The shared mental model is a term used in industrial and occupational psychology. Within the discipline of sport psychology (SP), a shared mental model is commonly referred to as a shared knowledge state. This is a state in which the knowledge held by each member of a sports team about the upcoming actions of the team is at least similar to other team members’ knowledge of these actions. In everyday language, the state can be described as team members being “on the same page” about their actions on the field. This topic is of interest to sport psychologists because the ability of a team to coordinate the actions of its members is thought to depend, in part, on the attainment and maintenance of this state. In this entry, the concept of a shared knowledge state and its role in achieving team coordination is outlined first. The processes by which a shared knowledge state is established prior to a given game, and established and updated during a given game, are then described. Note that there have been few studies to date on team coordination and the role of shared knowledge in achieving team coordination, within the sport domain. As such, while reference to such studies is made in the sections that follow, the content of these sections is largely theoretical.
Shared Knowledge States and Their Role in Achieving Team Coordination
A challenge for teams is to achieve coordination. Team coordination is the process of arranging team members’ actions so that, when they are combined, they are in suitable relation for the most effective result. Team members’ actions must be arranged so they are correctly related on three dimensions of action: type, timing, and location. First, achieving a team action often requires that each team member undertakes a specific type of action. If a point guard in basketball lobs the ball in an alley-oop pass through the air toward her offensive basket, her teammate might prepare to jump to catch the ball in midair and dunk it in the basket. If at the last moment the point guard makes a bounce pass on the ground, her teammate might not have time to adapt to the bounce pass because she has already begun to jump. This change in action type would likely result in an incomplete pass. Second, achieving a team action often requires each team member to undertake an action at a specific time. In the basketball example, if the point guard makes the alley-oop pass too early, her teammate might be unable to jump fast enough to catch the ball and dunk it. Third, achieving a team action often requires each member of the team to undertake an action at a specific location. If the point guard does not throw the alley-oop pass close enough to the rim of the basket, her teammate will not be in position to catch the ball to dunk it.
While the basketball examples are hypothetical, research has provided a real example of the importance of action timing. Studies of multiperson rowing show that rowers’ strokes must be synchronized, requiring that they deliberately coordinate their actions during races. In a recent study of elite-level two-person rowing, Carole Sève and colleagues, a rower (Marion) was shown a film of her and her teammate’s rowing performance and asked to comment on their performance. Within her comments, Marion highlighted a problem of coordinating her strokes with those of her teammate (Lucy), feeling that she (Marion) was being “pushed” by Lucy. This report suggests a problem with action timing; Lucy is stroking too fast for Marion. A subsequent analysis of the rowers’ movements confirmed this: The length of the stroke phase (i.e., the oar in the water) was shown to be shorter for Lucy than Marion, which meant that Marion needed to “catch up” to Lucy during the recovery phase (i.e., the oar out of the water). Having proposed that achieving coordination can be challenging for teams, we now consider the reasons why attaining this outcome can be challenging.
When one individual performs a task, the task being undertaken is controlled by only one executive (i.e., brain). However, when a team attempts a task, there are as many executives as there are team members. Each person possesses unique knowledge about how to perform a given task, so people placed into teams tend to select the type, timing, and location of their actions at their own discretion. The performance of a team suffers when team members select actions at their own discretion because, as proposed previously, team members’ actions must be related in terms of type, timing, and location to actions being undertaken by other team members. Establishing a shared knowledge state, in which team members share (i.e., hold in common) similar knowledge about how to undertake their team task, means that team members can draw on that similar knowledge to undertake this task. As a consequence, team members are able to select the type, timing, and location of their actions in a way that affords effective team coordination.
The Process by Which a Shared Knowledge State Is Established and Updated
In this section, we will distinguish between shared knowledge states established prior to a given game and those established and updated during a given game, as both types of shared knowledge state underpin team coordination. Prior to a given game, teams acquire a shared knowledge state in two ways: through actual play, in both practice and competition settings, and via explicit planning.
Shared knowledge acquired through play is often considered knowledge of “situational probabilities”—that is, knowledge of what the team and its individual members are likely to do in response to a given game situation. This type of shared knowledge can be acquired simply through experience of playing the sport. Players learn from practicing and competing within their sport what teams and individual team members are likely to do in general in a given situation. Consider a newly formed soccer team where, previously, every team member has played the sport but no team member has played on this particular team. As a result of their experience playing the sport, all the players on this team know that when a midfield player accidentally turns the ball over to the opposition, defenders on his or her team are likely to respond to the turnover by moving back toward their own goal into a defending position.
Shared knowledge of situational probabilities also arises from experience of playing alongside members of one’s own team. Players come to know through playing on a specific team what their team and its individual team members are likely to do in a given situation. Consider how team members playing together on the same ice hockey team for several years would learn that the right winger on their team likes to feint with his stick twice before passing the puck across the goalmouth for an assist. Coaches often design practice sessions and drills, such as “2 vs. 1” drills in soccer, as well as scrimmages with the goal of accelerating players’ learning of the situational probabilities related to their team and individual teammates. In other words, coaches place players in specific game (or gamelike) situations involving coordination so they will get to experience, and in turn learn, how those situations “play out.”
