“In solitude the mind gains strength and learns to lean upon itself.”
— Laurence Sterne
Constant communication may be limiting your productivity. In the modern world of cell phones and the internet, communication is easier than ever. We’ve been harvesting the advantages of this for years. We communicate with people on the other side of the world, stay in touch with colleagues outside the office, and immediately share work outputs with anyone who asks. But we have failed to spot one possible disadvantage this whole time: seamless team communication may prevent individuals from contributing their full potential.
We have suspected that isolated decision-making may be valuable for a long time. The classic “wisdom of crowds” effect has shown us that a total lack of communication between individuals can be a good thing, for example, when guessing the quantity of jelly beans in a large jar. If everyone throws their own independent estimate into the ring, you can take the average of all those estimates, and reach a surprisingly accurate answer. This is primarily because extreme guesses can fall either side of the truth: some people will guess too low, others too high, but the average will balance out those effects to approach the true number.
If people communicate in guessing the number of jelly beans, effects of social influence will tend to pull estimates in a particular direction. The most convincing member of the group may succeed in attracting people toward their point of view, even if they are dramatically wrong. With their estimate acting as the new anchor for other estimates, the average of all guesses can drift away from the true number. If history has taught us anything, it is that when leaders are wrong, they take others down with them.
However, in iterative tasks characterized by several rounds, social information can be helpful if it spreads among a decentralized group where each participant is equally connected to others. This is because the individuals with more accurate estimates tend to be the ones who attract other estimates toward them in successive rounds of the task. Instead of the loudest or most charismatic people becoming the anchor for other people’s estimates, the smartest people become the influencers.
* * *
Jelly beans are one thing, but when it comes to typical problem-solving tasks at work, it may be hard to believe that communication among team members could ever a bad thing. But a recent experiment published in August 2018 presents exactly this conclusion.
These researchers in Massachusetts developed a tricky problem-solving task to compare performance between different teams. Participants saw a basic map on a computer screen with a number of cities distributed across it, and had to find the optimal route around the map, visiting each of those cities only once and returning to their starting point. The task was complex enough to rule out the simplest solutions and strategies, so good performance depended on careful thinking and planning. On top of that, participants had less than a minute to attempt a solution in each of 17 rounds for a specific problem. However, these multiple rounds meant that they had a chance to gradually improve their solutions and strategies over time for each problem.
The participants did not know each other but were randomly assigned to complete the task in anonymous teams of three. Different teams had to stick to different communication rules. One type of team was allowed constant communication, which meant that team members could see each other’s previous solutions during every round of the task. Another type of team only had intermittent communication, in which they could see each other’s previous solutions only every three rounds. The final type of team were in fact not really a team at all because they were allowed no communication whatsoever; they had to complete the task without using any information from other people’s solutions.
So how did these different teams compare in solving problems? First, consistent with some previous evidence, the researchers found that people who did not communicate at all ended up finding the optimal solution more often than teams who constantly communicated (44% vs 33% of problems). The lack of communication led to greater diversity in thinking styles and responses, increasing the chances that at least one of the independent individuals would land on the optimal solution.
In contrast, the teams who constantly communicated performed better than non-communicators when researchers calculated the quality of their average solutions. Communication allowed people with bad solutions to improve the quality of their decision-making by looking at the successes of their teammates, raising the average standard for the team overall. So at the cost of reduced creativity and diversity compared to the independent individuals, the communicators could better build on the best solutions among them in each consecutive round of the task.
But what about the intermittent communicators? In principle, they could achieve the best of both worlds or the worst of both worlds. The results showed that they found the optimal solution to 48% of the problems, matching the independent non-communicators in their performance, and beating the constant communicators. Interestingly, when the researchers turned their attention to average performance rather than total optimal solutions, intermittent communicators also matched the high-performing constant communicators, and beat the non-communicators.
So according to the evidence, the intermittent communicators enjoyed the benefits of both non-communicators and constant communicators. They could make use of the diversity associated with isolated work and the cooperative development associated with communication. Their biggest improvements came when seeing the solutions of their teammates after a period of disconnected decision-making.
* * *
“One can be instructed in society; one is inspired only in solitude.”
— Johann Wolfgang von Goethe
We may finally have an answer to the long debate about whether independent thinkers or highly interconnected teams make better decisions. The unexciting answer is that each of them has their strengths and downfalls. Independent thinkers can have flashes of ingenuity uncontaminated by the thoughts or interruptions of others, while communicative groups achieve a higher average standard by sharing the strengths of each person. The exciting answer is that we may be able to do something about this dichotomy: we can build teams characterized by intermittent rather than ongoing communication.
Isolation has been a useful tool throughout our history. Many historical geniuses have had their moments of insight and inspiration while working alone in their creative spaces. While they may have stood on the shoulders of giants, they didn’t have those giants standing beside them during their most critical moments of thinking and problem-solving. They found their foundation in the books that they read and theories they came across, and then gave themselves the necessary period of solitude to allow their most creative sparks to take hold and spread.
There may even be parallels in biological evolution. New species tend to emerge when old species are physically separated into multiple groups. Those isolated groups then evolve their own unique adaptations to their environments and gradually become different enough to the original species that they can no longer mate with each other. In other words, the isolation that leads to greater biological diversity over the long-term may be analogous to the isolation that leads to greater decision-making diversity over the short-term.
We have entered an era of hyperconnected communication in every aspect of our lives. We check our social network feeds and the news headlines every morning, we frequently video-chat with friends and colleagues, and we have access to all the existing knowledge in the world through a quick Google search. I remain appreciative of this extraordinary situation, but no longer sit with the illusion that there are no major costs. We may well be sacrificing elements of creativity and diversity that were previously so valuable for innovative progress.
With a little effort, it may be possible for us to break away from our communication rituals and stop corrupting some of our best ideas with the perspectives of others. But it may require a dose of internet withdrawal when we most need to be creative and alone with our thoughts. If we want to design a solution better than both a horse and a camel, we may need to temporarily suspend the committee.