Neurofeedback Strengthens Executive Function in Children with ADHD

Executive Summary

For many children with attention-deficit/hyperactivity disorder (ADHD), the central struggle is not simply paying attention but regulating thought and behavior. A 2025 systematic review and meta-analysis by Zhong et al., published in Scientific Reports (a Nature Portfolio journal), pooled data from 17 randomized controlled trials and 939 children to address a practical question: Does neurofeedback measurably improve executive function?

The answer is a qualified yes.

Across studies, neurofeedback produced small but statistically significant gains in inhibitory control and working memory, the two executive domains most reliably impaired in ADHD. The benefits depended heavily on dose, emerging mainly after roughly 1,260 minutes (about 21 hours) of training, and several gains persisted for 6 to 12 months.

For clinicians, the message is measured optimism: neurofeedback is a credible adjunct, not a quick fix, and it rewards sustained, well-designed protocols (Zhong et al., 2025).

Why Executive Function Sits at the Heart of ADHD

ADHD is among the most common neurodevelopmental conditions of childhood (Polanczyk et al., 2007; Zhong et al., 2025). Increasingly, researchers frame it less as a pure attention problem and more as a disorder of self-regulation. Executive function — the family of higher-order processes that let a child inhibit impulses, hold information in mind, and shift flexibly between tasks — is central to that view (Diamond, 2013). Roughly half of children with ADHD show measurable executive deficits (Zhong et al., 2025).

Two domains stand out. Inhibitory control, the capacity to suppress impulsive or inappropriate responses, underlies a child’s ability to wait, resist distraction, and think before acting. Working memory, the ability to hold and manipulate information briefly, supports following multi-step instructions and staying on task.

A third domain, cognitive flexibility — shifting between mental sets or strategies — proved harder to study and, as we will see, remained underexamined here (Zhong et al., 2025).

These deficits do not stay confined to the testing room. They ripple outward into academic struggles, emotional dysregulation, and strained peer relationships, which is precisely why effective intervention matters.

How Neurofeedback Is Supposed to Work

Neurofeedback training (NFT) is a form of biofeedback in which a child sees or hears real-time information about their own brain activity and gradually learns to nudge it in a healthier direction.

The learning mechanism is operant conditioning: desired brain states are rewarded, so they become more frequent. Most ADHD protocols rely on electroencephalography (EEG), which records the brain’s electrical activity from the scalp; a smaller number use real-time functional magnetic resonance imaging (fMRI) to target specific circuits (Zhong et al., 2025).

The premise is that repeatedly self-regulating neural activity promotes neuroplasticity — the brain’s capacity to reorganize with experience — and strengthens the networks behind executive control.

Protocols differ in what they train. The most common is theta beta ratio (TBR) training, which addresses the elevated slow-wave (theta) relative to faster (beta) activity often seen in ADHD.

Other approaches include slow cortical potentials (SCP) training, sensory motor rhythm (SMR) training, peak alpha frequency training, and individualized paradigms (Zhong et al., 2025). The shared hope is that skills practiced in the training room transfer to real-world attention and behavior.

What the Researchers Asked and How They Tested It

Zhong et al. (2025) searched PubMed, EBSCO, and Web of Science for studies published between January 2000 and January 2024. They included randomized controlled trial (RCT) designs and related controlled studies of children aged 6 to 18 who met DSM or ICD criteria for ADHD, provided each used validated cognitive or behavioral measures of executive function.

Seventeen trials cleared the bar, comprising 939 children; about half received EEG or fMRI neurofeedback, and the rest formed control groups receiving no treatment, treatment as usual, physical activity, cognitive training, electromyographic biofeedback, behavior therapy, or medication.

The trials spanned eleven countries — the United Kingdom, Italy, Canada, the Netherlands, Germany, Poland, Switzerland, Iran, Spain, India, and the United States — with total training ranging widely from 119 to 2,400 minutes. That spread in dose later proved important.

The methodology was deliberately rigorous. The review followed PRISMA reporting standards and was preregistered through PROSPERO, which reduces the risk of selective reporting (Zhong et al., 2025).

Two reviewers independently screened and extracted data, resolving disagreements with a third. Study quality was rated with the PEDro scale, a validated tool for appraising clinical trials; scores ran from 6 to 10 (average 7.76).

One quality component, allocation concealment — keeping group assignment hidden until enrollment — was met by 12 of the 17 trials.

To combine findings measured on different scales, the authors computed the standardized mean difference (SMD), an effect size that expresses each result in common units.

They gauged heterogeneity — variability across studies — with the Q statistic and Higgins’ I², using fixed-effects models when studies agreed and random-effects models when they did not (Zhong et al., 2025).

What the Numbers Showed

The headline result is that neurofeedback produced statistically significant improvements in both inhibitory control and working memory.

For inhibitory control, 12 studies (640 participants) yielded an SMD of 0.36 (95% CI, 0.18 to 0.53; p < 0.0001), with low heterogeneity (I² = 26%).

