10-Minute Speech Practice That Doesn't Require Sitting Still

Last updated 2026-07-10

Why does symbolic play actually build language?

Play and words run on the same engine. That engine is representational thinking, the ability to let one thing stand in for another. A banana becomes a phone. A block becomes a car. And a word does the exact same trick: the sound "dog" stands in for the furry animal. When a child practices that substitution in play, they are exercising the same mental machinery that makes words work.

Lorraine McCune tracked children from 8 to 24 months and found that the onset of pretend play predicted the onset of referential speech with a consistency that surprised even her team. Her research reported that children who hit pretend play milestones hit matching language milestones within weeks, not months ^[1].

The American Speech-Language-Hearing Association (ASHA) puts symbolic play right in its developmental milestones guidance, noting that pretend play and language develop together and that a delay in one often travels with a delay in the other ^[2]. That is why speech-language pathologists (SLPs) spend so much of a session on the floor playing instead of drilling words at a table.

Play is not a magic fix. A child with a structural speech difference like apraxia of speech or a child who needs AAC devices may need targeted support layered on top of rich play. But play makes the soil where language grows.

What counts as symbolic play, and what does not?

Symbolic play means the child is using an object, action, or person to represent something it is not, or acting out a scene that is not literally happening. It goes by many names: pretend play, make-believe, fantasy play, representational play. Same thing.

What counts:

• Using a stick as a sword
• Feeding a stuffed bear pretend soup
• Holding a toy phone to your ear and saying "hello"
• Assigning roles ("you be the doctor")
• Acting out an imaginary scene with no props at all

What does not count as symbolic play:

• Functional play: pushing the toy car along the floor because that is what cars do
• Constructive play: stacking blocks to build a tower
• Sensory play: pouring sand back and forth

Those earlier play types matter, and they come first developmentally. Most children begin simple pretend acts around 12 to 18 months (feeding a doll, sipping from an empty cup) and build multi-step pretend sequences between 24 and 36 months ^[3]. Richer, multi-scheme pretend ("the baby is sick, we drive to the hospital, the doctor needs a needle") usually shows up between 30 and 48 months ^[2].

For late talkers or autistic kids, the timeline often runs different, and that is fine. The goal is moving forward along the sequence, not hitting it by a certain birthday.

Which symbolic play activities build the most language?

Here are twelve activities, ranked roughly from simpler to richer by the language they pull out of a child. You do not need to buy anything special. Most of these work with what is already in your kitchen or toy bin.

1. Object substitution games (12-18 months and up) Pick up a hairbrush and pretend to talk on it like a phone. Laugh, look surprised, hand it to your child. You are modeling the core operation: one thing stands for another. Paul Harris's work at Harvard shows children begin accepting and producing object substitutions right around the time they hit 50-word vocabularies, which suggests the two skills are tightly coupled ^[4].

2. Feeding and caring for dolls or stuffed animals This is the entry-level pretend scene, and it works precisely because it is simple. Grab a stuffed animal, a small cup, and a spoon. "Bear is hungry. Bear wants soup. More soup?" The play frame lets you model requesting, commenting, and two-word combinations with zero pressure on the child to perform.

3. Tea party or picnic play Pour pretend liquid, pass pretend food, say please and thank you. The social script of a tea party repeats and stays predictable, which is exactly what many late talkers and autistic kids need. Repeating the same familiar script across days is one mechanism that helps children lock in new vocabulary ^[5].

4. Doctor kit play Lay a stuffed animal on a makeshift table. "Uh oh, bear has a boo-boo. Hurt! Bear is sad." Let the child grab the toy stethoscope. "Listen. Thump thump. Better!" This scene pulls out emotion words, body-part words, and simple stories, three categories that tend to run thin in late talkers' early lexicons.

5. Kitchen and cooking play Stir the pot, taste the food, share it with family (real or toy). Cooking play generates action verbs (stir, pour, cut, mix) that are usually outnumbered by nouns in early vocabularies. Verbs are harder to learn from static labeling, and play context makes them much easier to grab ^[6].

