Critique of the Polyvagal Theory
Critique of the Polyvagal Theory-min

Critique of the Polyvagal Theory

Main features of the Polyvagal Theory

Since its first description by Stephen Porges in 1995 [1,2], the polyvagal theory (PVT) has  received much attention among mind-body therapists including osteopaths worldwide,  especially with regard to the treatment of trauma patients. PVT is an attempt to explain the  relationship between parasympathetic activity and behavior from an evolutionary  perspective [3]. It aims to provide an understanding of the connections between brain and  body processes [1,2]. 

The term ,,polyvagal,“ first employed in the polyvagal theory, refers to 2 vagal circuits

  • One is supposedly the phylogenetically older unmyelinated system represented by  the dorsal motor nucleus of the vagus, which mainly innervates subdiaphragmatic  organs (mainly the gastrointestinal tract), but also purportedly the heart, and is  supposedly associated with immobilisation and dissociation (Fig. 1). 
  • Evolutionarily later, according to Porges, a second younger vagal pathway is thought  to have developed, which is claimed only to be observed in mammals, not in  reptiles, and to have the ability to down-regulate immobilization and fight-and-flight  behavior. According to Porges, the anatomical structures of this component of the  vagus interact in the brainstem with structures innervating the striated muscles of  the face and head to create an integrated system of social engagement [4]. This  younger system is represented in particular by the nucleus ambiguus (Fig. 1). ln PVT,  it is associated with the other branchiomotor (special visceroefferent) nuclei of the  Vth, Vll, IXth, and Xlth cranial nerves referred to as the ventral vagal complex l5]. This  system regulates the heart and lungs via myelinated nerve fibers to allow resting  states and is thought to be related to social behavior and “feeling safe”. 

The focus of PVT is on the proposed phylogenetic shift between reptiles and mammals that putatively resulted in specific changes in the vagal pathways regulating the heart.  According to the PVT–but not necessarily accurate–primary vagal efferent pathways  regulating the heart shifted from the dorsal nucleus of the vagus in reptiles to the nucleus  ambiguus in mammals, establishing a face-heart connection with properties of a social  engagement system that allows social interactions to influence visceral state and visceral  dysfunction manifested in neural regulation of the heart [7].

Comparative anatomical and functional studies related to PVT

Comparative anatomical and functional studies argue against the proposed phylogenetic  basis of PVT. It is undisputed that in mammals, myelinated cardioinhibitory axons arise from  the nucleus Ambiguus. However, already in cartilaginous fish (elasmobranchs, e.g., sharks),  which have existed for 400 million years, the cardioinhibitory vagus neurons are myelinated  and conduct at speeds between 7 and 35 m/s (corresponding to the B fibers of mammals).  Moreover, their cell bodies are located at 2 different sites in the brainstem (dorsal vagus  nucleus and primordium of the nucleus Ambiguus) [8, 9]. Thus, cartilaginous fishes are  already ,,polyvagal“. 

Lungfish, evolutionary precursors of air-breathing animals, also have a myelinated cardiac  vagus nerve originating in dorsal and ventrolateral brainstem nuclei [10]. These myelinated,  fast-conducting axons enable beat-to-beat slowing of the heart rate, which is mandatory for  the cardiorespiratory interactions observed in these ancient vertebrates, similar to  mammalian respiratory sinus arrhythmia [10, 11]. The unmyelinated cardiac neurons of the  dorsal vagal nucleus do not have any significant influence on heart rate and thus cannot be responsible for bradycardia such as that observed in freezing states. They seem to  influence ventricular inotropy and might protect cardiomyocytes from ischemia [12a]. 

Response patterns in PVT

PVT assigns responses to perceived risks to 3 categories: feeling safe, being in danger, or  perceiving a threat to life. These categories are hypothesized to follow one another in  phylogenesis. They are proposed to relate to the adaptive behaviors of social  communication (facial expressions, speech, listening), which are proposed to be controlled  by the nucleus ambiguus. On the other hand, defensive behavior in terms of mobilization  (fight, flight) and immobilization responses (vasovagal syncope, dissociation or emotional  freezing state) is claimed to be mediated by the dorsal vagal nucleus [1, 6, 12-14]. 

