
Last updated 2026-07-11
TL;DR
Teaching preference communication starts with noticing what a child already does to show wants, then building a ladder from gestures and eye gaze up through pictures, AAC, and words. It takes months, not days, and the research is clear: responding to every attempt, no matter how small, is what drives progress. No single method works for every child.
What does 'communicating preferences' actually mean for young children?
Preference communication is any behavior a child uses to show what they want, like, or don't want. That includes a reach, a gaze, a push-away, a point, a picture exchange, a word, or a symbol on a device. All of these count. The goal is not to get a child saying "I want the red one" as fast as possible. The goal is to build a reliable, respectful signal between your child and the people around them.
The American Speech-Language-Hearing Association defines communication as "any act by which one person gives or receives information from another person" and explicitly includes nonsymbolic forms like eye contact, body movement, and vocalizations as real communicative acts [1]. That framing matters because a lot of parents don't realize their child is already communicating preferences. They're just doing it in ways adults aren't always trained to notice.
For late talkers and autistic children, preference communication is often where therapy starts. Before you can teach vocabulary, you need a child who believes their signals will be responded to. That trust is the actual foundation.
Why is learning to express preferences so hard for some kids?
Several things can make preference communication difficult. Motor planning issues (like in apraxia of speech) can mean a child knows what they want but can't organize the movements to say it [2]. Sensory processing differences can make the environment so overwhelming that forming a preference signal gets lost in the noise. Social motivation differences in autistic children can mean the usual "talk to get the thing" loop is less compelling, not because the child doesn't have preferences, but because social interaction itself isn't a natural reward.
There's also a learned helplessness problem. When adults consistently anticipate a child's needs before the child signals, the child learns that signaling is unnecessary. This is well-intentioned but counterproductive. A study in the Journal of Applied Behavior Analysis found that creating communication opportunities by withholding preferred items slightly (a temptation setup) produced more spontaneous communication attempts than free-access conditions [3].
And then there's the response gap. If a child's early attempts to signal preferences (a reach, a look, a vocalization) get ignored or misread too often, those attempts fade out. The child stops trying. This is why responsiveness from caregivers is treated in the research as the single highest-leverage variable in early communication development.
What are the stages of preference communication, from earliest to most advanced?
Think of preference communication as a ladder. Every child starts somewhere on it, and the job is to move up one rung at a time. Skipping rungs doesn't work.
| Stage | What it looks like | Typical age range (approximate) |
|---|---|---|
| Pre-intentional | Reflexive reactions (crying, stiffening, calming) | Birth to ~8 months |
| Intentional non-symbolic | Eye gaze, reach, push-away, body movement toward or away | ~8-12 months |
| Conventional gestures | Point, reach with open hand, head nod/shake | ~10-18 months |
| Symbolic: objects/pictures | Handing an object, exchanging a picture (like PECS) | ~12 months onward |
| Symbolic: single words or AAC symbols | Saying or selecting one word/symbol for a want | ~12-24 months, varies widely |
| Multi-word combinations | "More juice," "want cookie," "no blue one" | ~24+ months, varies widely |
| Full sentence preference statements | "I want the green cup, not the red one" | ~36+ months, varies widely |
Age ranges here are approximate norms from ASHA developmental milestones and should not be used diagnostically [1]. Children with speech and language disorders may reach these stages much later, and that's exactly what therapy addresses.
Here's the key idea from this ladder: you never skip a stage because you're in a hurry. If a child doesn't have reliable pointing yet, introducing a voice output device before they have the concept of "I signal, you respond" is going to be hard. Some children do learn symbolic communication before conventional gesture, but that's the exception, and even then they need the response-loop concept to be solid.
How do you actually teach a child to show what they want? (The core techniques)
Start where the child is. Watch them for a week and catalog every single behavior they use to approach or avoid things. That list is your baseline. Every behavior on it is a communicative act you can build on.
Respond to everything, immediately. When a child looks at the crackers, say "crackers? you want crackers" and give them one. You're teaching them that their signal (the look) has power. Do this hundreds of times. It feels repetitive. It is. That repetition is how neural pathways form.
