Your Child Talking More Is Not the Good News You Think It Is

In an 11-week experiment comparing three groups of children, the amount of talk rose the same in all of them. What diverged was the architecture of the story

Introduction: What Do We Look At to Feel Reassured?

The first thing a parent worried about language development checks is usually quantity. How many words the child uses, how many sentences they string together, whether they talk less than their peers. At a preschool conference, “their speech has really picked up lately” passes as good news.

But is the amount of talk really an index of language development?

Behind this study sits a somewhat heavy statistic. The team notes that the proportion of preschool and early-elementary children showing speech and language difficulties has risen sharply in recent years, and that the absolute number of children clinically diagnosed with language disorders has increased more than sixfold over the past two decades. Meanwhile time spent reading has fallen and screen exposure has risen.

So many intervention programs have appeared. The most widely used is dialogic reading — not simply reading aloud, but asking questions and expanding on the child’s answers. The effect is verified. Vocabulary rises; so does story comprehension.

But the team leaves one reservation here.

These approaches give children little opportunity to generate or transform content themselves. Spontaneity is reduced as a result, and creative engagement falls off.

Book reading is, in the end, an activity of following a story that is already finished. It is not the experience of owning a story. So what about an activity where the child builds the story themselves — that is, role play? And how does an adult joining that play differ from leaving children to play among themselves?

The Core Question

A Russian team (Monash University and Moscow State University) put an approach called PlayWorld to the test. The question:

“Does reading a fairy tale together and having an adult step inside that world to play build narrative ability more than children playing freely on their own?”

Specifically, they split it three ways:

1. Does the length of the story a child produces increase?
2. Does the story’s macrostructure — its architecture — improve?
3. Does the story’s microstructure — its vocabulary and grammar — grow more sophisticated?

Separating those three is the heart of the study. And the result shows that the three do not move together at all.

Macrostructure and Microstructure: Two Different Abilities

Before the results, these two terms need pinning down. When a child looks at pictures and builds a story, there are two separable layers to evaluate.

Macrostructure — Is the Story Story-Shaped?

Whether the goal – attempt – outcome structure is present, and whether the child grasps causality.

• Weak macrostructure: “There’s a cat. There’s a bird. There’s a tree.” (listing pictures)
• Intact macrostructure: “The cat climbed the tree to eat the bird’s egg, but the bird came and chased it off.” (intention → action → resolution)

Microstructure — Are the Sentences Sophisticated?

Whether vocabulary is accurate and varied, and whether grammar and syntax are error-free and diverse. Saying the same thing, “the cat went up and the bird came” and “the hungry cat climbed the trunk, and the startled mother bird flew in” score differently on microstructure.

What matters is that these are separate abilities. The power to frame a story’s skeleton and the power to clothe that skeleton in precise language grow independently. And this study found that the two grow under different conditions.

What the Interventions Were

PlayWorld Group: The Adult Takes a Part

PlayWorld is an educational approach rooted in Vygotsky’s cultural-historical theory. Its core is that the adult participates as a genuine play partner, not as observer or director.

A session ran like this:

1. Fairy tale reading. The adult reads a short chapter of a tale aloud. Sessions 1–11 used Pinocchio; sessions 12–22 used The Wizard of Oz.
2. Talking about the story. Questions like “do you like this story?”, “who do you like best?”, “why do you think Pinocchio did that?”
3. Passing through the ‘portal.’ The children cross an imaginary gateway into the story world, choosing from colored cloths, ribbons, and hats to suit their part.
4. Playing together. The children assign roles, and the adult takes a part as well. The adult’s job is to sustain the plot and the richness of the play.

One condition here matters. There was no requirement to follow the original. Pinocchio could go to Oz; the ending could change. The tale was a starting point, not a correct answer. The play materials, too, were things that could become anything — boxes, string, plastic bowls, chairs, cushions, cloth, sticks, colored paper, pebbles.

Free Play Group: The Adult Only Helps It Start

The same materials were provided. The adult helped the children get the play going — settling on a theme, dividing up roles — and after that,

did not intervene.

The adult only watched to see that the children played safely. This group exists to separate “the effect of play itself” from “the effect of an adult being in it.” Play duration, materials, and group size were identical to PlayWorld.

Control Group: As Usual

No special play sessions; ordinary preschool life.

