Your critical thinking gym
Home
The Commandments
Join the Club
Your critical thinking gym
● Cognitive bias
← From Latin confirmare, "to strengthen." Named by psychologist Peter Wason, 1960s.
The habit of noticing, seeking out, and remembering whatever agrees with what you already believe, while barely registering what doesn't.
Also called: myside bias · confirmatory bias. Not the same as cherry picking, a propaganda technique that deliberately hands you one-sided evidence (Section 01).
The tendency to search for, interpret, favor, recall, and give more weight to information that confirms a belief you already hold, while giving disproportionately less attention and scrutiny to information that contradicts it.
Plain version
Your brain likes finding proof it's already right, and kind of tunes out proof it's wrong. Not because you're lying to yourself on purpose. It usually doesn't feel like anything at all. It just feels like paying attention to "the good stuff."
Aliases
Myside bias and confirmatory bias are used interchangeably with confirmation bias in most research.
Not the same as cherry picking
This is the pairing worth getting right: confirmation bias is what happens inside your own head, often without you noticing, you just happen to search for, notice, and remember the confirming stuff more. Cherry picking (a propaganda technique) is what happens when someone else deliberately hands you only the confirming evidence, on purpose, to steer your conclusion. One is an internal habit. The other is an external weapon aimed at that habit. See Section 03 for how the two connect.
Scope
Gets confused with
Cherry picking (the propaganda technique, not the bias): precise distinction above and in Section 09.
One of the hardest biases on the whole site to catch in yourself. Searching for evidence that agrees with you doesn't feel like bias, it feels like "doing your research." There's no internal alarm, a totally reasonable cover story is always available ("I just looked into it"), and being right feels good. Nothing here trips a wire.
Why your brain does this
Constantly re-testing every belief you hold from scratch, symmetrically, every time new information shows up, would be exhausting and slow. Most of the time, your existing beliefs are actually right, so a quick check for "does this fit what I already know?" is a cheap, usually-reasonable shortcut, not a design flaw. This isn't a broken brain. It's a fast one.
Where it misfires
The shortcut breaks down exactly when it matters most: when a belief is wrong, strongly held, or tied to your identity or group. In those cases the "quick check" stops being a genuine check at all. It turns into a one-sided search that can never actually find out you're wrong, because it was never really looking.
The live academic debate, briefly
The traditional "heuristics and biases" program, built by psychologists Daniel Kahneman and Amos Tversky, treats patterns like this one as errors against a rational standard. German psychologist Gerd Gigerenzer's ecological-rationality research offers a different frame: shortcuts like this one are often well-suited to real environments where you don't have unlimited time or information, even when they'd fail a textbook logic test, making confirmation-seeking a reasonable strategy misapplied outside the conditions it works well in, not a bug. A separate, French-tradition angle from cognitive scientists Hugo Mercier and Dan Sperber's argumentative theory of reasoning suggests human reasoning didn't evolve mainly to find truth when you're alone. It evolved partly to win arguments and justify decisions to other people, so gathering "ammunition" that supports your side is, on this view, working exactly as designed for a social, argumentative animal. This page doesn't take a side between these views; all point to the same practical lesson: this isn't stupidity, it's a shortcut built for a different job than "get to the objective truth alone."
How it activates, step by step
1. You hold an existing belief or preference.
2. New information shows up: a claim, an article, a data point.
3. Without deciding to, you search harder for reasons the new information fits what you believe, and generate reasons to dismiss or discount anything that doesn't.
4. Your belief stays the same or gets stronger, even if the new evidence, weighed fairly, should have moved it.
Confirmation bias isn't just something that happens to you passively. It's a specific target that propaganda techniques are built to hit.
Feeds you only the evidence that already agrees with a conclusion. Your own confirmation bias does the rest of the work. You don't need to be tricked into believing a one-sided case, because a one-sided case is exactly what your search habits were already primed to accept.
Creates the appearance that "everyone" already agrees with a claim. Confirmation bias treats that appearance as new supporting evidence, even when the "everyone" is bots or coordinated accounts.
Agenda setting & algorithmic bias
Control what you're exposed to in the first place. If you never encounter the disconfirming information, your confirmation bias never even gets the chance to ignore it. The filtering happens one level upstream.
Ad nauseam (repetition)
Repeating a claim across many channels can look, to a confirmation-biased brain, like independent confirmation from multiple sources, even when it's the same claim recycled through one source many times.