Create opportunities without being manipulative. Put the preferred toy slightly out of reach. Pause before giving a drink. Open a container and wait instead of pouring automatically. Speech-language pathologists call this "communicative temptation" or "sabotage" and it's a standard technique for increasing spontaneous communication [3]. You're not withholding to punish. You're creating a reason to communicate.
Offer real choices with real options. Hold up two objects. "Apple or cracker?" Wait at least 5 to 10 seconds. If the child looks at one, reaches for one, vocalizes, or does anything that might be a response, treat it as a choice and act on it. Over time you can make the wait slightly longer and the required signal slightly more explicit.
Match the modality to the child. Some children will point naturally. Others do better handing you an object. Others benefit from picture exchange. Autistic children with limited speech often make faster progress with AAC devices than with verbal-only approaches [4]. The goal is communication, not speech specifically.
Model, model, model. Use the same system you're asking the child to use. If you're teaching PECS, exchange pictures yourself. If you're using a speech-generating device, touch the symbols when you talk. This is called aided language stimulation, and a 2018 systematic review in the American Journal of Speech-Language Pathology found it consistently increased symbol use in children with complex communication needs [5].
Keep the environment predictable enough to create contrast. If a child always gets apple juice, they have no reason to prefer it over something else. Rotating options gives preferences something to push against.
How long does it take for a child to learn to communicate preferences?
Honest answer: it depends on where they start, how consistently adults respond, and what underlying factors are affecting communication. There is no good population-level data on this specific timeline, partly because preference communication is a component of broader language development rather than a separate milestone that gets tracked in studies.
What the research does say: children who receive early intervention services before age 3 show significantly better language outcomes than those who start later [12]. A 2021 meta-analysis in Pediatrics found that early speech-language intervention had a moderate to large effect on expressive language outcomes, with the largest effects when intervention started before 24 months [6].
In clinical practice, speech-language pathologists often see meaningful gains in intentional preference communication within 8 to 16 weeks of consistent, structured practice at home and in therapy. But "meaningful gains" might mean moving from pre-intentional reactions to reliable eye gaze, not from silence to sentences. Celebrate the rung, more than the top of the ladder.
One honest caveat: consistency matters more than intensity. Two hours of focused, responsive interaction per day, seven days a week, beats one heavy therapy session that doesn't carry into daily routines. Parents and caregivers are the intervention, most of the day.
Does AAC help children communicate preferences earlier?
Yes, and it doesn't delay speech. This is one of the most stubborn myths in parent circles and the research directly contradicts it. A 2012 systematic review in the journal Augmentative and Alternative Communication found no evidence that AAC suppresses speech development and some evidence it supports it [4].
AAC gives children a way to signal preferences before their speech motor system is ready. A child who can touch a symbol for "more" or "stop" has preference communication. That success builds the very motivation that drives further communication, including speech.
AAC devices range from low-tech (a laminated board with two pictures) to high-tech (a full speech-generating device with hundreds of symbols). You don't need a $6,000 device to start. A paper choice board with two photos of preferred items is a legitimate AAC system. Start there.
For autistic children specifically, a 2017 review in Autism found that speech-generating devices produced faster acquisition of functional communication compared to picture exchange in some populations, but both outperformed verbal-only approaches for children with minimal speech [7]. The right answer depends on the individual child, which is why an autism spectrum speech therapy evaluation matters.
What should parents do at home between therapy sessions?
Carry-over is where progress actually happens. Research consistently shows that parent-implemented strategies produce better outcomes than clinic-only therapy. The Hanen Centre's "It Takes Two to Talk" program, one of the most studied parent-implemented approaches, showed significant gains in child language when parents learned to follow the child's lead, wait for communication, and respond consistently [8].
Practical things to do every day:
Turn off the background noise sometimes. A noisy environment makes it harder for a child to attend, form a preference, and signal it.
