Researchers from the ColMOOC team in the University of Valladolid have compared different configurations of conversational agents (either more focused on knowledge support or more oriented to promoting social interactions) with pre-service teachers enrolled at that University. ColMOOC agent software was successfully used to carry out that study.
The results show that students who received knowledge-based prompts significantly exchanged more messages with evidence of explicit reasoning and were more satisfied with the agent and their discussion during the task.
The study proposes a combination of different pedagogical agent behaviors that best adapt to the nature of the task and the student behavior. This research will be presented at the European Conference on Technology Enhanced Learning (ECTEL2020) on September 16 by a member of the research team, Konstantinos Michos. If you are interested, the full paper can be accessed here: https://link.springer.com/chapter/10.1007/978-3-030-57717-9_16.
The research community is excited to know more about conversational agents and the ColMOOC project, as the conference organization has nominated this research for the best paper award (see https://ectelfi20.org/best-paper-candidates-published).