Breathing mode selectively modulates brain-wide functional connectivity.

While respiration is known to rhythmically modulate brain activity, how different breathing modes (nasal vs. oral) affect frequency-specific large-scale neural connectivity in humans remains unexplored. We used resting-state functional magnetic resonance imaging (fMRI) to examine how nasal and oral breathing modulate functional brain connectivity, focusing on blood oxygenation level-dependent (BOLD) fluctuations in the intermediate frequency band of 0.1-0.2 Hz in 20 healthy male participants. A fully data-driven ROI-based inference approach across 133 whole-brain ROIs revealed that nasal and oral breathing significantly activated the olfactory region and brainstem, respectively. Seed-based connectivity (SBC) analysis, using nonparametric permutation testing (10,000 iterations) and cluster-wise false discovery rate (FDR) thresholding (p-FDR < 0.05), based on these seeds, revealed distinct patterns of network engagement depending on breathing mode. Nasal breathing was associated with greater functional connectivity within higher-order brain networks, including the salience, somatosensory, default mode, and frontoparietal networks. Conversely, oral breathing increased connectivity centered on the brainstem, engaging subcortical regions involved in autonomic regulation and survival functions. Despite these differences, both conditions recruited stable respiratory core regions comprising the hippocampus, amygdala, and insula. These findings suggest a novel framework, the respiration-entrained brain oscillation network (REBON), defined by three operational criteria: (1) it is frequency-specific to the 0.1-0.2 Hz band (centered around ~0.16 Hz); (2) the activity of its principal regions, the olfactory region and brainstem, alternates in dominance depending on the mode of breathing; and (3) it includes a stable core of limbic and interoceptive structures, such as the hippocampus, amygdala, and insula. Understanding this network may have implications for future therapeutic strategies aimed at supporting cognitive functions, emotion regulation, and the integrity of large-scale brain networks in both clinical and wellness contexts; however, these translational implications require validation in future experimental studies.