Of all the cognitive biases documented by psychologists over the past fifty years, confirmation bias is probably the one with the greatest real-world consequences — and the one most people are least willing to acknowledge in themselves. It feels like clear thinking. It is its opposite.
Confirmation bias is the tendency to search for, interpret, favour, and recall information in a way that confirms what you already believe. It operates at every stage of cognition: what evidence you seek out, how you interpret ambiguous evidence, and what you remember about evidence you've already encountered.
How it operates
In information seeking
When people want to test a hypothesis, they typically ask questions likely to confirm it rather than disconfirm it. In a famous series of experiments by Peter Wason in the 1960s, participants were given the sequence 2-4-6 and asked to discover the rule governing it by proposing other sequences. Nearly all proposed sequences that confirmed their initial hypothesis (e.g., even numbers ascending). Very few tested disconfirming sequences. The rule was simply "any ascending sequence" — most participants never found it.
In interpretation
Identical information is interpreted differently depending on prior belief. In one classic study, participants who supported and opposed capital punishment were given the same mixed evidence on its effectiveness. Both groups rated the evidence supporting their prior view as more convincing, and both groups emerged from the study more confident in their original position. The same data made both sides more extreme.
"It is the peculiar and perpetual error of the human understanding to be more moved and excited by affirmatives than negatives."
In memory
People remember information that confirms their beliefs more readily than information that challenges them. This means that even when you do encounter disconfirming evidence, it is less likely to be available to you later when forming a judgment.
Why it's so hard to correct
Confirmation bias is not a failure of intelligence. Studies consistently find it in scientists, doctors, lawyers, and judges — people trained and professionally motivated to think clearly. It persists because it is, in many situations, efficient. Using what you already know to filter new information is fast. The problem arises when the belief being confirmed is wrong, or when the stakes of the decision are high.
The practice: When forming an opinion, deliberately seek out the strongest version of the opposing argument — not to be converted, but to test whether your view survives contact with the best counterevidence. If you can't articulate the steelman case against your position, you probably don't understand the issue well enough to hold a confident view.
The deeper problem
Confirmation bias is especially dangerous in combination with motivated reasoning — the tendency to reason toward a desired conclusion. When you want something to be true, you apply confirmation bias with particular intensity: the bar for evidence supporting the belief gets lower; the bar for evidence against it gets higher. We are all capable of this, and most of us do it most of the time without realising it.