We explore decision-making under uncertainty using a framework that decom- poses uncertainty into three distinct layers: (1) risk, which entails inherent ran- domness within a given probability model; (2) model uncertainty, which entails subjective uncertainty about the probability model to be used; and (3) model mis- specification, which entails uncertainty about the presence of the correct probability model among the set of models considered. Using a new experimental design, we show that there exist distinct attitudes towards the three layers of uncertainty and examine the role of each of them in characterizing attitudes towards ambiguity. In addition to providing new insights into the underlying processes behind ambiguity aversion, we provide the first empirical evidence of the role of model misspecification in decision-making under uncertainty.

Practical information