A small new study asks whether 15 minutes of self-guided practice a few times a week can change how caregivers talk on their child’s AAC app

If you’ve ever handed a caregiver an AAC device and said “just model on it as much as you can,” you already know the gap between that advice and what actually happens at snack time. Finding the right word on a 100-plus-icon grid, three steps deep, while your child is losing interest — that takes more than good intentions. It takes fluency. And fluency, like any motor skill, is built through practice.

A new single-subject study by Elena Dukhovny and Clara Ackerman, published in Perspectives of the ASHA Special Interest Groups (2025), tested a simple idea: what if caregivers practiced their child’s actual AAC app on their own, in short private sessions, before trying to model with their child at all?

The problem the study is solving

Aided language modeling (ALM) — caregivers and other partners using the child’s own AAC system to demonstrate language, not just prompt the child to use it — is one of the better-supported strategies in AAC intervention. It’s linked to gains in vocabulary, turn-taking, and utterance length. But most caregiver-coaching research has quietly skipped over a basic prerequisite: before a caregiver can model language on the device, they need to be able to find the words on the device, quickly, without stopping to hunt through menus.

Most coaching research has also tended to use simplified, single-tap practice grids built for the study rather than children’s real, often large-vocabulary displays. That makes the coaching easier to evaluate, but it sidesteps the actual skill caregivers need: navigating a multi-step, densely populated grid in real time, the way their child’s app is actually organized.

What the researchers did

Three caregivers — Larry (father to Orlando), Raina (mother to Anton), and Maya (mother to Patrick) — took part. Each child used a different real-world AAC app (LAMP Words for Life or Proloquo2Go), with grid sizes ranging from 30 to 84 icons per screen. Nothing about the children’s displays was simplified or modified for the study.

The protocol had three parts:

• One brief consultation. A licensed SLP met with each caregiver for 30–45 minutes to help them choose a starter set of words based on the child’s interests and everyday activities, and to set a realistic practice schedule.
• Independent practice. Caregivers practiced producing at least 10 self-chosen words per session, four 15-minute sessions a week — alone, without their child, purely to build their own motor fluency on the app.
• No further coaching. After that single consultation, caregivers ran the entire protocol themselves. No ongoing therapist involvement was required.

Researchers then filmed caregivers playing and interacting with their children for 20 minutes at a time, before and after the practice period, and measured three things in the caregivers’ aided (device-based) language: total words produced, different words produced, and the number of multiword combinations.

What they found

Two of three caregivers showed real gains. Larry and Maya each showed large-to-very-large increases in both the total number of words and the variety of different words they produced on the device during play — changes that held up statistically and, for Maya, persisted into a maintenance check weeks later.

One caregiver didn’t change — and that may not be a problem. Raina, who already had the highest word output at baseline, showed no measurable increase. The authors suggest this may reflect a ceiling effect rather than a failure of the protocol: she may already have been producing about as much aided language as was useful in that interaction, especially since AAC guidance also emphasizes waiting for the child, not just modeling more.

Caregivers only used a small fraction of what they’d practiced. On average, caregivers’ practiced words made up somewhere between 9% and 14% of what they actually produced during play with their child. That’s a useful reality check: this protocol didn’t work by caregivers reciting a rehearsed script. It worked — where it worked — by building underlying comfort and speed with the device that carried over into spontaneous, real interaction.

Multiword combinations didn’t reliably improve. Producing more single words came more easily than producing more multiword phrases like “you play blocks.” That’s a meaningful limitation: fluency with individual words is a foundation, not a guarantee of longer combinations.

Caregivers liked it. In follow-up surveys, participants reported feeling more comfortable with the device, more confident helping their child make longer phrases, and said they’d recommend the approach to other families.

What this means in practice

Separate “getting fluent” from “modeling with my child.” The protocol’s core insight is almost mundane: caregivers practiced the device by themselves, with no expectation of a perfect interaction, before trying to model in the moment with their child. That removes a lot of pressure and lets the motor learning happen on its own track.

Fifteen minutes, a few times a week, is a realistic dose. This wasn’t hours of therapist-led coaching — it was one consultation plus caregiver-driven practice on their own schedule. For SLPs balancing large caseloads, that’s a meaningfully lower-cost way to support a real skill gap.

Word choice matters, but perfection doesn’t. Caregivers used relatively few of their specifically rehearsed words during actual play. The goal isn’t a caregiver who has memorized a script — it’s a caregiver who no longer has to hunt for the grid.

This addresses fluency, not the full picture of ALM. The protocol didn’t teach recasting, expectant waiting, or other well-established aided-modeling strategies — and multiword combinations didn’t reliably increase. Fluency with word-finding is a foundation to build on, best paired with the broader ALM coaching families may already be receiving.

An honest caveat

This was a very small study: three caregivers, no control group, and results that varied by caregiver in ways the authors themselves flag as needing more research (baseline word output, choice of activity, and the child’s own language level are all plausible factors). All three families were also receiving other community-based services at the same time, which the single-subject design accounts for but doesn’t fully rule out as a contributor. This is best read as promising early evidence for a low-cost, easy-to-replicate idea — not a proven protocol.

Still, the underlying logic is hard to argue with: caregivers can’t model fluent language on a device they aren’t yet fluent in themselves. Building that fluency deliberately, in caregivers’ own time, may be a small, practical piece worth adding to AAC support.

Sources

Dukhovny, E., & Ackerman, C. (2025). Effect of practice protocol on caregiver fluency with augmentative and alternative communication applications. Perspectives of the ASHA Special Interest Groups, 10, 1982–1996. https://doi.org/10.1044/2025_PERSP-25-00122

Biggs, E. E., Carter, E. W., & Gilson, C. B. (2019). A scoping review of the involvement of children’s communication partners in aided augmentative and alternative communication modeling interventions. American Journal of Speech-Language Pathology, 28(2), 743–758. https://doi.org/10.1044/2018_AJSLP-18-0024

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. https://doi.org/10.3109/07434618.2015.1052153

Moorcroft, A., Scarinci, N., & Meyer, C. (2019). Speech pathologist perspectives on the acceptance versus rejection or abandonment of AAC systems for children with complex communication needs. Augmentative and Alternative Communication, 35(3), 193–204. https://doi.org/10.1080/07434618.2019.1609577

O’Neill, T., & Wilkinson, K. M. (2020). Preliminary investigation of the perspectives of parents of children with cerebral palsy on the supports, challenges, and realities of integrating augmentative and alternative communication into everyday life. American Journal of Speech-Language Pathology, 29(1), 238–254. https://doi.org/10.1044/2019_AJSLP-19-00103