Cognitive bias

A cognitive bias is a systematic pattern in judgment, memory, attention, or choice relative to a stated normative or statistical benchmark. The benchmark matters: expected value, Bayesian inference, factual accuracy, consistency, and an experimental criterion are different standards.

Some biases arise when a heuristic is mismatched to a task. Others involve motivation, affect, memory reconstruction, social incentives, or how information is framed. Treating all biases as “mental shortcuts” hides those differences.

Research normally estimates group-level patterns under defined conditions. It does not show that every participant displays the effect, that a person is globally irrational, or that the magnitude transfers unchanged across populations and cultures.

Small directional errors can matter when the same judgment is repeated in medicine, hiring, forecasting, finance, policy, or everyday choice. The consequences depend on the stakes, frequency, reversibility, and whether errors accumulate or cancel.

Ability and deliberation can help on some tasks, but there is no simple rule that intelligence either eliminates bias or is unrelated to it. Performance depends on knowledge, motivation, numeracy, cognitive reflection, task representation, and the specific bias being measured.

The practical question is not whether someone “has biases.” It is which predictable failure modes are plausible in this decision, what benchmark matters, and whether a process change improves the outcome without creating a larger problem.

Heuristic-task mismatch. A fast rule can be useful in one environment and systematically depart from a normative answer in another.

Attention and memory. What is noticed, encoded, retrieved, or reconstructed changes which evidence enters a judgment.

Motivation and identity. Preferences, goals, group commitments, and self-protection can shape how ambiguous evidence is sought or interpreted.

Framing and reference points. Equivalent information can produce different responses when the representation, default, sequence, or comparison point changes.

Social and institutional context. Incentives, hierarchy, correlated information, and repeated exposure can create patterns that are not reducible to one person’s mental shortcut.

Popular lists can be organized into overlapping descriptive families, but there is no single authoritative taxonomy. Mitigation evidence is task- and context-specific; the checks below are candidates, not category-level guarantees.

• Information-search patterns. Confirmation bias concerns preferential search for or interpretation of supporting evidence. A possible check is to specify in advance what credible disconfirming evidence would change the decision.
• Anchoring and adjustment. Initial values can influence later estimates. Independent estimates or estimating before seeing a reference value are testable process options, not guaranteed corrections.
• Availability-related judgments. Ease of recall can influence frequency or risk judgments in studied tasks. Relevant base-rate information may provide a useful comparison when it is reliable and applicable.
• Framing and reference points. Equivalent descriptions can produce different responses in some tasks. Restating a problem in more than one equivalent form can reveal sensitivity to the frame.
• Hindsight and overconfidence. Retrospective inevitability and miscalibrated confidence are distinct effects. Recording predictions and comparing confidence with outcomes can make calibration measurable.
• Sunk-cost and commitment patterns. Prior investment can influence forward-looking choices even when it should not affect the relevant benchmark. A possible check is to compare future costs and benefits without counting unrecoverable investment.
• Self-assessment patterns. The bias blind spot describes judging oneself as less biased than others under specified conditions. External feedback or independent review may expose disagreements, but their effects still require evaluation.

Bruine de Bruin, Parker, and Fischhoff (2007) reported that higher adult decision-making competence was associated with fewer self-reported negative decision outcomes. That correlational result does not establish that a brief bias score predicts retirement saving, medical adherence, or every other real-world domain.

Morewedge and colleagues (2015) tested specific interventions: an instructional video and an interactive game. Performance improvements on the biases covered by those interventions were observed immediately and at least two months later. The study does not show that naming a bias is sufficient, that spaced reminders caused the persistence, or that the effects transfer to every bias and decision.

Cognitive ability is not a universal shield, but neither is susceptibility independent of intelligence across all tasks. Associations vary by measure; knowledge, numeracy, reflection, motivation, and task representation also matter. Interpret claims at the level of the specific bias task rather than as a global statement about IQ or rationality.

