Complexity Aversion

This paper proposes a model of decision-making under uncertainty in which an agent is constrained in her cognitive ability to consider complex acts. We identify the complexity of an act according to the corresponding partition of state space. The agent ranks acts according to the expected utility net of complexity cost. A key feature of this model is that the agent is able to update her complexity cost function after the arrival of new information. The main result characterizes axiomatically an updating rule for complexity cost function, the Minimal Complexity Aversion representation. According to this rule, the agent measures the complexity cost of an act conditional on the new information by using the cost of another act that gives exactly the same partition of the event but with the lowest ex-ante cost.