Neurofeedback helped children with ADHD, and so did a convincing fake

The pitch often starts with a picture. A clinician fits a cap of sensors on your child’s head, records a few minutes of brain activity and shows you a colorful map with one area flagged: too much slow-wave activity, not enough fast. The plan that follows runs to dozens of sessions in which your child plays a game that responds to their brainwaves. The ADHD trials described below used 30 to 38.

Families who go through it often see real change; in the trials below, most children and adults improved. A carefully blinded, federally supported American trial saw that too, and then found something harder to explain away. Children whose sessions were secretly driven by another child’s recorded brainwaves, and who never got genuine feedback at all, improved just as much.

This article is about the method. Diagnosis and medication for ADHD itself sit with the library’s article on executive function.

What happens in a neurofeedback session

Neurofeedback is a type of biofeedback, which feeds people a real-time readout of some bodily signal and lets them practice shifting it. Here the signal comes from an electroencephalogram, or EEG, which picks up the brain’s electrical rhythms through scalp sensors. Software turns those rhythms into something a person can follow, such as a game character that moves only when the targeted pattern appears. The logic is operant conditioning: behavior that gets rewarded tends to recur.

For ADHD, the usual target is the theta/beta ratio. In the definitions used by the largest US trial, theta waves run at 4 to 8 cycles per second and the beta band studied runs from 13 to 21. The investigators’ registry entry gives the rationale: many people with ADHD show excess theta power, low beta power and a high ratio between the two, and conditioning might correct that imbalance. Training aims to push the ratio down.

Physical risk appears low. In the children’s trial below, adverse events were spread evenly between real and sham groups, and the adult trial reported no serious ones. The costs that matter are time, money and whatever a family sets aside while the sessions run.

A trial built around a convincing fake

The real question is whether the brainwave feedback causes the improvement, as opposed to everything that comes with it. Answering that takes a control group with the same sessions, equipment and attention, and with the feedback disconnected from the child’s own brain.

The International Collaborative ADHD Neurofeedback trial, known as ICAN, built exactly that. Sponsored by Ohio State University with support from the National Institute of Mental Health, it ran at Ohio State’s Nisonger Center in Columbus and at the University of North Carolina Asheville. Its outcome papers appear in the Journal of the American Academy of Child and Adolescent Psychiatry under the byline of the Neurofeedback Collaborative Group.

The team enrolled 144 children aged 7 to 10 with carefully diagnosed, moderate to severe ADHD and a high theta/beta ratio, and randomly assigned them in a 3-to-2 ratio. The neurofeedback group trained to lower that ratio across 38 sessions. Control sessions looked identical, but their rewards came from a prerecorded EEG of another child, with the participant’s own artifacts, the electrical blips from blinking and moving, layered on top to make the fake harder to spot. An off-site co-investigator assigned the versions over the internet, so children, parents and site staff were all kept unaware.

What the results showed

The disguise held: at 25 months, only about a quarter could correctly identify the control treatment.

On the main measure, inattention rated by parents and teachers, both groups improved by an effect size of 1.5. That is measured in standard deviations, the typical spread of scores, and is a big shift by any yardstick. The gap between the groups was an effect size of 0.01 at the end of treatment and 0.23, not statistically significant, at 13 months. About 61% of the neurofeedback group and 54% of controls were rated as responders, a difference chance could easily explain.

The 25-month follow-up, published online in December 2022, told the same story. Parent-rated inattention did not differ between groups, though both had improved substantially. By then the controls had the higher response rate, 66% against 58.6%, again not a significant difference.

One result is easy to read as a point in neurofeedback’s favor. At 13 months, the neurofeedback group needed significantly less ADHD medication. That was a secondary finding, and it weakened: at 25 months, 38.6% of the neurofeedback group and 57.1% of controls needed medication, a gap that fell short of statistical significance.

The investigators concluded that most of the large improvement earlier unblinded studies had credited to neurofeedback reflects nonspecific effects, not the training of the brainwave ratio. A 2026 analysis of training data from 100 of the trial’s children, by University of Helsinki researchers working with colleagues from both trial sites, added a twist. Most participants did learn to regulate their theta/beta ratio, but how well they learned did not reliably predict how much their symptoms improved. Engagement and training context predicted outcomes better.

Where the improvement came from

None of this makes the children’s progress imaginary. Teachers and parents saw genuine change in both groups. The investigators’ registry entry describes the rival explanation the trial was designed to test: the plain benefit of sitting down several times a week for a structured task with a warm adult who praises a child for paying attention.

