Pretence, the act of treating something as if it were something else, is one of the earliest and most cognitively demanding things young children do. A toddler who lifts an empty cup to a doll’s lips or announces that the floor is lava is performing a mental operation that researchers have spent decades trying to fully explain: holding two representations of reality in mind at once, the actual state of the world and an imagined alternative. Far from being idle entertainment, pretend play turns out to be deeply connected to how children learn to regulate their own behavior, understand other people’s minds, and develop language. It also appears to have roots that extend millions of years into our evolutionary past.
When Pretend Play Emerges
Most children begin showing signs of pretend play between about 18 and 24 months of age. Early forms are simple: a child might hold a banana to their ear like a phone or “feed” a stuffed animal with an empty spoon. Over the next year or so, pretence becomes increasingly elaborate. Children start stringing pretend actions into sequences, assigning roles to other people or toys, and eventually creating entirely imaginary objects or scenarios with no physical prop at all.
This developmental progression has been tracked in detail. In a study following younger siblings of children with autism spectrum disorder alongside typically developing toddlers, children in all groups showed clear growth in pretend play between 22 and 34 months, engaging in more bouts of pretence and scoring higher on tasks requiring them to transform objects or act out scenarios as they got older.1PubMed Central. Developmental Changes in Pretend Play from 22- to 34-Months in Younger Siblings of Children with Autism Spectrum Disorder The trajectory matters: children who lagged behind in this progression, particularly those later diagnosed with autism, showed measurably less pretend play across contexts and less social engagement with parents and examiners.2PubMed Central. Pretend Play and Social Engagement in Toddlers at High and Low Genetic Risk for Autism Spectrum Disorder That link between pretence and social connection shows up early and persistently.
Understanding Other Minds
One of the most studied questions about pretend play is how it relates to children’s ability to understand what other people are thinking and feeling. To pretend that a block is a cake, you have to mentally represent the block as something it isn’t while knowing it’s still really a block. That dual-representation skill looks a lot like what’s needed to understand that another person can hold a belief that differs from reality, what psychologists call false belief understanding.
Research bears this out, though the picture is nuanced. A study examining the relationship between pretend play, mental-state language, and false-belief understanding in preschoolers found reliable connections between false belief performance and specific pretence categories, particularly object substitution (using one thing as another) and role assignment (declaring “you be the doctor”). Children who used more mental-state terms in their speech also tended to perform better on both false-belief tasks and these particular kinds of pretence.3British Journal of Developmental Psychology. An investigation of pretend play, mental state terms and false belief understanding: In search of a metarepresentational link Not every type of pretend play showed this connection, though. Simple pantomimes where the child used their own body as a stand-in for an object, like pretending a hand is a telephone receiver, were actually negatively correlated with mental-state language. The researchers concluded that some components of pretend play depend on the same capacity for thinking about mental representations that false-belief understanding does, but pretence is not a monolithic skill.
Pretend play also shares cognitive territory with counterfactual reasoning, the ability to think about what would have happened if circumstances were different. A study using confirmatory factor analysis found a significant correlation between children’s pretend play abilities and their counterfactual reasoning, with the two constructs sharing about a quarter of their variance.4PubMed Central. A plausible role of imagination in pretend play, counterfactual reasoning, and executive functions Both seem to draw on a shared imaginative capacity: the ability to mentally construct a scenario that isn’t currently real and reason about it coherently.
How Children Navigate the Pretend-Real Boundary
A persistent puzzle in the study of pretence is why young children don’t seem to get confused by it. If a two-year-old can pretend that water is tea, why don’t they also worry that their real tea might secretly be water? The answer is more interesting than “kids just know.” Research suggests that young children are actually quite conservative about what they accept as real, sometimes even too conservative.
A review of the evidence on how children judge reality status concluded that young children tend to be “naïve skeptics” rather than credulous fantasists. Rather than accepting too much as real, they often reject things that are genuinely real but unfamiliar to them. The researchers proposed that immature abilities to evaluate the scope and relevance of their own knowledge lead children to over-rely on what they already know when deciding what’s real, making them skeptical of things outside their experience.5PubMed Central. Revisiting the fantasy-reality distinction: children as naïve skeptics With development, children get better at drawing on a wider range of sources to inform their reality judgments.