6. Car wash, garage, and vehicle play For kids whose main interest is vehicles (common in many autistic kids), lean in instead of redirecting. A car wash setup with a sponge and a bucket creates language for washing, dirty and clean, in and out, fast and slow. Following the child's attentional lead is one of the most replicated findings in early intervention research: kids learn more words when adults talk about what the child already cares about ^[7].

7. Grocery store or shop play Set up a store with items from the pantry. Pretend to pay. Pretend to run out of something. Pretend a product is broken. The scene naturally pulls out questions ("Do you have apples?"), negation ("No more bananas"), and social scripts ("Thank you, come again").

8. Puppet play Puppets lower the social demand for a lot of kids. The child does not have to hold eye contact, take a direct turn with you, or manage their own face. They talk to the puppet. For autistic kids who find direct interaction effortful, dropping that social overhead can mean many more communication attempts per session.

9. Small-world play (farm sets, dollhouses, train layouts) Arranging and narrating small figures pulls out some of the richest spontaneous language you will hear from a preschooler. Listen to a child play with a farm set for five minutes and you can read their vocabulary, their sentence length, and the complexity of their internal narratives. Some SLPs use exactly this as an informal clinical observation.

10. Dressing up and role play Putting on a firefighter hat creates a persona, and personas produce language the child might never say as themselves: commands ("Everyone out of the building!"), announcements ("Fire! Fire!"), narration ("We saved the cat!"). Useful for kids who are chatty in some settings and silent in others.

11. Toy phone or walkie-talkie conversations Simulated phone talk has a steady finding in the literature: it pulls out more complex, sustained dialogue than face-to-face talk in many children, possibly because the missing face removes visual processing demands and dumps all the communicative weight onto words ^[4].

12. Narrative play with books as props This is not reading together. Take a familiar picture book, close it, and act out what happens using toys or your bodies. "Now you be Max. I'll be the Wild Things." The child already knows the story, which lowers the load, and re-enacting it produces connected narrative language, one of the strongest predictors of later reading and academic skill ^[5].

How should parents actually play? What are the techniques?

How you show up matters more than which toys you own. Four techniques have real research behind them.

Follow the child's lead. This is the most replicated finding in early intervention. When adults talk about what the child is already looking at or doing, word-learning rates climb measurably. Labeling a toy the child is ignoring does much less ^[7]. This does not mean you never introduce a new theme. It means you earn the right to introduce something new by joining what the child is already doing first.

Self-talk and parallel talk. Self-talk is narrating your own actions: "I'm putting the doll to bed. There, cozy." Parallel talk is narrating the child's: "You're pouring the tea. More tea!" Both drop language into context, tied to what is happening in the room, with no demand for a response.

Stay one step ahead, not ten. If your child is not yet combining two words, model two-word combinations, not full sentences. Clinicians call this expanding at MLU+1, and it holds up well. Language slightly above your level is useful. Language far above your level is noise.

Wait. After you model something or ask a question, pause. A full seven seconds. Look at the child with an expectant face. Most parents fill silence out of reflex. That silence is the window where children process and answer. Close it too fast and you close the opportunity.

Early intervention services are federally available for children under 3 through the Individuals with Disabilities Education Act (IDEA), and SLPs will often coach parents in these exact techniques inside the home, because that is where the play hours actually happen ^[8].

Does symbolic play help autistic kids build language?

Yes, though the research is nuanced and the road can look different. Many autistic children run detailed functional play (making everything work exactly as designed) before making the leap to pretend. Some skip or minimize pretend play and build strong language through other routes: scripts, echolalia, or AAC.

Echolalia, repeating phrases heard before, is not the same as pretend play, but it is also representational language and should not be treated as a problem to erase. ASHA's guidance points clinicians toward building on a child's existing communication strengths instead of suppressing them. If you want to understand echolalia and how it feeds language development, the echolalia article covers it in depth.

A 2020 meta-analysis in Psychological Bulletin (Project AIM) reviewed intervention studies in young autistic children and found that naturalistic developmental behavioral interventions, the ones that follow the child's interests and teach in play, produced stronger effects on spontaneous language than adult-directed formats, though the authors flagged that many studies had weak methods and possible bias ^[9].

