Working memory (WM), which is a core component of higher cognitive functions, is vital for complex mental abilities, including problem solving, reasoning, and learning (Baddeley and Hitch, 1974; Chiesa et al., 2011). Changes in WM are associated with normal neurocognitive aging and various psychiatric disorders, such as schizophrenia, attention-deficit/hyperactivity disorder (ADHD), and Alzheimer’s disease (Karatekin and Asarnow, 1998; Park and Reuter-Lorenz, 2009; Grady, 2012; van Ewijk et al., 2015; Manoach and Stickgold, 2019). In prior studies, drugs (Zhao et al., 2019) and behavioral interventions (Hargreaves et al., 2015; Ackermann et al., 2018) have been used to improve WM. In recent years, non-invasive neuromodulation techniques, such as transcranial direct current stimulation (tDCS) (Andrews et al., 2011; Zivanovic et al., 2021), transcranial alternative current stimulation (tACS) (Benussi et al., 2021), transcranial random noise stimulation (tRNS) (Murphy et al., 2020), and transcranial magnetic stimulation (TMS) (Hulst et al., 2017), have gradually become mainstream clinical treatment approaches for cognitive modulation. Of the various neuromodulation methods available, our group has mainly focused on the effects of transcutaneous auricular vagus nerve stimulation (taVNS), a peripheral nerve stimulation technique, on cognition—especially WM performance. This method offers three main benefits. Firstly, taVNS is cost effective, low risk, convenient and acceptable for most people (Badran et al., 2019). Secondly, a potential mechanism explaining the influence of taVNS on cognitive functions has been proposed, since the vagus nerves are intimately linked to perception and regulation of the central nervous system (CNS) with “bottom-up” functions in cognition and clinical disorders (Aniwattanapong et al., 2022). Some brain imagining studies further have shown that taVNS influences a range of cortical and subcortical regions important for cognitive functions, including the contralateral post-central gyrus, bilateral insula, frontal cortex, right operculum, left cerebellum, insula, hippocampus, amygdala, and thalamus (Yakunina et al., 2017; Badran et al., 2018). Thirdly, several studies have reported that taVNS improves several cognitive abilities (Beste et al., 2016; Colzato et al., 2018; Fischer et al., 2018; Adair et al., 2020; Neuser et al., 2020). In the last 2 years, our team has confirmed the positive effects of taVNS on WM performance in healthy cohorts (Sun et al., 2021a; Zhao et al., 2022).

Recent innovations in medicine and wellness are rediscovering and validating various practices such as yoga, meditation, and breathing techniques from traditional cultures that may confer physiological and psychological benefits (Carter and Carter, 2016; Niu et al., 2019). Among these, breath-based practices, especially slow-paced breath (SPB)-based techniques, a major component of Eastern practices, have attracted attention. These practices have gradually become “Westernized” and are now routinely suggested for better health and emotion, and cognitive functioning (Brown and Gerbarg, 2005). For example, breath qigong, a traditional Chinese exercise, is considered a form of mind-body medicine that coordinates gentle exercise with relaxation through slow breathing and meditation. Recent finding support that breath qigong improves cognitive performance in patients with vascular cognitive impairment (Niu et al., 2019). Moreover, slow and rhythmic breath-based meditation (e.g., Sudarshan Kriya) has shown significant effects on reducing post-traumatic stress disorder (PTSD) symptoms, hyperarousal, and anxiety in United States military combat veterans (Seppala et al., 2014; Schulz-Heik et al., 2022), and breath-based mindfulness has been shown to have a positive influence on participants’ cognitive performance—especially WM performance (Mrazek et al., 2012; Quach et al., 2016). Even a simple slow breathing intervention can improve participants’ heart rate variability (HRV) and multitasking test performance (Bonomini et al., 2020). Taken together, this evidence suggests that slow rhythmic breathing has the potential to restore or improve cognitive function. However, some studies have reported that the influence of SPB depends on long-term training and that a single intervention did not a induce significant effect (Goldberg et al., 2021; Quek et al., 2021). Thus, it is important to explore the immediate effect of SPB intervention and further improve the effects of SPB.

