Cognitive bias
A systematic, predictable error in judgment, often caused by the mental shortcuts (heuristics) people use to simplify decisions, rather than a random mistake.

On this page· 5 sections
Everyone misjudges things. A misjudgment becomes a cognitive bias when it is not random: it points the same way, in the same kind of situation, for most people[3].
That is what makes biases worth naming. A trader cannot plan around a one-off slip, but a lean that repeats can be spotted, described and checked against.
How it works
Amos Tversky and Daniel Kahneman showed in 1974 that people judge probabilities and values with a small number of mental shortcuts. The shortcuts turn hard questions into simpler ones and are quite useful in general, but they sometimes produce severe and systematic errors[4]. Looking back in 1983, the authors described the shortcuts as highly economical and usually effective, with predictable biases as the price[1].
A Financial Conduct Authority paper explains the mechanism with Kahneman’s two modes of thought. Intuition is fast, automatic and effortless; reasoning is slow and deliberate[3]. Most everyday choices run on intuition, and biases arise when intuition leads people astray in systematic and predictable ways[3].
One way it goes wrong is substitution. Intuition swaps a hard question for an easier one without noticing. In the FCA’s example, the question of whether a financial adviser is good turns into whether the adviser comes across as a nice person[3].
CFA Institute’s curriculum sorts biases into two groups[2]:
- Cognitive errors come from faulty reasoning: statistical, information-processing or memory errors. Examples include confirmation, illusion of control, hindsight, anchoring, mental accounting, framing and availability.
- Emotional biases come from impulse or intuition, reasoning shaped by feelings. Examples include loss aversion, overconfidence, self-control and regret aversion.
Example
Say six traders each estimate the fair price of an asset worth $50.
- Group A’s errors are random: +$3, −$3, +$1, −$1, +$2, −$2. The average error is $0, so the average estimate is $50.
- Group B shares one shortcut that pulls every estimate up: +$3, +$4, +$2, +$5, +$3, +$1. The average error is +$3, so the average estimate is $53.
Asking more people fixes Group A, because random errors cancel. It does not fix Group B, because everyone leans the same way. That is the difference between noise and a bias.
Watch out
Knowing the list does not protect you. The FCA notes that people are mostly blind to their own biases and that even people familiar with them find it hard to see how a bias affects a particular decision[3].
The shortcut is not the enemy. The same heuristics that produce biases are usually effective[1], which is why people keep using them.
Labels overlap. The CFA curriculum notes that one bias can have aspects of both types, with one dominating, and that cognitive errors are easier to correct than emotional biases[2].
Sources
- ResearchUniversity of Pennsylvania (Eugene Garfield library)This Week's Citation Classic: Judgment under uncertainty: heuristics and biases (Tversky & Kahneman, Current Contents, 4 April 1983, on their Science 185:1124-31, 1974)Checked November 9, 2026
- Professional bodyCFA InstituteThe Behavioral Biases of Individuals (2026 Level I refresher reading)Checked November 9, 2026
- RegulatorFinancial Conduct AuthorityApplying behavioural economics at the Financial Conduct Authority (Occasional Paper No. 1, April 2013)Checked November 9, 2026
- ResearchUniversity of Washington (course copy of the journal article)Judgment under Uncertainty: Heuristics and Biases (Tversky & Kahneman, Science 185(4157), 1974, pp. 1124-1131)Checked November 9, 2026
