Breathing in synchrony: Exploring neural-visceral effects in Su-soku breath counting

Meditation is a mental practice centered on focused attention and awareness. It often involves mindful breathing: attending to the flow of air into and out of the body. Mindful breathing is associated with improvements in mental well-being and cognitive function. However, mindfulness meditation can be challenging, especially those with mental health difficulties. In this study we investigated whether a single session of “Su-soku” – a traditional breath-counting exercise – would have measurable effects on central and autonomic correlates of attentional focus. Our primary prediction that was that Su-soku would tend to increase induced theta oscillations during an auditory oddball task. Forty one healthy participants performed a pair of probe tasks (auditory oddball; eyes-open rest) three times: at baseline; after the first “intervention” task (Su-soku or a control task); and after the second “intervention” task (intervention task order counterbalanced across participants). Frontal-midline central nervous system activity was recorded using electroencephalography (EEG). Heart Rate Variability (HRV) indices of autonomic function were computed from electrocardiogram traces recorded via wrist electrodes. We found no evidence that induced theta during the oddball task increased after Su-soku, but did find evidence of increased cardiac deceleration, indicating greater parasympathetic responsivity. HRV indices from the eyes-open rest period also indicated a greater shift towards parasympathetic dominance after Su-soku than after the control task. Frontal-midline oscillatory power in theta, alpha, and beta frequencies, and indices of parasympathetic activity, increased during Su-soku meditation relative to the previous period of Eyes-Open rest, and tended to peak towards the end of exhalation. These results indicate that even a single short session of Su-soku practice can shift autonomic balance in the direction of parasympathetic dominance. The study also provides insights into neural-respiratory interactions during focused breathing, potentially laying the groundwork for future accessible mindful breathing therapeutic interventions in mental health. Names and contact info for all co-authors and ORCiDs Natasha A Elliott: ORCID: 0000-0002-8669-3703. Email: [email protected] Affiliation: University of Nottingham Elizabeth B Liddle: ORCID: 0000-0002-9299-094X Email: [email protected] Affiliation: University of Nottingham Dawn M Walker: ORCID: 0000-0003-2135-1363 Email: [email protected] Affiliation: University of Southampton Peter F Liddle: ORCID: 0000-0001-6473-7640 Email: [email protected] Affiliation: University of Nottingham Author Contributions: NAE made a major contribution to the data preprocessing; statistical analysis including coding analysis scripts; interpretating the findings; writing the original draft; editing the final draft. EBL made a major contribution to the conceptualisation of the study; contributed to the writing of the ethics application; coded the tasks; collected the data; made a major contribution to the data preprocessing including coding the HRV preprocessing pipeline; conducted and supervised statistical analyses; contributed to the interpretation of the findings, and made a contribution to writing the original draft, and editing the final draft. DMW initiated the original study concept; made a major contribution to the writing of the ethics application; collected the data; contributed to the editing of the final draft. PFL made a major contribution to the conceptualization of the study; contributed to the design of the study protocol; contributed to the writing of the ethics application; supervised the statistical analyses and the writing of the original draft; contributed to the interpretation of the findings, and made a major contribution to editing the final draft.

We acknowledge the valuable contribution made by MSc students Anna Yates and Yamini Negi, and medical students Liam Musselbrook and Priya Asi to recruitment and data collection. Funding: Funding for participant payments was provided by the Institute of Mental Health. Nottingham. NAE was funded by NIHR BRC PhD scholarship plus a COVID19 stipend from the University of Nottingham

Meditation is a mental practice centered on focused attention and awareness. It often involves mindful breathing: attending to the flow of air into and out of the body. Mindful breathing is associated with improvements in mental well-being and cognitive function. However, mindfulness meditation can be challenging, especially those with mental health difficulties. In this study we investigated whether a single session of “Su-soku” – a traditional breath-counting exercise – would have measurable effects on central and autonomic correlates of attentional focus. Our primary prediction that was that Su-soku would tend to increase induced theta oscillations during an auditory oddball task. Forty one healthy participants performed a pair of probe tasks (auditory oddball; eyes-open rest) three times: at baseline; after the first “intervention” task (Su-soku or a control task); and after the second “intervention” task (intervention task order counterbalanced across participants). Frontal-midline central nervous system activity was recorded using electroencephalography (EEG). Heart Rate Variability (HRV) indices of autonomic function were computed from electrocardiogram traces recorded via wrist electrodes. We found no evidence that induced theta during the oddball task increased after Su-soku, but did find evidence of increased cardiac deceleration, indicating greater parasympathetic responsivity. HRV indices from the eyes-open rest period also indicated a greater shift towards parasympathetic dominance after Su-soku than after the control task. Frontal-midline oscillatory power in theta, alpha, and beta frequencies, and indices of parasympathetic activity, increased during Su-soku meditation relative to the previous period of Eyes-Open rest, and tended to peak towards the end of exhalation. These results indicate that even a single short session of Su-soku practice can shift autonomic balance in the direction of parasympathetic dominance. The study also provides insights into neural-respiratory interactions during focused breathing, potentially laying the groundwork for future accessible mindful breathing therapeutic interventions in mental health.