For working memory, seven studies (370 participants) yielded an SMD of 0.37 (95% CI 0.01 to 0.74, p < 0.05), although heterogeneity was high (I² = 65%), signaling that not all protocols delivered equally (Zhong et al., 2025).

Global executive function, captured mainly by the Behavior Rating Inventory of Executive Function (BRIEF), was examined in only three studies. The pooled effect favored neurofeedback (SMD = −0.44, 95% CI −0.81 to −0.07, p < 0.05); because lower BRIEF scores indicate better functioning, the negative value reflects improvement (Zhong et al., 2025).

These findings align with much of the prior literature, even as they run counter to at least one earlier meta-analysis that reported no significant executive benefit (Louthrenoo et al., 2022).

The Dose-Response Story: Why Minutes Matter

One of the most clinically useful findings concerns dose. Splitting the trials at the median of 1,260 minutes (about 21 hours), the authors compared shorter and longer courses. Below that threshold, neither inhibitory control (six studies, SMD = 0.22, p > 0.05) nor working memory (four studies, SMD = 0.31, p > 0.05) improved significantly (Zhong et al., 2025).

Above the threshold, the picture changed. Longer courses produced significant gains in inhibitory control (five studies, SMD = 0.30, 95% CI 0.10 to 0.58, p < 0.01) and working memory (three studies, SMD = 0.441, 95% CI 0.12 to 0.71, p < 0.01).

The pattern suggests that executive change requires extended, repeated practice before it transfers to behavior — a point worth emphasizing with families from the first session (Zhong et al., 2025).

Do the Gains Last?

Persistence is where neurofeedback becomes genuinely interesting. Six to twelve months after training ended, working memory gains remained robust (three studies, SMD = 0.63, 95% CI 0.19 to 1.07, p < 0.01), while inhibitory control showed a marginally sustained effect (two studies, SMD = 0.35, 95% CI 0.00 to 0.69, p = 0.05).

This durability mirrors earlier work showing that neurofeedback’s effects on attention and impulsivity can outlast the training period (Van Doren et al., 2019; Zhong et al., 2025).

The implication is that gains may reflect lasting neural adaptation rather than temporary task practice — a meaningful distinction when justifying a demanding protocol to a family.

Mechanisms: Different Protocols, Different Targets

The review also connects protocols to plausible mechanisms.

TBR training may sharpen response inhibition by reshaping the theta and beta dynamics tied to cognitive control. SCP training may strengthen cortical regulation and attentional allocation by modulating the contingent negative variation (CNV), an event-related potentials (ERP) marker of anticipation and preparation (Zhong et al., 2025).

Outcomes were measured with familiar laboratory tasks — the Continuous Performance Task (CPT) and Go/No-go task for inhibitory control, and span and n-back tasks for working memory. This is a useful reminder that “executive function” is only as meaningful as the tools used to measure it, and that task choice partly explains the heterogeneity seen across studies.

Strengths, Limitations, and Honest Caveats

The review’s strengths are real. It pooled a relatively large evidence base of 17 controlled trials, followed PRISMA and preregistration, systematically appraised study quality, and went beyond immediate outcomes to examine both dose and durability (Zhong et al., 2025). Those last two questions are exactly the ones clinicians and families ask.

The limitations deserve equal candor. Cognitive flexibility could not be meaningfully pooled because too few trials assessed it. Working memory results were highly heterogeneous, and the authors detected publication bias specifically within the long-duration subgroup analyses, where positive trials may be overrepresented. Effect sizes, though significant, were small to moderate — neurofeedback is an adjunct, not a cure (Louthrenoo et al., 2022; Zhong et al., 2025).

What This Means for Your Practice

For psychologists and counselors weighing neurofeedback, several practical steps follow.

First, set expectations early: meaningful executive gains appear to require sustained training, so frame neurofeedback as a multi-month commitment rather than a brief trial.

Second, plan for an adequate dose, since protocols delivering well beyond 1,260 minutes are the ones most likely to produce durable change.

Third, integrate rather than isolate. The evidence is strongest when neurofeedback complements cognitive training, cognitive-behavioral therapy, behavioral parent work, or medication, not when it stands alone (Zhong et al., 2025).

Fourth, measure what you treat by selecting validated tasks and rating scales at baseline and follow-up, so progress is visible to you and the family.

Finally, communicate the durability data honestly, because the prospect of gains lasting many months can sustain motivation through a demanding course of training.

Bringing It Together

Zhong et al. (2025) move the neurofeedback conversation forward by quantifying not just whether it works, but how much, for how long, and under what conditions. The signal is consistent: small but real improvements in inhibitory control and working memory, contingent on a sufficient dose, with working-memory gains that tend to endure.

None of these positions neurofeedback as a standalone remedy. Instead, it supports a pragmatic, multimodal model in which neurofeedback earns its place as one experience-dependent tool for strengthening self-regulation in children with ADHD.

