Before sharing anything, ask yourself one question, "Is this accurate?", because most sharing of false content comes not from belief but from inattention to accuracy, and the question alone measurably improves what people pass on.

How to do it

1. 1Make the question a fixed step between reading and sharing: before you hit share, retweet, forward, or send to the family chat, rate the headline's accuracy for yourself on a 1-to-4 scale.
2. 2Notice what you were actually about to share it for. Anger, humor, "this proves my point", and "my friends will like this" are the usual reasons; none of them is accuracy.
3. 3If your accuracy rating is below "probably true", do not share, or share with the doubt stated ("I have not verified this").
4. 4Put the prompt in your environment: a note on the monitor, a browser extension that delays posting, or a group-chat rule that links come with a one-line "how I know this is real".
5. 5If the self-check is hard to trigger, rate one unrelated headline first; the studies used a single unrelated accuracy judgment to shift attention, and the effect carried over.

When to use it

• •Any time you are about to share news, a screenshot, a statistic, or a quote on social media or in a group chat.
• •Forwarding health, safety, or emergency information to family.
• •When content arrives that is perfectly aligned with what you already believe and makes the other side look absurd.

Counters

Evidence and how strong it is

Pennycook et al. (2021) found a disconnect between judgment and sharing: participants rated true headlines as far more accurate than false ones, yet accuracy barely predicted what they said they would share. A single prompt to rate the accuracy of one unrelated headline reduced sharing intentions for false headlines, and a field experiment delivering the prompt by direct message to thousands of Twitter users who had previously shared low-quality news improved the quality of the news they subsequently shared. Pennycook & Rand (2022) pooled twenty experiments and found the effect replicable across partisan groups and headline sets, with a reduction in false sharing of roughly ten percent relative to control. Evidence strength: strong for the direction, small for the size. A preregistered direct replication (Roozenbeek, Freeman & van der Linden 2021) found a substantially smaller effect than the original COVID-19 study, and the intervention shifts sharing of false content, not belief in it.

1. Pennycook, G., Epstein, Z., Mosleh, M., Arechar, A. A., Eckles, D., & Rand, D. G. (2021). Shifting Attention to Accuracy Can Reduce Misinformation Online. Nature, 592, 590-595 · link
   The survey experiments and the Twitter field experiment showing that an accuracy prompt improves the quality of shared content.
2. Pennycook, G., McPhetres, J., Zhang, Y., Lu, J. G., & Rand, D. G. (2020). Fighting COVID-19 Misinformation on Social Media: Experimental Evidence for a Scalable Accuracy-Nudge Intervention. Psychological Science, 31(7), 770-780
   The original demonstration of the accuracy nudge on COVID-19 headlines.
3. Roozenbeek, J., Freeman, A. L. J., & van der Linden, S. (2021). How Accurate Are Accuracy-Nudge Interventions? A Preregistered Direct Replication of Pennycook et al. (2020). Psychological Science, 32(7), 1169-1178
   A direct replication finding a much smaller effect, the basis for rating the effect size as small.
4. Pennycook, G., & Rand, D. G. (2022). Accuracy Prompts Are a Replicable and Generalizable Approach for Reducing the Spread of Misinformation. Nature Communications, 13, 2333
   Pooled analysis of twenty experiments showing the effect replicates across groups with a modest average size.