Modeling has been researched to be one of the most effective strategies and the easiest to implement, to facilitate communication. 

Click here to know more about What is modelling or Aided Language Input/ Stimulation? 

How do I do modeling ?

Whenever you speak to the child, use Avaz simultaneously and point to the words you are speaking. Make sure that you have the attention of the child while using Avaz. 

  • Attention: Gain the attention of the child 
  • Speak: Speak out the message while pointing to it on Avaz
  • Point: Point to the icons on Avaz while forming a message to communicate 
  • Continue: Use this method constantly in every interaction 

Watch this video of Nimisha Muthaiah modeling the use of Avaz with her student. Note the following in the video: 

  1. She sits besides the child and keeps Avaz in front of him, so that he is able to see the words she is tapping and learn what each image stands for
  2. As she navigates through the folders, she describes what she is tapping in the app - e.g. "Ok, I'm going into the Toys folder" as she taps the Toys folder.
  3. She is using Choice making effectively for different activities.
  4. As she speaks each option aloud,  she simultaneously taps it or points to it in Avaz
  5. She reinforces his selection immediately with the actual object - e.g playing bubbles or ball or singing his favourite song immediately after the request is made by the child. This provides motivation. 

Courtesy and thanks to Nimisha for letting us use this video:  

Nimisha Muttiah-Fonseka 

Ph.D., CCC-SLPSpeech-Language Pathologist/Speech & Language Therapist

Senior Lecturer II, Department of Disability Studies, 

Faculty of Medicine, University of Kelaniya

PO box 6, Thalagolla 

Road,Ragama, Sri Lanka. 

How many words do I model - single words or full sentences?

You don't need to model every single word you speak. You model only the CORE words of the sentence. You can model at a level that is one or two words above the level of your child. E.g. if the child communicates in single words, you may model 2-3 words while speaking to the child. 

For example, when you want to tell the child "Let's eat some cookies" :

If the child is new to AAC, you can start by modeling one word - tap the word "eat" on Avaz, while speaking this entire sentence aloud. 

If the child is already at a single word level, you can model two words - tap the words "eat" and "cookie", while speaking the entire sentence aloud. 

Do's and Don'ts 

What to speak while modelling one word at at a time, e.g.  cake

mmm.. cakeWhere's cake on the device?
Let's eat cakeShow me cake 
It looks like you want cakeFind cake , press the switch
You can tell me if you want cakeTell me cake on the device
I think you want to eat cakeI won't give you cake if you don't show it on the device
I think you're trying to tell me you want cakeTap cake on the device. I know you can do it. 
Maybe you want cake ?You showed me cake yesterday, you're just stubborn today and not showing me
I wonder if you want cake?

Courtesy: Carole Zangari, PrAACtical AAC. 


Note the usage of the words MAYBE ..  , I WONDER .., I THINK ... Using these words ensures that the caregiver is not arriving at a conclusion but instead making an assumption about the child's intended meaning and so is asking if the assumption is correct. 

Important to bear in mind

  • Always model without expectation of a response
  • Model everyday, regularly and consistently across different contexts and environments.
  • If the child does not respond, continue to model anyway, even after repeated attempts
  • If the child attempts to communicate through any means , always respond to the attempt- see Response strategies
  • Start modelling  at ‘1 word’ level to begin with and progress to using 2 or 3 words per sentence.
  • Always model at a level that is 1 or 2 words more than the child's current level. E.g. if the child is communicating at a one-word level (e.g. park) , model two words. (e.g. go park)
  • Don't tap every word that you speak. Point only to the key words (Core words) in the sentence.
  • It may take upto a YEAR of modelling continuously before the child responds spontaneously
  • Communication should not feel like work for the child and for "testing" the child's knowledge
  • Remember that it is okay to make mistakes while modelling to the child, as this will also teach the child repair strategies if they press a word by mistake.

Tips for Modeling

  • Use a slow pace to talk while pointing to vocabulary on Avaz
  • Emphasize on the specific word(s) being taught
  • Continue modelling even if the communicator is not actively watching.
  • Model the same words multiple times.
  • The modelled words need not be grammatically correct. E.g. "play blocks", "go car" etc. 
  • Keep the device at your child’s eye level while you are using it. This helps him learn the vocabulary, its symbol, location and navigation.
  • No matter what your child selects on the device, RESPOND. Provide natural consequences to whatever your child says, even if it seems totally unrelated and seems like a mistake. This is called "attributing meaning" to the child's attempt to communicate. Click here to know more about How to attribute meaning to the child's attempt to communicate?
  • Example: 
    • Parent: I’m cooking potatoes for dinner.
    • Child: taps 'TV'
    • Parent: It’s not TV time yet. We can watch TV after dinner.
  • After your child says something, expand on it. Determine your child’s current language abilities and model one or two words beyond that level. If the child says “cookies”, model “want cookies”, “I want cookies” and so on.
  • Use props such as finger puppets and pencil toppers to capture your child’s interest initially, while pointing to the pictures.

  • Modelling helps the child 

Strategies for modeling

Click here to know more about Modelling Strategies

Related articles:

See video of SLP Amanda Hartman using AAC for Bugs activity