Confirmation Bias
The tendency to search for, interpret, and recall information in ways that confirm what we already believe — and to discount evidence that contradicts those beliefs.
Confirmation bias is one of the most thoroughly documented patterns in cognitive psychology. Once we hold a hypothesis — even a casually formed one — we unconsciously give more weight to evidence that supports it, ask questions whose answers will confirm it, and remember the wins while glossing over the misses. Peter Wason's classic 2-4-6 task showed people would invent elaborate rules to fit a hidden pattern but rarely tested cases that could disprove their guess. The bias shows up everywhere: in scientists clinging to a favored theory, in juries weighing testimony, in social-media feeds that quietly thicken into echo chambers. Nickerson's 1998 review in Review of General Psychology catalogued dozens of forms it takes — biased search, biased interpretation, biased recall — and noted that simply being warned about the bias is rarely enough to neutralize it. The mitigation that does work: deliberately constructing the case against your own position before you act on it.
What is confirmation bias?
Confirmation bias is the tendency to seek, interpret, recall and weight information in ways that confirm existing beliefs while underweighting evidence that would falsify them. The founding empirical demonstration is Peter Wason's 1960 2–4–6 task: participants were given the triple "2, 4, 6" and asked to discover the underlying rule by proposing additional triples that the experimenter would label as fitting or not. The actual rule was simply "any ascending sequence," but most participants generated narrow hypotheses such as "evens ascending by two" and proposed only confirming triples — never trying a sequence that would rule out their initial guess. Raymond Nickerson's 1998 Review of General Psychology survey assembled four decades of evidence into a single construct, distinguishing motivated from cognitive variants and arguing that confirmation bias is less a single bug than a family of related search and interpretation tendencies operating across working memory, attention and long-term retrieval.
Why it matters
Confirmation bias is the failure mode that turns intelligence into an amplifier rather than a check on belief. Charles Lord, Lee Ross and Mark Lepper's 1979 polarization study showed that participants exposed to mixed evidence on capital punishment ended up more confident in their original positions, not less — the same data, read selectively, hardened opposing camps. In medicine, Pat Croskerry's diagnostic-error work places confirmation bias at the top of cognitive-error taxonomies; in finance, behavioral-investing literature documents how investors over-weight news that supports an existing position; in science, Brian Nosek's reproducibility work surfaced the bias's role in selectively reporting confirming results. Brendan Nyhan and Jason Reifler's 2010 Political Behavior work originally suggested a "backfire effect" in which corrections strengthen false beliefs, but several pre-registered replications (Wood and Porter 2019; Nyhan 2021) have substantially narrowed the effect — a useful reminder that even bias research is itself subject to bias.
How Fokiq tests it
The Fokiq Daily embeds confirmation-bias probes inside the logic slice: decision-making items where the obvious confirming answer fails a single falsifying test, rule-induction tasks modeled on Wason's 2–4–6 paradigm, and forced-disconfirmation tasks that reward generating contradictory evidence. Difficulty scales with the cognitive load you handled correctly in earlier rounds, so what arrives tomorrow depends on what you cleared today. Track the logic bar in your evolution chart, or jump to the standalone logic-puzzle test for an isolated read. Bible Q19 walks the Wason 2–4–6 result and Q53 covers the broader prospect-theory frame that often coexists with confirmation, and the logical-deduction hub describes the practice patterns most aligned with falsification-style reasoning.
Common misconceptions
The first misconception is that confirmation bias is the same as motivated reasoning. The two overlap but dissociate: confirmation bias operates even on neutral, low-stakes hypotheses, while motivated reasoning kicks in when the conclusion has affective or identity stakes. The second is that smarter people are immune. Dan Kahan's cultural-cognition work shows the opposite — higher numeracy and analytic skill amplify confirmation bias on identity-protected questions, because more capable reasoners can construct more compelling justifications for the conclusion they already prefer. The third is that the "backfire effect" is settled science. Nyhan and Reifler's 2010 finding has been substantially narrowed by replications (Wood and Porter 2019), and the appropriate stance is to cite both the original and the corrective literature. The fourth is that disconfirmation strategies are easy once you know about them — Wason's task replicates with educated adults today. Awareness of the bias does not eliminate it; structural prompts (asking what would change my mind? before reading) outperform mere knowledge of the bias's existence.
Where to learn more
Pair confirmation bias with cognitive bias for the umbrella construct, with anchoring and availability heuristic for sister biases that frequently co-occur, with cognitive flexibility for the executive layer that supports rule revision, and with metacognition for the monitoring layer that catches confirmation in the act. Brain-types The Analyst and The Strategist profile the falsification-leaning ability mix, and the logical-deduction hub walks through the practice patterns most aligned with disconfirmation-style reasoning. Curated reading lives in the research corner, and the founder note explains why Fokiq treats falsification as a first-class skill alongside raw recall and speed.
Sources
- (1960). On the failure to eliminate hypotheses in a conceptual task. Quarterly Journal of Experimental Psychology, 12(3), 129–140.
- (1979). Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence. Journal of Personality and Social Psychology, 37(11), 2098–2109.
- (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2(2), 175–220.
- (2019). The elusive backfire effect: Mass attitudes’ steadfast factual adherence. Political Behavior, 41(1), 135–163.
Frequently Asked Questions
What is a simple example of confirmation bias?
You believe a coworker is always late. You start noticing every time they arrive after you, and forget the days they got in early. Within a few weeks the belief feels confirmed — but you were only counting evidence that fit it. Tracking the data without filtering is the corrective.
How do you reduce confirmation bias in your own thinking?
The most effective technique researchers have found is the consider-the-opposite prompt: before locking in a judgment, write down what evidence would change your mind, and actively look for it. Pre-committing to a falsification test — the same logic scientists use — is harder to skip than a vague intention to "be objective".
Where was confirmation bias first systematically studied?
Peter Wason's 1960 rule-discovery experiments are the classical reference. Raymond Nickerson's 1998 review article in Review of General Psychology is the standard modern survey, cataloguing the bias across reasoning, memory, social judgment, and scientific practice.