nucleation

THEORY-910: Brown, Price and Dombrovski propose, without a model run or data, that in aversive environments avoiding uncertain states keeps their uncertainty high and so perpetuates avoidance, that anxiety strengthens this loop by misreading noise as change, and that exposure works by reducing the uncertainty rather than an excess of learned threat

3 min read

Properties6
StatusActive
Version2
Tagspsychopathology-and-treatment, learning-and-conditioning
SourceLIT-1479 — Anxiety as a disorder of uncertainty: implications for understanding maladaptive anxiety, anxious avoidance, and exposure therapy
Bears onQUESTION-325 — Is persistent avoidance in anxiety maintained by an inflated uncertainty estimate about the avoided states, as Brown, Price and Dombrovski propose, or by a pessimistic evaluation of one's own future avoidance, as Zorowitz, Momennejad and Daw propose, rather than by an inflated threat value?
CorroboratesCLAIM-446 — Some persistent anxiety may be governance-maintained rather than belief-maintained: anticipated threat grants defensive processes authority over action, attention and evidence acquisition; under the avoidance that authority selects, the hypotheses that the situation is dangerous and that it is safe predict the same observation, so repeated safety need not correct the threat estimate; relief reinforces the defense without being evidence of danger; and the regime can outlast a corrected belief, its pathology, where present, being defensive authority insensitive to corrective evidence at avoidable cost

Graph View

Source

Brown, Price & Dombrovski, LIT-1479, read in full in NOTE-1120 from the NIH author manuscript (PMC10475148): the sections from “The explore-exploit dilemma and avoidance” to “Practical advantages of incorporating uncertainty into anxiety treatment”, and Figures 3 and 4.

What was actually shown

Nothing was shown in the sense of a result. The paper argues:

  • In aversive environments people shift toward avoiding short-term losses, and “normative behavior in uncertain, aversive MDPs will show avoidance of uncertain states from increased exploitation of safer states. This avoidance maintains high uncertainty and perpetuates avoidance by preventing exploration of more uncertain states.”
  • In anxiety, misreading noise as estimation or unexpected uncertainty raises exactly the uncertainty that reduces exploration under threat: “Continued avoidance of uncertain states will maintain inappropriately elevated uncertainty estimates, ensuring continued avoidance.”
  • Exposure is forced exploration of the avoided states; it lowers their uncertainty, and lower uncertainty in turn favours exploration. Because values update at the end of an episode, the update shows between sessions, which the authors offer as the reason between-session fear reduction predicts outcome.

What could come out otherwise, if tested: avoidance might track the expected threat of the avoided option and not its uncertainty, or lowering uncertainty with threat fixed might leave avoidance in place.

What this does not say

  • That the loop has been demonstrated. No agent is simulated and no participant is tested. The figures are schematics with illustrative values.
  • That the threat estimate is wrong. The account is built so that avoidance can persist with a correct expected threat and an inflated uncertainty estimate. It is not an excess-threat account.
  • That anything other than a belief maintains the loop. The variable that keeps avoidance going is the uncertainty estimate. There is no separate process granted influence over action or sampling.
  • That it covers anxiety in general. Fear-based disorders only; the authors say it does not explain exposure in GAD or eating disorders.

Connections

  • CLAIM-446. Supports the step “the defense keeps the uncertainty unresolved”, and the claim’s caution that defensive policy can be rational. It is also the belief-maintained rival the claim’s defeater names: a loop of the same shape with no authority variable. A test of the claim has to hold the uncertainty estimate fixed as well as the represented threat.
  • QUESTION-045. A candidate maintaining link (avoidance of uncertain states) and a candidate cut (exposure), at the level of verbal theory.
  • THEORY-285. The same self-sealing shape in active inference, where an over-precise prior makes sampling look uninformative. Both are model-level; neither has individual data.

Backlinks