AAC for Speech-Language Pathologists

AAC Modeling

We teach by showing. We use AAC to talk.

What is Modeling AAC?

Modeling is a strategy that falls under the broader category of Aided Language Stimulation. This technique draws from the research on natural language development. When an infant is born, we model language, but don’t expect a response for at least a year. That doesn’t stop us from talking to our babies. We have entire conversations. When a baby coos, the parent coos back. We establish a conversational give and take in a relaxed, socially reinforcing manner. What we don’t do is to ask tons of questions and demand responses. We do not drill. When a baby vocalizes, we infer their intent and respond. 

MODEL poster for AAC

There are teaching methodologies that focus on drilling information with lots of repetition.  AAC modeling is quite different. Building language is about building relationships. We don’t use “drill” because we don’t want learning to talk to feel like work. When communication becomes a task demand, we risk an increase in task avoidance behaviors. We don’t want the communication to become associated with feelings of anxiety. 

Modeling can be done with light tech or high tech systems. We use the system to communicate, too. We show our students where vocabulary is located and we activate key words. This helps teach the location and meaning of the icons. We also strive to do this in meaningful communicative contexts.

Instead of drilling picture cards, we find a topic that the student wants to talk about. We might use a little environmental sabotage to give them a reason to initiate communication. And we respond to all of their attempts. Just as during natural language development, we infer intent. Even if the student pushes a button that might appear to be random, we still respond. We learn words by pairing their use with the context and the response they receive. 

Some Keys Practices

  • model one step ahead of the AAC learner. If they use single words, model 2-3 word combinations
  • Give the learner time to respond. Count to 20 and model again
  • Use a least to most prompt hierarchy (see the posters below)
  • Don’t force the learner to respond. If you are modeling, they are learning
  • Model comments, greetings, and questions.  Not just requests

Also be sure to leave words in the same location on the device! We don’t move them around to “test comprehension.” When we learn to communicate, we need to develop consistent motor plans for the production of words. This is how communication becomes faster and more efficient.

Modeling has been the subject of research and has been demonstrated to support language acquisition. In a nutshell, research has shown that modeling can…

  • increase symbol comprehension and production
  • increase responsiveness and use of AAC 
  • facilitate the comprehension of vocabulary items
  •  improve syntactic performance 
  • Increase the consistent production of multi-symbol messages

The evidence shows that modeling works. Use language to teach language.

The typically developing child with reach language mastery around 9-12 years. at twice a week, the alternate symbol user will need 701 years (paraphrase)


Least to Most Prompt Poster

High Contrast Least to Most Poster

Link to MODEL poster

High Contrast MODEL poster

Link to Go the Distance Poster

High Contrast Go the Distance Poster