14 The Power of the Team Mind

Team decision making includes properties that we might never predict if we study only individuals—for example, the ability of a team to come up with ideas that are beyond the skills of any single team member.¹ This chapter examines the concept of the team mind to explain how teams can think.² The purpose of using this concept is to force us to consider a team as an intelligent entity and to center our attention on the way the team thinks rather than on the individual members.

The Concept of Team Mind

Usually we try to use metaphors based on phenomena that we understand well in order to help make sense of fuzzy phenomena. Computers make good metaphors. So do biological processes. But the concept of mind is a terrible metaphor, and thus a poor candidate to help us study anything. We do not know what it is, or where it is, or even if it is.

And yet the mind of a team can be easier to study than the mind of an individual. Consider the types of information used to make inferences about what is going on in a person’s mind.

First, we infer mental activities from behavior. You can study the behavior of a team as easily as that of an individual. For example, aircrews are emitting behaviors, the behaviors of their airplane. The changes in flaps, the radio calls, and the engine adjustments are all observable behaviors emitted by aircrews. The first officer might change the flaps and the captain might make the radio call, but the behaviors are the actions of the crew.

Second, we infer mental processes from our own conscious experience. We see and feel things, and the contents of awareness fill and occupy our minds. This is murky and hard to bring to the surface when studying individual consciousness. Yet there is no difficulty in studying the consciousness of a team; you just watch and listen. The matters the team talks about, even the gestures that team members make to each other, are the contents of the team’s consciousness. We can call this the collective consciousness of the team.³ An outside observer who is familiar with the task can have as clear an understanding of the collective consciousness as any of the team members.

Third, we infer mental processes from activities of which a person is not aware—perhaps physiological or electrical events in the brain, or part of the person’s unconscious, to be ferreted out by interpreting dreams and Freudian slips. They can be indicated by brief facial gestures or other nonverbal signs. We have trouble making sense of these in trying to understand an individual’s thinking. For the team, it is as easy as interviewing the team members separately to see what they knew that they did not share with the others. We can uncover the ideas that never made it into the collective consciousness. This is the preconscious level of the team mind.⁴

Example 14.1 presents another simulated malfunction from a study we did for NASA.

Example 14.1
The Case of the De-Generators

The B-727 has three generators; it needs at least two generators to fly normally and at least one to fly safely. Each of the three engines on the 727 has its own generator.

A malfunction begins early in the flight when one of the generators fails. This is not unusual or even cause for concern. But as the airplane is making its initial descent prior to landing, the oil pressure on a second engine drops down to 35 psi (pounds per square inch), the borderline value for either operating the engine or shutting it down. The malfunction lets us see how the aircrews would make the decision.

The argument in favor of leaving the engine on is that the more engines that are available, the more power is available. The argument for shutting it down is that the engine can be damaged if used without enough oil. And if the oil pressure decreases more, the crew will have to go through the procedure for shutting it down. If that happens during the landing when everyone is already busy, it will add to the confusion and stress and reduce the margin for error.

One of the aircrews we observed hits on a compromise solution: leaving the troublesome engine on standby so it is not being used but is available in case they need the extra power. The decision is reached only after a lot of discussion, and as an observer, I am not sure of the status of all the engines upon landing. Afterward, when we gather the aircrew for the postflight interview, I remind them of this malfunction and ask them how many generators they had when they landed. “Two,” says the copilot, who was flying the plane when they landed. His expression shows that he is not too confident of that answer.

“One and a half,” says the captain, meaning the one good engine plus the other on standby; the copilot did not realize it was on standby.

“One,” says the flight engineer, reminding the captain that it takes a minute or two to bring an engine back from standby, so the generator was not available when they were landing.

This story seems amusing—that the three have different answers—except that the person with the worst answer was flying the airplane. If he had needed extra power, it would not have been available.

There was disagreement between the contents of the team’s preconscious and its collective consciousness. The team had failed to clarify critical information about the status of the machine and at the conscious level had never realized there were such different interpretations.

The Functions of a Team Mind

A team mind has the following functions:

• Working memory. This is the ability to hold information such as telephone numbers for brief periods of time. Applied to a team mind, information is presented and gets some discussion; then the team moves on to another topic and forgets the first one.
• Long-term memory. This ability to store information permanently applies to teams as well as individuals (Wegner, 1987). Teams have to store information to retrieve later. If a team member has particular information and drops out of the team, the information is lost. According to Wegner, teams find it helpful to store the information redundantly, so more than one person knows it.
• Limited attention. We can attend to only one thing at a time. Teams also can discuss only one thing at a time. They have to be careful in directing their attention to make sure that what gets the spotlight is worthwhile.
• Perceptual filters. We experience the world through the sensory mechanisms that convert mechanical energy into patterns of neural activity. Similarly, teams do not have direct experience but must depend on secondhand reports, which can introduce inaccuracies.
• Learning. Teams need to learn in many ways, such as acquiring new procedures, discarding inefficient behaviors, and figuring out how to become more effective.