Meditation is a mental practice centered on focused attention and awareness. It often involves mindful breathing, which entails paying attention to the breath as it flows in and out of the body. Practicing mindful breathing is associated with improvements in mental well-being and cognitive function, even following brief training (Basso et al., 2019; Wu et al., 2019; Zeidan et al., 2010). Mindfulness-based therapies have also shown promise in alleviating symptoms across a range of mental health conditions, including mood and anxiety disorders (Rodrigues et al., 2017), and psychosis (Jansen et al., 2020). However, the underlying physiological mechanisms supporting these effects remain unclear. In their neurovisceral integration model, Thayer and Lane (2000) proposed that the brain and body are closely connected in a dynamic system that supports adaptive behaviour, especially emotional and cognitive self-regulation. Within this framework, the central nervous system regulates the autonomic nervous system (ANS) activity in response to environmental demands. For instance, Thayer and Lane point to work done by Clifton and Graham (1966) showing that while Heart Rate (HR) acceleration is associated with the defensive “startle” response, preparing the organism for “fight or flight”, parasympathetically mediated HR deceleration in response to a novel stimulus is part of the orienting response (Bradley et al. 2012). Heart Rate Variability (HRV) - changes in heart-rate over time - reflects activity in both the sympathetic and parasympathetic branches of the ANS, and can be a useful tool in assessing how effectively this dynamically integrated system is functioning. Short-term fluctuations in heart-rate (high frequency) HRV largely reflect rapid changes in vagal nerve activity, thus indexing activity in the parasympathetic branch of the ANS, while longer-term lower-frequency fluctuations reflect variation in sympathetic activity. Mindful breathing tends to increase high frequency HRV and shift autonomic balance away from sympathetic-dominant “fight-or-flight” mode towards parasympathetic “rest-and-recover” mode (Natarajan, 2023). High-frequency HRV reflects a strong capacity to adapt rapidly to changes, suggesting effective emotional regulation, while low-frequency HRV has been linked to anxiety, depression, and reduced ability to cope with stressors (Chalmers et al, 2014; Mather & Thayer, 2018). Respiratory sinus arrhythmia (RSA) is a normally adaptive physiological mechanism that modulates variation in heart rate across the respiratory cycle in a manner that optimises gas exchange between the lungs and the cardiovascular system. As the lungs fill during inspiration, pulmonary stretch receptors inhibit vagal activity, allowing heart rate to rise and take advantage of oxygen rich air, while during expiration, vagal nerve activity increases in response to increased activity in arterial baroreceptors, reducing heart rate again (Komisaruk and Frangos 2022). Importantly, the RSA relies on fluctuations in parasympathetic vagal activity, rather than sympathetic activity. By employing a Zen practice known as Su-Soku, which involves simply sitting quietly and counting one’s breaths, Kubota et al. (2001) found that participants frequently exhibited increased frontal midline theta (FM-theta) activity. HRV analyses during these episodes of increased FM-theta episodes showed concurrent increases in both sympathetic and parasympathetic activity, with theta power negatively correlated with sympathetic activation, suggesting an interaction between medial frontal attentional networks and autonomic regulation. Broader investigations into mindfulness practices using EEG have found increases in both theta and alpha activity in individuals with and without mental health diagnoses (Lomas et al., 2015). Lomas et al. (2015) suggest that the co-occurrence of elevated alpha and theta may reflect a state of “relaxed alertness” that supports mental health. However, theta activity in isolation—especially in the absence of higher-frequency oscillations – can also signal drowsiness rather than focused relaxation (Makeig et al., 2000). While frontal-midline theta is often elevated during cognitively demanding tasks, the theta/beta ratio has been proposed as a marker of cognitive control and may index resilience to stress (Putman et al., 2014). Sargent et al (2024) reported that in healthy participants in an eyes-open resting state, high frequency oscillations in heart rate (HF-HRV) were coupled to neural oscillations recorded over frontal channels in all frequency bands from delta to gamma. Granger causality analysis indicated that stronger influence of the heart on the brain than vice versa in all frequencies except gamma. Furthermore Melnychuk et al. (2021) found that theta-beta ratio was synchronized with respiratory rate. Several studies suggest the benefits of meditation may arise from the functional integration of the central and peripheral nervous systems (Critchley & Garfinkel, 2018; Zelano et al., 2016; Herrero et al., 2018). While mindfulness practices confer both mental and physical health benefits, they can be difficult for people with serious mental health conditions to engage with, and may even have adverse effects, such as hyperarousal and dissociation (Britton et al., 2021). In order to better understand the mechanisms underlying the effects of mindfulness practices, we investigated the effects of a short single session of Su-soku on neuro-visceral processes. Study Aims Originally conceived as a small feasibility study of the potential usefulness of Su-soku practice in patients with schizophrenia, we later expanded the original study to allow us to investigate the effects of a single session of Su-soku practice in a more substantial sample of participants recruited from the general population. This paper reports findings from this non-clinical sample. Using a single-session counterbalanced design, we set out to compare effects of two brief interventions: a Su-soku-based breath-counting task and a cognitively demanding timing task. Each intervention task lasted approximately 15 minutes total. Before and after each one, participants completed an auditory oddball target-detection task to assess task-related neural and behavioral responses, then a brief eyes-open rest period to assess effects on awake relaxation. Objectives/hypotheses Given evidence that people with schizophrenia exhibit a marked reduction in induced frontocentral theta oscillations—associated with attentional control - during auditory target detection tasks (Doege et al., 2009), alongside deficits in parasympathetic activity (Stogios et al., 2021), our primary hypothesis was that event-related fronto-central theta oscillations during the auditory oddball task would exhibit a greater increase following the Su-soku task than following the Timing task. We also hypothesized that Su-soku practice would induce a greater parasympathetic responsivity during the oddball task, reflected in greater heart-rate deceleration following task stimuli. Additionally, we expected Su-soku to modulate autonomic and oscillatory indices associated with a focused attentional state during the subsequent eyes-open rest period. To assess these effects, we examined key autonomic markers, including RMSSD (Root Mean Square of Successive Differences), a time-domain index of HRV that reflects vagally mediated changes in heart rate (Shaffer & Ginsberg, 2017), as well as the Cardiac Vagal Index (CVI) and Cardiac Sympathetic Index (CSI), which represent vagal and sympathetic variability, respectively (Toichi et al., 1997). The indices of interest included: • Oscillatory power in theta, alpha, and beta frequency bands • Theta/beta power ratio • Parasympathetic-dominant HRV markers of autonomic balance, specifically: • Increased RMSSD and/or CVI • Decreased CSI and heart rate (HR) • Enhanced amplitude of Respiratory Sinus Arrhythmia (RSA) Finally, we were also interested in how these indices measured during the Su-soku task itself changed relative to the preceding period of eyes-open rest; whether they changed over the course of the Su-soku task; and whether the oscillatory amplitude fluctuations during the Su-soku task would be entrained by the respiratory cycle.

