Effects of open- versus closed-skill exercise combined with mindfulness training on inhibitory control in children with ADHD: protocol for a three-arm randomised controlled trial with fMRI

Abstract

Introduction

Impaired inhibitory control is a core neurocognitive deficit in children with attention deficit hyperactivity disorder (ADHD). Physical exercise and mindfulness training are each promising non-pharmacological approaches for ADHD; however, whether combining mindfulness with different exercise modalities, particularly team-based and individual-based open-skill exercise versus closed-skill exercise, yields differential effects on inhibitory control and its neural substrates remains unknown. This study aims to compare these intervention approaches in children with ADHD while exploring associated neural correlates.

Methods and analysis

This is a single-centre, assessor-blinded and statistician-blinded, three-arm parallel-group randomised controlled trial. 60 children with ADHD aged 6–8 years will be randomly assigned (1:1:1) to one of three 8-week intervention groups: (1) team-based open-skill exercise (basketball) combined with mindfulness training, (2) individual-based open-skill exercise (badminton) combined with mindfulness training or (3) closed-skill exercise (structured aerobic activity) combined with mindfulness training. The co-primary outcomes are behavioural inhibitory control, assessed using the Go/No-Go task and Stroop Colour–Word Test at baseline, postintervention and 3-month follow-up. Secondary outcomes include attention, ADHD symptom severity, executive function, health-related quality of life and resting-state functional MRI indices.

Ethics and dissemination

This trial has received ethical approval from the Ethics Committee of the Affiliated Changzhou Children’s Hospital of Nantong University (Approval No.: 2024-029; Approval Date: 2 January 2025) and will be conducted in accordance with the Declaration of Helsinki (2024 revision). The protocol has been prospectively registered with the Chinese Clinical Trial Registry. Findings will be disseminated through peer-reviewed publications and academic conferences.

Trial registration number

ChiCTR2500108809.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This trial employs a three-arm parallel-group randomised controlled design with concealed allocation, blinded outcome assessors and a blinded statistician.

• Outcomes are assessed at three time points (baseline, postintervention and 3-month follow-up) using a multimodal protocol that combines validated behavioural tasks, resting-state functional MRI and parent-reported measures.

• An active control condition matched for session duration and frequency is included to minimise confounding from non-specific training effects.

• The single-centre design and the inability to blind participants and intervention providers may limit external validity and introduce performance bias.

• The sample size is powered for three-group comparison but not for pairwise comparisons between open-skill subgroups, and the combined design precludes isolation of individual intervention components.

Introduction

Attention deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder characterised by inattention, hyperactivity and impulsivity, with impaired inhibitory control widely recognised as a core neurocognitive deficit.1,3 Deficits in inhibitory control contribute to academic and social difficulties.^{4 5} Although pharmacotherapy remains first-line treatment, its long-term use is often limited by adverse effects and poor adherence,⁶ highlighting the need for effective non-pharmacological interventions.^{7 8}

Physical exercise has emerged as a promising non-pharmacological strategy for ADHD.⁴ Network meta-analyses demonstrate that open-skill activities such as basketball and badminton, which require continuous adaptation to dynamic environments, produce greater benefits for inhibitory control than closed-skill activities such as running and cycling.9,11 However, open-skill exercise is not homogeneous.¹⁰ Team-based sports demand distributed attention and rapid decision-making across multiple players, whereas individual sports emphasise sustained, focused attention on a single opponent.¹² These distinct cognitive demand profiles may differentially influence inhibitory control, yet no trial has directly compared them in children with ADHD.¹³

Mindfulness training is a complementary non-pharmacological approach that targets top-down cognitive regulation through meta-awareness and cognitive flexibility—processes directly relevant to inhibitory control and impaired in ADHD.^{14 15} Evidence from intervention studies indicates that mindfulness can improve inhibitory control in children with ADHD,¹⁶ with neuroimaging findings linking these effects to enhanced prefrontal cortical activity,¹⁴ which parallels and potentially complements the neural pathways modulated by physical exercise.¹⁷ The integration of mindfulness with exercise is supported by three converging rationales: (1) both modulate overlapping prefrontal–striatal circuits; (2) mindfulness may enhance cognitive engagement during open-skill exercise and (3) mindfulness components may improve adherence in paediatric populations.^{14 18 19} In the present trial, mindfulness is delivered as a standardised component across all groups, enabling isolation of exercise modality effects.