There are many other correspondences between a team mind and an individual one, and there are dissimilarities as well, since no metaphor can be an exact match. The value of a metaphor is to help illuminate a phenomenon of interest; the value of the team mind metaphor is to put some structure on our understanding of teams to see how far we can use the findings of cognitive psychology.

How a Team Mind Develops

To learn how a team mind increases in ability, Marvin Thordsen and I have compared some of the teams we had studied. We contrasted the best and the worst teams.

Example 14.2
The Best Teams: Wildland Firefighters

The U.S. Forest Service is responsible for handling forest fires. It uses an incident command structure that is similar to the one used by the military. The incident commander is supported by a staff of specialists in planning, operations, logistics, personnel, and so forth.

Marvin Thordsen was on location during a forest fire in Idaho, a large one that covered six mountains. He watched the command staff assemble a team of 4,000 firefighters, drawing them from all over the country. They put together a working organization in only a few days and sent them out to fight the fire. It is hard to manage an intact organization of 4,000 people, to give directions and make policies, even in stable and safe bureaucratic settings. Here, in less than a week, they were building that organization and trusting it enough to risk lives.

Why are they so good?

Experience. They get lots of experience. There are always fires to be fought. We read only about the dramatic ones, but Idaho alone can count on many fires each season. Unlike the military, which does not fight many wars and has to rely on training to keep sharp, the people fighting forest fires have plenty of first-hand experience. Furthermore, they are fighting an adversary that does not change tactics or add new weapons, so the experience gained one year applies the next. The members of the top command and control echelons have decades of experience.

Sharpness. Some members of the strike team fight forest fires twelve months a year. They work the western states during the summer and fall, then travel to Australia and New Zealand to catch the summer and fall fire seasons down there. The rest of the team is engaged in firefighting in the United States for six months each year.

Stability. The members of each core incident command team work side by side every fire season, for decades. This stands in contrast to a military system, where soldiers rotate assignments every few years.

Promotion from within. Everyone starts at the bottom. Everyone has moved up through the ranks. The lowest crew members realize that their leaders have been in their boots and know what it will feel like to carry out the orders they give. This understanding adds to confidence, since the line crews appreciate that their leaders are more skilled and competent than they are.

Networks. This is a closely knit community. Many commanders and their staff members have worked together before—even the ones from different states who may have joined forces on large-scale fires in the past.

Together, these factors resulted in a calm and competent team. Despite the complexity of assembling a large force overnight, the commanders and planners had done it before, even if on a smaller scale. The different team members knew their assignments, and actions were taken smoothly and purposefully. There was not much wasted energy. The tough decisions they had to face were ones they had faced in the past. For example, a common decision was where to set up a firebreak—a bulldozed semicircle in advance of a fire, designed to stop a fire by depriving it of fuel. The temptation is to draw this tightly around the leading edge of the fire because the smaller the firebreak, the faster it can be built. The disadvantage is that the smaller it is, the greater the risk; if the wind shifts, the fire can slide by, and the team has wasted its time. These kinds of decisions are made quickly, using fragmentary information about terrain, weather, characteristics of the wood, extent of underbrush, competence of the bulldozer operators, and so forth.

The command staff met twice a day to make the difficult decisions. After years of working together, the team members knew how to plan together. They did not waste time on politeness, and their egos were strong enough to take criticism without bristling. They were also sensitive to issues of morale. Someone who disagreed with the commander’s action would confront the commander in the meeting only if it was necessary. Otherwise, the disagreement would be expressed in private. They did not want to waste staff time on lower-priority fights or create a feeling of divisiveness.

There was another characteristic that helped their teamwork: their expectation that they would be working together for years to come. In the midst of fighting fires, they were also conducting on-the-job training. They were aware of opportunities for providing new challenges to different people to help them grow in skill. Rather than remove someone doing a poor job, they tried to find an alternate responsibility from which the person could learn. This was part of their culture: ensuring continuity by developing the skills of their successors.⁵

Compare these teams to the worst ones we have observed: crisis management teams. Typically a manufacturing company handling dangerous materials will set up a crisis management team to prepare for emergencies—fires, explosions, and even terrorist attacks. Marvin and I observed several of these teams established by companies manufacturing nuclear weapons (these efforts were sponsored by the Department of Energy). We observed these teams during training exercises, and we were surprised by their incompetence.

One reason that crisis management teams fall apart is that the heads of these teams have to be the corporate executives, because they are the ones who are legally responsible. But these executives have no background in crisis management. They might go through a one- to two-day crisis training exercise a few times a year. During the exercise, they are still learning their basic tasks. The companies do have a director of security, but that person often has low status because the function is a drain on profits. Besides, there are few emergencies. In some industries, the security officer knows that the chance of a single emergency during a twenty-year career is less than 50 percent.

Example 14.3
The Reaction Time of a Crisis Management Team

The training scenario is that some right-wing terrorists have infiltrated a plant that manufactures nuclear materials and are holding some secretaries hostage. The crisis manager is worried that this is a diversion as other terrorists break into sector H, where nuclear materials are stored. Therefore he tells his director of security to make sure the number of guards there is doubled. Later, the crisis management team receives word that the terrorists have just invaded sector H. The crisis manager is furious. He complains that the training exercise is unrealistic. He has just given the order to increase security, and it had no effect.