Ethical approval for the study was obtained from the East Midlands National Research Ethics Service (NRES) committee (REC reference 05/Q2404/45). Participants In our original study protocol, we planned to recruit at least 18 (max 50) participants per group (patients with schizophrenia; control participants), allowing for a dropout rate of 10%. This sample size was to provide 80% power to detect a meaningful increase in induced theta in the patient group (effect size estimated from data in Doege et al. 2009). Unanticipated recruitment challenges meant we were able to recruit only five patients, only three of whom provided usable data. However, we were able to recruit 42 control participants (21 female), giving >80% power to detect a medium effect size for our primary hypothesis in the control participants. Eligibility criteria for control participants were: • No personal history of psychotic, medical or neurological illness. • No first-degree relative with a history of psychotic illness. • No history of drug abuse (according to the DSM-IV criteria) within the last three months or any illicit drug use 48 hours before the EEG session. • No substantial change to nicotine or caffeine use before the EEG session. The 42 participants were recruited from the surrounding Nottinghamshire area using advertisement flyers. All participants were naïve meditators with little to no experience with meditation. Participants received £10 in compensation for travel and inconvenience. Study design The single-session counterbalanced design of the study is illustrated in Figure 1. During the experimental session, participants completed two Intervention tasks (Su-soku; Timing task) and two Probe tasks (Auditory Oddball; Eyes Open Rest). The pair of Probe tasks were presented three times: at the beginning of the session, and after each Intervention task. Intervention task order was randomly counterbalanced across participants. Task order was included as a control factor in tests for any intervention effects. Figure 1: Diagram showing the experimental protocol, with the Probe tasks shown in green. After the EEG cap and electrodes had been applied, the participant was asked to count their breaths over one minute. The software running the task sequence was then started, and their breathing rate entered as input to determine the pacer stimulus rate for the Su-soku task. Task order was then randomly allocated, and participants practiced each task in their allotted task order. The pacer stimulus was then adjusted if the participant had found it too fast or slow. Brackets indicate the comparisons used to test the study hypotheses. Procedure After obtaining written informed consent from the participant, and applying the EEG cap and electrodes, the participant was asked to sit quietly and count their breaths for one minute to determine their comfortable breathing rate. The software running the experimental session was then started, and the participant’s breathing rate input to determine the rate of the pacer stimuli during Su-soku. This was set at a maximum of 17 breaths per minute. After randomly allocating the participant to one of the two counterbalanced task sequences, the participant then practiced short versions of all tasks, in the allotted task order. If they found the breathing pace for the Su-soku task uncomfortable, this was adjusted prior to beginning the experimental task sequence. The whole task sequence, including practice procedures, took about one hour. EEG acquisition EEG was recorded continuously during the entire series of activities, using a 128-channel (ABC layout) BioSemi Active Two system (Amsterdam, Netherlands) at a sampling rate of 512Hz. Reference electrode were placed on both ear lobes. Horizontal electrooculogram (HEOG) data were recorded from electrodes located laterally to the outer canthus of the left and right eyes, and vertical EOG (VEOG) from an electrode below the right eye and scalp electrode FP2. Electrocardiographic (ECG) data were recorded from electrodes applied to each wrist. Probe tasks Oddball task For the Auditory Oddball task, low-pitched Standard tones (1000 Hz) and less frequent Oddball tones (1500 Hz) were presented in random order, with a probability ratio of 9:51 (.15 Oddball probability). Participants were instructed to press a button in response to the Oddball tones, but to ignore the Standard tones. Each task presentation consisted of 60 trials, which always included 9 Oddball tones. Interstimulus intervals were randomly drawn from a uniform distribution of 60 values between 2000ms and 3500ms. The participant was given an opportunity to take a short rest after 30 trials. Eyes Open Rest For the Eyes Open Rest task, participants were asked to quietly view a static image of a woodland scene for two minutes, keeping their eyes open, and their eyes fixed on an unobtrusive fixation cross at the focal point of the scene to minimise eye-movements. A looped soundtrack of natural woodland sounds was played throughout. Intervention tasks Su-soku task The Su-soku task consisted of three consecutive blocks of breath-counting, each lasting about four minutes (rounded to an integral number of respiratory cycles). Participants were presented with a visual pacer stimulus set to their comfortable breathing pace. This consisted of a coloured circle that expanded and contracted changing in colour from blue to yellow as it expanded, and back to blue as it contracted. Markers were sent to the EEG recording computer at the onset of each circle expansion. Participants were asked to breathe in time with the pacer, and to count their breaths. The number of pacer stimuli presented per block was randomly varied by a small amount. At the end of each block, the researcher asked the participant how many breaths they had taken. Timing task The Timing task involved learning, by trial and error, the time interval between an auditory stimulus (a clicking sound) and the appearance of a visual stimulus (a cartoon “bug”). Participants were asked to press a button just as the bug appeared (“zap the bug”). Feedback was provided after each trial. There were five blocks, each with a different time interval to be learned. The task was designed to engage both attention and inhibitory control. EEG analysis EEG data was processed offline using Brain Vision Analyzer software (BrainVision Analyzer, Brain Products GmbH, Gilching, Germany). Pre-processing Raw EEG data was referenced to ear lobe channels, then passed through Butterworth Zero Phase filters with a low cut-off of 0.5 Hz and high cut-off of 45 Hz to reduce spectral leakage and aliasing effects. A notch filter of 50 Hz was also applied to minimise 50 Hz electrical noise. Data were then segmented into epochs of interest (Su-soku task; Eyes-open Rest; Oddball task) for further processing. Artefacts within segments were identified using a semi-automatic approach using the following criteria: Allowing a maximal voltage step of 50µVs, a maximal allowed difference of values of 200µV in intervals of length 200ms, the minimal allowed amplitude of -200µV, the maximal allowed amplitude of 200µV and lowest allowed activity in intervals at 0.5µV. Oculomotor artefacts were corrected using an automated Gratton and Coles method (Gratton et al., 1983) using HEOG and VEOG data. Oscillatory measures We selected data recorded at Fz, and from electrodes approximating to positions FCz, FC1 and FC2 (see Figure 2), as defined by the American Clinical Neurophysiology Society Guidelines for Standard Electrode Position Nomenclature (Acharya et al, 2016). We defined Frontal-midline (FM) power as the signal averaged across these four electrodes (see supplementary materials). Data from noisy or missing electrodes were estimated by topographical interpolation, using spline interpolation. Figure 2: Electrode locations. Fz, Cz, C3 and C4 are defined on the BioSemi ABC cap layout. In addition to data from Fz, we analysed data from an electrode midway between Fz and Cz, approximating to FCz, and electrodes to either side, approximating FC1 and FC2. The colour intensity represents a topographical map of theta amplitudes from a typical participant recorded during the Su-soku task. For spectral analysis of oscillatory power during each of the three consecutive Su-soku blocks and the three Eyes-open Rest tasks, data were first segmented into 10 second epochs. Absolute power (µV²) was then calculated using fast-Fourier transform (FFT) with a resolution of 1Hz, 10 percent Hanning window length, full spectrum, normalized segments and variance correction (see supplementary materials). For each participant, the spectra were then averaged across the 10 second epochs and broadband values extracted at each of the four selected electrodes for: theta: 4-7 Hz, alpha: 8-12 Hz, beta: 12.5–20 Hz. These values were then averaged across electrodes, then transformed to amplitudes (which have a less skewed distribution) prior to statistical analysis by taking the square root. Theta/beta power ratios were calculated by dividing theta power by beta power for each electrode for each task block. These ratios were then log 10 transformed to produce an interval scale. For time-series analyses of oscillatory amplitude changes over the respiratory cycle, a continuous wavelet transform (CWT) was applied to the Su-soku task data recorded at Fz using the Morlet complex wavelet function between 2 to 30 Hz frequencies, using 