key issue of identifying the optimal clinical approach.
Methods
This study will use a single-center, double-blind, randomized controlled trial design. A total of 60 patients with post-stroke cognitive impairment will be randomly assigned in a 1:1:1:1 ratio to four groups: three intervention groups receiving computerized cognitive training combined with transcranial direct current stimulation at different timings (pre-training, concurrent, or post-training), and one sham control group. All interventions will be administered five times a week for two consecutive weeks. Follow-up assessments will be conducted at 4 and 12 weeks post-intervention.
Discussion
This trial will systematically evaluate the optimal timing strategies for combining tDCS and CCT in PSCI. The hypothesized mechanisms are: (1) synchronous stimulation enhancing cognitive processing efficiency through immediate synaptic changes, (2) pre-training stimulation promoting plasticity via long-term potentiation-like modifications, and (3) post-training stimulation consolidating training effects by modulating neuronal synchrony. High-sensitivity cerebral oxygenation parameters will be used to assess microcirculatory improvements, linking physiological changes to cognitive recovery. However, ceiling effects and test biases will potentially limit the accuracy of long-term results, while sample size and lack of stratification based on brain lesion characteristics may introduce heterogeneity. Future studies will need to incorporate multimodal neuroimaging for stratified analysis and should develop personalized interventions based on lesion severity, disease stage, and timing to address therapeutic challenges and advance precision rehabilitation.
Trial registration: ChiCTR2500102565.[Chinese Clinical Trial Registry (ChiCTR), https://www.chictr.org.cn/] [Registered on May 16, 2025]
Citation: Wei X, Xia J, Luo R, Hu P, Wu Y, Liu J, et al. (2026) Combining transcranial direct current stimulation and computerized cognitive training for post-stroke cognitive recovery: Study protocol for a randomized controlled trial with optimized temporal intervention sequences. PLoS One 21(3): e0342761. https://doi.org/10.1371/journal.pone.0342761
Editor: Roya Khanmohammadi, Iran University of Medical Sciences, IRAN, ISLAMIC REPUBLIC OF
Received: September 26, 2025; Accepted: January 27, 2026; Published: March 4, 2026
Copyright: © 2026 Wei et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: No datasets were generated or analysed during the current study. All relevant data from this study will be made available upon study completion.
Funding: This work was supported by the Nanshan District Health Bureau of Shenzhen [Grant Number NSZD2023002] to MingHong Sui. The funder’s website is available at http://www.szns.gov.cn/wsjkw/. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: PSCI, Post-stroke cognitive impairment; tDCS, transcranial direct current stimulation; CCT, computerized cognitive training; fNIRS, functional near-infrared spectroscopy; MoCA, Montreal Cognitive Assessment; MMSE, Mini-Mental State Examination; BI, Barthel Index; HAMA, Hamilton Anxiety Rating Scale; HAMD, Hamilton Depression Rating Scale; DLPFC, dorsolateral prefrontal cortex; PFC, Prefrontal Cortex; HbO, oxygenated hemoglobin; VFT, Verbal Fluency Task
Introduction
Stroke, the second leading cause of death worldwide, significantly affects patients and their families due to its high incidence and severe long-term consequences [1].Survivors often experience lingering motor, sensory, cognitive, and bowel/bladder dysfunctions, which reduce their quality of life, impose a significant economic burden on society, and cause emotional strain on families [2,3].International studies show that the incidence of post-stroke cognitive impairment (PSCI) in the Chinese population can reach 78.7% within 3 months post-stroke [4].Cognitive impairments, including severe declines in memory, attention, and executive function, make it difficult for patients to perform daily tasks. The distress caused by these impairments can sometimes outweigh that of physical disabilities, complicating rehabilitation and hindering overall recovery. Given the multidimensional impact of cognitive dysfunction on rehabilitation, cognitive recovery should be prioritized as highly as physical restoration [5,6].
The James Lind Alliance, a UK-based partnership for setting research priorities, has identified PSCI as a top priority in stroke research [7].The 2021 joint guidelines from the European Stroke Organisation and European Academy of Neurology highlight the ongoing therapeutic limitations of cognitive training for PSCI. To overcome this challenge, the guidelines emphasize the need for prioritizing randomized controlled trials in future research [8].Computerized cognitive training (CCT) is a software-based intervention delivered through digital platforms, known for its engaging interface, portability, and real-time progress tracking. This approach has gained popularity in clinical practice and shown significant therapeutic efficacy in cognitive impairment rehabilitation [9].Its potential benefits may arise from the interaction between cognitive-psychological changes induced by CCT and dopamine neurotransmitter release [10].However, its use as a standalone treatment for PSCI patients shows limited clinical efficacy in certain populations [11].This limitation arises from CCT’s unimodal design (e.g., visual tasks), which restricts the activation of multimodal neural circuits and hinders neuroplasticity. Additionally, standalone CCT provides short-term cognitive benefits but struggles to generalize domain-specific improvements to overall functional gains [12].