Pause before you act. You know your child's routines. That knowledge is a gift, and it is also quietly reducing their need to communicate. Pausing for three seconds before pouring the drink creates space for them to signal.
Label preferences aloud when you see them. "You moved away from the loud toy. You don't like that one." You're giving language to what they're already doing. This is called mapping, and it speeds up vocabulary development.
Use two-choice offers consistently, instead of open-ended "what do you want?" questions. Open-ended questions require the child to generate the options and the signal. Two-choice offers only require a selection. That's a much more reachable task for a child who is still building the communication ladder.
If you're working with a speech therapist, ask them to write down the three specific techniques they want you to use at home that week. Verbal instructions fade. Written ones stick.
How do you teach a child to say 'no' and to reject things they don't want?
Teaching "no" is as important as teaching "yes." Children who can only signal wants but can't reject unwanted things are half-equipped. And practically, a child who can't signal "no" reliably will often escalate to crying, hitting, or self-injurious behavior because those are the only tools that work.
Start by honoring proto-rejections. If a child turns their head away from a food, that's a "no." Name it. "You turned away. No broccoli." Remove the item. You're teaching them that a small signal works so they don't need a big one.
Then build toward a more explicit signal. A head shake, a "no" picture on a choice board, a "no" symbol on a device. Teach it the same way you teach any other signal: model it, create opportunities to use it, and respond every time.
One nuance worth catching: some autistic children use echolalia as a form of communication, including to indicate preferences and rejections. A child who quotes a movie line when offered something unfamiliar may be communicating "this doesn't fit my expectation" in their own way. Those attempts deserve the same respect and response as a conventional "no." Understanding echolalia meaning can help parents decode these signals correctly.
When should you be worried and ask for a professional evaluation?
The American Academy of Pediatrics recommends developmental surveillance at every well-child visit and formal developmental screening at 9, 18, and 30 months, with autism-specific screening at 18 and 24 months [9]. If you're concerned about communication at any age, that concern alone is enough reason to ask for a referral.
Specific flags that warrant prompt evaluation: a child who has lost communicative behaviors they previously had (regression), a child who by 12 months has no back-and-forth social communication, a child who by 18 months uses no words or clear symbolic communication of any kind, or a child at any age whose communication is not meeting their needs and causing frustration.
You can request a free evaluation through your state's early intervention program if your child is under 3. For children 3 and older, the public school system is required to evaluate and provide services under IDEA (Individuals with Disabilities Education Act, 20 U.S.C. § 1400 et seq.) at no cost to families [10]. You do not need a diagnosis first. You just need to make the request in writing.
If you want to supplement or bridge while waiting for services, online speech therapy is a legitimate option. Telehealth speech-language pathology is recognized by ASHA and has been shown to produce comparable outcomes to in-person therapy for many communication targets [11].
How does preference communication develop differently in autistic children?
Autistic children often have strong preferences and clear internal experiences of liking and disliking things. The challenge is rarely the preference itself. The challenge is the signaling system.
Many autistic children show preferences through approach and avoidance behaviors long before they develop a conventional signal like pointing or words. They might stim more around preferred items, move toward them without reaching, or go quiet and still around disliked things. These are real preference signals. Adults who learn to read them build the trust that makes more explicit communication possible later.
The joint attention piece is worth understanding. Joint attention (looking at an object, then at a person, then back at the object, to share the experience) develops differently in many autistic children, and it's part of how preferences usually get communicated socially. A child without reliable joint attention can still communicate preferences very effectively through direct reaching, picture exchange, or device use. Joint attention is not a prerequisite.
For autistic children who use echolalia, scripted language, or idiosyncratic communication, the core principle doesn't change: respond to every signal, build on what's already there, and don't require the signal to look neurotypical before honoring it. There's more on this in our section on autism spectrum speech therapy.
Tools like the Little Words app are designed around this kind of flexible, multimodal communication. If you want a quick sense of whether an app-based approach fits your child's current profile, take the quiz at littlewords.ai/start to see what level of support makes sense.
What's a realistic progress plan for the first six months?
