In a recent paper, researchers use agent-based models to explore how hierarchical groups perform under varying conditions. (image: Stan Rhodes)

Most organizations operate under command hierarchies: workers, who know the ground reality, report to managers, who know the big picture. If these views conflict, what happens to organizational performance? In a study published on March 31 in the Journal of Artificial Societies and Social Simulation, Stan Rhodes (Colorado State University) with SFI External Professor Stefani Crabtree (Utah State University), and Jacob Freeman (Utah State University) use an agent-based model to explore how the performance of hierarchical groups varies with changing environments.

The researchers simulated hierarchical and non-hierarchical organizations as they responded to situations at two extremes: when local conditions changed synchronously, or at staggered times. In most scenarios, teams with a hierarchical structure performed better than those without, with one crucial caveat: workers must have the autonomy to judge the manager’s input when deciding what to do. Worker autonomy, the study finds, allows a hierarchical organization to learn.

Read the paper “An Agent-Based Model of Hierarchical Information-Sharing Organizations in Asynchronous Environments” in the Journal of Artificial Societies and Social Simulation (March 31, 2024). DOI: 10.18564/jasss.5328