To overcome these limitations, combining therapy with non-invasive brain stimulation has become a key research focus. Transcranial Direct Current Stimulation (tDCS) is especially prioritized due to its safety profile—it does not induce action potentials, which significantly reduces the risk of seizures and ensures treatment safety [13].Additionally, tDCS effectively modulates cortical excitability, addressing CCT’s limitations by regulating neuroplasticity in a targeted manner, thereby creating a ‘neuroactivation-cognitive enhancement’ feedback loop [14,15].tDCS also exhibits significant long-lasting aftereffects, with 20–30 minutes of stimulation maintaining neural network modulation for up to 90 minutes post-treatment, thereby providing a neurobiological basis for sustained cognitive enhancement [16].However, most devices lack real-time monitoring and adaptive parameter adjustments (e.g., stimulus intensity), which limits their regulatory efficacy beyond simple intervention-based stimulation.
Current evidence suggests that combined therapy typically enhances efficacy compared to monotherapy, though study conclusions remain inconsistent.Chen et al. [17]observed significantly improved vascular vasomotor function in the combined therapy group via transcranial Doppler ultrasonography, linked to optimized hippocampal activation patterns and a bidirectional beneficial cycle between vascular and cognitive functions. Notably, Kolskar et al. [18] reported no additional benefits of tDCS-CCT combinations over CCT alone in high-reliability fMRI-based brain activation analyses.Meanwhile, studies confirm significant differences in clinical outcomes between combined strategies.Naro et al. [19]found both concurrent tDCS-RAGT and sequential RAGT-then-tDCS interventions in stroke patients significantly improved balance (p < 0.001) and moderately enhanced gait endurance (p = 0.04), with the sequential approach showing particularly sustained effects—contradicting Leon’s null findings [20].Current investigations into tDCS-CCT combinations highlight potential cognitive benefits, yet their efficacy remains inconsistent [21].A meta-analysis confirmed significant improvements in working memory with CCT in stroke patients, but there were insufficient tDCS studies for analysis, and meta-regression identified no significant moderators.Given that tDCS and CCT exhibit complementary mechanisms of action and that controversy exists regarding the efficacy of combined tDCS at different time sequences., we will propose combining these interventions to address cognitive impairments in stroke patients [22].
Given the significant gap in research on combined tDCS-CCT intervention strategies, particularly the lack of systematic exploration into the differential effects between simultaneous and sequential intervention modes, this study will investigate the cognitive intervention effects of tDCS in stroke patients undergoing CCT. The primary objective will be to compare efficacy differences across various temporal combinations of tDCS-CCT therapies to identify the optimal synergistic protocol. Rooted in the theoretical framework linking tDCS-mediated cerebral hemodynamic modulation to cognitive enhancement, the study will specifically examine the association between prefrontal cortical oxygen metabolism and cognitive improvement. Functional Near-Infrared Spectroscopy (fNIRS) will be used to dynamically monitor pre- and post-treatment hemodynamic parameters in the prefrontal cortex, aiming to elucidate the neurophysiological mechanisms underlying cognitive recovery through the lens of neurovascular coupling.
Methods and design
The study protocol was approved by the Scientific Research Ethics Committee of Nanshan District People’s Hospital, Shenzhen (Approval Number: Scientific Ethics Review ky-2024 103101, Approval Date: March 5, 2025), and registered with the Chinese Clinical Trial Registry (ChiCTR) prior to the initiation of the study (Registration Number: ChiCTR2500102565, Registration Date: May 16, 2025)https://www.chictr.org.cn/. All study activities will be conducted in accordance with the requirements of the ethics committee, and the validity of the study protocol will cover the entire duration of the research (until March 5, 2026).
This study will be in the participant recruitment phase, which is expected to be completed between June 2025 and July 2026. At the time of submission, data collection has not yet commenced. As a study protocol, this manuscript will aim to describe the planned study prior to the generation of any results.
Study design
This is a single-center, double-blind, randomized controlled exploratory trial. Sixty patients will be randomly assigned to one of four groups (n = 15 per group). The protocol will be reported following Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) schedule statement(S1 File).Participant enrollment, randomization, assessment, and follow-up schedules are summarized in Fig 1.For a detailed visualization of the study workflow, Fig 2, which outlines the sequential steps from participant recruitment to final follow-up.
