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

Last updated 2026-07-11

Why do parents and some programs focus so much on nouns?

Nouns are easy to teach, so they get taught first. You hold up a ball and say "ball." The child looks at it, maybe says something close, and everyone feels progress. Objects are concrete, nameable, and easy to drill. Flashcard apps, picture books, and a lot of early-intervention worksheets lean hard on object labels for exactly this reason.

The bias even shows up in how researchers have historically counted vocabulary. Early word inventories often prioritized object nouns because they are easy to measure: you can point to a cup, photograph a cup, and ask a parent whether their child has ever said "cup." Social words like "more," "go," "stop," or "mine" are harder to inventory and easier to miss.

Some older theories also proposed that children naturally acquire nouns first and verbs later, and that this order was fixed. That framing gave cover to programs that loaded up on object labels and deferred everything else. The evidence for that sequence was drawn heavily from English-speaking, object-rich homes, and it did not hold across languages or across children with language delays ^[1].

The noun-first approach has a long institutional history. That does not make it right.

What does the research actually say about noun bias in early language?

The "noun bias" hypothesis, the idea that all children universally acquire nouns before verbs, has taken hits since the 1990s. Kathy Hirsh-Pasek and Roberta Michnick Golinkoff reviewed cross-linguistic data and found that children learning Mandarin, Korean, and several other verb-favoring languages did not show the noun-dominant early lexicons seen in English learners ^[1]. The structure of the input language, not a hard-wired priority for objects, drove the pattern.

More to the point for late talkers: a study by Rescorla found that late talkers had smaller verb vocabularies than age-matched peers, and that verb vocabulary size at age 2 predicted later sentence length better than noun vocabulary alone ^[2]. Knowing fifty object labels does not help a child say "I want more" or "Daddy go," because those utterances need words from categories other than nouns.

The American Speech-Language-Hearing Association (ASHA) recommends that early vocabulary targets include a mix of word classes: nominals (nouns), action words (verbs), modifiers, personal-social words, and function words ^[3]. Leaning on one category at the expense of the others is not consistent with current evidence-based practice.

Noun-heavy training can produce a child who names everything in the room and cannot ask for a single thing they want. That is the whole problem in one sentence.

What word types actually move late talkers toward two-word combinations?

Two-word combinations, the stage most parents are chasing, need at least two different word classes working together. "More juice" is a modifier and a noun. "Dog go" is a noun and a verb. "My ball" is a determiner and a noun. None of those are possible if the child's entire vocabulary is nouns.

Here is a rough breakdown of the word categories that research and clinical practice flag as high-priority for late talkers:

Paul and Norbury emphasize that functional vocabulary, words a child actually needs and wants to use in daily life, generalizes faster than decontextualized noun drilling ^[4]. A child who learns "open" can use it for doors, containers, bags, and tablets. Compare that to learning the word "rhinoceros."

The goal is a vocabulary diverse in word class, more than large in number. A child with 50 words that are all nouns runs a higher risk of staying stuck than a child with 30 words spread across verbs, social words, and a handful of nouns ^[2].

How does verb learning work differently from noun learning, and why is it harder?

Nouns map onto objects that stick around. You can point at a dog and say "dog" three times across three minutes, and the dog sits there obligingly. Verbs map onto events that unfold in time and then vanish. By the time you say "jump," the jump is over.

So verbs need a different teaching strategy. Michael Tomasello's work showed that children learn verbs more reliably when the adult uses the verb at the moment of the action and repeats it across different events ^[5]. Saying "jump" while jumping, while a toy jumps, while another person jumps, and while watching a video of jumping builds a verb the child can generalize. A flashcard of a stick figure jumping does not do the same job.

For late talkers, the input needs to be even more deliberate. Self-talk (narrating what you are doing: "I'm pouring. I'm opening. I'm pushing.") and parallel talk (narrating what the child is doing: "You're pushing. You're eating. You're running.") are two techniques speech-language pathologists use to flood the environment with verbs in context, without requiring the child to say anything back ^[3].

