Empathy is one of the few skills that compounds across a lifetime. It shapes who your child becomes as a friend, a sibling, a partner, eventually a coworker and a parent. And here is what most parenting books underplay: empathy is not taught directly. You cannot sit a 4-year-old down and explain how to feel what someone else is feeling. That is not how children learn it.
Empathy is built through play. Specifically, through three kinds of play that map onto different developmental windows between roughly ages 3 and 9.
The first is doll play, where a child takes care of a small representation of another being. The second is pretend or imaginative play, where a child takes on a role and tries on a different perspective. The third is open conversation, where a child practices putting their inner world into words and listening to someone else do the same.
These three modes are not phases that replace each other. They overlap. A 5-year-old might do all three in a single afternoon. But each mode is most influential during a specific window, and parents who understand which mode is doing which work can structure a child's environment in a way that quietly builds empathy without it ever feeling like a lesson.
This guide walks through all three.
Why Empathy Matters (And Why It Is Built, Not Taught)
Decades of research on child development converge on the same point: empathy is the foundation skill underneath nearly every other social and emotional outcome that matters. Children who develop strong empathy in early childhood do better academically, form healthier friendships, navigate conflict more effectively, and report higher life satisfaction as adults.
But unlike literacy or math, empathy is not built through instruction. Lecturing a 4-year-old about “thinking about how Sarah felt when you took her toy” rarely changes behavior. What does change behavior is repeated, low-stakes practice in the kind of imaginative reasoning that empathy requires: imagining another being's experience, taking a perspective that is not your own, articulating what someone else might want or feel.
That practice happens during play. The age range from 3 to 9 is when the cognitive scaffolding for empathy gets built. After 9 or 10, kids continue to refine empathy, but the underlying capacity is largely set. Which means the years between 3 and 9 are not just any years. They are the years.
Doll Play (Ages 3 to 6): Where Empathy Begins
Doll play is the first form of structured empathy practice most children encounter. It looks simple from the outside. A child holds a small figure, names it, dresses it, talks to it, feeds it, puts it to sleep. From the inside, something far more sophisticated is happening.
When a child cares for a doll, they are practicing what developmental psychologists call theory of mind: the ability to recognize that another being has its own inner state, distinct from your own, and that you can act on that being's behalf even when its needs are different from yours. The doll is hungry. The doll is tired. The doll is sad. None of these are the child's own feelings, but the child practices noticing them and responding.
This is exactly the cognitive move empathy requires.
The kind of doll matters. Researchers and practitioners have long noted that open-ended dolls (those without preset personalities, scripted dialogue, or battery-powered features) produce richer pretend play than electronic toys that direct the child's behavior. A doll that does nothing on its own forces the child to do everything, which is the entire point of the exercise.
For parents looking to invest in dolls that support this developmental work, look for a few specific things. Soft enough to hold and sleep with. Realistic enough to project emotions onto, but neutral enough to be many different characters across many sessions of play. Durable enough to survive years of being carried, dressed, undressed, and tucked under arms. The Bebenita doll collection is one example of a doll line designed with these criteria in mind, made from materials that hold up to genuine play and with expressions open enough to support the wide range of caregiving scenarios children imagine.
If you want a deeper read on the developmental science behind doll play, our guide on dolls in child development as empathy builders covers the research foundation in more detail.
Pretend and Imaginative Play (Ages 4 to 7): Practicing Other People's Lives
If doll play teaches a child to take care of another being, pretend play teaches them to inhabit one. The shift is subtle but significant. In doll play, the child is the caregiver and the doll is the cared-for. In pretend play, the child becomes the doctor, the teacher, the firefighter, the parent, the bus driver. They step into a perspective that is not their own and try it on.
This is the developmental engine of perspective-taking, which is the most cognitively demanding aspect of empathy. Empathy is not just feeling for someone else. It is understanding what the world looks like from where they stand. Pretend play is hours and hours of low-stakes practice in exactly that.
Watch a 5-year-old playing “doctor.” They are not just arranging blocks or reading a book. They are reasoning about what a doctor knows, what a doctor would say, how a doctor would react when a patient is scared. They are running a real-time simulation of another mind. The fact that the patient is a stuffed bear or a younger sibling or an imaginary friend is irrelevant to the cognitive work. What matters is that the child is practicing being someone else.
The best pretend play environments give children minimal but suggestive props. A wooden kitchen, a doctor's bag, a costume box, a few small figures. The less prescriptive the prop, the more the child has to fill in with their own thinking. Our post on 7 playtime ideas for playing with dolls and toys that spark creativity is a useful resource for setting up this kind of play environment without going overboard on stuff.
The window for pretend play is roughly 4 to 7. Before 4, most children do not yet have the cognitive infrastructure to sustain extended role-play. After 7, they start to drift toward more rule-bound games and structured social play. The 4 to 7 window is the heart of imaginative play, and it is when the perspective-taking muscle gets built.
Open Conversation (Ages 5 to 9): When Kids Can Tell You What They Think
Around age 5, something new becomes possible. Children develop enough language and self-awareness to articulate their inner world out loud. Not perfectly, not always coherently, but well enough to have a real exchange with a parent. This opens the third mode of empathy practice: open conversation.
Open conversation does for empathy what doll play and pretend play could not. It lets a child practice naming their own feelings, listening to someone else name theirs, and noticing the gap between the two. This is the most direct form of empathy work, and it is the form that most parents most consistently miss.
