Understanding The Roles Of Central And Autonomic Activity During Sleep in The Improvement Of Working Memory And Episodic Memory Part 2

Sleep Modulates ANS Activity

The transition from wake to sleep produces the largest shift in autonomic activity we experience every day. Sleep is not one uniform event, and its characterization into organized stages shows specific profiles in central and autonomic activity during each stage. Over a night of sleep, the human brain cycles through two primary phases: NREM and rapid eye movement (REM) sleep. 

In recent years, the connection between sleep and memory has attracted much attention from researchers. Research shows that sleep plays an important role in people's memory function.

Human sleep is divided into two states: REM (rapid eye movement sleep) and NREM (non-rapid eye movement sleep). Among them, NREM sleep is a very important stage, which includes four stages: one, two, and three.

During NREM sleep, the human brain strengthens memory through repeated stimulation and transfers information to long-term memory. Therefore, it is very important for people who want to improve their memory to get enough NREM sleep.

At the same time, NREM sleep can also promote the body's metabolism, help restore body functions, and enhance the body's immunity. Therefore, ensuring adequate NREM sleep is not only beneficial to improving memory but also a necessary condition for maintaining good health.

To ensure adequate NREM sleep, we should develop good sleep habits, such as controlling dinner time and reducing strenuous exercise and strenuous evening activities. In addition, relaxing before falling asleep can help us enter NREM sleep faster.

All in all, NREM sleep plays an indispensable and important role in human memory and physical health. Improving sleep quality and ensuring adequate NREM sleep will have a positive impact on our lives and work. Let us start now to develop healthy sleeping habits and welcome a better tomorrow! It can be seen that we need to improve our memory. Cistanche deserticola can significantly improve memory because Cistanche deserticola is a traditional Chinese medicinal material with many unique effects, one of which is to improve memory. The efficacy of minced meat comes from the various active ingredients it contains, including acid, polysaccharides, flavonoids, etc. These ingredients can promote brain health in various ways.

Click know 10 ways to improve memory

NREM sleep is further divided into stages 1, 2, and 3 (or slow-wave sleep) (53). Stage 1 sleep is a transitional state from wake to sleep, making up 3% of adult nocturnal sleep. About 60% of adult sleep is stage 2 sleep, which is marked by distinct electrophysiological events named sleep spindles and K-complexes. Stage 3, or slow-wave sleep (SWS), makes up about 20% of sleep, and it is marked by slow, high-amplitude oscillations called slow oscillations (SOs, <1 Hz) and slow-wave activity (SWA, 0.5 to 2 Hz). In NREM sleep, cholinergic systems in the brainstem and forebrain become markedly less active; firing rates of LC–NE and serotonergic Raphe neurons are also reduced, compared to waking levels
(54). 

However, the conventional dogma about the relative quiescence of LC–NE neurons during NREM sleep has been challenged by evidence of a transient increase in LC–NE activity during NREM sleep (55). A simultaneous electroencephalogram (EEG)–functional MRI (fMRI) study in humans further revealed that the increased phasic activity of the LC nucleus is temporally related to SO down-to-up transitions (55, 56), suggesting a more complex neuromodulator dynamic during NREM sleep. The activity of LC–NE neurons during NREM sleep is potentially relevant in understanding how autonomic activity during this sleep period may contribute to cognitive enhancement.

After a bout of SWS, the brain shifts into REM sleep, which makes up about 20% of human sleep and it is marked by sudden bursts of eye movements and faster, low-amplitude alpha (8 to 12 Hz) and theta (4 to 8 Hz) oscillations. During REM sleep, both aminergic populations are strongly inhibited, while cholinergic systems become more active compared to waking levels (57). The transition through stage 2, SWS, and REM occurs in 90- to 100-minute cycles across the night, with the first half of the night dominated by SWS and the second half of the night dominated by REM sleep (58).

