EveryBOTy Counts: Examining Human–Machine Teams in Open Source Software Development

Abstract

In this study, we explore the future of work by examining differences in productivity when teams are composed of only humans or both humans and machine agents. Our objective was to characterize the similarities and differences between human and human–machine teams as they work to coordinate across their specialized roles. This form of research is increasingly important given that machine agents are becoming commonplace in sociotechnical systems and playing a more active role in collaborative work. One particular class of machine agents, bots, is being introduced to these systems to facilitate both taskwork and teamwork. We investigated the association between bots and productivity outcomes in open source software development through an analysis of hundreds of project teams. The presence of bots in teams was associated with higher levels of productivity and higher work centralization in addition to greater amounts of observed communication. The adoption of bots in software teams may have tradeoffs, in that doing so may increase productivity, but could also increase workload. We discuss the theoretical and practical implications of these findings for advancing human–machine teaming research.

Introduction to the Emerging Cognitive Science of Distributed Human‐Autonomy Teams

“…They may remain a team but fluctuate between high and low degrees of teamness (Cooke et al, 2022). Newton, Saadat, Song, Fiore, and Sukthankar (2024) have explored all human teams and human-bot teams in open-source software development and find that human-bot teams are associated with more productive humans and more centralized work than are human-human teams. They also pay attention to some dimensions of teams that may be critical such as team size and team expertise.…”

“…We speculate that the spatiotemporal spread of trust and distrust as a form of social information is amenable to interactive and real-time team cognition analysis with, for example, influence (communicative or behavioral) being a generalizable mechanism of spread Huang et al, 2021;Zhou et al, 2023). Newton et al (2024) analyzed human-BOT interactions relying on metrics aggregated over time using a Macrocognition in Teams lens. Macrocognition in Teams, we think, is closely related to distributed cognition in which cognitive processing occurs across human and technological system elements (Hutchins, 1991), making it a particularly suitable framework for assessing team cognition in HATS.…”

“…Newton et al. (2024) analyzed human–BOT interactions relying on metrics aggregated over time using a Macrocognition in Teams lens. Macrocognition in Teams, we think, is closely related to distributed cognition in which cognitive processing occurs across human and technological system elements (Hutchins, 1991), making it a particularly suitable framework for assessing team cognition in HATS.…”