{
  "study_id": "NCT05351541",
  "agent": "framework",
  "run": 3,
  "p_primary_hypothesis_supported": 0.8,
  "effect_direction": "favours intervention",
  "effect_size_90pct_interval": {"low": 2, "high": 20, "metric": "Mean change in BPI interference, 0-70 scale, baseline to 1 month (positive = reduction)"},
  "moderator_prediction": "Psilocybin + placebo (and modafinil) arms respond more than zolpidem-containing arms: zolpidem sedation blunts attention and the intensity of the session, lowering the impact term that makes the single session a deep write; pain interference is a felt, self-rated outcome that the render layer reaches.",
  "claim_ids_used": ["PM-0369", "PM-0401", "PM-0092", "PM-0090", "PM-0485", "PM-1286", "PM-0159", "PM-0146", "PM-0946", "PM-0104", "PM-0110"],
  "rationale": "The primary test is effectively a pooled pre-post change in a self-rated felt outcome after a single psilocybin session with preparation and integration, in a therapeutic setting. The framework treats psychedelics as a critical-faculty bypass that relaxes priors, and one saturated session as a heavy write; setting is causal; pain interference sits at the render layer and is self-reported, where the Prover reads ambiguous signals as improvement. Add regression to the mean and expectancy in an unblinded-to-psilocybin sample and a significant pooled drop at 1 month is likely even with n~30. Risks: variable dosing (1 mg floor), zolpidem blunting, small n, and fade by 1 month without daily verification."
}