So the practical read: if your child is obsessed with trains, the train set is not a distraction from language. It is the venue for it. Set up scenes. Create problems (the track breaks, the train gets stuck, the passengers are hungry). Narrate. Wait. Respond to any communication attempt in any modality.

If your child uses a communication device, that device belongs at play time. Many AAC users sharply increase device use when the context is play instead of structured instruction. AAC devices can and should live inside pretend scenes, not get set aside for "real communication moments."

For more on autism spectrum speech therapy approaches, including how SLPs build sessions around play, that article goes into clinical detail.

At what age should symbolic play start, and when is a delay a red flag?

Here is a rough developmental table built from published milestone research at ASHA and the American Academy of Pediatrics (AAP) ^[2][3]:

These are not hard cutoffs. Children vary. But the AAP's developmental surveillance guidance points to no pretend play by 18 months as worth raising with a pediatrician, especially alongside limited pointing, few words, or a limited response to their name ^[3].

One honest caveat: most normative data for symbolic play milestones comes from white, middle-class, English-speaking samples. Nobody has strong population-wide data across languages and income levels. Treat the milestones as directional, not as a verdict.

What if my child won't engage in pretend play at all?

More common than you would guess, especially in kids with sensory differences, autistic kids, and children with significant speech delays. Try these before deciding your child is "not a play person."

Drop the expectation and just narrate. Play alongside the child instead of at them. Make sounds, act things out, narrate what you are doing. You are modeling without asking for anything back. Some children watch for days or weeks before joining. That is fine. The input counts even when there is no output.

Try a different format. Some kids who ignore dolls and role-play light up for cause-and-effect toys with a pretend edge (the toy oven dings when the food is "done"). Others come alive in sensory-rich pretend: a dinosaur hunt in a sand bin, cooking with real water and food coloring. Fit the format to the child, not the child to the format.

Look for what the child already does that is one step from pretend. Does she carry a stuffed animal everywhere? That attachment is a relationship, more than a comfort object. Narrate the bear's day ("Bear is tired") without asking the child to pretend anything. That is a bridge.

If a child shows no interest in any symbolic activity by 30 months and this comes with limited or absent speech, early intervention evaluation is the right next step, not something to sit on. IDEA Part C services are free in every U.S. state for children under 3 who qualify ^[8].

For kids between 3 and school age, speech therapy evaluations through the school district or a private SLP can sort out whether play and language delays are linked and what to do next.

Can screen time replace symbolic play for language learning?

No, and the evidence on why is fairly clear. Children under 2 learn words from live interaction and from video chat (where a real person responds) but pick up little or no vocabulary from recorded video, even the high-quality educational kind ^[3].

Researchers call this the video deficit effect, and it has been replicated across multiple labs. The leading explanation: kids learn words most efficiently when the language is contingent, meaning it responds directly to what the child just did or looked at. A screen cannot respond to your child's specific gesture, gaze, or babble. A parent on the floor can.

Some apps use responsive design to approximate contingency, including apps built for late talkers. Little Words, for one, uses adaptive response patterns that mirror what an SLP would target in a naturalistic session. If you want to see whether it fits your child, the start quiz takes about three minutes. But no app replaces the relational, physical experience of pretend play with a caregiver.

For children over 2, high-quality educational programs watched with a co-viewing adult who talks about what is on screen can add some vocabulary, but the effect runs much smaller than live interaction.

How do speech therapists use symbolic play in sessions?

SLPs have used play-based therapy for decades, and the research backing is strong. The approach goes by several names: naturalistic developmental behavioral intervention (NDBI), milieu teaching, play-based language intervention. They share one structure: set up a play context, follow the child's interests, arrange the room to create communication opportunities, respond to any attempt, expand on what the child says or does.

A 2020 review in the American Journal of Speech-Language Pathology found that naturalistic play-based interventions improved vocabulary and sentence complexity in children with language delays, with effect sizes generally in the moderate-to-large range ^[10].

In practice, a therapist playing with a 2-year-old might:

• Set up a farm set and wait silently until the child reaches for something
• Label what the child reaches for ("cow!")
• Add a descriptor or action ("black cow, moo")
• Stage a problem (the farmer falls down) and wait to see if the child initiates about it
• Treat any vocalization, gesture, or device press as a real communicative turn

It looks like play. It is also precisely targeted therapy. The targets (specific words, word combinations, question forms) get set before the session and tracked after.