Investigations into the physiological benefits of SPB have uncovered significant effects on the respiratory, cardiovascular, cardiorespiratory, and autonomic nervous systems (Russo et al., 2017). Numerous studies suggest that a slow breathing rate of around 6 breaths/min leads to cardiorespiratory coupling, which can induce greater arterial oxygen saturation, hemodynamic fluctuations, and respiratory sinus arrhythmia (RSA) sensitivity, compared to breathing at a typical rate (Bernardi et al., 1998; Chang et al., 2013; Ben-Tal et al., 2014; Li et al., 2018). These changes may have a broad influence on both physical state and mental functions. Furthermore, both arms of the autonomic nervous system, i.e., the parasympathetic and sympathetic nervous systems, are under control of the central respiratory centers. Specifically, autonomic outflows are inhibited during inspiration and disinhibited during expiration, which is termed respiratory gate theory (Badra et al., 2001; Eckberg, 2003). Prior studies have found that the stimulation effects of expiration on sympathetic bursts are significantly smaller than those on parasympathetic action, and that the greatest parasympathetic activity occurs at a low respiratory frequency (about 6 breaths/min) (Eckberg, 2003; Elghozi and Julien, 2007; Wang et al., 2013); in the other words, vagal nerves are the primary parasympathetic efferent activated by SPB. Expiration is thus important for maintaining a healthy physical and mental state in humans and may have a similar influence as taVNS. Moreover, since changes in autonomic afferent activity are believed to be the primary mechanism of SPB’s effects on psychophysiological state (Noble and Hochman, 2019), SPB and taVNS may interact in the CNS, where signals from both autonomic afferents and the auricular branch of the vagus nerves are processed and integrated (Baekey et al., 2010), leading to larger modulative effects on cognitive performance—especially WM performance. Considering the unique effects of taVNS and SPB on cognition and their potential interaction, employing them simultaneously may lead to greater improvement in WM performance, i.e., a synergistic effect. Of note, a recent review commented that combining taVNS with SPB might produce positive effects via both interventions, which is certainly worth exploring (Szulczewski, 2022).

To the best of our knowledge, no prior studies have explored the role of combined taVNS and SPB on cognitive functions, specifically WM. Thus, the current study used four intervention conditions, i.e., taVNS, SPB, taVNS + SPB, and a sham condition, to ascertain whether taVNS and SPB modulate WM performance in healthy adults and lead to a syncretic effect.

Previous studies have revealed that both taVNS and SPB have positive modulative effects on participants’ cognitive performance, especially WM capacity (Carter and Carter, 2016; Niu et al., 2019; Sun et al., 2021a; Zhao et al., 2022). These two interventions may exert their influence on participants’ physical health and mental state via autonomic nerves, especially the parasympathetic nervous system, i.e., vagal nerves. Thus, it is worth investigating the synergistic effects of these two interventions on WM. To do so, this study employed a within-subject design with four conditions (i.e., taVNS, SPB, taVNS + SPB, and sham) in a large sample. The results confirmed the ability of taVNS to facilitate spatial 3-back accuracy, as reported in our previous studies (Sun et al., 2021a; Zhao et al., 2022). However, SPB did not improve participants’ WM performance and there was no significant difference between the taVNS and taVNS + SPB interventions. To the best of our knowledge, this is the first study to directly compare the modulation efficiency of taVNS, taVNS + SPB and SPB interventions on WM performance. In the following sections, we discuss the potential reasons underlying the lack of effect of SPB and the limited synergistic effects of taVNS and SPB. We also propose some suggestions for practice and future research into the combined taVNS and SPB neuromodulation technique.

