SPECIAL ISSUE ARTICLE Reflection in the heat of the moment: The role of in‐action team reflexivity in health care emergency teams Jan B. Schmutz 1 | Zhike Lei 2 | Walter J. Eppich 3 | Tanja Manser 4 1 Department of Management, Technology, and Economics, ETH Zürich, Zürich, Switzerland 2 The Graziadio Business School, Pepperdine University, Malibu, California, U.S.A. 3 Northwestern University Feinberg School of Medicine, Chicago, Illinois, U.S.A. 4 FHNW School of Applied Psychology, University of Applied Sciences and Arts Northwestern Switzerland, Olten, Switzerland Correspondence Jan B. Schmutz, Department of Management, Technology, and Economics, ETH Zurich, Weinbergstr. 56/58, 8092, Zürich, Switzerland. Email: jschmutz@ethz.ch Funding information Swiss National Science Foundation, Grant/ Award Number: PP00P1_128616 Summary Team reflexivity (TR)–defined as a team's conscious reflection on their objectives, strate- gies, and processes—is an important team process that fosters adaptation and information processing. However, traditional conceptualizations frame TR as a process that occurs in periods of downtime to reflect on past, terminated performance, largely ignoring reflective team processes occurring during intense performance events of action teams. To address this gap, we conceptualize TR as a team process that occurs not only during periods of downtime after the action but also during performance events as brief TR moments. We elaborate on the concept of in‐action TR and explore it by delineating its relationship to task type and timing during a performance event. Further, we test a team level contin- gency model of in‐action TR, namely, team size and performance. Using behavior observa- tion, we test our hypothesis with 70 medical teams responding to simulated in‐hospital emergencies. Task type is related to in‐action TR and reflection tends to increase as action progresses. Further, in‐action TR is related to team performance and is especially impor- tant for larger teams. Our study is the first to investigate in‐action TR and provides theo- retical and practical implications on how in‐action TR operates in extreme action teams. KEYWORDS extreme environments, performance, reflection, reflexivity, teamwork, team size 1 | INTRODUCTION Fighting the World Trade Center inferno, rescuing the 33 Chilean miners trapped underground for more than 2 months, or resuscitating a trauma patient after a multicar collision present examples of unusual team endeavors under extreme and dynamic conditions. These extreme environments, characterized by tremendously high levels of velocity, complexity, ambiguity, and consequences, all possess extraordinary physical, psychological, and interpersonal demands and challenges (Klein, Ziegert, Knight, & Xiao, 2006; Manzey & Lorenz, 1998). Due to the dynamic nature of these extreme environments, teams must adapt to changing circumstances and process emerging information in the midst of active task performance to perform effectively. Research suggests that team reflexivity (TR)—defined as a team's conscious reflection on their objectives, strategies, and processes (West, 2000)—is an important team process fostering adaptation and informa- tion processing (Konradt, Otte, Schippers, & Steenfatt, 2015; Schippers, Edmondson, & West, 2014). Although considerable theory and research links TR to adaptive team performance (Konradt et al., 2015; Konradt & Eckardt, 2016; Vashdi, Bamberger, & Erez, 2013), this work suffers from three major limitations especially related to action teams. First, existing team research has conceptualized TR as a transition process (e.g., Schippers, Den Hartog, Koopman, & van Knippenberg, 2008; Schippers, West, & Dawson, 2015), which refers to “periods of downtime where team members can reflect upon past events and prospect future events” (DeChurch & Haas, 2008, p. 544) taking place after completion of taskwork and during episodes of low activity in the team (Fernandez, Kozlowski, Shapiro, & Salas, 2008). These periods of downtime represent debriefings or after‐action reviews (Tannenbaum & Cerasoli, 2013). Due to this focus on TR as an elaborate and time‐ The authors thank Ellen Heimberg and Florian Hoffman and the PAEDSIM net- work (www.paedsim.org) for their collaboration. Also, we thank Laurenz L. Meier for its support with the statistical analysis. Received: 15 February 2017 Revised: 4 April 2018 Accepted: 30 April 2018 DOI: 10.1002/job.2299 J Organ Behav. 2018;39:749–765. Copyright © 2018 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/job 749