Five Takeaways

1. Neurofeedback produced small but statistically significant improvements in inhibitory control and working memory in children with ADHD.

   
   

2. Dose mattered: gains emerged mainly after about 1,260 minutes (roughly 21 hours) of training, not in shorter courses.

   
   

3. Working memory gains were the most durable, persisting six to twelve months after training ended.

   
   

4. Different protocols likely target different mechanisms, so protocol choice should match the executive deficit of interest.

   
   

5. Neurofeedback works best as part of an integrated treatment plan rather than as a standalone therapy.

Glossary

allocation concealment: a procedure in randomized trials that hides upcoming group assignments until a participant is enrolled, reducing selection bias.

Behavior Rating Inventory of Executive Function (BRIEF): a standardized questionnaire that rates everyday executive functioning, where lower scores indicate better function.

cognitive flexibility: an executive function involving the ability to shift between mental sets, tasks, or strategies.

Continuous Performance Task (CPT): a computerized task used to assess sustained attention and inhibitory control.

contingent negative variation (CNV): a slow event-related brain potential associated with anticipation, preparation, and attentional regulation.

effect size: a statistical estimate of the magnitude of a treatment effect, independent of sample size.

electroencephalography (EEG): a technique that records scalp electrical activity of the brain, used clinically to evaluate cortical function.

event-related potentials (ERP): time-locked electrical brain responses tied to specific sensory, cognitive, or motor events.

executive function: a set of higher-order cognitive processes supporting planning, inhibition, working memory, and goal-directed behavior.

functional magnetic resonance imaging (fMRI): a neuroimaging method that measures blood-oxygen-level changes associated with neural activity.

Go/No-go task: a paradigm that measures response inhibition by requiring participants to execute or withhold responses.

heterogeneity: variability across studies in a meta-analysis arising from methodological, clinical, or statistical differences.

inhibitory control: an executive function involving the suppression of impulsive or inappropriate responses.

meta-analysis: a statistical procedure that combines results from multiple independent studies to estimate an overall effect.

neurofeedback training (NFT): a form of biofeedback in which a person learns to regulate neural activity using real-time feedback from brain signals.

neuroplasticity: the capacity of the nervous system to reorganize its structure or function in response to experience or training.

operant conditioning: a learning process in which behaviors are shaped through reinforcement or feedback.

PEDro scale: a validated instrument for rating the methodological quality of clinical intervention trials.

PRISMA: a standardized framework for transparent reporting of systematic reviews and meta-analyses.

randomized controlled trial (RCT): an experimental design in which participants are randomly assigned to intervention or control conditions.

sensory motor rhythm (SMR): an EEG frequency band associated with sensorimotor regulation and attentional stability.

slow cortical potentials (SCP): slow shifts in cortical electrical activity linked to the regulation of cortical excitability and attention.

standardized mean difference (SMD): a meta-analytic metric that compares effects across studies using different measurement scales.

theta beta ratio (TBR): an EEG-derived metric comparing slower theta activity to faster beta activity, often elevated in ADHD.

working memory: an executive function responsible for briefly holding and manipulating information during cognitive tasks.

References

Diamond, A. (2013). Executive functions. Annual Review of Psychology, 64, 135–168. https://doi.org/10.1146/annurev-psych-113011-143750

Louthrenoo, O., Boonchooduang, N., Likhitweerawong, N., Charoenkwan, K., & Srisurapanont, M. (2022). The effects of neurofeedback on executive functioning in children with ADHD: A meta-analysis. Journal of Attention Disorders, 26(7), 976–984. https://doi.org/10.1177/10870547211045738

Polanczyk, G., de Lima, M. S., Horta, B. L., Biederman, J., & Rohde, L. A. (2007). The worldwide prevalence of ADHD: A systematic review and metaregression analysis. American Journal of Psychiatry, 164(6), 942–948. https://doi.org/10.1176/ajp.2007.164.6.942

Van Doren, J., Arns, M., Heinrich, H., Vollebregt, M. A., Strehl, U., & Loo, S. K. (2019). Sustained effects of neurofeedback in ADHD: A systematic review and meta-analysis. European Child & Adolescent Psychiatry, 28(3), 293–305. https://doi.org/10.1007/s00787-018-1121-4

Zhong, X., Yuan, X., Dai, Y., Zhang, X., & Jiang, C. (2025). Neurofeedback training for executive function in ADHD children: A systematic review and meta-analysis. Scientific Reports, 15, 28148. https://doi.org/10.1038/s41598-025-94242-4

About the Author

Dr. John "Dusty" Raymond Davis is an adjunct lecturer in the Department of Psychiatry and Behavioural Neurosciences at McMaster University's Faculty of Health Sciences. His scholarly contributions include research on EEG changes in major depression and case studies on neurological conditions. ​

Support Our Friends

Neurofeedback Strengthens Executive Function in Children with ADHD.pdf

Download PDF • 7.87MB