Electroencephalographic Biomarkers of Relaxation: A Systematic Review and Meta-analysis

Abstract

The human electroencephalogram (EEG) is composed of synchronous oscillations within characteristic frequency bands, including 8-13 Hz alpha oscillations that often appear during relaxation. Relaxation is critical for physical and mental health, but the extent to which various EEG components reflect relaxation remains uncertain. This systematic review and meta-analysis investigated associations between EEG components and concurrently measured reference indices of relaxation in healthy adults. A comprehensive database search and screening using preset criteria identified 38 studies involving 1,120 participants published between January 1940 and January 2022 for qualitative synthesis. These studies used various reference relaxation measures such as electrocardiographic indices related to parasympathetic nervous system activity and introspective indices obtained through questionnaires. Risks of bias were evaluated following the risk of bias assessment tool for nonrandomized studies. A meta-analysis of 23 studies using a random-effects model revealed positive correlations between relaxation index and the power of alpha oscillations at central channels (Fishers z-transformed correlation coefficient and [95% confidence interval]: 0.23 [0.10-0.36]) and frontal channels (0.16 [0.02-0.31]). The correlation was significantly higher for the central channel compared with other channels. No significant correlations were detected between relaxation indices and other EEG frequencies or channels. The causal relationships between relaxation and alpha power at central and frontal channels warrant further study.