As I review my notes on the exercise, I see that the crisis manager issued his order thirty-one minutes before receiving the message that sector H had been invaded. I mentally review the chain of events. It begins with the manager issuing the order. Relaying the order to the people at sector H takes maybe ten minutes. The security director has to get through on the busy telephones. The guard on duty has to locate the unit chief and get him to the telephone. The order is communicated, but they may have other things to talk about; for example, the sector H security chief will want a quick briefing on the hostage situation. So it is a minimum of ten minutes before the sector H security chief starts to carry out the order. And how long does that take? He cannot double his force by asking the people to clone themselves. He has to make more telephone calls to line up the extra staff, determine where the low-priority areas are to draw from, call people at their homes. Then he has to arrange the transportation to bring everyone over to sector H. He might even need route planning if he is worried about ambushes by terrorists. It might take two hours to double the staff, and probably more. Finally the terrorists attack sector H. This is a dangerous situation, so only when there is a free moment will anyone pick up the telephone to call headquarters. Now add another five to ten minutes from the start of the terrorist attack in sector H until the phone call is made. What does that add up to? At least a few hours.

Yet the crisis manager is amazed that his orders hadn’t been obeyed thirty-one minutes later. He complains that it is a flaw in the exercise. He clearly does not know how long it takes to carry out an order such as the one he gave. He doesn’t know the reaction time of his own team.

This incident is reminiscent of an infant who reaches for a slowly bouncing ball. The infant reaches for the ball, but the reaction time is so slow that by the time the child gets to where the ball was, the ball has moved past. So the infant reaches again, aiming directly at the point where he or she sees the ball, again not taking reaction time into account. And again the ball has moved before the infant’s hand gets there. This is challenging to infants. It is amusing to their parents. This is not a game of catch but a different game, a precursor that enables the infant to learn his or her own reaction time, to learn how long it takes to get that hand out.

The crisis management team also did not know its own reaction time. Consequently, it was trying to control events inside the time horizon for responding.

We saw other problems with reaction time in this crisis management team. They tried to micromanage events in sector H, relying on photographs showing the terrorists. Again, they ignored the time lag in taking and sending the photographs, and the time lag in deliberating about actions and issuing orders. The team was trying to direct the security forces in sector H using photographs rather than leaving these decisions to the people on the spot. These problems compounded. Before long the crisis management team had done such a poor job that the exercise controller had to replace the crisis manager to prevent the team from becoming totally demoralized.

The Development of Team Decision Making

Building on the metaphor of a team mind, Klein, Zsambok, and Thordsen (1993) identified four features of child development to map onto teams: the development of competencies, identity, cognitive skills, and metacognition. Figure 14.1 presents these four features. It also provides a set of questions that we have found useful in appraising different teams for the degree to which they have matured along these four dimensions. We have found it useful to apply this framework in assessing teams.

Figure 14.1 Advanced team decision-making model

Team Competencies

Newborns have few competencies beyond their reflexes. In time, they learn to use their arms and fingers, and all the rest, automatically. They learn how to grasp objects and push other objects away and scoot back in a car seat to let their parents strap them in. Teams too are limited by the competencies of their members. Any evaluation of what to expect from a team must consider the individual levels of skill, particularly if the team members keep changing. Everyone is trying to gauge the level of competency of the other members.

Teams also rely on shared practices and routines. Mature teams are supposed to be automatic in performing their basic procedures.

Team Identity

A newborn does not know where it ends and the rest of the world begins. It does not know it has arms. It sees finger-shaped objects drift in front of its face and does not know that it can take direct control of them. Similarly, the immature teams we have watched lack a sense of what they control. The team members themselves are still learning how to do their individual jobs. The advanced teams have already worked this out. The team members think of the team’s requirements as much as their own.

If you try to explain to people on an immature team what the overall goals are, they are likely to respond, “Don’t bother me with that. Just tell me what you want me to do and let me do my job.” In contrast, the members of experienced teams want to know as much as possible about the overall status of the team. They realize that they may have to compensate for others, ask for help, or pitch in for team goals even if they have to abandon their own tasks temporarily. Experienced teams have integrated identities; the members identify themselves in relationship to the whole team. Inexperienced teams have fragmentary identities and focus on individual assignments more than team requirements.⁶

Identity develops slowly. The team members have to learn their own jobs. Then they have to understand something about the jobs their teammates are doing. Then they have to develop more automatic ways of coordinating and working together. Finally, when they have the basics down, they can free up attention to see the challenges facing the team as a whole. Even when the team members are experienced and have worked on other teams, it still takes time to learn everyone’s characteristics. Team identity continues to grow for a long time.

When teams have not developed sufficient identity, they are confused about roles and functions—about who is responsible for what. Some team members tune out, letting others carry the load. Some may be insensitive to the needs of others, as when one person struggles, with no help from others. In an exercise with a crisis management team, the security director and two assistants were greatly overloaded, trying to make phone calls all over the plant. At the next table, the public relations unit sat without anything to do for several hours until someone realized they could help with the security calls.

