There are many different kinds of context information, with different demands on representation. In this paper we discuss the representation of the context information needed to support intelligent dialogue management in interactive speech systems. We argue that simple types of context information should be represented in a very simple form to allow efficient processing, while other types of context information require sophisticated logics for articulate representation and reasoning. For dialogue management in spoken natural language dialogue systems, a complication is that some types of context information most of the time have a very simple structure, but occasionally may be quite complex, and require articulate representation. We will propose a way to handle this problem, and show how a dialogue manager can operate on elements of context information that may be represented in different formalisms.