The reason most parents miss it is that direct questions usually fail. “How was school?” reliably gets “fine.” “What did you do today?” reliably gets “I don't remember.” The standard parental questions activate the kind of summary-mode answer that gives the child no reason to actually think.
What works is the opposite. Specific, slightly unexpected, low-stakes questions that bypass the summary instinct. “If you could pick one rule to change at school, which one?” “Which kid in your class would you want as a sibling for a day?” “What is the best thing that happened on the playground today?” These questions invite a real answer because they require thinking, not summarizing.
One product designed specifically around this insight is Tell Me Cards, a 107-card conversation deck for families with kids ages 5 to 9. It was created by a founder whose wife is a practicing psychologist. Over the course of her work with adult clients, she kept noticing the same pattern: many of them grew up in homes where their inner world was never really invited out. Not because their parents did not care, but because no one had given them the right kind of questions to ask. The deck was built around open-ended prompts designed to bypass the "how was school" reflex and pull something real from a child.
But the broader point matters more than any specific deck. Children between 5 and 9 are at the most receptive point of their lives for honest conversation with a parent. They have the language to say what they think. They are still young enough to want to. And that combination does not last forever. Around age 10 or 11, most kids start guarding their inner world more carefully, especially from parents. The window is real, and it is now.
How the Three Modes Overlap
A common mistake parents make is thinking of these three modes as sequential. Doll play first, then pretend play, then conversation. That is not how children develop. The reality is that all three modes coexist for years, with the dominant mode shifting gradually as the child's cognitive capacity grows.
A 4-year-old who is deep in doll play is also already starting to do basic role-play. A 6-year-old who is deep in pretend play is also starting to have real conversations about what is fair and what is not. An 8-year-old who is having sophisticated conversations is still happily playing with their dolls when no one is watching.
The right way to think about this is layering, not replacement. Each mode adds a layer of empathy practice to the previous ones, and none of the layers fully retire. The developmental work compounds.
What this means practically is that parents of a 7-year-old should not phase out doll play or pretend play just because their child is verbal enough for conversation. All three modes can coexist in a single afternoon. The doll play that looked simple at age 4 still has value at age 7, even if the child is also having more articulate conversations than they could two years earlier.
Practical Ways to Weave All Three Into Daily Routines
The good news for parents is that you do not have to schedule any of this. The three modes of empathy-building play happen naturally when the environment supports them. A few practical patterns that work:
Keep dolls and pretend-play props accessible, not stored away. Children gravitate toward whatever is in their visual field. A doll on the bedroom floor invites doll play. A doll in a closet does not.
Resist the urge to direct the play. The cognitive work of empathy depends on the child doing the imagining. If a parent steps in and starts narrating (“Now Mr. Bear is feeling sad because…”), the child stops generating the empathy themselves.
Build conversation moments into specific anchors of the day. Dinner, the car ride to school, bedtime. The repeating context lowers the cognitive load of starting a conversation. Over time, your child will start to expect conversation in those moments, which is half the battle.
Ask questions that cannot be answered with one word. “What was the most surprising thing that happened today?” works better than “did anything good happen?” The harder a question is to summarize, the more thinking your child has to do, and the more empathy practice they get out of the answer.
Mix the modes. There is no rule that says doll play has to be silent, or that conversation has to happen at the dinner table. A 6-year-old who is acting out a scene with their dolls is one good question away from telling you something real about their own life.
What the Research Says About Play and Empathy
The connection between play and empathy development has been studied for decades. Researchers like Martin Hoffman, Nancy Eisenberg, and more recently Angeline Lillard at the University of Virginia have built a substantial literature showing that early empathy emerges from a combination of biological readiness, modeling by caregivers, and structured opportunities to practice perspective-taking. Play is the primary mechanism for that practice.
Studies on doll and pretend play specifically (going back to Vygotsky in the 1930s and continuing through more recent work on imaginative play) consistently find that children who engage in extended imaginative play score higher on measures of perspective-taking and emotional understanding. The effect is largest when the play is open-ended rather than scripted.
Conversation-based interventions have a shorter research history but a similar pattern. Studies on dialogic reading, where parents engage with children about books rather than reading at them, find that the conversational element predicts vocabulary and emotional understanding more reliably than the reading itself. The same principle applies to any structured conversation. The form is less important than the practice.
In short: the research supports what most experienced parents and educators already know intuitively. Empathy is built through repeated, low-stakes practice in imagining other minds. Doll play, pretend play, and open conversation are the three highest-leverage forms of that practice during the 3 to 9 window.
Conclusion
The three modes of play in this guide are not competing tools. They are layers that build on each other, and the work each layer does does not fully end when the next one begins.
A 9-year-old who has spent years on doll play, pretend play, and real conversation has been practicing empathy for thousands of hours, mostly without realizing it. They have rehearsed thousands of small moments of imagining another mind. By the time they need that capacity in real social situations (with a sibling, a friend, a classmate, eventually a partner), the muscle is already trained.
The window is shorter than most parents realize. The 3 to 9 years are when the foundation gets laid. The good news is that none of it requires expertise, expensive programs, or a developmental science degree. It requires having the right tools accessible, leaving room for the child to do the imaginative work, and being willing to ask better questions when the moment is right.
