So, ‘having a bunch of smart people in a group doesn’t necessarily make the group smart’.
This is one of the conclusions from new research on collective intelligence by Carnegie Mellon & MIT (full story below). While this seems counter to what we might expect at a superficial level, on reflection it makes perfect sense. Any of us who’ve been in group ideation or creative processes can attest to the fact that sometimes things just seem to click, other times the group or team just never quite gets up to speed, even if all the smart people have been painstakingly fought for, gathered, briefed, fed Haribo & put to work.
What the research team found is that the emotional intelligence of the group was as important as its raw cognitive power. And that the better group members perceived each other’s emotions, the better the performance of the group (ceteris paribus). They also found that having women in the group made the group more effective as a group (again, ceteris paribus), because of what they suggest is the higher emotional sensitivity displayed by women, on average.
My own extremely amateur efforts in this area support this thesis entirely. At BBH we were working on improving the way we collaborate – especially in more creative phases of projects – for some time. We frequently convened ‘ideas sessions’ when working on the various Google projects that BBH handled. Anything between 10 to 14 people would get together for perhaps three hours in carefully briefed and run creative sessions. We’d try and include creatives and senior people (clients) from the Creative Lab. We’d have a mixture of core Google team, plus interesting and smart people from every discipline. What we found as we progressed through a year of trialling this approach was (a) there was a definite ceiling of around 12-14 attendees before it became ineffective and inefficient; (b) that carefully casting the attendees (to the point of sometimes not including people on the core team) was critical to success (it’s just an unavoidable fact that not everyone’s good at this); and that (c) we got much much better at these sessions as we did more of them, learned to trust each other, learned to relax into them, and learned about each other. I guess that might be called ‘emotional intelligence’, in some way.
Every situation requires a bespoke solution, and some creative teams both prefer to work alone and, indeed, are more effective when working alone. So I would hesitate to prescribe the above approach for all situations. It depends. But if you are in a creative agency or business in which collaborative creativity *is* something you like to utilize, here are three observations – weaving in the research results from Carnegie Mellon / MIT – on how things might be made more effective:
1. CASTING: When casting a team for a collaborative session, the smartest people in the room are not always the best people to have in the room (or, at least, not all of them); many times in agencies over the years I’ve been in group creative or strategy sessions that have imploded because of ‘smarts overload’. While not being selected for a group session can be disappointing, it’s better to disappoint one or two than waste the time of 12 or 14. It’s relatively easy to get to very good. Somewhat harder to get to awesome.
2. TRAINING: Spend time training people to thrive in and enjoy group situations. At BBH we found the Master Classes run by Hyper Island to be especially effective at this. People (me included, I confess) turn up to these classes thinking they were going to get a crash course in digital creativity; in fact, 90% of the focus was on the ‘how’ not the ‘what’. On tearing down the invisible walls that separate silos and permit egos to run amok. Priceless stuff.
3. TOOLS: Explore new online collaboration tools. We used a tool called Kluster a fair amount within the Google team, especially in late 2009, and had started to develop our own bespoke collaboration engine when I left. The advantages here are many: participants can escape the physical (& some of the emotional) confines of a face-to-face ideas session; a more enduring and iterative idea development platform is created (it’s the start of something, not merely a punctuation); the right people can be cast from any part of a business (any division, timezone, office); and the whole endeavor is automatically recorded / archived as you go (so you lose less value).
As I’ve noted before, one of the things Nigel Bogle used to say that resonated with me most during my time at BBH was ‘none of us is as good as all of us’. I think this remains spot on, but needs qualifying. It might be more accurate (if less elegant) to suggest that while none of us is as good as all of us, all of us might be more effective if we let some of us organize all of us properly.
While process surely isn’t as glamorous as hiring big name rockstars, it’s actually even more important. Get the operating system right and the software will perform.
Original article from Carnegie Mellon website is below.
