My boss showed me a list of the members of a project team. On paper, their inclusion made sense: Each person was talented, creative, and experienced. 

But each person -- and I mean this affectionately -- was also basically a jackass. Especially Mike. Great operator, but boy -- and I also mean this affectionately, since he and I got along great -- he was a jerk.

"You know they're going to kill each other," I said.

"No doubt," my boss said. "Left alone, they would be& be a disaster. So I'm going to add Rita to the team and put her in charge."

On paper, that also made sense. Layering in someone like Rita -- empathetic, intuitive, possessor of off-the-charts emotional intelligence -- is a standard team-formation move. Smart leaders try to include at least one bridge-building, collaboration-fostering, interpersonal skills superstar to every team. 

Unfortunately, adding Rita didn't work. They didn't kill each other, but the project went nowhere. 

Time-honored management traditions aside, failure shouldn't have come as a surprise, since it turns out a team's collective intelligence and effectiveness doesn't hinge on its most emotionally intelligent member (like Rita). A team's collective intelligence doesn't even hinge on its average emotional intelligence.

Or as Temple University professor Ravi Kudesia says, "Our collective intelligence is limited by jerks!"

A group's collective intelligence reflects its capacity to perform well across a variety of cognitive tasks and it transcends the individual intelligence of its members. Our empirical results indicate that collectively intelligent groups are those in which the least socially sensitive group member has a rather high score on social sensitivity.

Differently stated, [socially sensitive] group members cannot compensate for the lack of social sensitivity of the other group members.

In our case, Mike was the anchor the group -- much less Rita -- could not raise. Having Mike on the team meant the group would basically operate at Mike's level. 

Why? As the researchers write:

When one of the group members is not able to "tune in" with the rest of the group, this might create coordination losses and eventually hamper the progress of the task. This is also in line with the study that shows the minimum score in personality traits is particularly relevant for tasks with intensive workflow patterns, where work unfolds freely and frequently among all team members.

Such workflow patterns require intense coordination, and while one highly sensitive member cannot compensate for the rest of the group, one highly insensitive member can disrupt the flow of coordination and diminish the level of collective intelligence of the group.

Yep: Jerks rule, and not in a good way.

So what should we have done when we put the team together?

For one thing, left Mike off the team. Brainstorming, problem-solving, idea generation...Mike's interpersonal insensitivity made him a terrible fit for those types of teams. (But that doesn't mean he couldn't be on any team. Mike was great on task-oriented teams -- his no-nonsense attitude tended to cause people to stay more focused and work harder.)

We also shouldn't have included Rita, who was only added in the hope her emotional intelligence would overcome all of the team's interpersonal dynamic shortcomings. Since "one highly sensitive member cannot compensate for the rest of the group," much less the least sensitive member of the group, we wasted her talents. 

Instead, we should have matched the difficulty and importance of the project with the emotional intelligence of its individual team members.

Say you have a tough project, one requiring substantial give-and-take, collaboration, encouragement, and inclusion among team members. Putting together a team of smart people matters, but putting together a team of socially intelligent people matters even more.

Especially since just one weak link can ruin the effectiveness of the entire team. As the researchers write, "Harmonious interpersonal interactions in groups, likely to be conducive for collective intelligence, require that even the least socially sensitive member scores relatively high on social sensitivity."

The next time you put together a team, don't assume adding one highly emotionally intelligent person will offset another team member's rough interpersonal edges. 

Take a look at the least socially sensitive members of the group, and make sure they have the interpersonal skills required for the team to succeed. If they don't?

Leave them off the team.

Because, at least in this case, science says one bad apple really does spoil the whole bunch.