Recent studies have suggested that breathing at a rate of approximately 10 s/breath can significantly improve respiratory, cardiovascular, cardiorespiratory and autonomic nervous system functions (Russo et al., 2017), thereby regulating emotions and cognition (Carter and Carter, 2016; Mather and Thayer, 2018). However, more than one prior study has reported that a single SPB intervention did not influence WM performance (Goldberg et al., 2021; Quek et al., 2021). However, participants who maintained breath training for at least 3 months showed significant improvements in WM performance (Bonomini et al., 2020; Quek et al., 2021). Physiological studies provide further evidence that SPB can regulate the autonomic nervous system, especially parasympathetic nerves, i.e., vagal nerves. To achieve a long-term shift toward parasympathetic dominance, prolonged SPB practice is necessary, as observed in a previous study that required healthy humans to practice SPB regularly for 3 months (Pal et al., 2004). These results suggest that the effects of SPB on WM accumulate over time. Likely for this reason, long-term training based on intentionally decreasing the respiratory rate has existed throughout history in Chinese (Roth, 1997), Japanese (Komori, 2018), Indian (Saoji et al., 2019), and European (Bernardi et al., 2001) cultures. Today, many cognitive-behavioral therapies (CBT) include a slow breathing component, such as HRV biofeedback and mindfulness training, that lasts for about 3 months (Lehrer et al., 2020). However, the current study did not find any significant improvements in the SPB intervention group, consistent with existing theory. To improve participants’ cognitive functions, future studies should employ a regular SPB training paradigm.

To investigate the neuromechanism underlying these neuromodulation techniques and further enhance efficiency, numerous studies have tried to combine interventions. For instance, anodal tDCS combined with repetitive peripheral nerve stimulation (rPNS) has been shown to promote motor hand recovery in stroke patients (Sattler et al., 2015). In another study, tDCS combined with tRNS, which is another method of transcranial stimulation, exhibited better facilitation of WM (Murphy et al., 2020). Synchronized intermittent theta burst stimulation (iTBS) and invasive vagus nerve stimulation (VNS) have been shown to be safe, feasible, and potentially effective depression treatments (George et al., 2020). In a recent study, we demonstrated that combined taVNS with tDCS evoked greater activation in a range of cortical and subcortical regions (Sun et al., 2021b), and increased the consistency of WM improvements in healthy participants (Zhao et al., 2022). Since SPB has been shown to have various benefits for physical and mental health and is growing more popular in western society, many researchers have tried to combine it with neuromodulation techniques. For example, Schlatter et al. (2021) found that combined tDCS with HRV biofeedback (breath at 0.1 frequency) significantly reduced participants’ subjective stress. An insightful review pointed out that, considering the similarity of the mechanism and effects of taVNS and SPB, and their possible interactions at the level of the central autonomic network, this kind of hybrid intervention might have potential applications in both clinical and non-clinical areas (Szulczewski, 2022). However, as the various neuromodulation techniques involve complex mechanisms, their combination might induce antagonistic effects (Schabrun et al., 2013). Although a single SPB intervention did not increase participants’ WM performance in the present study, there were no antagonistic effects with taVNS in the taVNS + SPB group, indicating that the underlying mechanism of SPB is not in conflict with taVNS. Thus, although the results of the current study did not reveal a synergistic effect of combined taVNS and SPB interventions, considering the positive influence of SPB on participants’ autonomic system activity (Carter and Carter, 2016; Russo et al., 2017; Niu et al., 2019) and improvements in cognitive performance (Mrazek et al., 2012; Seppala et al., 2014; Quach et al., 2016; Bonomini et al., 2020; Schulz-Heik et al., 2022), this kind of combined intervention (i.e., taVNS + SPB) is still worth investigating in the future. A new combination technique—namely combined respiratory-gated taVNS (RAVNS) with SPB and an intervention-sensitive participant cohort, such as clinical or chronic disease patients—may be two breakthroughs in this area. Furthermore, collecting physiological or neurophysiological indicators, like HRV or electroencephalogram (EEG) data, may be important for quantifying the effects and understanding the underlying mechanism.

