The freeze response: when doing nothing is the most expensive survival strategy

Freeze is not absence.
It is a precise piece of work.

The freeze response is the third arm of the defense cascade, alongside fight and flight, and it is the one that gets deployed when the other two have no positive expected value. If running is impossible because the threat is faster, larger, or already in physical contact, and fighting is impossible because the threat is overwhelming or because resistance will increase the harm, the nervous system selects freeze. This is a calculation, not a failure. The system is choosing the least costly available error.

What freeze actually does, at the physiological level, is reduce the metabolic and motoric signature of the organism. Heart rate drops. Muscle tone shifts from active engagement to tonic immobility. Pain perception attenuates. Vocalisation suppresses. The dorsal vagal complex takes a more active role in autonomic regulation, often described, accurately, as a kind of biological shutdown (Porges, 2007). In the original adaptive context, this is precisely the right answer. A predator orients to motion. A perpetrator escalates against resistance. A child confronts a parent who is the source of safety and the source of threat at once and has no other available regulatory partner. Freeze is what the nervous system has when nothing else will work.

The defense cascade literature (Kozlowska et al., 2015) describes freeze as one node in a sequence that the system runs in fractions of a second, mostly outside cortical awareness, and that includes attentive immobility, fight-or-flight, tonic immobility, collapsed immobility, and quiescent immobility as related but distinct states. From the inside, all of these can feel like the same thing: a sudden inability to act, often accompanied by a sense of distance from the body, often described later as “I just went blank.” From the outside, the patient looks calm. From the architecture, the system is running its most conservative protocol, with full physiological commitment, against a perceived threat the cortex has not yet caught up with.

The contraction can stay
long after the event.

What happens to a freeze response after the threat ends is the part most clinical conversations underestimate. In the original event, freeze was selected, deployed, and produced an outcome. The patient survived. The prediction circuit registered, correctly, that this strategy worked. The pattern was archived. The next time a sufficiently similar cue arrives, the pattern fires again, because the system has no reason to consider any other strategy.

The cue does not need to look like the original event. It needs to share the structural properties the prediction circuit was tracking. A particular tone of voice. A particular spatial configuration. A particular asymmetry of power in a room. The cortex sees a polite conversation. The subcortical archive sees the precondition of the original event, and it deploys the response that worked then.

This is why patients with strong freeze histories report going blank in mundane settings. Job interviews. Doctors' offices. Conversations with parents. Sometimes intimate ones. The blank is not random. It is the prediction circuit reading something the cortex did not consciously register, and producing the response that, at the moment the pattern was set, was the right one.

The cost is the part patients carry without naming it. Energy goes into holding the contraction that has not been signaled to release. Sleep does not restore. Body wakes already braced. Decision-making slows in the categories the pattern is monitoring. None of this looks like trauma in the screening room. All of it is trauma running its assigned protocol, in a context where the protocol is no longer adaptive but has not been retired.

The release signal does not come
from cognitive insight.

Knowing that you freeze does not retire the freeze pattern. This is the same structural problem described in the related mechanism literature: cortical recognition does not propagate down to a subcortical pattern, because the pattern is not weighted to update on cognitive content (the architecture of the gap is described here in detail). The patient who can name the response in real time, who can describe the trigger, who can predict the duration of the blank, still goes blank.

What modifies a freeze pattern is a felt sense of completion that the original event did not provide. The freeze was deployed because action was not possible. The system never registered the discharge of activation that a successful fight or flight produces. The activation was loaded and held, indefinitely. Releasing the contraction requires giving the system the signal it never received: the signal that the activation can now move through, in some channel, with the patient in the room and not overwhelmed.

This is precise work. It cannot be done by talking about the freeze, no matter how accurately. It cannot be done by relaxation techniques imposed on top of the contraction, which the system will register as another override, not as completion. It is done by introducing a small, specific motoric or somatic experience that the prediction circuit can read as the missing discharge. The form this takes varies. The principle does not.

What changes when the work happens
at the right layer.

In a small clinical-phenomenological study (Laugman, 2026), patients whose freeze patterns had been continuously running for years described a recognisable set of shifts when the work targeted the subcortical layer rather than the cognitive surface. The specific markers showed up before any change in self-report on standard scales.

Breath deepened. Not because anyone instructed it to. The intercostal muscles, which the contraction had been holding, released slightly, and the breath moved further down. Warmth returned to extremities that had been chronically cool. Small motor impulses arose that the system had been suppressing, often described as a wish to push, to stretch, to make sound. The cortex registered these changes as “something is happening that was not happening before,” usually without language for what.

What changed structurally was the prediction. The system, having received the discharge signal it never had, began to weight the pattern slightly differently. The next sufficiently similar cue did not produce the same intensity of freeze. The release was not total, and not all at once. Over weeks, the system moved from defaulting to freeze under marginal threat to defaulting to a wider range of responses, including ones the patient had not been able to access in years.

The contraction releases when it receives the signal that the event is over and that the body can now do what it was holding. That signal is what the original moment did not provide, and what the years of cognitive work, however precise, were not built to deliver. It comes from a different layer, and the work that reaches that layer has its own protocol. The freeze pattern, treated on its own terms, does not need to be argued with or interpreted. It needs to be given the closure the original moment denied it. Once it has that, the system stops defending against an event that ended a long time ago, and the energy that was holding the contraction becomes available for everything else.