Children use make-believe aggression and violence to manage bad-tempered peers

Academics from the University of Cambridge believe that the tendency for children to introduce aggressive themes in these situations -- which seems to happen whether or not they are personally easy to anger -- may be because they are 'rehearsing' strategies to cope with hot-headed friends.

The finding comes from an observational study of more than 100 children at a school in China, who were asked to play with toys in pairs. Children whose play partners were considered bad-tempered by their peers were 45% more likely to introduce aggressive themes into their pretend play than those whose partners were reckoned to be better at controlling their temper.

Importantly, however, a child's own temperament did not predict the level of make-believe aggression. Instead, children often appeared to introduce these themes specifically in response to having an irritable playmate.

This may mean that in certain cases aggressive make-believe play actually helps children's social and emotional development. The paper's authors stress, however, that further research will be needed before they can provide definitive guidance for parents or practitioners.

Dr Zhen Rao, from the Centre for Research on Play in Education, Development and Learning (PEDAL), at the Faculty of Education, University of Cambridge, said: "If children have a friend who is easily angered, and particularly if they haven't coped well with that behaviour, it's possible that they will look for ways to explore it through pretend play. This gives them a safe context in which to try out different ways of handling difficult situations next time they crop up in real life."

Aggressive pretend play has been the subject of considerable wider research, much of which aims to understand whether it predicts similarly aggressive real-life behaviours. Most of these studies, however, tend to focus on whether these associations are linked to the child's own temperament, rather than that of the children they are playing with.

The Cambridge study aimed to understand how far aggressive pretend play is associated with a play partner's anger expression. It also distinguished between aggressive pretend play and its 'non-aggressive, negative' variant: for example, pretend play that involves imagining someone who is sick or unhappy.

The research was carried out with 104 children, aged seven to 10, at a school in Guangzhou in China, as part of a wider project that the team were undertaking in that region.

Participants were asked to organise themselves into pairs -- many of them therefore picking friends -- and were then filmed playing for 20 minutes. The toys they were given was deliberately neutral in character (for example, there were no toy weapons), and the children could play however they wanted.

The researchers then coded 10-minute samples of each pair in 120 five-second segments, earmarking instances of pretend play, aggressive themes, and non-aggressive negative themes.

Separately, they also asked peers to rate the children's tendency to become angry. Each of the 104 children in the study was rated by, on average, 10 others, who were asked to decide whether they were good at keeping their temper, easily angered, or 'somewhere in between'.

The researchers then analysed the data using a statistical model called an Actor-Partner Interdependence Model, which is a means of measuring and testing the influence that two individuals have on one another. This allowed them to work out how far children were playing a certain way of their own volition, and how far they were being influenced by their partner.

On average, the children spent only about a fifth of their recorded session participating in pretend play, of which around 10% involved aggressive themes and 8% involved non-aggressive negative themes. Pretend play was observed in all children. More than half (53.5%) showed at least one instance of aggressive pretend play, and 43% of the children showed at least one instance of negative pretend play.

The children's own ability to control their temper, as reported by their peers, did not significantly predict how much their pretend play involved aggressive themes. If they had a play partner who was considered quick to anger, however, they were 45% more likely to create pretend situations that involved some sort of aggressive element. This percentage is to some extent shaped by how the data was segmented, but nonetheless indicates a greater likelihood that children will do this if they are playing with someone peers regard as easy to anger.

There was no evidence to suggest that either child's temperament influenced the frequency of non-aggressive, negative pretend play. The researchers also found that boys were 6.11 times likelier to engage in aggressive pretend play than girls.

The theory that children may introduce these themes to rehearse ways of handling bad-tempered peers is only one possible explanation. For example, it may also represent an attempt to stop playmates becoming angry by giving them a pretend situation in which to 'let off steam', or simply to keep them playing by appealing to their nature.

"Our study highlights the importance of taking into account a social partner's emotional expression when understanding aggressive pretend play," Rao added. "Further research is clearly needed to help us better understand this in different social contexts. The possibility that children might be working out how to handle tricky situations through pretend play suggests that for some children, this could actually be a way of developing their social and emotional skills."

The research is published in the British Journal of Developmental Psychology. Dr Rao's research is funded by an ESRC postdoctoral Fellowship.