Every family starting AAC hears the same advice: model, model, model. Use the device yourself. Show your child how it works by using it in real conversations.
The advice is sound. Modeling is the single most effective strategy for teaching AAC use. But it raises a practical question that research has only recently started to answer: how much modeling is actually enough?
What Is AAC Modeling?
Before we get into dosage, let's be clear about what modeling means. AAC modeling (also called aided language stimulation or aided language input) is when a communication partner uses the AAC system while talking to the AAC user. You tap the symbols on the device as you speak, showing the child what the words look like in their system.
For example, while playing with blocks, you might tap "want" + "more" + "block" on the AAC device while saying "Do you want more blocks?"
You're not asking the child to repeat what you did. You're showing them how the system works by using it yourself, the same way speaking parents teach spoken language by talking around their children for years before expecting full sentences back.
What the Research Says About Dosage
Binger and Light (2007)
Binger and Light conducted one of the earliest studies specifically looking at the amount of modeling needed to produce results. They studied aided AAC modeling with preschoolers who had developmental disabilities and found that children began producing multi-symbol messages after relatively modest amounts of modeling.
Their key finding: children showed improvements in multi-symbol message production after as few as 10 to 15 modeling sessions, each lasting about 15 to 20 minutes. The sessions were structured around play activities and book reading.
This was significant because it suggested that families didn't need to model for hours every day to see progress. Consistent, focused modeling during natural activities was enough to produce measurable change.
Sennott, Light, and McNaughton (2016)
Sennott and colleagues reviewed the research on AAC modeling interventions and found consistent evidence that aided language input improves both comprehension and production of AAC-based communication. Their review identified several important patterns:
- Modeling was effective across different age groups, from toddlers to adults
- Both high-tech and low-tech AAC systems benefited from modeling
- Effects appeared even when modeling was delivered by parents with no professional training
- Consistent modeling over weeks produced better outcomes than intensive but short-term modeling
They emphasized that the research supports AAC modeling as an evidence-based practice, but noted that the "optimal dose" varies by individual and has not been precisely defined.
Kent-Walsh, Murza, Malani, and Binger (2015)
This meta-analysis looked at communication partner instruction (training parents and others to model AAC) and found large effect sizes for both the partners' modeling behavior and the children's communication outcomes. Importantly, they found that when partners were taught specific modeling techniques, their children's communication improved significantly compared to control groups.
The practical takeaway: it's not just about quantity. Learning to model well matters as much as how often you do it.
Biggs, Carter, and Gilson (2018)
Biggs and colleagues conducted a systematic review of aided AAC modeling interventions and found that modeling was effective across a range of partners and settings. Their review included evidence that non-specialist partners, including peers, could deliver effective AAC modeling. This finding matters because it shows that modeling doesn't have to come exclusively from trained therapists. The more people modeling, the more input the AAC user receives.
Quality vs. Quantity
Research consistently suggests that how you model matters more than how long you spend doing it. Here's what quality modeling looks like:
Match the child's level (plus one)
If the child is at the single-word stage, model one to two word combinations using core words. If they're combining two symbols, model three-word phrases. Modeling at a level just slightly above the child's current ability gives them a reachable next step.
Model in context
Modeling works best during activities the child is already engaged in and interested in. Playtime, meals, bath time, getting dressed. These daily routines provide repeated, predictable opportunities for the same vocabulary.
Model without expecting imitation
This is hard for parents, but it's critical. Modeling is input, not a test. If every time you model a word you then look at the child expectantly and wait for them to copy you, you're turning communication into a performance task. Model, keep talking, and let the child respond (or not) on their own timeline.
Use the actual AAC system
Pointing to symbols on the child's actual device is more effective than just saying the words. The child needs to see where words live in their system. Abstract pointing or verbal-only models don't teach navigation.
