Neurofeedback offers minimal improvements for ADHD symptoms

A systematic review and meta-analysis published in JAMA Psychiatry has found no evidence that neurofeedback meaningfully improves ADHD symptoms at the group level. However, small but statistically significant effects emerged when standard neurofeedback protocols targeted processing speed, a neuropsychological measure. Despite these limited findings, neurofeedback appears to offer minimal benefit for clinical or cognitive symptoms of ADHD overall.

ADHD is a neurodevelopmental disorder characterized by persistent inattention, hyperactivity, and impulsivity, which can significantly impair daily functioning. While medications, particularly stimulants, are effective in reducing symptoms, not all individuals respond to or tolerate pharmacological treatments. Nonpharmacological interventions like neurofeedback have been proposed as alternatives. Neurofeedback is a technique that uses a brain-computer interface to train individuals to regulate their brain activity in real time, often through electroencephalogram (EEG) feedback. Proponents argue that by learning to normalize ADHD-related brain patterns, individuals can alleviate their symptoms and improve cognitive performance.

Although neurofeedback has been studied for decades, its clinical efficacy remains controversial. Many past studies were limited by small sample sizes, inadequate blinding, or suboptimal protocols. Recent advances, including the use of standardized neurofeedback methods and neuroimaging technologies, provided an opportunity to reassess its potential. The new meta-analysis sought to determine whether neurofeedback offers meaningful benefits for ADHD symptoms and related cognitive impairments when evaluated through rigorous, blinded research.

“Neurofeedback has been proposed as a treatment for ADHD and its widely used. It has a lot of advocates. However, the evidence as to effectiveness was mixed,” said study author Edmund J.S. Sonuga-Barke, a professor of developmental psychology, psychiatry, and neuroscience at King’s College London.

A meta-analysis is a research method that combines the results of multiple studies on a particular topic to draw a more comprehensive conclusion. By statistically analyzing data from different studies, a meta-analysis identifies patterns, evaluates the overall effectiveness of an intervention, and provides a more precise estimate of outcomes than individual studies alone.

In the current study, the meta-analysis included 38 randomized clinical trials published up to July 2023. These trials included 2,472 children, adolescents, and adults diagnosed with ADHD. Participants underwent neurofeedback interventions that primarily targeted EEG activity, with a smaller subset involving newer techniques such as functional magnetic resonance imaging (fMRI) or near-infrared spectroscopy (fNIRS).

The meta-analysis focused on outcomes measured after participants completed neurofeedback training. ADHD symptoms were assessed using reports from individuals close to the intervention setting (e.g., parents, teachers) and probably blinded sources, such as independent clinicians.

“Probably blinded” refers to situations in research where the person reporting or assessing outcomes is likely unaware of which treatment or intervention participants received, but complete certainty about their blinding cannot be guaranteed. Probably blinded contrasts with fully blinded (where blinding is explicitly confirmed) and unblinded (where the assessor clearly knows the treatment).

The researchers also examined neuropsychological outcomes, including measures of attention, working memory, and processing speed. Sensitivity analyses explored whether results differed when standard neurofeedback protocols were used or when evidence of learning to regulate brain activity was demonstrated.

The results showed that neurofeedback did not significantly improve core ADHD symptoms, such as inattention or hyperactivity, when evaluated using probably blinded reports. This finding held true across nearly all sensitivity analyses. Small but statistically significant effects were observed in studies that adhered to well-established standard neurofeedback protocols. Specifically, these protocols appeared to produce reductions in total ADHD symptom severity, but the effects were not large enough to be clinically meaningful.

“Concerning the substantial number of well designed trials, where raters are blinded to intervention allocation – there was no evidence that the use of neurofeedback led to clinically meaningful reductions in ADHD symptoms at the level of the whole group,” Sonuga-Barke told PsyPost. “There was also almost no impact on neuropsychological processes, which might be considered a more proximal target for treatment.”

Neurofeedback only showed promising results in one specific area: processing speed. Participants in neurofeedback trials demonstrated small but significant improvements in this cognitive domain, suggesting that the intervention may enhance certain aspects of neuropsychological functioning. Importantly, these gains in processing speed were also observed at longer-term follow-up assessments, indicating some durability of the effects. However, neurofeedback did not lead to significant improvements in other cognitive outcomes, such as attention or working memory.

When compared to other treatments, neurofeedback was less effective than stimulant medication in reducing ADHD symptoms, based on outcomes reported by individuals close to the intervention. Additionally, neurofeedback did not perform better than other nonpharmacological interventions, such as cognitive training or physical exercise.

“This finding was in line with our previous work so we were not especially surprised, but rather disappointed that recent trials of neurofeedback have not shown better outcomes,” Sonuga-Barke said. “We desperately need effective non-pharma treatments.”

The study, “Neurofeedback for Attention-Deficit/Hyperactivity Disorder: A Systematic Review and Meta-Analysis,” was conducted by Samuel J. Westwood, Pascal-M. Aggensteiner, Anna Kaiser, Peter Nagy, Federica Donno, Dóra Merkl, Carla Balia, Allison Goujon, Elisa Bousquet, Agata Maria Capodiferro, Laura Derks, Diane Purper-Ouakil, Sara Carucci, Martin Holtmann, Daniel Brandeis, Samuele Cortese, and Edmund J. S. Sonuga-Barke, on behalf of the European ADHD Guidelines Group (EAGG).