Advancing
Neurofeedback
Through Science

About the CURATOR Project

Outcomes of neurofeedback sessions vary widely because protocols are rarely tailored to the unique neural dynamics of each individual. As a result, it is difficult and time-consuming to provide personalized treatment recommendations for everyone. Brain activity patterns, personal goals (e.g., anxiety reduction, sleep improvement), age, lifestyle, and comorbidities all influence treatment response, yet current approaches lack objective tools to guide personalization. CURATOR will develop objective, computational methods to optimize the design of neurofeedback treatment. We will (1) systematically test feedback modalities (e.g., visual, auditory, interactive) to identify stimulus types that best engage and benefit individuals, and (2) derive computational biomarkers from EEG spectral features, event-related potentials, and brain connectivity metrics that can predict suitable treatment responses and inform protocol selection. State-of-the-art Machine Learning algorithms and Large Language Models will be applied to integrate multimodal feedback signals in order to enable precise, data-driven delivery of personalized reports and treatment recommendations. In sum, this project enables closed-loop neurofeedback systems that adjust protocols in real time based on cognitive and neurophysiological data. Among other benefits, personalized reporting will enable clinicians to track progress, predict outcomes, and detect relapse risks early. By relying on objective measures rather than trial and error, CURATOR will advance the current state of the art in neurofeedback technology.

In Plain Language

Neurofeedback is a type of brain training that teaches people how to change their own brain activity using real‑time feedback from their neurophysiological signals. The goal is to improve mental health, thinking, sleep, and overall well‑being. However, current neurofeedback programs are often "one‑size‑fits‑all" even though no two brains are alike. As a result, some people benefit greatly, while others see little effect. Factors such as age, lifestyle, personal goals, and mental health challenges play a role in how well someone responds to the treatment selected by the clinician. The CURATOR project will make neurofeedback more personal and effective. We will test different types of feedback (such as images, music, videos, or games) to discover which approach works best for each individual. At the same time, we will search for brain activity patterns, known as biomarkers, that can predict how someone will respond to training. These biomarkers will be derived from detailed EEG signal analyses and connectivity measures, then processed using advanced Machine Learning algorithms and Large Language Models to provide personalized outcomes. CURATOR will make it possible to tailor the treatment programs to each person's unique brain and to monitor their progress automatically. Our ultimate goal is to create next‑generation neurofeedback tools that adapt in real time, helping people learn to regulate their own brain activity more effectively and giving clinicians powerful new ways to guide treatment.