If retracting even high‑profile fraudulent studies doesn’t topple theories, that can mean core findings are supported by many independent results. The right lesson isn’t that a field is empty, but that single studies—however flashy—aren’t load‑bearing in a cumulative science.
— This reframes the replication crisis narrative and guides media, funders, and policymakers to judge fields by the strength of converging evidence rather than the fate of headline papers.
2025.10.07
70% relevant
By estimating that only about 17.7% of significant psychology results are likely false despite publication bias and p-hacking, the paper supports the notion that core findings in a field can remain robust even amid fraud and replication failures, rather than implying wholesale collapse.
Paul Bloom
2025.08.19
100% relevant
Bloom’s response to Adam Mastroianni cites Diederik Stapel’s fraud and contrasts it with robust effects like the word‑frequency advantage.
← Back to All Ideas