How do you spot fake news before you share it?

Summary: Most people can already spot fake news when they stop and ask whether a headline is true. The trouble is that we rarely ask before we share, so the most useful habit is a short pause with a few checks, run every time a headline makes you want to react.

The usual advice treats fake news as a detective problem: Learn the clues, spot the fraud. Research points somewhere else. When people are asked to judge whether headlines are accurate, they sort true from false fairly well. When they're asked what they'd share, that skill goes quiet. Spotting fake news depends less on what you know than on whether you bother to check in the moment that counts.

Why do people share headlines they don't believe?

In a 2021 study published in Nature, psychologists Gordon Pennycook, David Rand, and their colleagues showed 1,015 Americans a mix of real and false headlines. Half the group rated each headline's accuracy. The other half said whether they'd consider sharing it.

The raters did well. True headlines were judged accurate 55.9 percentage points more often than false ones. Whether a headline matched the reader's politics mattered far less, at 10.1 points.

The sharers flipped that pattern. Whether a headline was true moved sharing by only 5.9 points, while political fit moved it by 19.3 points. Participants were more than twice as likely to consider sharing false headlines that flattered their side (37.4%) as they were to rate those same headlines accurate (18.2%). Most of them said, when asked at the end, that sharing only accurate content was important to them.

So the readers in this study weren't fooled. They knew a false headline looked false. They shared it anyway, because a feed rewards speed, agreement, and a quick laugh, and none of those prompts the question "Is this true?"

What happens when someone asks that question first?

The same researchers ran the obvious test. Before the sharing task, some participants rated the accuracy of a single unrelated headline. That one prompt made their later choices more careful than those of people who skipped it: The gap between how often they'd share true headlines and how often they'd share false ones grew 2.0 times larger in one experiment and 2.4 times larger in another.

They then sent a similar one-question message to 5,379 people on a large social platform who had recently shared links from sites that fact-checkers rate as unreliable. Afterward, the sites those users shared rated slightly higher for quality on average than the sites they'd shared before the message. The shift was small, and it came from a single message nobody was required to answer.

The prompt worked because people already had the skill. It only turned their attention toward accuracy for a moment.

Is fake news really that easy to catch?

Some fake news is hard to catch, even with full attention. Plenty of misleading content is built to pass a quick look: Real photos with false captions, accurate numbers from the wrong year, opinion dressed as reporting. A 2018 Pew Research Center survey of 5,035 US adults asked people to sort 10 statements into facts and opinions. A majority got at least 3 of 5 right in each set, which Pew noted was only slightly better than chance. Even among the most politically aware, only 36% labeled all 5 factual statements correctly, against 17% of the least aware. Both Republicans and Democrats were more likely to call a statement factual when it favored their side.

A pause can't supply knowledge you lack. If a forged document looks real to experts, a moment of reflection won't expose it. Who spreads fakes also depends on more than attention. A 2019 study in Science Advances by Andrew Guess, Jonathan Nagler, and Joshua Tucker linked survey answers to social media sharing during the 2016 election. Only 8.5% of the linked users shared any article from a fake news site, and users over 65 shared nearly 7 times as many of those articles as the youngest group. The age gap held after the authors accounted for ideology, education, and how many links people shared overall.

Those limits are real. We'd bet the pause still catches most of what reaches you, because the falsehoods that go viral tend to spread on emotion and speed, the exact condition the pause interrupts. For the sophisticated fakes, the pause gets you to the slower checks, like searching the claim somewhere else. Skip the pause and those checks never happen.

How do you spot fake news in practice?

Each check takes under a minute, whether the headline arrives on a front page, in a group chat, or at the top of a search result.

1. Ask "Is this accurate?" before anything else. Say it in your head before you share, forward, or repeat a claim. This is the step the Nature study tested, and it does most of the work.

2. Notice the feeling. A headline that makes you furious or vindicated deserves more doubt than one that bores you. Strong agreement is the moment confirmation bias does its best work.

3. Leave the page. Open a new tab and search the claim, the outlet, or the author. Fact-checkers call this lateral reading, and it beats studying the page itself for clues.

4. Find the original. Look for who said it first, in what document, and on what date. A real quote, report, or dataset can be located. A claim that only exists in screenshots of other posts is a warning.

5. Separate the fact from the frame. Ask whether each sentence could be proven or disproven. "The bill passed 52 to 48" can be checked. "The bill is a disaster" is an opinion, however confident it sounds.

6. Wait when you're unsure. Early reports on breaking news often change as facts come in. Holding a share for a day costs you nothing, and early errors often get corrected by then.

The same steps also help with the subtler problem our post on misinformation describes: False claims spread by honest people who never meant to mislead anyone.

Does reading less news help?

Partly. Every headline you meet in a rush is one more chance to skip the pause. Reading fewer headlines, in a calmer setting, gives each one more of your attention than it would get in a feed of hundreds. A morning paper read at the kitchen table does this. So does a short list of sources you check once a day instead of a feed you check 40 times.

That's the idea behind informed.now. It's one text every weekday morning with 6 news summaries, picked with one transparent filter, and we publish every source we cite. You read it once, when you have a minute to think, instead of meeting the news in fragments between other tasks.

The skill is already there

Most people who share fake news could have caught it with one question they never asked themselves. The research shows the ability sitting unused, switched on by a single prompt. Ask that question before you share, and most of the work of spotting fake news is done.