When Teams Shift Among Processes: Insights From Simulation and Optimization

• 成果类型: Article
• 署名作者: Kennedy, Deanna M.; McComb, Sara A.
• 署名单位: University of Washington; University of Washington Bothell; Purdue University System; Purdue University; Purdue University System; Purdue University
• 刊物名称: JOURNAL OF APPLIED PSYCHOLOGY
• ISSN/ISSBN: 0021-9010
• DOI: 10.1037/a0037339
• 发表日期: 2014
• 页码: 784-815
• 关键词: team; transition process; interventions; genetic algorithm; simulation

摘要

This article introduces process shifts to study the temporal interplay among transition and action processes espoused in the recurring phase model proposed by Marks, Mathieu, and Zacarro (2001). Process shifts are those points in time when teams complete a focal process and change to another process. By using team communication patterns to measure process shifts, this research explores (a) when teams shift among different transition processes and initiate action processes and (b) the potential of different interventions, such as communication directives, to manipulate process shift timing and order and, ultimately, team performance. Virtual experiments are employed to compare data from observed laboratory teams not receiving interventions, simulated teams receiving interventions, and optimal simulated teams generated using genetic algorithm procedures. Our results offer insights about the potential for different interventions to affect team performance. Moreover, certain interventions may promote discussions about key issues (e. g., tactical strategies) and facilitate shifting among transition processes in a manner that emulates optimal simulated teams' communication patterns. Thus, we contribute to theory regarding team processes in 2 important ways. First, we present process shifts as a way to explore the timing of when teams shift from transition to action processes. Second, we use virtual experimentation to identify those interventions with the greatest potential to affect performance by changing when teams shift among processes. Additionally, we employ computational methods including neural networks, simulation, and optimization, thereby demonstrating their applicability in conducting team research.