Abstract
INTRODUCTION: Growing evidence shows that voluntary breathing maneuvers modulate cortical oscillations, yet the precise frequency-specific signatures of functional connectivity (FC) remain unclear. METHODS: This study investigated the impact of different respiratory conditions on brain FC using EEG recordings. Three respiratory conditions were collected and analyzed: self-paced breathing (SB), breath-holding (BH), and computer-paced breathing (PB). The power spectral density (PSD), phase-locking value (PLV), and brain network characteristics were analyzed for these different conditions. RESULTS: The results all showed significant differences. The PSD analysis revealed increased low-frequency (δ and θ) activity during SB and higher high-frequency (α and β) activity during BH conditions. The PLV analysis demonstrated significant differences in FC between conditions, indicating specific modulation of brain networks by respiratory state. The brain network properties analysis uncovered frequency-specific changes in clustering coefficient (CC), global efficiency (GE), local efficiency (LE), and degree centrality (DC), reflecting alterations in brain network organization. The three-class classifier showed superior performance in the α band, suggesting its potential as a biomarker for distinguishing respiratory conditions. Correlation analysis with forced vital capacity (FVC) revealed significant associations between brain connectivity and FVC metrics. DISCUSSION: These findings highlight the complex interplay between respiratory conditions and brain FC. These findings suggest that controlled and uncontrolled breathing patterns can influence brain network organization, a mechanistic observation that may inform future respiratory-based interventions aimed at enhancing cognitive function, although behavioural or affective outcomes were not assessed here.