The World Health Organization (WHO) reports that “depression is a leading cause of disability worldwide” and the condition affects at least 322 million people. Even more surprising is that 10 million people with depression have thought about suicide and 3 million have actual suicide plans. Those who actually seek help are diagnosed through the administration of questionnaires and subjective interviews. To date no specific physiological or biochemical markers are considered in current clinical practice, and no physiological measurement objectively distinguishes among the different subtypes of depression. Economically-speaking, the cost of lost productivity due to depression in the EU has been estimated at over €70 billion per year (World Health Organization Regional Office for Europe, 2018; Jaffe et al., 2019; Johnston et al., 2019; Greenberg et al., 2021).

According to the Diagnostic and Statistical Manual of Mental Disorders Fifth Edition (DSM-5 American Psychiatric Association, DSM-5 Task Force, 2013), if depressive symptoms are not attributable to the physiological effects of a substance or other medical or neurological conditions, the depression and related subtypes are to be considered mental disorders, i.e., disorders somehow targeting the central nervous system (CNS) exclusively.

But why do we assume that these changes result from intrinsic functioning of the brain only? Why is it that more than 60% of patients with major depression may show no response to first-line antidepressant treatment mainly targeting brain neurotransmitters (Johnston et al., 2019)? Why do recent meta-analyses show inconsistency between studies that only focus on brain dynamics and depression (e.g., Müller et al., 2017)? These questions suggest that previous literature on depression has mainly focused on brain dynamics, while giving inadequate attention to body dynamics.

The brain and heart have long been studied and treated individually at the cortical/subcortical or neuro-peripheral level, including cardiovascular, blood pressure and respiration dynamics (as part of non-invasive monitoring) using specific techniques focused on specific system dynamics. However, the physiological system is complex. In other words, because of the numerous interactions of many subcomponents, the system as a whole exhibits characteristics that the individual components cannot act on. The CNS and autonomic nervous system (ANS) control all body organs simultaneously through anatomical and biochemical/functional connections that have a lasting impact on health and especially disease.

In piecing together segregated parts of research and in following the development of a comprehensive vision on whole-nervous system mental health and related science, here we pose the following fundamental hypotheses:

- In the absence of medical or neurological illnesses, brain changes in depression are not only the cause but also the consequence of a mood disorder

Scientific evidence supporting these hypotheses leverage upon the definition of the Central-Autonomic Network (CAN), which comprises brain regions involved in autonomic control (Benarroch, 1993; Beissner et al., 2013; Valenza et al., 2019; Valenza et al., 2020), as well as Brain-Heart Interplay (BHI), which comprises the functional links between CNS and ANS through electrical, biochemical, and physical communications. In this perspective, depression is envisioned as a manifestation of a dysregulation of cardiac vagal and sympathovagal dynamics sustaining CAN dysfunctions, which in turn reflects on functional BHI. Consequently, a “functional-vagal theory” of depression is stated, suggesting that depression treatment should also target brain-heart dynamics and, especially, act at a nervous-system-wise level.

The aforementioned literature proves that there is a close association between emotional regulation and processing, ANS and CNS control over the body’s internal state. Nonetheless, it seems that contemporary psychiatry and clinical psychology largely assume that the functional and dynamic interactions within the human nervous system during emotional processing, regulation, and dysregulation are primarily driven by brain-based processes rather than by bottom-up processes, such as depression and its various subtypes. To this extent, it is worth mentioning the scientific debate on the nature of emotions that has lasted over a century (James, 1880). Recent research shows that neural control over heartbeat dynamics creates and initiates emotional responses (Candia-Rivera et al., 2022; Hsueh et al., 2023). This scientific finding undermines the “classical” emotion theories that suggest emotions are solely functional states of the brain. Consequently, the current grand challenge is to demonstrate the significant causal involvement of dynamical vagal and sympathovagal activities in depression and its subtypes, which may be reflected in changes in the brain.

So why are vagal activity levels lower when a patient has parasympathetic-dominant hypoarousal symptoms?