Trial schedule of enrollment, intervention, and assessments.MoCA,Montreal Cognitive Assessment; MMSE,Mini-Mental State Examination; HAMA,Hamilton Anxiety Scale; HAMD,Hamilton Depression Rating Scale;MBI,Modified Barthel Index; tDCS,Transcranial Direct Current Stimulation; CCT,Computerized Cognitive Training; fNIRS,Functional Near-Infrared Spectroscopy; VFT,Verbal Fluency Test.
Study setting and recruitment
Patient recruitment will be conducted at the Department of Rehabilitation Medicine, Shenzhen Nanshan People’s Hospital, from June 2025 to July 2026, using a dual approach to optimize eligibility assessment and engagement. Outpatient recruitment will utilize posters in the rehabilitation clinic, enabling interested patients to notify therapists for referral to the principal investigator for preliminary screening. Concurrently, resident physicians in the Neurology and Rehabilitation Medicine Departments will systematically screen inpatients against predefined criteria and refer eligible candidates to the research team for formal enrollment. All referring clinicians will undergo standardized protocol training to ensure consistency, and written informed consent will be obtained prior to participant inclusion. Eligibility criteria are detailed below.
Participants
Inclusion criteria.
- (a). Diagnosed with stroke (intracerebral hemorrhage or cerebral infarction) confirmed by head CT/MRI, per the 1995 Fourth National Cerebrovascular Disease Conference criteria;
- (b). Alert, hemodynamically stable, and capable of completing cognitive assessments;
- (c). First-time stroke within 3 months post-onset;
- (d). MoCA score between 10–26 (inclusive);
- (e). Voluntary informed consent.
Exclusion criteria:.
- (a). History of other neurological disorders (e.g., Parkinson’s disease, tumors, hydrocephalus, or trauma);
- (b). Cognitive impairment attributable to non-stroke conditions;
- (c). Significant auditory/visual deficits, aphasia, or uncooperativeness;
- (d). Pre-stroke cognitive decline.
Randomization
In this study, we’ll use blocked randomization and IBM SPSS Statistics 26.0 (International Business Machines Corporation, Armonk, NY, USA) for a 1:1:1:1 balanced allocation across four groups (pre-, concurrent-, post-training stimulation, and sham control).First, set a fixed random seed (based on the date for reproducibility). With N = 60 and balance needs, define block length as 12 (3 per group) and generate sequences for 5 blocks.Then, use rv.random(1,4) for random integers and sort randomly within blocks.Finally, match sequences with subject numbers (1–60) via match files and sort cases, assigning groups by block and order to control time trend interference.
Allocation concealment
To maintain allocation concealment, a pre-generated randomization sequence will be embedded in sequentially numbered, opaque, sealed envelopes. These envelopes will be securely stored and accessed exclusively by designated therapists responsible for stratified randomization of eligible patients into study arms.
Blinding
Throughout the trial, participants, outcome assessors, and data analysts will remain blinded to group assignments until final database lock.Unblinding will be permitted solely under emergency circumstances (e.g., serious adverse events). No interim unblinding procedures are required, as all data analyses will be conducted following trial completion.
Sample size estimation
Prior to study initiation, we will perform sample size estimation using G*Power 3.1.9.7 (Heinrich Heine-Universität Düsseldorf, Düsseldorf, Germany). The primary outcome will be the MoCA score, and the effect-size parameter will be set at Cohen’s f = 0.708, as reported in a peer-reviewed publication [17]. The statistical model will be a linear mixed model testing the group × time interaction. The following settings will be used: power = 0.95, α = 0.05, four groups, three repeated measurements (baseline, post-treatment, and follow-up), and a within-subject correlation of 0.30 (reflecting an expected increase in between-subject variability at follow-up). The software will indicate that a total of 48 participants will be required. Anticipating a 20% dropout rate, we will divide 48 by 0.8 and round up to the nearest multiple of four, resulting in a planned recruitment of 60 participants (15 per group).
Interventions
All participants will receive five sessions per week for two weeks, totaling 10 sessions. Adverse events will be monitored during each session, with researchers administering tDCS stimulation according to Brunoni et al.‘s protocol [23] to assess patient discomfort, pain, and any causality related to tDCS. Four treatment groups will be established:
- Group A (tDCS-before-CCT): Patients will receive 20 minutes of tDCS, followed by 30 minutes of CCT after a 4-hour interval.
