To make the layers of theoretical explanation (LTE) more concrete and to highlight the depth of explanation, prediction, and control provided by computational process theory (CPT) compared to construct theory (CT), we developed an agent-based simulation (ABS) examining teams operating under autocratic versus delegative leadership. The ABS defines operational rules for processes such as team member evaluation; task assignment; team members deciding whether and how to perform tasks; team members performing tasks; and updating expertise, priorities, and social influence as performance episodes unfold. The ABS generates simulated construct and process data, which, in turn, is used to compare the quality of explanation (LTE), prediction, and control offered by CT versus CPT.
In this example, the CT employs a typical mediation model in which team member motivation serves as the explanatory mediator for the effect of leadership style on team performance (i.e., leadership style  team member motivation  team performance). We found the indirect effect accounted for much of the total effect, suggesting team member motivation fully mediates the relationship between leadership style and team performance. However, using our CPT, we examined a key process: evaluating team member expertise. Autocratic leaders make work decisions single handedly, including evaluating all team members. Although autocratic leaders may not make biased evaluations across team members, they do make biased evaluations about specific team members. Under delegative leadership, team members average their individually biased evaluations for a teammate. As a result, their average evaluation leads to less biased evaluation of specific team members. By considering the evaluation process in depth, the CPT provides a more complete understanding of how differences in team performance emerged from the two leadership styles. In turn, this more complete coverage of the LTE translates into better predictions of, and interventions on, team performance, which our presentation explores.