So the familiar reasoning, my child got better and therefore the brainwave training worked, skips a step. The improvement can be entirely real while the brainwave component adds little.

What pooled trials show, and why the rater matters

A 2016 meta-analysis led by Samuele Cortese, then at the University of Southampton, combined 13 trials with 520 children and adolescents. Neurofeedback looked effective when symptoms were scored by the raters closest to the treatment, the least blinded measure, with an effect size of 0.35. With ratings from people probably unaware of who got what, or against an active or sham comparison, the effect was no longer significant.

A larger 2025 update in JAMA Psychiatry, led by Samuel J. Westwood of King’s College London, pooled 38 randomized trials with 2,472 participants aged 5 to 40. On probably blinded ratings of total ADHD symptoms, the pooled effect was 0.04, indistinguishable from zero. Trials using established standard protocols showed a small, significant effect of 0.21, and of five lab measures, only processing speed improved. The King’s College team concluded: “Overall, neurofeedback did not appear to meaningfully benefit individuals with ADHD, clinically or neuropsychologically, at the group level.”

A 2022 meta-analysis by researchers in Taiwan found the same pattern on attention tests. Across 14 trials, the benefit for sustained attention was 0.32, but in the two trials that kept participants unaware of their assignment, the overall benefit on attention tests fell to 0.05.

Adults fare no differently. A triple-blind trial at Germany’s University of Tübingen, published in Lancet Psychiatry in 2017, assigned 118 adults with ADHD to 30 neurofeedback sessions, to a sham condition (15 sham sessions followed by 15 real ones) or to 12 sessions of metacognitive group therapy. Self-reported symptoms fell substantially in all three groups through six months of follow-up, with no significant difference among them.

Neurofeedback has scientific defenders. A 2019 review in Current Psychiatry Reports argued that three standard protocols, theta/beta training among them, are effective and specific. It predates the 25-month results and the 2025 meta-analysis, and one of its authors lists a company affiliation.

On the library’s five-step evidence scale, the claim that the brainwave training itself reduces ADHD symptoms is tested and not supported. The small standard-protocol signal is, in our reading, measured but below the threshold that matters, since the review’s own authors found no meaningful group-level benefit.

Where pediatric and federal health bodies stand

The American Academy of Pediatrics’ 2019 ADHD guideline for children and adolescents, still the version linked from the academy’s pediatrician resource page as updated in January 2026, states: “Some nonmedication treatments for ADHD-related problems have either too little evidence to recommend them or have been found to have little or no benefit.” Its next sentence puts EEG biofeedback on that list. The passage sits in the guideline’s discussion, not in a graded recommendation, so it describes the evidence rather than formally recommending against. The same document describes the evidence behind established medication and behavioral treatments as strong.

The National Institute of Mental Health draws a similar line in its public summary, last reviewed in August 2026. Medication and psychosocial care such as parent training and school-based support are the standard options. Neurofeedback appears among newer approaches, alongside cognitive training, mindfulness and neuromodulation.

Insomnia, PTSD and other uses

Outside ADHD, the controlled evidence thins quickly.

Insomnia. In a double-blind study published in Brain in 2017, researchers at the University of Salzburg gave 25 people with primary insomnia 12 sessions of real neurofeedback and 12 of placebo feedback, with nine nights of lab sleep recordings. The protocol, which the paper calls practitioners’ gold standard for overarousal, aimed to boost a rhythm over the brain’s sensorimotor strip. Real and placebo feedback improved sleep complaints equally, and objective sleep measures did not change. The authors credited unspecific factors such as trust, care and empathy from the staff, and said neurofeedback cannot be recommended in place of cognitive behavioral therapy for insomnia, the standard non-drug treatment, which the library’s article on sleep and dementia explains. Verdict: tested and not supported, on one small sham-controlled trial.

PTSD. A 2025 meta-analysis by researchers at the Defense Health Agency’s Psychological Health Center of Excellence and the Walter Reed Army Institute of Research found moderate to large benefits for EEG neurofeedback in five trials that compared it with passive conditions such as wait lists or usual care. Neurofeedback looked better against passive comparisons than active ones, and two sham-controlled trials of neurofeedback guided by functional MRI scans showed no improvement. The authors rated their confidence as very low to low.