In peripheral sites, the transition from wake to SWS is associated with a significant drop in heart rate and blood pressure, as well as increased dominance of HF-HRV (59, 60). The blood pressure plunge during NREM sleep compared to wake is beneficial for cardiovascular health, leading some experts to describe sleep as a “cardiovascular holiday” (61). SWS, in particular, is a period of cardiovascular quiescence and may represent an opportunity for the cardiovascular system to recuperate from daytime insults, such as stress-induced blood pressure surges. Indeed, one study comparing amounts of SWS and subclinical markers of cardiovascular disease found that participants who experienced greater SWS showed lower markers of cardiovascular disease after cardiovascular stress (62), suggesting that SWS may buffer autonomic responses to daytime stress that may modify disease risk.

Additionally, sleep, rather than circadian effects, appears to influence ANS activity, as similar HRV profiles have been shown in the daytime and nighttime sleep (63), which may also indicate that daytime naps serve as a mini cardiovascular holiday. Furthermore, a study comparing HRV profiles during a 50-minute nap versus waking rest in a supine position reported parasympathetic dominance during sleep only, and not during quiet rest, indicating that the cardiovascular benefits are specific to sleep (64).

Studies have revealed a consistent interdependency between the heart and brain activity, with temporally coincident changes in EEG delta (0.5 to 4 Hz) power and ANS activity (65–70). Modulations in HRV are so closely associated with the onset of SWS that they can be used as a parameter to automatically detect SWS (71). Furthermore, delta band power, a marker of homeostatic sleep drive that dissipates across successive NREM periods, shows inverse coupling with LF/(LF+HF) ratio during nighttime sleep. Generally, the LF/(LF+HF) ratio increases during REM sleep and decreases during SWS (Fig. 1A) (66), indicating greater sympathetic activity during REM sleep, with heart rate and blood pressure levels reaching values similar to wake (72). 

Parasympathetic/vagal activity during NREM sleep and sympathetic activity during REM sleep can exceed average levels of quiet wakefulness (72). Fig. 1B demonstrates the power spectrum of RR intervals during quiet wake, stage 2, SWS, and REM sleep (Fig. 1B) (73). More recently, causally increased SWA via acoustic SOs stimulation resulted in increased vagal activity (measured by HF-HRV and SDNN) during SWS compared to sham stimulation (74), suggesting a strong interdependency between vagal activity and slow EEG oscillatory events during SWS.

Considering more temporally precise levels of analysis, coupling has been shown between individual ANS and CNS events, such as heartbeats and EEG SOs in deep sleep (75). Using a cross-correlation approach, Thomas et al. (69) showed a temporal relation between SWA and high-frequency cardiopulmonary (0.1 to 0.4 Hz) coupling, an ECG-derived biomarker of stable sleep, during NREM sleep. Several studies have also reported on the coupling between autonomic and central events whereby short bursts of heart rate are temporally coincident with transient increases in SOs during NREM sleep(73, 76). Rembado et al. recorded vagal-evoked potentials, manifested as the vagal afferents to the cerebral cortex in responses to VNS, in macaque monkey brains during different consciousness states (77). Vagal-evoked potentials were reported to be 300 to 500% larger during NREM sleep, compared to REM sleep and wakefulness (Fig. 1C) (77), and, critically, vagal-evoked potentials during NREM were larger for stimuli delivered at the depolarized phase of ongoing delta oscillations, suggesting a close temporal coupling between ANS and CNS events. These findings demonstrate that CNS–ANS dynamics support the interdependency between cortical and cardiac function during sleep. Moreover, taken together with findings from wake HRV studies, natural surges in parasympathetic activity during SWS suggest that HRV profiles during sleep might account for some degree of cognitive enhancement.

