Table 2. Procedures and main findings of the included studies investigating the effects of NB and OB on brain function.

First author (year)	Performed analyses	Main findings
Salimi et al.14 (2023)	Three random experiments with EEG acquisition were performed to evaluate DMN: 1) NB, 2) OB, 3) OB + nasal air puff. EEG data were recorded using a 32-channel active electrode system. EEG electrodes were attached according to the 10-20 system, with reference and ground channels placed respectively on the left mastoid and right earlobe. Nasal air puffs consisted of a brief puff of odorless air delivered to the nasal cavity via a nasal cannula for 3 min (7-10 L/min; 1.1 bar, frequency 0.2 Hz).	NB had a higher DMN power, particularly in the gamma range, than OB. OB + nasal air puff significantly increased signal power compared to OB. In the frontal area, the power of signals was enhanced during OB + nasal air puffing (not observed during NB). NB and OB + nasal air puffs were associated with increased signal coherence compared with OB. OB + nasal air puff increased the number of synchronized channels compared to OB.
Mollakazemi et al.15 (2023)	8-channel EEG recording from the scalp to evaluate whether NB or OB affect emotions triggered by music delivered to participants through a pair of circumaural headphones. The music comprised three two-minute songs (one happy, one peaceful, and one sad song).	Participants found songs more relaxing during NB than OB (P = 0.00013), and felt more aroused than during OB (P = 0.036). During NB, participants found the songs happier (P = 0.069), more exciting (P = 0.063), and less boring (P = 0.082) than during OB. During NB, the respiratory rate (P < 0.001) and heart rate were higher than during OB.
Zaccaro et al.16 (2022)	SNB and SOB recordings were compared with EEG recordings to investigate olfactory epithelium stimulation’s role in disentangling its effects from those related to respiratory vagal stimulation. For SOB, participants were asked to breathe only through their nostrils for 15 min at a respiratory rate of 2.5 breaths per min; A respiratory cycle consisted of four consecutive phases, each lasting 6 s (inspiration-pause-expiration-pause). For SNB, participants were asked to breathe only through their mouth for 15 min at a respiratory rate of 2.5 breaths per min. Nostrils were closed using a clinically approved nasal clip.	A higher power spectral density in the theta and delta bands was observed after SNB compared to post-SOB in the prefrontal and frontal areas. Higher theta band connectivity during post-SNB compared to post-SOB. The connectivity increase was mostly lateralized, involving the left hemisphere (from prefrontal to occipital regions). Increase of theta-high-beta coupling after SNB compared with baseline and post-SOB. Significant increases were found in midline prefrontal/frontal areas and midline posterior regions with the theta phase modulating the high-beta amplitude. Post-SNB was accompanied by an increase in experienced positive emotions compared with post-SOB (P < 0.04). During post-SNB, participants experienced a heightened perception of being in an altered state of consciousness. The post-SNB phase was associated with lower physical and psychological tension than the post-SOB phase, although the differences were not significant (P < 0.32 and P < 0.06).
Jung and Kang21 (2021)	Determine the differences in active brain regions and functional connectivity between the NB and OB groups during a 2-back working memory task using fMRI. On the same day, participants underwent one brain structure scan and two fMRI scans for the working memory task during NB and OB.	Fifteen and 10 regions were activated during NB, while 10 were activated during OB. Among the 15 regions during NB, five (inferior parietal gyrus, insula, cerebellum, precentral gyrus and middle frontal gyrus) appeared in both hemispheres. The functional connection decreased significantly during a working memory task in the OB group compared with the NB group. Functional connections of the left cerebellum and left and right inferior parietal gyrus were observed only during NB but not during OB. Brain areas closely related to working memory function were less active during OB.
Hong et al.22 (2021)	EEG analysis of changes in brain oscillatory activity caused by breathing. Measurements were performed using a 32-channel EEG in a soundproof room. To examine the EEG signal differences in working memory performance during NB and OB tasks, the EEG signals were measured for 5 min each in the rest, 1-back, and 2-back task states. During the rest state with closed eyes, the tasks included NB, OB, and OB with O2 supply.	The working memory accuracy did not differ significantly between NB and OB (P = 0.711). An additional O2 supply during OB is recognized as NB, at least by the brain waves.
Lee et al.10 (2020)	EEG recording was performed to analyze the physiological changes associated with NB and OB. A multi-parameter patient monitor was used to measure the physiological data, SpO2, ETCO2, and RR during resting and n-back working memory tasks.	Compared with NB, discomfort scores were significantly higher during OB in all three states (resting P < 0.001, 0-back P < 0.001, 2-back P < 0.001). ETCO2 was significantly increased during OB (P = 0.0064). The delta wave power increased during OB in the 2-back task. The beta and gamma wave power decreased significantly in the 2-back task during OB (β P = 0.0031, γ P = 0.0057).
Zelano et al.23 (2016)	iEEG from depth electrodes inserted into the PC, amygdala, and hippocampus of seven patients with surgical epilepsy during natural breathing (five with PC coverage; all seven with amygdala and hippocampal coverage).	The inspiratory phase of NB was associated with increased power in the delta frequency range in five patients in PC and seven patients in the amygdala and hippocampus. The nasal route of respiration provides an entry point to limbic brain areas for modulating cognitive function. Air plumes periodically entering the nose at a quiet breathing rate may elicit slow and rhythmic neuronal oscillations that propagate throughout limbic brain networks.

EEG: electroencephalography, DMN: default mode network, NB: nasal breathing, OB: oral breathing, SNB: slow nasal breathing, SOB: slow oral breathing, fMRI: functional magnetic resonance imaging, SpO₂: oxygen saturation, ETCO₂:end-tidal CO₂, RR: respiratory rate, β: beta, γ: gamma, iEEG: intracranial electroencephalography, PC: piriform cortex