Skip to content
Mind Flaws

Judgment & Decision · Quick Test

Conjunction Fallacy

Can two things together be more likely than one alone?

The experiment

Read the description and pick the option you think is more likely.

Linda is 31 years old, single, outspoken and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also took part in anti-nuclear demonstrations. Which is more likely?

What happened?

The second option seems to fit Linda better. But it is a special case of the first: everyone who is a “bank teller and feminist” is also a “bank teller”. So the first option can never be less likely than the second.

Why does your brain do this?

Instead of working out the odds, your mind compares the description with each option and asks: “which one sounds more like Linda?”. The “feminist” detail increases the resemblance and, with it, the feeling of probability. This shortcut is called the representativeness heuristic.

When does this shortcut work well?

Judging by resemblance usually works when the profile really helps predict something, like recognizing a friend by the way they walk or guessing what music someone likes. The trouble starts when the question is about probability and each extra detail should lower the odds, not raise them.

Where does it show up in daily life?

Conspiracy theories and rumors full of details feel more convincing precisely because they are detailed, even though each extra detail lowers the chance that the whole story is true. Very specific forecasts (“it will rain at 3 pm and the game will be postponed”) also sound more reliable than they are.

Deep Dive

Open Nerd Mode

Amos Tversky and Daniel Kahneman published the “Linda problem” in 1983. In the two-option version, 85% of 142 university students chose the conjunction, breaking the rule that P(A and B) is never greater than P(A).

The evidence badge is mixed: the effect shows up consistently in the classic format, but its size and interpretation depend on how the question is asked. Hertwig and Gigerenzer (1999) argued that many people read “probable” as “plausible” or “fits the description”, not in the mathematical sense; when the question is phrased in frequencies (“out of 100 people like Linda, how many…”), the fallacy drops sharply. Later studies debate how much of the effect survives these changes.

Technical terms

  • Probability
  • Representativeness
  • Heuristics

Discovery

Described by Amos Tversky e Daniel Kahneman in 1983.

Level of evidence

Mixed evidence

The effect exists, but its size, conditions or interpretation vary across studies.

Basis: Hertwig, R.; Gigerenzer, G. (1999). The 'conjunction fallacy' revisited: How intelligent inferences look like reasoning errors.

References

  1. Original studyTversky, A.; Kahneman, D. (1983). Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment. Psychological Review, 90(4), 293–315.
  2. ReplicationHertwig, R.; Gigerenzer, G. (1999). The 'conjunction fallacy' revisited: How intelligent inferences look like reasoning errors. Journal of Behavioral Decision Making, 12(4), 275–305.

Review

This text has not been reviewed by a volunteer yet. Want to help review it?

Found a mistake? Write to [email protected]