Design and Participants

• Population: children aged 5–6 attending public preschools in Moscow
• Numbers: 90 initially → 67 in the final analysis (PlayWorld 22, free play 25, control 20)
• Attrition: 23 children (25.6%) were excluded — 13 missing post-test data, 10 unable to attend at least half the sessions. A measles outbreak in Moscow preschools in December 2024 was the main cause
• Duration: 11 weeks, twice weekly, 20–30 minutes per session, 22 sessions total
• Facilitators: three trained researchers with psychology and education backgrounds. Each researcher ran one PlayWorld subgroup and one free-play subgroup simultaneously — a design that keeps individual facilitator skill from mixing into the results
• Instrument: MAIN (Multilingual Assessment Instrument for Narratives), the Russian-validated narrative measure. Six pictures composing three episodes are shown, with the prompt “tell me what happened here, as much as you can”
• Testing points: pre-test one week before the intervention, post-test within 10 days of its end
• Blinding: the experts who transcribed and scored the recorded narratives did not know the study’s purpose. One person, trained directly by MAIN’s developers, scored everything

At baseline the three groups did not differ statistically in narrative ability (word count p=.494, macrostructure p=.563, microstructure p=.31). They started level.

Results: The Amount of Talk Showed No Difference at All

The most important result in this study may sit in the item that looks least interesting.

Word Count: All Three Groups Rose, and There Was No Difference

The difference in improvement across the three groups was not statistically significant (χ²(2)=2.03, p=.362). The control group that did nothing actually rose the most — though that difference isn’t significant either.

Speech rate was the same story. Everyone sped up slightly over time (p=.056, trend level), with no group difference.

That is: the stories produced by children who spent 11 weeks playing in a fairy-tale world with an adult were exactly as long as those produced by children who did nothing.

Macrostructure: The Two Groups With Play Pulled Ahead

The time × intervention interaction was significant (F(2,51)=8.049, p<.001, η²=.141).

• PlayWorld: significantly improved (p=.044)
• Free play: significantly improved (p=.028)
• Control: significantly declined (p=.032)

In the post-test between-group comparison, both PlayWorld and free play were significantly higher than control (p<.001 and p=.006 respectively).

Microstructure: Here the Two Play Groups Diverged

The interaction was again significant (F(2,51)=8.808, p<.001, η²=.148). Both play groups improved significantly against themselves (p=.003 each).

But in the comparison against control, only PlayWorld was significantly ahead (p<.001). The free play–control difference was not significant. The team puts it this way:

The development at the lexical and grammatical level seen in free role play did not differ significantly from natural development without any intervention.

What Has to Be Said Honestly: The Decisive Comparison Stopped at “Trend”

Here we have to pause. The temptation to summarize this study as “an adult has to play along for it to work” is strong, but the data does not go that far.

The head-to-head comparison of PlayWorld against free play was not statistically significant. Macrostructure p=.062, microstructure p=.052 — both narrowly failing to clear the threshold of statistical significance (p<.05), at trend level.

So the conclusion that “vocabulary and grammar grow only when an adult plays along” was not established by directly comparing the two conditions. It is an interpretation drawn from the indirect pattern that PlayWorld beat control while free play did not. The direction points consistently toward PlayWorld, but with a sample of 67 you cannot call it conclusively proven.

One more thing. The control group’s macrostructure score falling 1.32 points, from 5.11 to 3.79, in 11 weeks is something the team could not cleanly explain either.

Without targeted linguistic stimulation, preschoolers may not have sufficient opportunity to practice coherent, structured speech within a general educational setting. Small regressions in skill development are not uncommon in early childhood and may reflect fluctuation in the natural developmental trajectory.

Honestly, that reads more like conjecture than explanation. The fact that the picture sets differed between pre-test and post-test (a nest scene before, baby goats after) may also be a variable. That said, the picture change applied equally to all three groups while only the play groups rose, which test difficulty alone does not explain.

Why Do a Story’s Skeleton and Its Flesh Grow Under Different Conditions?

The most interesting point in this study is the contrast: macrostructure grew from play alone, but microstructure grew only with an adult present. The team’s explanation is persuasive.

Role Play Is Itself a Story

The reason macrostructure grew from play alone is simple. Role play is narrative construction by nature. When children play, two layers of dialogue run at once.

• In-role talk: “I am the evil wizard! I’ve come to get you!”
• Talk about the play: “No, you died before. Be someone else now.”

The second is called metacommunication. As they play, children continuously negotiate who does what and what happens next. That is exactly the training for assembling goal–attempt–outcome. It happens perfectly well among children with no adult present.

But Sophisticated Language Has to Come From Somewhere

Microstructure is a different matter. A room of children playing alone contains only children’s vocabulary and grammar. The team argues that adult participation brings three things.

First, culturally accumulated language. The adult pulls a role’s intonation, varied vocabulary, and above all the literary language of the fairy tale into the play. A wizard who says “my blood runs cold” instead of “scary!” now exists inside the game.