Again, the proposed association of these behavioral phenomena with the old unmyelinated  or the new myelinated vagus nerve is misleading. The mammalian nucleus ambiguus contains, in addition to cardioinhibitory neurons, primarily the branchiomotor (special visceroefferent) neurons for laryngeal, pharyngeal, and striated esophageal muscles [1 5],  but does not control facial expression (mimic muscles are innervated by the nucleus facialis,  nF) nor hearing via the middle ear muscles (tensor tympani muscle, innervated by the motor  branch of the trigeminal nerve, and stapedius muscle, innervated by the facial nerve) nor  other head and neck muscles, as suggested by PVT. Conversely, the facial nucleus (nF) also  does not affect the nucleus ambiguus. 

All these motor nuclei, including the hypoglossal nucleus, are coordinated by premotor  networks in the lateral parvocellular and intermediate reticular formation [16-20]. The  intermediate reticular formation, located between the medial magnocellular and lateral  parvocellular areas, also houses the neuronal networks for cardiovascular regulation  (vasomotor center) and the central generators for respiratory rhythm (pre-Bötzinger  complex, respiratory center) and for swallowing and vomiting.

The dorsal vagal nucleus and the nucleus ambiguus are anatomically and functionally embedded in these networks,  but as output elements rather than coordinators. Vagal afferents are connected via the  solitary nucleus (nucleus tractus solitarii) not only to the motor vagus nuclei (dorsal vagal  nucleus and nucleus ambiguus), but also to the premotor networks of the reticular  formation and the circulatory and respiratory centers [20]. However, trigeminal and upper  cervical spinal afferents, which are also fed to the premotor reticular networks, are equally  important for the coordination of the entire head-cervical motor system. 

Role of the mesencephalic periaqueductal gray (PAG)

A bilateral periventricular nucleus in the ventral mesencephalon, showing a similar location  to the mammalian PAG, has already been described in the lamprey, which belongs to the  oldest group of vertebrates living today [21]. Behavioral states such as fight and flight,  immobilization or freezing state, and risk assessment—together with associated motor,  autonomic, and endocrine effects—are coordinated by the mesencephalic periaqueductal  gray (PAG) [22-25]. The PAG is connected to the hypothalamus and limbic system (primarily  the amygdala and prefrontal cortex) [23,24] as well as to various premotor and autonomic  brainstem nuclei that coordinate respiration and the emotional motor system [25]. The PAG  receives afferents from almost all sensory systems—not least of all the nociceptive  system—and modulates their processing [24]. 

Undoubtedly, the vagus nerve has a significant influence on emotions and various  behavioral states due to its large afferent component. Vagal afferents, which constitute  about 80% of its axons, are transmitted through the nucleus tractus solitarii to the PAG,  hypothalamus, amygdala, and insular, cingulate, and prefrontal cortex, where they are  integrated into emotional and cognitive processes [26-29]. Recent studies suggest that  subdiaphragmatic vagal afferents influence innate fear, learned fear, and other behaviors  [30, 31]. Moreover, vagal afferents modulate spinal nociceptive processes in several  experimental models [32, 33]. 

It is true that Porges [34 ] mentions the representation of neuroanatomically– already long  known–relations of the limbic system and PAG with bidirectional connections to the vagus  complex. However, since it is not the ventral vagus complex but the PAG in association with  

limbic and other brainstem networks that is responsible as a coordinator for these  behavioral states and, moreover, numerous brain areas, if not the entire brain, function as a  system of social engagement, the term ,,polyvagal“ to characterize it appears to be a  misleading misnomer. 


As Grossman and Taylor [11] have already shown, phylogenetic references are questionable  as a basis for PVT. Facts of cranial nerve anatomy are also sometimes incorrectly  represented in PVT. Instead of extending the concept of the ventral vagal complex to all  branchiomotor nuclei, it would be more appropriate to leave them their independence and  emphasize their coordination by a network of brainstem neurons. 