Apraxia is worth flagging here. If a child struggles to produce multisyllabic words or makes inconsistent errors on the same word across attempts, verb learning may be tangled up with a motor speech problem rather than, or alongside, a vocabulary one. A speech-language pathologist can sort that out. There is more on that overlap in our article on apraxia of speech.

What are social words, and why do late talkers need them?

Social or pragmatic words manage interaction instead of labeling things. "More," "no," "help," "mine," "stop," "go," "hi," "bye." Each one earns its keep for a child with a small vocabulary, because a single word can work across dozens of situations.

A child who learns "more" can use it at snack, at the park, in the bathtub, in the middle of a game. That is one word doing the communicative work of many object labels. Same for "help" or "open" or "done."

For children with autism or significant social communication differences, pragmatic words may matter even more than they do for children with a simple language delay. ASHA's guidance on autism communication intervention treats functional communication, the ability to express wants, needs, and social bids, as the primary goal for many children with autism-related speech delays ^[6]. Teaching 40 animal names before a child can say "help" or "no" has the priorities backwards.

There is more on how communication intervention fits autism specifically in our article on autism spectrum speech therapy.

Does this mean parents should stop teaching nouns entirely?

No. Nouns are part of the vocabulary a child needs, and some noun learning happens on its own because the world is full of objects. The problem is exclusivity and proportion, not nouns themselves.

Here is the practical shift. For every noun you practice, target at least one action word and one social word in that same play session. Playing with a toy car? "Car" is a noun, fine. "Go" and "stop" and "crash" are verbs with huge generalization potential. "More" and "my" cover social and possessive meaning that makes the interaction richer.

Speech-language pathologists who follow current evidence aim for what is sometimes called a balanced lexicon. Paul and Norbury's textbook (used widely in graduate SLP programs) describes this as ensuring early vocabulary targets include at least 40 to 50 percent non-noun words for children beginning to combine ^[4]. That is a useful rough benchmark for parents too.

If you are working with a therapist and the homework they send home is nothing but noun flashcards, ask the question: "Are we also targeting verbs and social words?" A good clinician will welcome it.

How does this apply to AAC users who are just starting out?

This is where the noun-only problem shows up most visibly and does the most damage. Many early AAC systems and communication boards, especially homemade ones, get organized around categories of nouns: food, toys, body parts, animals. The logic is that the child needs vocabulary to ask for things, and things are nouns.

But research on AAC intervention keeps landing on the same point: core vocabulary, a small set of high-frequency words that appear across many contexts, should anchor any AAC system. Core vocabulary is dominated by verbs, pronouns, and social words, not nouns. Words like "want," "go," "more," "stop," "help," "I," "you," "it," "do," "can," and "that" turn up in nearly every study of core vocabulary in AAC and are the building blocks of generative language ^[7].

Fringe vocabulary (specific nouns like a favorite food or a family member's name) matters too, but it should supplement a core vocabulary base, not replace it. An AAC device or board that is 90 percent nouns limits a child to requesting objects. It cannot support commenting, protesting, asking questions, greeting, or describing.

If you are just starting to explore AAC, this is one reason working with a speech-language pathologist who specializes in it pays off. See our overview of AAC devices for a starting point on options.

What does 'balanced vocabulary' look like in everyday play?

Theory aside, here is what this looks like at home on a Tuesday afternoon.

You are playing with blocks. The typical noun-only script: "Block. Red block. Big block. Stack the blocks." Fine, but it is heavy on nouns and modifiers with almost no verbs or social words.

A balanced script for the same activity: "Go! Build. More blocks? I want more. My turn. Your turn. It fell. Uh oh. Help? Push. Stack. Done." You just introduced action verbs (build, push, stack, fell), social words (more, my, your, uh oh, done), a request frame (I want more), and a pragmatic signal (help). The nouns "blocks" and "tower" can absolutely live in there too. The difference is distribution, not the absence of nouns.

