Explainer · 2 min read
Why should a discovery be found again?
Published on
An experiment produces a surprising result. The calculations are correct and the paper is published. Why ask another team to repeat it?
Because checking a calculation and finding a phenomenon again are two different operations.
In the first case, researchers use the same data to check that the analysis produces the reported result. In the second, they collect new data, with other participants or at another time, to see whether the effect reappears.
The SCORE programme examined these questions in the social and behavioural sciences. In the part involving new experiments, around 55% of the 274 claims tested produced a statistically significant result in the original direction. This figure does not represent the proportion of theories that are true: it depends on the studies selected and the criterion for success.
The challenge is the same in psi research. Researchers need to state what they want to find again: a departure from chance, an effect size or a change linked to a specific condition.
Replication helps establish what we can rely on and under what conditions. A failed replication calls for an investigation of the protocol, the precision of the measurements and the strength of the original result.
Further reading: Bulletin of 25 September 2026.
Sources: Tyner et al. (2026), “Investigating the replicability of the social and behavioural sciences”, Nature, article ; Miske et al. (2026), “Investigating the reproducibility of the social and behavioural sciences”, Nature, article.