29 linear frequency steps. Wavelet normalization was set to instantaneous amplitude with a Morlet parameter of 5. The output values of this analysis were the absolute values of the complex output, giving a time series of spectral amplitudes at each frequency band in µV units. To investigate consistently respiration-locked changes across the broad 2 to 30 Hz spectrum, the 29-band time frequency spectrograms (TFS) for the Su-soku data were first z-scored within each participant, at each frequency, across all three Su-soku blocks, then, within blocks, segmented between pacer markers, and finally averaged across segments to give a z-valued mean TFS for each participant across their respiratory cycle. As the assigned breathing rate was different for each participant, these averaged TFS amplitude matrices were then binned into 1000 bins, regardless of the participant’s respiratory cycle length. To analyse respiration-locked changes in the three broader frequency bands of interest, amplitude time series for each band were computed by averaging across the TFS bands: theta (4-7 Hz), alpha (8-12 Hz) and beta (12.5–20 Hz). The time series for each band was then segmented between pacer markers, and as with the TFS time series, averaged across segments for each Su-Soku block, then binned into 1000 bins, regardless of the participant’s respiratory cycle length. Oddball task: Induced theta The Oddball EEG data recorded at Fz were segmented into the three Oddball sessions (two blocks per session), then filtered to within 6 to 8Hz by a CWT using a Morlet Complex wavelet with parameter 3, in Brain Vision Analyzer, and the event markers and complex output exported, for further processing in Matlab. Absolute values of the complex-valued wave-forms were then segmented into two windows: a baseline reference from between 300ms and 200ms prior to stimulus onset, and a window of interest from between 200ms to 500ms after stimulus onset (Doege et al., 2009). Induced theta was calculated as the difference between peak theta amplitude in the window of interest and mean theta during the reference period, divided by the reference period mean, to give an induced theta value for each trial (Pfurtscheller and Lopes da Silva, 1999). These were then averaged across Oddball and Standard trial types respectively, for each of the three Oddball task sessions. ECG analysis ECG data recorded from wrist electrodes during each task of interest were preprocessed using an in-house interactive MatLab script with a graphical user interface. ECG data for each task block were bandpass filtered to between 8 and 40 Hz, squared, and presented graphically to the user to select the cleanest channel, and then to select visibly noisy segments for excision. The user then selected the highest and lowest values for plausible R wave peaks from which to calculate candidate interbeat interval (IBI) values i.e. the elapsed interval between each R wave. These IBIs were then presented graphically in a bar chart, with each R-R interval bar located at the time point at which the terminating R value had occurred. This allowed outlying large (e.g. representing missed R waves or excluded data) or small values (e.g. representing misidentified R waves) to be identified and excluded, as well as further selection of visibly implausible segments of IBIs for exclusion. An interpolation function (MatLab function interp1, method ‘pchip’) was then used to a generate continuous IBI-based wave form at the original ECG sampling rate (512 Hz), which was then converted to a continuous Instantaneous Heart Rate (IHR) time series by dividing into 60,000, for presentation to the user. To estimate the amplitude of the RSA, a CWT was then applied to the continuous IBI series with frequency limits of 0.05 and 2.5 Hz, and a frequency spectrum computed as the median value in each frequency band across the task block. A graphical representation of this spectrum was then presented, with frequency bounds bracketing the assigned breathing rate (half and double the assigned rate), and the peak identified as the maximum value within those bounds. Users could adjust the boundaries to isolate a local maximum within those bounds where a higher “shoulder” value had been identified as the maximum. Outputs from this process were then fed as inputs to another in-house function which returned a the mean HR value as well as a set of HRV indices. These included: RSA peak frequency; RSA peak amplitude; the Root Mean Square of Successive Differences (RMSSD) between IBIs; the two indices proposed by Toichi et al. (1997): the Cardiac Vagal Index (CVI); and the Cardiac Sympathetic Index (CVI). For the Oddball task data, we additionally used the interpolated IBI time series to calculate mean stimulus-locked HR change rates for Target and Standard stimuli for each of the three Oddball blocks. The continuous IBI time series was first segmented from -500ms to +500ms relative to each stimulus. Baseline IBI was calculated as the mean IBI from -500ms to -250 ms, and Poststimulus mean IBI as the mean from 250 to 500ms. To give an index of HR change (positive for acceleration; negative for deceleration) for each trial type, change in IBI was then computed for each stimulus as (Baseline-Poststimulus)/Baseline, then averaged across each trial type. Statistical analysis As this study was only powered for one primary outcome (effects of Su-soku on induced theta during the oddball task), we treat all other null hypothesis tests as exploratory, and present p values as indicators of model fit only. For these exploratory tests, following the second recommendation by Benjamin and Berger (2019), we provide both p value and upper Bayes Factor Bound (uBFB) on the odds in favour of the alternative to the null hypothesis. Upper BFBs were calculated from p values using the sample-size adjusted method proposed by Held & Ott (2016), and interpreted using their proposed strength-of-evidence descriptors. Linear statistical analyses were conducted using IBM SPSS 29. For repeated measures ANOVAs, Huynh-Feldt adjustments to degrees of freedom were used where the assumption of sphericity was violated. Circular statistics were computed either in Python using imported R functions, or in MatLab using the CircStat toolbox (Berens, 2009) to estimate phase angle distributions (mean, median, SD) and test against a von Mises distribution. Effects of Su-soku vs Timing task on Oddball and Eyes-Open outcomes. To assess the effects of each task on changes in each outcome measure, we computed the difference between values recorded prior to each task (Su-soku; Timing Task) from those recorded after task completion. We then conducted two-way mixed ANOVAs, with these change scores as dependent variables, Task Type (2 levels) as within-subject factor and Task Order as a between-subjects factor to control for and identify any effects of intervention task order. For effects on HRV measures we entered change scores for RMSSD, HR and CSI as multivariate dependent variables, and report doubly-multivariate ANOVAs outputs. We selected RMSSD rather than CVI as a measure of parasympathetically mediated HRV, as it is a more widely reported index. Effects of Su-soku blocks relative to preceding Eyes-Open resting state To assess changes in outcome measures over the three Su-soku blocks, relative to each other and to values recorded during the prior Eyes-Open resting state block, we conducted two-way mixed ANOVAS, with Block (4 levels: resting state prior to the Su-soku task; the three Su-soku blocks) as within-subjects factor and Task Order as a within-subject factor. Where evidence indicated change over time, orthogonal contrasts (Helmert) were used to compare each successive block with the mean of the remainder. Respiratory entrainment of oscillatory amplitudes To assess how consistently participants’ oscillatory amplitudes across the broad spectrum from 2 to 30hz varied across the oscillatory cycle during the Su-soku task, for each block, we conducted one-sample t-tests (test value = 0) across participants on the z-valued spectrograms at each time point and frequency. This gave a t -valued spectrogram representing how consistently, across participants, oscillations at each frequency deviated from the participant’s mean at each point in the respiratory cycle. To evaluate the point in the respiratory cycle at which oscillatory power within each frequency band of interest. for each participant, Su-soku block and frequency band, we computed the mean phase angle and resultant value for each point in the 1000 bin time series, treating the j th bin as an phase angle of j *2 π /1000 radians, and the radius as the oscillatory amplitude at that bin. A custom Python script was used to check whether the data met the assumptions for the Rayleigh’s test, using a p-p plot was used to compare the distribution of oscillatory amplitude across the respiratory cycle with a von Mises distribution, the circular equivalent of a normal distribution. The consistency of the mean phase angles for across participants, for each frequency band and block, was then tested using Rayleigh’s test. Within-subject changes in the magnitude of the resultant values was across the three Su-soku blocks was tested using a mixed repeated measures ANOVA, with block as the within-subjects factor (3 levels) and Task Order as a between-subjects factor, to control for any task-order effects.

