Using Agent-Based Modeling to Test and Integrate Process-Oriented Perspectives of Leadership Emergence

By Bryan Acton in Research

May 22, 2020


As organizations utilize less hierarchical forms of leadership, the study of how leadership emerges within teams continues to grow in importance. Despite many theoretical perspectives used to study leadership emergence, little is understood about the actual process by which a collective structure emerges. In the current work, I address two of the primary limitations within this literature: imprecise theoretical perspectives and methodological challenges in studying emergence. Specifically, although there are many conceptual works that describe the leadership emergence process, these descriptions do not have enough precision to be able to design a model with formal rules, a necessary requirement for studying emergence. Additionally, studying leadership emergence requires the study of newly formed teams frequently over time, which is challenging to accomplish using existing methods. To address the two above limitations, in the current work, I translate two dominant process-oriented perspectives of leadership emergence (social interactionist and social cognitive) into formal theories that include a series of testable hypotheses. In doing so, these theories outline the essential elements and process mechanisms of each theoretical perspective. Next, I use these theories to design two agent-based models to simulate the process by which leadership emerges within teams, under each perspective. Using the software NetLogo, I simulate 500 newly formed teams over the initial period of 500 dyadic interactions (i.e., hours). Finally, after simulating these models, I use the resulting data to test the predictions from each theoretical perspective. In addition to testing the hypotheses from each model, I also utilize agent-based modeling to systematically test the relative importance of the unique individual-level elements and process mechanisms from each model. From this entire process, I generate results about (1) how well the agent-based models represent the respective perspectives, and (2) the relative influence each perspective’s unique elements and mechanisms have on team outcomes. Overall, results generally supported the core concepts from each perspective, but also identified areas where each perspective needs to revisit for theory on leadership emergence to advance. Specifically, the results illustrated that certain individual-level elements were most influential for leadership emergence. For the social interactionist perspective, it was the comparison between implicit leadership theories and self-prototypical leadership characteristics. For the social cognitive perspective, it was leader self-schemas. Additionally, results indicated that future work may need to revisit the conceptualization of both leadership structure schemas, as well as the dynamic process of weighting implicit leadership theories. Finally, predictions about the rate of leadership emergence over time within the social cognitive perspective were the only predictions that were not supported. From these results, I present multiple themes as a conceptual road map for the advancement of leadership emergence theory. I argue that the lack of support regarding leadership emergence trajectories presents opportunities for a reconceptualization of emergence at the event level, as well as new modeling procedures to capture emergence as it occurs. I also present future study ideas that can directly test the competing assumptions from each perspective. In total, I argue that this work advances the study of leadership emergence by adopting a method that helped integrate two dominant perspectives of leadership emergence, possibly laying the groundwork for the development of a combined formal theory.

Posted on:
May 22, 2020
3 minute read, 525 words
leadership emergence
See Also:
Putting emergence back in leadership emergence: A dynamic, multilevel, process-oriented framework