Months 1-2: Observation and foundation. Catalog every communicative behavior the child already shows. Get a baseline. Start responding to every signal within two seconds, every time. Introduce two-choice offers at meals and play. If the child is not yet with a therapist, get on the waitlist and start home techniques now.
Months 3-4: Build one explicit signal. Pick one. A reach-and-look combination, a head shake for "no," a picture card for one highly preferred item. Use it in every relevant situation. Don't add a second signal until the first is reliable (meaning the child uses it unprompted at least 3 to 4 times across different contexts). This is slower than it sounds. That's fine.
Months 5-6: Expand and generalize. Add one more signal. Work on using the signals with different people (other parent, grandparent, teacher). Generalization is the hard part. A child who signs "more" only for their primary caregiver in the kitchen has learned a conditioned response, not a communication skill. They need to learn the signal means the same thing everywhere.
Along the way, keep a simple log. Nothing elaborate. Date, what signal the child used, whether they used it spontaneously or with a prompt. Looking back over six months of even rough notes will show you the progress that's invisible when you're in it day to day.
If you want a structured way to support this at home, Little Words offers guided activities built around this kind of step-by-step preference communication building. You can find out if it fits your child's stage at littlewords.ai/start.
Frequently asked questions
At what age should a child be able to say what they want?
By 12 months most children point to things they want, and by 18 months typically say 10 to 20 words, including some to request preferences. By 24 months, two-word combinations like "more juice" are expected. But range is wide, and any child not meeting those milestones deserves an evaluation, more than a wait-and-see response. ASHA milestone charts are free at asha.org.
What if my child screams or melts down instead of asking for things?
Screaming and meltdowns are communication. They mean the child's other signals haven't worked, or they don't yet have another signal. The response is to honor the underlying need while also teaching a smaller signal that works earlier in the escalation. Over time, if the smaller signal consistently gets results, the meltdown frequency drops. This is a standard behavior-language approach in speech and applied behavior analysis work.
Does using pictures or AAC mean my child will never talk?
No. Multiple systematic reviews, including a 2012 review in Augmentative and Alternative Communication, found no evidence that AAC suppresses speech and some evidence it supports it. AAC gives children a working communication system while their speech develops. Many children who start with picture exchange or speech-generating devices add verbal speech alongside it or transition to speech over time.
How do I know if my child is choosing a picture on purpose or just grabbing randomly?
Look for consistency across multiple trials, across different settings, and across different people. A child making intentional choices usually shows some differentiation: they take the cookie picture when cookies are available but not when only vegetables are. Speed of response and whether they look at the picture before grabbing it also matter. If you're unsure, a speech-language pathologist can do a preference assessment.
What is a preference assessment in speech therapy?
A preference assessment is a structured observation where a therapist systematically presents items and watches how the child responds. More detailed versions present multiple items in rotating pairs. The result is a rank-ordered list of what the child finds most motivating. That list drives everything: which items are used in teaching, which rewards are offered, and what vocabulary to target first.
Can a child communicate preferences without pointing?
Yes. Pointing is one signal among many. Eye gaze, reaching, moving toward or away from objects, handing items to adults, vocalizing while orienting toward something, and selecting pictures or symbols are all valid preference communication. Many autistic children develop strong, reliable preference communication using systems that don't rely on pointing at all.
How do I teach my child to choose between two things?
Hold up both items clearly within the child's visual field. Name them. Then wait silently for 5 to 10 seconds. Honor any signal: a look, a lean, a reach. Give the item they seemed to move toward. Repeat across many, many trials. Over weeks, the signal usually becomes more deliberate because the child learns it works. Keep choices real, not hypothetical, at first.
What if my child just picks the same thing every time?
That's information, not a problem. A consistent preference is still a preference being communicated. Work with it. Use the highly preferred item as a reward for trying to communicate about less-preferred items. Gradually introduce novelty alongside the familiar favorite. Forced variety before trust is built usually backfires. Start where the child is, always.
Is it okay to honor every preference, even unhealthy ones?
