Abstract
<jats:p>This paper proposes that persistent maladaptive fixation is characterised not by the presence of a pathological content but by a reduction in the dimensionality of an individual’s response space within a specific content domain. Under overwhelming demand, a state capable of supporting multiple context-dependent responses is projected onto a single response and loses its sensitivity to context. What distinguishes fixation from an ordinary strong reaction is that this projection persists after the demand ends: the state does not re-open. Fixated material is therefore not an unresolved state awaiting collapse but an already-collapsed one, and therapeutic action is reframed as the restoration of degrees of freedom rather than the removal of a stored entity. This proposal responds to a gap. Attention-guided and somatically oriented interventions for persistent stress-related patterns are widely practiced and frequently report effects, yet the field lacks a unifying theoretical account of what such interventions would be changing. We offer one and specify how it could be shown to be wrong. The account is developed in the quantum-like formalism established in quantum cognition — Hilbert-space state representation, non-commuting observables, and contextual measurement — used as a modelling language for psychological state dynamics. No physical quantum process in neural tissue is claimed or required, and entanglement-based accounts of interpersonal therapeutic action are excluded on principled grounds. Reduced contextual sensitivity is already predicted by the existing compatibility account for extensively rehearsed material. The framework’s distinctive claim is therefore not about contextual sensitivity alone but about its joint pattern with autonomic complexity: fixated and merely familiar domains, indistinguishable by order effects, are predicted to be separable by state-dependent heart rate variability complexity. Falsification conditions are stated, and the decisive test targets this claim.</jats:p>