When teams have not developed sufficient identity, their leaders can be diverted by low-level tasks without understanding what can happen when they let their responsibilities go. For an illustration of micromanagement, we again turn to the crisis management teams.

Example 14.4
The Crisis Manager Who Became Less Than a Secretary

The company had just installed a new center filled with computers, command desks, and the other paraphernalia. They were conducting their first training exercise in the facility. The crisis manager sat at the head table with four staff members in charge of security, operations, and the like. In the middle of the exercise, he got an idea that he wanted to put into action immediately and dramatically. He made his own phone calls to arrange for support helicopters from a nearby U.S. Air Force base. He didn’t tell anyone what he was doing. Instead, he left the head table and went over to where a secretary was sitting and entering messages into the computer system, to notify everyone of key events. He wrote out his own message describing the helicopters and handed it to the secretary. Then he stood behind her, proofreading to catch her typographical mistakes.

The crisis manager wasn’t at his post, making himself available to members of his team who needed help. He wasn’t even a secretary, entering his own messages. He was a secretary’s assistant for the length of time it took to type and revise his message.

Experienced teams have seen instances where roles and functions have broken down, and they know the consequences. One fireground commander working in an urban setting told us about a milk box he brought with him to fires. He explained that when he first became a commander, he was prone to rushing about to offer help, dashing into buildings to assist with rescues, and being an all-around resource. Of course, whenever he left his post, his crew members did not know where to find him if they needed a decision made, and they wasted time searching for him.

Gradually it dawned on him what it meant to be in charge and what his responsibilities were. He knew about his impulsive nature, so he made it a practice to bring the milk box, to serve as his station, and there he stayed. He sat on it, stood on it, kept a foot touching it, remained within sight of it. He had realized the functions he served for his team and where he fit into the jobs of everyone else.

Team Cognition

A few primary questions can be used to distinguish teams with high and low conceptual levels. The first is, How do the teams describe their goals and intents? The answer was covered in chapter 13. Second, to what degree does the team share an understanding of the situation? This can include giving team members a chance to voice divergent views, as well as pulling these ideas together so the members know the assumptions guiding their actions. Some teams take extra steps to keep the members informed as the situation changes.

A third category is the time horizon: How much effort does a team make to look ahead and anticipate problems? Some teams focus on events too close in for them to affect. Example 14.3, on the reaction time of a crisis management team, shows what happens to those who follow events rather than prepare for them.

Fourth, how does the team manage uncertainty, keeping track of gaps and ambiguities? Because the windows of opportunity will close if a team waits to get every last bit of information, good teams must live with uncertainty. At the same time, they have to be sensitive to the assumptions and guesses they have been making in case these turn out to be wrong.

Thus far we have been concentrating on planning teams. When the plans are executed, the team’s priorities shift. Now the team needs everyone to have the same overriding goals and to make sure members share the same understanding of the situation.

Recognizing a situation is important, and teams have to work to communicate this recognition, to achieve a shared understanding. Situation awareness incorporates the nature of the goals, and teams have to work to communicate intent. It also highlights critical cues. Teams have to ensure that the right members are searching for these cues and interpreting and communicating them. They also have to make sure some members are looking ahead while others are carrying out the plans. Finally, teams must manage uncertainty and may have to reconcile opposing viewpoints; each member may be certain about what is happening but disagree with other members.

Team Metacognition

Metacognition refers to the concept of thinking about thinking. It emerged from research with children to describe how they learn to take their own thinking strategies into account. They learn the limits of their memory and acquire strategies for working around these limits, such as knowing when to reread something because they are not sure they understood it. Children cannot develop good skills for metacognition until their behaviors become sufficiently stable and predictable for them to anticipate what will happen and take the necessary steps. Metacognition depends on a clear identity.

For teams, as well as individuals, metacognition is an important development. A team can learn its own capabilities and acquire a strong identity and a shared understanding of the situation. Once this awareness is in place, the team can monitor its own performance and select strategies to avoid weaknesses and capitalize on strengths.

One of the pleasures of watching a child grow up is to see the child learn to handle ideas and juggle different concepts. A critical aspect of a child’s cognitive development is to manage the flow of ideas.

The same holds for teams. The ability to manage the flow of ideas is one of the central skills that distinguishes immature from experienced teams. Members of an immature team may struggle to come up with any ideas, and often the ideas take the team in all sorts of directions—some useful, some irrelevant. Time winds down, and too quickly the team has to make sense of everything that was said. The team members are so excited to tell everything they know that they do not pay attention to whether their comments fit into their task. Experienced teams are more careful; they try to build on the comments of others and create linkages all along. They bring up new perspectives where appropriate, but are cautious about going off on tangents. During a team decision-making session, the ideas have to be managed. Too few, and the product is disappointing. Too many, and the team is overwhelmed.

This is the power of the team mind: to create new and unexpected solutions, options, and interpretations, drawing on the experience of all the team members to generate products that are beyond the capabilities of any of the individuals.

