The Effect of Cognitive Load on a Statistical Dialogue System

In recent years statistical dialogue systems have gained significant attention due to their potential to be more robust to speech recognition errors. However, these systems must also be robust to changes in user behaviour caused by cognitive loading. In this paper, a statistical dialogue system providing restaurant information is evaluated in a set-up where the subjects used a driving simulator whilst talking to the system. The influences of cognitive loading were investigated and some clear differences in behaviour were discovered. In particular, it was found that users chose to respond to different system questions and use different speaking styles, which indicate the need for an incremental dialogue approach.