Of the participants recruited one did not complete the protocol, owing to discomfort with the EEG set-up procedure. Data from one further participants were excluded from further analysis due to noisy electrophysiological recordings, leaving at most 39 participants with usable data for analysis: 20 in the Su-soku first group (13 female) and 19 in the Timing task first group, 8 female). Pearson’s correlations across participants, between prescribed respiratory rate and peak RSA frequency measured in each of the Su-soku blocks were at least r =.96, indicating that high compliance with the prescribed breathing rate. Su-suku vs Timing Task 1: effects on Oddball outcomes Task performance Overall, participants’ accuracy on the auditory oddball task was high: mean accuracy was over 99% for all three oddball sessions, with minimum scores of 95% on the first session and 98% on the second and third sessions. Individual median RT values were positively skewed across participants; we therefore applied a log 10 transform before statistical testing. We found moderate evidence that RTs to Target stimuli increased more after Su-soku than after the Timing task, F (1,38)=24, p =.017, upper BFB = 5.8 (Figure 3, E and F). There was no evidence for any effect of task order (Su-soku first vs Timing first) Induced theta Consistent with Doege et al. we found substantial evidence that Targets induced greater theta amplitudes than Standards in the late window (200ms to 500ms post stimulus) in frontal electrodes, F (1,37)=8.59, p =.006, upper BFB =11.8. However, we found no evidence for any difference between Su-soku and the Timing task on changes in induced theta amplitude for either trial type, nor any evidence for any increase in induced theta following either intervention (See Figure 3, A and B). HRV There was strong evidence of post-stimulus HR deceleration across both Oddball trial types, F (1,37)=71.71, p =68.5, and moderate evidence for greater deceleration following Targets than standards, F (1,37)=6.44, p =.015, uBFB >6.4. There was also moderate evidence that phasic HR deceleration increased more after Su-soku than after the Timing task, F (1,37)=5.06, p =.031, uBFB =4.1, and moderate evidence of a greater reduction in mean HR after Su-soku than after the Timing task, F (1,37)=5.38, p =.026, uBFB =4.1 (see Figure 3 C and D). Figure 3: Effects of Su-soku vs Timing on Oddball task. A: Induced theta in each group for each trial type and session; B: Change in induced theta before and after each intervention task (Su-soku vs Timing); C: Post-stimulus HR change in each group for each trial type and session. Negative values indicate post-stimulus deceleration; positive values indicate acceleration. D: Change in post-stimulus HR deceleration before and after each intervention task. Post stimulus HR deceleration tended to increase after Su-soku in contrast to a tendency to increase after Timing. E: Reaction times for correct responses for each group and session. F: change in RT before and after each intervention task: RTs tended to lengthen after Su-soku and shorten after Timing. Error bars indicate 95% confidence intervals. Su-soku vs Timing Task 1: effects on Eyes-Open rest Theta There was no evidence that either frontal theta amplitudes were greater following Su-soku than after the Timing task. HRV (N=38) A doubly-multivariate test of the effect of intervention type on RMSSD, mean HR, and CSI, provided strong evidence, ( F (4,33)=12.40, p <.001, uBFB <68.8 of correlated effects of intervention type on these three measures of autonomic state, in the direction of greater increase in parasympathetic dominance after Su-soku than after the Timing task. The univariate effects are depicted in Figure 4: After Su-soku, RMSSSD tended to increase, and HR and CVI decrease relative to changes in the opposite direction after the Timing Task. There was no evidence of any effect of intervention order. Figure 4: Changes in HRV variables assessed during rest before and after Su-soku vs before and after Timing. Error bars indicate 95% confidence intervals. Su-soku compared to preceding rest period Theta There was substantial evidence of a change in FM theta power over Susoku blocks, relative to the preceding Eyes-open Rest block, indicated by a main effect of Block, F(1.7, 64.1)=5.7, p=.007, uBFB =11.8. Inspection of the change in mean values indicated a monotonic increase in oscillatory power over time (blue line,Figure 5A). Planned orthogonal contrasts provided moderate evidence, F (1, 37)=5.7, p =.015, uBFB =6.4, of greater FM theta amplitudes during the Su-soku session than during the prior rest period, as well moderate evidence that theta continued to increase after the first block: amplitudes over the second and third Su-soku blocks were higher than in the first block, F (1, 37)=5.23, p =.028, uBFB =3.9. There was no evidence of any effect of intervention order. Figure 5: Change in oscillatory power across the three Su-soku blocks, relative to baseline (preceding Eyes-Open Rest period.) A: changes in theta, alpha and beta power. B: Changes in theta-beta ratio. Error bars indicate 95% confidence intervals. Alpha There was also strong evidence of a change in FM alpha power over Susoku blocks, relative to the preceding Eyes-open Rest block, indicated by a main effect of Block, F (2.2, 80.0)=21.87, p 68.5. Inspection of the change in mean values again indicated a monotonic increase in oscillatory power over time (orange line, Figure 5A). Planned orthogonal contrasts provided strong evidence, F (1, 37)=32.23, p <.001, uBFB =68.5, of greater FM alpha amplitudes during the Su-soku session than during the prior rest period, as well substantial evidence of continued increase: over