Frontal gamma as a marker of effective training during neurofeedback to improve memory in patients with mild cognitive impairment

Abstract

Currently, more than 23 million people in the United States are affected by early Alzheimer's disease (AD), mild cognitive impairment (MCI), or subjective cognitive decline (SCD). Despite widespread awareness of the high costs to individuals and communities, treatment options for those at risk of, or with, AD is limited. To help meet this urgent need for new treatments, we identified frontal gamma activity, a neural signature of optimal memory function, as a promising treatment target. More specifically, patients with AD and MCI exhibit deficient frontal gamma activity, which we can help restore using electroencephalographic (EEG) neurofeedback (NFB). In brief, our MATLAB/EEGLAB-based brain-computer interface (BCI) converts frontal gamma coherence into re-inforcement signals, enabling patients with MCI to directly modulate frontal gamma activity. Preliminary results from our current double-blind, placebo-controlled randomized clinical trial (RCT) demonstrate that, compared to patients receiving placebo-NFB, patients receiving active-NFB (gamma-NFB, 30 min, 2/week, 12 weeks) exhibit significantly-increased frontal gamma coherence during training. Furthermore, among active gamma-NFB patients, baseline/pre-NFB gamma power at F4 (but not at other electrodes) significantly correlates with slope of training-related increases in frontal gamma coherence. These results support a model where, in patients with MCI, frontal gamma EEG-NFB specifically engages and increases frontal gamma activity. Furthermore, baseline/pre-NFB gamma power at F4 may serve as a predictor of training efficacy, a promising step towards more efficient and personalized treatment protocols for improving memory in patients with MCI or related neuropsychiatric difficulties.