The Science of Doll Play: Barbie, Empathy, and the Trouble with Branded Neuroscience
Doll play is easy to dismiss if you are determined to be dreary about it.
Adults often file it under cute, domestic, childish, gendered, nostalgic, or “things found under the sofa with one shoe missing.” But pretend play with dolls is psychologically more interesting than that. A child moving figures around a room is not just passing time. They may be staging arguments, rehearsing care, inventing motives, assigning feelings, testing power, managing conflict, and conducting tiny social experiments in which everyone has plastic hair and no concept of tax.
That does not make dolls magical. A child does not become empathic because Barbie unlocks a secret empathy gland. The more sensible claim is that dolls invite children to practise social imagination. They give children bodies to speak through, characters to care for, conflicts to resolve, and minds to invent.
That is worth taking seriously.
It is also worth taking carefully, because some of the most publicised research on doll play has been conducted in collaboration with Mattel, the company behind Barbie. Privately funded research is not automatically bad. Universities collaborate with companies all the time. But when a toy company funds or promotes neuroscience showing the developmental benefits of its kind of product, we should read the claims with both interest and one eyebrow politely raised.
The science of doll play is interesting.
The marketing version is where things start dressing themselves as certainty.
Doll play is social rehearsal
Pretend play often involves perspective-taking. A child does not only move a doll from one place to another. They decide what the doll wants, knows, feels, fears, likes, remembers, or misunderstands. They may create scenes of friendship, danger, family life, school, illness, rescue, conflict, jealousy, comfort, and dramatic betrayal over who gets the pink cup.
That matters because early social development involves learning that other people have minds of their own. They can believe things, want things, hide things, misunderstand things, and feel differently from us. Psychologists often talk about this as part of theory of mind: the developing ability to understand mental states in oneself and others.
Dolls provide a useful play format for this because they are both objects and characters. They do not do anything by themselves, which is precisely the point. The child supplies the voice, motive, feeling, and relationship. The doll becomes a placeholder for imagined inner life.
This is also why doll play should not be reduced to “girls’ play” or treated as training for domesticity. Children of any gender can use dolls to practise social thinking. A doll can be a baby, superhero, patient, teacher, monster, friend, parent, villain, lost explorer, or deeply unreasonable café customer. The psychological value is not in the doll being feminine-coded. It is in the child using figures to create social worlds.
Dolls may matter less as products and more as social placeholders.
The Cardiff and Mattel studies
The research that brought doll play into wider public discussion came from a collaboration between Cardiff University researchers and Mattel.
In a 2020 study, researchers examined children aged four to eight while they played with dolls and tablets, both alone and with others. They used functional near-infrared spectroscopy, or fNIRS, to monitor activity in brain regions associated with social processing. Cardiff reported that solo doll play was linked to greater activation in the posterior superior temporal sulcus than solo tablet play, suggesting that children may engage social-processing systems while playing with dolls even when they are alone. The study was conducted in collaboration with Mattel, which is important context for how the findings were later promoted.
A follow-up study looked at children’s use of internal state language: words and phrases about thoughts, feelings, desires, intentions, and knowledge. This matters because internal state language gives researchers a window into whether children are talking about minds, not just actions. If a child says, “She thinks the puppy is lost” or “He’s sad because nobody came,” they are not only moving toys around. They are constructing mental states.
The findings suggested that children used more internal state language when playing with dolls than when playing with tablets. Again, that is interesting. Doll play seemed to encourage children to talk about what characters think and feel, which is exactly the kind of social imagination that developmental psychologists care about.
More recently, a 2026 randomised-control trial reported that children aged four to eight who engaged in doll play over six weeks improved more in false-belief reasoning than children assigned to creative tablet play. False-belief understanding is a key part of theory of mind: it involves understanding that someone else can hold a belief that is different from reality. The study’s authors described this as evidence that doll play may support social-processing skills, while also noting the need to understand mechanisms and generalisability more fully.
This newer work strengthens the case that doll play may do more than simply correlate with social language. It suggests there may be a causal effect, at least under the conditions studied. That is a meaningful step.
Still, the careful phrase is “may support,” not “proves Barbie builds empathy,” despite what a marketing department might understandably prefer.
The neuroscience needs a little less glitter
The phrase “brain regions associated with empathy and social skills” is the kind of line that travels well in press releases.
It sounds impressive. It feels concrete. It gives a familiar toy the glow of laboratory seriousness. But neuroscience language can easily make ordinary developmental claims sound more definitive than they are.
If a brain region becomes more active during a task, that does not mean the task directly builds a trait in a simple way. Brain activation is evidence of engagement, not an automatic certificate of long-term developmental benefit. It can tell us something about what the brain is doing during play, but it does not settle every question about empathy, social skill, or real-world behaviour.
This is not a dismissal of the research. The use of fNIRS with young children is clever, and the findings are genuinely interesting. The problem is the temptation to turn “doll play engages social-processing regions” into “dolls make children empathic,” and then into “this specific branded doll is practically a neuroscience intervention with accessories sold separately.”
That is the jump to watch.
A child talking through dolls may be practising social cognition. That is plausible and supported by the broader play literature. But the strongest interpretation is about pretend character play, not one particular corporate universe of dolls, dreamhouses, careers, and very ambitious footwear.
Barbie is not the mechanism
One of the key questions is whether the developmental value comes from Barbie specifically, dolls generally, or a broader category of character-based pretend play.
The most reasonable answer is that Barbie is probably not the active ingredient. The likely mechanism is not the brand, the body shape, the wardrobe, or the fact that Barbie has had more careers than most employment agencies. The mechanism is probably that dolls invite children to create social scenarios involving characters with minds.