Both SWS and Vagal Tone Benefit WM

Compared to wakefulness, sleep between WM training sessions may be critical for enhancing WM performance (78–82), potentially due to the effect of SOs (83–85) (SI Appendix, Fig. S2A). Moreover, a recent study using acoustic SO stimulation during nighttime sleep reported that stimulation improved WM as a result of enhanced SWA, compared to individuals whose SWA was not enhanced (86). Not all studies, however, find a positive association between EEG features of SWS and WM improvement (78, 81). One potential reason for the lack of consistent findings may in part be related to the fact that few studies measure ANS activity during sleep and therefore miss the ANS’s contribution to the performance change. One recent study accounted for both autonomic and central activity across a sleep period in which subjects’ WM was tested before and after a sleep period (78) (SI Appendix, Fig. S2B). In this study, participants performed the operational-span task in the morning and evening. During the interest period, subjects either experienced a nap or quiet wake with EEG and ECG recorded (78). Results showed that HF-HRV during SWS predicted WM improvement in healthy young adults, to a greater extent than HF-HRV during wakefulness or than other SWS features, including SOs or SWA. These findings suggest that along with EEG events of SWS, naturally elevated vagal activity during SWS also supports WM improvement in young adults.

Although the sleep-dependent mechanisms of WM improvements remain elusive, based on the current picture, SOs and vagal activity may facilitate WM improvement by increasing prefrontal efficiency and automaticity in prefrontal–subcortical inhibitory networks. Specifically, vagal inputs increase prefrontal phasic-NE activity while SOs up-regulate cortical synapse potentiation. More interestingly, Eschenko et al. showed that LC neurons fire during down-to-up transition, while medial prefrontal neurons fire at the peak of SOs (87), which corresponds to the time of maximal cortical excitability during up-states, suggesting that in prefrontal–LC interactions, LC leads the prefrontal activity. It has been hypothesized that SOs potentiate synapses that were depressed due to persistent activity during the previous day and that potentiation provides a physiological basis for sleep-dependent memory improvement (88). Taken together, optimal levels of NE during sleep might support next-day WM performance by potentiating WM prefrontal activity and thus enhancing prefrontal–striatal WM efficiency.

Another possibility is that SOs and vagal activity facilitate WM by clearance of tau/beta-amyloid plaques in the prefrontal–subcortical executive function networks, driven by phasic LC–NE activity (vagal afferents) and SOs (35). Even among individuals who perform within the normal range of cognitive functioning, mesial temporal tau accumulation is associated with worse cognitive performance (89). This study employed a standardized test to measure cognitive performance, which does not allow us to dissociate the impact of tau on WM per se. Further research is needed to understand whether glymphatic clearance is truly involved with day-to-day sleep-dependent WM improvements in the healthy brain, or whether other mechanisms may also be involved, including SO-mediated restoration of synaptic homeostasis (90) or increasing maximal synaptic efficacy (91).

We now turn to the question of LTM, for which ample evidence supports the role of SWS specifically in the consolidation of hippocampal-dependent episodic memories. Yet, evidence for a potential role of ANS activity during sleep in this process remains scarce, a gap that may have explanatory value.

CNS and ANS Contributions to Long-Term Memory

Hippocampal-dependent, episodic memory refers to the conscious recollection of information specific to the time and place of acquisition (92). A growing literature supports the role of sleep in the consolidation of episodic memories and has identified a critical role for specific electrophysiological events during NREM sleep (93). Although there is much debate as to how recent experiences are represented and transformed in cortical and subcortical long-term stores (94–96), there is a consensus that the hippocampus is a fast-learning system that binds recent experiences into representations across different cortical structures during encoding (93). 

During consolidation, repeated reactivation stabilizes and strengthens memory traces, with sleep being an optimal offline period for consolidation as it facilitates the dialogue between the hippocampus and the neocortex (93). Specifically, during NREM sleep, the memory trace is reactivated by hippocampal sharp-wave ripples nested within thalamic spindles, which are in turn nested within the down-to-up transition of the SO, providing a pathway for neural communication between neocortical and hippocampal cell assemblies. Spindles have recently been recognized as playing a causal role in hippocampal-dependent memory consolidation via pharmacology (97) and targeted memory reactivation (98). Mednick et al. (97) compared zolpidem, a short-acting GABA-A agonist, with a placebo and a positive control hypnotic (sodium oxybate) across a night of sleep. Compared with controls, zolpidem increased sleep spindles and enhanced hippocampal-dependent, episodic verbal memory, and the spindle boost mediated memory improvements. Other pharmacology studies have corroborated these findings (99), implicating GABAergic modulations of the thalamocortical network as important for LTM formation.

