Does the Body Keep the Score?
A Question Worth Asking Twice
Originally published on LinkedIn
August 26, 2026
Written by Amelia Barili — Teacher of Meditation, Volunteering and Positive Neuroplasticity; Ancient Wisdom and Neuroscience at UCB OLLI; and of Embodying Safety: Integrating Yoga, Qigong and Polyvagal Theory at the POLYVAGAL INSTITUTE.
A paper in FRONTIERS in Systems Neuroscience this year set out to challenge one of the most influential ideas in trauma care of the past decade. Its title: “The body does not keep the score” — a direct answer to Bessel van der Kolk’s The Body Keeps the Score. One of its authors, Karl Friston, is by several measures the most-cited living neuroscientist. This deserves a closer look than the headline gave it — not to defend or attack the book, but because the paper’s title itself, I think, is quietly at odds with the very theory it’s built on.
For the past decade, my work has centered on translating between ancient embodied wisdom traditions — yoga, qigong, Taoist practice — and the neuroscience increasingly finding mechanisms underneath them, teaching this integration at the University of California, Berkeley and the OLLI at UC Berkeley (Osher Lifelong Learning Institute) and at the Polyvagal Institute, in direct dialogue with Dr. Stephen Porges. I’m not a clinician. I’m an educator and translator, holding ancient wisdom and current science honestly side by side. Long before I had a name for systems thinking, the advanced masters I studied with in India and China had already shown me what those traditions have always known: body, mind, and spirit were never separate. My dear friend, the physicist Fritjof Capra — with whom I had the pleasure of collaborating for decades — later gave that knowing a scientific language, through our conversations and co-taught workshops built on exactly this kind of bridge-building. So when a debate like this one surfaces, I experience it as an invitation to look closer, the way those ancient traditions and Fritjof taught to look.
What the trauma paper argues. Kotler, Mannino, Fox, and Friston propose that trauma is not tissue-level storage but a brain-level prediction gone rigid: “the brain dynamically reenacts it through maladaptive inference.” The nervous system over-weights old danger signals and keeps generating bodily states — racing heart, braced gut — that match a threat no longer present. They’re careful about van der Kolk himself, crediting his use of Damasio and his attention to interoception. The paper itself frames this as less an outright refutation of van der Kolk than an updated reframing — one loop, described from two ends: his bottom-up, felt account of the body; theirs, top-down and computational.
Here is where I think the argument gets into trouble with itself. The title sets up a contrast — brain versus body, prediction versus storage — and that contrast only works if “body” means something separate from the nervous system. But predictive coding, as Friston has developed it since the mid-2000s, is not a theory of a brain computing in isolation. It is a theory of continuous exchange between brain and periphery — interoceptive signals from the viscera, proprioceptive and pressure signals from connective tissue, constantly informing and revising the brain’s predictions. The neuroscientist Anil Seth, working in this same theoretical tradition, has shown that interoceptive inference — the brain’s ongoing prediction of internal bodily states — treats signals from the viscera as primary data, not background noise. Under this model, the brain isn’t opposed to the body. It is in permanent conversation with it, updating its model on the body’s own reporting.
So when the title says “the body does not,” it borrows the everyday sense of “body” — physical tissue, separate from thought — precisely to make the sentence land. But that’s not the sense of “body” the paper’s own theory requires. The title, as a piece of rhetoric, quietly reintroduces a split their own framework was built to dissolve.
This is where fascia and the viscera become useful — not as a rebuttal, but as evidence. The paper never names fascia. Its actual phrase is “non-innervated tissue,” shorthand for a body-as-container, holding content apart from the nervous system. Real aneural tissue does exist — cartilage, for instance, has no nerve supply at all. But nobody claims trauma is held in a kneecap. The tissue the popular narrative actually points to — muscle, fascia, “held tension” — is exactly the tissue that is richly innervated. Fascia is now documented as one of the most densely innervated tissues in the human body: recent anatomical work (Fede et al., 2020, ) places its nerve density second only to skin. Viscera are an even clearer case — visceral afferents, traveling mainly through the vagus, are the primary signal interoceptive-inference models depend on. The phrase points to the wrong tissue — and the separate body it describes is proving hard to find.
There’s a genealogy worth naming here too. I drew on that same lineage — Maturana, Varela, Capra directly — in a paper I wrote in 2007, arguing that “the mind/body split is artificial… mind is not a thing but an embodied process.” Friston’s own framework descends from that same tradition. His 2013 paper, “Life as we know it,” cites Maturana and Varela’s foundational work on autopoiesis. And in 2019, he went further still, publishing a paper arguing that his own theory of the predicting brain and the older tradition of the embodied mind are, in his words, one and the same thing. So it’s worth noting, nearly two decades later, that a title separating brain from body is quietly stepping back from a lineage its own author has elsewhere insisted cannot be separated at all.
Whichever side of this argument you land on, some things hold regardless: trauma is felt in the body, and breath, posture, touch, and safety remain legitimate targets for healing under either framework. And van der Kolk’s phrase, even where its literal claim is now contested, still does something no diagram can — it validates a survivor’s lived sense that trauma is not simply in their head, which is exactly what motivates the body-based work both sides agree is worth doing. What changes is the language we reach for next: not that something is permanently damaged and waiting to be found, but that an old prediction is still running and can, with patience, be updated. That distinction isn’t just semantic — it’s the difference between feeling broken and feeling in process.
What this means for embodied practice. This isn’t only an argument I teach — it’s one I’ve lived. I know, from my own experience of significant trauma, what it means to carry an old danger the body won’t stop rehearsing, and what it takes, slowly, to convince a nervous system otherwise. That knowledge, long before I had this language for it, is what sent me toward ancient wisdom traditions — yoga, qigong, breath, stillness — and it’s what I bring into the room now, alongside the neuroscience: with clinicians and educators, people healing from deep trauma or chronic illness, and anyone seeking real, embodied tools for resilience, aging well, and inner steadiness.
Autopoiesis, the tradition Maturana and Varela gave us, describes a living system continuously creating and renewing itself — not just physically, but emotionally, and across an unfolding journey of consciousness. Slow, attentive movement, breath, and touch are how I’ve learned, in my own body first, to practice that continuity directly: feeding the nervous system present-moment evidence of safety that can compete with an outdated prediction of danger, while learning, experientially, that a state is a state — not a fixed identity or a permanent scar. Neuroscience now confirms what these traditions taught me by other names, long before I could name the mechanism: the body was never separate from the mind predicting through it, and a nervous system convinced of old danger can be gently, patiently convinced otherwise.