IOE 333 Chapter Notes - Chapter 14: Confirmation Bias, Loss Aversion, Decision Tree Learning

• School: U of M
• Department: Industrial And Operations Engineering
• Course: IOE 333
• Professor: Clive D'Souza

Document Summary

Human decision making models: traditional optimal or normative models, descriptive. Expected value theory (evt): compare expected value of outcomes, based on probabilities of event occurring. Satisfice: pick option that works satisfactorily, for now. Tendency to prefer avoiding sure loss, or a guaranteed sure thing (loss aversion heuristic) Humans should make rational choices, selecting optimal outcome. Identifying optimal value using costs, benefits, probabilities of outcome. Specifies what people ideally should do, not what they actually do. Limited psychological resources: human as information processor, limits on working memory, long-term memory, attention. Utilities cannot be estimated over the long term. Heuristic and biases: since people have limited psychological and informational resources, they use short-cuts to make decisions with those resources limitations, which lead to predictable biases (errors) in the decisions people make. Ex: gambler"s fallacy: flipping a coin has heuristic representativeness. Representativeness: people overgeneralize the representativeness of info. Availability: people rely on things that are easy to call to mind.