Publication Date


Document Type


Committee Members

William S. Horton (Committee Member), Ion Juvina (Committee Member), Valerie L. Shalin (Advisor), Scott N. J. Watamaniuk (Committee Member)

Degree Name

Doctor of Philosophy (PhD)


Human-human communication is a coordinated dance (Clark, 1996) that requires each participant to consider the other participants. The majority of this coordination centers on the conversational grounding process that develops and maintains the common ground, or shared understanding between the individuals (Clark and Schaefer, 1989). Conversational grounding is also a crucial process for human-computer interaction using language-based methods, such as spoken dialogue systems. Previous work has tied grounding processes to the performance outcomes in collaborative tasks (Reitter and Moore, 2014; Gergle et al., 2013, 2004; Clark and Krych, 2004), making it a high priority for increasing capabilities of spoken dialogue systems. The model of grounding for human-computer interaction should be informed by human-human dialogue. However, the processes involved in human-human grounding are under dispute within the research community. Three models have been proposed: alignment, a simple model that has been influential on dialogue system development, interpersonal synergy, an automatic coordination emerging from interaction, and audience design, a strategic interaction based on intentional coordination. Interpersonal synergy and audience design are two different types of coordination models. Previously, only one study has tested both the alignment and coordination models simultaneously. Fusaroli and Tylen (2016) introduced communication models based on recurrence quantification analysis to model the amount of repetition between speakers. The current research extended their models to differentiate between the types of coordination. Throughout, the current research applied Fusaroli and Tylen's methods to richer stimuli/tasks that generate longer dialogues with larger vocabulary and more influences on performance outcomes. Through analysis of four different dialogue tasks, the current work also examined how common ground processes change as a function of the task characteristics. Subsequent analyses investigated the validity of the nascent recurrence models. The results showed strong support for the coordination model over the alignment model in human-human communication, Additional results suggested that coordination is the audience design variant. These results place a new requirement on the design of human-computer interaction mechanisms.

Page Count


Department or Program

Department of Psychology

Year Degree Awarded