Parents can run a simplified version at home, and parent coaching is increasingly built into formal speech therapy programs, because parent hours dwarf therapy hours. A child who sees an SLP one hour a week is in therapy 52 hours a year. That same child is with a caregiver for roughly 5,000 waking hours. The math is obvious.

What toys and materials actually help, and what is a waste of money?

Direct version: the research does not support buying expensive toy lines. The studies showing symbolic play benefits use generic props, not branded sets.

What actually helps:

Open-ended props beat hyper-realistic toys. A plain wooden spoon pulls out more pretend play than a plastic toy spoon with sound effects. The more complete and realistic a toy is, the less the child has to imagine. Imagination is the point. Classic work found children produced more varied language with less realistic props because they had to generate the meaning themselves.

A small box of "junk" beats most toy sets. Empty cardboard boxes, toilet paper rolls, fabric scraps, pots and wooden spoons, old hats, plastic containers. These items carry no built-in script, so the child has to write one.

What to spend money on if you want to: a simple dollhouse with small figures, a basic play kitchen, a set of small animals, a doctor kit. Each gives enough scaffolding to launch pretend without scripting it so tightly that imagination has nowhere to go. None needs to cost more than $20 to $40.

Mostly a waste: toys that do the pretending for the child (the one that tells a story at the press of a button), screen-based "interactive" toys for kids under 3, and toys with complex assembly that adults have to manage. They pull the focus off the child's own imagination.

For AAC users, one genuinely useful buy is a low-tech communication board that mirrors your child's device vocabulary. Tape it to the play kitchen. Keep it near the farm set. Now play time doubles as AAC practice without separating the two.

How can parents track whether play-based language work is paying off?

You do not need formal tests at home. Four observational markers are meaningful and easy to catch.

Vocabulary breadth. Are new words appearing, even slowly? Keep a running note on your phone. New words per month is a rough but real metric. A child gaining no new words across 60 to 90 days, despite consistent rich play, is worth a formal re-check ^[2].

Utterance length. Are sentences getting longer? Moving from one word to two is a real marker. Two to three is another. Track it by noticing the longest thing your child said today.

Play complexity. Is the pretend getting more elaborate? More characters, more steps, more invented problems? Play complexity and language complexity move together. When play gets richer, language usually follows.

Communication initiation. Is your child starting interactions more often, even without words? A child who brings you a toy to show you, or who tugs your hand toward the play kitchen, is communicating on purpose. That counts.

If your answer to all four is "no change after 3 months of consistent effort," that is a signal to call an SLP rather than run more of the same at home. Nobody has a clean evidence base for exactly how long to wait before escalating, but most clinical guidance suggests no progress over 3 months in a child under 3 warrants a formal evaluation ^[8].

For families using Little Words between therapy sessions, the app tracks communication patterns over time and can hand parents a cleaner picture of trends that are easy to miss day to day. That data is useful to bring to an SLP appointment.

Are there specific play activities for children who use AAC?

Yes, and it is an underserved area. Most play guidance reads as if speech is the only channel, which leaves AAC families without a map.

The core principle holds: follow the child's lead, create scenes, wait for communication, respond to every attempt. What changes is the setup.

Bring the device into the scene. If your child uses a speech-generating device, it should be physically present and reachable during play, not on a shelf. Model it yourself. Say "more" on the device when you want more pretend tea. Show that the device is for play, more than for requesting snacks.

Core vocabulary matters in play. Words like more, stop, go, mine, help, no, yes, want, like, and finished show up in nearly every play scene. These are core vocabulary, and they are usually the first words programmed on an AAC system for exactly this reason. A child who can say "no" in doctor play is practicing pragmatics as much as vocabulary.

Allow alternative access. Some children use eye gaze, switches, or partner-assisted scanning. Play has to be paced to give the device user time to build a message. The seven-second wait rule matters even more here, and it may need to stretch to fifteen.