- Group B (tDCS-after-CCT): Patients will complete 30 minutes of CCT first, then will receive 20 minutes of tDCS after a 4-hour interval.
- Group C (tDCS-with-CCT): tDCS will be administered concurrently during the last 20 minutes of the 30-minute CCT session.
- Group D (CCT-only/Sham): Patients will undergo CCT with simultaneous placebo tDCS (sham stimulation).
Prior to the treatment intervention, patients will undergo a baseline assessment to determine eligibility. Enrollment will be based on the Montreal Cognitive Assessment (MoCA) [24]. Assessors will be blinded to group allocation. Post-treatment assessments will be conducted immediately after treatment completion, with follow-up assessments scheduled at 4 and 12 weeks post-treatment.The use of any non-invasive brain stimulation protocols other than those explicitly specified in this study (e.g., transcranial magnetic stimulation [TMS]) is strictly prohibited throughout the research period.
tDCS.
The MBM-1 intelligent electrostimulator (Jiangxi Huaheng Jingxing Medical Technology Co., Ltd.) will be used, with 4.5 × 5.5 cm isotonic gelatin sponge electrodes soaked in saline (0.9% NaCl). Stimulation will target the dorsolateral prefrontal cortex (DLPFC) based on the EEG 10–20 system, with F3 as the anode and F4 as the cathode (Fig 3). Parameters: 1 mA current intensity (selected for safety and demonstrated behavioral effects) [25].Each session will last 20 minutes, with 5 sessions per week for 2 weeks.Ramp-Up/Ramp-Down: The current will gradually increase to 1 mA over the first 30 seconds and decrease to 0 mA during the final 30 seconds of each session. For sham stimulation, the device will activate for 30 seconds before automatically shutting down.
CCT.
The PASS-CATS 1.0 software will simulate cognitive tasks targeting learning/memory and executive inhibition.Learning/Memory Module: Object relocation, pattern recognition, novel character identification, numerical matching, mental arithmetic recall, and facial memory tasks.Executive Inhibition Module: Error detection, selective attention, symbol-number matching, linguistic inference, pattern reasoning, and numerical sequence tasks.Training sessions will last 30 minutes, administered 5 times per week for 2 weeks. The software will automatically record performance data, which will be scored by a blinded therapist.
Outcome measures
Primary outcomes.
- MoCA: will be administered at baseline, 4 weeks, and 12 weeks, with a focus on detecting mild cognitive improvements. The cutoff scores for cognitive impairment will be determined based on educational attainment (e.g., ≤ 13 points for primary education, ≤ 19 points for secondary education, and ≤24 points for tertiary education, as per the MoCA-Chinese version criteria).
- Prefrontal Cortex (PFC) Activation (fNIRS): Task-related oxygenated hemoglobin (HbO) signals will be measured using the NirSmartⅡ-3000D system (Huichuang Medical, Danyang), with wavelengths of 730 nm/850 nm and a sampling rate of 11 Hz. The channel configuration will consist of 22 channels using 8 emitters and 8 detectors (3 cm separation), aligned with the EEG 10–20 system (Fig 4).The protocol includes 8 minutes of resting-state data (eyes closed, seated comfortably in a quiet room). The Verbal Fluency Task (VFT) will involve 30 seconds of pre-task rest, 60 seconds of task phase (generate words starting with |d|, |g|, or |h|; 20 seconds per phoneme), and 75 seconds of post-task rest. The outcomes will include PFC activation intensity, resting-state functional connectivity (left/right PFC), and word-generation accuracy.
Secondary outcomes.
- Mini-Mental State Examination (MMSE): evaluates cognitive function (scored 0–30) and will be administered at baseline, 4 weeks, and 12 weeks. A score of ≥23 indicates normal cognition.
- Barthel Index (BI): evaluates activities of daily living (ADL) with dependency levels classified as follows: independent (100 points), mild dependency (61–99 points), moderate dependency (41–60 points), and severe dependency (≤40 points).
- Hamilton Anxiety Scale (HAMA):Scores ≥25 = severe anxiety; 21–24 = moderate anxiety; 14–20 = mild anxiety; 7–13 = borderline anxiety; < 7 = no anxiety.
- Hamilton Depression Scale (HAMD): Total scores: 0–54 (higher = worse). ≥ 25 = severe depression; 18–24 = moderate depression; 7–17 = mild depression.
- Follow-Up: MoCA and MMSE will be repeated at 4 and 12 weeks post-treatment to assess sustained cognitive improvements.
