Thalia R. Goldstein Ph.D.

The Mind On Stage

Drama Games Help Kids Gain Emotional Control

Guiding children through pretend play games causally helps their emotion skills.

Posted Oct 06, 2017

Please see the full scientific report here, or watch a short video about the paper here.

What can pretend play, drama, and role play do for very young children? When kids pretend they are using their imaginations, obviously, but what other kinds of social, emotional, and cognitive skills might they use, and how might such use lead to increased levels of those skills? There are many different options, including language, social understanding of others, emotional regulation, empathy, and compassion. There has also been a considerable amount of research in developmental psychology looking at the relationship between children’s involvement in pretend play and their levels of these skills. However, as detailed by an important and well-read paper published by Angeline Lillard at the University of Virginia and her group, most of the research on the relationships between pretend play and these outcomes lacks scientific controls or rigorous and careful research necessary in order to be able to make a causal claim. That is, the research isn’t strong enough to be able to say, for example, "Yes! Playing pretend makes you better at empathy!" 

Thalia R. Goldstein
Source: Thalia R. Goldstein

When it comes to claims about drama games and acting training, the ideas are often even grander, and with much less scientific study. While there is some good evidence to support the idea that drama games can lead to increased vocabulary and narrative abilities, the links between engaging in drama and increases in social skills are not clearly delineated. Drama teachers, students of drama, and actors have long talked about what theatre and role play does for them psychologically, and the anecdotes from classrooms and productions of play are many. But strong, well-controlled scientific evidence is still rare.

What does it mean to provide clear, causal evidence for activity X causing outcome Y? Without getting too much into research methodology and scientific philosophy, the basic premise is that you must run an experiment (called a Random Control Trial, or RCT) in which:

1. You have children involved in activity X, and compare them to a closely related activity, in order to isolate that activity X is causing change, rather than any activity at all (this is the “control” part)

2. The children involved in the experiment do not get to choose which of these activities they are going to be involved in (this is the “random” part)

3. The people running the activity don’t know what you’re trying to find, so they can’t “teach to the test”

4. The people doing the testing don’t know what you’re trying to find, and also don’t know whether the children they are testing were doing activity X or some other activity.

This is the gold standard. It probably sounds very similar to the types of studies you hear about when reading about new medicines or diets. That’s because it is.  RCTs are the ultimate scientific way to determine what input causes which output. Without good control conditions, and people who do not have preconceived notions about the outcomes of a study, it’s hard to truly isolate the causal path from point X to point Y.

As recently published in the journal Developmental Science, this is what Matthew Lerner from Stony Brook University and I set out to do. I developed an eight-week dramatic game pretend play intervention. These games were based on Viola Spolin’s improvisational exercises that have come to form the bases of modern improvisation (think Whose Line is it Anyway, for the preschool crowd), but the eight weeks/ 24 sessions also included more standard pretend play activities, like dressing up in Chef’s Hats and “making a birthday cake”, or pretending to be a statue and having a classmate pose you in different ways. Kids were directed through these games by a group leader who knew how to work with the children and how to guide them through a task, but who did not know what we were studying or why. And these group leaders didn’t just teach the drama games. They also taught two control conditions.

To claim that these games were causally changing children’s emotional and social abilities, we have to isolate what was specific to these games—outside of the fact that children were simply physically moving around, interacting with a teacher, interacting with each other, or thinking about characters—all things that children can do in lots of activities, not just in dramatic pretend play games. So, in parallel, I developed two control conditions to test alongside the drama games. One control condition was building with blocks. The children in this group were still working in a group, physically moving around the space, interacting with a group leader, and working towards a goal. But, there was no physicalizing of characters, no embodiment of emotions and mental states. The second control condition was reading stories. The children in this group were still thinking about narratives and characters and still working with each other and a group leader to answer questions, but there was again no physical embodiment of characters, emotions, or mental states.