Preview familiar scenes. Some AAC users do better knowing what vocabulary a scene will need before it starts. A quick preview ("We're going to play grocery store. Let's find the words for that") can lower the load and raise the child's participation.

For a fuller picture of AAC approaches and devices, the AAC devices article covers the landscape. For children with childhood apraxia of speech, which often runs alongside significant language delay, AAC plus play-based therapy is frequently part of the plan.

Frequently asked questions

What age does symbolic play typically start?

Simple symbolic play, like pretending to drink from an empty cup, usually begins between 12 and 18 months. By 24 months most children pretend with dolls or stuffed animals. Multi-step pretend narratives develop between 30 and 48 months. The American Academy of Pediatrics recommends flagging absent pretend play at 18 months to a pediatrician, especially if it comes with limited speech or pointing.

Does pretend play really help speech development?

Yes. Symbolic play and language share one foundation: representational thinking. Lorraine McCune's research found that pretend play milestones predicted language milestones with high consistency in children 8 to 24 months old. ASHA lists the two as developing together and notes that a delay in one often travels with a delay in the other.

My child won't do pretend play. Should I be worried?

Not automatically, but pay attention. Some children, particularly autistic kids, show different or delayed pretend play timelines. If your child shows no pretend play by 18 to 24 months alongside limited speech and limited joint attention (pointing, showing, following a gaze), bring it up with your pediatrician. An SLP evaluation can clarify whether it is a delay worth treating or a different developmental path.

How is symbolic play different from regular play?

Regular functional play means using objects as designed: rolling a toy car, stacking blocks. Symbolic play means using an object or action to represent something else: the block becomes a car, the banana becomes a phone. That representational substitution is the key cognitive move. It is the same move that makes words work, which is why symbolic play and language development are so tightly linked.

What are the best symbolic play activities for autistic children?

Follow the child's interests instead of imposing a theme. A child obsessed with trains benefits more from train-themed pretend scenes (broken tracks, stuck passengers) than from being pushed toward dolls. Puppet play, small-world sets, and parallel play with narration all lower direct social demands and work well for many autistic kids. A 2020 meta-analysis found naturalistic play-based approaches tied to child interests improved spontaneous language more than adult-directed drill.

Can I use toys we already have, or do I need to buy special ones?

Use what you have. Research suggests less realistic, open-ended props (cardboard boxes, wooden spoons, old hats) often pull out more pretend play than expensive realistic toy sets, because the child has to imagine more. A generic dollhouse and a set of small animals cover most scenes. Neither needs to cost more than $20 to $40.

How long should I do symbolic play with my child each day?

No published study sets an exact daily dose, so any specific number you see elsewhere is someone's opinion. What the research supports is consistency over duration: frequent short sessions (even 10 to 15 minutes daily) where you follow the child's lead and narrate produce more language exposure than occasional long ones. Quality of interaction beats clock time.

Can symbolic play activities help with speech sound problems?

Symbolic play builds language (vocabulary, sentences, communication initiation) more directly than speech sounds. A child with articulation difficulties or apraxia may need specific motor-based speech practice on top of play. That said, play gives a low-pressure context for speech practice, because the child is motivated and not being formally tested. SLPs often embed speech targets inside play scenes for exactly this reason.

Should I correct my child's grammar or pronunciation during play?

Generally no, especially during child-led play. Direct correction tends to shut down a child's willingness to communicate. Use expansions instead: repeat what the child said with one small addition. If the child says "dog go," you say "yes, the dog is going." That confirms you understood, models the correct form, and keeps things positive. Clinicians call it recasting, and it has consistent research support.

What if my child uses echolalia instead of original language during play?

Echolalia during play is not a problem to fix. For many children, scripted phrases from shows, books, or past interactions are a real communication tool and a bridge to more flexible language. If your child quotes a cartoon while playing doctor, they are using representational language in context. Build on it rather than suppress it. The echolalia article on this site covers this in much more depth.

How is symbolic play used in early intervention?

IDEA Part C guarantees free early intervention for children under 3 with developmental delays in every U.S. state. SLPs in early intervention often run sessions in the home and use the child's own toys and routines as the therapy context. Parents get coached in techniques like following the child's lead, parallel talk, and expansions. Play is the delivery vehicle because it is where children under 3 spend most of their waking time.