Despite the growing list of studies demonstrating a role for ANS activity in executive function and WM, links to LTM are sparsely reported, and no studies have endorsed a role for ANS during sleep in hippocampal-dependent episodic LTM specifically. One study showed that overnight improvement in nonhippocampal-dependent procedural memory was correlated with LF-HRV and SDNN during sleep (100). In addition, parasympathetic activity (HF-HRV) during REM sleep strongly predicted improvement in implicit priming in a creativity task (101). Studies examining the impact of HRV during wakefulness on episodic LTM also show mixed results. One study demonstrated that people with poor vagal autonomic functioning (low resting HRV) show greater false memory errors (102). In addition, the cardiac vagal tone has been shown to positively correlate with better memory for emotionally charged stimuli (103, 104), albeit with no relation with memory for neutral stimuli. In contrast, several studies showed that HRV during wakefulness does not predict episodic memory performance (105–107). Taken together, the emerging picture of the role of CNS and ANS inputs for cognitive enhancement is that SWS is an optimal brain state for the stabilization of episodic, long-term, nonemotional memories, as well as for the improvement of executive function, but not necessarily via the same mechanisms.

How to Do Working And Long-Term Memory Interact During Sleep?
While studies have shown that both SWS (including SWA and SOs) and vagal activity during SWS contribute to WM and that SO–spindle-ripple complexes contribute to episodic LTM, the relation between WM and episodic LTM remains unclear. On one hand, studies have shown positive associations between WM and LTM, such that WM increases LTM recognition and WM capacity constrains LTM encoding (108). On the other hand, Hoskin et al. demonstrated that episodic memory reactivation during wake intrudes on WM maintenance (109). Specifically, participants learned word lists, and they either got tested immediately after encoding (short-term WM) or performed a distractor task during which the episodic memory was reinstated (LTM reactivation). They showed that the delay period introduced additional information that negatively influenced WM maintenance. This suggests coordinated activity patterns across a broad swath of cortical regions, including the prefrontal cortex, triggered by memory reactivation may steal resources from WM maintenance. Thus, WM and episodic memory may be supported by separate and potentially competitive neural mechanisms, namely, the LC–NE prefrontal–subcortical network and the GABAergic thalamocortical hippocampal network, respectively.

At the neuromodulatory level, along with LC–NE enhancement of WM (VNS studies), animal studies have implicated this system during and immediately following encoding novel experiences (110, 111) and in wake-dependent gene expression regulating synaptic potentiation that supports learning (112). However, while reversible inactivation of LC during the Morris water maze task demonstrated significant impairments in spatial memory encoding and WM, consolidation, and retention of spatial memory were not affected (113). Together, these findings suggest that the LC–NE system may play an important role in the early acquisition of new experiences and cognitive control efficiency, but not in the consolidation and retrieval of LTM.

In humans, recent findings in the tVNS literature have corroborated a selective functional role of the LC–NE system in cognitive control, but not in LTM. A meta-analysis of 19 tVNS studies (114) showed significant effects of acute tVNS on cognitive inhibitory control, particularly as task difficulty increases, but no evidence supporting the effectiveness of tVNS on LTM performance, attention, or other cognitive domains. Specifically, Mertens et al. (115) found that tVNS did not affect either immediate or delayed word recognition memory in young and middle-aged adults. Furthermore, using pharmacology, Lozano-Soldevilla et al. (116) administered a GABAergic benzodiazepine (lorazepam) to healthy adults and reported dose-dependent decreases in WM, driven by decreased gamma and alpha power, suggesting an antagonistic relation between GABA and LC–NE prefrontal–subcortical networks.

