Working Paper: NBER ID: w15084
Authors: Nicola Gennaioli; Andrei Shleifer
Abstract: We present a model of judgment under uncertainty, in which an agent combines data received from the external world with information retrieved from memory to evaluate a hypothesis. We focus on what comes to mind immediately, as the agent makes quick, intuitive evaluations. Because the automatic retrieval of data from memory is both limited and selected, the agent's evaluations may be severely biased. This framework can account for some of the evidence on heuristics and biases presented by Kahneman and Tversky, including conjunction and disjunction fallacies.
Keywords: Judgment under uncertainty; Heuristics; Cognitive biases
JEL Codes: D03; D81
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| automatic retrieval of scenarios from memory (D84) | biased evaluations of hypotheses (C52) |
| representativeness of recalled scenarios (C83) | decision-maker's assessment (D91) |
| biased evaluations of hypotheses (C52) | overestimation or underestimation of probabilities (D80) |
| limited and selective memory retrieval (D91) | biases such as base rate neglect and insensitivity to predictability (D91) |