To explain the autonomic correlates observed in depression, it is here assumed that depression is a disease involving the parasympathetic nervous system, therefore involving dysregulation of the dynamical CAN through vagal control and sympathovagal interplay. Accordingly, let us assume that time-varying vagal activity results from the complex interaction between body- and brain-related dynamics, and pathological ANS dysfunctions are associated with significant CAN changes and emotional dysregulation. According to the Fourier theory, every time series can be represented as a linear combination (i.e., sums and differences) of sinusoidal functions that have specific amplitude and frequency. Figure 1.

Depression is often characterized by a similar or reduced vagal tone and increased emotional reactivity compared to healthy individuals, coupled with parasympathetic-dominant hypoarousal symptomatology. In any case, depression may be associated with pathological vagal dynamics. In order to transition from pathological vagal dynamics to healthy ones, the amplitude of the sinusoid, i.e., the emotional reactivity (Garcia et al., 2016), must be reduced, and the mean of the sinusoid, i.e., the vagal tone, must be increased. This approach is hereby referred to as the “functional vagal theory of depression” While this theory may be useful in understanding the underlying pathophysiology of depression and its relationship to neurocardiovascular dysfunctions, the non-specificity of brain and neurocardiovascular dynamics (Saul and Valenza, 2021), along with the nonlinear and complex nature of the cardiovascular system, makes its direct translational clinical application challenging.

Indeed, although numerous physiological conditions and pathologies are associated with HRV-derived and brain-signal-derived biomarkers that correlate with either the severity of the pathology or the physiological state, these correlations may not be particularly specific, especially when experimental conditions and methodological approaches vary. Consequently, it is possible that a healthy individual in a particular neuro-autonomic state may exhibit similar dynamics to a subject with pathology in a different psycho-physiological state. The nonlinearity of the cardiovascular system poses challenges since an increase in vagal activity could potentially result in an increase in heart rate, as the effect of vagal stimulation on heart rate strongly depends on the level of sympathetic stimulation occurring concurrently (Sunagawa et al., 1998). This phenomenon, known as accentuated antagonism, is partly attributed to the inhibitory effect of the vagally released ACh on the release of NorEpinephrine from nearby sympathetic nerve endings. Additionally, the complexity of the cardiovascular system, resulting from the many interactions of numerous sub-components, means that the system as a whole exhibits properties that the individual components acting alone cannot demonstrate. The system’s nonlinearity, coupled with the multiple feedback mechanisms of sympathetic and vagal activity on cardiovascular control, makes the system extremely sensitive to input, even infinitesimal changes (Sunagawa et al., 1998). Therefore, we do not explicitly endorse the use of the mean of the sinusoid, i.e., the so-called vagal tone, to detect depression. Rather, it is the author’s opinion that brain-heart related biomarkers are more promising in targeting specific oscillations and spatial information over the scalp or brain region associated with afferent or efferent peripheral activity (Pfurtscheller et al., 2017; Catrambone et al., 2021; Pfurtscheller et al., 2021; Pfurtscheller et al., 2022; Pfurtscheller et al., 2023).

Prospectively, the diagnosis and treatment of depression and its subtypes should move from a symptomatic- and brain-centred view exclusively to a functional and whole-body framework where vagally-mediated CAN dynamics play a crucial and causal role. In other words, modern medicine for this highly disabling mood disorder should reverse the current diagnostic and treatment framework for depression by moving from a symptomatic-only and brain-dominated view to a new functional-vagal and whole-body conception. To this extent, science needs to achieve the following goals:

Gain Fundamental Knowledge on SympathoVagal activity & Functional Brain-Body interplay. There is the need to develop effective biomarkers of dynamical sympathovagal activity and associated CAN/brain-heart interplay to characterize physiological sympathovagal activity and functional BHI in healthy conditions. These biomarkers may be defined from generic multivariate signal processing methods and subsequent feature extraction in the time, frequency, and/or nonlinear and complexity domains; Characterise brain-heart-mediated emotional responses. There is the need to characterise a dynamical emotional profile in the healthy through time-varying autonomic- and CAN-related metrics. Gender, age, and other socio-demographic factors should be considered because they are known to affect heartbeat dynamics and, therefore, CAN-related and BHI-related metrics; Characterise brain-heart-mediated biomarkers specific of emotional dysregulation and depression. An impaired dynamical emotional profile of depression should then be characterised throughout treatment by mapping and comparing associated autonomic and CAN and BHI patterns to a healthy profile. Experimental paradigms including standardised emotional elicitation as well as personalised recall scripts combined with non-invasive brain stimulation should be employed to stimulate the emergence of dysfunctional emotion regulation. The desired outcome may be achieved through identifying personalized autonomic and BHI-related features that are distinct to each individual and contribute to their depressive symptomatology. By utilizing these features, the pace and intensity of brain and vagal stimulations can be adjusted accordingly, enabling concurrent progress towards remission and recovery.

