What the data actually shows
One of the most robust findings is confirmation bias: people seek, notice, and remember information that fits their existing beliefs and tend to overlook or discount what contradicts them. Reviews of this literature (such as Raymond Nickerson's) describe it as pervasive and largely automatic — we don't usually decide to ignore disconfirming evidence; the filtering happens before we're aware of it. The result is that beliefs, once formed, tend to gather supporting evidence and resist correction.
Repetition alone can manufacture belief. The illusory truth effect — documented in research going back to Hasher, Goldstein and Toppino (1977) and replicated many times since — is the finding that statements feel more true simply because we've encountered them before. Familiarity gets mistaken for accuracy, and strikingly, the effect can occur even when people actually know better, which is part of why repeated false claims are so sticky.
We also reason in service of conclusions we're motivated to reach. Motivated reasoning research (associated with Ziva Kunda and others) finds that when we want something to be true, we apply easier standards of evidence to it and harder standards to what we'd rather reject — building a one-sided case while feeling like we're being fair. This sits inside a broader picture of the brain as a predictive system that fills gaps with expectations: useful for speed, but a recipe for confident error when the expectations are wrong.
Why this feels different from how it actually is
False beliefs don't feel false from the inside — that's the core of the problem. A belief produced by familiarity or motivated reasoning arrives with the same sense of obviousness and conviction as a well-evidenced one, because the brain doesn't tag where a belief came from. So 'it feels true' is not reliable evidence that it is, even though it's exactly what we tend to trust.
The shortcuts are also invisible while they operate. You don't experience yourself filtering out inconvenient facts or lowering your evidence bar for a comforting claim; you just experience arriving at a conclusion. Because the bias happens beneath awareness, the conclusion feels like the product of clear thinking rather than of a process tilted before you started.
And efficiency usually rewards the shortcut. For most of daily life, going with the familiar and the expected works fine and saves enormous effort, so the habit is constantly reinforced. The cases where it produces a serious false belief are a minority of a generally useful process — which is precisely why we don't notice the machinery and tend to over-trust it.
False beliefs don't feel false from the inside — the brain doesn't tag where a belief came from, so 'it feels true' isn't evidence that it is.
What the research says to do about it
The research consistently favours actively seeking disconfirming evidence over gathering more support. Because confirmation bias makes us collect what already fits, deliberately asking 'what would change my mind?' and 'what's the strongest case against this?' is one of the few moves that counteracts the default — though it takes effort and rarely feels natural.
Slowing down on important judgments helps, because many biases run on the fast, automatic system and recede when you engage slower, more deliberate thinking. Treating 'this feels obviously true' as a prompt to check rather than a conclusion — especially for claims you've heard many times or that you want to be true — guards against both the illusory truth effect and motivated reasoning.
Outsourcing some checking to other people and to good sources is realistic where self-correction is hard. Since it's difficult to catch your own biases from the inside, exposure to people who think differently and to reliable evidence does part of the work for you. Forewarning — simply knowing in advance about a specific bias — also has modest support as a way to reduce it.
What the research says does not help
Believing that bias is something only other people fall for is itself a documented trap (the 'bias blind spot'): people readily spot distortions in others while feeling clear-eyed about their own reasoning. Assuming you're the exception tends to make you more susceptible, not less.
Simply being told a claim is false often doesn't undo it, and sometimes backfires. Corrections can fail to dislodge a familiar belief, and merely repeating a myth in order to debunk it can increase its familiarity and therefore its felt truth. Loudly restating the false claim is one of the least effective ways to fight it.
More intelligence or more information is not a reliable fix on its own. Motivated reasoning research finds that greater cognitive ability can be turned toward defending existing beliefs more effectively, and more data gives confirmation bias more material to cherry-pick. Without the deliberate habit of seeking disconfirmation, raw brainpower and extra facts don't guarantee more accurate beliefs.
Being clever can make some biases worse, because a sharper mind is better at constructing reasons for what it already wants to think.
What this looks like in real life
The claim that feels true because it's familiar
A statement you've encountered many times arrives with a sense of obviousness that has nothing to do with evidence — familiarity gets mistaken for accuracy. Strikingly, the illusory truth effect can occur even when you actually know better, which is part of why repeated false claims are so sticky, and why repeating a myth in order to debunk it can quietly increase its felt truth.
Building a one-sided case while feeling fair
When you want something to be true, you apply an easier standard of evidence to it and a harder one to what you'd rather reject — gathering support, discounting the counterpoints, and experiencing the whole process as clear-eyed reasoning. A sharper mind can make this worse, because it's better at constructing reasons for what it already wants to think.
Real numbers in context
This is a topic where the honest evidence is about robust, repeatedly replicated effects rather than a single headline statistic, so it's worth saying plainly: there's no meaningful percentage for 'how often people believe false things.' What the research strongly supports is that confirmation bias, the illusory truth effect, and motivated reasoning are pervasive, normal, and largely automatic features of human cognition — found across education levels and intelligence.
The practical, non-numeric takeaway is that the feeling of certainty is not a reliable gauge of accuracy. Because the brain optimises for efficient prediction rather than truth, a belief can feel completely obvious and still be wrong — and that's the expected behaviour of a normal mind, not a sign of a defective one. The fact that this applies to everyone, including the most confident and the most informed, is itself the central finding.