At the neurobiological level, inhibitory control deficits in ADHD are associated with prefrontal–striatal–cerebellar dysfunction. Resting-state functional MRI (rs-fMRI) provides sensitive indices of intervention-induced neuroplasticity, particularly amplitude of low-frequency fluctuations (ALFF) and degree centrality (DC). However, no study has examined how different exercise modalities combined with mindfulness influence these neural markers in children with ADHD. Therefore, this three-arm randomised controlled trial aims to compare team-based open-skill exercise, individual-based open-skill exercise and closed-skill aerobic exercise, each combined with mindfulness training, on behavioural measures of inhibitory control as co-primary outcomes and associated neurobiological mechanisms using rs-fMRI in children with ADHD. We hypothesise that team-based open-skill exercise will produce the greatest improvements, followed by individual-based open-skill exercise, then closed-skill exercise, mediated by enhanced prefrontal–striatal neural activity. By integrating behavioural and neuroimaging assessments, this study seeks to inform more targeted, evidence-based exercise prescriptions for children with ADHD.

Methods and analysis

Design

This is a single-centre, three-arm, parallel-group randomised controlled trial with a 1:1:1 allocation ratio. The protocol adheres to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT 2013) Statement; the completed SPIRIT checklist is provided in online supplemental file 1. The anticipated participant flow is illustrated in figure 1.

Figure 1. Anticipated participant flow diagram. fMRI, functional MRI.

Participants, recruitment and eligibility criteria

The study will be conducted in Changzhou, Jiangsu Province, China. Participants will be children with ADHD recruited via the outpatient clinics of Changzhou Children’s Hospital, with supplementary recruitment through local primary schools and community health centres to ensure broad access within the target region.

Written informed consent will be obtained from parents or legal guardians before study commencement. Detailed information is provided in online supplemental file 2. Developmentally appropriate assent, either verbal or written based on the child’s cognitive maturity, will also be obtained from all child participants in accordance with paediatric research ethics guidelines.

Key inclusion criteria

• Children aged 6–8 years.

• Diagnosis of ADHD (combined or inattentive presentation) confirmed by two independent paediatric specialists in accordance with the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria.

• Full-scale intelligence quotient (FSIQ) ≥90 as measured by the Wechsler Intelligence Scale for Children, Fourth Edition.

• Medication-free from ADHD medications for ≥4 weeks before enrolment, with written agreement to abstain from ADHD pharmacotherapies throughout the study period.

• Basic motor skills such as running, jumping and balancing confirmed by physician assessment to ensure safety during exercise interventions.

• Willingness to complete all study procedures, including rs-fMRI scanning, assessed via a pre-enrolment practice session.

Key exclusion criteria

• Comorbid neurodevelopmental disorders such as autism spectrum disorder, intellectual disability or specific learning disorder, as well as Tourette syndrome or epilepsy.

• Diagnosis of schizophrenia, bipolar disorder or elevated suicide risk as defined by a Columbia-Suicide Severity Rating Scale score ≥2 in the past month.

• Untreated major depressive disorder, defined as a Children’s Depression Inventory score ≥20, or post-traumatic stress disorder (PTSD). PTSD was screened during the baseline clinical interview conducted by a licensed child psychiatrist based on DSM-5 criteria.

• Significant uncontrolled comorbidities affecting the cardiopulmonary, metabolic, musculoskeletal or neuromuscular systems that may pose safety risks during exercise.

