A single-centre cross-sectional study has revealed that dynamic alpha-band electroencephalography network balance differs significantly across clinically defined cognitive groups. In examining brain activity during resting states, researchers found that global network segregation–integration balance demonstrated an ordered association across normal cognition, subjective cognitive decline, and mild cognitive impairment.
The underlying clinical challenge addressed by the study is that both subjective cognitive decline and mild cognitive impairment represent heterogeneous conditions whose dynamic electrophysiological organization remains insufficiently understood. To explore these neurophysiological patterns, the investigators recruited a clinical cohort consisting of 88 participants, which included 14 individuals with normal cognition, 42 with subjective cognitive decline, and 32 with mild cognitive impairment.
Electrophysiological data were captured using eyes-closed resting electroencephalography, with functional network connectivity assessed through alpha-band weighted phase-lag connectivity.
Primary statistical modeling was conducted using age-adjusted HC3-robust regression to estimate ordered-category and pairwise diagnostic relationships. Lower global segregation–integration balance was significantly associated with the ordered diagnostic categories, yielding a regression coefficient of negative 0.418 with an HC3 95 percent confidence interval from negative 0.760 to negative 0.076, an HC3 p-value of 0.0167, and a studentized permutation p-value of 0.0206 across 5,000 Freedman–Lane iterations.
Pairwise group comparisons revealed that only the contrast between mild cognitive impairment and normal cognition survived three-contrast false discovery rate correction, producing an estimated coefficient of negative 0.926 with a 95 percent confidence interval from negative 1.669 to negative 0.184 and a corrected q-value of 0.0435. In contrast, exploratory regional cognitive associations adjusted for participant sex and educational attainment remained statistically unsupported after local false discovery rate correction.
The authors conducted thorough sensitivity analyses to evaluate demographic adjustments, age overlap, conventional electroencephalography measures, temporal signal stability, and multilayer configuration parameters. Across all twelve tested combinations of gamma and omega multilayer parameters, the association coefficients remained negative. When applying strict age-restricted analyses to participants aged 50 to 73 years, estimates remained negative, though both ordered-category and pairwise HC3 intervals encompassed zero, producing permutation p-values of 0.0532 and 0.0746.
Overall, the researchers concluded that lower dynamic alpha-band network balance characterizes an ordered cross-sectional pattern across these stages, appearing most clearly between mild cognitive impairment and normal cognition. Although this balance metric warrants continued investigation alongside conventional electroencephalography as an experimental tool for measuring network organization, the small, younger normal cognition group and the statistical uncertainty in restricted-age analyses currently prevent any diagnostic or prognostic application.
Subject of Research: Dynamic alpha-band EEG network organization across clinical cognitive stages
Article Title: Dynamic alpha-band EEG network balance differs across clinically defined cognitive groups: a cross-sectional study
Article References: Chen, Y., He, R., Xu, M., Liu, X., & Qing, K. (2026). Dynamic alpha-band EEG network balance differs across clinically defined cognitive groups: a cross-sectional study. BMC Neurology. https://doi.org/10.1186/s12883-026-05595-w
Image Credits: AI Generated
DOI: 10.1186/s12883-026-05595-w
Keywords: electroencephalography, alpha-band oscillations, subjective cognitive decline, mild cognitive impairment, normal cognition, network segregation-integration balance, temporal multilayer community detection, weighted phase-lag connectivity, resting-state EEG, cross-sectional study, electrophysiological organization, HC3-robust regression, Freedman-Lane permutations, false discovery rate correction, cognitive neuroscience, brain network dynamics, neurophysiology
News Source: Cassandra Pierce. (November 19, 2026). Dynamic Alpha Brain Wave Network Balance Differs Across Clinical Cognitive Groups. Scienmag.



