{
  "study_id": "NCT07133529",
  "agent": "framework",
  "run": 5,
  "p_primary_hypothesis_supported": 0.27,
  "effect_direction": "favours intervention",
  "effect_size_90pct_interval": {"low": -0.2, "high": 0.5, "metric": "Cohen's d (within-subject, crossover) for uncertainty minus certainty in buffet intake"},
  "moderator_prediction": "Stress eaters (high SSES) and those with larger subjective stress/cortisol rises under uncertainty eat more, because an aroused, threat-primed receiver prices the scarcity signal higher; interaction likely underpowered at n=45.",
  "claim_ids_used": ["PM-0585", "PM-0321", "PM-0601", "PM-0845", "PM-0113", "PM-0289", "PM-1077"],
  "rationale": "Withholding meal timing is a mild scarcity/threat signal; threat weighs heaviest (PM-0585) and an unfed imprint widens desire (PM-0321), so the framework predicts more intake under uncertainty. But the manipulation is weak: a lab visit with a known buffet reaching no deeper than an opinion-level expectation (PM-1077), in healthy, fasted women where hunger dominates both arms (PM-0113). Buffet intake is noisy, n~45 crossover, and 28 primary outcomes mean the effect may land on adjacent measures like stress or craving rather than grams eaten (PM-0289). Arousal amplifies pricing only for responsive participants (PM-0601, PM-0845). Expect a small positive d (~0.15), likely non-significant."
}