Models and Anti-Models: The Structure of Payoff-Dependent Social Learning

摘要: We conducted an experiment to describe how social learners use information about the relation between payoffs and behavior. Players chose twontechnologies repeatedly. Payoffs were random, but one technology was better because its expected payoff higher. divided into two groups:n1) individual who knew their realized after each choice andn2) learners, had no private feedback own payoffs, in period could choose learn which behavior produced lowestnpayoff among or highest payoff. When know producing payoff, a model of imitating this successful matches data verynclosely. lowest they tended opposite early periods, while increasingly choosing same late periods. This kind rapidntemporal heterogeneity has received little attention theoretical study learning.