Naturalistic language intervention approaches, including Hanen's It Takes Two to Talk program and the Milieu Teaching framework studied by Ann Kaiser and colleagues at Vanderbilt, emphasize following the child's lead, commenting on their actions with a variety of word classes, and setting up situations that pull for functional language ^[8]. These approaches have a stronger evidence base than drill-based noun training for late talkers.

One specific move: put a desired item just out of reach or in a container the child cannot open alone. Then wait. The pressure of that moment pulls for "help," "open," "want," and "please" far harder than any flashcard session will.

When should a parent stop doing home strategies and get a professional evaluation?

Home strategies matter. They can also mask a delay long enough to lose the window of fastest neural change, which is why the referral thresholds are worth knowing cold.

The American Academy of Pediatrics and ASHA both publish surveillance and referral benchmarks. By 12 months, a child should have at least one word with meaning. By 18 months, 10 to 20 words is the general expectation, though the range is wide. By 24 months, 50 words and the start of two-word combinations is the standard milestone. A child who misses these does not automatically have a disorder, but they warrant a speech-language pathology evaluation rather than more home practice ^[9].

Wait-and-see is not an evidence-based approach for most late talkers. ASHA's guidance on late language emergence is direct that early identification and intervention produces better outcomes than watchful waiting, particularly for children showing other developmental differences alongside the language delay ^[3].

If cost or access is the obstacle, early intervention services under IDEA Part C (for children under age 3) are free or low-cost in the United States, and you do not need a physician referral to request an evaluation. Your state's early intervention program must evaluate within 45 calendar days of your referral ^[10]. There is more on how this works in our article on early intervention.

For children over 3, the school district's Child Find obligation under IDEA Part B requires a free evaluation if there is reason to suspect a disability affecting educational performance ^[10].

Is there a difference between how this applies to late talkers versus children with autism?

Late talker is a descriptive term for a child whose language is delayed without a known cause. Children with autism often have language delays too, but the mechanism can be different, and the communication targets may need even more weight on social-pragmatic functions.

For a child with autism, the "why" of communication is as important as the "what." Barry Prizant's SCERTS work shows that many autistic children have particular difficulty with communicative intentionality and joint attention, the foundational skills that support vocabulary learning in any word class ^[11]. A child who is not yet reliably using eye contact, pointing, or showing to share experience may need a stretch of joint attention and social engagement work before or alongside vocabulary intervention.

The core problem with noun-only instruction applies just as much here, arguably more. A child who can label 100 objects but cannot say "stop" or "help" or "I want" is in a functionally hard spot. The noun-object flashcard approach can also reinforce rote labeling rather than real communicative use, which runs against what most autism communication research recommends ^[6].

Some autistic children who are not developing speech may also be candidates for AAC. The evidence is clear that AAC does not suppress speech development and may support it ^[7]. If verbal speech has not emerged by age 4 to 5 despite intervention, an AAC evaluation is warranted rather than something to put off.

Some children with language delays also show echolalia, repeating words or phrases heard elsewhere. That is worth understanding on its own terms, especially in the context of autism. See our pieces on echolalia and echolalia meaning for detail.

What should a speech therapy home program actually include?

If you are doing home practice between sessions, or waiting for an evaluation and want to do something useful in the meantime, here is a framework built on current evidence rather than commercial flashcard logic.

Target words from at least three categories every week: one or two new verbs (go, eat, put, give, want), one or two social or function words (more, help, no, mine, all done), and one or two nouns that genuinely matter to your specific child, not arbitrary word-list vocabulary.

Use responsive interaction. That means commenting and narrating instead of quizzing. "You're pushing the car. Go! It went fast" lands differently than "What is that? Say car. Say car." The second approach creates pressure and can cut down on communication attempts in children who are already struggling.

