Social Science Job Candidate

What do we notice and how does this affect what we learn and come to believe? I present a model of an agent who learns to make forecasts on the basis of freely available information, but is selective as to which information he attends because of limited cognitive resources. I model the agent's choice of whether to attend to and encode information along a dimension as a function of his current beliefs about whether such information is predictive, taking as given that he attends to information along other dimensions. If the agent does not attend to and encode some piece of information, it cannot be recalled at a later date. He uses Bayes' rule to update his beliefs given encoded information, but does not attempt to fill in missing information. I show that, as a consequence of selective attention, the agent may persistently fail to recognize important empirical regularities, make biased forecasts, and hold incorrect beliefs about the statistical relationship between variables. In addition, I identify factors that make such errors more likely or persistent. The model sheds light on a set of systematic biases in inference, including the difficulty people have in recognizing relationships that prior theories do not make plausible, and the overattribution of cause to salient event features. The model is applied to help understand the formation and stability of erroneous stereotypes, as well as discrimination based on such stereotypes.