Outside politics
Sales relies on this constantly and non-maliciously in its mildest form: a good salesperson lets you "discover" the features you already wanted, so the decision feels self-confirmed rather than sold to you. In its predatory form, investment scams use fabricated testimonials specifically to give victims a feeling of external confirmation for a decision they were already inclined to make.
What increases it
Strong: belief strength and identity stakes. The more a belief is tied to who you are or what group you belong to, the stronger the effect.
Strong: social reinforcement. Being surrounded by people who already agree with you amplifies it, since their agreement itself becomes "confirming evidence."
Moderate: time pressure and cognitive load. Less mental bandwidth means less capacity to slow down and genuinely check the other side.
What does NOT predict it
Research by psychologists Keith Stanovich and Richard West found that the size of a person's confirmation bias ("myside bias" in their terminology) is essentially unrelated to intelligence or cognitive ability. Smart people aren't meaningfully better at avoiding it. They're often just better at building a more sophisticated-sounding justification for the conclusion they already wanted.
State-dependent factors
Time pressure and cognitive load (see above) aren't just standing traits, they're state-dependent: the same person is measurably more prone to one-sided searching when rushed, tired, or multitasking than when given time to slow down and deliberately look for disconfirming evidence.
The pattern across all three: confirmation bias doesn't require bad faith or low intelligence. It shows up in careful academic subjects, in politically engaged adults, and in ordinary shopping decisions alike, because the mechanism is the same everywhere the stakes involve an existing belief.
First named
Psychologist Peter Wason coined "confirmation bias" following his 1960 study, "On the failure to eliminate hypotheses in a conceptual task" (Quarterly Journal of Experimental Psychology).
The landmark study
Wason gave participants a sequence of three numbers, "2, 4, 6," and told them it fit a secret rule. Participants proposed their own triples and were told whether each fit. Most people formed an overly specific hypothesis (like "even numbers increasing by two") and only ever tested triples that fit their own guess, never triples designed to prove themselves wrong. The actual rule was simply "any ascending sequence," far broader than almost anyone guessed, because almost no one tried to falsify their own idea.
Since then
Psychologist Robert Nickerson's widely cited 1998 review, "Confirmation Bias: A Ubiquitous Phenomenon in Many Guises" (Review of General Psychology), consolidated decades of follow-up research showing the effect appears across memory, evidence-search, and interpretation, not just in Wason's original lab task.
Cross-tradition note
See Section 02 for how Gerd Gigerenzer's German ecological-rationality tradition and Hugo Mercier and Dan Sperber's French argumentative theory of reasoning each offer a distinct framing of why this bias exists at all, genuinely additive here, not just flavor.
Score: 1 out of 5, deliberately one of the lowest on the whole site (see the breakdown in the card after Section 01).
The in-the-moment question
Before you settle on a conclusion: "Have I actually gone looking for a good reason I might be wrong, not just braced myself against one, but actually looked?" If the honest answer is no, that's the signal, not any feeling of doubt (there usually isn't one).
Why this is so hard to catch
Every other bias on this site has at least one thing going for the person trying to catch it: a felt discomfort, a visible trigger, something. Confirmation bias has almost none of that. Searching for confirming evidence feels identical, from the inside, to "doing your research properly." That's the entire reason it scores a 1.
Debiasing research note
The research showing these interventions work is real but modest. "Consider the opposite" reliably reduces the effect size in lab studies; it doesn't eliminate it. Be wary of any source that claims a simple trick makes confirmation bias go away.
Commonly co-occurs with
Motivated reasoning and cognitive dissonance also compound closely with this one but don't yet have their own Filter pages to link to.
Often mistaken for
The broader pattern, not the same thing. Motivated reasoning is reasoning in a direction that serves what you want to be true, across memory, evidence-search, and even effort spent (working harder to find flaws in unwelcome conclusions). Confirmation bias is the specific, well-studied evidence-search piece of that larger pattern, not a separate thing.
Compounding effect
Confirmation bias typically comes first and feeds cognitive dissonance next: once you've built a belief through one-sided search, encountering strong contrary evidence produces the uncomfortable tension dissonance describes, and dissonance-reduction often resolves by doubling down, which is confirmation bias again, now with more emotional investment behind it.
Primary sources
Accessible reading