A Military Health System evidence summary dated August 2025 reports that the 2023 VA/DoD PTSD guideline judged the evidence too thin to recommend neurofeedback either way, and says clinicians may consider it for patients who haven’t responded to, or don’t want, the treatments the guideline recommends. Verdict: insufficient, because the trials that exist are poorly controlled.

Everything else. Insurer policy is the clearest available signal. Aetna’s biofeedback policy, last reviewed in July 2026, classes biofeedback as unproven for ADHD, anxiety disorders, autism, depression, insomnia, PTSD and substance use disorders, among others, and calls home biofeedback unproven for any use.

Brain maps, devices and what FDA has reviewed

The map

Brain maps come from quantitative EEG, or qEEG, in which software splits the signal into frequency bands and compares them with reference norms. Two problems follow.

First, the numbers depend on the software. ICAN investigators found that the theta/beta ratio from the program used to screen children differed from the one computed by the training program, because the packages calculate band power differently. They advised practitioners to know their system’s algorithm and use comparison norms collected on that same system.

Second, the claims outrun the evidence. Aetna’s qEEG policy, reviewed in April 2026, calls it unproven for ADHD, depression, anxiety, concussion, learning disability, autism, PTSD and insomnia, covering it only as an add-on to standard EEG in narrow situations such as epilepsy and intensive-care monitoring. A 1997 assessment by the American Academy of Neurology and the American Clinical Neurophysiology Society, since retired, had already called qEEG investigational for attention disorders, learning disability, depression and post-concussion syndrome.

The ADHD test FDA authorized is not a treatment

In July 2013, FDA authorized an EEG-based aid for assessing ADHD in patients aged 6 to 17. It uses the theta/beta ratio from a single electrode, combined with a clinician’s evaluation, and FDA’s review says it is not to be used on its own to evaluate or diagnose ADHD. FDA flagged misdiagnosis as a key risk: a false positive could start ADHD medication that isn’t needed, and a false negative could delay treatment. The device measures. It trains nothing, and its authorization says nothing about whether neurofeedback works.

Hardware and wellness gadgets

Under federal regulation, biofeedback instruments, including those that read brain waves, are Class II devices, and a prescription, battery-powered unit meant for relaxation training and muscle re-education is exempt from premarket notification. The library’s piece on relaxation methods explains what such a clearance does and doesn’t establish.

Consumer gadgets sit under a looser policy. FDA’s general wellness guidance, in its January 2026 edition, counts promises of relaxation, stress relief or better concentration as low-risk wellness claims, and the agency says it won’t check whether low-risk products limited to such claims are devices at all. The shelter ends where disease claims begin; the guidance’s examples include help for an anxiety disorder. A brainwave headband advertised only for calm or focus could fit the wellness category, which says nothing about whether it works.

Credentials, coverage and cost

The best-known credential comes from the Biofeedback Certification International Alliance (BCIA), a private certifying body. Its entry-level neurofeedback certification requires a bachelor’s degree in a health discipline BCIA accepts, a neuroanatomy or neurophysiology course, a 36-hour accredited training program, mentoring that spans 25 contact hours and 100 client sessions, and a supervised online exam with roughly 100 multiple-choice items.

The certificate comes in two versions: one for people licensed to treat diagnosed medical or psychological disorders, and one for unlicensed practitioners in peak-performance settings, where, by BCIA’s description, there are no patients. Unlicensed holders working with anyone who has a diagnosed disorder must be under legal supervision. BCIA’s home page adds the deciding point: its designations don’t change which services a person can legally provide, because each state’s practice act determines that.

Coverage is thin. Medicare’s national biofeedback rule, listed as effective since 1966, pays only to retrain specific muscles or treat spasticity, disabling spasm or weakness after conventional treatments fail, and doesn’t mention neurofeedback. Aetna notes that some of its plans exclude biofeedback entirely. Authoritative price data are scarce; the ICAN investigators wrote in their registry entry that one neurofeedback treatment can cost as much as a month of medication.

Questions to ask a neurofeedback clinic

These questions are our own, built from the evidence above.

1. What will be trained, and what should change? A named protocol and an outcome you can observe, such as teacher reports or a sleep diary, beat a promise to balance the brain.
2. What do sham-controlled trials show for this exact problem? Testimonials and before-and-after scores can’t separate brainwave training from the structure and encouragement that surround it.
3. Which state license does the person running sessions hold? Certification and licensure are different things.
4. Where did the brain map’s numbers come from? Ask which software produced them and whether its norms were collected on the same system.
5. What is the full price, and what happens if nothing changes by the midpoint? Get the package cost and your insurer’s written policy before the first session.
6. How will this fit with current care? A clinic that discourages existing treatment, or won’t share progress with your child’s doctor, is a warning sign.