the second and third Su-soku blocks FM alpha was higher than in the first block, F (1, 37)=10.40, p =.003, uBFB =24.0). Again, there was no evidence of any effect of intervention order. Beta There was also strong evidence of a change in FM beta power over the baseline and Su-soku blocks: main effect of Block, F (1.9, 75.0)=11.26, p68.5. Inspection of the change in mean values once more indicated a monotonic increase in oscillatory power over time (blue line, Figure 5A). Planned orthogonal contrasts provided strong evidence, F (1, 37)=18.93, p <.001, uBFB =68.5, of greater FM beta amplitudes during the Su-soku session than during the prior rest period, but only weak evidence of continued increase, F (1, 37)=3.70, p =.062, uBFB =24.0). There was no evidence of any interaction between intervention order and change over time, but weak evidence, F (1, 37)=4.50, p =.041, uBFB =3.0, that mean FM beta was lower overall in those who did Su-soku before the Timing task than in those who did it after. Theta/Beta ratio There was only weak evidence for an effect of time over the baseline and Su-soku blocks, F (1.1,42.2) = 3.52, p =.062, uBFB =2.2. Figure 5B shows a slight monotonic decrease in theta/beta ratio over the four blocks. Helmert contrasts indicated moderate evidence of a decrease between the second and third Su-soku block, F (1,37) = 5.73, p =.062, uBFB =4.7. Heart Rate Variability A doubly multivariate repeated measures ANOVA to investigate changes in the three HRV variables (RMSSD, mean HR, and CSI) during three Su-soku blocks relative to the preceding Eyes-Open Rest period provided strong evidence for correlated changes, F (9,28)=8.52, p <.001, uBFB = 68.8 over time. The patterns of change (Figure 6) indicate that relative to Eyes-open Rest, Su-soku produced an autonomic shift towards sympathetic, rather than parasympathetic dominance. Helmert contrasts indicated that HR and CSI were elevated during Su-soku relative to the preceding Eyes-open period, while RMSSD increased between Su-soku blocks 2 and 3. Figure 6: Change in HRV variables over the three Su-soku blocks relative to baseline. Respiratory entrainment of oscillatory amplitudes Broadband effects The t -valued spectrograms representing consistency across participants in the degree to which their oscillatory amplitudes varied systematically across their respiratory cycle are shown in Figure 7. Respiratory phase is represented on the horizontal axis, starting at the signal to begin inspiration, with oscillatory frequency on the vertical axis, and colour represents the t value. Large positive or negative t-values (dark blue or bright orange regions of time-frequency space) indicate deviations from mean amplitude that are consistent across participants. The most prominent effects are for alpha and low beta to peak around the end of expiration/beginning of inspiration. Figure 7: Time frequency spectrograms showing change in oscillatory amplitudes from 2 to 30 hz, over the respiratory cycle, represented on the horizontal axis, starting at the signal to inhale. The amplitude data at each frequency band was z scored within each participant then then binned into 1000 bins per respiratory cycle to allow comparison across participants. These spectrograms show t values for one-sample t tests across participants at each bin and frequency, thus indicating how consistently oscillations at each frequency deviated from the participant’s mean at each point in the respiratory cycle. Respiratory entrainment within each frequency band of interest

from the between-subjects Rayleigh’s tests of participants’ mean respiratory phase angle for each frequency and Su-soku block and frequency band of interest are shown in Table 1. Polar histograms illustrating the distributions across participants are shown in Figure 8. A Respiratory phase angle of 0 radians would indicate that peak oscillatory amplitude occurs at the time of the pacer circle expansion (signalling “inhale”), while an angle of π radians would indicate peak at the start of circle contraction (signalling (exhale). For all frequencies and blocks, mean respiratory angle is within one radian of the start of inspiration. EVIDENCE STRENGTH Table 1: Rayleigh’s test results for tests of mean respiratory phase angle for theta, alpha and beta amplitudes. A phase angle of 0 represents a peak at the onset of the signal to inhale. A small negative phase angle represents a peak near the end of expiration. | Mean (SD) | Rayleigh’s z | p value | Mean (SD) | Rayleigh’s z | p value | Mean (SD) | Rayleigh’s z | p value | | | Theta | -0.82 (1.30) | 0.94 | .394 | -0.47 (1.15) | 4.41 | .011 | -0.82 (1.17) | 3.93 | .019 | | Alpha | -0.36 (1.11) | 5.89 | .036 | -0.85 (1.16) | 4.19 | .014 | -0.50 (1.00) | 9.69 | <.001 | | Beta | -0.84 (1.13) | 5.22 | .030 | -0.39 (1.21) | 2.93 | .053 | -0.63 (1.07) | 7.22 | <.001 | Figure 8: Polar histograms showing distributions of oscillatory amplitude peaks across participants, for each Su-soku block. A mixed ANOVA on the participants’ resultant lengths for each frequency and blocks, task-order as a between subjects factor, provided strong evidence, F (1.6, 51.1)=15.64, p 70.2 that resultant length tended to decrease in successive Su-soku blocks, indicating a tendency for the the timing of the oscillatory peak to be either broader or more variable with successive blocks (Figure 9). There was no evidence for differences in resultant length between frequencies, nor any evidence for any effect of task order. Figure 9: Resultant length changes (within-participants) across the three Su-soku blocks. Resultant length tended to decrease, suggesting that variability in the timing of the oscillatory peaks across blocks tended to increase.