“Every bias comes from a heuristic.” Heuristics are one route. Motivation, affect, memory, social processes, and task design also contribute.

“A long list means there are hundreds of independent mechanisms.” Lists often contain overlapping labels, boundary cases, and variants of broader effects.

“Smart people are immune.” Cognitive ability helps on some tasks and is weakly related on others. Knowledge, reflection, incentives, and representation all matter.

“Knowing the name fixes the bias.” Awareness may cue a useful check, but transfer is limited. Defined process changes and feedback are more testable than a general instruction to think harder.

“A bias finding diagnoses an individual.” Most findings describe average patterns under study conditions. Individual assessment requires a valid measure and an appropriate interpretation.

Suppose a hiring manager favors a candidate who resembles a former high performer. Availability, representativeness, rapport, or job-relevant evidence could each contribute; the observation alone cannot identify which mechanism produced the preference.

Predefined criteria, structured questions, work samples, and independent scoring can make the basis of comparison more explicit. Whether those safeguards improve validity and fairness must be evaluated in the actual selection process. Awareness of a bias label is neither necessary nor sufficient evidence that the process is protected.

• Heuristic — a simplified strategy that can succeed or fail depending on the environment.
• Bounded rationality — the broader framework of constrained decision procedures.
• Decision hygiene — preventive process safeguards for noise and some bias.
• Decision support system — software that can structure a decision but may also introduce automation bias.
• Confirmation bias, anchoring, and sunk cost — individual effect families with narrower definitions and evidence bases.

The LifeByLogic Cognitive Bias Test presents brief tasks adapted from published judgment paradigms, including work associated with the Heuristics-and-Biases Inventory, Adult Decision-Making Competence, and rational-thinking assessments. Evidence for those source instruments does not automatically validate this implementation or support one stable, overall bias trait.

Reliability, validity, scoring, and interpretation are task-specific. Treat the result as educational self-reflection, not a clinical assessment or a basis for high-stakes decisions, and review the tool’s own methodology and limitations.

Take the Cognitive Bias Test →

Bias claims are benchmark- and task-dependent. The sources include foundational experiments, individual-difference work, intervention evidence, and a recent organizational review.

• Tversky and Kahneman (1974), Judgment under Uncertainty — foundational heuristics-and-biases paper.
• Bruine de Bruin, Parker, and Fischhoff (2007) — development of adult decision-making competence measures and an association with fewer self-reported negative decision outcomes.
• Morewedge and colleagues (2015) — tests of an instructional video and an interactive game, with effects on covered biases measured immediately and at least two months later; not evidence for awareness alone or universal transfer.
• Fasolo, Heard, and Scopelliti (2025) — review of approaches to mitigating cognitive bias in organizational decisions.
• Stanovich, West, and Toplak (2016), The Rationality Quotient — assessment framework for rational thinking and its relation to cognitive ability.

This entry is intended as a citable scholarly reference. Choose the format that matches your context. The retrieval date should reflect when you accessed the page, which may differ from the entry's last-reviewed date shown above.

LifeByLogic. (2026). Cognitive Bias: Definition, Types, and Examples. https://lifebylogic.com/glossary/cognitive-bias/

LifeByLogic. "Cognitive Bias: Definition, Types, and Examples." LifeByLogic, 12 July 2026, https://lifebylogic.com/glossary/cognitive-bias/.

LifeByLogic. 2026. "Cognitive Bias: Definition, Types, and Examples." July 12. https://lifebylogic.com/glossary/cognitive-bias/.

@misc{lblcognitivebias2026,
  author = {{LifeByLogic}},
  title = {Cognitive Bias: Definition, Types, and Examples},
  year = {2026},
  month = {jul},
  publisher = {LifeByLogic},
  url = {https://lifebylogic.com/glossary/cognitive-bias/},
  note = {Accessed: 2026-07-12}
}