sLORETA Neurofeedback for Cognitive Health

Researchers Viviane Dasilva, Diana Poli, and Olimpia Pino have proposed a highly personalized model for cognitive rehabilitation. Their theoretical framework, detailed in the International Journal of Environmental Research and Public Health, connects advanced computational neuroscience with clinical practice to create a potential new protocol for Mild Cognitive Impairment (MCI).

Mapping the Faulty Brain Network with qEEG and sLORETA

The model starts with a specific problem in MCI: dysfunction in the Default Mode Network (DMN). This brain network, active when we are at rest or daydreaming, is often disrupted in early cognitive decline. The researchers propose using quantitative electroencephalography (qEEG) combined with a source localization technique called sLORETA. This allows clinicians to move beyond general brainwave readings and pinpoint the exact dysfunctional nodes within the DMN, such as the precuneus and posterior cingulate cortex. Identifying these “putative pathological nodes” turns them into precise targets for intervention.

This approach to localization is similar in concept to advanced neurofeedback techniques explored for other conditions. For instance, targeted sLORETA neurofeedback is being investigated for its potential to retrain specific brain regions.

Isolating the Pure Brain Signal from the Noise

A central innovation in this framework is its method for calculating an individual’s alpha frequency. Alpha waves are a key biomarker for relaxed alertness. Traditional methods can be biased by the brain’s background electrical “noise,” known as the aperiodic 1/f component. The proposed model uses spectral decomposition to isolate and remove this noise. The result is a “pure” individual alpha frequency (IAF). This purified IAF is then used to recalibrate what the authors term “Weber’s Cognitive Threshold,” essentially setting a more accurate, personalized baseline for neurofeedback training. The goal is to make the initial brain state assessment as precise as possible.

A Bayesian Brain Trainer That Adapts in Real Time

The core of the rehabilitation protocol is a dynamic neurofeedback system. It employs Bayesian algorithms and stochastic modeling—mathematical tools for dealing with probability and uncertainty. This creates a “Dynamic Weight Change” mechanism. In practice, the system constantly adjusts the difficulty of the neurofeedback task based on the user’s performance.

Its objective is to maintain a specific 70% success rate. This rate is chosen as a strategic balance. It is high enough to be rewarding and promote the synaptic changes of Long-Term Potentiation via Hebbian learning, which relies on a positive Reward Prediction Error. Simultaneously, it is low enough to anticipate and mitigate neural fatigue, keeping the user engaged. The system is designed to be a responsive, adaptive brain trainer.

How to Measure If It Works: MoCA, Memory, and Brain Coherence

As a theoretical paper, this model requires empirical validation. The authors propose clear pathways for that testing. Clinical improvement would be measured using standard cognitive assessments, such as the Montreal Cognitive Assessment (MoCA) and the Rey Auditory Verbal Learning Test (RAVLT). Beyond behavioral tests, a successful intervention should show “normalization of cortical coherence in the Default Mode Network.” This means the targeted brain regions would begin to communicate in a more synchronized, healthy pattern again, observable through follow-up qEEG/sLORETA analysis.

Implications for Hearing and Sound Sensitivity Disorders

While focused on MCI, this computational framework underscores a broader principle in neurology: maladaptive brain patterns can be identified and potentially retrained through precise, measurement-guided plasticity. This principle directly resonates with research on tinnitus and sound sensitivity conditions.

Tinnitus is increasingly understood as a network disorder involving hyperactivity and altered connectivity in specific brain circuits, not just the ear. The concept of using neurofeedback to promote beneficial neuroplasticity is actively explored in tinnitus management strategies. Furthermore, conditions like hyperacusis and misophonia involve abnormal limbic and autonomic responses to sound, suggesting dysfunctional network activity. Research comparing misophonia vs hyperacusis seeks to identify their distinct neural signatures—a step analogous to identifying pathological nodes in the DMN for MCI.

The model by Dasilva and colleagues exemplifies a move toward highly individualized, biomarker-driven neurorehabilitation. It maps individual biology onto a mathematical model to guide treatment. This approach, if validated, could inform not only cognitive rehabilitation but also future precision therapies for auditory and sensory processing disorders where maladaptive brain plasticity is a core feature.

Source: Dasilva V, Poli D, Pino O. A High-Precision Theoretical Computational Neurorehabilitation Framework for Mild Cognitive Impairment. Int J Environ Res Public Health. 2026;23(5):624. doi:10.3390/ijerph23050624.