In many dynamic economic settings, a decision maker finds it challenging to quantify the uncertainty or to assess the potential for mistakes in models. We explore alternative ways of acknowledging these challenges by drawing on insights from decision theory as conceptualized and implemented in statistics, engineering, and economics. Building on prior research, we suggest tractable and revealing ways to incorporate behavioral responses to uncertainty, broadly conceived. Our analysis adopts recursive intertemporal preferences for decision makers that allow them to be ambiguity averse and concerned about the potential misspecification of subjective uncertainty. By design, these representations have revealing implications for continuous-time environments with Brownian information structures. Problems where uncertainty’s structure is obscure such as macroeconomics, finance and climate change are promising areas for application of these tools.