The concept of a system of social engagement is plausible and seems to be relevant to  practice. However, it is misleading in the formulation of the polyvagal assertions, and linking the concept with the „old unmyelinated or new myelinated vagus“ and the term,  „polyvagal“ should be avoided. In addition, the hypoglossal nerve, which is not a  branchiomotor nerve but innervates the socially important tongue muscles, should also be  included in the concept of social engagement. 

The mesencephalic trigeminal nucleus and other sensory trigeminal nuclei are also  important in the coordination of orofacial motor activity. The vagus nerve, efferent as well  as afferent, is certainly an important factor in the social engagement system. However,  because the [supposedly] “new“ vagal nucleus in the form of the nucleus ambiguus does  not exert a coordinating function on the other branchiomotor nuclei (V, Vll, lX, Xl)–even  though vagal afferents are fed into these coordination networks via the nucleus tractus  solitarii–PVT turns the causal relationships upside down. Consequently, the term  „polyvagal“ is a misleading misnomer. The functional construct of the social engagement  system should not be associated with the term „polyvagal.” Possibly a clarifying new  designation would be indicated. 

Translated (Paul Grossman, Winfried Neuhuber, from:  Deutsche Zeitschrift für Osteopathie (German Journal of Osteopathy) 2021 19:34-37 


[1] Porges SW. The polyvagal theory; phylogenetic substrates of a social nervous system. Int J Psychophysiol 2001: 42 (2): 123-146 

[2] Porges SW. The polyvagal theory: phylogenetic contributions to socialbehavior. Physiol  Behav 2003:19 (3): 503-5 1 3 

[3] Porges SW. The polyvagal perspective. Biol Psychol 2001;14 (2):116-143 

[4] Porges SW, Dana D. Clinical Applications of the Polyvagal Theory: The Emergence of  Polyvagal-lnformed Therapies. New York: WW Norton; 2018 

[5] Porges SW. Social engagement and attachment: a phylogenetic perspective. Ann NYAcad  Sci 2003; 1008:31-47 

[6] Porges SW. The Polyvagal Theory: Neurophysiological foundations of emotions,  attachment, communication, and self-regulation. New York, NY: Norton; 201 1 

[7] Porges SW, Kolacz l. Neurocardiology through the lens of the polyvagal theory. ln: Celpi  Rl, Buchholz B. Neurocardiology: Pathophysiological Aspects and Clinical lmplications.  London: Elsevier; 2018 

[8] Barrett Dl, Tayior EW. The location of cardiac vagal pregan-glionicneurones in the brain stem of the dogfish Scyliorhinus canicula. J Exp Biol 1985; 117: 449-458 

[9] Barrett DJ, Taylor EW. The characteristics of cardiac vagal pre-ganglionic motoneurones  in the dogfish. I Exp Biol 1985; 117:459-470 

[10] Monteiro DA, Taylor EW, Sartori MR et al. Cardiorespiratory interactions previously  identified as mammalian are Present in the primitive lungfish. Sci Adv 2018; 4(2): eaaq0800

[11] Grossman P, Taylor EW. Toward understanding respiratory sinus arrhythmia: relations  to cardiac vagal tone, evolution and biobehavioral functions. Biol Psychol 2007: 74: 263-285 

[12] Porges SW. Love: an emergent property of the mammalian autonomic nervous system.  Psychoneuroendocrinology 1998; 23:837-861 

[12a] Gourine AV, Machhada A, Trapp S, Spyer, KM. Cardiac vagal preganglionic neurones: an  update. Auton Neurosci 2016; 199:24-28 

[13] Porges SW. The polyvagal perspective. Biol Psychol 2001;14:116-143 

[14] Porges SW. The polyvagal theory: new insights into adaptive reactions of the autonomic  nervous system. Cleve Clin J Med 2009:76 (suppl 2): 586-590 

