As work becomes more decentralized and more informally coordinated, members of organizations increasingly span the boundaries of multiple work teams and organizational units. Building a knowledge base about learning in these arrangements of teams and smaller work units is thus critical, as they constitute the basis for learning throughout the organization. Here, we have highlighted three different perspectives on conceptualizing and measuring team learning. Although these three streams of research involve different theoretical orientations, research methods, and samples, it would be shortsighted to say they represent the full taxonomy of learning teams.
Learning teams are concerned with improving outcomes, coordinating knowledge, and developing effective processes to one degree or another. Further, a team’s focus may shift from one goal to another as the team’s work evolves, integrating mechanisms that have traditionally been the focus of different lines of research. Similarly, team-learning researchers have begun to consider more than one stream in their research. For example, some research in the learning curve stream focused on improving efficiency by building effective group processes (Bunderson and Boumgarden, 2011). Another study combines the group process and task mastery streams by investigating the association between team learning behaviors (and its antecedents) and mutually shared cognition on team effectiveness (Van den Bossche, Gijselaers, Segers, and Kirshner, 2006). Therefore, we suggest that these streams are useful as organizing frameworks for considering the progression of team learning research, and are not meant to be prescriptive divisions for team learning research in the future. Rather, they highlight some of the differences in perspectives, assumptions, and terms that inhibit coherent progress in understanding these important phenomena.
The central argument in this chapter, as discussed at the outset, is that team learning increasingly takes place across teams rather than just within teams, because many organizations have become more complex and less hierarchical. Yet, to date, much of the research on team learning cited here has focused on the learning within focal teams as opposed to collective learning among teams. Scholars have begun to explore the theoretical implications of studying learning across teams (for example, Huckman and Staats, 2009; Woolley, 2009; Nembhard and Edmondson, 2006). Edmondson and Roberto (2003) refer to this type of learning as a team-based learning infrastructure.
Indeed, each stream of research has something to contribute to our understanding of learning across teams: the learning curve stream shows how learning is associated with experience, the task mastery stream outlines cognitive mechanisms for organizing team knowledge and skill, and the group process stream explores the numerous conditions associated with how to learn in teams. In addition, studies within each stream have indirectly examined some of the factors associated with learning across teams.
A few studies in the learning curve stream examined learning and knowledge transfer across teams. In a study of thirty-six stores in a service organization franchise, Darr, Argote, and Epple (1995) found that all the stores improved with experience, but only the stores with the same owner shared ideas and transferred knowledge across stores. In addition, Edmondson and colleagues (2003), in their study of cardiac surgery teams, showed that codified knowledge transferred across teams whereas tacit knowledge was difficult to transmit across teams. These studies reveal some important features of learning across teams such as: (1) it could be important to have an individual or individuals on both the original team and the team to which learning is to be transferred, and (2) teams should attempt to surface and codify tacit knowledge in order to transfer learning.
In the task mastery stream, researchers often performed experiments on teams by training them on tasks together—then reshuffling half the teams to explore what happened when members were required to work with new teammates. In one such study, Lewis, Lange, and Gillis (2005), found that teams performed equally well on the task irrespective of whether they were trained together and kept intact, or reshuffled. This study provided early evidence that team members can transmit certain types of knowledge and experience to new teams. More recently, researchers showed that direct experience yields higher team performance than indirect experience (such as learning by watching another team perform a similar task), further complicating methods of multi-team learning (Gino, Argote, Miron-Spektor, and Todorova, 2010). Another study in this stream investigated how teams leverage knowledge introduced by visitors from other ‘foreign’ teams. Gruenfeld, Martorana, and Fan (2000) found that although the ‘indigenous’ team was likely to use ideas from the foreign team member, they were less likely to surface unique ideas from the indigenous team members. This study suggests that knowledge transfer and learning across teams is difficult due to social processes that determine how team-level memory systems are formed.
Finally, the group process stream of research outlines team learning behaviors related to leveraging learning and knowledge across teams. Building on boundary-spanning research, Ancona and Caldwell (1992) identified three conceptually distinct forms of learning behaviors: experiential team learning, vicarious team learning, and contextual team learning. Both vicarious and contextual team learning refer to behaviors that leverage extra-team knowledge from other teams within the organization and outside of the organization, respectively. Similarly, Wong (2004) delineated the differences between local learning (learning from internal team activities) and distal learning, or learning from ideas, feedback, or help from external parties. When engaged in task mastery, distal learning actually hampered the team’s performance whereas when engaged in innovation, distal learning enhanced team performance. Thus, the nature of the task can influence multi-team learning.
However, studies such as the ones mentioned have not examined learning across teams as the primary focus, but rather as a side-effect of the study design. Team scholars have called for more careful consideration of learning across teams (Mathieu, Maynard, Rapp, and Gilson, 2008). In particular, individual team members serve as conduits of learning through their simultaneous membership of more than one team (O’Leary, Mortensen, and Woolley, 2011). As mentioned above, researchers estimate that knowledge workers are members of more than one team anywhere between sixty-five and ninety-five percent of the time (Martin and Bal, 2006; Zika-Viktorsson et al., 2006). Given the considerable prevalence of multiple team membership, it is critical for organizational learning scholars to understand how MTM can foster the cross-fertilization of learning across teams.
Furthermore, as knowledge and specialization become commodities for employees—above and beyond job training and skills—team members are starting to consult extra-team members with task-relevant knowledge as a form of internal consulting for team learning (Ancona and Bresman, 2007; Bresman, 2010; Edmondson et al., 2001; Nembhard and Edmondson, 2006; Zellmer-Bruhn, 2003). Individual team members are valued more and more for depth of knowledge over breadth of knowledge, increasing the level of specialization of team member knowledge. As such, transactive memory systems are expanding beyond the single team unit to include strategies to leverage knowledge from local experts across teams.