Safety and adverse event monitoring
Participant safety will be monitored continuously throughout the 1-week intervention period and the 2-month follow-up phase. All adverse events (AEs) and serious adverse events (SAEs) – whether reported by participants or observed by study investigators – will be systematically documented in accordance with established clinical research protocols.To monitor and manage adverse effects, patients will report sensations of itching, tingling, burning, headache, or other discomforts using a 4-point scale (1 = absent, 2 = mild, 3 = moderate, 4 = severe). The causal relationship between symptoms and tDCS stimulation will be assessed on a 5-point scale (1 = unrelated, 5 = strongly related).In the event of adverse events (AEs) or serious adverse events (SAEs), investigators shall provide immediate appropriate medical intervention and conduct follow-up assessments until clinical resolution or stabilization is confirmed. SAE data will be formally reported during regular meetings of the Data Safety Monitoring Board (DSMB) in compliance with institutional safety protocols.
Data management
Standardized eCRFs will capture baseline demographics, clinical parameters (stroke subtype, lesion site, cognitive status), labs, and treatment history from EMRs post-enrollment. AEs/SAEs will be tracked throughout treatment, with 3-month cognitive follow-up (MoCA, MMSE).Primary source data for this study will be maintained through secure archiving of paper CRFs containing participant information in locked cabinets within controlled-access facilities, while two trained data clerks under independent QA (JL and YYW) supervision will enter all CRF data into the EDC system, which automatically generates immutable audit trails that will document user identity and timestamps for all entries, modifications, and deletions.
Optimizing retention and follow-up adherence
Participants will receive comprehensive information regarding trial procedures and expectations during the study period, with particular emphasis on the critical importance of follow-up compliance. A structured 3-month follow-up protocol will be provided to all patients,who will be strongly encouraged to attend scheduled clinic visits for face-to-face assessments. Two days prior to each follow-up appointment, designated research staff will proactively contact participants via telephone, WeChat, or alternative preferred communication channels to confirm attendance. For individuals unable to complete in-person evaluations, remote cognitive assessments will be facilitated through validated software platforms [https://doctor.shenzhenheling.com/],enabling flexible self-administration at the participant’s convenience. All remotely collected data will undergo immediate verification by research personnel, with subsequent outreach to be conducted as necessary to resolve missing or inconsistent responses. Participants retain the right to withdraw from follow-up procedures at any time without consequence to their standard medical care.
Statistical methods
Clinical data analysis.
Statistical analyses will be conducted using SPSS 26.0, with a two-tailed significance level of p < 0.05. Categorical variables will be presented as frequencies and percentages, and compared using chi-square tests. Continuous variables will be assessed for normality using the Shapiro-Wilk test; normally distributed data will be reported as mean±SD, and non-normally distributed data as median [IQR].For repeated measures (e.g.,cognitive assessments at multiple timepoints), linear mixed models (LMM) will be used to include covariates (e.g., age, baseline scores, education level) and manage missing data. Between-group comparisons at individual timepoints will be conducted using one-way ANOVA for normally distributed data or Kruskal-Wallis tests for non-normal data, with multiple comparisons adjusted using the false discovery rate (FDR) method. Correlations will be analyzed using Pearson’s correlation for normally distributed data and Spearman’s correlation for non-normal data.The primary outcome analysis will be conducted on an intention-to-treat (ITT) basis [26]. Missing data will be addressed using complete case analysis if the missing rate is less than 5%, or multiple imputation if the missing rate is 5% or higher [27].
FNIRS data analysis.
The fNIRS data will be systematically analyzed using the NirSpark software package (v1.7.5, Danyang Huichuang Medical Equipment Co., Ltd., China). The preprocessing pipeline will include motion artifact correction based on moving standard deviation (Moving SD) detection combined with cubic spline interpolation to eliminate signal jumps caused by head movements. Subsequently, a bandpass filter with a cutoff frequency of 0.02–0.20 Hz will be applied to remove physiological noise such as heartbeats and respiration. The preprocessed optical density signals will be then converted into relative changes in HbO and HbR)concentrations using the modified Beer-Lambert law. HbO will be selected as the primary analytical indicator due to its higher signal-to-noise ratio. For the VFT phase, the average HbO concentration will be calculated over a 60-second time window following task onset, with baseline drift will be corrected using linear fitting. To extract group-level hemodynamic features, the individual HbO signals across all 19 channels from both experimental groups will be averaged at the group level. Finally, HbO concentration topographic maps and functional connectivity maps of the right prefrontal cortex (channels 1–11) and left prefrontal cortex (channels 12–19) will be generated using the NirSpark software to reflect inter-regional brain interactions.
