From coalition governments to teams of scientists, the notion that “two heads are better than one” is the en vogue approach to problem-solving these days.  

The premise is simple: We perform better and make more sensible decisions by putting our heads together than by working alone.  For example, let’s say that there are two stock traders – Warren and Ben.  The notion is that if they work together to pick stocks they will make more profit than if they work in isolation.

But is this true? And, if so, is this the case all of the time or are there some situations in which they’ll benefit more from collaboration than others? What factors influence this? And what will maximize the success of the collaboration?

A new study offers novel insights into these questions. Published in the journal Science last week, it was led by Bahador Bahrami of the Interacting Mind Project (University College London and Aarhus University in Denmark). In this research, two participants worked together or in isolation to make decisions about visual images. Their task was a little bit like a scientist’s version of Where’s Waldo: They had to pick the image that contained a target. They initially decided on an answer alone. Then, if the duo gave different answers, they had to discuss who gave the right answer and form a joint decision. To test whether two heads really are better than one, these collective decisions were then compared to performance when each person worked alone.

The first results showed that, yes, two heads were indeed better than one. The volunteers combined their information “optimally” and reached a level that none of the players could possibly achieve as an individual (good news for Warren and Ben!).  In effect, the volunteers were able to combine weak neuronal activities residing in two separate brains to maximize performance.  There was, however, a twist in the tale.  

The key to the success was communication – about how they felt about their answer and how confident they were in their decision. When they were not allowed to communicate with each other about their confidence, they couldn’t do any better than the best solo player.  So, even if Warren found a fantastic junior mining stock that he thought was a ‘sure thing’, it would only help to maximize profits if he could share his level of confidence in the stock with Ben.     

It was also critical that both players reported their confidence reliably. If one of them was poor at the task but didn’t know it, the team performance only got worse. Put differently, if Warren’s ‘sure thing’ was actually a mediocre option, then making a decision together would not make their team better at all. In fact, it would have a detrimental effect.
This implies that two heads may be better than one, but only when we can competently discuss our different perspectives.  If one person in the team has flawed information -- or is less competent -- then the outcome can be negative and perhaps you should completely ignore them.

Bahrami’s study tells us that what’s important for successful collaboration is the ability to estimate and report our own ability accurately. However, this is not always easy, especially for incompetent individuals. In psychology, there is a known cognitive fallacy called the Dunning-Kruger effect. The most incompetent individuals often overestimate their skills and think they are all above average, though that’s logically impossible. Having such a person in your team would severely damage performance. Simply put, if you are not sure about your competency in a team, the most productive thing to do is to tell your team members -- though in reality, of course, this is not easy.

Another well known cognitive fallacy is the overconfidence effect, which is the tendency for our subjective confidence to be larger than our objective accuracy.  Alongside the Dunning-Kruger effect, these effects would imply that successful collaboration has some problems to overcome before we can reap the benefits.  They might also point to the construct of confidence as being a difficult indicator of accuracy, but studies like Bahrami’s suggest that by engaging in a discussion about our confidence judgements we may be able to refine them to reach an optimal decision.

Finally, it is interesting to consider the extent to which the principle that two heads are better than one extends to three, four or more heads.  Let’s say that in addition to Warren and Ben, we introduce Mervyn and Vince to the equation.  What would happen to performance?  In this context, the phenomenon of "social loafing" may be informative.  Increasing the number of group members can sometimes reduce the social pressure on each individual in the group, and actually reduce a given person’s contribution.

For example, Warren might be a great stock trader when he is being assessed on his work in isolation, but he may work at only half of his potential when the focus is on the team’s performance.  
So, are two heads really better than one? It seems the answer is yes, but success depends on the competency and confidence of the individuals involved. And by the way, if you’re investing your money with Warren and Ben, you might want to take very careful note of how well they communicate with each other.