Competition between these networks may be especially prevalent during offline sleep (117, 118). During hippocampal ripples, signatures of LTM replay, and consolidation, Logothetis et al. (117) demonstrated deactivations in brainstem regions regulating the ANS. These fascinating results may mean that during sleep-dependent memory consolidation, ripple/spindle complexes may orchestrate a privileged interaction state between hippocampus and cortex by silencing the output from diencephalic, midbrain, and brainstem regions (SI Appendix, Fig. S3). Interestingly, the deactivation of the basal ganglia, the pontine region and the cerebellar cortex, is consistent with prior evidence of competition between episodic and procedural memory systems (119). In the reverse direction, Novitskaya et al. (118) experimentally increased NE by phasic-LC stimulation and blocked the generation of ripple-associated cortical spindles, thus interfering with spatial LTM consolidation. Moreover, Marzo et al. (120) have shown that phasic stimulation of LC in anesthetized rats transiently suppressed spindles while evoking sustained spiking in the medial prefrontal cortex that resembled NE-dependent prefrontal activity during the delay period of WM tasks. It is also well documented that NE input shifts the thalamocortical network from a synchronized state associated with SOs and spindles to a desynchronized state characterized by increased neuronal responsiveness to synaptic inputs, which is more optimal for encoding and sensory processing (121).

Thus, a bidirectional, mutual antagonism between thalamic–hippocampal–cortical and prefrontal–subcortical–autonomic systems during NREM might underlie the reported behavioral trade-off between episodic LTM and WM. To test this possibility, Chen et al. used a GABA-A agonist, zolpidem, to investigate the directionality of the interplay between the vagal and spindle systems during overnight sleep and their effect on WM improvement and episodic memory consolidation (122). Along with enhancing spindle activity and episodic LTM consolidation, zolpidem also showed vagal suppression and less WM improvement compared with placebo (SI Appendix, Fig. S4A). Vagal activity during SWS positively correlated with WM improvement (SI Appendix, Fig. S4B) but negatively correlated with LTM retention (SI Appendix, Fig. S4C). Effective connectivity estimation (123, 124) demonstrated that cortical oscillations in the spindle range showed causal suppression over vagal autonomic activity both in the zolpidem and placebo conditions, the magnitude of which was associated with a trade-off between enhanced episodic LTM at the cost of reduced WM improvement. Additionally, spindle–SO coupling was associated with a similar behavioral trade-off, with greater spindle–SO coupling associated with less WM improvement. These findings are consistent with the notion that GABAergic hippocampal– thalamocortical communication and LC–NE frontal–subcortical– autonomic communication compete for resources during SWS sleep and that this interaction can be biased toward LTM consolidation by increasing spindles/sigma, in this case pharmacologically, and presumably other methods would have predictable effects. For example, we hypothesize that VNS would bias the interaction in the opposite direction by enhancing vagal activity and WM improvement at the expense of spindles coupled with SOs and LTM performance.

The Slow Oscillation Switch Model

The Slow Oscillation Switch Model (Fig. 2) proposes that the brain switches between separate and nonoverlapping During the LTM state, consolidation occurs via spindle-coupled SOs, which leads to increased hippocampal–thalamocortical communication, increased ripple–spindle complexes, while reducing the prefrontal–autonomic inhibitory networks. During the WM state, SOs without spindles promote greater neural efficiency within the prefrontal–subcortical–autonomic network. The competitive dynamics between these networks are theoretically guided by animal studies showing antagonistic relations between brain regions that regulate autonomic activity versus memory replay (117, 118) and evidence for a periodic switch between spindles and autonomic activity (126).