Discussion
PSCI remains a significant and unmet challenge in rehabilitation, greatly affecting functional recovery and quality of life. This randomized controlled trial will rigorously evaluate the clinical efficacy of sequential tDCS-CCT interventions for PSCI. Based on previous literature, the sequence of combined interventions may lead to different therapeutic outcomes through distinct neuroplastic mechanisms, as the effects of tDCS are time-dependent, and the impact of stimulation timing exhibits complexity and bidirectionality [28–30]. When the training task is synchronized with the tDCS-induced state of neural excitation, tDCS can optimize neuronal firing synchrony, thereby enhancing synaptic efficiency and cognitive processing through activity-dependent synaptic plasticity. If tDCS is administered prior to CCT, it may initiate neuroplastic networks through a synapse remodeling process similar to LTP; however, the potential interference from homeostatic plasticity mechanisms must be considered. According to the Bienenstock–Cooper–Munro (BCM) theory, the enhancement of postsynaptic activity induced by tDCS increases the modification threshold (MT), thereby reducing the likelihood of LTP and increasing the probability of long-term depression (LTD), which may ultimately impair learning performance [31].In contrast, administering tDCS after training may stabilize newly formed neural connections through activity-dependent synchronization. We hypothesize that the synchronous strategy, by maximizing the immediate induction of LTP, yields the most optimal intervention outcome.
The brain’s high metabolic demand—accounting for 20% of total oxygen consumption despite representing only 2% of body weight—necessitates close monitoring of cerebral hemodynamics [32,33]. Chronic hypoperfusion triggers pathological cascades, including microvascular dysfunction, neuronal injury, and cortical atrophy, ultimately disrupting the functional networks essential for cognition [34]. Functional neuroimaging confirms task-induced neurovascular coupling, wherein activated neurons drive local arteriolar dilation [35,36]. Chen et al.’s tDCS-CCT study demonstrated enhanced cerebrovascular reactivity and regional blood flow, highlighting the utility of HbO as a sensitive biomarker for microvascular changes that predict cognitive recovery.
Methodological limitations must be considered: neuropsychological testing is susceptible to ceiling and practice effects in longitudinal designs. High-functioning patients may experience premature plateaus (“ceiling effect”), while low-functioning individuals may show artificial improvements due to repeated exposure (“practice effect”), complicating the assessment of true intervention effects.Notably, inter-individual variance in lesion site and severity will modulate treatment response; we will therefore integrate baseline imaging—lesion location and volume—for subgroup analyses to craft interventions that concurrently respect lesion signature, disease stage and temporal order. As evidence on the tDCS–CCT pairing will remain sparse and prior cohorts will have been small, heterogeneity will persist; thus, the forthcoming findings will generalize only to the single-center Chinese cohort to be recruited here, and subsequent larger, geographically dispersed trials will be required to validate this sequential strategy.
Composition of the data monitoring committee, its role, and reporting structure
In accordance with recommendations from the Data Safety Monitoring Board (DSMB) of Shenzhen Nanshan People’s Hospital, a dedicated DSMB has not been appointed for this study. As all intervention protocols in this trial constitute standard clinical practices within our department, DSMB interim assessments would not confer additional safety benefits. In the event of a serious adverse event (SAE), the department director will be notified within 48 hours to evaluate whether the SAE is definitively or possibly treatment-related. Should a plausible causal relationship be identified, subsequent safety measures will be implemented following consultation with the designated safety officer, in accordance with established risk management protocols.
Frequency and plans for auditing trial conduct
The research group will convene on a monthly basis to review the progress of the trial and address any potential issues that may arise. Furthermore, an independent monitor will annually verify the presence and integrity of the trail documentation, which include informed consent, inclusion and exclusion criteria, as well as source data collection.
Supporting information
S1 File. SPIRIT checklist.
The completed Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2013 checklist for this study protocol.
https://doi.org/10.1371/journal.pone.0342761.s001
(PDF)
Acknowledgments
We sincerely thank all participants who engage in this study.
Role of sponsor
This is an investigator initiated clinical trial. Therefore, the funders will play no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.
