Attention deficit hyperactivity disorder (ADHD) affects millions of children and can lead to challenges in learning, behavior, and daily life. As treatment options grow, researchers are exploring how brainwave data—measured through electroencephalography (EEG)—might help predict which children respond best to specific interventions like neurofeedback.
A recent study examined this idea using data from the International Collaborative ADHD Neurofeedback (ICAN) trial. This large, multi-site trial focused on how theta/beta ratio neurofeedback (TBR-NFB) affects symptoms of inattention in children aged 7 to 10. Using machine learning, the research team explored how EEG connectivity and neurofeedback in ADHD could work together to support treatment decisions.
What Is Neurofeedback Therapy?
Neurofeedback is a non-invasive therapy that helps individuals learn to regulate their brain activity. In ADHD, it often involves training children to shift their brainwave patterns—particularly by reducing theta waves (linked to daydreaming) and increasing beta waves (associated with focus and attention).
The ICAN trial used TBR neurofeedback to help children with ADHD improve focus. But not every child responds the same way. That’s where EEG connectivity came into play.
EEG Connectivity and Neurofeedback in ADHD: A Predictive Tool?
The researchers looked at EEG data recorded before treatment began. They used machine learning to analyze how different brain regions were communicating—what scientists call “connectivity.” Their goal was to see whether these patterns could predict which kids would improve most on teacher and parent rating scales.
Here’s what they found:
- Children with more dysregulation in attention-related brain networks (especially the ventral and dorsal attention networks) were more likely to show improvement.
- Certain brainwave frequencies, like delta and high beta (hibeta), played an important role in predicting treatment response.
- These patterns were different between kids who received active neurofeedback and those in the control group.
Interestingly, these connectivity signals were able to predict improvement even when other variables like ADHD subtype or comorbidities (e.g., anxiety, slow cognitive tempo) were taken into account.
Why This Matters for ADHD Treatment
The findings support the idea that a personalized approach to ADHD treatment is possible. Instead of a one-size-fits-all model, clinicians could use a child’s baseline EEG data to guide therapy choices—potentially saving time, money, and frustration.
While the predictive power was modest, the study opens the door to integrating EEG and machine learning into everyday clinical practice. A future where treatment is tailored to a child’s unique brain patterns could lead to better outcomes for kids and their families.
The Future of EEG and Neurofeedback in Mental Health
This study contributes to a growing movement toward precision psychiatry—using technology and data to customize care. In ADHD and beyond, EEG-guided neurofeedback could become a valuable tool in the mental health toolkit.
As machine learning algorithms become more sophisticated and accessible, they may help identify the best candidates for treatments like TBR-NFB. With continued research, we might soon see a shift toward highly individualized treatment plans driven by real-time brain data.
References
- Kerson, C., Yazbeck, M., Shahsavaripoor, B., et al. (2025). EEG Connectivity as Predictor of ICAN ADHD Children’s Improvement After Completion of Theta Beta Ratio Neurofeedback: Machine Learning Analyses. Applied Psychophysiology and Biofeedback. https://doi.org/10.1007/s10484-025-09713-1
- Arns, M., Heinrich, H., & Strehl, U. (2014). Evaluation of neurofeedback in ADHD: The long and winding road. Biological Psychology, 95, 108–115. https://doi.org/10.1016/j.biopsycho.2013.11.013