When Correcting Someone Makes Things Worse — And What's Actually Going On

Picture this: your uncle is absolutely convinced that a certain health supplement cures whatever ails you. You show up to Thanksgiving with printed studies, a calm demeanor, and genuine good intentions. Two hours later, he's more convinced than ever and you're stress-eating pie in the kitchen. What just happened?

For a while, psychologists thought they had a clean answer: the backfire effect. The theory, popularized around 2010, suggested that presenting people with facts that contradict their beliefs doesn't just fail — it actively strengthens those beliefs. The wrong idea digs in deeper. It was a tidy, alarming story, and it spread like wildfire through science communication circles, TED talks, and newsrooms.

There's just one problem. When researchers tried to replicate those original findings at scale, the effect mostly evaporated.

The Study That Started It All — And Why It Didn't Hold Up

The original backfire effect research, conducted by Brendan Nyhan and Jason Reifler in the mid-2000s, showed participants political misinformation followed by corrections. In a handful of cases, the corrections seemed to make certain beliefs stronger. The finding was striking enough to generate enormous buzz.

But science is supposed to be self-correcting, and in this case it was. Nyhan himself was part of later, much larger studies that tried to reproduce the effect across a wide range of political topics. The result? The dramatic backfire effect was largely absent. Most people, when shown accurate corrections, actually updated their stated beliefs — at least somewhat.

So the scary headline version — "facts make people more wrong" — doesn't really describe how most human minds work most of the time.

Good news, right? Not so fast.

What's Actually Happening Instead

Just because the backfire effect is mostly a myth doesn't mean people are eager to change their minds. What researchers are finding instead is a messier, more human picture.

People frequently update their stated beliefs when corrected — they'll acknowledge the fact you've presented — but that update often doesn't travel very far. It doesn't necessarily change related beliefs, downstream attitudes, or future behavior. You might get someone to admit that a specific claim was wrong while they walk away still holding the same general worldview that made that claim appealing in the first place.

That's sometimes called the belief echo — the original misinformation leaves a residue even after the correction lands. Think of it like spilling red wine on a white tablecloth. You can blot it out pretty well, but the faint stain tends to linger.

There's also the issue of identity-protective cognition. When a belief is deeply tied to who someone is — their community, their politics, their sense of self — the emotional stakes of being wrong go way up. Correcting the belief isn't just a factual exchange anymore. It feels like an attack on the person. And people are remarkably good at defending themselves from perceived attacks.

The Motivated Reasoning Machine

Here's where things get genuinely humbling: your brain isn't built to find truth. It's built to survive and belong. Those goals overlap with truth-seeking a lot of the time, but not always.

Motivated reasoning is the process by which people unconsciously work backward from a conclusion they want to reach, assembling whatever evidence supports it and dismissing what doesn't. It's not lying. It's not stupidity. It's just how human cognition operates under certain conditions — especially when emotions, identity, or social belonging are involved.

What this means practically is that the way a correction is delivered matters enormously. Research in science communication suggests that leading with facts and evidence — especially when someone feels defensive — often triggers exactly the kind of identity-protection response that makes people dig in. Not a full backfire, but a kind of emotional shutdown.

Approaches that tend to work better:

• Affirming the person's values first. If someone feels respected and understood, they're more open to new information.
• Asking questions instead of making statements. "What would it take to change your mind about that?" is more powerful than "Here's why you're wrong."
• Separating the belief from the person. "That claim has been looked at pretty carefully and here's what turned up" lands differently than "You believe something false."
• Patience over victory. Minds rarely change in a single conversation. The goal is to plant a seed, not win a debate.

Why This Matters for How We Learn

At SkeptiCamp, we spend a lot of time thinking about what genuine intellectual growth looks like. And one of the most uncomfortable truths is that being a critical thinker doesn't immunize you from any of this. Motivated reasoning doesn't check your IQ or your familiarity with logical fallacies before it kicks in.

The backfire effect myth was, in a way, its own kind of motivated reasoning for science communicators. It gave us a convenient story: "People who reject evidence are psychologically broken in this specific way." It let us off the hook from examining whether we were actually communicating well, meeting people where they are, or respecting the social and emotional dimensions of belief.

The real picture is harder but more useful. People are mostly capable of updating their beliefs. The conditions under which that happens, though, are specific. Trust matters. Tone matters. Relationship matters. The source of the correction matters. Whether someone feels attacked matters.

So What Do You Do With Your Uncle?

Probably not nothing. But also probably not "show up with a stack of studies and a point to win."

If you genuinely want to help someone reason more carefully — whether it's a family member, a coworker, or a student at one of our workshops — the most effective thing you can do is build the kind of relationship where being wrong doesn't feel catastrophic. That's a longer game than a single Thanksgiving argument.

Ask what drew them to the belief in the first place. Listen for the underlying concern — often there's a real worry buried under the bad information. Share your own uncertainty about things. Model what it looks like to say "I'm not sure, I should look into that more."

The goal isn't to be the person who knows more. It's to be the person who makes thinking feel safe.

Because here's the thing the backfire effect research got right, even if it got the mechanism wrong: changing minds is genuinely hard, and blunt factual correction is rarely the best tool for the job. The science just tells us why it's hard — and that turns out to be a lot more interesting, and a lot more actionable, than a simple story about beliefs that dig in when challenged.

Curiosity is contagious. So is intellectual humility. Those might be your best tools after all.