References
- 1. GBD 2019 Stroke Collaborators. Global, regional, and national burden of stroke and its risk factors, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Neurol. 2021;20(10):795–820. pmid:34487721
- 2. Kulesh A, Drobakha V, Kuklina E, Nekrasova I, Shestakov V. Cytokine Response, Tract-Specific Fractional Anisotropy, and Brain Morphometry in Post-Stroke Cognitive Impairment. J Stroke Cerebrovasc Dis. 2018;27(7):1752–9. pmid:29610037
- 3. Geng S, Liu N, Meng P, Ji N, Sun Y, Xu Y, et al. Midterm Blood Pressure Variability Is Associated with Poststroke Cognitive Impairment: A Prospective Cohort Study. Front Neurol. 2017;8:365. pmid:28804475
- 4. He A, Wang Z, Wu X, Sun W, Yang K, Feng W, et al. Incidence of post-stroke cognitive impairment in patients with first-ever ischemic stroke: a multicenter cross-sectional study in China. Lancet Reg Health West Pac. 2023;33:100687. pmid:37181529
- 5. Rajda CM, Desabrais K, Levin MF. Relationships Between Cognitive Impairments and Motor Learning After Stroke: A Scoping Review. Neurorehabil Neural Repair. 2025;39(2):142–56. pmid:39606925
- 6. French MA, Cohen ML, Pohlig RT, Reisman DS. Fluid Cognitive Abilities Are Important for Learning and Retention of a New, Explicitly Learned Walking Pattern in Individuals After Stroke. Neurorehabil Neural Repair. 2021;35(5):419–30. pmid:33754890
- 7. Hill G, Regan S, Francis R, Stroke Priority Setting Partnership Steering Group. Research priorities to improve stroke outcomes. Lancet Neurol. 2022;21(4):312–3. pmid:35305334
- 8. Quinn TJ, Richard E, Teuschl Y, Gattringer T, Hafdi M, O’Brien JT, et al. European Stroke Organisation and European Academy of Neurology joint guidelines on post-stroke cognitive impairment. Eur Stroke J. 2021;6(3):I–XXXVIII. pmid:34746430
- 9. Zhou Y, Feng H, Li G, Xu C, Wu Y, Li H. Efficacy of computerized cognitive training on improving cognitive functions of stroke patients: A systematic review and meta-analysis of randomized controlled trials. Int J Nurs Pract. 2022;28(3):e12966. pmid:34036682
- 10. Bäckman L, Lindenberger U, Li S-C, Nyberg L. Linking cognitive aging to alterations in dopamine neurotransmitter functioning: recent data and future avenues. Neurosci Biobehav Rev. 2010;34(5):670–7. pmid:20026186
- 11. Orgeta V, McDonald KR, Poliakoff E, Hindle JV, Clare L, Leroi I. Cognitive training interventions for dementia and mild cognitive impairment in Parkinson’s disease. Cochrane Database Syst Rev. 2020;2(2):CD011961. pmid:32101639
- 12. Eve M, O’Keeffe F, Jhuty S, Ganesan V, Brown G, Murphy T. Computerized Working-Memory Training for Children Following Arterial Ischemic Stroke: A Pilot Study With Long-Term Follow-Up. Appl Neuropsychol Child. 2016;5(4):273–82. pmid:26980059
- 13. Bikson M, Grossman P, Thomas C, Zannou AL, Jiang J, Adnan T, et al. Safety of Transcranial Direct Current Stimulation: Evidence Based Update 2016. Brain Stimul. 2016;9(5):641–61. pmid:27372845
- 14. Nissim NR, O’Shea A, Indahlastari A, Kraft JN, von Mering O, Aksu S, et al. Effects of Transcranial Direct Current Stimulation Paired With Cognitive Training on Functional Connectivity of the Working Memory Network in Older Adults. Front Aging Neurosci. 2019;11:340. pmid:31998111
- 15. Antonenko D, Hayek D, Netzband J, Grittner U, Flöel A. tDCS-induced episodic memory enhancement and its association with functional network coupling in older adults. Sci Rep. 2019;9(1):2273. pmid:30783198
- 16. Nitsche MA, Paulus W. Sustained excitability elevations induced by transcranial DC motor cortex stimulation in humans. Neurology. 2001;57(10):1899–901. pmid:11723286
- 17. Chen Y, Zhao Z, Huang J, Wang T, Qu Y. Computer-aided cognitive training combined with tDCS can improve post-stroke cognitive impairment and cerebral vasomotor function: a randomized controlled trial. BMC Neurol. 2024;24(1):132. pmid:38641827
