Joint visual attention and collaboration in Minecraft

In the service of a project organizing social futures (Penuel & O’Connor, 2018) in virtual worlds, this paper develops methods for analyzing collaboration in Minecraft through joint visual attention (JVA) and other multimodal learning analytics, and contributes a software package for doing so. We continuously calculate each player’s position, gaze vector, and the block upon which their gaze falls based on server logs, revealing when players are looking at the same place at the same time. We show that JVA functions similarly in online worlds as offline, and provide examples of how JVA and other analytics can supplement qualitative analysis of online collaboration. In contrast to previous research on joint visual attention using dual eye-tracking glasses, JVA research in virtual worlds allows for larger collaborative groups, freedom of movement, and the opportunity to observe how players establish and negotiate genres of multimodal interaction through new forms of embodiment.

Project repository (GitHub)

Download full text


Proctor, C. & Muller, D. (2022). Joint visual attention and collaboration in Minecraft. In A. Weinberger, W. Chen, D. Hernández-Leo, & B. Chen (Eds.). (2022). Proceedings of the 15th International Conference on Computer-Supported Collaborative Learning - CSCL 2022. (pp. 226-233). Hiroshima, Japan: International Society of the Learning Sciences. (Best paper nomination. 46% accepted.)


  title = {Joint Visual Attention and Collaboration in {{Minecraft}}},
  booktitle = {Proceedings of the 15th {{International Con-}} Ference on {{Computer-Supported Collaborative Learning}} - {{CSCL}} 2022},
  author = {Proctor, Chris and Muller, Dalia Antonia Caraballo},
  editor = {Weinberger, A and Chen, W and {Hern{\'a}ndez-Leo}, D and Chen, B},
  year = {2022},
  pages = {226--233},
  publisher = {{International Society of the Learning Sciences}},
  address = {{Hiroshima, Japan}},
  langid = {english}