Taking a functional-vagal perspective and a broader view of the nervous system into account can help clinicians better understand the clinical symptoms of depression that drive current practice and allow for objective neurophysiological assessment. Instead of changing the current clinical paradigm, this approach can complement the existing diagnostic framework and provide new diagnostic criteria that take into account the essential role of aetiology and pathophysiology in diagnostic decision-making. By adopting quantitative assessments based on specific BHI-related disease biomarkers, modern psychiatry can align itself with other medical specialties, including cardiology and neurology. This approach will enable psychiatry to achieve a more comprehensive understanding of the biological underpinnings of mood disorders, bringing it closer to the diagnostic and treatment practices of these related fields. This holistic view may contribute to the update of the so-called Research Domain Criteria (Morris and Cuthbert, 2012), which proposes to study five psychopathological domains including negative and positive valence, cognitive systems, social processes, and arousal/regulatory systems to better understand mood disorders.

It is important to note that a comprehensive investigation of dysfunctional neural dynamics in depression should also consider sympathetic dynamics. In terms of symptomatology, sympathetically dominated hyperarousal is marked by emotional hyperactivity, reactivity, impulsivity, anxiety, and anger, accompanied by parasympathetic hyposensitivity. Evidence concerning the correlation between sympathetic activity and depression is relatively scarce compared to vagally-related measures, likely due to challenges in quantitatively assessing sympathetic activity (Valenza et al., 2018). Nevertheless, it is posited that dysfunctional sympathetic dynamics may contribute to the increased risk of cardiac complications observed in depression (Nicholson et al., 2006; Barton et al., 2007; Dong et al., 2012). Indeed, exposure to risk or trauma has been found to stimulate the ANS, resulting in sympathetic and/or parasympathetic hyperalertness, leading to a pathological mood state in which an individual is unable to experience positive emotional states (Corrigan et al., 2010). When uncontrolled ANS (i.e., sympathetic-dominant hyperarousal and/or parasympathetic-dominant hypoarousal) cannot control heightened emotional or depressive states, patients often report being unable to cope with emotional and physiological arousal (Ogden, 2006; Corrigan et al., 2010). The “window of tollerance” model of autonomic arousal states that there is a “window” of healthy autonomic tolerance when intense emotions and a state of calm or relaxation can be integrated and integrated throughout the body. The periaqueductal grey area, which is also part of the CAN, is thought to be involved in these mechanisms (Corrigan et al., 2010; Beissner et al., 2013). Identifying CNS-ANS or brain-heart markers of depression—ideally under naturalistic conditions—can help maintain or re-establish a healthy “Window of Tolerance,” for example, through personalised concurrent brain-body stimulation.

Considering the swiftly growing body of scientific evidence highlighting dysfunctional BHI dynamics in a range of pathological conditions (Taggart et al., 2011; Tahsili-Fahadan et al., 2017; Joyner, 2019; Riching et al., 2019; Riganello et al., 2019; Liu et al., 2020; Méloux et al., 2020; Costagliola et al., 2021; Nowroozpoor et al., 2021; Tumati et al., 2021; Vaccarino et al., 2021; Bekała et al., 2022; Grassi et al., 2022; Kumral et al., 2022; Lazaridi et al., 2022; Liu et al., 2022; Seligowski et al., 2022; Wang and Peng, 2022; Xue et al., 2022), it is possible that cardiologists will run EEG/psychometric/emotional assessments before their standard clinical evaluation. Likewise, psychiatrists and clinical psychologists may need to conduct EEG/MRI scans and 24-h cardiac holter monitoring to consider depression as a neurocardiovascular disorder.”