Why Specific Seems More Likely Than General
A conjunction (A AND B) can never be more probable than either element alone (A or B).
The orange circle (feminist bank tellers) fits entirely inside the green circle (all bank tellers). It can never be larger!
Amos Tversky and Daniel Kahneman presented the Linda problem to participants of varying statistical sophistication. The results were stunning:
The effect persisted even when participants were explicitly warned about the conjunction rule. Our intuition about "fit" overrides our logical knowledge.
Adding details creates a more vivid, coherent mental image. "Bank teller" is bland and generic. "Bank teller active in the feminist movement" paints a specific person we can imagine—and that imagined person matches Linda's description perfectly.
This is the representativeness heuristic: we judge probability by similarity to a prototype rather than by base rates and logical rules.
Detailed scenarios seem more plausible to juries than general claims
Specific diagnoses feel more likely than general categories
Detailed market narratives seem more credible than uncertainty
Specific forecasts feel more believable than vague ones
When evaluating probability claims, always ask: "Is this a subset of a larger category?" If someone says X AND Y is likely, remember that X alone must be at least as likely. Be especially skeptical of vivid, detailed predictions—they may feel more real, but they're mathematically constrained to be less probable.