Dialogue Cohesion Sharing and Adjusting in an Enhanced Multimodal Environment

Abstract

Multimodality is a powerful concept for dealing with dialogue cohesion in a Human-Computer NL-centered system. This work is a little step in better exploiting the potential large bandwidth of communication provided by this situation. Light cognitive load feedback and direct manipulation are proposed so that user and system can cooperate in mutually conveying the cohesion of the ong?ing dialogue. Main points are: (i) use of an appropriate dialogue structure to constrain inference in the anaphor resolution process; (ii) use of a graphical representation of the structure, to limit the problem of opacity; (iii) allow the possibility of direct manipulation on this representation, to avoid the necessity of operating linguistically at the meta-level. The context of the work is within a NL-centered multimodal information access system, in which basic entities are pairs (most commonly question and answer). A dialogue model simple enough to be shown in an intuitivereading way on the screen, and powerful and accurate enough is provided by a modified version of the centering model.