Computational Interpersonal Communication: Communication Studies and Spoken Dialogue Systems

With the advent of spoken dialogue systems (SDS), communication can no longer be considered a human-to-human transaction. It now involves machines. These mechanisms are not just a medium through which human messages pass, but now occupy the position of the other in social interactions. But the development of robust and efficient conversational agents is not just an engineering challenge. It also depends on research in human conversational behavior. It is the thesis of this paper that communication studies is best situated to respond to this need. The paper argues: 1) that research in communication can supply the information necessary to respond to and resolve many of the open problems in SDS engineering, and 2) that the development of SDS applications can provide the discipline of communication with unique opportunities to test extant theory and verify experimental results. We call this new area of interdisciplinary collaboration “computational interpersonal communication” (CIC)