--- title: "Confirmation Bias" description: "Confirmation bias is the tendency to seek, notice, and remember information that supports what you already believe, while scrutinizing or overlooking evidence against it. It operates without any intent to deceive yourself, which is what makes it hard to catch: the search feels thorough because th..." last_updated: "2026-08-02" --- # Confirmation Bias Confirmation bias is the tendency to seek, notice, and remember information that supports what you already believe, while scrutinizing or overlooking evidence against it. It operates without any intent to deceive yourself, which is what makes it hard to catch: the search feels thorough because the results feel convincing. The tendency to seek, interpret, and remember information that confirms what you already believe while discounting evidence that contradicts it. *Example: Convinced a diet works, you remember the days you felt great and forget the days you felt awful.* ## What it looks like **In a hiring decision** > I liked her in the first five minutes, and everything after that confirmed it. *Why it fails:* An early impression steered which answers got weight. Strong responses were counted as evidence and weak ones excused as nerves, so the interview tested the impression rather than the candidate. **In a debugging session** > I'm sure it's the database, so I've been reading the query logs all afternoon. *Why it fails:* The hypothesis determined where to look, so hours of evidence accumulate about one subsystem while the actual fault sits somewhere never examined. **In following the news** > Every article I read this week confirmed what I already thought about this policy. *Why it fails:* The sources were chosen before the reading began. A feed assembled around existing views produces agreement that reflects the selection, not the state of the evidence. ## Practice ### Question 1 You believe a new process improved team velocity. Which approach best guards against confirmation bias? A. Collect as many examples as you can find of the process working well B. Decide in advance what result would show the process did NOT help, then check for that specific evidence C. Ask team members whether they think the process helped D. Compare this quarter's output to the same quarter last year **Answer: B.** Specifying a disconfirming result in advance is the core defence, because it removes your freedom to reinterpret whatever you find as support. Option A is confirmation bias as a method. Option C invites the same bias in others and adds social pressure. Option D is a reasonable measurement but says nothing about how you will interpret it, which is where the bias operates. ### Question 2 In Wason's selection task, people are shown cards reading A, K, 4, 7 and asked which to turn over to test the rule "if a card has a vowel on one side, it has an even number on the other." Most people choose A and 4. Why is choosing 4 a mistake? A. Because 4 is even, and even numbers are irrelevant to the rule B. Because turning 4 can only confirm the rule — whatever is on its back, the rule survives — while turning 7 could disprove it C. Because the rule mentions vowels, so only letters matter D. Because 4 and 7 are both numbers, so only one needs checking **Answer: B.** The correct choices are A and 7. Turning 4 cannot break the rule: the rule says vowels have even backs, not that even numbers have vowel backs, so any letter behind 4 is permitted. Turning 7 can break it — a vowel there falsifies the rule outright. The pull toward 4 is confirmation-seeking: people look where the rule could be confirmed rather than where it could fail. ### Question 3 Which is the clearest sign confirmation bias is operating in your own reasoning? A. You feel strongly about the conclusion B. You apply noticeably harder scrutiny to evidence that challenges your view than to evidence that supports it C. You have held the belief for a long time D. Other people disagree with you **Answer: B.** Asymmetric scrutiny is the diagnostic signature — asking "is this study well designed?" of contrary findings while accepting supportive ones at face value. Strength of feeling is not a reliable indicator: you can hold a well-evidenced view strongly. The asymmetry is checkable in a way that conviction is not. ### Question 4 Why does confirmation bias make online research feel more thorough than it is? A. Because search engines are deliberately biased B. Because query wording encodes the expected answer, and a search returns matching results, so extensive reading can still sample only one side C. Because online sources are less reliable than books D. Because people read too quickly online **Answer: B.** Searching "why is X harmful" and searching "is X harmful" return substantially different reading lists, and the first still produces pages of citations. The volume of confirming material creates a genuine feeling of diligence. Deliberately searching for the strongest case against your view is the practical counter. ### Question 5 What is the difference between confirmation bias and motivated reasoning? A. They are identical B. Confirmation bias favours what you already believe; motivated reasoning favours what you want to be true — the two often coincide but the driver differs C. Motivated reasoning is deliberate; confirmation bias is not D. Confirmation bias applies only to scientific questions **Answer: B.** The distinction is between prior belief and preferred outcome. You can be confirmation-biased toward something you find unwelcome but expect, and motivated toward a conclusion you have no prior belief about but badly want. They frequently overlap — most people want their existing beliefs to be right — but the mechanisms are separable. ## Frequently asked ### Can you eliminate confirmation bias by being aware of it? Awareness alone helps surprisingly little — knowing about the bias does not stop it operating. What helps is structural: writing down in advance what would change your mind, seeking out the strongest opposing case rather than the weakest, and inviting critique from people who disagree. ### Is confirmation bias the same as being closed-minded? No. Closed-mindedness is a refusal to consider alternatives; confirmation bias operates in people who are genuinely trying to be fair. That is what makes it significant — it distorts the reasoning of people actively attempting to weigh evidence, not just those who have stopped trying. ### How does confirmation bias relate to the backfire effect? The backfire effect — where contrary evidence supposedly strengthens a belief — was proposed as an extension of this pattern, but later replication work has found it far less common and less general than early reporting suggested. Confirmation bias itself is well established; the backfire effect should be treated as contested. ## Related - [Motivated Reasoning](https://www.criticalthinkingexercise.org/pages/glossary/motivated-reasoning.md) - [Falsifiability](https://www.criticalthinkingexercise.org/pages/glossary/falsifiability.md) - [Steelmanning](https://www.criticalthinkingexercise.org/pages/glossary/steelmanning.md) - [Common Cognitive Biases](https://www.criticalthinkingexercise.org/pages/exercises/common-cognitive-biases.md) - [Overcoming Your Biases](https://www.criticalthinkingexercise.org/pages/exercises/overcoming-your-biases.md) - [Lesson: traps-and-pitfalls/cognitive-biases-decision-social](https://www.criticalthinkingexercise.org/pages/learn/traps-and-pitfalls/cognitive-biases-decision-social.md) ## Sources - [Confirmation Bias: A Ubiquitous Phenomenon in Many Guises (Nickerson, 1998)](https://doi.org/10.1037/1089-2680.2.2.175) — Review of General Psychology - [Reasoning about a Rule (Wason, 1968)](https://doi.org/10.1080/14640746808400161) — Quarterly Journal of Experimental Psychology ## Sitemap See the full [sitemap](https://www.criticalthinkingexercise.org/sitemap.md) for all pages.