What makes the process so difficult is that no one knows in advance which ideas will be put forth and how they will coalesce. No one, not even the leader, knows what is in everyone’s head. No one can be sure of which suggestions or experiences team members will offer. The team’s ideas depend on who shows up that day, who is feeling alert and who is sleepy, who is prepared and who is distracted, which ideas any of the team members articulate, the order of these ideas, and the ways the ideas are combined. Any session could go in different directions if different ideas are voiced, or are joined in different ways. The team can try to manage the flow of ideas, to keep them coherent and linked. But even a good planning team would have some difficulty managing the flow of ideas, since no one knows in advance what is going to come out.

It is not enough to be aware of strengths and weaknesses. A team has to change its approach if it sees itself running into difficulties. Consider the following example, from an interview of the incident commander of a very large forest fire in northern California.

Example 14.5
The Firefighters Who Realized They Weren’t Supposed to Fight Fires

The fire, at an order of magnitude greater than anyone had ever seen, is entirely out of control. The commanders pull in all kinds of fire crews, outfitting them and sending them to different parts of the state. Yet the news keeps coming back that they aren’t making very much progress.

When the command team meets to figure out what is going wrong, they realize their problem is that they are fighting fires, yet their job is to put fires out. And they simply aren’t putting any fires out.

They decide to stop fighting every fire in the state. They list all the fires, and select the one that will be easiest to put out with available resources. Then they move to the next easiest fire, and so on. In this way, they can send their crews where they will have the most impact. For the first time, they start to put fires out.

Their shift in strategy isn’t easy. The hardest part is to let some fires go. The crews have been working hard to keep these fires checked. Now they are told that the Forest Service is going to let those fires rage uncontrolled, with the crews transferred elsewhere. It feels like a betrayal. Friendships are broken, some permanently.

A few fires are left burning through the fall, into winter. Some fires, like the one at Klamath, have gotten so hot that when spring arrives, they spontaneously start up again. This time the crews can go after them.

Unlike these wildland firefighters, most teams have trouble correcting their problems. Teams can sometimes go too far in the opposite direction, tying themselves up in an attempt to overcontrol. To illustrate a team that did not know how to manage itself, we turn again to our favorite target, the crisis management teams.

Example 14.6
The Crisis Management Team That Loved to Caucus

The crisis manager has obviously been on teams where things have gotten out of hand and is determined to keep a tight rein. He informs the team that they will caucus regularly so everyone can stay informed. His strategy sounds reasonable—except that he doesn’t know what he is doing.

During the first three hours of the scenario, as the crisis swirls around, the crisis manager stands up and loudly announces, “Caucus,” interrupting everyone at work. He calls for a caucus every nine minutes, on the average. (I am timing him.) To make matters worse, he lets the caucus ramble on, and then trail off, and usually doesn’t announce it is over. People just start to go back to their work.

The unit with the toughest job is the security force. They are constantly on the telephones, trying, with limited success, to talk with the people they want to reach. Just as they are making contact, the leader calls for a caucus, and they have to hang up. Then the leader asks if they have spoken with so-and-so, and they admit they haven’t, so the crisis manager berates them and calls more caucuses to check on their progress.

Sometimes leaders become frustrated by the job of managing the team’s progress, diagnosing its problems, and taking corrective actions. We have seen leaders disengage in many ways.

Example 14.7
The Leader Who Vanished

The job of the team is to develop a plan over the two and a half days scheduled for the work. The problem is that two or three members insist on dominating the discussion, and the other members are ignored. Worse, the discussions are so chaotic that some members simply leave the room and wander over to another team meeting in parallel. The team leader is frustrated by his inability to create an orderly atmosphere, yet each time the free-for-all starts, he is one of the people fighting hardest to give his own speeches.

On the last day, we are scheduled to work in the morning and give our briefing after lunch. The morning session is more of the same. Around 10:00 we take a break, then reassemble for the final discussion. We wait for the leader, but he doesn’t come back. One hypothesis is that he has gone up to his room to take a nap. We call, but he has already checked out. Another hypothesis is that he has taken a walk outside the hotel and has gotten mugged. If that has happened, his luggage will still be in the hotel. We check, but his luggage isn’t there. A third hypothesis is that he has been murdered by someone on the team who cannot face yet another session together. This one has some plausibility; the motive is believable, and a murderer might be smart enough to hide the leader’s luggage. The fourth hypothesis is that he is fed up and has flown back to California without telling anyone. I call him the following week, and that was the case.

This was a situation where metacognition was needed to change the way the team was doing business, but the leader did not know how to provide the metacognition. The continual interruptions and posturing that made it so hard to work together also prevented us from finding a better way to structure our discussions.

In Figure 14.1, the four features are linked by a set of arrows. One of the central functions of metacognition is to detect strains in the other aspects of teamwork (competencies, identity, and cognition) and make adjustments. Competencies are linked to team identity because until the team develops to a certain level of proficiency, it may have trouble getting the members to identify themselves with the output of the entire team; as the team develops a stronger identity, the team members will monitor each other and help each other out in ways that improve coordination of basic tasks. Competencies are linked to team cognition because they serve as a limiting factor; as long as the team members are still struggling with basic procedures, they will not be able to pay sufficient attention to larger issues such as situation awareness. As they develop a better sense for time horizons and managing uncertainty, they will find that performance of basic procedures becomes smoother. And team identity is linked to team cognition because a stronger identity helps a team determine if their goals are congruent and if they have a shared situation awareness. As the cognitive processing of the team matures, the individuals gain a stronger sense of identity.

