People with Parkinson’s disease often have difficulty controlling leg movements, such as starting to walk or keeping a steady rhythm. A new study by Lena Salzmann from ETH Zurich, Oliver Bichsel and colleagues at University Hospital Zurich and collaborating institutions shows that patients can learn – within a single session – to consciously reduce excessive brain activity using neurofeedback based on deep brain stimulation. This self-regulation was linked to measurable improvements in the quality of lower-limb movements, highlighting neurofeedback as a promising addition to current therapies.
Parkinson’s disease is associated with excessive beta oscillations – rhythmic brain activity in the 13–30 Hz range – in the subthalamic nucleus, a deep brain structure crucial for movement control. These oscillations are a key target of deep brain stimulation (DBS), but it remains unclear whether patients can voluntarily reduce them and whether this has measurable motor benefits. To address this question, the researchers used DBS‑based neurofeedback, allowing patients to see and actively downregulate their own beta activity in real time. Ten patients with implanted DBS systems completed a single session of DBS-based neurofeedback. Movement quality was assessed using inertial measurement units (IMUs) – wearable sensors that objectively capture speed, acceleration, and pauses of movement.
The results show that neurofeedback can translate neural self‑regulation into behavioral change. On average, patients reduced subthalamic beta power by about 12%. Importantly, stronger beta reductions were tightly linked to better leg performance: after neurofeedback, participants showed faster foot movements, higher peak velocities, greater acceleration, and fewer brief halts during a foot‑stomping task. No clear improvements were seen in comparable hand movements, likely reflecting factors such as task order, fatigue, and protocol design. Notably, patients often used different mental strategies to achieve beta reduction, suggesting that neurofeedback learning is individual and can rely on different cognitive approaches.
Together, these findings suggest that DBS‑based neurofeedback can induce short‑term improvements in movement quality by directly targeting pathological brain oscillations. More broadly, the study highlights the value of combining neurofeedback with objective, sensor‑based movement measures, which can reveal subtle changes that may be missed by traditional clinical ratings. While future work is needed to test longer training periods and lasting functional benefits, the results point toward neurofeedback as a promising, personalized complement to existing Parkinson’s therapies.
Reference: Salzmann L, Bichsel O, Rohr‑Fukuma M, Naef AC, Stieglitz L, Oertel MF, Bujan B, Jedrysiak P, Lambercy O, Imbach LL, Gassert R. Lower limb motor effects of DBS neurofeedback in Parkinson’s disease assessed through IMU‑based UPDRS movement quality metrics. Scientific Reports. 2025. https://doi.org/10.1038/s41598-025-28378-8
Main image: Stefan Schneller, RELab, ETHZ