This matters because toy marketing loves specificity. “Doll play supports social thinking” is a developmental claim. “Our doll supports social thinking” is a sales claim. Those two claims sit close together, but they are not identical.
Children can use many kinds of figures for social imagination: dolls, action figures, stuffed animals, peg people, puppets, animal figures, Lego characters, or frankly two spoons and a worrying amount of commitment. The question is whether the play encourages the child to imagine perspectives, emotions, relationships, and intentions.
If it does, then the exact toy may matter less than the form of play. A superhero figure can support internal state language. So can a baby doll. So can a dinosaur who is apparently upset because nobody came to his birthday. Children are generous like that. They will give an inner life to almost anything not nailed down.
The developmental value is in the child’s social imagination, not in the product’s brand identity.
Tablets are not automatically the villain
The studies often compare doll play with tablet play, which makes sense experimentally, but it can encourage a slightly lazy reading: dolls good, screens bad.
That is too simple.
Some tablet play is passive, repetitive, and not especially social. Some is creative, collaborative, language-rich, and imaginative. A child playing alone on a tablet may engage differently from a child co-playing with an adult, making a story, designing characters, or narrating what is happening. The medium matters, but the activity matters too.
Likewise, not all doll play is equally rich. A child can play with dolls in a socially imaginative way, or they can simply undress them, lose the shoes, and move on with the emotional decisiveness of a small landlord. The developmental value depends on what the child actually does with the toy, whether there is narrative, whether mental states are being imagined, whether adults or peers join in, and how free the play is.
The best takeaway is not that every doll is superior to every screen. It is that open-ended pretend play, especially play involving characters and relationships, gives children opportunities to practise social thinking.
Dolls are one good route into that.
They are not the only route, and the tablet is not automatically waiting in the corner wearing villain music.
The funding problem
Industry-funded research needs careful interpretation.
That does not mean the studies are invalid. The Cardiff/Mattel work has been peer-reviewed, conducted by academic researchers, and has produced findings that fit with broader theories of pretend play and social development. It would be lazy to dismiss it simply because Mattel is involved.
But it would also be naive to ignore the commercial context. Mattel benefits if parents, educators, and journalists come away believing that Barbie is not merely a toy but a developmental good. The company has every reason to promote findings in the most flattering possible light. That does not make the findings false. It means the public interpretation may become too tidy.
The question is not “was this funded by a company, therefore is it bad?” The better question is: what exactly did the study show, what did it not show, and how has the finding been translated into public claims?
Good science often speaks in careful terms: associated with, may support, suggests, under these conditions, further research is needed. Marketing prefers cleaner verbs. Builds. Boosts. Improves. Proven. Helps your child become socially brilliant and presumably easier at restaurants.
The gap between those two languages is where scepticism belongs.
What parents and educators can actually take from it
The practical takeaway is refreshingly ordinary: children benefit from rich pretend play.
That does not require expensive dolls, branded accessories, or a panic purchase because a headline mentioned neuroscience. It means children need time, space, and permission to create imaginary social worlds. Dolls and figures can help because they give children characters to think with.
Adults can support this without hijacking it. Ask gentle questions. “What does she think is happening?” “Why is he upset?” “What will they do next?” “Does the other character know that?” These kinds of prompts encourage children to think about perspectives, beliefs, emotions, and intentions.
But adults should also know when to leave the play alone. Pretend play is not always improved by turning it into a developmental workshop. Children are allowed to make Barbie fall off a sofa for reasons that remain narratively unclear. Not every moment needs a learning objective lurking behind the curtains.
The best support is often simple: provide open-ended toys, allow time for unstructured play, join in when invited, and take children’s imaginary worlds seriously without colonising them.
A doll does not need to be sold as neuroscience to be valuable.
Sometimes it is enough that it gives a child someone to speak for.
Simply Put
Doll play may support children’s social development because it invites them to imagine minds.
When children play with dolls, they often create characters with thoughts, feelings, desires, conflicts, and relationships. That kind of play can support social language, perspective-taking, and theory of mind. The Cardiff and Mattel research programme offers interesting evidence for this, including brain-imaging work, internal state language findings, and newer randomised evidence on false-belief reasoning.
But the claims need careful handling.
The sensible conclusion is not that Barbie is a small plastic empathy machine. It is that character-based pretend play can give children a useful space to practise social imagination. Dolls are one route into that, but not the only one.
The branded neuroscience angle is the bit to watch. Research funded or promoted by a toy company is not automatically wrong, but it should not be allowed to turn a nuanced developmental finding into a sales pitch with brain scans.
Dolls can matter.
Not because they are magic, and not because a company says so, but because children use them to do something psychologically rich: invent inner lives, stage social worlds, and rehearse what it means for other people to have minds of their own.
References
Hashmi, S., Vanderwert, R. E., Paine, A. L., & Gerson, S. A. (2022). Doll play prompts social thinking and social talking: Representations of internal state language in the brain. Developmental Science, 25(2), e13163. https://doi.org/10.1111/desc.13163
Hashmi, S., Vanderwert, R. E., Price, H. A., & Gerson, S. A. (2020). Exploring the benefits of doll play through neuroscience. Frontiers in Human Neuroscience, 14, 560176. https://doi.org/10.3389/fnhum.2020.560176
Kristen, S., Sodian, B., Licata, M., Thoermer, C., & Poulin-Dubois, D. (2012). The development of internal state language during the third year of life: A longitudinal parent report study. Infant and Child Development, 21(6), 634–645. https://doi.org/10.1002/icd.1767
Table of Contents
