Next generation learners engage in self-paced individual and collaborative learning with digital tools, games, intelligent systems, and virtual environments more than ever before. The focus of this paper is on understanding how problem-solving is distributed and negotiated among high school learners and simulated Personal Assistants for Learning (PALs). In our research we envision a PAL as a coach that guides and supports the user's learning experiences, e.g., an early instantiation could be an intelligent tutor that can function across domains.

The OpenSimulator virtual environment (Imaginarium) we use in our study utilizes a factual complex problem scenario that takes place in the Pacific Northwest. Learners virtually explore how culture and the environment together act as a system that impacts sustainability science while interacting with a simulated PAL.

Their actions are monitored by the virtual environment, ADL experience API (xAPI), Learning Record Store, and the hybrid chat bot/human PAL. This paper is relevant to the I/ITSEC community because it explores the design and use of an open source virtual environment to 1) understand the future implications of a personal assistant for learning and 2) begin to formulate human-computer interaction (HCI) guidelines for next generation learning systems. This paper describes our study, reports preliminary findings, and addresses the following research questions: How is problem-solving negotiated between high school learners and a simulated PAL? Can rapid prototyping in virtual environments expose assumptions about instructional best practices and improve the design of technology-mediated learning experiences? While human interactions with future PALs cannot be accurately predicted, simulating these phenomena in an immersive environment allows for exploration of the effects of technology on human behavior.