Even within pretend play itself, children’s grasp of what’s real and what’s performed develops gradually. Young children can distinguish pretending from joking by age two; in a study of 72 toddlers, two-year-olds protested more against jokes than against pretend acts, showing they understood these as different kinds of intentional behavior.6Child Development. Two-Year-Olds Distinguish Pretending and Joking But understanding that the emotions expressed during pretence might not match the person’s real feelings takes considerably longer. A study of over 300 children aged four to twelve found that four-year-olds were not aware that emotions expressed in pretend play might differ from internal emotions. This distinction didn’t emerge until around age six, and even then, children found it easier to recognize when a pretended negative emotion was fake than when a pretended positive emotion was.7PubMed. Do children realize that pretend emotions might be unreal?
Self-Regulation and Impulse Control
One of the most practically significant findings about pretend play is its connection to self-regulation, the ability to control your own behavior, resist impulses, and follow rules. This link has theoretical roots in the work of Lev Vygotsky, who argued that pretend play forces children to act against their immediate impulses because the role they’re playing has its own rules. A child pretending to be a soldier, for instance, has to stand still even though they want to run around. That internal tension, Vygotsky proposed, is where self-regulatory skill is built.
Modern research supports this idea. A study examining children’s play from a cultural-historical perspective found that children’s play skill and their self-regulation abilities were significantly linked: children who performed their assumed role better in play also showed higher self-regulation. The children’s playing skills, along with their gender and age, predicted their self-regulation abilities.8PubMed Central. Pretend play as the space for development of self-regulation: cultural-historical perspective
The type of pretend play matters too. A study testing Vygotsky’s assumptions directly found that cooperative sociodramatic play, the kind where children jointly construct a pretend scenario and negotiate roles, predicted improvements in self-regulation during cleanup periods. Solitary dramatic play, by contrast, was negatively correlated with improved self-regulation. Most striking was the finding that the relationship between sociodramatic play and improved self-regulation was especially strong for highly impulsive children and essentially nonexistent for children who were already low in impulsivity.9Early Childhood Research Quarterly. Self-regulation in young children: Is there a role for sociodramatic play? In other words, the children who needed the most help with impulse control seemed to benefit the most from collaborative pretend play. This is an important finding for parents and educators, because it suggests that pretend play isn’t just reflecting pre-existing self-regulation skills but may actively help develop them in children who are behind.
Pretend Play and Language Growth
Pretend play and language development are intertwined in ways that go beyond the obvious observation that children talk while playing. The sophistication of a child’s pretend play appears to predict later language abilities, and play-based educational approaches seem to boost language skills more effectively than traditional instruction.
A longitudinal study found that the quality of preschool children’s pretend play, specifically how elaborately they sequenced pretend events and how flexibly they used symbols, predicted up to about a fifth of the variation in their semantic organization and narrative retelling skills up to five years later.10PubMed. Quality of pre-school children’s pretend play and subsequent development of semantic organization and narrative re-telling skills That’s a notable proportion of variance for a single predictor measured years earlier. The ability to use symbols flexibly in play was the strongest predictor of narrative skills, which makes intuitive sense: both require holding a representation in mind and communicating it to someone else.
Another study found more specific concurrent and longitudinal links between pretend play tasks and language competence in preschoolers, though not all pretend play measures predicted all language outcomes equally. A telephone task, where children pretended to make a phone call, predicted later receptive vocabulary.11Cognitive Development. A multidimensional investigation of pretend play and language competence: Concurrent and longitudinal relations in preschoolers The pattern that emerges across studies is that pretend play’s relationship to language isn’t generic; specific forms of pretence connect to specific language skills.
On the educational side, children in a play-based curriculum showed significant improvement in pretend play skills, narrative language ability, and grammar knowledge over six months, while children in a traditional classroom did not.12Journal of Early Childhood Research. An investigation into the effect of play-based instruction on the development of play skills and oral language This is one of the more practically relevant findings for schools debating how much structured play to include in early education.
Evolutionary Roots of Pretence
Pretend play isn’t unique to humans, and its presence in other species suggests it has deep evolutionary origins. The evidence comes from two directions: observations of play signaling in wild primates and experimental work with language-trained apes.