Consistent with the effects of breath on the autonomic nervous system, i.e., respiratory gate theory, recent studies have demonstrated that respiratory-gated taVNS, i.e., exhalation-locked taVNS, exerts cardiovagal modulative effects (Garcia et al., 2017; Sclocco et al., 2019) because the nucleus tractus solitarii (NTS), as the primary synaptic target of afference over the vagus nerve, receives an inhibitory influence during inhalation and a facilitatory influence during exhalation (Miyazaki et al., 1998, 1999; Baekey et al., 2010). Several brain imaging studies have confirmed the RAVNS effects, namely that exhalation-locked taVNS can lead to larger regulation of brainstem activity and cardiovagal modulation than inhalation-locked taVNS (Sclocco et al., 2019). Notable, a few of these studies combined RAVNS with a slow respiratory frequency. To date, two studies have examined taVNS with SPB (Frokjaer et al., 2016; Juel et al., 2017), and one combined RAVNS with SPB (Keute et al., 2021); all of these studies reported an increment in HRV. Therefore, one reason for the lack of synergistic effects of taVNS and SPB in this study may be that the independent taVNS intervention without the respiratory phase locked limits interaction and cooperation SPB. To further explore the synergistic effect of taVNS and SPB, both expiration-locked taVNS, i.e., RAVNS, and respiratory frequency should be considered in future studies.

Slow-paced breathing, as one kind of CBT intervention, influences the respiratory system (Bernardi et al., 1998; Nattie and Li, 2012), cardiovascular system (Hsieh et al., 2003; Dick et al., 2014; Ovadia-Blechman et al., 2017), and autonomic nervous system (Lopes and Palmer, 1976; Badra et al., 2001; Eckberg, 2003). Given the complex processes, it may take a long time for effects to accumulate and influence advanced cognitive abilities. This might restrict the instantaneous effects of combined SPB and taVNS intervention on WM performance. Besides, for healthy participants, most of their physiological indicators are normal and stable, which makes it hard for them to reap benefits from the gentle changes in the cardiovascular system and/or autonomic nervous system induced by SPB. In contrast, previous studies suggest that participants suffering from diseases or bad emotion states, such as chronic pain (Jafari et al., 2017; Lehrer et al., 2020; Reneau, 2020), cardiovascular diseases (Russo et al., 2017; Zou et al., 2017; Lehrer et al., 2020), anxiety, depression or stress (Goessl et al., 2017; Zaccaro et al., 2018; Pinter et al., 2019; Lehrer et al., 2020) were sensitive to the effects of SPB. Therefore, to further investigate the synergistic effects of taVNS and SPB, it may be important to recruit different participant cohorts—especially for clinical patients—and evaluate various behavioral measures that include, but are not limited to, ECG, electroencephalogram, emotional tasks, and cognitive tests.

With a growing body of studies supporting the cognitive modulatory effects of taVNS over the last 10 years, many have focused on the underlying mechanisms. However, the detailed processes and mechanisms remain unclear. Regulation of a range of cortical and subcortical regions, including the thalamus, hippocampus, amygdala, and insula (Goehler et al., 2000; Saper, 2002), as well as the ability to activate the locus coeruleus (LC) and cholinergic neurons in the nucleus basalis to further release noradrenaline (NE) and acetylcholine (Hassert et al., 2004; Roosevelt et al., 2006; Nichols et al., 2011), are two plausible mechanisms of taVNS’s cognitive modulatory effects. However, there are likely many undiscovered mechanisms. As a team that has been focusing on taVNS-based WM performance modulation (Sun et al., 2021a,b; Zhao et al., 2023), we have investigated the synergistic effects of taVNS and tDCS and found significant improvement in the modulation effects (Sun et al., 2021a,b). Therefore, the current study explored the synergistic effects of taVNS and SPB on WM performance and we intend to continue investigating their combined effects. We believe that combined taVNS with other neuromodulation techniques or CBT interventions will have great potential for cognitive modulation.