We further hypothesize that spindles may act as a gating mechanism that regulates SO resources for other processes. Given that ∼20% of SOs during NREM are spindle-coupled (127), this leaves plenty of resources to be divided among other processes. Recent work analyzing the spatiotemporal characteristics of SOs has established that they can be categorized into three types, global, frontal, and local SOs, with global SOs more often co-occurring with spindles (127). Global SOs have been shown to support long-range communication, increase spindle coupling, and predict LTM consolidation (128), whereas frontal SOs have lower amplitude than global SOs, spread only within the frontal cortex, and may be promising candidates for modifying prefrontal–subcortical–autonomic networks that support WM. We hypothesize that frontal SOs may also be a carrying wave for restorative processes that facilitate prefrontal glymphatic clearance, as studies have reported that SOs are temporally coupled with and precede cerebrospinal glymphatic clearance (129, 130), but these studies have not distinguished between SO categories. Frontal SOs may regulate vagal activity, thereby promoting phasic LC neuron firing and tau clearance (35). A recent study showed that auditory stimulation not only boosts frontal SOs but also increases vagal HRV (131). Together, vagal activity might increase prefrontal efficiency and automaticity of WM functional networks by glymphatic clearance that preferably occurs during frontal SOs. These speculations are experimentally testable, and more research is needed to further elucidate sleep’s complex and potentially competitive, dynamics that support a wide range of cognitive domains.

Concluding Remarks and Future Directions

WM is the ability to hold a small amount of information active and relevant for a short amount of time, whereas episodic memory is a seemingly unlimited bank of autobiographical experiences, each of which can be explicitly evoked. Sleep, specifically brain activity during SWS, has been shown to influence both types of memory processes. Additionally, vagal activity measured by HF-HRV is associated with WM, while studies investigating the effect of ANS on episodic memory yield inconsistent results. The mechanism underlying how CNS and ANS activity during sleep coordinate to facilitate both types of cognitive processes is unknown. To bridge this gap, we present the Slow Oscillation Switch Model, which states that sleep is a competitive arena in which episodic LTM and WM vie for limited resources. Although this model is thought to occur naturally, the inner workings of the system are revealed experimentally in studies that amplify one side of the trade-off (e.g., pharmacologically increasing spindles or ripple-triggered stimulations of LC) with predictable reductions in vagal activity and postsleep WM performance or ripple/spindle coupling and LTM.

If sleep acts like a switch that toggles between the LC–NE prefrontal–subcortical autonomic processes and the GABAergic thalamocortical–hippocampal replay, one question is what determines the priority of the switch mechanism. We consider two possibilities—a natural periodic switch versus an experience-dependent bias. Lecci et al. (126) demonstrated a periodic alternating pattern between spindle bursts and heart rate accelerations, occurring every 50 s, supporting a possibility that the Slow Oscillation Switch mechanisms alternate periodically under tonic conditions. Alternatively, learning, emotional experiences, novelty, or cognitive load might determine prioritization. Consistent with this idea, more demanding memory tasks show a greater number of spindles during subsequent sleep (132). Similarly, intensive WM training can increase frontal SOs and vagal activity to a higher degree, compared to less intense WM training (22, 84). Future studies investigating how our brain and body coordinate to control this switch would allow further understanding of the competitive dynamics between different memory domains.

WM has traditionally been considered an unmodifiable trait, yet WM training studies have demonstrated that executive function in general and WM specifically does improve (20), and recent studies show that sleep supports this improvement (78, 82). The underlying mechanisms of this benefit, however, are unclear. Furthermore, we know little about how prefrontal–subcortical autonomic networks might coordinate SOs and vagal activity to facilitate WM. One study suggested an association between cognitive control, vagal activity, and automaticity in the prefrontal–subcortical autonomic networks. Lin et al. showed that cognitive training decreased functional connectivity in the bilateral striatum–prefrontal networks while increasing vagal activity, thereby facilitating performance in trained and untrained tasks, with fewer resources needed for successful cognitive inhibitory control (22). However, how such dynamics are modulated during sleep or SOs remains unexplored. Future neural imaging studies with simultaneous EEG–fMRI are crucial to allow an understanding of the neural mechanisms underlying WM plasticity during specific sleep events.