In early communication training, yes, mostly. The goal in the early stages is to build the signal-response loop, and that requires you to honor signals. A child who asks for cookies and gets rejected will learn that asking doesn't work. Once communication is more established, you can introduce natural limits while still validating the preference itself. "You want cookies. We're having dinner first. Then cookies." The preference is heard. The answer is sometimes no.
How do I get my child's school to support preference communication the same way I do at home?
Put it in writing. If your child has an IEP (Individualized Education Program), the communication goals, strategies, and response protocols should be listed there. Ask the team to document the specific signals your child uses and how adults should respond. Under IDEA, the IEP must include a statement of measurable annual goals and services needed to meet them, which can include communication support strategies.
What's the difference between a late talker and a child who needs AAC long-term?
Late talkers are children who are slow to develop verbal speech but have no other obvious developmental differences and often catch up. Children who benefit from long-term AAC typically have motor, cognitive, or neurological factors affecting speech production or language more broadly. The distinction matters for planning but not for starting: both groups benefit from early communication support. A speech-language pathologist makes this determination through evaluation, not age alone.
How important is following the child's lead in teaching preferences?
Very. Child-led interaction is one of the strongest predictors of communication growth. When adults follow the child's attention rather than redirecting it, vocabulary acquisition is faster and communication attempts are more frequent. The Hanen Centre's research on parent-implemented naturalistic approaches consistently shows this effect. Following the lead doesn't mean no structure. It means the structure starts from what the child is already attending to.
What's the role of motivation in preference communication?
Motivation is the engine. A child who doesn't care about the item being offered has no reason to communicate for it. This is why preference assessments matter: you need to know what the child actually wants. Teaching communication with neutral or low-value items produces slow progress. Teaching with highly preferred items produces fast progress. Find the things your child genuinely wants and make those the currency of every teaching interaction.
Sources
- American Speech-Language-Hearing Association (ASHA), Communication Milestones: ASHA defines communication as any act giving or receiving information and includes nonsymbolic forms like eye contact, body movement, and vocalizations as real communicative acts
- ASHA, Childhood Apraxia of Speech practice portal: Childhood apraxia of speech involves difficulty planning and programming the movements needed for speech
- Journal of Applied Behavior Analysis, communicative temptation / establishing operations research: Creating communication opportunities by slightly withholding preferred items (temptation setup) produces more spontaneous communication attempts than free-access conditions
- Augmentative and Alternative Communication, Millar et al. systematic review on AAC and speech: A 2012 systematic review found no evidence that AAC suppresses speech development and some evidence it supports it
- American Journal of Speech-Language Pathology, aided language stimulation systematic review: A 2018 systematic review found aided language stimulation consistently increased symbol use in children with complex communication needs
- Pediatrics, meta-analysis of early speech-language intervention outcomes: A 2021 meta-analysis found early speech-language intervention had moderate to large effect sizes on expressive language, with largest effects when started before 24 months
- Autism (journal), review of AAC and speech-generating devices in minimally verbal autistic children: A 2017 review found speech-generating devices produced faster acquisition of functional communication compared to picture exchange in some autistic populations with minimal speech
- Hanen Centre, It Takes Two to Talk program research summary: Parent-implemented It Takes Two to Talk program showed significant gains in child language when parents learned to follow the child's lead, wait for communication, and respond consistently
- American Academy of Pediatrics (AAP), developmental surveillance and screening policy: AAP recommends formal developmental screening at 9, 18, and 30 months with autism-specific screening at 18 and 24 months
- U.S. Department of Education, IDEA Individuals with Disabilities Education Act, 20 U.S.C. § 1400: Under IDEA, public schools are required to evaluate and provide services at no cost to families; families do not need a diagnosis first and must make the request in writing
- ASHA, Telepractice practice portal: ASHA recognizes telehealth speech-language pathology as producing comparable outcomes to in-person therapy for many communication targets
- ASHA, Early Intervention practice portal: Children who receive early intervention services before age 3 show significantly better language outcomes than those who start later