A Practical Modeling Schedule for Families
Based on the research, here's a realistic schedule that balances effectiveness with the reality of family life:
| Time of Day | Activity | Modeling Focus | Duration |
|---|---|---|---|
| Morning | Getting dressed | "want," "help," "on," "off," clothing words | 3 to 5 min |
| Breakfast | Eating | "want," "more," "all done," "eat," "drink," food words | 5 to 10 min |
| Playtime | Free play or structured activity | Core words + activity-specific vocabulary | 10 to 15 min |
| Lunch | Eating | Same as breakfast | 5 to 10 min |
| Afternoon | Outside, errands, or quiet time | "go," "stop," "look," "want," location words | 5 to 10 min |
| Dinner | Eating | Same as breakfast | 5 to 10 min |
| Bedtime | Books, bath, bedtime routine | "more," "all done," "want," story vocabulary | 5 to 10 min |
That adds up to roughly 40 to 70 minutes per day, but notice that none of these are separate "AAC sessions." They're all embedded in activities that are already happening. You're not adding time to your day. You're adding modeling to your existing routine.
Start Small
If 40 to 70 minutes feels overwhelming, start with one routine. Pick the activity where your child is most engaged and motivated, maybe mealtime or playtime, and focus all your modeling energy there. Once that feels natural, add another routine. Then another.
The research from Binger and Light (2007) suggests that even 15 to 20 minutes of focused daily modeling can produce results. You do not have to do everything at once.
Why Imperfect Modeling Works
Here's the most important thing: you will mess up. You'll tap the wrong symbol. You'll forget to model during lunch because you're tired. You'll go three days without touching the device because life happened. If the pressure of consistent modeling starts to feel unsustainable, you're not alone. Caregiver burnout is common in AAC families, and it's worth reading about before it catches up with you.
That's fine. Here's why:
Children are resilient learners
Hearing children learn spoken language from imperfect models. Parents use baby talk, make grammatical errors, speak in fragments, and mumble. Children figure it out anyway. AAC users are equally capable of learning from imperfect input.
Consistency beats perfection
Modeling three words imperfectly every single day is more valuable than modeling perfectly for one hour once a week. Language learning depends on repeated exposure over time. Frequency and consistency matter far more than flawlessness.
Your modeling teaches more than vocabulary
Even when you hit the wrong button, your child is learning something. They're learning that the device is for communication. They're learning that adults use it too. They're learning that mistakes happen and communication continues anyway. These are all essential lessons.
Research confirms this
Sennott et al. (2016) found that parent-delivered modeling, even without extensive training, produced positive outcomes. Parents who received just basic instruction in modeling techniques still saw improvements in their children's communication. The bar for "good enough" modeling is much lower than most families fear.
Common Modeling Mistakes (and Quick Fixes)
Modeling too many words at once. If you're tapping out full sentences while your child is at the one-word stage, scale back. Match their level plus one word.
Only modeling during therapy. One hour of SLP-led modeling per week cannot replace daily exposure from communication partners. Modeling needs to happen throughout the day, across settings.
Asking "What does this say?" instead of modeling. Testing is not teaching. Use the device to communicate, and your child will learn by watching.
Giving up after a few weeks. Spoken language takes years to develop. AAC language development follows a similar timeline. Expect months, not days, before you see consistent spontaneous use.
The Bottom Line
How much modeling is enough? The honest answer is: more than none, and less than you think.
The research supports 15 to 20 minutes of focused daily modeling as a starting point. Embedding that modeling into existing routines makes it sustainable. And imperfect modeling, delivered consistently, is better than perfect modeling that only happens once in a while.
Your child needs to see their AAC system being used by the people they care about. That's you. You don't need to be perfect. You just need to be present.
Download SabiKo free and start modeling today.
References
- Binger, C., & Light, J. (2007). The effect of aided AAC modeling on the expression of multi-symbol messages by preschoolers who use AAC. Augmentative and Alternative Communication, 23(1), 30-43.
- Sennott, S.C., Light, J.C., & McNaughton, D. (2016). AAC modeling intervention research review. Research and Practice for Persons with Severe Disabilities, 41(2), 101-115.
- Kent-Walsh, J., Murza, K.A., Malani, M.D., & Binger, C. (2015). Effects of communication partner instruction on the communication of individuals using AAC: A meta-analysis. Augmentative and Alternative Communication, 31(4), 271-284.
- Biggs, E.E., Carter, E.W., & Gilson, C.B. (2018). Systematic review of interventions involving aided AAC modeling for children with complex communication needs. American Journal of Speech-Language Pathology, 27(1), 1-12.