Objective:To explore effective schemes for improving cognitive function in patients with subjective cognitive decline (SCD) by EEG neurofeedback (NF) technique.Methods:Ten SCD patients recruited in Department of Neurology, Xuanwu Hospital, Capital Medical University from April 2019 to August 2019, were chosen in our study; all patients were subjected to neuro-regulation of mismatched negativity (MMN) for 5 times in 2 weeks by EEG NF technique. By using the midlines of frontal zone (Fz), central zone (Cz) and peak zone (Pz) as training electrodes, auditory tones (1000 Hz as standard stimulus, individual hearing discrimination threshold as deviation stimulus) were used to induce MMN characteristics of event-related potential (ERP) via Oddball paradigm. SCD patients received visual disc stimulation feedback with dynamic neural signal activity. ERP waveforms and standard MMN characteristics (amplitude and latency) before and after training were used to evaluate the changes in neural activity of these patients; nine N-Back working memory tasks (three types×three gradients) and hearing discrimination ability tests were used in these patients to evaluate the improvement of cognitive competence.Results:ERP analysis showed that the MMN waveforms in SCD patients after training were obviously different as compared with those before training; their standard stimulus and deviant stimulus in the point-by-point paired t-test analysis were significant different ( P<0.05). The MMN amplitudes at the Fz, Cz and Pz electrodes in these patients after training were significantly higher than those before training ( P<0.05). The accuracy of auditory tone 3-Back task in SCD patients was statistically significant before and after training ( P<0.05). Conclusion:NF based on MMN can significantly improve the MMN amplitude and accuracy of auditory working memory task in patients with SCD, which might provide a potentially effective cognitive intervention method for patients with early AD and the elderly with normal memory decline.

Neurofeedback may improve cognitive function and behaviour by regulating the cerebral neural activities. Neurofeedback works as a kind of therapy for patients with Parkinson's disease, primarily based on electroencephalogram signals and hemodynamic signals. It regulates abnormal neuronal rhythm oscillations in the cortical-spinal system by electroencephalogram neurofeedback and sensory motor rhythm as the main adjustment frequency band, and modulates activation in the cortical-basal ganglia-thalamic-cortical network of brain with the supplementary motor area as the target area via functional magnetic resonance imaging neurofeedback. Neurofeedback effectively improves the motor symptoms of Parkinson's patients, both dynamic and static. However, the number of samples is limited in these researches, and the assessments before and after neurofeedback training are not yet comprehensive. Neurofeedback technology might be a potential effective rehabilitation for patients with Parkinson's disease.