Engineer your environment to create communication opportunities. Offer two choices held up where the child can see them. Put desired items in sight but out of reach. Pause expectantly in familiar routines. These moves come from milieu teaching and Hanen research and have good evidence behind them for late talkers ^[8].

If you are working with a therapist, ask them straight: "What word classes are we targeting this month?" and "How much of the target vocabulary is verbs and social words versus nouns?" The answers tell you fast whether the program is balanced.

Families who want a more structured home practice option can also look at tools like Little Words, which includes a quiz to identify where a child is and what word categories to prioritize, built on the same balanced-lexicon principles described here.

For children whose delays may have a motor speech component, a separate evaluation for childhood apraxia of speech may be warranted, since the intervention for apraxia looks quite different from vocabulary-based approaches.

And if you are wondering whether online or telehealth speech therapy is realistic for your family, see our guide to online speech therapy for what the evidence says about effectiveness.

What is the most common mistake parents make when trying to build vocabulary at home?

Asking too many questions is probably the single most common one. "What's that? What color is it? What is the dog doing?" feels like engagement, and it is, but it puts the child in the position of being quizzed rather than being supported in spontaneous communication.

For a child who already struggles to produce words, a question carries implicit demand. The child knows an answer is expected. That pressure can actually cut the number of communication attempts they make, especially if they have had experiences of failing to answer correctly.

Comments require nothing back. "Oh, a dog. Big dog. Dog's running" gives language exposure without demand. Over time, hearing varied vocabulary in low-pressure moments builds a richer internal lexicon than quiz-style talk.

The second most common mistake: practicing words in one context and expecting them to generalize. A child who has drilled "apple" from a flashcard may not say "apple" when handed a real apple at snack, because the flashcard and the snack do not feel connected. Teaching new words across multiple real contexts, multiple people, multiple settings, is what produces functional use.

Third, and directly relevant to this whole piece: choosing vocabulary based on what is easy to teach rather than what the child actually needs. A child who loves trains does not need "rhinoceros" or "helicopter" right now. They need "go," "stop," "crash," "more," "my train," and "help" for the moments the track falls apart. Motivation-matched vocabulary that spans word classes will beat an arbitrary noun list every time.

Frequently asked questions

At what age should I worry if my child only says nouns?

If your child is 18 to 24 months old and their vocabulary is mostly or entirely nouns, raise it with a speech-language pathologist. By 24 months, children are typically combining words, which requires verbs and social words alongside nouns. If the vocabulary is large but noun-dominated and no combinations are showing up, a formal evaluation is appropriate rather than waiting.

How many verbs should a toddler know before they start combining words?

Children typically begin combining words reliably once they have around 50 words total, including a meaningful share of verbs and social words. Rescorla's work on late talkers found that verb vocabulary at 24 months predicted later language outcomes. There is no hard cutoff for verb count, but fewer than 5 to 10 action words at 24 months alongside limited nouns is a clinical concern.

What are the best first words to teach a late talker?

Focus on words that are functional, motivating, and span word classes. High-priority first words include action words (go, eat, open, want, help), social words (more, no, mine, hi, bye, done), a few key pronouns (I, me, you), and a small number of nouns that genuinely excite your specific child. The most useful early words work across many situations rather than naming a single object.

Can noun-only flashcard apps hurt a late talker's progress?

They are unlikely to cause harm, but they can eat practice time that would do more good spent on balanced, functional vocabulary. Time is not unlimited, and children with language delays benefit most from targeted input. If flashcard apps are the main vocabulary tool at home, the child ends up with a noun-heavy lexicon that does not support word combination. Supplement with verb-rich naturalistic play.

My child knows 50 words but they're all nouns and they're not combining. What now?

This is exactly the pattern this article describes. A child with 50 nouns has a large reference vocabulary but lacks the functional word classes needed to build sentences. A speech-language pathology evaluation is warranted to assess vocabulary composition, comprehension, and word-combining readiness. In the meantime, shift home practice heavily toward action words and social words instead of adding more nouns.

