Polyvagal Theory Wasn't Debunked. But That's Not Really the Point.
What a heated scientific debate reveals about how we should treat all clinical models, and why research should refine our approach, not define it.
**Note: This piece draws from my original ideas, research, hooks, and metaphors. For editing and some wording, I’ve used AI tools trained on my own books and style, always blending technology with my hands-on curation and oversight.
I saw the debate raging online and just shook my head.
To listen to the discourse, you would think the entire field of trauma therapy had collapsed overnight. Polyvagal Theory has been debunked. Everything you learned is wrong. The researchers have spoken. Practitioners who built careers around this framework were either defending it like scripture or abandoning it like a sinking ship.
I set the articles aside and actually read them.
Here’s my take. And here’s what all of this means for those of us doing clinical work.
First, something I should tell you upfront: I don’t actually use Polyvagal Theory in my clinical work. Not exactly. What I use is something I’ve assembled over decades from multiple research streams. Memory reconsolidation. ACE and PTSD research. Attachment theory. The literature on stuck nervous system states. Interoception. Predictive processing. I’ve borrowed language from Polyvagal Theory where it was useful and built my own working model around it..... a model I know is not perfectly accurate neuroanatomically, and one I’m refining continuously as both the research and my clinical experience evolve.
That process of ongoing refinement..... holding a working model lightly, updating it when the evidence demands it, not treating it as finished.... is exactly what this debate is now asking the entire field to do.
So when I read these papers, I didn’t feel like something I believed in was under attack. I felt like the refinement process was being done publicly, loudly, and in a way that should force a lot of practitioners to get more honest about what they actually know versus what they’ve been assuming.
That’s worth paying attention to.
What Is Actually Being Debated
Let me be precise here, because the online conversation has collapsed something genuinely complex into a simple narrative: Polyvagal Theory has been debunked. That framing does more damage than it does good.
The Grossman et al. critique, published across two papers in Clinical Neuropsychiatry and Biological Psychology, targets specific anatomical and measurement claims embedded in the theory. These include, most centrally: whether the dorsal vagal motor nucleus (DMNX) is capable of mediating the kind of massive bradycardia the theory proposes during states of shutdown or freeze; whether respiratory sinus arrhythmia (RSA) is actually a reliable pathway-specific index of ventral vagal tone originating in the nucleus ambiguus; and whether the evolutionary story that polyvagal maps onto.... the idea of reptilian versus mammalian vagal organization... holds up against modern comparative biology.
Grossman’s argument, in brief: the evidence strongly suggests that virtually all cardiac vagal control in mammals runs through the nucleus ambiguus (the ventral vagal nucleus), not the dorsal motor nucleus. Which means the clean distinction polyvagal draws between these two pathways, and the behavioral correlates tied to each, may be built on shaky neuroanatomical ground.
Porges’ response, also published in Clinical Neuropsychiatry, is extensive and technical. His core rebuttal: the critique is attacking a reconstructed version of the theory, not the theory as actually formulated. Polyvagal never claimed anatomical exclusivity. RSA is not meant as a global vagal tone measure but as a pathway-specific, respiratory-gated index of ventral vagal influence. And the evolutionary framing is about functional reorganization, not rigid hierarchical layering of ancient versus modern circuits.
Who is right? That’s a harder question than either side’s certainty would suggest. And for our purposes as clinician-integrators, it may be the wrong question to lead with.
The Map and the Territory
Here is something I keep returning to: every model in clinical psychology is a metaphor. Every framework is a map. And the map is never the territory.
This isn’t a dodge. It’s an epistemological reality that clinical practice has to sit with every single day. We work with models that are necessarily incomplete representations of staggeringly complex biological systems. The DSM is a map. Attachment theory is a map. The stress-response literature is a map. The HPA axis as a simplified narrative of threat and cortisol is a map.
Polyvagal Theory is a map. A detailed one, a generative one, one that helped practitioners actually help people in ways that older models couldn’t quite account for. But still a map.
The debate about polyvagal theory’s anatomical premises is not a debate about whether co-regulation matters. It’s not a debate about whether the nervous system scans for cues of safety and danger. It’s not a debate about whether relational attunement, tone of voice, facial expression, and felt sense of safety are therapeutically central.
Even Grossman acknowledges that polyvagal hypotheses borrowed substantially from prior existing theories including attachment theory, arousal models, and the social engagement literature. His critique is that the polyvagal framework acquired its unique claims by linking those well-grounded observations to specific anatomical mechanisms that may not hold.
That’s a meaningful scientific critique. It deserves engagement.
But here’s what it doesn’t do: it doesn’t erase the clinical utility of the relational and regulatory work being done under the polyvagal umbrella. That work is grounded in older, more solid traditions even if the mechanistic story needs updating.