New Study by Carnegie Mellon, MIT and Union College Shows Collective Intelligence
Of Groups Exceeds Cognitive Abilities of Individual Group Members
Groups Demonstrate Distinctive ‘Collective Intelligence’ When Facing Difficult Tasks
PITTSBURGH—When it comes to intelligence, the whole can indeed be greater than the sum of its parts. A new study co-authored by Carnegie Mellon University, MIT and Union College researchers documents the existence of collective intelligence among groups of people who cooperate well, showing that such intelligence extends beyond the cognitive abilities of the groups’ individual members, and that the tendency to cooperate effectively is linked to the number of women in a group.
Many social scientists have long contended that the ability of individuals to fare well on diverse cognitive tasks demonstrates the existence of a measurable level of intelligence in each person. In a study to be published in an upcoming issue of the journal Science, the researchers applied a similar principle to small teams of people. They discovered that groups featuring the right kind of internal dynamics perform well on a wide range of assignments, a finding with potential applications for businesses and other organizations.
“We set out to test the hypothesis that groups, like individuals, have a consistent ability to perform across different kinds of tasks,” says Anita Williams Woolley, the paper’s lead author and an assistant professor at Carnegie Mellon’s Tepper School of Business.
“Our hypothesis was confirmed,” continues Thomas W. Malone, a co-author and Patrick J. McGovern Professor of Management at the MIT Sloan School of Management. “We found that there is a general effectiveness, a group collective intelligence, which predicts a group’s performance in many situations.”
That collective intelligence, the researchers believe, stems from how well the group works together. For instance, groups whose members had higher levels of “social sensitivity” were more collectively intelligent. “Social sensitivity has to do with how well group members perceive each other’s emotions,” says Christopher Chabris, a co-author and assistant professor of psychology at Union College in New York.
“Also, in groups where one person dominated, the group was less collectively intelligent than in groups where the conversational turns were more evenly distributed,” adds Woolley. And teams containing more women demonstrated greater social sensitivity and in turn greater collective intelligence compared to teams containing fewer women.
To arrive at their conclusions, the researchers conducted studies at MIT’s Center for Collective Intelligence and Carnegie Mellon, in which 699 people were placed in groups of two to five. The groups worked together on tasks that ranged from visual puzzles to negotiations, brainstorming, games and complex rule-based design assignments. The researchers concluded that a group’s collective intelligence accounted for about 40 percent of the variation in performance on this wide range of tasks.
Moreover, the researchers found that the performance of groups was not primarily due to the individual abilities of the group’s members. For instance, the average and maximum intelligence of individual group members did not significantly predict the performance of their groups overall.
Only when analyzing the data did the co-authors suspect that the number of women in a group had significant predictive power. “We didn’t design this study to focus on the gender effect,” Malone says. “That was a surprise to us.” However, further analysis revealed that the effect seemed to be explained by the higher social sensitivity exhibited by females, on average. “So having group members with higher social sensitivity is better regardless of whether they are male or female,” Woolley explains.
Malone believes the study applies to many kinds of organizations. “Imagine if you could give a one-hour test to a top management team or a product development team that would allow you to predict how flexibly that group of people would respond to a wide range of problems that might arise,” he says. “That would be a pretty interesting application. We also think it’s possible to improve the intelligence of a group by changing the members of a group, teaching them better ways of interacting or giving them better electronic collaboration tools.”
Woolley and Malone say they and their co-authors “definitely intend to continue research on this topic,” including studies on the ways groups interact online, and they are “considering further studies on the gender question.” Still, they believe their research has already identified a general principle indicating how the whole adds up to something more than the sum of its parts. As Woolley explains, “It really calls into question our whole notion of what intelligence is. What individuals can do all by themselves is becoming less important; what matters more is what they can do with others and by using technology.”
“Having a bunch of smart people in a group doesn’t necessarily make the group smart,” concludes Malone.
In addition to Woolley, Malone and Chabris, the other co-authors were Alexander Pentland, the Toshiba Professor of Media Arts & Science at the MIT Media Lab; and Nada Hashmi, a doctoral candidate at MIT Sloan. The study received funding from the National Science Foundation, the Army Research Office and Cisco Systems.