When to involve a clinician

• Attention or behavior problems in a child who hasn’t been evaluated. Whether it’s ADHD, and what to try first, is a question for a pediatrician or child mental health specialist, not for a clinic selling one method.
• Thinking of pausing ADHD medication during a neurofeedback course. That choice belongs with the prescriber, since the blinded trials don’t show neurofeedback can stand in for it.
• Insomnia that has lasted for months. A clinician can look for underlying causes and discuss a referral for cognitive behavioral therapy for insomnia.
• PTSD symptoms. The first conversation is about the treatments the VA/DoD guideline recommends; the Military Health System summary frames neurofeedback as an option clinicians might weigh for patients those treatments haven’t helped, or who don’t want them.
• Suicidal thoughts, or a sense that nothing will get better. Calling or texting 988, or chatting at 988lifeline.org, connects to the 988 Suicide & Crisis Lifeline at no cost, confidentially, at any hour. For immediate danger, call 911.

The takeaway

In blinded trials, children and adults improved during neurofeedback, and people whose feedback was faked improved by about the same amount. The improvement is real; the brainwave training is an unproven explanation for it. Before paying for a package, the most useful question to put to any clinic is what sham-controlled trials show for your specific problem, and which state license the person delivering the sessions holds.

Sources

1. National Institute of Mental Health. “Attention-Deficit/Hyperactivity Disorder (ADHD) (last reviewed August 2026).” nimh.nih.gov
2. American Academy of Pediatrics. “Attention-Deficit/Hyperactivity Disorder: Resources for Pediatricians (last updated January 14, 2026).” aap.org
3. ClinicalTrials.gov, National Library of Medicine. “NCT02251743: Double-Blind 2-Site Randomized Clinical Trial of Neurofeedback for ADHD.” clinicaltrials.gov
4. U.S. Food and Drug Administration, Center for Devices and Radiological Health. “De Novo decision summary K112711: neuropsychiatric interpretive EEG assessment aid for ADHD.” accessdata.fda.gov
5. U.S. Food and Drug Administration. “Letter granting De Novo request K112711 (July 15, 2013).” accessdata.fda.gov
6. U.S. Food and Drug Administration. “General Wellness: Policy for Low Risk Devices (guidance, January 6, 2026).” fda.gov
7. Legal Information Institute, Cornell Law School. “21 CFR 882.5050, Biofeedback device.” law.cornell.edu
8. Centers for Medicare & Medicaid Services. “National Coverage Determination 30.1, Biofeedback Therapy.” cms.gov
9. Military Health System (health.mil). “Neurofeedback for Posttraumatic Stress Disorder (evidence summary dated August 2025, posted January 9, 2026).” health.mil
10. Aetna. “Quantitative EEG (Brain Mapping), Clinical Policy Bulletin 0221 (last reviewed April 14, 2026).” aetna.com
11. Aetna. “Biofeedback, Clinical Policy Bulletin 0132 (last reviewed July 16, 2026).” aetna.com
12. Biofeedback Certification International Alliance. “Neurofeedback Entry Level.” bcia.org
13. Biofeedback Certification International Alliance. “BCIA home page (certification designations and state practice acts).” bcia.org
14. 988 Suicide & Crisis Lifeline. “988 Suicide & Crisis Lifeline.” 988lifeline.org
15. Pediatrics. “Wolraich ML, Hagan JF, Allan C, et al. "Clinical Practice Guideline for the Diagnosis, Evaluation, and Treatment of Attention-Deficit/Hyperactivity Disorder in Children and Adolescents." Pediatrics, 2019. PubMed. DOI: 10.1542/peds.2019-2528.” pubmed.ncbi.nlm.nih.gov
