Cognitive biases in decision making

A cognitive bias is a systematic pattern in how people notice, interpret, remember or act on information that can move judgement away from an appropriate standard. A hiring manager may, for example, favour a candidate who attended the same school because familiarity feels relevant even when it predicts little about performance. Biases arise from useful mental shortcuts as well as social and motivational pressures, so understanding them improves decisions without assuming that human intuition is always defective.

When to use it

• Use bias awareness to examine how you reach consequential decisions and reduce preventable error.

• Use it to understand why reasonable people interpret the same evidence differently.

• Use it to design meetings, evidence reviews and choice environments that produce more independent judgement.

Origins

Research on judgement has deep roots, but psychologists Amos Tversky and Daniel Kahneman established the modern heuristics-and-biases programme through experiments begun in the late nineteen sixties. Their landmark Science article “Judgment under Uncertainty: Heuristics and Biases” showed how efficient rules of thumb can produce predictable errors when people estimate likelihood and interpret evidence.

Their later prospect theory challenged important assumptions of expected-utility models. People evaluate outcomes relative to a reference point, generally dislike a loss more than an equivalent gain and often become risk-averse over gains but risk-seeking when trying to avoid a sure loss. The work helped connect experimental psychology with economics and influenced medicine, politics and management. Kahneman received the 2002 Nobel Memorial Prize in Economic Sciences for integrating psychological research into economic decision-making; Tversky had died earlier and Nobel prizes are not awarded posthumously. The research programme has since generated more than 1,000 studies and applications across decision domains.

What it is

The term covers many mechanisms rather than one defect. Some influence group decisions, such as groupthink, in which pressure for agreement suppresses challenge. Others affect probability judgement: the representativeness heuristic can make a case seem likely because it resembles a stereotype while base rates are neglected. Illusory correlation creates a perceived relationship between events that are not reliably connected.

Memory is reconstructive, so consistency bias can make earlier beliefs appear closer to current ones than they really were. Motivation also matters: egocentric and self-serving biases protect a positive self-image by assigning disproportionate credit to oneself or interpreting evidence favourably. These patterns are often adaptive in ordinary conditions, but become dangerous when the shortcut does not fit the decision.

How to use it

Awareness is a starting point, not a complete remedy, because people can name biases and still remain subject to them. For an important product-launch decision, structure the process before the group knows which option senior leaders prefer. Ask:

• Have sponsors searched for disconfirming evidence, or has confirmation bias shaped the market estimate? Would the recommendation change if gains and losses were framed differently?

• Did participants form views independently before discussion? Were relevant base rates and alternative explanations considered? Could minority voices raise concerns safely, or did status and early consensus silence them?

Build safeguards around the risks identified. Ask an independent analyst to reproduce disputed evidence, require a pre-mortem, use a devil’s advocate, compare the proposal with a reference class and record assumptions and decision thresholds in advance. The chair should invite lower-status or dissenting participants before senior voices dominate and distinguish a lack of objection from informed agreement.

Apply the same discipline to performance reviews, recruitment, forecasts and customer conversations. Use explicit criteria, multiple sources and contemporaneous records to reduce reliance on vivid recent events or affinity. Focus on the few biases most likely in the decision rather than diagnosing every possible label after the fact, and examine your own incentives before attributing error to others.

Top practical tip

Before discussion, ask every participant to write an independent judgement and the evidence behind it. Reveal those views before open debate so the first speaker does not anchor the entire room.

Top pitfall

Do not turn bias language into amateur diagnosis or analysis paralysis. Match safeguards to the stakes and reversibility of the choice, and remember the bias blind spot: invite others to challenge your reasoning because recognising error in yourself is harder than finding it in them.

Further reading

• Tversky, A. and Kahneman, D. (nineteen seventy-four). “Judgment under Uncertainty: Heuristics and Biases.” Science.
• Kahneman, D. (twenty eleven). Thinking, Fast and Slow. Farrar, Straus and Giroux.