SIGdial 2014

This paper explores dialogue adaptation over repeated interactions within a taskoriented human tutorial dialogue corpus. We hypothesize that over the course of four tutorial dialogue sessions, tutors adapt their strategies based on the personality of the student, and in particular to student introversion or extraversion. We model changes in strategy over time and use them to predict how effectively the tutorial interactions support student learning. The results suggest that students leaning toward introversion learn more effectively with a minimal amount of interruption during task activity, but occasionally require a tutor prompt before voicing uncertainty; on the other hand, students tending toward extraversion benefit significantly from increased interaction, particularly through tutor prompts for reflection on task activity. This line of investigation will inform the development of future user-adaptive dialogue systems.