nucleation

THEORY-912: In a picture-based avoidance task, treatment-seeking patients with OCD or PTSD avoided a safety cue more than healthy controls and, after effective avoidance, still expected the aversive picture more; patients with panic disorder with agoraphobia or PTSD reported more relief when it was omitted; the groups did not differ in avoiding the danger cues

3 min read

Properties6
StatusActive
Version2
Tagspsychopathology-and-treatment, learning-and-conditioning
SourceLIT-1490 — Impaired action-safety learning and excessive relief during avoidance in patients with anxiety disorders
Bears onQUESTION-331 — In a second, preferably pre-registered sample with a cost attached to avoiding, do anxiety-disorder patients again avoid a safety cue more than controls but danger cues no more, keep a higher threat expectancy after effective avoidance, and report more relief at omissions, with rated aversiveness matched or controlled?
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

De Kleine, Hutschemaekers, Hendriks, Kampman, Papalini, Van Minnen & Vervliet, LIT-1490, read in full in NOTE-1123 from the publisher’s version of record (CC BY 4.0): pp. 3–8, Table 1 and Figures 2–3.

What was actually shown

  • Avoidance. A Group × CS interaction, driven by OCD (p = .004) and PTSD (p = .028) patients pressing the avoidance button more than controls to the safety cue. “No significant differences between groups were found in reaction to the CS+ ’s” (p. 5).
  • Expectancy. OCD and PTSD patients rated the aversive picture as more expected than controls did after the safety cue and the avoidable danger cue. They also did so on trials where they had pressed the button to the avoidable cue (OCD p = .045, PTSD p < .001; p. 6).
  • Relief. PDA and PTSD patients reported more relief at omissions than controls, and PTSD patients more than the other two patient groups. This held with rated aversiveness as a covariate (p. 7).
  • Before avoidance. In the Pavlovian phase the groups did not differ in expectancy, but relief to the safety cue already differed in the same direction (p. 5).

What could have come out otherwise: patients could have avoided the danger cues more, as the authors predicted for the unavoidable one. Or they could have learned the action-safety contingency as well as controls, with any difference confined to avoidance rates.

What this does not say

  • That excess relief reinforces avoidance. That is the authors’ account (p. 7). Relief was rated, and reinforcement was not measured. Relief was elevated before avoidance existed in the task.
  • That the pattern maintains the disorders. The design is cross-sectional. The authors: it “does not allow testing whether the observed effects and proposed mechanism are causal to the disorder” (p. 8).
  • That it is diagnosis-specific. Two of three groups showed each effect, and comorbidity crossed the groups.
  • That learning, rather than threat appraisal, differs. Patients rated the pictures as more aversive, and the authors raise this as a rival explanation (p. 7).
  • Anything about costly avoidance. Pressing the button cost nothing.

Connections

  • CLAIM-446. It supports the clause that relief reinforces the defence without being evidence of danger, weakly and only through the authors’ interpretation. It complicates the clause about identifiability: patients avoided without trusting the avoidance, which is a different failure from learning that avoidance works while learning nothing about the world without it.
  • THEORY-285. A different failure from the model’s. There the agent stops sampling. Here patients sampled as often as controls and drew less from the outcomes.

Backlinks