# Timing matters when correcting fake news

An interesting aticle in PNAS: https://www.pnas.org/content/118/5/e2020043118 / [DOI]([https://doi.org/10.1073/pnas.2020043118](https://doi.org/10.1073/pnas.2020043118))

## Objective
> Countering misinformation can reduce belief in the moment, but corrective messages quickly fade from memory. We tested whether the longer-term impact of fact-checks depends on when people receive them.

## Conclusions

> Providing fact-checks after headlines (_debunking_) improved subsequent truth discernment more than providing the same information during (_labeling_) or before (_prebunking_) exposure.

> (debunking) reduced misclassification of headlines 1 wk later by 25.3%, compared to an 8.6% reduction when tags appeared during exposure (labeling), and a 6.6% increase or 5.7% reduction when tags appeared beforehand (prebunking).

This is interesting. My guess would be that labelling would work the best. I thought that when the information is "baked into the memory" of a person, it's harder to change.

Perhaps, post-fact debunking/knowledge revision actually has a more profound effect on an individual because if it works, it probably does not only fix the incorrect information received, but also stains the source from where the person received it.

Still, it's notable that while the effect of fact-checking was significant, by far it did not have the effect most people would hope for: once you learn that what you believed was wrong, you fix yourself - i.e., 100% success rate. For false information, debunking it later fixed only a minority of misinformation cases.

They used Amazon Mechanical Turk for the experiment, so it might also be there are some biases stemming from self-selection of the participants.