Even though significant progress has been made over the past decade, there is still much to understand about the role of sleep in different cognitive domains. Here, we present a scenario in which episodic LTM and WM are supported by separate circuitry that vie for limited resources during sleep. Importantly, we highlight the potential that SOs could be further divided into subcategories implicated in different functions, as electrophysiological events that share the same frequency may have separate functions, a possibility recently explored by Ngo and colleagues, showing a functional dissociation between delta and SOs, with delta waves facilitating forgetting whereas SOs are more likely to couple with spindles and facilitating episodic LTM consolidation (133, 134). We propose that identifying autonomic-central biomarkers during sleep for different cognitive processes and understanding their competitive dynamics may facilitate novel insights into memory models and provide new targets to combat neurodegenerative disease.

Data, Materials, and Software Availability.

There are no data underlying this work.

ACKNOWLEDGMENTS.

This work was supported by NIH Grant R01-AG062288 to S.C.M.

Reference

1. M. A. Hagenaars, M. Oitzl, K. Roelofs, Updating freeze: Aligning animal and human research. Neurosci. Biobehav. Rev. 47, 165–176 (2014). 

2. K. Miki, M. Yoshimoto, Role of differential changes in sympathetic nerve activity in the preparatory adjustments of cardiovascular functions during freezing behavior in rats. Exp. Physiol. 95, 56–60 (2010). 

3. S. E. Taylor et al., Biobehavioral responses to stress in females: Tend-and-befriend, not fight-or-flight. Psychol. Rev. 107, 411–429 (2000). 

4. L. N. Whitehurst, A. Subramoniam, A. Krystal, A. A. Prather, Links between the brain and body during sleep: Implications for memory processing. Trends Neurosci. 45, 212–223 (2022). 

5. S. Breit, A. Kupferberg, G. Rogler, G. Hasler, Vagus nerve as a modulator of the brain-gut axis in psychiatric and inflammatory disorders. Front. Psychiatry 9, 44 (2018). 

6. M. Kalia, J. M. Sullivan, Brainstem projections of sensory and motor components of the vagus nerve in the rat. J. Comp. Neurol. 211, 248–265 (1982). 

7. K. K. Sumal, W. W. Blessing, T. H. Joh, D. J. Reis, V. M. Pickel, Synaptic interaction of vagal afferents and catecholaminergic neurons in the rat nucleus tractus solitarius. Brain Res. 277, 31–40 (1983). 

8. F. Shaffer, R. McCraty, C. L. Zerr, A healthy heart is not a metronome: An integrative review of the heart’s anatomy and heart rate variability. Front. Psychol. 5, 1040 (2014). 

9. J. F. Thayer, R. D. Lane, Claude Bernard, and the heart-brain connection: Further elaboration of a model of neurovisceral integration. Neurosci. Biobehav. Rev. 33, 81–88 (2009). 

10. R. E. Kleiger, P. K. Stein, J. T. Bigger, Jr, Heart rate variability: Measurement and clinical utility. Ann. Noninvasive Electrocardiol. 10, 88–101 (2005). 

11. S. Laborde, E. Mosley, J. F. Thayer, Heart rate variability and cardiac vagal tone in psychophysiological research - recommendations for experiment planning, data analysis, and data reporting. Front. Psychol. 8, 213 (2017). 

12. M. Malik, Heart rate variability. Ann. Noninvasive Electrocardiol. 1, 151–181 (1996). 

13. V. Menon, M. D’Esposito, The role of PFC networks in cognitive control and executive function. Neuropsychopharmacology 47, 90–103 (2021). 

14. A. Baddeley, Working memory. Science 255, 556–559 (1992). 

15. T. S. Braver, J. R. Gray, G. C. Burgess, “Explaining the many varieties of working memory variation: Dual mechanisms of cognitive control” in Variation in Working Memory, A. Conway, C. Jarrold, M. Kane, A. Miyake, J. Towse, Eds. (Oxford University Press, 2007), pp. 76–106.

For more information:1950477648nn@gmail.com