Medicine Technology 🌱 Environment Space Energy Physics Engineering Social Science Earth Science Science
Science 2026-02-26 3 min read

Children Who Aren't Friends Cooperate 25% Better When Given a Shared Goal

Study of 148 children aged 6-8 found goal-directed play raised non-friend connectedness from 44% to 55%, while friends showed almost no change

Give children toys and tell them to play nicely. They will, in all likelihood, sit next to each other and largely play alone. Give them a task to complete together - something with a defined goal and shared materials - and something different happens, particularly if the two children don't already know each other well.

A study from the Universities of Cambridge and Sussex, published in the journal Infant and Child Development, put this dynamic to a systematic test with 148 children aged six to eight. The results were clear: goal-directed play raised cooperative communication among non-friends by around 25 percent. Among established friends, almost nothing changed.

Measuring connectedness, not just proximity

The researchers organized children into pairs of friends and non-friends, identified through each child's own assessment of their three best friends. Each pair participated in two activities: a free play session with a Playmobil treehouse set, and a collaborative drawing task in which both children had to work together to complete a single picture using shared materials.

The measurement was "connectedness" - the proportion of conversational turns in which a child's statement was linked to what the other child had just said. A connected exchange reflects active coordination: listening, responding, building on the partner's contribution. Pairs were given a connectedness score - a percentage measure of connected conversation rates across each activity.

Across the full sample, average connectedness rates were higher during the drawing task than during free play. The difference was statistically significant at around four percentage points. But when the researchers split the sample into friends and non-friends, the aggregate finding decomposed dramatically: non-friends' connectedness jumped from 44 percent to 55 percent. Friends' connectedness barely moved - 48 percent during free play, 50 percent during goal-directed play.

Why friends don't need the scaffold

Lead researcher Dr. Emily Goodacre, from the PEDAL Research Centre at Cambridge, offered an explanation: "Friends have shared experiences and an intuitive understanding of how to play together, but non-friends lack that familiarity and might benefit from being set a goal." Friends communicate through non-verbal cues, implicit understandings, and a shared history that allows them to coordinate without the verbal scaffolding that a shared task provides. Non-friends have none of that. They are, in effect, strangers who have been handed toys and told to get along.

A shared goal changes the dynamic for non-friends because it provides a common reference point - something both children need to attend to and respond to. The drawing task required genuine interdependence: with only one pad and a few pens, children had to negotiate whose turn it was and what the image should look like. That structure replaced the implicit social knowledge that friends bring to unstructured play.

Implications for classrooms and group activities

The study used data from five UK primary schools. The findings align with a broader body of research from the PEDAL centre suggesting that cooperative behavior in children depends less on individual social skills than on contextual factors - who they're with and what they're doing.

For teachers and parents, the practical implication is specific: simply providing toys and instructing children to play together is a less effective strategy for promoting cooperation between non-friends than assigning a collaborative task with a defined shared goal. The goal provides the structure that non-friends need but friends already supply through their relationship.

Limitations of the study include the UK school setting - different cultural contexts or classroom compositions could produce different results. The study measured connectedness as a proxy for cooperation rather than directly observing outcomes or relationship development over time. Whether the increased connectedness observed during goal-directed tasks translates into stronger social bonds between non-friends in the longer term was not assessed in this design.

Source: Goodacre E et al. Published in Infant and Child Development. DOI: 10.1002/icd.70089. Researchers from the PEDAL Research Centre, University of Cambridge, and University of Sussex. Contact: Thomas Kirk, University of Cambridge, tdk25@cam.ac.uk, 7764-161923.