The primary purpose of this paper was to investigate the effects of a single session of Su-Soku breathcounting on central and autonomic nervous system correlates of cognitive performance on an auditory oddball task, as well as effects on oscillatory and HRV indices of autonomic state during a period of eyes-open rest. We found no evidence to support our prediction that Su-soku would produce a greater increase in induced theta activity for Oddball Targets compared to Standards, relative to the control Timing task. Similarly, there was no evidence of a greater increase in oscillatory power in theta, nor in alpha or beta bands during the Eyes-Open Rest period following Su-soku compared to after the Timing task. However we did find evidence that HRV indices shifted more in the direction of parasympathetic dominance after Su-soku than after the Timing task. This effect was evident in an increase in cardiac slowing in in response to both stimulus types during the oddball task, as well as in HRV indices recorded during the Eyes-Open Rest period. These findings align with the notion that increased parasympathetic activity may support inhibitory control by modulating the balance of autonomic inputs during cognitive performance, and may be consistent with Thayer and Lane’s (2000) neurovisceral integration model, which posits that enhanced parasympathetically mediated HRV is associated with greater inhibitory control over prepotent responses. During Su-soku, FM oscillatory power in theta, alpha and beta bands were significantly elevated compared to the preceding Eyes-Open Rest period. Although there was only weak evidence of a reduction in theta-beta ratio, it is of note that there was no evidence of an increase, suggesting that the increase in theta power was not an indication of increased drowsiness, but rather of maintained cognitive control. Moreover HRV autonomic indices suggested a shift to more sympathetic nervous dominance during Su-soku than during the prior eyes-open rest period, indicating that effort was required to perform the Su-soku task. Interestingly, however, there was evidence between the second and final Su-soku blocks RMSSD tended to rise, which might be consistent with Kubota et al.’s (2001) finding of increased sympathetic and parasympathetic indices during periods of increased theta activity in participants performing the Su-soku task. Further analyses revealed that oscillatory amplitudes in theta, beta and alpha frequencies, systematically varied across the respiratory cycle. Oscillatory power in all three frequency bands tended to peak towards the end of exhalation. However, it is possible that the association between oscillatory power and respiratory phase may be at least in part due to cognitive effects of the pacer stimuli. It is of interest in this regard that the resultants tended to reduce across the three blocks, suggesting a loosening of the coupling between respiration and neural activity over time, which may possibly be related to reduced effort required to maintain the assigned breathing rate, and a reduced requirement to attend to the pacer. A few studies report that paced breathing, as opposed to spontaneous breathing, can influence cognition; which suggests that attending to internal signals via paced breathing could be the primary contributor to the effects reported in the current study (Laborde et al., 2022). Future studies into the synchronization of brain oscillations and the respiratory cycle during a Su-soku exercise that do not make use of a pacer stimulus, would shed light on this possibility, particularly if a respiratory belt is used to record the respiratory cycle. Other limitations of the study include the counter-balanced design and choice of comparator. In our design, the same probe tasks were treated as baseline for one intervention condition and outcome for the other. A study in which participants were randomized to either Su-soku or a non-active comparator would have allowed more confident conclusions. It would also have useful to have investigated effects lasting longer than the few minutes of task and rest immediately following the Su-soku session. A larger sample size would have allowed us to investigate whether effects differ depending on the participant’s comfortable breathing rate. Nonetheless, the findings from this study suggest that, even in naive meditators, enhancing interoceptive awareness through paced breathing exercises such as Su-soku modulates brain activity in a manner consistent with a calm-attentive state reflected in enhanced parasympathetic activity. The enhanced cardiac slowing following target stimuli during the subsequent Oddball Target detection task is encouraging, while the greater shift towards parasympathetic dominance in the following rest period indicates that the enhancement of parasympathetic activity associated with the Su-soku task persists in a period of rest beyond an intervening attention-demanding task. The systematic variation of the amplitude of theta, alpha and beta oscillations across the respiratory cycle during Su-soku is consistent with Thayer and Lane’s (2000) concept of neurovisceral integration.

1. Acharya, J. N., Hani, A., Cheek, J., Thirumala, P., & Tsuchida, T. N. (2016). American Clinical Neurophysiology Society Guideline 2: Guidelines for standard electrode position nomenclature. Journal of Clinical Neurophysiology, 33 (4), 308–311. https://doi.org/10.1097/WNP.0000000000000316 Basso, J. C., McHale, A., Ende, V., Oberlin, D. J., & Suzuki, W. A. (2019). Brief, daily meditation enhances attention, memory, mood, and emotional regulation in non-experienced meditators. Behavioural Brain Research, 356, 208–220. https://doi.org/10.1016/J.BBR.2018.08.023 Benjamin, D. J., & Berger, J. O. (2019). Three recommendations for improving the use of p‑values . The American Statistician, 73(S1), 186–191. https://doi.org/10.1080/00031305.2018.1543135 Berens, P, (2009). CircStat: A Matlab Toolbox for Circular Statistics, Journal of Statistical Software, 31(10),1-21. https://doi.org/10.18637/jss.v031.i10 Bradley, M. M., Keil, A., & Lang, P. J. (2012). Orienting and emotional perception: Facilitation, attenuation, and interference . Frontiers in Psychology, 3, Article 493. https://doi.org/10.3389/fpsyg.2012.00493 Britton, W. B., Lindahl, J. R., Cooper, D. J., Canby, N. K., & Palitsky, R. (2021). Defining and measuring meditation-related adverse effects in mindfulness-based programs. Clinical psychological science : a journal of the Association for Psychological Science, 9 (6), 1185–1204. https://doi.org/10.1177/2167702621996340 Chalmers, J. A., Quintana, D. S., Abbott, M. J.