Imagine a classroom in the United Kingdom. A Playmobil treehouse sits between two kids at a table. They are merely classmates and don’t really know each other. Behind a camera, a researcher silently observes. The kids pick up pieces, put them down, and circle around one another for a while in that somewhat uncomfortable manner that strangers do when they are confined to the same area. Not much takes place. After that, they are given a drawing of a tree trunk and some coloring pens, and they are instructed to construct a treehouse out of it. Almost instantly, something changes.
A recent study from the Universities of Cambridge and Sussex, published in February 2026, revolves around that little moment, and the results are worth considering. Researchers divided 148 kids between the ages of six and eight into pairs, some of whom were friends, and observed what transpired during two distinct forms of play. Children were given a set of toys and allowed to play freely. During goal-directed play, they were tasked with working together to finish a drawing using limited materials. The instinctive belief that cooperative play is primarily a personality trait that a child either possesses or lacks is challenged by what the researchers discovered.
During the goal-directed task, children’s connectedness—the researchers’ term for how frequently they link their ideas in conversation, responding to each other rather than talking past one another—rose significantly. The thing that stuck, though, was that not every child experienced the same level of improvement. The non-friend pairs were almost the only ones driving it. Their connectedness scores increased by about 25%, from 44% to 55%. Friend pairs hardly moved at all. It’s possible that a formal goal doesn’t really affect close friends because they rely so heavily on nonverbal cues like shared history, a glance, or a half-sentence. Lacking that shorthand, non-friends seemed to require an external anchor for their interactions. a reason. A reason to pay attention.
Working on the study through Cambridge’s PEDAL Research Centre, Dr. Emily Goodacre acknowledges that she was initially taken aback by the results. “When I first saw the results, I thought: this doesn’t make sense.” It is actually comforting when a researcher expresses honest confusion because it indicates that the data was doing something unexpected rather than just confirming what everyone already knew. Once she sat with it, her interpretation was pretty obvious: friends have scaffolding built in. A common objective can give non-friends something else to cling to.
None of this is particularly startling to anyone who has observed kids working on a group project in the classroom. Most adults seem to already be aware of this intuitively: putting people together and hoping chemistry develops seldom works as well as providing them with something tangible to strive for. This is useful because, instead of using a corporate team-building metaphor, researchers have now quantified it in a controlled environment using actual children and a drawing of a tree trunk.
The question of how to plan group activities in the classroom has subtle but significant ramifications. Structure may be a social bridge rather than merely a management tool if cooperative communication between unfamiliar children depends more on what they’re doing than on who they are. It’s still unclear how teachers should apply this to their everyday work and whether the effect persists over longer periods of time and in different age groups. However, the fundamental message is difficult to overlook. The next time a parent tells their kids to “play nicely,” it might be worthwhile to give them a problem to solve as a group. It turns out that the playing nicely usually follows.