• Colour vision deficiency or uncorrected severe visual or auditory impairment that may affect task performance or intervention engagement.

• Known contraindications to MRI, including metallic implants or severe claustrophobia.

• Participation in ADHD behavioural interventions within 4 weeks before enrolment or concurrent participation in another interventional study.

Trial status

The protocol (v1.0) was finalised in December 2024 and approved by the local Ethics Committee in January 2025. Internal procedural testing was conducted between February and August 2025 to refine intervention protocols and assessment procedures. No participants were enrolled in the formal randomised trial before registration, and no data from internal procedural testing will be included in the formal analysis. The trial was prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2500108809) in September 2025, and formal participant recruitment commenced immediately thereafter, concluding in December 2025. As of submission, the 8-week intervention is ongoing (January–June 2026), with T1 postintervention assessments scheduled between March and August 2026 and T2 follow-up assessments scheduled between June and November 2026.

Patient and public involvement

Patients and the public were not involved in the design, conduct, reporting or planning of dissemination for this research protocol. The study design and intervention components were determined by the clinical research team based on existing empirical evidence and paediatric guidelines. A lay summary of the trial results will be provided to participants and their guardians on completion of the study.

Randomisation and blinding

A non-affiliated biostatistician, with no role in participant enrolment, intervention delivery or outcome assessment, will generate the randomisation sequence using a computerised random number generation tool. Stratified block randomisation (block size=6) will be applied with a 1:1:1 allocation ratio, stratified by sex and baseline FSIQ (dichotomised at the median) to ensure balanced group characteristics.

Allocation concealment will be achieved via a centralised, password-protected web-based system. After baseline assessments, an independent study coordinator will enter participant details, and the system will generate a fixed group assignment (A, B or C).

Blinding of participants, guardians and intervention providers is not feasible due to the nature of the interventions. To minimise bias, outcome assessors and data analysts will remain masked to group allocation. Assessor blinding will be evaluated at study completion using a standardised questionnaire.

Interventions

All participants will receive an 8-week intervention (5 sessions/week), consisting of 40 min of exercise followed by 20 min of mindfulness training. Participants will be allocated to one of three groups: (1) basketball combined with mindfulness, (2) badminton combined with mindfulness or (3) structured aerobic exercise combined with mindfulness. Exercise intensity will be monitored using heart rate (Polar H10) and perceived exertion scales. All groups will follow a progressive training structure with intensity and complexity increasing every 2 weeks. Detailed session-by-session protocols are provided in online supplemental file 3. The slightly lower heart rate target for group C, 60%–75% of maximum heart rate (HRmax), reflects the typical intensity profile of closed-skill aerobic activities; all groups maintain moderate-intensity exercise.

Mindfulness training component

A standardised 20-minute mindfulness session will be delivered after each exercise session across all groups. Drawing on a child-adapted mindfulness protocol validated for ADHD and specifically adapted for Chinese children with ADHD, the session uses a prerecorded audio script grounded in Mindfulness-Based Stress Reduction and Mindfulness-Based Cognitive Therapy frameworks. Core practices include mindful breathing, body scan and meta-awareness of thoughts and emotions, contextualised with developmentally appropriate metaphors to enhance engagement.²⁰ The full protocol and script are provided in table 1 and online supplemental file 3.

Table 1. Combined intervention protocol.

