Engineering Active Learning in 3D Virtual Worlds

Abstract

AbstractTo advance education in the 21st century, experts in neuroscience argue for more ‘active learning’ through experience. A practical implementation for such active, experiential and, we propose, multi-disciplinary learning is for students to consider, analyze, solve and make personal meaning from engaging problems. Collaborative problem-solving promotes communication involving creative and interpretive meaning-making, analysis and reflective judgement. Due to advances in computer technologies, educators can now consider a wider range of collaborative problems and extensive solutions which can be implemented in both real and virtual worlds. This paper will summarise five scenarios of virtual worlds from our research project: Second Life, OpenSim, OpenQwaq, Unity 3D and Unity 3D with Oculus Rift. In each virtual world participants from Japan, UK and USA have been engaged in remote collaborations to solve problem-based tasks requiring the programming and manouvering of virtual robots and real-world LEGO Mindstorms robots. Robot Mediated Interactions are analyzed to determine robot task complexity, student learning and task immersion. We have determined that for effective active learning involving 3D virtual worlds specifically focusing upon the development of beginner programming knowledge, the challenge is for educators to design tasks of zero (or close to zero) Task Fidelity and for students to become fully immersed within the tasks. The paper explains how this has been achieved, summarizes the benefits, obstacles, limitations and challenges of 3D virtual world implementations, and suggests future directions in educational research to engineer active learning in 3D virtual worlds.KeywordsComputer science educationVirtual worldsRobot task complexityLearning