We can use these ideas about a team mind to examine one of the mysteries of the USS Vincennes shootdown, described in chapter 6. The captain was monitoring the battle with the surface ships while the Airbus was approaching. At a critical point, the captain asked, “What is 4474 doing?” referring to the original track number used to designate the Airbus, forgetting that it had been changed to 4131. Some crew members punched “4474” into their key pad and found that the airplane currently using that track number was descending. Others used the current number, 4131, and found the airplane was ascending. In the confusion and ambiguity that followed, the captain made the decision to shoot down the unknown airplane.

Something did not happen during this sequence: when the captain asked, “What is 4474 doing?” no one corrected him. Some crew members knew that the track number had been changed but did not remind the captain. If they had, the decision might have been different. As a result, the inconsistent use of track numbers stayed at the preconscious level, where it was inaccessible to scrutiny. There were so many causes contributing to the accident that we cannot single out one and say this was where the team went wrong. But it certainly did not help to have an uncorrected mistake that led half the team to claim the airplane was descending.

There are several possible reasons that no one corrected the captain. Maybe no one realized the mistake. Maybe the crew members who remembered the change did not understand the implications of letting the error stand. Most likely, some did notice the error but adhered to U.S. Navy culture of not correcting a superior officer. Other naval accidents have been documented where crew members failed to inform a captain of an error he was making.

The purpose of the concept of a team mind is to help us see the team and not be distracted by the individuals. A team can have a poor leader, but if others compensate, the job will get done. Or the team may have brilliant individuals who get so caught up in debate that the job does not get done. If we are evaluating a team, we can try to ask how mature it is. If the team was a child, how developed would that child be? Is he or she still clumsy at basic procedures, or has it gotten them down pat? Is the child able to think effectively, or does it get stumped or even overloaded with ideas? Is the child aware of self, or does it have no idea about the way he or she is working? By looking at the team as a thinking organism, we can get a different appreciation of its abilities.

The Chaotic Nature of the Team Mind

The team mind often functions in an “accidental” manner. That is, teams do not reliably follow one idea with its successor, and action-directed teams do not regularly have a clear idea of the reasons for their actions. The word chaotic refers to the observation that teams usually do not think in systematic and predictable ways. Their ideas and flow of attention are unpredictable, they have a delusion of controlling their own thoughts and their own actions, and sometimes the ideas control the team.

Unpredictable Ideas

Unless the leaders of a team deliberately stage a meeting, it is virtually impossible to figure out in advance what ideas are going to be brought forward. What gets entered into collective consciousness is only a small part of what all the team members are thinking about. There are many good ideas that never get spoken—and many good ideas that could be combined into real breakthroughs. None of the team members can know what the others have not said, so the team is unaware of what it misses.

Unpredictable Flow of Attention

In most operational settings, teams are exposed to all sorts of interruptions and distractions. In his observations at Fort Hood, Marvin Thordsen made a transcript of the five-hour planning session, then counted how many times one element, one thought, was followed by a related one. Of the sixty-four segment transitions he identified, only five (8 percent) were categorized as natural transitions. Another twenty-six transitions could be seen as having been within the context of the discussion segment. More than half of the transitions (thirty-three) were unconnected. The most common category, out-of-context questions, accounted for nineteen transitions (30 percent); the planning became diverted to new, unrelated topics for a variety of irrelevant reasons. Someone would receive a phone call, or a messenger would tell them they needed to use transparencies for their briefing and they turned to a discussion of where they were going to make the transparencies. More important, Marvin found that once the interruption was dealt with, the team usually did not return to its original discussion. It moved on to a different topic. The flow of the discussion was driven by random associations people brought up, not by an agenda.

The Delusion of Controlling Their Own Thoughts

After the fact, team members may describe their interactions as consistent and well directed. Once they know what approaches they arrived at, they can trace how those approaches evolved and make the story seem tidy. However, observers watching them struggle to find any approach, decision, or course of action know about the blind alleys, the stumbling, and the confusion.

Interruptions

Even when interruptions do not come from the outside, the team makes its own distractions. One person has just read an interesting article and wants the others to look at it; another person wants to know how to fill out the new time sheets.

The process resembles free association. There is a general expectation that one person’s comments will be marginally related to the preceding comments. In the short run, members find or make up connections between their comments and the preceding ones; however, these connections can be varied and tenuous. The members will often wonder how they got on the topic. In immature teams, even an agenda may not be sufficient to keep the discussion on track.

The Delusion of Controlling Their Actions

Often in our observations of a team, a meeting would be ending and someone, usually the leader, would say something like, “Okay. I think we’ve talked that over long enough. Now let’s go out and do it.” These comments struck us as odd in cases where no one had defined what the team was supposed to do. Nevertheless, the team members were satisfied to pack up their pads and materials and prepare to swing into action. When asked later, they had trouble describing their goals. Once the action started, the team members took their cues from what others were doing, adjusting and coordinating on the spot. Afterward, when the team leaders see what they have achieved, they can try to weave an explanation showing how their activities were aimed in this direction all the time.

