Human-Agent Teamwork in Collaborative Virtual Environments

The growing interest in using Collaborative Virtual Environments (CVEs) increases the need to create heterogeneous teams comprised of humans and Intelligent Virtual Agents (IVAs). The main aim of my PhD is to study the factors that tend to improve team performance and foster collaboration between humans and IVAs in CVEs. To reach this aim, a framework, namely MACVILLE, was designed for multi-agent collaborative virtual learning environment based on Activity Theory was proposed. In addition, an agent architecture integrated with a human-agent teamwork communication model, namely HAT-COM, was developed. Three experiments were conducted to investigate: a) the impact of multimodal communication, verbal and nonverbal, on the development of a Shared Mental Model (SMM), task-based and teambased, between humans and IVAs; b) the effect of the development of a SMM on human-IVA team performance; c) criteria that are likely to break the developed SMM in human-IVA teams.