MANAGEMENT SCIENCE Vol. 50, No. 11, November 2004, pp. 1534–1544 issn 0025-1909  eissn 1526-5501  04  5011  1534 inf orms ® doi 10.1287/mnsc.1040.0210 ©2004 INFORMS Effects of Adaptive Behaviors and Shared Mental Models on Control Crew Performance Mary J. Waller A. B. Freeman School of Business, Tulane University, 7 McAlister Drive, New Orleans, Louisiana 70118, mwaller1@tulane.edu Naina Gupta Department of Business Administration, University of Illinois at Urbana–Champaign, 1206 South Sixth Street, Champaign, Illinois 61820, ngupta@uiuc.edu Robert C. Giambatista College of Business Administration, Texas Tech University, P.O. Box 42101, Lubbock, Texas 79409-2101, rgiambatista@ba.ttu.edu C ontrol crews are highly trained teams responsible for monitoring complex systems, performing routine procedures, and quickly responding to nonroutine situations. Previous literature suggests that higher- performing control crews engage in adaptive behavior during high-workload or crisis situations. Other work suggeststhathigher-performingcrewsuseperiodsoflowerworkloadstoprepareforfutureproblems.Tounder- stand which behaviors performed during which situations better differentiate lower- from higher-performing crews, we conducted a study of 14 nuclear power plant control room crews and examined adaptive behaviors and shared mental model development in the crews as they faced monitoring, routine, and nonroutine situa- tions.Ourresultssuggestthatfewdifferencesinadaptivebehaviorsexistbetweenhigher-andlower-performing crews during monitoring or routine situations, but that information collection and shared mental model devel- opment activities, and intracrew processes used during model development, differ significantly between lower- and higher-performing control crews during nonroutine situations. Key words : group dynamics; teams; shared cognition; mental models; nuclear power History : Accepted by Linda Argote, organization performance, design, and strategy; received March 1, 2001. This paper was with the authors 16 months for 2 revisions. Introduction In many high-reliability organizations, overall sys- tems operations are the responsibility of a small crew that controls multiple systems and makes decisions concerning system functioning. These control crews are teams of highly trained individuals who work and train together to monitor system interfaces and keep systems at equilibrium (Waller and Jehn 2000). Control crews are particularly prevalent in aviation, naval operations, petroleum and chemical refining, nuclear power, and other environments in which reli- abilityofperformanceiscrucial.Thesecrewsareoften referred to as the “brains” of the complex systems they manage, because control crews are responsi- ble for interpreting information from multiple sys- temsandmakingaccuratedecisionsduringabnormal, time-pressured, high-workload situations. Existing literature concerning control crews and their organizations suggests two central themes regarding control crew performance. First, because control crews’ actions during abnormal situations are so critical in high-reliability contexts, much of the research on control crews and the training they receive focuses on their actions during abnormal, high-workload situations (see Waller 1999, Weick 1993), and includes the actions taken by crews and organizations to manage and learn from crises (Roberts 1993, Weick and Sutcliffe 2001). Work in this area focuses on the idea that successful crews must “think on their feet and do the ‘right thing’ in novel situations” (Weick and Roberts 1993, p. 358). Second, while crew behaviors during nonroutine situations are clearly critical to crew and organiza- tion performance, other work in this area suggests thatwhatcrewsdoduringlower-workloadlevelssig- nificantly impacts their overall performance. Specifi- cally, this work suggests that higher-performing con- trolcrewsprepareforpossibleabnormalsituationsby planning for such situations during lower-workload periods(HueyandWickens1993,Orasanu1993,Pepi- tone et al. 1988, Stout et al. 1999). Thesetwocentralthemesaboutcontrolcrewactions motivate our research regarding the timing and con- tent of control crew actions. Taken together, these themessuggestthatthepatternofcrewbehaviorsover time and across variable workload conditions might 1534