Research in neurophysiology has shown that individuals can have the delusion that they are controlling their own thinking when this is not the case. To illustrate this delusion of controlling the actions taken, consider the following example.

Example 14.8
The Illusion of Rationality

The neuropsychologist Michael Gazzaniga (1985) conducted a study of patients whose uncontrollable epilepsy was so severe that the treatment was to cut the connections between the two hemispheres of the brain so the epileptic seizures would not have a path for spreading throughout the brain.

Working with these patients individually, Gazzaniga presented a written message to the visual field that fed into the right hemisphere, asking the patient to perform some simple act, like standing up and walking around. Next, he presented a written message to the left hemisphere, which controls speech, asking the person why he or she was walking around. Invariably, the person could make up an answer, such as “I just felt like stretching my legs,” or “I felt thirsty and wanted to find a Coke.” The left hemisphere could not know the true reason for the action, but it showed no hesitation in making up a plausible reason. One speculation is that we have some sort of rational engine whose job it is to observe our behaviors and make inferences about rational causes.

If individuals can form a delusion of rational control over their actions, we should not be surprised to find teams doing the same thing.

Ideas That Control the Team

Skilled rowers refer to a phenomenon they call swing, in which all four or eight rowers catch the water at the same instant, and it feels as if the boat has gone flying out of the water. The rowers stop worrying about their individual actions and try to synchronize their movements, to gain the power of coherent focus like light waves that become lasers when they are brought into coherence.

During a team meeting when individuals are waiting for an opening to speak and preparing what they will say, it sometimes happens that an idea is articulated that captures everyone’s attention and refocuses the discussion. We can say that the idea captured the team. It brought the thoughts of the team into coherence. This usually does not last very long, and in most meetings it does not happen at all. When it does occur, it offers a glimpse of the team mind.

The Team Mind as a Metaphor for Thinking

I have been using the concept of an individual’s mind to understand team decision making. Now I want to use what we have learned about teams to make more sense of our own individual minds, with team interactions as a metaphor for thinking.

The next time you are in a group or take part in a team discussion, consider the possibility that you may be seeing the way your own mind works from the inside. The chaos, the accidents, the inhibited thoughts, the chance connections, the serendipity: that is what is going on inside your own head. You do not realize it because we cannot bring to consciousness all the fluctuations in our brains. We cannot become aware of the thoughts we suppress. Our thinking usually looks so orderly, so purposive, so clean. Watching a team think is perhaps the closest we will get to being inside a mind.

Each of the “accidental” characteristics of team cognition listed above also seems to describe individual minds, resembling features of our own thinking. Now we have a way of appreciating how accidental the formation of new ideas is. We can be amused at our delusions that we always think clearly, controlling the flow of our thoughts and rationally controlling our actions. We can understand how thoughts can be generated and actions taken without any general awareness of the dynamics of a situation. We can know how different lines of thought can go on in silence and in parallel until one is ready to enter into consciousness. We can see how the other line of thought never gains awareness.

Applications

The impetus for studying teams, and for the team mind concept, was to develop ideas for training. We found an interesting disconnect about training in many organizations we studied, particularly in the Department of Defense. Some high-ranking officers insisted that they already offered team training, while others admitted that this was one of their weak points. In trying to figure out the reason for the disconnect, we found it was based on an ignorance of what team training was.

Team training requires a certain process:

1. Identify a set of functions and processes that teams should master, within a given setting (e.g., how to communicate intent or compensate to help each other out).
2. Evaluate how well the teams in this setting perform those functions and processes.
3. Identify any areas of weakness.
4. Provide specific training in the form of exercises tailored to provide experience or remedy weaknesses.

The officers who were unhappy with team training knew that this type of training never occurred. Yet no one tried to find out why teams went wrong, and then design new scenarios to build the necessary skills.

The officers who were satisfied that they trained teams meant that they trained people in situations where they had to work as teams. If a team fell apart, the observers might comment on the poor outcome but not on what went wrong inside the team. These officers did not know they could do more by examining a team’s strategies or its competence. They did not know what they did not know, so they were unaware that they were missing something.

Once at the Army War College I was talking about how the school could adapt its exercises to provide more focused team training. One instructor, a major, took exception to my comments. “How can you claim that we do not already know how to train teams?” he demanded. I replied, then finished my presentation. At the end, he still did not want to make any changes, but for a different reason. “What makes you think we can evaluate these team processes?” he demanded. “We need a lot more preparation than we’ve been given in this school.”

In working in different settings, we realized that lots of training exercises already went on involving teams. It would be easy to add some additional material and make these into directed team training opportunities. We could use the existing exercises for team decision training as well as for familiarity with the tasks themselves.