Three Roles, One Ecosystem
I want to offer a framework I’ve been sitting with for years. I think conflict around debates like this one almost always comes from confusing or collapsing three distinct roles.
The first is the Researcher. The researcher’s job is to stress-test models, expose mechanistic overreach, narrow or revise claims, and occasionally open up new measurement strategies or constructs. Grossman is doing exactly this. It’s valuable work. Without researchers willing to push back on beloved frameworks, clinical practice calcifies around stories that may not survive contact with data.
The second is the Clinician. The clinician’s job is different in kind. Clinicians are at the front edge of practice. They notice patterns the literature hasn’t catalogued yet. They invent and refine techniques based on what actually works in the room with actual human beings in actual states of dysregulation. They build frameworks and metaphors that help real people understand their own experience. Clinicians, by necessity, operate ahead of the research. They have to. The person sitting across from them is in distress right now, not in two years when the meta-analysis comes out.
The third role is the one I locate myself in: the Clinician-Integrator. This is the rare and underappreciated bridge position. The clinician-integrator takes what researchers are finding and translates it into something usable and humane without distorting it into something the research didn’t actually say. They keep the clinically powerful elements of a framework, soften or update mechanistic claims where evidence is weak, and feed practice-based questions back into the research pipeline.
When a debate like the polyvagal controversy erupts, each role tends to see it differently. The researcher says: the mechanistic claims are wrong, therefore the framework is compromised. The clinician says: but it works, therefore the critique is irrelevant. The clinician-integrator holds the tension between those two positions without collapsing into either of them.
That’s harder than it sounds. Most people don’t want to hold tension. They want resolution.
What a Clinician-Integrator Does With This Debate
Here is what I am not going to do: burn down the clinical language of polyvagal theory because some anatomical premises are contested. That would be epistemically sloppy and practically harmful.
Here is what I am going to do: hold the clinical tools more carefully and teach the mechanistic claims more honestly.
The state maps..... the idea of a ventral vagal state of social engagement, a sympathetic mobilization state, and a dorsal vagal state of shutdown..... are still clinically useful as organizing metaphors, even if the underlying anatomy is more complicated than originally described. People recognize themselves in these maps. That recognition is therapeutic. The map is working even if the legend needs revision.
Co-regulation is not a polyvagal invention, but polyvagal gave it language and neurophysiological scaffolding that made it accessible to practitioners who don’t read developmental psychology journals. Losing that language carelessly costs something real.
Neuroception..... the idea that the nervous system is evaluating cues of safety and danger outside of conscious awareness..... is consistent with predictive processing models, with interoception research, with a substantial literature on implicit threat detection. The specific claim about which vagal nucleus is doing what may be contestable. The broader point about subcortical threat evaluation preceding conscious appraisal is not.
So I’ll keep using the concepts that are well-supported. I’ll soften the anatomical certainty in how I teach. I’ll be more precise about what RSA does and doesn’t tell us. And I’ll say, honestly, that this is an evolving framework, not a finished one. That’s not a concession. That’s intellectual honesty.
Research Should Refine Our Approach, Not Define It
This is the bigger principle underneath the polyvagal debate, and it matters far beyond this particular controversy.
Clinical psychology has a complicated relationship with its own evidence base. The official standard of evidence-based practice calls for integrating best research evidence with clinical expertise and patient preferences. That three-part formulation is important. It’s not just research. It’s research plus experience plus the actual human being in front of you.
But in practice, we sometimes treat research findings as if they are clinical verdicts. A new meta-analysis drops and suddenly everyone is scrambling to revise their entire framework. A theory gets critiqued and practitioners who built careers on that framework feel personally attacked. A measurement question becomes a referendum on whether an entire clinical tradition is legitimate.
That’s research defining practice. And it tends to produce a kind of epistemic whiplash that serves no one well.
Compare this to how a skilled clinician-integrator actually works: they use research to sharpen their questions, not replace their judgment. They update their mechanistic language when evidence calls for it. They don’t abandon clinical tools that work simply because the theoretical explanation for why they work is being revised. They hold the model lightly.....
Because every clinical model is provisional. Every one. The ones we trust most today will look different in twenty years. That’s not a failure of the field. It’s how knowledge works.
I’ve been here before, on a much smaller scale. Years ago, a well-known researcher took issue publicly with how I talked about hormones and stress in my clinical work. The framing was essentially: this isn’t real medicine, this language isn’t scientifically legitimate. My answer then is the same as my answer now. Clinical language and research language serve different masters. When I talk about cortisol patterns and hormonal states with a perimenopausal woman who has been dismissed by three doctors and is trying to understand why her body feels unrecognizable to her, I am not writing a journal article. I am building a bridge between her experience and a framework she can use. Collapsing that distinction doesn’t make the science more rigorous. It just makes the clinician less useful.