-A., & Kemp, A. H. (2014). Anxiety disorders are associated with reduced heart rate variability: A meta‑analysis. Frontiers in Psychiatry, 5 (80). https://doi.org/10.3389/fpsyt.2014.00080 Critchley, H. D., & Garfinkel, S. N. (2018). The influence of physiological signals on cognition. In Current Opinion in Behavioral Sciences (Vol. 19, pp. 13–18). Elsevier Ltd. https://doi.org/10.1016/j.cobeha.2017.08.014 Doege, K., Jansen, A., Mall, A., Wüstenberg, T., Juckel, G., & Gallinat, J. (2009). The modulation of prefrontal cortex activity by the Nogo task in schizophrenia—a near-infrared spectroscopy study. Psychophysiology, 46 (4), 681–688. https://doi.org/10.1111/j.1469-8986.2009.00785.x Garfinkel, S. N., Seth, A. K., Barrett, A. B., Suzuki, K., & Critchley, H. D. (2015). Knowing your own heart: Distinguishing interoceptive accuracy from interoceptive awareness. Biological Psychology, 104, 65–74. https://doi.org/10.1016/j.biopsycho.2014.11.004 Gratton, G., Coles, M. G., & Donchin, E. (1983). A new method for off-line removal of ocular artifact. Electroencephalography and Clinical Neurophysiology, 55 (4), 468–484. https://doi.org/10.1016/0013-4694(83)90135-9 Graham, F. K., & Clifton, R. K. (1966). Heart‑rate change as a component of the orienting response . Psychological Bulletin, 65 (5), 305–320. https://doi.org/10.1037/h0023258 Held, Leonhard, & Ott, Manuela. (2016). How the maximal evidence of P -values against point null hypotheses depends on sample size. The American Statistician, 70 (4), 335–341. https://doi.org/10.1080/00031305.2016.1209128 sciencedirect.com+12 Herrero, J. L., Khuvis, S., Yeagle, E., Cerf, M., & Mehta, A. D. (2018). Breathing above the brain stem: volitional control and attentional modulation in humans. 119 (1), 145–159. https://doi.org/10.1152/JN.00551.2017 Jansen, J. E., Gleeson, J., Bendall, S., Rice, S., & Alvarez-Jimenez, M. (2020). Acceptance-and mindfulness-based interventions for persons with psychosis: a systematic review and meta-analysis. Schizophrenia research, 215, 25-37. Komisaruk, B. R., & Frangos, E. (2022) . Vagus nerve afferent stimulation: Projection into the brain, reflexive physiological, perceptual, and behavioral responses, and clinical relevance. Autonomic Neuroscience, 237, Article 102908. https://doi.org/10.1016/j.autneu.2022.102908 Kubota, Y., Sato, W., Toichi, M., Murai, T., & Okada, T. (2001). Frontal midline theta rhythm is correlated with cardiac autonomic activities during the performance of an attention demanding meditation procedure . Brain Research. Cognitive Brain Research, 11 (2), 281–287. https://doi.org/10.1016/S0926-6410(00)00086-0 Laborde, S., Allen, M. S., Borges, U., Dosseville, F., Hosang, T. J., Iskra, M., Mosley, E., Salvotti, C., Spolverato, L., Zammit, N., Javelle, F., & others. (2022). Effects of voluntary slow breathing on heart rate and heart rate variability: A systematic review and a meta-analysis. Neuroscience & Biobehavioral Reviews, 138, Article 104711. https://doi.org/10.1016/j.neubiorev.2022.104711 Lomas, T., Ivtzan, I., & Fu, C. H. (2015). A systematic review of the neurophysiology of mindfulness on EEG oscillations. Neuroscience & Biobehavioral Reviews, 57, 401-410. Makeig, S., Jung, T. P., & Sejnowski, T. J. (2000). Awareness during drowsiness: dynamics and electrophysiological correlates. Canadian Journal of Experimental Psychology/Revue canadienne de psychologie expérimentale, 54 (4), 266. Mather, M., & Thayer, J. F. (2018). How heart rate variability affects emotion regulation brain networks. Current Opinion in Behavioral Sciences, 19, 98–104. https://doi.org/10.1016/j.cobeha.2017.12.017 Melnychuk, M. C., Robertson, I. H., Plini, E. R. G., & Dockree, P. M. (2021). A bridge between the breath and the brain: Synchronization of respiration, a pupillometric marker of the locus coeruleus, and an EEG marker of attentional control state. Brain Sciences, 11 (10), 1324. https://doi.org/10.3390/brainsci11101324 Natarajan, A. (2023). Heart rate variability during mindful breathing meditation. Frontiers in physiology, 13, 1017350. Pfurtscheller, G., & Lopes da Silva, F. H. (1999). Event-related EEG/MEG synchronization and desynchronization: Basic principles. Clinical Neurophysiology, 110 (11), 1842–1857. https://doi.org/10.1016/S1388-2457(99)00141-8 Putman, P., Verkuil, B., Arias-Garcia, E., Pantazi, I., & van Schie, C. (2014). EEG theta/beta ratio as a potential biomarker for attentional control and resilience against deleterious effects of stress on attention. Cognitive, Affective, & Behavioral Neuroscience, 14, 782-791. Rodrigues, M. F., Nardi, A. E., & Levitan, M. (2017). Mindfulness in mood and anxiety disorders: a review of the literature. Trends in psychiatry and psychotherapy, 39 (3), 207-215. Sargent, K. S., Martinez, E. L., Reed, A. C., Guha, A., Bartholomew, M. E., Diehl, C. K., Chang, C. S., Salama, S., Subotnik, K. L., Ventura, J., Nuechterlein, K. H., Miller, G. A., & Yee, C. M. (2024). Oscillatory coupling between neural and cardiac rhythms . Psychological Science, 35 ( …), 517–528. https://doi.org/10.1177/09567976241235932 Shaffer, F., & Ginsberg, J. P. (2017). An overview of heart rate variability metrics and norms. Frontiers in Public Health, 5, 258. https://doi.org/10.3389/fpubh.2017.00258 Stogios, N., Gdanski, A., Gerretsen, P., Chintoh, A. F., Graff-Guerrero, A., Rajji, T. K., Remington, G., Hahn, M. K., & Agarwal, S. M. (2021). Autonomic nervous system dysfunction in schizophrenia: Impact on cognitive and metabolic health. NPJ Schizophrenia, 7 (1), Article 22. https://doi.org/10.1038/s41537-021-00151-6 Thayer, J. F., & Lane, R. D. (2000). A model of neurovisceral integration in emotion regulation and dysregulation. Journal of Affective Disorders, 61 (3), 201–216. https://doi.org/10.1016/S0165-0327(00)00338-4 Toichi, M., Sugiura, T., Murai, T., & Sengoku, A. (1997). A new method of assessing cardiac autonomic function and its comparison with spectral analysis and coefficient of variation of R-R interval. Journal of the Autonomic Nervous System, 62 (1–2), 79–84. https://doi.org/10.1016/S0165-1838(96)00112-9 Wu, R., Liu, L. L., Zhu, H., Su, W. J., Cao, Z. Y., Zhong, S. Y., … & Jiang, C. L. (2019). Brief mindfulness meditation improves emotion processing. Frontiers in neuroscience, 13, 482990. https://doi.org/10.3389/fnins.2019.01074 Zeidan, F., Johnson, S. K., Diamond, B. J., David, Z., & Goolkasian, P. (2010). Mindfulness meditation improves cognition: Evidence of brief mental training. Consciousness and Cognition, 19 (2), 597–605. https://doi.org/10.1016/J.CONCOG.2010.03.014 Zelano, C., Jiang, H., Zhou, G., Arora, N., Schuele, S., Rosenow, J., & Gottfried, J. A. (2016). Nasal Respiration Entrains Human Limbic Oscillations and Modulates Cognitive Function. Journal of Neuroscience, 36 (49), 12448–12467. https://doi.org/10.1523/JNEUROSCI.2586-16.2016 Information & Authors Information Version history Copyright This work is licensed under a Non Exclusive No Reuse License. Authors Metrics & Citations Metrics Article Usage 217views 95downloads Citations Download citation Natasha A Elliott, Elizabeth B Liddle, Dawn M Walker, et al. Breathing in synchrony: Exploring neural-visceral effects in Su-soku breath counting. Authorea. 16 August 2025. DOI: https://doi.org/10.22541/au.175534414.40177581/v1 DOI: https://doi.org/10.22541/au.175534414.40177581/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.