Contained: Using Multiplayer Online Games to Quantify Success of Collaborative Group Behavior


  • Damian Debkowski Rutgers University
  • Andrew Marrero Rutgers University
  • Nicole Yson Rutgers University
  • Li Yin Rutgers University
  • Yichen Yue Rutgers University
  • Seth Frey Dartmouth College
  • Mubbasir Kapadia Rutgers University



Every day, millions of people gather on online game servers to collaborate in real time toward shared goals. What may seem like frivolous activity may, when investigated more closely, provide revolutionary opportunities to advance the science of teamwork. Teamwork is an important part of modern society, however, collaboration between people is often made difficult due to differing ideals, opinions, and personality types. We leverage a popular self-hosted multiplayer online game environment to design a framework for developing and deploying tasks that elicit different kinds of teamwork. We propose to use these to capture fine-scale details of individual and group performance across environments. The game in which we implement this system, Minecraft, is ideal because it is heavily modifiable and already enjoys a vast user base of surprising gender, age, and ethnic diversity. We heavily modify the game in order to introduce new mechanics that facilitate collaboration, monitor activity, and manipulate group composition, all to provide the groundwork for deeper quantitative insights into effective teams.




How to Cite

Debkowski, D., Marrero, A., Yson, N., Yin, L., Yue, Y., Frey, S., & Kapadia, M. (2021). Contained: Using Multiplayer Online Games to Quantify Success of Collaborative Group Behavior. Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, 12(1), 149-154.