[15] Fritzsch B, Elliott KL, Clover JC. Caskell revisited: new insights into spinal autonomics  necessitate a revised motor neuron nomenclature. Cell Tissue Res 2017; 370: 195-209 

[16] Fay RA, Norgren R. ldentification of rat brainstem multisynaptic connections to the oral  motor nuclei using pseudorabies virusc l: masticatory muscle motor systems. Brain Res Brain  Res Rev l997; 25 (3):255-275 

[17] Fay RA, Norgren R. ldentification of rat brainstem multisynap-tic connections to the oral  motor nuclei in the rat using pseudorabies virus: ll: Facial muscle motor systems. Brain Res  Brain Res Rev 1997;25 (3):216-290 

[18] Fay RA, Norgren R. ldentification of rat brainstem multisynap-tic connections to the oral  motor nuclei using pseudorabies virus: lll: Lingual muscle motor systems. Brain Res Brain Res Rev 1 997; 25 (3): 291 -31 1 

[19] Travers lB, Norgren R. Afferent projections to the oral motor nuclei in the rat. I Comp  Neurol 1983; 220 (3):280-298 

[20] Büttner-EnneverJA, Horn AKE. Solitary nucleus (SOL). ln: Büttner-Ennever JA, Horn AKE.  Olszewski and Baxter’s Cyto-architecture of the Human Brainstem. 3. Aufl. Basel: Karger; 2014 

[21] Olson l, Suryanarayana Sl\4, Robertson B, Crillner S. Criseum centrale, a homologue of  the periaqueductal gray in the lamprey. IBRO Rep 2011;2:24-30 

[22] Deng H, Xiao X, Wang Z. Periaqueductal gray neuronal activ-ities underlie different  aspects of defensive behaviors. J Neurosci 2016; 36 (29): 7580-7588 

[23] Roelofs K. Freeze for action: neurobiological mechanisms in animal and human freezing.  Philos Trans R Soc Lond B Biol Sci 2011: 312 (l718): 20160206

[24] Jänig W. The integrative action of the autonomic nervous system. Cambridge:  Cambridge University Press; 2006 

[25] Holstege C. The periaqueductal gray controls brainstem emotional motor systems  including respiration. Prog Brain Res 2014; 209: 319-405 

[26] Craig AD. How do you feel? lnteroception: the sense of the physiological condition of  the body. Nat Rev Neurosci 2002; 3(8):65s-666 

[27] Critchley HD, Mathias Cl, Dolan Rl. Neuroanatomical basis for first and second-order  representations of bodily states. Nat Neurosci 2001: 4 (2): 201 -212 

[28] Saper CB. The central autonomic nervous system: consciotrs visceral perception and  autonomic pattern generation. Annu Rev Neurosci 2002:25: 433-469 

[29] Berthoud HR, NeuhuberWL. Functional and chemical anatomy of the afferent vagal  system. Auton Neurosci 2000; 85 (1-3):1-17 

[30] Klarer M, Arno d M, Cünther L et al. Cut vagal afferents diffe-rentially modulate innate  anxiety and learned fear. I Neurosci 2011: 34 (21): 1061 -1066 

[31] Klarer M, Krieger JP, Richetto J et al. Abdominal vagal afferents modulate the brain  transcriptome and behaviors relevant to schizophrenra. J Neurosci 2018; 38: 1634-1641 

[32] Foreman RD, Qin C, lou CJ. Spinothalamic system and viscerosomatic rnotor reflexes:  functional organization of cardiac and somatic input. In: l(ing HH,länig W, Patterson MM.  The science and clinicai application of manual therapy. Edinburgh: 

Churchil Livingstone; 201 1: 11-127 

[33] Jänig W Creen P. Acute inflammation ln the joint: its control by the sympathetic  nervous system and neuroendocrine systems. Auton Neurosci 2014: 182: 42-54 

[34] Porges SW. Social engagement and attachment: a phylogenetic perspective. Ann NY Acad Sci 2003; 1008:31-47