Why are People Bad at Detecting Randomness? Because it is Hard

摘要: People often detect structure and patterns in data that is random. This difficulty accurately evaluating randomness manifests itself mistaken beliefs a fair coin has bias towards heads or tails, detection of causal relationships between variables randomly co-occur, observation illusory correlations continuous variables. A computational analysis an optimal reasoner’s performance on these three tasks suggests this does not arise simply because people have irrational disposition to see meaning randomness, but the underlying inference problem intrinsically hard– for both statistical human intuition. produced inherently ambiguous, which can also be by systematic process. An experiment reported provides evidence inferences about are difficult.