Are there symbolic play activities specifically for late talkers with small vocabularies?

Yes. For children with fewer than 50 words, the goal is not long pretend sequences but single-act pretend with familiar objects. Feeding a doll, putting a bear to sleep, cooking soup in a pot. Use simple two-word models ("bear eat," "more soup") and wait after each one. Repeating the same simple scene across several days locks in new words far better than constant novelty.

When should I stop worrying and just wait, versus getting an evaluation?

Honest answer: if you are reading this and your gut is already worried, an evaluation is rarely the wrong call, and waiting rarely helps. ASHA recommends evaluation if a child has fewer than 50 words by 24 months, is not combining two words by 24 months, or shows no pretend play by 18 months. Evaluations are diagnostic, not commitments to treatment, and they are free under IDEA for children under 3.

Does online or app-based play support count, or does it need to be in person?

Live video interaction (like online speech therapy with a real SLP) produces language learning in young children, unlike recorded video. Apps can supplement in-person play by building vocabulary, tracking progress, or coaching parents, but they do not replace the contingent, physical experience of play with a caregiver. Think of apps as a tool between sessions, not a swap for face-to-face play and real early intervention.

Sources

1. McCune, L. (1995). A normative study of representational play at the transition to language. Developmental Psychology, 31(2), 198-206.: Pretend play milestones predicted referential speech milestones within weeks in children 8 to 24 months old
2. American Speech-Language-Hearing Association (ASHA), speech and language developmental milestones: ASHA states that pretend play and language develop together, and delays in one often accompany delays in the other; milestones for symbolic play listed by age
3. American Academy of Pediatrics (AAP), developmental surveillance and screening: AAP recommends flagging absent pretend play at 18 months alongside limited speech or pointing; children under 2 learn words from live interaction but show video deficit with recorded video
4. Harris, P.L. (2000). The Work of the Imagination. Blackwell.: Children begin accepting and producing object substitutions in play around the time they reach 50-word vocabularies; toy phone conversations elicit more complex dialogue than face-to-face in many children
5. Hirsh-Pasek, K., Golinkoff, R.M., et al. (2015). Putting education in 'educational' apps. Psychological Science in the Public Interest, 16(1), 3-34.: Repetition across familiar play scripts helps children internalize new vocabulary; narrative play with re-enactment predicts later reading and academic language skills
6. Gentner, D. (1982). Why nouns are learned before verbs. In S. Kuczaj (Ed.), Language development, Vol. 2. Erlbaum.: Action verbs are underrepresented in early vocabularies compared to nouns; play context makes action verbs more salient than static labeling
7. Tomasello, M., & Farrar, M. J. (1986). Joint attention and early language. Child Development, 57(6), 1454-1463.: Children learn more words when adults talk about what the child is already focused on (following attentional lead); labeling a toy the child ignores is significantly less effective
8. U.S. Department of Education, IDEA Part C (Infants and Toddlers with Disabilities), 20 U.S.C. 1431-1444: IDEA Part C guarantees free early intervention services for children under 3 with developmental delays in every U.S. state; SLPs can deliver services in natural environments including the home
9. Sandbank, M., et al. (2020). Project AIM: Autism intervention meta-analysis for studies of young children. Psychological Bulletin, 146(1), 1-29.: Naturalistic developmental behavioral interventions following child interests produced stronger effects on spontaneous language than adult-directed formats, though many studies had weak methods and possible bias, in a meta-analysis of intervention studies in young autistic children
10. Waddington, H., et al. (2020). Play-based interventions for children with language delays. American Journal of Speech-Language Pathology.: Naturalistic play-based interventions produced significant improvements in vocabulary and sentence complexity for children with language delays, with moderate-to-large effect sizes
11. ASHA, Augmentative and Alternative Communication practice portal: AAC users increase device use when communication context is play; core vocabulary words are foundational for AAC programming
12. Centers for Disease Control and Prevention (CDC), Learn the Signs. Act Early. developmental milestones: CDC milestones reference: pretend play with dolls or stuffed animals by 24 months; multi-step pretend sequences by 36 months