{
  "study_id": "NCT07611422",
  "agent": "framework",
  "run": 1,
  "p_primary_hypothesis_supported": 0.27,
  "effect_direction": "favours intervention",
  "effect_size_90pct_interval": {"low": -0.15, "high": 0.6, "metric": "Cohen's d, positive-information vs neutral-information on PSS-10 at 4 weeks (positive = less stress with positive information)"},
  "moderator_prediction": "Cold-shower participants respond more to positive information: the cold is an arousing body render that gives the rationale something felt to confirm, and 'cold builds resilience' fits existing cultural belief, whereas a resilience rationale for ordinary warm showers is less plausible and has no salient sensation to verify it.",
  "claim_ids_used": ["PM-0056", "PM-0128", "PM-0102", "PM-1353", "PM-0851", "PM-0159", "PM-1286", "PM-0113", "PM-0110", "PM-0218"],
  "rationale": "The expectancy manipulation is a single spoken rationale to alert adults, which is one small deposit paying the normal tax (PM-0056, PM-0128). Two things help it: four weeks of repeated enactment with a daily diary (PM-0102, PM-1353, PM-0851), and a self-rated outcome that the Prover can read as confirming (PM-0159, PM-1286). So the direction should favour positive information. But PSS-10 is a general stress-appraisal measure, and the effect is probably small (d about 0.2). With about 60 per level and snowball recruitment, power to detect a main effect at alpha 0.05 is low. The effect is likely concentrated in the cold arm (PM-0113, PM-0110, PM-0218), which dilutes the main effect."
}