However, as research from each of the three streams shows, there are significant differences between the types of tasks that benefit from the forms of learning studied (for example, learning curve research is focused on repetitive tasks whereas task mastery research examines novel, if well-structured, tasks). As such, the challenge of learning across teams (and of studying learning across teams) is further complicated by the variety of tasks and goals linked to these tasks. In particular, team tasks often reflect different types of learning goals, such as improving team productivity or fostering innovation. The nature of teamwork is correspondingly different for these goals as well. But, what is the relationship between productivity and learning across teams? How do individuals manage different goals as members of different teams? Thus, we will consider how these two linked, but distinct, team goals shape the experience of learning across multiple teams.
Tension Between Productivity and Learning
Learning and productivity are often related (and sometimes conflated), but conceptually distinct and often in tension (Sessa and London, 2006; Singer and Edmondson, 2008; Wilson, Goodman, and Cronin, 2007). For example, Bunderson and Sutcliffe (2003) provide evidence about how team learning can both hurt and help team effectiveness, and Edmondson, Dillon, and Roloff (2007) note how learning and execution are often at odds. The same is true at the individual and organizational levels of analysis. Thus, although there is the potential for a reciprocal relationship between productivity and learning, there are also features of the work environment that can foster one at the expense of the other.
Productivity is generally defined as the quantity of output produced with a given amount of resource (time, personnel, etc.). MTM evolved in some organizations initially in an effort to distribute employees’ time across multiple smaller contracts (Mortensen, Woolley, and O’Leary, 2007) thus increasing organizational productivity. As the practice of MTM intensifies, particularly for knowledge work, individuals tend to deepen their knowledge base in a particular area and become deep subject matter experts whose knowledge is leveraged across an increasing number of teams. Initially this can provide a vehicle for sharing knowledge and skills across organizational units; however, when individuals belong to too many teams at once, this can severely limit the availability of slack resources (in particular, time) needed to schedule meetings and draw benefits from individuals’ knowledge, resulting in the opposite effect. Moreover, the opportunity that teams provide to capitalize on unexpected opportunities for learning is also limited by multiple memberships. As the average number of team memberships increases, the ability of teams to meet or talk synchronously declines, often leading members (or project managers) to seek opportunities to carve projects into smaller parts so that progress can be made by members asynchronously. Consequently, while MTM can initially promote learning by exposing workers to a broader array of problems and encouraging expertise sharing, when left unchecked MTM can inadvertently encourage individuals to greater levels of specialization, and lead teams to create systems with lower levels of interdependence, undermining learning.
In summary, features of an MTM environment that lead to diversity of experience, people, and settings may foster learning at the expense of productivity, while conditions that encourage narrower and deeper individual task specialization, emphasizing efficient practices and reducing slack resources, foster productivity at the expense of learning. This tradeoff is not inevitable, however. For example, trends in medicine that have encouraged deeper specialization and a decline in the ability of collectives to learn have been addressed through interventions that enhance coordination, sometimes practices as simple as keeping surgical teams intact (Edmondson, Bohmer, and Pisano, 2001; Edmondson et al., 2003), conducting a weekly meeting (Gersick, 1989), or using a checklist that prompts conversations (Gawande, 2007; Pronovost et al., 2006). Thus, understanding the features that enhance team learning, coupled with understanding how MTM environments can inadvertently undermine it, can be helpful in developing interventions that can preserve and enhance team learning practices and, thus, the organizational level learning that can result.
Practical Implications and Future Research Directions
Given the fast pace of modern team learning environments, simply keeping up with the pace of change is challenging, and producing high performance necessarily requires continuous learning. Yet, the pace of business in contemporary global economies often demands results in the short term, creating pressures for organizations to sacrifice learning for productivity. Indeed, by their nature, MTM environments may involve a greater emphasis on productivity. Therefore, we suggest that managers should create work climates that foster learning and openness to increase the chance that quality feedback occurs. Studies on psychological safety and team learning have investigated teams in environments where productivity is paramount, such as operating rooms, manufacturing facilities, and research and development teams. In particular, as organizations are increasingly interested in leveraging the skills of highly specialized individual team members across groups, creating teams where team members feel safe to speak up is critical for success.
Team learning researchers must also perform studies in the same complex environment. To capture the dynamics of learning across teams through MTM, researchers should implement both quantitative and qualitative data collection strategies. Quantitative data collection will allow researchers to connect MTM to important outcomes such as performance, productivity, and efficacy—in the learning curve tradition—whereas qualitative data collection will allow researchers to begin to understand the complex inter-group processes and team member experiences, as in the group process tradition. Edmondson and MacManus (2007) suggested that both quantitative and qualitative methods of data collection are appropriate for ‘intermediate theory,’ or theories that draw new connections across potentially unconnected research paradigms by using established constructs in new ways. As such, research on MTM will marry the team and organizational learning literatures with established work in sociological fields such as social network theory.
In terms of data analysis, studying the link between team learning and organizational learning will require the use of multi-level analyses. It will be important to understand the individual’s experience of learning across teams (MTM skills, knowledge, and ability), as they are nested in both team-level phenomena (for example, transactive memory systems) and in their connection to a greater notion of organizational learning (for example, adaptation to change). These models will help researchers to understand how team learning is embedded in the larger context of organizational learning.