How to Check a Viral Science Claim in Five Minutes

How do you check a viral science claim? Start by leaving it. Professional fact-checkers don’t stare harder at a suspicious page. They open new tabs and look up who is behind it. In five minutes you usually can’t prove a claim true or false. But you can often learn whether it deserves your trust, your worry or your share button.

Think of it like buying a used car. You don’t just trust the seller’s ad. You look up the car’s history, ask a mechanic and read what other owners say. Here is a friendly checklist for science claims, plus four simple boxes for sorting what you find.

Who is behind the claim, and what do others say about them?

Researchers at the Stanford History Education Group, now called the Digital Inquiry Group, watched 45 experienced internet users judge real websites. They included 10 professional fact-checkers, 10 historians with PhDs and 25 Stanford undergraduates. The historians and students tended to read “vertically.” They stayed on a site and were often fooled by official-looking logos and web addresses. The fact-checkers read “laterally.” They glanced at a site, then left it to see what other sources said about it. The study found that the fact-checkers reached better-supported conclusions in a fraction of the time.

The good news is that lateral reading can be taught. In a later field study in one city school district, high school students who had six class lessons in it improved significantly at judging online content, compared with similar students who didn’t. The group’s Civic Online Reasoning lessons are free.

Try it now. Search the name of the website, group or “expert,” perhaps adding words like “funding” or “criticism.” Who pays for it? Does it sell something tied to the claim?

Is there an original source, and has anyone checked it?

A primary source is the original: the study, the data or the official statement. A news story or video about a study is secondary. Try to find the original. Was it published in a peer-reviewed journal, meaning independent experts reviewed it first? Or is it a preprint, shared publicly before review? Preprints can be valuable, but they are first drafts.

Then ask whether anyone else has repeated the result. In 2011 the OPERA experiment reported that neutrinos, tiny ghostly particles, sent from CERN near Geneva to a lab in Italy about 730 kilometers away, seemed to arrive about 60 billionths of a second early. That would mean they moved faster than light. The team did the responsible thing: it asked for scrutiny and independent measurements. By June 2012, four experiments, including OPERA itself, had measured travel times consistent with the speed of light. The original result was traced to a faulty part in the experiment’s fiber-optic timing system. (For more, see Can Anything Travel Faster Than Light?)

Also look for independent experts, meaning scientists who were not part of the study and have nothing to sell. If they are cautious and the headline is certain, trust the caution.

Is that picture what it claims to be?

Many viral posts pair a claim with a real photo from somewhere else: an old storm, another country, a movie set, or an artist’s illustration passed off as a photograph. A reverse image search helps. You upload a picture, or paste its link, and the tool finds where else it appears online. Google’s image search and TinEye are free. Look for the earliest copy and its original caption. If an image seems to have no traceable origin at all, treat it with extra care.

What was seen, and what was guessed?

This is the heart of the check. Sort what you have found into four boxes. Here they are, filled in with a famous example: the “Face on Mars.”

1. What was observed? On July 25, 1976, NASA’s Viking 1 orbiter photographed a hill in the Cydonia region of Mars. In that fuzzy image, the angle of the sunlight and the shadows made it look like a human face. According to ESA, a NASA press release said it resembled a human head, though NASA scientists had already read it as a trick of light and shadow.
2. What was inferred? Some people concluded it was an artificial monument, perhaps built by aliens. That is an interpretation, not an observation.
3. Does the evidence support it? No. Much sharper images from NASA’s Mars Global Surveyor in 1998 and 2001, and from ESA’s Mars Express in 2006, show an eroded natural hill. NASA describes it as one of thousands of buttes, mesas, ridges and knobs in that region.
4. What is unresolved? Very little about the “face” itself. The bigger question it stirred, whether Mars ever had life, remains open. NASA’s Perseverance rover is searching for signs of ancient microbial life. In 2025 its team reported a rock sample with “potential biosignatures,” clues that need more study before anyone can say whether life was there. A face-shaped hill doesn’t answer that question either way.

You can use the same four boxes on any claim, from a new diet study to a strange light in the sky.

What can’t a five-minute check tell you?

A quick check is a filter, not a verdict. It can’t redo an experiment or reanalyze data. Some claims are brand new, and nobody has had time to test them. Sometimes honest experts disagree, and the true answer is “we don’t know yet.” A claim can also fail a quick check and still turn out to be right later. And a claim can come from a trustworthy source and still be wrong. OPERA was a serious team using serious equipment.

What would change your mind?

Before you decide, ask one more question: what evidence would make me change my view? Good reasons to shift include independent teams getting the same result, a primary source that supports the headline, and experts with no stake in the outcome agreeing. On the other side, a failed repetition, a missing original source or an image found in an older, unrelated context should lower your trust. For a deeper look, see What Would Count as Evidence?

We try to hold this site to the same standard. For example, the HD-Blast idea in Ricardo Maldonado’s book is his proposal that our hot early universe may be the local aftermath of a brief higher-dimensional blast. It has not been confirmed, so we label it as the author’s hypothesis. You are welcome to run the four boxes on it at our hypotheses page. You can also read our editorial standards.

Curiosity is a good instinct. So is a pause before sharing. Next time a claim makes you gasp, take five minutes, open a few tabs, and ask what was really seen. Then keep asking.