Supporting Collaborative Online Science Education With a Transferable and Configurable Conversational Agent

Teachers in online science laboratories frequently struggle to provide timely discussion support to different groups.Collaborative conversational agents have been proposed to bridge this gap and foster productive talk in small-group discussions.Current implementations are usually highly configurable, but rather limited when modeling users' intent in a transferable way.In this paper, we propose a new conversational agent that supports productive talk while addressing these requirements.Also, we discuss lessons learned from a pilot study in a science classroom with secondary students.Findings illustrate the agent's potential to support productive talk among students during online science lab activities.