Conjunction Fallacy
Judging a detailed scenario as more likely than a simpler one
What is it?
The conjunction fallacy is the error of judging that two events together are more likely than one of them alone, when a combination can never be more probable than either of its parts. In Amos Tversky and Daniel Kahneman's well-known "Linda" problem, a woman is described as outspoken and concerned with social justice; many people then rate "Linda is a bank teller and active in the feminist movement" as more probable than "Linda is a bank teller". The detailed version fits the description better, so it feels more likely, even though each added detail can only lower the odds. In decisions, this makes rich, coherent stories seem more credible than plain ones: a business plan with many specific steps, a vivid threat scenario or a precise forecast. It tends to get worse when the details are plausible and consistent with each other, and when you judge a story by how well it hangs together rather than by how many things must all go right.
Example
Believing a startup will win because its plan describes a precise sequence of partnerships, hires and launches. Worrying more about a specific break-in scenario than about burglary in general. Trusting a candidate's detailed five-year career story more than a vague but equally likely one.
References
Tversky, A., & Kahneman, D. (1983). Extensional Versus Intuitive Reasoning: The Conjunction Fallacy in Probability Judgment. Psychological Review, 90(4), 293-315.
Hertwig, R., & Gigerenzer, G. (1999). The 'Conjunction Fallacy' Revisited: How Intelligent Inferences Look Like Reasoning Errors. Journal of Behavioral Decision Making, 12(4), 275-305.
How to Prevent It
Doxa uses AI and can make mistakes. How it's built
How many separate things must all happen for this scenario to come true?
Does this story feel likely because it is probable, or because it is coherent?
What is the simpler version of this claim, and is it more likely?
Did adding a detail make me more confident rather than less?
Which single link in this chain is the weakest?
Break the scenario into its steps and estimate the odds of each one.
Multiply rough step probabilities to see how fast the total shrinks.
Compare the detailed forecast with the broad outcome it belongs to.
Think in frequencies: out of 100 similar cases, how many fit every detail?
Ask a colleague to strip a plan down to its essential assumptions.