Why a correction travels slowly

In 2018 a group of researchers tracked how stories spread across a large social network. Their central finding was easy to state and hard to accept: false stories reached more people, and reached them faster, than true ones. The gap was not small. False political stories were roughly six times quicker to reach fifteen hundred people.

The explanation most readers reach for is automation, and it is wrong. The researchers checked. Removing the automated accounts barely changed the pattern, because the automated accounts shared true and false stories at similar rates. The acceleration came from people.

Why would people behave this way? The researchers' answer concerns novelty. False stories were, on average, more surprising than true ones, and surprise is the quality that makes a person want to pass something on. A story that confirms what everyone already believes is not worth sharing. A story that overturns it is.

This has an awkward consequence for corrections. A correction is, by its nature, unsurprising: it returns the world to how it was. It carries none of the energy that moved the original claim, so it reaches a fraction of the same audience. Newspapers have known this for a century — the correction appears on page nineteen because almost nobody will look for it — but the pattern is sharper online, where sharing is the whole mechanism of distribution.

The practical lesson is not that readers should be ashamed of being interested in surprising things. Curiosity is not a fault. The lesson is that the moment a story feels remarkable is exactly the moment to slow down and ask who is reporting it, because that feeling is doing the work the evidence has not yet done.