The polyvagal debate is a case study in this. If you treat polyvagal as a sacred text, Grossman’s critique is devastating. If you treat it as an evolving clinical framework built on a foundation of older, more robust traditions, then what changes is not the work. What changes is the story we tell about why the work works.
I don’t know. Maybe that’s enough.
The Rewrite, Rewire, Retrain Lens
Within the M.U.D. framework (Misguided Unconscious Decisions), I’ve written about how the work of transformation operates across three levels: rewriting subconscious stories and identity structures, rewiring emotional and attachment patterns held in the body, and retraining the nervous system toward more adaptive regulation.
The polyvagal debate lives entirely in that third layer. It’s a dispute about the specific neurophysiological mechanisms by which nervous system regulation operates and how accurately we can measure it.
What the debate doesn’t touch: whether safety matters. Whether co-regulation matters. Whether the felt sense of being met by another human being changes something in a traumatized nervous system. Whether relational attunement is a therapeutic mechanism. The evidence for all of that is robust, cross-theoretical, and older than polyvagal theory itself.
What the debate does touch: whether the anatomical story we’ve been telling to explain those effects is accurate. Whether RSA is measuring what we’ve claimed it’s measuring. Whether the evolutionary narrative is scientifically sound.
The rewrite layer in transformation work is about updating outdated stories. The polyvagal debate is asking us to update some stories we’ve been telling about the nervous system. That’s not a crisis. It’s the field doing its job.
Final Thoughts
I keep coming back to that moment of shaking my head at the noise online, stepping back, and actually reading the papers.
There was a man I worked with years ago..... I’ll call him D..... who came to me after what felt like his life had finished. Divorce finalized. Job gone. He was staying in a spare room at his brother’s place. He told me he felt like a switch had been thrown somewhere inside him. That he couldn’t feel anything. Couldn’t respond. Couldn’t fight. He just..... wasn’t.
He described it as being underwater. Looking up through glass at a world he couldn’t reach.
The model I used with him wasn’t pure polyvagal. It was my assembled version..... the language of stuck states, of a nervous system running a survival strategy that made perfect sense once and had just never been updated. The specific anatomy underneath that explanation may be more complicated than the story I told. But the story was true enough to open a door. And he walked through it.
That’s what I mean when I say research should refine our approach, not define it. The refinement happening in the polyvagal debate right now is good. It’s necessary. It will make the maps better. Some of the language will need to change, some of the mechanistic certainty will need to soften, and the field will come out of this more honest than it went in.
But the work itself..... the co-regulation, the attunement, the slow careful business of helping a human nervous system learn that the world can be safe again... that work doesn’t wait for the research to settle. It never has.
I’ve been building and rebuilding my own framework for decades. Every new paper is a chance to refine it. Every client is a chance to test it. The polyvagal debate is just a louder, more public version of the process I’ve been running quietly in my own thinking for years.
The debate continues. Good. That’s exactly what a living field looks like.
PS: If you’re ready to break free of the patterns keeping you stuck in dysregulation and become the kind of person who naturally reads and responds to your own nervous system with skill, explore my Next Level Human coaching program today. Spots are limited... don’t wait. 👉 http://www.nextlevelhuman.com/human-coaching
References
Porges, S. W. (2026). When a Critique Becomes Untenable: A Scholarly Response to Grossman et al.’s Evaluation of Polyvagal Theory. Clinical Neuropsychiatry, 23(1), 113-128. doi.org/10.36131/cnfioritieditore20260111
Grossman, P. (2023). Fundamental challenges and likely refutations of the five basic premises of the polyvagal theory. Biological Psychology, 169, 108342. https://pubmed.ncbi.nlm.nih.gov/37230290/
Grossman, P., et al. (2026). Why the Polyvagal Theory is Untenable. An international expert evaluation of the polyvagal theory and commentary upon Porges (2025). Clinical Neuropsychiatry, 23(1).
Porges, S. W. (2025). Polyvagal Theory: Current Status, Clinical Applications, and Future Directions. Clinical Neuropsychiatry, 22(3). https://pmc.ncbi.nlm.nih.gov/articles/PMC12302812/
Kolacz, J., & Porges, S. W. (2025). Polyvagal theory: A journey from physiological observation to neural circuits, application, and new questions. Frontiers in Behavioral Neuroscience, 19, 1659083.
Swisher, A. K. (2010). Using Evidence to Guide but Not Dictate Practice. Cardiopulmonary Physical Therapy Journal, 21(1), 4-5. https://pmc.ncbi.nlm.nih.gov/articles/PMC3104930/
Cook, J. M., et al. (2017). Evidence-Based Psychotherapy: Advantages and Challenges. Neurotherapeutics, 14(3), 537-545. https://pmc.ncbi.nlm.nih.gov/articles/PMC5509639/
Tolin, D. F., et al. (2016). Evidence-Based Practice and Psychological Treatments. Frontiers in Psychology, 7, 1170.