Dynamic Volitional Respiratory Modulation Increases Large-Scale Brain Network Integration

Poster Session F - Tuesday, April 1, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom

Jude Hammoud¹ (jhammoud@mgh.harvard.edu), Suk-tak Chan², Kenneth K. Kwong³, Clemens C. Bauer^1,4,5,6, David Magone⁷, Susan Whitfield-Gabrieli⁶, Karestan Koenen⁸; ¹Center for Precision Psychiatry, Massachusetts General Hospital, ²Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, ³Department of Radiology, Massachusetts General Hospital, Harvard Medical School, ⁴Department of Psychology, Northeastern University, ⁵Department of Brain and Cognitive Sciences and McGovern Institute for Brain Research, Massachusetts Institute of Technology, ⁶Department of Psychiatry, Harvard Medical School, ⁷Down Under School of Yoga, Boston, MA, ⁸Department of Epidemiology, Harvard T.H. Chan School of Public Health, Harvard University

Volitional breath control modulates nervous system activity, promoting emotional regulation and psychological flexibility—similar to effects seen with psychedelics and meditation. Its neurophysiological mechanisms, particularly the interplay between the Default Mode Network (DMN) and Frontoparietal Control Network (FPCN)—key systems in self-referential processing and cognitive control, remain underexplored. We developed the Dynamic Volitional Respiratory Modulation (DVRM) paradigm, incorporating brief hyperventilation and breath retention, to investigate its effects on resting-state connectivity. Five participants experienced in DVRM completed three rounds of hyperventilation (10, 6, and 6 minutes), each followed by a 1-minute breath-hold, with the final round including up to a 3-minute breath-hold. Resting-state fMRI scans were acquired pre- and post-paradigm. Functional connectivity analyses targeted resting-state networks using masks from Yeo et al. (2011) and Choi et al. (2012). Multiple comparisons were corrected using false discovery rate (FDR). For all participants, post-task DMN-FPCN connectivity increased, with r values ranging from 0.665 to 0.959 (p less than 0.0001), reflecting 28.7 to 618.4 percent increases from baseline. Post-task Somatomotor Network (SMN)-DMN connectivity increased, with r values from 0.360 to 0.904 (p less than 0.0001), corresponding to 19.8 to 381.1 percent increases from baseline. Pre-task SMN-DMN connectivity was non-significant for one participant (p equal to 0.065). Results reflect large-scale network integration, a pattern observed in psychedelic and meditative states. In post-task interviews, participants reported altered self-perception, emotional catharsis, and vivid sensory experiences. DVRM appears to facilitate resting-state network integration, consistent with mechanisms underlying these altered states, suggesting its potential as a complementary strategy in mental health interventions.

Topic Area: METHODS: Neuroimaging