- 18. Kolskår KK, Richard G, Alnaes D, Dørum ES, Sanders A-M, Ulrichsen KM, et al. Reliability, sensitivity, and predictive value of fMRI during multiple object tracking as a marker of cognitive training gain in combination with tDCS in stroke survivors. Hum Brain Mapp. 2021;42(4):1167–81. pmid:33216408
- 19. Naro A, Billeri L, Manuli A, Balletta T, Cannavò A, Portaro S, et al. Breaking the ice to improve motor outcomes in patients with chronic stroke: a retrospective clinical study on neuromodulation plus robotics. Neurol Sci. 2021;42(7):2785–93. pmid:33159273
- 20. Leon D, Cortes M, Elder J, Kumru H, Laxe S, Edwards DJ, et al. tDCS does not enhance the effects of robot-assisted gait training in patients with subacute stroke. Restor Neurol Neurosci. 2017;35(4):377–84. pmid:28697574
- 21. Leo A, De Luca R, Russo M, Naro A, Bramanti P, Calabrò RS. Role of tDCS in potentiating poststroke computerized cognitive rehabilitation: Lessons learned from a case study. Appl Neuropsychol Adult. 2016;23(3):162–6. pmid:26506950
- 22. Kazinczi C, Kocsis K, Boross K, Racsmány M, Klivényi P, Vécsei L, et al. The effect of computerized cognitive training and transcranial direct current stimulation on working memory among post-stroke individuals: a systematic review with meta-analysis and meta-regression. BMC Neurol. 2024;24(1):314. pmid:39232643
- 23. Brunoni AR, Valiengo L, Baccaro A, Zanao TA, de Oliveira JF, Vieira GP, et al. Sertraline vs. ELectrical Current Therapy for Treating Depression Clinical Trial--SELECT TDCS: design, rationale and objectives. Contemp Clin Trials. 2011;32(1):90–8. pmid:20854930
- 24. Freitas S, Simões MR, Alves L, Vicente M, Santana I. Montreal Cognitive Assessment (MoCA): validation study for vascular dementia. J Int Neuropsychol Soc. 2012;18(6):1031–40. pmid:22676901
- 25. Nikolin S, Martin D, Loo CK, Boonstra TW. Effects of TDCS dosage on working memory in healthy participants. Brain Stimul. 2018;11(3):518–27. pmid:29361442
- 26. Moher D, Hopewell S, Schulz KF, Montori V, Gøtzsche PC, Devereaux PJ, et al. CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c869. pmid:20332511
- 27. White IR, Royston P, Wood AM. Multiple imputation using chained equations: Issues and guidance for practice. Stat Med. 2011;30(4):377–99. pmid:21225900
- 28. Miniussi C, Harris JA, Ruzzoli M. Modelling non-invasive brain stimulation in cognitive neuroscience. Neurosci Biobehav Rev. 2013;37(8):1702–12. pmid:23827785
- 29. Karabanov A, Ziemann U, Hamada M, George MS, Quartarone A, Classen J, et al. Consensus Paper: Probing Homeostatic Plasticity of Human Cortex With Non-invasive Transcranial Brain Stimulation. Brain Stimul. 2015;8(3):442–54. pmid:26050599
- 30. Hurley R, Machado L. Using tDCS priming to improve brain function: Can metaplasticity provide the key to boosting outcomes? Neurosci Biobehav Rev. 2017;83:155–9. pmid:29020606
- 31. Rezaei S, Khanmohammadi R. Comparison of short- and long-term effects of neurofeedback and transcranial electrical stimulation on the motor learning in healthy adults. Behav Brain Res. 2025;476:115263. pmid:39307285
- 32. Rolfe DF, Brown GC. Cellular energy utilization and molecular origin of standard metabolic rate in mammals. Physiol Rev. 1997;77(3):731–58. pmid:9234964
- 33. Bélanger M, Allaman I, Magistretti PJ. Brain energy metabolism: focus on astrocyte-neuron metabolic cooperation. Cell Metab. 2011;14(6):724–38. pmid:22152301
- 34. de Groot M, Ikram MA, Akoudad S, Krestin GP, Hofman A, van der Lugt A, et al. Tract-specific white matter degeneration in aging: the Rotterdam Study. Alzheimers Dement. 2015;11(3):321–30. pmid:25217294
- 35. Chen W-L, Wagner J, Heugel N, Sugar J, Lee Y-W, Conant L, et al. Functional Near-Infrared Spectroscopy and Its Clinical Application in the Field of Neuroscience: Advances and Future Directions. Front Neurosci. 2020;14:724. pmid:32742257
- 36. Owens CD, Pinto CB, Mukli P, Gulej R, Velez FS, Detwiler S, et al. Neurovascular coupling, functional connectivity, and cerebrovascular endothelial extracellular vesicles as biomarkers of mild cognitive impairment. Alzheimers Dement. 2024;20(8):5590–606. pmid:38958537