Second, the power to hold an imaginary situation open longer. This part is especially striking.

Adult participation sustains the imaginary situation for longer. In free role play, by contrast, children quickly ‘fall out’ of the imaginary situation and shift to impulsive, improvised action.

Left to themselves, children easily move on after five minutes to “let’s stop this and play tag.” The play ends before the story deepens enough for sophisticated language to emerge. The adult acts as the pillar keeping the story from collapsing.

Third, emotion experienced together. The team invokes the Russian concept of perezhivanie — intensely lived emotional experience.

Common experience, jointly felt emotion, and dramatic conflict between role characters stimulate a shared play idea and shared perezhivanie. This means the children are inside the problem situation, engaged both emotionally and cognitively.

An adult and a child who have read a tale together and entered that world together tense at the same things and feel relief at the same things. Speech that comes out of that engagement does not float free.

The Zone of Proximal Development

All of these explanations converge on one Vygotskian concept. Rich, well-organized play demands a great deal of a child — planning, inventing an interesting story, choosing suitable props, arranging the play space, resolving conflict with playmates, sustaining dramatic tension, developing a plot.

These abilities sit in the region where a child cannot yet manage alone but can manage with help — the zone of proximal development (ZPD). Hence the need for adult guidance.

Practical Takeaways

The Amount of Talk Is Not an Index of Development

Word counts rose similarly in all three groups here. The group with no intervention rose the most in raw numbers. A child talking more and a child’s language growing are different stories.

If you want to check at home, ask this way. Ask your child to tell you about yesterday, and listen for whether “why” and “so” are in the story. “I went to preschool. I ate. I came home” and “a friend sat in my spot and I was upset, but the teacher made another spot so it was okay” have similar word counts and are completely different stories.

Playing With Them and Watching Them Are Different

The common wisdom that leaving children to play alone is good for autonomy is half right. In this study, children playing among themselves did grow the architecture of their stories. That is a real effect.

But for vocabulary and grammar to grow too, adult language had to be inside the story. What matters here is that the adult’s role was not teacher and not referee, but a play partner holding a part. Not the adult who says “that’s not how you do it,” but the adult who takes the wizard and delivers the lines in earnest.

Tell Them the Story Can Change

Following the original was optional in this study. Re-enacting Pinocchio after reading Pinocchio was not the goal. The tale was raw material.

If, after reading a picture book at home, you move to “what do you think happened next?” or “what would you have done if you were the hero?”, the child shifts from the seat of consuming a story to the seat of making one. The difference between following a finished story and owning a story was this study’s starting point.

Twenty-Two Times, 20–30 Minutes Each

Not a burdensome scale. Twice a week for 11 weeks, 20–30 minutes a session. That it isn’t a daily obligation matters practically.

Limitations of the Study

The constraints the team named directly:

• Sample loss. 25.6% dropped from 90 to 67. The measles outbreak was unavoidable, but it lowers confidence in the statistical conclusions. The three groups’ pre-test scores were, however, confirmed statistically equivalent.
• Insufficient control variables. Language fluency, cognitive ability, and bilingual status were not measured alongside. Other factors affecting narrative ability may be mixed into the results.
• No follow-up. Only the effect immediately after the intervention was measured. Whether the improvement remains months later is unknown.

Some additions. As noted, the direct comparison of PlayWorld and free play was not significant, and the cause of the control group’s decline was not established. The only prior experiment on PlayWorld was a single quasi-experimental study of 20 children in 2005. The accumulated evidence for this approach is, in other words, thin.

And this study was conducted in Moscow public preschools with a Russian-language instrument. Whether it applies as-is to Korean children and the Korean language needs separate verification.

Closing Thoughts

The question this study poses comes down to one thing. What are we looking at when we look at a child’s language?

Whether talk has increased, whether sentences have lengthened, whether the vocabulary beats their peers’ — these are easy to count. So we look at them. But in this experiment, the amount of talk failed entirely to distinguish the children who were played with from the children who were not.

What diverged over 11 weeks was not how much the child talked, but whether the child’s story had a skeleton.

And that skeleton grew from children playing among themselves — while clothing it in precise language seemed to require an adult who came inside the story with them. The research did not prove it conclusively, but the direction the data pointed was consistent.

Between watching from beside your child with your phone in hand, and taking the part of the evil wizard and delivering the lines in earnest, there is probably about that much difference.

Source: Veresov, N. N., Veraksa, A. N., & Plotnikova, V. A. (2025). Using PlayWorld to promote narrative development: Evidence from a double-blind control experiment. Psychology in Russia: State of the Art, 18(3). https://doi.org/10.11621/pir.2025.0309