To test the effects of these dramatic pretend play games, we conducted an experiment in a preschool summer program. After their parents gave permission, a group of 4- and 5-year-olds in a HeadStart Program in New York City were randomly assigned, by lottery, to participate in one of the three groups. Group leaders, trained in the interventions, but not in the outcomes, led the 24 sessions. A different group of lab research assistants, training to become clinical psychologists, tested the children in the week before the intervention started, and then again a week after the intervention ended. None of the group leaders or research assistants knew the goals of the study, although of course, we live in a culture which talks constantly about educational and social outcomes for children, as well as what kinds of play or testing activities may best help those outcomes.

Before the children had received any of the “intervention” (the different types of play), and then again after the end of the 24 sessions, we tested their emotional control (how well could they calm their own distress in the face of others’ distress, and how often did they report becoming overwhelmed by their own emotions), as well as on their empathy, helping, and compassion. We found that only in the dramatic pretend play group, and not in the block play or story time group, children got better at emotional control over the eigh weeks. From this finding and the methods we used to discover it, we can make the claim that it was something about the drama games, specific to embodying characters, emotions, and mental states (because we tried to control for everything else), that caused the children to get better at controlling their overwhelming emotions. Importantly, though, we did not find that the children in the dramatic pretend play group improved their understanding of others, empathy or compassion any more than the children in the other groups. The effect of the dramatic pretend play was unique to emotional control.

Why this effect, and why so specifically?

The control conditions of block play and story time controlled for interacting with a teacher, interacting with other children, and goal-oriented activities. Storytime shared a focus on narrative and characters with dramatic pretend play games, and block time shared physical movement and working towards a single goal in a group. Unique in the dramatic pretend play game group was embodiment. Only in this group did children embody emotions, mental states, characters, and goals. And it’s this embodiment that may have taught children how to control their emotional states, not to overreact or become overwhelmed by others.

Why did we not find higher levels of empathy or understanding others?

A lot of previous theory and work has focused on drama, theater, and pretend as a way to orient children to other people and allowing them to feel more empathy for others. We didn’t find that here. One possibility is that it’s not something specific about drama that increases empathy. Perhaps anytime children have to work in a group towards a goal, as they did in block play and in drama games, they increase empathy. Or perhaps, simply learning about narratives and emotions is enough, and therefore there was no differential effect between drama games and story time.

But another possibility is about the group of children we worked with, and the effects we did find. You cannot feel the emotions of another person if it causes you to get overwhelmed. It’s hard to have empathy for others if you’re too busy with your own distress. Therefore, the ability to control and lessen your own emotional state is often considered a precursor to the ability to understand others’ emotional states and feel empathy and sympathy for them. If a child is unable to control their emotions, they won’t feel empathy (and, frankly, neither will adults!).

So it’s possible that with this group of low SES children, they began with lower than average levels of emotional control (a claim that would require a group of high SES children as a comparison group to validate as true). What the drama games uniquely did, then, is to increase their emotional control. Perhaps if we had kept going, we would have seen later increases in empathy and sympathy.  Work to think about for the future!

As a final note, I do not want to claim in any way that the reason to have children engage in drama classes or in pretend play is to give them better emotional control. That’s not the purpose of drama or play. I’m interested in what drama can do for children, as well as basic questions as to why it exists (despite so many forces, financial and time constraints, that seem to work against it). Emotional control may be one benefit, for this group of children, but other benefits may come out depending on the type of drama activities, the type of children, and the needs of each group. This, I think, is the real bonus and benefit of drama and acting training. It’s incredibly flexible. Different acting techniques, games, and activities can be shaped and used with different populations—from Theatre of the Oppressed techniques for communities dealing with subjugation—to classical Shakespearean training for prisoners to work through issues of emotional expression and anger. We’re only beginning to scientifically scratch the surface of all that performance, embodiment of character and emotion, and playing stories can do.  

*This work was funded by a generous grant from The John Templeton Foundation. All opinions expressed herein are the author’s own.

Please see the full scientific report here, or watch a short video about the paper here.


Goldstein T.R., & Lerner M.D. (2017). Dramatic pretend play games uniquely improve emotional control in young children. Developmental Science. e12603.

Lillard, A. S., Lerner, M. D., Hopkins, E. J., Dore, R. A., Smith, E. D., & Palmquist, C. M. (2013). The impact of pretend play on children's development: A review of the evidence. Psychological bulletin, 139(1), 1.

More Posts