What is "Beginning AAC"?

In May this year David Niemeijer from AssistiveWare and I did a presentation at the AGOSCI Conference on the Do's and Don'ts of Implementing Real Communication Through AAC. We created a poster to accompany that blog post - and this is one of the follow ups to the poster.


One of the things I do every 6 months (or so), is I download all the AAC apps I can find from the App store, evaluate them and then add them to my list of AAC apps (Updated to add: the AAC apps list has been discontinued).

Every time I do this update, one or more of the new apps describes itself as a "beginning AAC app" or an app for "beginning communicators".  I frequently give these apps, along with the apps that describe themselves as "for the lowest functioning" one star - the lowest rating I give.  This sometimes results in an email from the app developer, questioning my rating of their app.  I'd like to explain why I rate them like this.

The phrase "Beginning AAC" in this context, often seems to me a euphemism for poor practice in AAC.  These apps often only have a small number of words; the vocabulary has a heavy emphasis on nouns; there is no symbol system included and the user has to import all their own images; the vocabulary is arranged in categories; and sometimes there are also limited options for expansion or customisation.  The app is often accompanied by phrases like "small number of symbols, as needed by beginning communicators" or "photograph based to simplify for early AAC users".

So much of what is contained in these apps is NOT beginning AAC - but AAC that doesn't promote language and communication development.  The statements in the app descriptions feed into this poor practice - and live in a world where the phrase "presume competence" has never been heard.

Having said all of that, what DO we know about beginning AAC?


There is no evidence that the often referred to "symbol hierarchy" has any relevance in AAC implementation - or even that it operates as some people think it do (and as I was introduced to it in my speech pathology degree nearly 30 years ago).

Romski and Sevcik (2005) refer to this as one of the myths of "Augmentative Communication and Early Intervention". They summarise this myth with the statement "during early phases of development, it may not matter if the child uses abstract or iconic symbols because to the child they all function the same." Porter and Burkhart (2010) greatly expand on this, putting the case very strongly in terms of us needing to do aided language stimulation to help an individual learn what a symbol means. DaFonte et al (2008) agree that "there does not seem to be a hierarchy of aided-visual symbols" and add that "experience plays a significant role in learning aided-visual symbols and generalizing their usage".

So - to go back the topic of beginning AAC there is absolutely no need to limit ourselves to an AAC system that can manage objects or photographs when making AAC decisions - and those AAC apps that state that there is are subscribing to a myth.

Amount of Vocabulary

At the ISAAC Conference last year in Lisbon, Professor Pat Mirenda gave a pre-conference workshop on "Autism Spectrum Disorder and AAC". During this workshop she reviewed the current state of what we know about AAC as it related to individuals with ASD.  One of her lines from her handout really rang true for me - and not just for individuals with ASD:

"WE USED TO THINK: Start with just a few (4-6) picture symbols and add a few more at a time, as the student with ASD shows that he or she can communicate appropriately with them usually by requesting

Now we think: Really? Where is the research that defends this practice?

This is certainly not how other kids learn new words and acquire language."

Mirenda, 2014

For individuals to learn language, we need to provide not just a few picture symbols - but a wide range of symbols that represent a robust vocabulary that supports them to learn how to put words together, supports them to contribute in every situation and supports them to develop into an autonomous communicator.  This vocabulary needs to consist of a range of parts of speech - they need adjectives, verbs, adverbs, pronouns, prepositions, conjunctions, interjections, determiners and even some nouns (but go easy on the nouns).  Try using a well designed core vocabulary system or Pragmatically Organised Dynamic Display (PODD)

To once more return to the topic of beginning AAC,  there is no evidence to support starting with only a few symbols and those AAC apps that say there is are adding to yet another myth.

So how does this turn into autonomous communication?

For any individual to learn how to use their well designed AAC system with robust vocabulary represented by picture symbols we need to make sure that we introduce AAC in a way that helps them to become an autonomous communicator.  This can happen with a range of strategies - but it has to include aided language stimulation (Goossens', Crain & Elder, 1992).  By using the individual's AAC system to initiate, respond, negate, continue, greet, wrap-up and all the other conversational turns we take, we show them how they can do it.  By using their AAC system for real reasons, we show them what the symbols mean. We might need to do this for some time before the individual starts doing it (remember - we talk to infants for around 18 months before they start speaking) - but aided language stimulation is absolutely part of the road to good communication.  It has been shown to increase receptive vocabulary - which totally makes sense (Dada & Alant, 2009) as well turn taking (Beck et al, 2009) and other communication skills.  And in a cyclical pattern that validates this - unless you have a good robust vocabulary you can't do aided language stimulation.

So - for any of those AAC apps that claim they are for a beginning AAC user we need to try having a conversation with them.  Can you have a chat with a partner or friend? Can you ask a question at school? Can you change topic? Can you keep this going through multiple communication turns?  If you can't use it yourself to talk with, how can you do aided language stimulation?  And how do we expect a beginning communicator to communicate with something a competent communicator can't use?

So what is beginning AAC?