One of the projects, run by Caroline Zsambok, was to provide team decision training. (This project was sponsored by the Army Research Institute.) Our contract monitor, Owen Jacobs, wanted us to apply the ideas of the team mind to train high-ranking officers, and we worked out arrangements with the Industrial College of the Armed Forces, part of the National Defense University in Washington, D.C. The school trains officers at the rank of lieutenant colonel (U.S. Army or Air Force) or captain (U.S. Navy). Caroline produced a module on advanced team decision making that was adopted by the Industrial College of the Armed Forces. Under her leadership (Zsambok, 1993), we transformed the initial ideas about a team mind into a set of primary dimensions and key behaviors along the lines of the model shown in figure 14.1. This advanced team decision making model has become part of the curriculum of the college’s course on strategic decision making. The faculty wanted its students to learn how to observe teams in action and make adjustments on the spot. It also wanted the students to anticipate difficulties and make the necessary changes and preparations in advance.

We have used this team decision-making model in other domains, such as emergency response organizations set up by nuclear power plants. The Nuclear Regulatory Commission (NRC) requires each nuclear power plant to set up an emergency response organization that it evaluates periodically. Members of the emergency response organization have primary jobs in the plant and usually come together only for drills.

In 1995 Duke Power Company hired Klein Associates to study the team decision making in the emergency response organization of one of its nuclear power generating stations. Dave Klinger led the project, which included an independent consultant, Doug Harrington, of Team Formation. They used the advanced team decision-making model as a diagnostic tool. They observed drills and interviewed several people in each of the key positions. (Because the emergency response organization must always be on call to activate, several people fill each position.)

The primary difficulties were that the crews were not clear about their roles and functions, and they were having trouble maintaining a shared situation awareness. Dave and Doug worked with the teams to redefine the roles and functions and to redesign the workspace layout in the emergency response organization. Another change was to conduct after-action reviews to cover team decision making. During the project, more than fifty changes were recommended in organizational structure, processes, and the physical environment of the emergency center.

Over the ten months that this project took, Dave and Doug drastically reduced the number of people in the emergency center. At first, there were more than eighty people crowded into one room. Because the workload was so heavy, Duke Power believed they needed to add even more people, but they weren’t sure how to fit them in. Dave and Doug found that the heavy workload was caused, in part, by having too many unnecessary people. They tried cutting out assistants and nonrelevant staff, and performance got better. By the end of the project, the staff was down to thirty-five, and workload had reduced, not increased.

In addition to staff reduction, the station manager and emergency planning director decided to put many of the major recommendations in place shortly before their annual drill with NRC observers—without practice. We were nervous about this. But the previous history of difficulties had raised the possibility that the plant would have to increase the number of practice drills from four a year to six a year (each drill can cost between $250,000 and $500,000) and go from one to two formally evaluated drills a year (each of these can cost between $500,000 and $1 million). The plant personnel were highly motivated to do well.

They decided to institute a new room layout. They placed individuals who must share information next to each other. They moved the status board so that all the major players at the command table could see it. They reorganized the board to show plant status, the status of teams in the plant, the status of equipment, the most recent events, and current priorities. The status board, which had been ignored during previous drills, could now be used during briefings to describe the problems and how they were being addressed. One other major recommendation was the definition of expectations by the team leader. During the drill, but prior to activating the center, the emergency director went around the command table and described to each team member what was expected of him or her and what was not expected. This placed the team in a position to move forward and handle the event without the crippling effects of redundant, ambiguous, and meaningless tasks and information.

Was it worth it? There were several obvious differences. The room was noticeably quieter. There was a shared perception that the problems presented in the drill were easier than in previous drills, whereas in actuality the exercise was more challenging. The improved teamwork just made it seem easier because the team members were no longer getting in their own way. After watching the team in action, the NRC reduced the number of required drills at the plant down to one every two years and wiped the slate clean regarding previous deficiencies. Now the plant was held up as a model of how an emergency response organization should work.

Key Points

• A team is an intelligent entity.
• The cognition of a team can be inferred from three sources: the team’s behavior, the contents of the team’s collective consciousness, and the team’s preconscious.
• The mind of a team shows some familiar features: it has a limited working memory, it has to store some information permanently, it has a limited attention span, it can process information in parallel, and it relies on filtered information.
• The team mind develops basic competencies and routines, forms a clear identity, learns to manage the flow of ideas, and learns to monitor itself to adjust its thinking when necessary.
• The idea of a team mind helps us understand the way individuals think, including the accidental nature of thinking.

Notes

1. Most of the work in this chapter, unless otherwise noted, was funded by the U.S. Army Research Institute for the Behavioral and Social Sciences.

2. Smith (1994) has used a related concept, collective intelligence.

3. I am indebted to Judith Orasanu, now at NASA Ames, for the term collective consciousness to refer to this aspect of team interaction. For a definition of team, see Orasanu and Salas (1993) and Duffy (1993).

4. Herb Colle suggested the term preconscious, using the Freudian framework as an analogue.

5. We have heard suggestions that these factors have been changing in the firefighting community as a result of budgetary pressures, downsizing, political factors affecting promotions, and their influences. The sense of community may not be as strong as it was in the past.

6. Jim Banks suggested that we add the category of team identity to the advanced team decision-making model. Driskell and Salas (1993) have empirically demonstrated the gain in performance that occurs when team members abandon their egocentric attitudes.