Does teaching too many nouns cause autism or make autism worse?

No. Teaching nouns does not cause autism or worsen it. The concern is narrower: a noun-only approach can limit the communication skills of any child with a language delay, including autistic children, because it does not address the full range of communicative functions. It is a teaching limitation, not a medical risk. Autism is not caused by how vocabulary is taught.

What is core vocabulary in AAC and why does it matter?

Core vocabulary is a small set of high-frequency words that appear across many communication contexts. Studies of English speakers consistently find that roughly 200 to 400 words account for the large majority of spoken language, and these words are mostly verbs, pronouns, prepositions, and social words. For AAC users, placing core vocabulary in a consistent, accessible spot on a device is standard practice and supports generative, more than labeling, communication.

Is the Hanen program scientifically supported for late talkers?

Yes, with some nuance. The Hanen It Takes Two to Talk program has been evaluated in randomized and quasi-experimental studies. Systematic reviews and ASHA's evidence maps rate it as having adequate to good evidence for improving parent-child interaction quality and child vocabulary, particularly for late talkers without additional diagnoses. Effect sizes vary. It is not a replacement for direct speech therapy when significant delays are present, but it is a well-supported parent-coaching approach.

What is the difference between a late talker and a child with a language disorder?

A late talker is a child whose expressive language is delayed but who has age-appropriate comprehension, play, and social skills, with no identified cause. Around 70 to 80 percent of late talkers catch up by age 3 to 4 without formal intervention, though the remaining 20 to 30 percent go on to have persistent language disorders. A language disorder involves a more pervasive, lasting difficulty with language. Only a speech-language pathology evaluation can tell them apart.

Should I use sign language or gestures to teach verbs to a late talker?

Yes. Sign support helps, particularly for action words that are otherwise fleeting in time. Research on total communication and augmentative communication in toddlers consistently shows that using gestures alongside speech does not suppress verbal development and can speed it up. Simple signs for go, stop, more, help, and eat give a child a motor and auditory version of the word at once. Always pair the sign with the spoken word.

What should I ask a speech-language pathologist about my child's vocabulary targets?

Ask specifically: what proportion of the target vocabulary is verbs, social words, and function words versus nouns? Ask how words are being selected: driven by the child's interests and daily needs, or by a standardized word list? Ask what the functional goal is this quarter: commenting, requesting, protesting, or greeting. A good clinician answers all three clearly and ties each target to a functional communication outcome.

My pediatrician said to wait and see until age 3. Is that still the standard advice?

It is no longer consistent with current AAP or ASHA guidance, which both recommend referral for speech-language evaluation rather than watchful waiting when a child is significantly behind milestones. Under the AAP's 2020 surveillance guidance, any concern about expressive or receptive language at 18 or 24 months should prompt a referral rather than a six-month delay. You have the right to request an early intervention evaluation directly, without a physician referral, under IDEA Part C.

Do bilingual children have a different noun-verb balance when they start talking?

Bilingual children's vocabulary spans two languages and cannot be accurately judged by looking at one language alone. Total vocabulary across both languages should be used for comparison to monolingual norms. Research on bilingual toddlers does not suggest a different noun-verb ratio is normal for them; the balanced-lexicon principle applies across languages. If a bilingual child shows a noun-dominant, non-combining vocabulary in both languages, that pattern warrants the same evaluation it would in a monolingual child.

What is the fastest way to introduce more verbs in daily life at home?

Self-talk and parallel talk are the fastest entry points. Self-talk means narrating your own actions aloud: 'I'm pouring. I'm opening. I'm pushing.' Parallel talk means narrating the child's actions: 'You're running. You're eating. You dropped it.' Neither requires the child to respond. Doing this for even 10 to 15 minutes of intentional play per day sharply increases the density of verb input a child receives, which research on Milieu Teaching and responsive interaction supports.

