Conversational Agents for Advanced Learning: Applications and Research

Learning/training is an important application domain for intelligent agents technologies supporting believable dialogues and realistic virtual environments. In this paper, we first review, in section 2, some of the current applications and research in technology-enhanced learning. In section 3 we discuss the impact of conversational agents on the design of more effective individual and group/community learning environments and processes. Three fundamental emerging approaches to integrate conversational agents in learning environments, at the individual as well as at the group/community/organizational level, are identified: (1) Conversational Agents as advanced help and learning process facilitation tools, (2) Conversational Agents as personal coaches equipped with pedagogical models as well as specific domain knowledge, and (3) Conversational Agents as roleplaying actors in simulated experiential learning environments. In the following sections of the paper we provide specific examples of such conversational agents, focussing particularly on the two last, most advanced categories. Section 4 provides a brief overview of K-InCA, a personal coach agent, supporting learning of knowledge management concepts, practices, and behaviours. Section 5 illustrates the application of Conversational Agents as role-playing actors in simulated experiential learning environments. Starting with a brief description of the EIS Simulation to provide a concrete context for experimentation with Conversational Agents in simulation environments, and then illustrates two planned extensions of the simulation: (1) C-VIBE (role of conversational agents to enhance and extend the learning experience to the domain of virtual teams coordination and decision making) and (2) extension to learning interpersonal and social skills in micro-domains (simulating organizational interactions / dialogues and meetings). In the conclusions we outline a number of different research questions to be addressed to conduct further research on the impact of Conversational Agents on learning.