This volume's goal is to begin to document the dialogue processes in naturally-occurring human tutoring, in the context of informing the design of intelligent tutoring systems, and of interactive systems in general. This project represents the first empirical study of human tutorial dialogue from a conversation analytic perspective — the conversational interaction is the focus of analysis rather than larger scale techniques for teaching. It is also the first study of tutoring to make use of large quantities of carefully transcribed tutoring conversations/dialogues.
The motivation for this focus comes from two sources: First, although all tutoring systems have implicit theory or theories of minute-level interaction built into them, little research has been done to form an empirical foundation for such theories. Therefore, current systems tend to be based on the designers' intuitions rather than on data. This fact almost certainly makes systems unnecessarily brittle in actual use. Second, of the small but growing collection of empirical studies of tutoring, almost all have been designed and carried out by computer scientists, whose training naturally leads them to be concerned with interaction at the level of knowledge transfer and teaching techniques. Fox's training as a linguist brings attention to the minute-by-minute details of the interaction, in particular to the processes that bring the interaction into existence and allow it to develop relatively smoothly.