Moderate evidence· Mixed findings
Across the wider neurofeedback-for-chronic-pain research base, which includes samples of people with headache, the reviewing researchers found neurofeedback has potential for reducing pain and improving related outcomes, but they stop short of endorsing any single protocol as the reliable choice.
Studied in Adults with chronic pain conditions including headache, fibromyalgia, and other pain presentations (24 pooled studies, individual samples of 7 to 71 participants)clinical studies
What complicates this
The review explicitly states current evidence 'does not provide definitive conclusions or allow for reliable recommendations on which protocols or methods of administration may be the most effective,' and the authors emphasise the need for higher-quality future research. The included samples are small and the population is chronic pain broadly, not migraine specifically.
Limited evidence· Mixed findings
Across 15 randomized trials in healthy adults, EEG neurofeedback training was associated with a small improvement in attentional performance (standardized mean difference 0.27), but within the eight trials that used a sham-neurofeedback comparison there was no significant pooled effect, and the review authors called for further sham-controlled trials.
Studied in Healthy adults; 41 randomized controlled trials in qualitative synthesis, 15 trials (569 participants) pooled, 8 of those sham-controlled.clinical studies
Limited evidence· Mixed findings
In a four-week randomized workplace trial of 92 employees, perceived stress fell significantly in all three arms, and adding neurofeedback to a mindfulness app did not outperform the mindfulness app alone on stress. The neurofeedback arm scored higher on resilience than the paper-based self-learning arm, and showed the largest gain in an EEG relaxation index at one-month follow-up.
Studied in Working adults in a corporate settingclinical studies
How strong the finding is
Three-arm randomized design with a one-month follow-up, but the observed stress reduction is not attributable to the neurofeedback component.
What complicates this
The resilience advantage reached significance only against the passive paper-based comparator, not against mindfulness training alone, so the added value of the neurofeedback layer on stress outcomes remains unestablished.
Limited evidence· Mixed findings
In a randomized double-blind trial of 38 adults, two weeks of twice-daily neurofeedback-assisted meditation with a wearable EEG headband produced a small but statistically significant advantage over a control device on perceived-stress scores, while stress hormones and quantitative EEG showed no difference between groups and the state-anxiety difference fell short of significance.
Studied in Adults using a consumer EEG headband for 12 minutes twice daily over two weeksclinical studies
How strong the finding is
Double-blind and device-controlled, which is unusual in consumer neurofeedback research, but very small (n=38) and short (two weeks), with the benefit confined to a self-reported measure.
What complicates this
Objective markers (stress hormones, qEEG) did not separate the groups, and subjective device satisfaction was significantly higher in the active arm, which leaves an expectation effect on the self-reported score open.
Limited evidence· Points toward a benefit
In an open clinical trial where participants chose their own study group rather than being assigned one, QEEG-guided neurofeedback training was followed by complete cessation of migraines in 54 percent of the neurofeedback group and a drop of more than 50 percent in migraine frequency for a further 39 percent, while the group that continued medication alone reported minimal change.
Studied in Adults aged 17 to 62 with recurrent migraine from a single neurological practice (n=71; 46 chose neurofeedback, 25 continued drug therapy)clinical studies
What complicates this
Not randomized: participants self-selected into the neurofeedback or medication-only group, so expectancy and selection effects cannot be ruled out. It is a single-practice study by a single author, and we found no blinded replication. The author's own summary language is considerably stronger than a non-randomized design can support.
Anecdotal only· Points toward a benefit
A published case report describes a single 25-year-old woman with migraine whose attack frequency and pain eased over 25 sessions combining sensorimotor neurofeedback with respiration and vasomotor biofeedback, and the authors themselves flagged the result as preliminary.
Studied in Single adult female migraine patient aged 25 (case report)expert opinion
What complicates this
A single uncontrolled case with no comparison group. The authors state that 'further study, have to be done with more patients and with placebo group' before the method can be considered scientifically supported. A case report cannot separate the intervention from natural fluctuation in migraine frequency.