Regarding shared knowledge acquired through explicit planning, coaches or other team members often provide the same information about the team’s intended actions to all team members by communicating plans of action (e.g., approaches, strategies, tactics, and plays) to those members. Coaches usually communicate plans initially via verbal communication and often demonstrate plans using visual aids such as a whiteboard. Subsequently, the coach asks the team to practice executing the plan (e.g., running a play) on the field. Ideally, all team members acquire similar knowledge of the plan from the coach’s communications and demonstrations and/or by practicing the execution of the requisite plan.
The result of (a) playing the sport generally, (b) playing in a particular team within the sport, and (c) creating and practicing the execution of team plans is a relatively stable form of shared knowledge state. This state is established prior to a given game and forms a cognitive resource upon which team members can draw during a given game to achieve coordination. Consider an example of a play in football. The play is planned by a coach and then discussed with, and demonstrated using a whiteboard to, players prior to a practice session. Next, the play is practiced on the field, and feedback on it is provided by the coach until players feel comfortable executing it. At this point, each player involved in the play knows (a) when to use the play (or what signal from the coaching staff or teammate denotes that the play should be used), (b) what actions those involved in the play are expected to perform, and (c) how his own actions fit with the actions of his teammates. Consequently, the play can be used during actual games.
In addition to the relatively stable shared knowledge state achieved prior to the game, it is usual that a relatively dynamic version of a shared knowledge state is established and updated by a team during actual games. This dynamic “in game” state is required because games in most team sports are characterized by rapidly unfolding and difficult-to-predict situations within which team members must adapt (e.g., a faked punt by the opposing team in football). Nonetheless, the establishment of this in-game state is often based in part on the shared knowledge state achieved prior to the game. Consider a soccer team that has developed a strategy of attacking up the wings of the field; in other words, the team has acquired a shared knowledge state in which this is the team’s go-to offensive strategy. Of course, the team members would likely also know, and share knowledge of, how to attack up the middle as a team. In other words, although attacking up the middle is not the team’s current go-to strategy, the team has used this strategy and thus knows how to implement it. Then consider that the team begins a game but soon realizes that their opponent is defensively strong on the wings but weaker in the middle. The team would likely switch from attacking up the wings to attacking up the middle. Thus, the in-game shared knowledge state of the team would need to be updated accordingly for the team to achieve the coordination required to effectively execute this new strategy—that is, a strategy that is new to this game, if not to this team.
The updating of this in-game state occurs via incidental and deliberate means. Regarding incidental means, when a given change in the game situation occurs, one or two team members may begin to adapt to this situation by switching to a different strategy. In the previous example, as the soccer team experiences a strong defense against its attacks up the wings, an attacking player (e.g., Emma) may begin to pass the ball to a teammate in the middle of the field rather than dribble it up the outside of the field. When other team members see the change in Emma’s attacking behavior, they infer (on the basis of their experience of attacking up the middle) that Emma is switching from the up-the-wing strategy initially prioritized to an up the-middle strategy. In turn, these team members infer that they are required to integrate their own actions with the up-the-middle strategy to achieve effective coordination.
Team members also often deliberately update the team’s in-game shared knowledge state by communicating intended changes to upcoming actions during games to other team members. They typically communicate this information using bodily movements (e.g., pointing) and verbal communication. In the soccer example that was just given, consider that when Emma realizes that the strategy of attacking up the wings is not working, she might communicate verbally to her teammates that they should attempt to attack up the middle. Even subtle facial expressions can be used for this purpose. In a recent study of table tennis doubles teams, Germain Poizat and colleagues reported that a particular stare by one table tennis player was enough to communicate intended changes to tactics to the other player.
In conclusion, team coordination is thought to depend in part on the team’s ability to establish a state in which knowledge of the team’s intended actions is shared by team members before a given game and establish and update an “in-game” version of this state during actual games. Establishing a shared knowledge state before a game is achieved by (a) general experience of practicing and competing in the sport, (b) experience of practicing and competing in a particular team, and (c) communication of team plans by coaches to players and practice executing these plans. Establishing and updating the in-game version of this state occur on the basis of observed changes in teammates’ actions, which signal a change of team plan, and deliberate communication by coaches or team members to other team members about intended changes to the team plan.
- Eccles, D. W. (2010). The coordination of labour in sports teams. International Review of Sport and Exercise Psychology, 3, 154–170.
- Eccles, D. W., & Tenenbaum, G. (2004). Why an expert team is more than a team of experts: A socialcognitive conceptualization of team coordination and communication in sport. Journal of Sport & Exercise Psychology, 26, 542–560.
- Eccles, D. W., & Tran, K. (2012). Getting them on the same page: Strategies for enhancing coordination and communication in sports teams. Journal of Sport Psychology in Action, 3, 1–10.
- Eccles, D. W., & Tran Turner, K. (in press). Coordination in teams. In M. Beauchamp & M. Eys (Eds.), Group dynamics in exercise and sport psychology: Contemporary themes (2nd ed.). London: Routledge.
- Poizat, G., Bourbousson, J., Saury, J., & Sève, C. (2012).Understanding team coordination in doubles table tennis: Joint analysis of first and third-person data. Psychology of Sport and Exercise, 13, 630–639.
- Sève, C., Nordez, A., Poizat, G., & Saury, J. (2013). Performance analysis in sport: Contributions from a joint analysis of athletes’ experience and biomechanical indicators. Scandinavian Journal of Medicine and Science in Sports, 23, 576–584. doi:10.1111/j.1600-0838.2011.01421.x