Sources

1. Hirsh-Pasek K & Golinkoff RM, 'The Origins of Grammar' (MIT Press, 1996); cross-linguistic noun bias review: The noun-bias in early vocabulary is not universal across languages; children learning verb-prominent languages do not show the same noun-dominated early lexicons seen in English learners, indicating language input structure rather than fixed cognition drives the pattern.
2. Rescorla L, 'Language and Reading Outcomes to Age 9 in Late-Talking Toddlers', Journal of Speech, Language, and Hearing Research, 2002: Verb vocabulary size at age 2 was a stronger predictor of later sentence length and language outcomes in late talkers than noun vocabulary size alone.
3. American Speech-Language-Hearing Association (ASHA), 'Late Language Emergence' practice portal: ASHA recommends that early vocabulary intervention targets include a mix of word classes: nominals, action words, modifiers, personal-social words, and function words, and that early identification and intervention produces better outcomes than watchful waiting.
4. Paul R & Norbury C, 'Language Disorders from Infancy through Adolescence', 4th ed., Elsevier, 2012: Functional vocabulary, words a child actually needs in daily life, produces faster generalization than decontextualized noun drilling; early vocabulary targets should include at least 40 to 50 percent non-noun words for children beginning to combine.
5. Tomasello M & Kruger AC, 'Joint Attention on Actions: Acquiring Verbs in Ostensive and Non-Ostensive Contexts', Journal of Child Language, 1992: Children learn verbs more reliably when adults use the verb at the moment of the action and repeat it across multiple different events, rather than in static or decontextualized presentation.
6. ASHA, 'Autism Spectrum Disorder: Interventions for Communication' practice portal: ASHA notes that functional communication, the ability to express wants, needs, and social bids, is the primary communication goal for many children with autism-related speech delays, and that AAC does not impede verbal speech development.
7. Beukelman DR & Mirenda P, 'Augmentative and Alternative Communication', 4th ed., Brookes Publishing, 2013; core vocabulary research summary: Core vocabulary in AAC systems is dominated by verbs, pronouns, and social words, not nouns; roughly 200 to 400 high-frequency words account for the large majority of spoken language across contexts, and these should anchor any AAC system.
8. Kaiser AP & Roberts MY, 'Parent-Implemented Enhanced Milieu Teaching with Preschool Children Who Have Intellectual Disabilities', Journal of Speech, Language, and Hearing Research, 2013: Milieu Teaching and responsive interaction approaches, including following the child's lead and creating communication opportunities, have a stronger evidence base for vocabulary growth in late talkers than drill-based noun training.
9. American Academy of Pediatrics, 'Developmental Surveillance and Screening Policy Statement', Pediatrics, 2020: The AAP recommends referral for speech-language evaluation rather than watchful waiting when a child is significantly behind language milestones at 18 or 24 months; a child should have 10 to 20 words by 18 months and 50 words with early two-word combinations by 24 months.
10. U.S. Department of Education, IDEA Part C Early Intervention regulations, 34 CFR Part 303: Under IDEA Part C, states must evaluate a child for early intervention eligibility within 45 calendar days of referral, and services are free or low-cost to families; no physician referral is required to request an evaluation.
11. Prizant BM et al., 'The SCERTS Model: A Comprehensive Educational Approach for Children with Autism Spectrum Disorders', Brookes Publishing, 2006: Many autistic children have particular difficulty with communicative intentionality and joint attention, foundational skills that support vocabulary learning; the SCERTS model emphasizes social engagement and functional communication before or alongside vocabulary intervention.
12. Hanen Centre, 'It Takes Two to Talk' program overview and evidence summary: The Hanen It Takes Two to Talk program has adequate to good evidence for improving parent-child interaction quality and child vocabulary in late talkers, with naturalistic play and responsive interaction techniques as core components.