In the wild, many primates engage in social play that includes aggressive-looking behaviors like biting, chasing, and wrestling. To keep these from escalating into real aggression, animals use signals to mark their actions as playful. A study of free-living Hanuman langurs tested the “metacommunication hypothesis,” which proposes that certain play signals modify the meaning of aggressive-looking actions. The researchers found that the play face, an open-mouthed expression distinct from genuine aggression, served exactly this function: when play face accompanied aggressive-looking elements at the start of a play bout, those bouts lasted longer rather than being terminated.13Behaviour. Metacommunication in social play: the meaning of aggression-like elements is modified by play face in Hanuman langurs (Semnopithecus entellus) This is metacommunication, a signal about a signal, and it represents a basic form of the “as if” framing that underlies human pretence.
More striking evidence comes from apes who have been raised in human-like communicative environments. A study of chimpanzees and bonobos with interspecies communication systems found that these animals achieved basically the same levels of representational play as human children, including the creation of imaginary objects and the ability to extend pretend scenarios initiated by their human caregivers.14Cognitive Development. The development of representational play in chimpanzees and bonobos: Evolutionary implications, pretense, and the role of interspecies communication A more recent experimental study pushed this further. Kanzi, a lexigram-trained bonobo, correctly identified the location of pretend objects, for example tracking where “juice” had been poured between empty containers, in response to verbal prompts. Across three experiments with controls for alternative explanations, the researchers concluded that the capacity to form mental representations of pretend objects is within the cognitive potential of at least an enculturated ape and likely dates back six to nine million years, to the common ancestor of humans and great apes.15PubMed. Evidence for representation of pretend objects by Kanzi, a language-trained bonobo
This doesn’t mean wild apes routinely pretend in the way human children do. The communication-rich environment these apes were raised in appears to be crucial. But the fact that the cognitive hardware to support pretence exists in our closest relatives suggests that the foundations were in place long before modern humans appeared.
What Happens in the Brain During Pretence
Studying pretend play in a brain scanner is tricky with toddlers, so some of the most interesting neuroscience on pretence comes from an unusual population: professional actors. A study using functional MRI scanned actors while they responded to questions either as themselves or in character as Romeo or Juliet. The results were surprising. Responding in character produced widespread reductions in brain activity, particularly in the cortical midline network of the frontal lobe, including areas associated with self-referential processing. Rather than adding extra mental activity on top of the actor’s normal self-processing, portraying a character appeared to involve a deactivation-driven process, something the researchers described as a “loss of self.”16PubMed Central. The neuroscience of Romeo and Juliet: an fMRI study of acting
This finding challenges the intuitive assumption that pretending should be more cognitively demanding than being yourself. Instead, it suggests that part of what happens during pretence is a suppression of your own default perspective to make room for an alternative one. Whether the same neural pattern holds for children engaged in everyday pretend play remains an open question, but the finding aligns with the broader idea that pretence involves managing competing representations of reality and selectively dampening one of them.
Pretend Play and Autism
Reduced pretend play has long been noted as one of the early behavioral markers of autism spectrum disorder. Toddlers who go on to receive an ASD diagnosis show less pretend play across different settings and less social engagement with both parents and examiners compared to low-risk toddlers and high-risk toddlers who don’t receive a diagnosis. Within the high-risk group, lower levels of pretend play and social engagement track with symptom severity.17PubMed Central. Pretend Play and Social Engagement in Toddlers at High and Low Genetic Risk for Autism Spectrum Disorder
The developmental trajectory of pretend play in children at heightened risk for autism follows the same general upward slope as in typically developing children, but with a persistent lag. Between 22 and 34 months, children later diagnosed with ASD engaged in fewer bouts of pretend play and scored lower on elicited pretend and transformation tasks than both low-risk children and high-risk children without a diagnosis.18PubMed Central. Developmental Changes in Pretend Play from 22- to 34-Months in Younger Siblings of Children with Autism Spectrum Disorder
However, the relationship between autism and pretence is more complex than a simple deficit story. Some autistic children do engage in rich imaginative play, and those who create imaginary companions appear to show distinct social advantages. A study of autistic children who had imaginary friends found that these children were better able to understand others’ minds and had stronger social skills than autistic peers without imaginary friends, and this difference was not explained by language ability.19PubMed Central. Autistic children who create imaginary companions: Evidence of social benefits Whether the imaginary companions are helping to build social understanding or whether children with stronger social cognition are more likely to create imaginary companions in the first place is unclear from a single study. But the finding complicates the older narrative that autistic children simply “can’t pretend.”
