nucleation

THEORY-1179: In an active-inference model of post-traumatic context inference, an overly precise prior that one is in a harmful context makes exploratory sampling look uninformative, so the agent attenuates the periphery, never receives the evidence that would correct the prior, and the belief sustains itself; lowering precision removes the symptoms without correcting the prior

2 min read

Properties5
StatusProposed
Tagspsychopathology-and-treatment, neuroscience, complex-systems
SourceLIT-983 — Impaired context disambiguation as a key determinant of post-traumatic psychopathology: A neurocomputational account
ExtendsTHEORY-285 — In an illustrative active-inference simulation of post-traumatic context inference, run once per phenotype, a prior that one is in a harmful context, made overly precise by a run of harmful trials under the "vulnerable" learning and forgetting weights, makes exploratory sampling look uninformative, so the agent attenuates the periphery, never receives the evidence that would correct the prior, and the belief sustains itself; lowering precision makes the attenuation policy less vigorous without correcting the prior, and the policy returns when precision is restored
Bears onQUESTION-303 — After cued traumatic recall, do people with persistent PTSD sample disambiguating peripheral context less than matched trauma-exposed controls, and does restoring that sampling during the same exposure produce lasting change that the exposure alone does not?

Graph View

Source

Linson, Anderson, Markham, Friston & Moutoussis (2025), bioRxiv, LIT-983, version 2, read in NOTE-766 (Results, Discussion and Online Methods; not the figures or the supplement). What the simulations showed is THEORY-285, which this entry extends.

The claim

The paper recasts post-traumatic stress disorder as a failure to tell which context one is in, not a failure of fear extinction. Under active inference an agent chooses policies by expected information gain, so a prior already confident that the context is harmful makes looking at the periphery seem pointless: exploration is “falsely believed to be uninformative due to overly informative prior beliefs”. The agent attenuates the periphery, learns nothing about the harmless context, and keeps the prior: “As long as the sensory attenuation policy is selected, no amount of observations will render prior beliefs obsolete.” The collapse to one basin is “self-maintaining because it precludes epistemic foraging”. Exposure that restores sampling rebalances the prior, while a precision-lowering drug suppresses symptoms that return when precision does, a contrast the authors say “closely matches” the EMDR–fluoxetine trial. They state predictions for people: after cued traumatic recall, peripheral sensing is attenuated more than focal sensing.

What THEORY-285 holds is narrower: one illustrative run per phenotype, under learning and forgetting weights the paper does not report, and a drug arm that makes the attenuation policy less vigorous rather than removing it. The match to the trial is asserted by citation, not fitted.

What would settle it

The promote_when, which QUESTION-303 also asks: the authors’ prediction confirmed in people with context sampling measured directly, or an exposure-matched contrast in which restoring sampling produces lasting change and a precision-lowering drug does not. More simulations cannot settle it.

Backlinks