Category	Component	Group A: Basketball	Group B: Badminton	Group C: Aerobic
Exercise training parameters	Warm-up	Dynamic stretching and joint mobilisation
	Main session	Progressive basketball training (dribbling, passing, small-sided games)	Progressive badminton training (racket skills, rally practice, modified games)	Structured aerobic activities (treadmill, cycling, rope skipping, brisk walking)
	Training goal	Reactive agility and decision-making and visuospatial judgement	Visuomotor coordination and anticipation and positioning	Moderate aerobic intensity with low cognitive load
	Cool-down	Static stretching and breathing
	Intensity	75% HRmax (220−age); RPE 5–6	75% HRmax (220−age); RPE 5–6	60%–75% HRmax (220−age); RPE 4–6
	Monitoring	Heart rate (Polar H10) and perceived exertion (Borg CR-10 scale)
	Session duration	Warm-up: 5 min; Main Session: 30 min; Cool-down: 5 min (total 40 min exercise per session)
	Frequency	5 sessions/week, 8 weeks (40 sessions total)
Mindfulness training parameters (all groups)	Timing	20 min mindfulness training, commencing within 5 min after exercise
	Structure	Guided introduction, core mindfulness practice and group discussion
	Core content	Mindful breathing, body scan and awareness of thoughts and emotions
	Delivery	Standardised audio protocol adapted from MBSR/MBCT frameworks
	Instructor	Certified instructors with ADHD-specific training

Within each group, the assigned exercise modality progresses through 2-week phases; exercise and mindfulness training are delivered sequentially within each session. Each session comprises 40 min of exercise followed by 20 min of mindfulness training, for a total of 60 min.

ADHD, attention deficit hyperactivity disorder; Borg CR-10 scale, Borg Category-Ratio 10 scale; HRmax, maximum heart rate; MBCT, mindfulness-based cognitive therapy; MBSR, mindfulness-based stress reduction; RPE, rating of perceived exertion.

Intervention personnel and training

All sessions will be delivered by trained exercise instructors and certified mindfulness practitioners who have completed study-specific training.

Intervention fidelity

Intervention fidelity will be ensured through standardised protocols, heart rate monitoring and perceived exertion ratings. 20% of sessions, selected by stratified random sampling by week and group, will be video-reviewed by independent raters using a standardised checklist (online supplemental file 4). Deviations exceeding 10% will trigger retraining. Adherence rates for exercise and mindfulness sessions will be recorded and reported.

Concomitant care and interventions

Participants will continue usual care but will be asked not to initiate new ADHD-related non-pharmacological interventions during the study. Medication changes will be recorded. Following completion of all study assessments, participants in group C will be offered an open-skill exercise session.

Outcomes

Primary outcomes

The co-primary outcomes are behavioural measures of inhibitory control, assessed at baseline (T0), postintervention (T1) and 3-month follow-up (T2).

Inhibitory control will be assessed using a computerised Go/No-Go task, in which children respond to frequent ‘Go’ stimuli and withhold responses to infrequent ‘No-Go’ stimuli. No-Go accuracy will serve as the primary index.²¹

Cognitive interference control will be assessed using the computerised Stroop Colour–Word Test. The interference effect, defined as the reaction time difference between incongruent and control conditions, will serve as the primary index.

Secondary outcomes

Secondary outcomes include attention, ADHD symptom severity, executive function, health-related quality of life and neuroimaging measures, assessed at T0, T1 and T2.

Attention will be assessed using the Integrated Visual and Auditory Continuous Performance Test (IVA-CPT), with the Response Control Quotient (RCQ) as the primary index.

ADHD symptom severity will be rated by parents using the Swanson, Nolan and Pelham Questionnaire-IV (SNAP-IV).²² Inattention and hyperactivity–impulsivity subscale scores will be reported.

Executive function will be assessed using the Behaviour Rating Inventory of Executive Function, Second Edition (BRIEF-2). The Behavioural Regulation Index, Metacognition Index and Global Executive Composite will be reported.²³

Health-related quality of life will be evaluated using the Pediatric Quality of Life Inventory 4.0 Generic Core Scales (PedsQL 4.0), which covers physical, emotional, social and school functioning. The total score will be reported.^{6 24}

Neuroimaging outcomes will include rs-fMRI measures, specifically ALFF and DC within predefined regions of interest, including the inferior frontal gyrus. Detailed acquisition and preprocessing procedures are provided in online supplemental file 5. The schedule of enrolment, interventions and assessments is presented in table 2.