16. Journal of the American Academy of Child and Adolescent Psychiatry. “Neurofeedback Collaborative Group. "Double-Blind Placebo-Controlled Randomized Clinical Trial of Neurofeedback for Attention-Deficit/Hyperactivity Disorder With 13-Month Follow-up." J Am Acad Child Adolesc Psychiatry, 2021 (online 2020). PubMed. DOI: 10.1016/j.jaac.2020.07.906.” pubmed.ncbi.nlm.nih.gov
17. Journal of the American Academy of Child and Adolescent Psychiatry. “Neurofeedback Collaborative Group. "Neurofeedback for Attention-Deficit/Hyperactivity Disorder: 25-Month Follow-up of Double-Blind Randomized Controlled Trial." J Am Acad Child Adolesc Psychiatry, 2023 (online 2022). PubMed. DOI: 10.1016/j.jaac.2022.07.862.” pubmed.ncbi.nlm.nih.gov
18. Applied Psychophysiology and Biofeedback. “Kuznetsova E, de Beus R, Arnold LE, Rudakov E, Cowley BU. "Learning in Neurofeedback is Heterogenous and Does Not Guarantee ADHD Symptom Improvement." Appl Psychophysiol Biofeedback, 2026. PubMed. DOI: 10.1007/s10484-026-09774-w.” pubmed.ncbi.nlm.nih.gov
19. Clinical EEG and Neuroscience. “Kerson C, deBeus R, Lightstone H, Arnold LE, Barterian J, Pan X, Monastra VJ. "EEG Theta/Beta Ratio Calculations Differ Between Various EEG Neurofeedback and Assessment Software Packages: Clinical Interpretation." Clin EEG Neurosci, 2020 (online 2019). PubMed. DOI: 10.1177/1550059419888320.” pubmed.ncbi.nlm.nih.gov
20. Journal of the American Academy of Child and Adolescent Psychiatry. “Cortese S, Ferrin M, Brandeis D, et al. "Neurofeedback for Attention-Deficit/Hyperactivity Disorder: Meta-Analysis of Clinical and Neuropsychological Outcomes From Randomized Controlled Trials." J Am Acad Child Adolesc Psychiatry, 2016. PubMed. DOI: 10.1016/j.jaac.2016.03.007.” pubmed.ncbi.nlm.nih.gov
21. JAMA Psychiatry. “Westwood SJ, Aggensteiner PM, Kaiser A, et al. "Neurofeedback for Attention-Deficit/Hyperactivity Disorder: A Systematic Review and Meta-Analysis." JAMA Psychiatry, 2025 (online 2024). PubMed. DOI: 10.1001/jamapsychiatry.2024.3702.” pubmed.ncbi.nlm.nih.gov
22. Child and Adolescent Psychiatry and Mental Health. “Chiu HJ, Sun CK, Fan HY, et al. "Surface electroencephalographic neurofeedback improves sustained attention in ADHD: a meta-analysis of randomized controlled trials." Child Adolesc Psychiatry Ment Health, 2022. PubMed. DOI: 10.1186/s13034-022-00543-1.” pubmed.ncbi.nlm.nih.gov
23. The Lancet Psychiatry. “Schönenberg M, Wiedemann E, Schneidt A, et al. "Neurofeedback, sham neurofeedback, and cognitive-behavioural group therapy in adults with attention-deficit hyperactivity disorder: a triple-blind, randomised, controlled trial." Lancet Psychiatry, 2017. PubMed. DOI: 10.1016/S2215-0366(17)30291-2.” pubmed.ncbi.nlm.nih.gov
24. Current Psychiatry Reports. “Enriquez-Geppert S, Smit D, Pimenta MG, Arns M. "Neurofeedback as a Treatment Intervention in ADHD: Current Evidence and Practice." Curr Psychiatry Rep, 2019. PubMed. DOI: 10.1007/s11920-019-1021-4.” pubmed.ncbi.nlm.nih.gov
25. Brain. “Schabus M, Griessenberger H, Gnjezda MT, Heib DPJ, Wislowska M, Hoedlmoser K. "Better than sham? A double-blind placebo-controlled neurofeedback study in primary insomnia." Brain, 2017. PubMed. DOI: 10.1093/brain/awx011.” pubmed.ncbi.nlm.nih.gov
26. Frontiers in Neuroscience. “Berman DE, Cowansage KP, Bellanti DM, et al. "Systematic review and meta-analysis of neurofeedback training efficacy and neural mechanisms in the treatment of posttraumatic stress disorder." Front Neurosci, 2025. PubMed. DOI: 10.3389/fnins.2025.1658652.” pubmed.ncbi.nlm.nih.gov
27. Neurology. “Nuwer M. "Assessment of digital EEG, quantitative EEG, and EEG brain mapping: report of the American Academy of Neurology and the American Clinical Neurophysiology Society [RETIRED]." Neurology, 1997. PubMed. DOI: 10.1212/wnl.49.1.277.” pubmed.ncbi.nlm.nih.gov