Conjunction fallacy
The conjunction fallacy is a cognitive bias where individuals assume specific conditions are more probable than a single general one. This often occurs when people mistake the conjunction of two events as more likely than one of the events alone, violating the basic rule of probability.
How it works
This bias arises because people often rely on heuristic thinking instead of probabilistic reasoning. When presented with a narrative that seems more representative of real-world knowledge, individuals tend to favor it, despite it being statistically less probable. This simplification caters to the human need to make quick judgements rather than complex analyses.
Examples
- A famous example introduced by Tversky and Kahneman involves a fictional woman named Linda. When participants are asked whether Linda, who is described as a philosophy major and socially active, is more likely to be a bank teller or a bank teller who is also a feminist, the majority incorrectly choose the latter despite it being a less likely conjunction of traits.
- In a medical context, a patient might believe they have a rare disease because their symptoms align with those described for that disease plus their unique traits, rather than the far more common condition with a few overlapping symptoms.
Consequences
This fallacy can lead to poor decision-making in various fields, as the reliance on representative narratives over probabilistic data may result in incorrect judgments and beliefs. It can impact areas such as law, where juries may be swayed by emotional stories rather than factual evidence, and finance, where investors might assume unjustified correlations between market events.
Counteracting
To counteract the conjunction fallacy, individuals should be encouraged to engage in probabilistic thinking, assessing the base rates and statistical likelihood rather than relying on intuitions or narratives. Education and training focused on critical thinking and statistical literacy can help mitigate this bias.
Critiques
Some argue that the conjunction fallacy arises not solely from cognitive errors but also from misunderstandings in the framing of problems. They suggest that some instances attributed to the fallacy may involve different interpretations by respondents, where they focus on relevance or typicality instead of strict probability.
Also known as
Relevant Research
Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment
Tversky, A., & Kahneman, D. (1983)
Psychological Review, 90(4), 293
The 'conjunction fallacy' revisited: How intelligent inferences look like reasoning errors
Hertwig, R., & Gigerenzer, G. (1999)
Journal of Behavioral Decision Making, 12(4), 275-305
Case Studies
Real-world examples showing how Conjunction fallacy manifests in practice
Context
A mid-sized asset management firm ran a high-conviction growth portfolio where investment ideas were debated in a weekly committee. Portfolio managers favored richly detailed narratives about companies — product launches, management turnarounds, and market-share wins — when making allocation decisions.
Situation
An analyst presented a thesis that Company X would both beat quarterly earnings and successfully launch a new subscription product that would accelerate revenue growth. The committee found the narrative compelling and increased the fund's position in Company X well above its typical position limits based on the combined story.
The Bias in Action
Committee members treated the conjunction — “beat earnings AND successful product launch” — as more plausible than the simpler event “beat earnings,” because the richer story felt more representative and persuasive. Several members reported that the product launch narrative made the earnings beat seem almost inevitable, and the committee overweighted the stock as a result. They failed to decompose the joint probability, effectively ignoring that P(A and B) cannot exceed P(A), and did not seek quantitative probability estimates for each event separately.
Outcome
Three months later the company beat earnings but the product launch missed key adoption targets, reducing forward guidance. The fund’s overweight position amplified losses: Company X returned -28% from the decision date while the sector returned -6%. The committee’s conviction in the combined narrative delayed rebalancing and increased drawdown.