Shared, Replicated, or Separated? a Comparative Study of Virtual Workspace Configurations for Collaborative Hands-on Learning

Most work on collaborative immersive systems mimics real-world settings, using fully shared virtual workspaces that foster close collaboration. However, recent work in educational contexts using remote desktop environments suggests these shared approaches may not be optimal for learning, as it showed that individual workspaces lead to better learning outcomes. In this paper, we investigate whether individual workspaces also lead to better outcomes in a collaborative VR learning environment. We compare three distinct workspace configurations in a problem-solving task: (1) a fully shared environment where two users work on the same materials, (2) a replicated environment where each user has their own copy of the materials but can still see their partner and their workspace, and (3) a separated environment where users cannot see each other nor each other's workspace and each has individual materials. We evaluate how these configurations influenced collaborative interaction, problem-solving strategies, and learning. Our results suggest the replicated workspace reduced social experience and did not improve learning outcomes compared to the shared one, however, it allowed broader exploration of the problem space.