Teams research is fragmented. Every discipline that studies groups — organizational behavior, social psychology, HCI, economics, operations — has its own favorite tasks: brainstorming, jury deliberation, prisoner’s dilemma, hidden profile, creative writing prompts, estimation games. Results on these tasks rarely talk to each other, and it’s genuinely hard to tell whether that’s because the findings conflict or because the tasks do. This problem extends beyond teh task to how tasks are operationalized and how experiments are parameterized and measured.

With Xinlan Emily Hu, Linnea Gandhi, Duncan Watts, and Abdullah Almaatouq, we introduce the Task Space: a framework that organizes the tasks teams do along dimensions that matter for how teams actually perform them, so findings from one task can be meaningfully compared with findings from another Management Science.

This sits on top of a stack of earlier work that kept pointing at the same gap. In Did It Have To End This Way? we showed that the same teams can produce very different outcomes depending on the task in front of them CSCW’19. In Parallel Worlds we re-convened the same teams without them knowing to see how much of their trajectory was locked in CSCW’20. And in Online Juries we found that for some decision tasks, teams are remarkably consistent — while for others, they’re essentially random CHI’21.

The Task Space takes the next step: rather than studying individual tasks and hoping the findings generalize, it maps the space those tasks live in so that generalization can be tested directly.