It’s good to talk, but periods of silence are essential for problem solving. - Jaime Bonetti Zeller

It’s good to talk, but periods of silence are essential for problem solving.

Article Published by:

If you – like many people across the world – are now working remotely, you might be worrying about loss of facetime with your team. Remote collaboration tools might help us to make up for this by allowing regular communication throughout the working day – and it would seem to make sense to take advantage of that connectivity with constant updates about your progress.

Yet the latest psychological literature suggests that less communication might actually be more: constant collaboration can in fact reduce ‘collective intelligence’ (a team’s joint problem-solving ability). Instead of always staying in touch with colleagues with continual chats on Slack, say, we should aim to concentrate group communication to short, intermittent bursts – a single daily video call, for example – to boost team problem solving and creativity.

Besides helping us to make better use of our time during the current crisis, these findings could help to shape the ways that we go about team decision making in the future. Even if we are in the office, we might all benefit from having a bit more me time and a bit less team time.

Jumping to conclusions

This new understanding of group communication has been years in the making. Jesse Shore at Boston University kicked things off in 2015, with a study that explored the role of people’s connectivity during group problem solving. Put simply, is it useful for everyone to be talking to everyone else, so we all know what’s going in in every part of the team? Or is it sometimes better to limit our communication to just a few individuals?

Shore and his colleagues set up 51 groups, each consisting of 16 people. They asked the groups to play an online whodunnit game, borrowed from one of the US Department of Defense’s R&D teams, in which participants find and piece together clues to discover to predict the who, what, where and when of a (fictional) terrorist attack.

The participants were given a user interface to share their findings with other team members, but the researchers controlled exactly how many people within each group would see each person’s updates. In some groups, each person’s messages were passed on to the majority of their teammates; this was a bit like a group chat channel, broadcasting info to everyone. In others, the messages went to just a couple of other members. Those members could then choose to share it with another couple of people, but no one could broadcast to the whole team at once. In this way, it was a little more like email, where you might be more likely to pass on information to select individuals rather than cc’ing everyone at every stage of the process. Messages can still pass through the whole group through a chain of communication, but each person is acting as a bit of a filter.

Initially, in the information-gathering ‘reconnaissance’ stage of the game, the more connected groups performed well: the ability to pass leads on to all the other members meant that the team quickly gathered lots of potential clues about the attack. But they soon lost that advantage when they had to piece that information together to form a coherent theory about the way the terrorist plot would take place. Although you might expect that these teams would struggle to come to an agreement, like a hung jury, the major problem was conformity: the team members quickly converged on a consensus without really exploring the other possibilities. “People aren’t brainstorming effectively – they’re not going off in their own direction,” explains Shore.

The less well-connected groups, in contrast, suffered a bit with the information gathering, but they were also less likely to reach a consensus too quickly. Without the immediate updates from all teammates, each member was instead more likely to build their own theories, meaning that there was a greater diversity of ideas available before the team as a whole settled on the best solution.

Alone time

This finding, by itself, would suggest that teams might consider limiting the communication among members. A bit of chit chat is good, but you don’t necessarily want to know what everyone else is doing during the more creative parts of the problem solving that require the generation and testing of lots of ideas.

Inspired by this result, Shore and his colleagues next examined how the rhythm of our communication can also influence our problem solving. Even if you are working in a small group, you have the option to receive constant updates, or limit your communication to a few regular catch-ups – but which is best?

To find out, the researchers asked participants to find solutions to a classic puzzle known as the “travelling salesman” problem, in which they were given a map of 25 different cities and needed to work out the shortest journey passing through them all. It’s best solved iteratively, as you play with different options – and often retracing your previous steps – to find the optimal path.

The participants were divided into groups of three. Using an online interface, some groups were allowed to interact constantly – they had immediate feedback on the solutions their teammates were trying. Others had only intermittent communication between periods of independent work. And some ‘groups’ had no internal communication at all – they spent all the time working by themselves.

The researchers measured the effects on two outcomes – the team’s average performance, and the performance of the top performer within each team. (That’s important in real-life work – you want everyone to perform better, but you don’t want the top individuals to be dragged down by less able individuals either all.)

Once again, moderation was key. The groups with continuous interaction had good average performance, but it flattened the performance of the top-performing individuals. As the researchers had also seen in the previous experiment, the team members all tended to conform to mediocre solutions. “The ‘out-there’ ideas aren’t really considered,” Shore explains. “And some of their ideas might actually be excellent, even if most of them might not be.”

The people working independently had the opposite problem – there was lots of variation among people, with some brilliant solutions, but the worst performers didn’t have the opportunity to benefit from others’ solutions, meaning that they dragged down the group average.

The intermittent communication turned out to be the ‘Goldilocks’ outcome – it was just the right about of interaction for both these measures: the average performance of the group was high, without reducing the performance of the best member. “The groups could build on the diversity of the ideas that were generated independently, and then integrate them,” says Shore. “It was the best of both worlds.”

‘Bursty’ communication

Shore’s results complement the research of Anita Williams Woolley of Carnegie Mellon University. Woolley studied 260 software workers, divided into groups of five, who were tasked with designing a new medical algorithm over a 10-day contest. She found that teams working with short “bursts” of communication, followed by longer periods of silence, performed better than less intense conversations stretched out over a long time. Once again, it seems that intermittent updates are best.

Woolley’s team suggest that “bursty” communication could help maintain momentum and motivation in various ways. One is obvious to anyone who has tried to focus on a task while receiving a steady stream of DMs: dealing with all your communication in one go reduces distraction.

Short but intense communication also helps to create feelings of enthusiasm, which can dissipate if you face a slow and steady drip feed of emails without immediate feedback. By synchroenising their communication to relatively short periods, the best groups could catch up on each other’s progress, share ideas, and coordinate their activity going forward, enthusing each other in those interactions while still leaving enough time for intense concentration afterwards.

Given these results, it might be tempting to look for a hard-and-fast rule for the optimal level of communication, but the exact ratio of active collaboration and solitude almost certainly depends on the task at hand and the medium. You have to judge it on your own team dynamics for whether a daily Zoom call is necessary, for example; whether Slack use should be limited to certain stages of the project; or whether you are better sending a single round-up email rather than a stream of one-line messages.

What’s clear, however, is that we probably don’t need to be connected to each other all the time. Whether we are working remotely or in the office, it’s good to talk – but sometimes our interactions with others are best taken in short doses.

About Jaime Bonetti Zeller

Jaime Bonetti Zeller is an investment professional and entrepreneur with businesses in multiple industries. He is president of Servicios Consulares Eurodom, the local partner in the Caribbean region for VFS Global, a leader global outsourcing and technology services specialist for diplomatic missions and governments worldwide. Jaime Bonetti Zeller also started the company Sofratesa de Panama inc., an organization in the engineering services industry located in Panama City.