Table 2. Schedule of enrolment, interventions and assessments.

Study period	Enrolment	Allocation	Baseline (T0)	Intervention (weeks 1–8)	Postintervention (T1)	Follow-up (T2)
Time point	−2 to 0 week	Week 0	Week 0	Weeks 1–8	Within 1 week after Week 8	3 months after T1
Enrolment
Eligibility screening	×
Informed consent	×
Allocation
Randomisation		×
Interventions
Group A: basketball+mindfulness				×
Group B: badminton+mindfulness				×
Group C: aerobic exercise+mindfulness				×
Assessments
Demographics			×
WISC-IV			×
Primary outcomes
Go/No-Go Task			×		×	×
Stroop Colour-Word Test			×		×	×
Secondary outcomes
IVA-CPT			×		×	×
SNAP-IV			×		×	×
BRIEF-2			×		×	×
PedsQL 4.0			×		×	×
rs-fMRI			×		×	×
Safety monitoring
AE	×	×	×	× (ongoing)	×	×

AE, adverse event; BRIEF-2, Behavior Rating Inventory of Executive Function-2; IVA-CPT, Integrated Visual and Auditory Continuous Performance Test; PedsQL 4.0, Pediatric Quality of Life Inventory 4.0 Generic Core Scales; rs-fMRI, resting-state functional magnetic resonance imaging; SNAP-IV, Swanson, Nolan and Pelham Questionnaire-IV; WISC-IV, Wechsler Intelligence Scale for Children-IV.

Sample size

Sample size was estimated using G*Power V.3.1 for a one-way analysis of variance (ANOVA) with three groups (α=0.05, two-sided; power=0.80; Cohen’s f=0.40), indicating a required sample size of 54 participants (18 per group). The effect size estimate (f=0.40) was informed by a meta-analysis reporting a pooled effect of Hedges’ g=0.84 for physical activity on executive function in paediatric ADHD populations.^{10 25} Given the use of an active comparator, this effect size was considered a conservative and plausible estimate. To account for an anticipated 10% attrition rate,^{26 27} we plan to recruit 60 participants (20 per group), ensuring at least 18 analysable participants per group for the primary outcome analysis. Trial feasibility will be evaluated based on recruitment, retention and adherence rates. Neuroimaging outcomes are considered secondary and exploratory.

Data collection and management

Data will be collected and managed in accordance with the Declaration of Helsinki and local data protection regulations, following an ethics-approved data management plan. All data will be de-identified at the point of collection and assigned a unique study ID.

Behavioural tasks (Go/No-Go Task, Stroop Colour–Word Test and IVA-CPT) will be administered in a quiet, standardised environment using computer-based protocols. Parent-reported questionnaires (SNAP-IV, BRIEF-2 and PedsQL 4.0) will be completed independently following standardised instructions. rs-fMRI data will be acquired using a 3-Tesla Siemens Prisma scanner with paediatric-adapted procedures. All outcome assessors and data collectors will receive standardised training and will remain blinded to group allocation throughout the study.

Neuroimaging data will be preprocessed using the Data Processing & Analysis for Brain Imaging toolbox following standard pipelines, including motion correction, spatial normalisation, and temporal filtering.²⁸ Participants with more than 20% of volumes exceeding framewise displacement >0.5 mm will be excluded from neuroimaging analyses.

All data will be securely stored on password-protected servers with restricted access. Paper records will be digitised and stored in encrypted formats. Data will be retained for a period of 10 years in accordance with institutional and regulatory requirements.

Data analysis

Statistical analyses will be performed using R (V.4.4.0 or later) and SPSS (V.26.0) with a two-sided α=0.05. Primary analyses will follow the intention-to-treat principle, including all randomised participants, with a per-protocol sensitivity analysis for those completing ≥80% of intervention sessions and all scheduled assessments. Baseline characteristics will be compared across groups using one-way ANOVA or Kruskal-Wallis tests for continuous variables and χ² tests for categorical variables.