Specifically, we found that the taVNS and taVNS + SPB interventions specifically improved mACC in the spatial 3-back task, but not the spatial 1-back or digit 3-back tasks. This phenomenon has been replicated in more than 300 participants in our previous studies (Sun et al., 2021a; Zhao et al., 2022, 2023), suggesting there may be some special mechanisms. A meta-analysis suggested that the verbal n-back, like the digit n-back task, was associated with enhanced activation in the left ventrolateral prefrontal cortex, whereas the non-verbal location n-back task, like the spatial n-back task, was associated with enhanced activation in a set of regions described as the spatial attention network, which includes the right dorsolateral prefrontal, lateral premotor, and posterior parietal cortex (Owen et al., 2005). Some studies have found that taVNS stimulation can improve selective attention (Sun et al., 2017), which may affect the spatial attention network and contribute to the effect of improved taVNS on spatial WM performance. Participants’ baseline performance may be another influential factor. The effect of modulation of electric field on WM depends on baseline performance (Assecondi et al., 2021). As a result, individuals or tasks with a lower baseline outcome, like the spatial 3-back task, are more likely to show greater improvement. To further understand this phenomenon, more research—especially brain function studies—are required in the future.

Lastly, there are some limitations of the current study. Firstly, although there is a clear physiological rationale for the relationship among taVNS, SPB and the autonomic nerves system, the current study did not test the exact changes of participants autonomic nervous activity, which makes it difficult to draw a conclusion about the changes under the combined interventions of taVNSand SPB and the underlying mechanism of them on WM performance. Thus, the specific mechanisms for taVNS, SPB, and their combined intervention still need research with neurophysiologic or biological indicators to explain in the future. Secondly, to continue exploring a more efficient taVNS technique for WM capacity modulation, we focused on the effects of taVNS, SPB and taVNS + SPB on WM performance. However, both taVNS and SPB influence a wide range of physiological and psychological functions, such as emotion regulation (Strigo and Craig, 2016; Russo et al., 2017; Zaccaro et al., 2018) and pain management (Hassett et al., 2007; Berry et al., 2014; Chakravarthy et al., 2015). Thus, taVNS combined with SPB may have a large potential in these areas. Thirdly, since this study focused on a healthy cohort, it is unknown whether these changes can predict the therapeutic effects of taVNS + SPB-based treatment in different patient groups. Thus, more research is urgently needed to evaluate clinical translation of this technique.

In summary, informed by previous research, this study concluded that the effects of taVNS on WM accuracy—especially spatial WM with high cognitive loads—were stable and could be replicated in different cohorts. Considering that, taVNS is convenient, easy to administer, and has fewer side-effects, it may be an effective WM modulation technique and thus of important clinical significance. SPB is also a convenient, popular, and effective intervention. Combining these interventions may have a larger effect for cognitive modulation. However, facilitation of SPB and its synergistic effect with taVNS on WM performance was not observed in this. However, we view this study as the beginning of research on this new synergistic technique, not the end. Further research combining RAVNS with SPB using different participant cohorts is required in the future.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

This study was approved by the Institutional Research Ethics Committee of Xijing Hospital of the Air Force Medical University. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

J-BS, RZ, C-ZT, N-GX, and WQ conceived the study. RZ, Q-QT, HD, X-JY, L-ML, and J-BS designed the study. Z-XY, P-HL, and Y-PC acquired the data. RZ, Q-QT, Z-XY, P-HL, and J-BS analyzed the data. RZ, Q-QT, CC, X-JY, and J-BS drafted the manuscript. All authors revised the manuscript, read, and approved the submitted version.