6 Cognitive Biases and Judgment

Mental shortcuts and judgment

People make judgments with limited time, attention, and information. They rely on heuristics, or quick rules of thumb, which often work well but can sometimes produce cognitive biases: systematic tendencies that pull judgment away from a careful assessment of evidence.

These tendencies are not proof that people are irrational. They are predictable risks in ordinary thinking, especially when a situation is uncertain or emotionally charged.

Biases in seeking and judging information

Confirmation bias is the tendency to seek, notice, or interpret information in ways that support an existing belief while overlooking or discounting information that challenges it. Someone who believes a new study method works, for example, may remember successful tests but fail to compare them with tests where the method did not help.

The availability heuristic leads people to judge how common or likely something is by how easily examples come to mind. A vivid news story about a rare danger may make that danger seem more frequent than it is; ease of recall does not establish actual frequency.

Anchoring occurs when an initial number or estimate influences later judgments, even if the starting point is arbitrary or only loosely relevant. After seeing a very high suggested price, a shopper may regard a merely high price as reasonable.

Biases in probability and interpretation

Representativeness and base-rate neglect describe judging likelihood by how closely a case resembles a familiar pattern while giving too little weight to background frequencies. If a person fits a stereotype of a profession, that resemblance alone does not show that the person is more likely to belong to that profession than to a much larger group.

The framing effect occurs when different wording or presentation changes how the same choice feels. A treatment described as having a “90% survival rate” may sound more appealing than one described as having a “10% death rate,” even though both figures express the same outcome.

Overconfidence is being more certain about a judgment than the available evidence warrants. Confidence is a feeling about a conclusion, not independent proof that the conclusion is correct.

Hindsight bias arises after an outcome is known: people may then see it as having been predictable all along. This can lead them to overestimate how much they knew before the result was revealed.

How biases shape belief

Biases can affect several stages of reasoning: which questions people ask, what evidence they notice, how they interpret it, what they remember, and how strongly they hold a conclusion. Motivation and identity can matter as well. Information that challenges an important belief may be scrutinized more harshly than information that supports it.

These effects can happen without deliberate dishonesty or awareness. Biases can also interact: a person may start with an anchor, search for confirming evidence, and then become overconfident in the resulting judgment.

Recognizing a bias is useful, but naming one does not by itself prove that a particular conclusion is false. The claim still needs to be assessed by examining its evidence and reasoning. Ask what evidence would change your mind and whether you have considered it fairly.

Practices for more careful judgment

1. State the claim clearly. Separate what is known from what is inferred or assumed.

2. Look for disconfirming evidence. Ask what observation would count against your favored explanation, and actively seek it.

3. Check alternatives and base rates. Compare an appealing explanation with other plausible explanations and relevant background frequencies.

4. Delay exposure to anchors when possible. Make an independent estimate before seeing a suggested price, forecast, or opinion.

5. Reframe choices. Express outcomes in more than one equivalent way, such as with both survival and mortality rates.

6. Calibrate confidence. Record estimates and confidence before outcomes are known, then review how accurate they were.

7. Use independent perspectives. Ask people to assess the evidence before discussing a group’s preferred conclusion.

These practices do not eliminate bias. They make judgments more testable by directing attention to evidence, alternatives, and uncertainty rather than relying only on an initial impression.