Primary outcomes (Go/No-Go accuracy and Stroop interference) will be analysed using linear mixed-effects models (LMMs) with fixed effects of group, time and group×time interaction, and participant as a random intercept, adjusting for baseline values, age and sex. Bonferroni correction will be applied for the two coprimary outcomes (α=0.025), with significant interactions, followed by post hoc pairwise comparisons. Secondary behavioural outcomes (IVA-CPT RCQ, SNAP-IV, BRIEF-2, PedsQL 4.0) will use the same LMM specification. Neuroimaging outcomes (ALFF, DC) at T1 will be compared using ANCOVA adjusting for baseline, age, sex and mean framewise displacement; T2 data will be explored to examine the persistence of neural changes. Whole-brain analyses will apply family-wise error (FWE) correction at p<0.05.

Missing data will be handled under the missing at random (MAR) assumption via multiple imputation (20 datasets) including all analysis variables and key predictors of missingness. Prespecified exploratory analyses include correlations between neuroimaging changes and behavioural outcomes, moderation analyses by baseline characteristics and subgroup comparisons between open-skill exercise groups, with results reported using exact p values.

Trial oversight, monitoring and safety

Safety monitoring

An independent Data and Safety Monitoring Board (DSMB) will be established before the trial commences, comprising a paediatric neurodevelopmental specialist, a biostatistician and a clinical ethicist, none of whom have conflicts of interest related to this study. The DSMB will meet quarterly to review unblinded aggregate data on recruitment, retention, adherence and safety, and will not conduct interim efficacy analyses. Ad hoc meetings may be convened at any time in response to urgent safety concerns. The DSMB is empowered to recommend trial continuation, protocol modification or early termination to protect participant welfare.

All adverse events (AEs) and serious adverse events (SAEs) will be monitored at enrolment, allocation, baseline (T0), throughout the 8-week intervention, at postintervention (T1) and at 3-month follow-up (T2) (see table 2). Events will be recorded in a password-protected electronic system, including onset date, severity, duration, management and outcome. Severity will be classified as mild, moderate or severe according to the Common Terminology Criteria for Adverse Events (v5.0 Paediatric Module). A blinded paediatrician will assess causality (related, possibly related, unlikely related or unrelated). SAEs must be reported to the principal investigator within 24 hours and subsequently to the Ethics Committee in accordance with institutional requirements, forwarded to the DSMB within 48 hours, and reported to the Chinese Clinical Trial Registry within 7 days.

Withdrawal, intervention modification and trial termination

Participants may be withdrawn if a new exclusion criterion arises during the study, if intervention attendance falls below 80%, if a SAE precludes continued participation, or if the family voluntarily withdraws in writing. On a case-by-case basis, a treating physician may modify the intervention (eg, reducing exercise intensity) when necessary for participant safety. The intervention will be permanently discontinued if continuation poses an unacceptable risk. Withdrawn participants will retain full access to routine medical care, as described under post-trial care.

The trial may be terminated early on written recommendation from the DSMB or the Ethics Committee. Grounds for early termination include SAE rates exceeding 10%, major protocol violations compromising data integrity or recruitment failure.

In addition, feasibility will be monitored in terms of recruitment, retention, adherence and data completeness. Recruitment is planned to enrol 60 participants between September and December 2025, with a reassessment conducted 6 weeks after initiation and adjustments made if necessary. Retention is targeted at ≥85% from randomisation to postintervention (T1), supported by regular parent contact, flexible scheduling and small incentives. Adherence is targeted at a mean attendance rate of ≥80%, with weekly monitoring and support as needed. Data completeness aims to limit attrition for primary neuroimaging outcomes to ≤10%, with ongoing quality monitoring and standardised procedures to minimise data loss.