Therapeutic Uses of Pretend Play
Play therapy, which frequently draws on pretend play as its central tool, is used with children facing a range of difficulties. A review of the literature found play therapy to be suitable for preschool and school-age children dealing with life transitions like parental divorce or loss, difficulties managing feelings, socializing challenges, behavioral problems, exposure to domestic violence, abuse history, upcoming major surgery, speech disorders, and developmental conditions including autism and ADHD. Research reviewed showed significant decreases in negative emotions and behaviors with its use.20PubMed Central. An Overview of Play Therapy
A randomized controlled trial tested child-centered play therapy specifically with children who had experienced adverse childhood experiences and were at risk for complex trauma. Children who participated in the therapy showed significant increases in empathy, social competence, and self-regulation, along with decreases in total behavior problems.21Journal of Counseling & Development. Child‐centered play therapy and adverse childhood experiences: A randomized controlled trial The pretend play context gives children a way to process experiences they may not have the vocabulary to talk about directly. A child who can’t articulate what frightened them can act it out with toy figures, giving a therapist access to the child’s internal world through a medium the child already knows.
Pretence Across Cultures
Pretend play appears in every culture that has been studied, but its form varies considerably. A cross-cultural comparison of pretending at home found that while the basic capacity for pretence is universal, the details differ. In some cultures, objects are central to pretend play; in others, the body and social interaction carry the pretence without much reliance on props. Who plays with the child varies too: in some cultural contexts, caregivers are active pretend play partners, while in others, siblings or peers fill that role. The extent to which children initiate social pretend play with their caregivers, the functions that pretend play serves in interaction, and the specific themes children choose all differ across cultural settings.22Child Development. Universal, Developmental, and Variable Aspects of Young Children’s Play: A Cross-Cultural Comparison of Pretending at Home
This is worth knowing because much of the research on pretend play comes from Western, middle-class samples with access to large numbers of toys and caregivers who have been told play is important. In cultures where children are expected to participate in real work from a young age, pretend play may take a different form and occupy a different social role, but it doesn’t disappear. The universality of pretence across radically different child-rearing contexts is itself evidence that it reflects something fundamental about human cognitive development rather than a culturally constructed luxury.
When Adults Pretend for a Living
Pretence doesn’t stop in childhood. Adults engage in it constantly, from white lies and social niceties to professional contexts where managing one’s displayed emotions is part of the job. In organizational psychology, this workplace pretence is called emotional labor, and it comes in two flavors: “deep acting,” where you genuinely try to feel the emotion your role requires, and “surface acting,” where you display an emotion you don’t actually feel. Surface acting is the adult equivalent of the pretend-emotion distinction that children begin to grasp around age six.
Unlike childhood pretend play, which appears to benefit development, chronic surface acting in adults carries measurable costs. A meta-analysis across multiple occupational groups found that surface acting correlated significantly with depression and anxiety.23PubMed. The impact of emotional labor on mental health: A systematic review and meta-analysis of multi-occupational groups In healthcare specifically, surface acting intensified experiences of job stress and emotional exhaustion, affecting not just individual professionals but the systems they work in.24PubMed Central. Exploring the Relationship between Surface Acting, Job Stress, and Emotional Exhaustion in Health Professionals: The Moderating Role of LMX
The contrast is striking. In childhood, pretence builds cognitive flexibility and social understanding. In adulthood, when pretence becomes obligatory and chronic rather than voluntary and playful, it erodes well-being. The difference may come down to agency and framing: a child who pretends to be sad while playing “hospital” chose the role and can drop it at any time. A nurse who has to project calm reassurance while feeling overwhelmed cannot. The “as if” framing that makes pretend play developmentally useful, the knowledge that this is not real, becomes a source of strain when the person performing it can’t step out of the performance.