Optimal Attention

Abstract

In many environments it is costly for decision makers to determine which option is best for them because learning about different options takes time, and time is valuable to decision makers. To accurately predict changes in behavior and estimate preferences in such settings it is crucial to understand what information agents choose to acquire. In the following three chapters I study costly information acquisition both theoretically and experimentally. In the first chapter I characterize the choice patterns that are consistent with two major models of costly learning when price changes and then test these theoretical results in a lab setting. The experiment shows that the heterogeneity of subjects is key for understanding aggregate behavior and that subjects are sophisticated when choosing both if and what to learn. In the second chapter I use an axiomatic foundation to create a new measure for the cost of information that allows for multiple perceptual distances in a single choice environment so that some events can be harder to differentiate between than others. In the third chapter I show that the new measure of uncertainty produced in the second chapter maintains the tractability of Shannon's classic measure but produces richer choice predictions, identifying a new form of informational bias significant for both welfare and counterfactual analysis.

The third chapter also establishes a new foundation for `non-compensatory' behavior and demonstrates novel predictions for the formation of consideration sets.