Compare two papers on the same question.
Where they agree, where they actually disagree (not where they use different words), and what would resolve the disagreement.
Two papers make different claims about a question you care about. Reading them both takes hours and you still might miss the specific point where they diverge. This prompt makes the disagreement explicit.
THE PROMPT
Compare two papers on [QUESTION]. Paste both below (abstracts + methods sections minimum; full text if you have it). Return four sections: **Same question, same words?** Restate what each paper actually claims to answer, in one sentence each. If they're not really answering the same question — flag it. **Where they agree** The specific findings where both papers land on the same conclusion, even if worded differently. **Where they actually disagree** The specific findings where their conclusions conflict. Name the numeric difference (e.g., "Paper A says X=0.7 CI [0.5-0.9], Paper B says X=0.3 CI [0.1-0.5]"). If they disagree only in emphasis rather than in findings, say that. **What would resolve the disagreement** The specific experiment, dataset, or method choice that would tell us which is right. If the disagreement is unresolvable from current data, say so plainly. Do not hedge. Do not add a "further research is needed" closer. Give a specific answer to which paper's methods you'd trust more on this question, and why. Papers: [PASTE PAPERS HERE]
**Same question, same words?**
Paper A asks whether daily creatine improves cognitive performance in healthy adults. Paper B asks whether creatine improves cognitive performance during sleep deprivation. Different questions — Paper B's population is stressed; Paper A's is not.
**Where they agree**
- Creatine dosing at 5g/day is well-tolerated with no significant adverse events at 12 weeks.
- Working memory shows the largest effect size across both populations.
**Where they actually disagree**
Paper A finds a 3% improvement on the digit-span task (CI [-1%, +7%], not significant at p<0.05). Paper B finds an 11% improvement (CI [6%, 15%], p<0.001) on the same task under sleep-deprived conditions.
**What would resolve**
The disagreement isn't actually a disagreement — both papers point at the same conclusion: creatine helps working memory more when cognitive reserves are depleted. A study that tests both conditions in the same subjects (crossover, ~4 weeks per arm) would confirm. Given current data, I'd trust Paper B's methods more (larger n, pre-registered, blinded outcome scoring).