Inattentive Inference

Abstract

This paper studies how people make inference about a state of the world when the information structure includes additional, payoff-irrelevant states. For example, learning about effort from observed performance may require accounting for the otherwise irrelevant role of luck. This creates an attribution problem that is common to all information structures with multiple causes. We report controlled experimental evidence for pervasive over-inference about states that affect utility given an action, providing an explanation for a collection of well-known but previously unconnected mis-attribution patterns. In studying why systematic mis-attribution arises, we consistently find that errors are not due to excessive task complexity or deliberate effort avoidance. Instead, people fail to notice the need to account for alternative causes. Mis-attribution responds to a variety of attentional manipulations, but not to changes in the costs of inattention.

Keywords: Belief, Attention, Bounded Rationality, Learning, Unawareness

JEL Classification: C91, D01, D83, D84