Given a vivid description, we judge by how well it fits a stereotype and ignore the underlying odds — the base rate. So we call the quiet poetry-lover a librarian, forgetting there are vastly more salespeople than librarians.
Base-rate neglect: when we have a specific, vivid description of a case, we judge probability by how well it fits a stereotype — and ignore the base rate, how common each possibility actually is.
Kahneman and Tversky described 'Steve': shy, withdrawn, helpful but with little interest in people, a love of order and detail. Is he more likely a librarian or a farmer? Most say librarian — he fits the stereotype. But there are far more farmers than librarians, so even if librarians are likelier to be shy, the sheer numbers make 'farmer' the better bet. The vivid portrait crowds out the base rate.
The base rate is just how frequent each option is to begin with. Our minds latch onto the specific story and treat the background frequencies as irrelevant — a deep flaw, because the base rate often dominates. A classic medical version: a test is 95% accurate for a disease that affects 1 in 1,000. You test positive — your real chance of having it is still under 2%, because the rare disease's tiny base rate swamps the test's accuracy. People (including doctors) routinely guess 95%, neglecting the base rate entirely.
The fix is to ask 'how common is this in the first place?' before letting the story decide. When something feels like a perfect match for a category, check whether that category is actually rare — a perfect fit for a rare thing is still usually something common. Start from the odds, then let the specific evidence adjust them; don't let a vivid description erase the math.
From medical tests to hiring to fearing rare dangers, we let a compelling description override the actual odds — and the base rate often matters far more than the story, so ignoring it produces confidently wrong judgments.
It's not 'ignore the details, only use base rates.' Specific evidence genuinely matters and should update your estimate. The error is letting the vivid description REPLACE the base rate instead of adjusting it. Good reasoning starts from the odds and moves with the evidence — it uses both.
Lock this idea into memory with a 5-minute active-recall session — the science of spaced repetition, no signup.
Try this idea free →