Ethics and dissemination

Ethical approval, registration and ongoing compliance

This trial has received ethical approval from the Ethics Committee of the Affiliated Changzhou Children’s Hospital of Nantong University and will be conducted in accordance with the Declaration of Helsinki (2024 revision) and the SPIRIT 2013 Statement. The protocol has been prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2500108809) prior to participant recruitment.

To ensure ongoing ethical compliance, progress reports covering recruitment status, AEs and protocol adherence will be submitted to the Ethics Committee after completion of the 8-week intervention. A mid-trial ethical review will be conducted at approximately week 4 to assess participant safety and trial conduct. A final report, including outcome data and safety summaries, will be submitted within 3 months after the T2 follow-up assessment. Any protocol amendments affecting study conduct will require Ethics Committee approval prior to implementation and will be communicated to investigators, updated in the trial registry and disclosed to participants and guardians where relevant.

Informed consent and participant protection

Written informed consent will be obtained from each participant’s parent or legal guardian prior to any study-related procedures. For children aged 6–8 years, developmentally appropriate assent will also be obtained using simplified, picture-assisted information sheets accompanied by verbal explanation. Participation is voluntary, and families may withdraw at any time without providing a reason and without affecting the child’s medical care or access to services.

The trial sponsor has obtained clinical trial insurance to cover compensation for trial-related harm. For AEs assessed as related or possibly related to the intervention, the research team will arrange medical referral and follow-up care at no cost until the condition resolves or stabilises. Compensation claims must be submitted within 12 months of the participant’s final study assessment.

Data sharing and dissemination

De-identified individual participant data generated during the trial will be available from the corresponding author on reasonable request after publication, subject to approval by the research team, the sponsor institution and relevant ethical requirements. No personally identifiable information will be shared. Trial results will be submitted for publication in peer-reviewed journals and presented at academic conferences regardless of the direction or significance of findings. Plain-language summaries will be provided to participating families free of charge.

Discussion

This study is designed to provide a rigorous comparison of different exercise modalities, each combined with mindfulness training, for improving inhibitory control in children with ADHD. Delivering standardised mindfulness training across all groups helps control for non-specific cognitive influences, while the three-arm design with blinded outcome assessment enables direct comparison between open-skill and closed-skill exercise modalities. The multimodal outcome framework, integrating behavioural tasks, parent-reported symptom and executive-function measures, quality-of-life assessment and rs-fMRI, provides a comprehensive assessment of behavioural, clinical and neurobiological outcomes. rs-fMRI measures are included as secondary mechanistic outcomes to explore neurobiological substrates of behavioural changes rather than as indicators of clinical efficacy. Intervention fidelity is supported by multiple safeguards, as detailed in the Methods section. Several limitations should be acknowledged. First, the single-centre design may limit generalisability. Second, the sample size (N=60) is not powered to detect differences between the two open-skill exercise subgroups; these comparisons are exploratory. Third, blinding of participants and intervention providers is not feasible because of the nature of the interventions, although this is mitigated by blinded outcome assessment and prioritisation of objective measures. Fourth, the combined design does not permit disentanglement of the independent effects of mindfulness training and exercise; future factorial designs may address this issue. Finally, given the young age of participants, motion artefacts in neuroimaging remain a recognised challenge, and fMRI findings will therefore be interpreted cautiously.

The funder had no role in study design, protocol development, data collection, analysis or the decision to submit for publication. This trial is sponsored by Affiliated Changzhou Children’s Hospital of Nantong University (958 Zhongwu Avenue, Tianning District, Changzhou, Jiangsu Province, China). Contact person: JIANG Kaihua, Chief of the Department of Science and Education; Email: kjc_czch@ntu.edu.cn; Tel: +86-519-69808062. As the sponsor, the hospital is legally responsible for trial conduct, quality assurance and safety reporting, with ultimate authority over trial design, data management and result dissemination. The definitive dataset will be owned by the sponsor institution.