Abstract:

Group meetings are ubiquitous, with millions of meetings held across the world every day. However, meeting quality, group performance, and outcomes are challenged by a variety of dysfunctional behaviors, unproductive social dynamics, and lack of experience in conducting efficient and productive meetings. Previous studies have shown that meeting facilitators can be advantageous in helping groups reach their goals more effectively. However, many groups do not have access to a human facilitator due to a lack of resources or other barriers. In this work, I leverage advances in conversational agent technology to provide some of the functionality of human facilitators in group decision-making using an automated system. I present a general computational framework for automated meeting facilitation that provides a real-time, co-located, synchronous, multiparty, personified interface to a group conducting a face-to-face meeting. The framework is designed to support a range of virtual and robotic embodiments of the agent, and provides facilitation functions based on group management science, including a) management of meeting structure; b) management of participation to avoid dominance; and c) management of conflict. Using this framework, I report on three prototypes and their evaluation to address a series of research questions. I first explored the overall acceptance of a conversational agent in the role of a meeting facilitator and compared reactions and outcomes to different agent embodiments in the meeting room, finding that embodiment improved participants’ rapport with and trust of the agent as well as their perceptions of the agent’s intelligence and power. Second, I developed a fully automated meeting facilitation robot that uses multimodal inputs to manage the multiparty conversation, enforce meeting structure, promote time management, balance group participation, and facilitate group decision-making processes. Results of a between-subject study showed that the robot facilitator was accepted by group members, was effective in enforcing meeting structure, and that users found it helpful in balancing group participation. Finally, I developed and evaluated conflict management strategies for the facilitation robot and evaluated them in a task setting designed to evoke task conflict. I found that participants complied with the robot’s intervention, performing active listening with each other when conflicts arose. I also report on qualitative findings from interviews with study participants and a focus group, highlighting the potential role for robots in meetings of the future.