Interestingly, good beginning AAC looks an awful lot like later AAC.  Every individual with complex communication needs requires a comprehensive vocabulary that lets them say what they want to say, wherever they want to say it, to whoever they want to say it. The difference is that in beginning AAC, everyone around the individual with complex communication needs to use the system, as much as possible, to demonstrate how the system can be used and to teach them the meaning of the symbols.  As times goes on the individual with complex communication needs begins using the system themselves - and this use increases.  Without these components of: 1. a robust vocabulary; 2. represented by symbols;  and, 3. lots of aided language input we are not doing the best we currently know how by each and every beginning communicator.  This may be different from what you believe - but if you think that beginning AAC involves only a few photographs or objects of reference ask yourself the same question Professor Mirenda asked at ISAAC - "Really? Where is the research that defends this practice?"

PS If you're really interested in how I rate AAC apps - it's at the bottom of every page of my AAC apps lists (Updated to add - the AAC apps list has been discontinued).


  • Beck,  A.  R.,  Stoner,  J.  B.,  &  Dennis,  M.  L.  (2009).    An  investigation  of  Aided  Language  Stimulation:  Does  it  increase  AAC  use  with  adults  with  developmental  disabilities  and  complex  communication  needs?    Augmentative  and  Alternative  Communication,  25 (1),  pp.  42–54.
  • Dada, S. & Alant, E. (2009). The impact of an aided language stimulation program on the receptive language abilities of children with little or no functional speech. American Journal of Speech Language Pathology, 18(1), 50-64.Da Fonte, M. A., Thurber, M., Chae, S., & Lloyd, L. L. (2008, November). Is there a hierarchy of aided visual symbols? A review of the literature. American Speech Language Hearing Association (ASHA) Convention: Celebrating the Winds of Change. Chicago, IL.
  • Goossens’, C., Crain, S., & Elder, P. (1992). Engineering the preschool environment for interactive, symbolic communication.  Birmingham, AL: Southeast Augmentative Communication Conference Publications.
  • Mirenda, P. (2014, ). Autism Spectrum Disorder and AAC. Session presented at the 2014 International Society on Augmentative and Alternative Communication Conference, Lisbon, Portugal.
  • Porter, G & Burkhart, L. (2010). Limitations with Using a Representational Hierarchy Approach for Language Learning. Retrieved from http://www.lburkhart.com/handouts/representational_hierarchy_draft.pdf
  • Romski, M.A. & Sevcik, R.A. (2005) Augmentative communication and early intervention: myths and realities. Infants and Young Children 18 (3), 174

Spread the word. Share this post!

Comments (19)

  1. Alicia Garcia


    I fully support the 'Presume Competence' paradigm but often struggle aligning this practice with Rowland's Communication Matrix model. Can you please elaborate what. in your view. constitutes a Beginning Communicator?

    • jane


      Hi Alicia, I always work with Beukelman and Mirenda's Participation Model in mind. This means I always make two decisions - I make a plan for today and a plan for tomorrow. So a beginning communicator might be someone who is communicating very informally - those familiar with them can work out what they mean but it's hard for an unfamiliar person. I look at what their system for today is and support it in any way I can - so I might do a personal communication dictionary to help everyone with consistency. But I would start implementing consistent aided language input as well - because that's getting ready for the system for tomorrow. Does that help?

      • khowery


        Thanks for this Jane Awesome post and it was like a (Canadian) Thanksgiving present. Just what I needed. One other thing about the participation model I think is that it forces us to look at the communicative context and environment. I think we need to loosen up our language around the communicator and look at the communicative environment, and the communication partners.
        Aided Language makes ever so much sense to me, but man it is hard to do and hard to 'sell' for kids who are in busy classrooms with people who are not trained (and supported) in doing it. Our beginning AAC may be the communication partners, not the kid.
        Also have you used the Communication Supports Inventory- Children & Youth (CSI-CY)? and if so how? and what do you think? Charity Rowland was a big part of this too I believe

        Again, thank you for the "present"

        • jane


          HI Kathy, I'm glad you thought the post was awesome! One of my friends said it was better than chocolate - which was HUGE praise from her!

          I absolutely agree with you about the communication environment and communication partner. We need to change the "systems" around the user to get AAC happening for so many people. I've written some blog posts about this - but this one is my favourite - http://www.janefarrall.com/aac-systemic-change-for-individual-success/. I wrote that after visiting two schools in a row where all the students with CCN had AAC but no-one was using it. This was, I felt, because AAC was seen as totally related to the individual and people hadn't considered the environment at all.

          And I haven't used the Communication Supports Inventory - I'll look into it! What do you think of it?

      • Alicia


        Thank you very much for your reply, Jane : ) My next question is, how far into the future you look when determining the system that will support the child's needs Tomorrow? For example, if the child has just started demonstrating intentional signals Today (e.g., sometimes she would stamp on footrest to ask for more) and your plan for today is to increase frequency, consistency and range of intentional signals, what system would you introduce for Tomorrow? ... a PODD? or would you start modelling/aided language with a system closer to her perceived level, such as a simple board to first establish 1:1 symbol-referent correspondence?

        • jane


          HI Alicia, I would give the student a comprehensive system. I know sometimes I have to bring the system size down because the team feels it isn't appropriate. Even though I disagree with them I know I'm going to get more buy-in if we compromise - so then I might start with aided language displays or a very high frequency core system. But I always have on my agenda that we are moving to a more comprehensive system as soon as the environment can support it. I think that answers what you asked - but let me know if I've missed the point.

  2. Jodi Altringer


    I work with toddlers and preschoolers. Great post. Highlights all that I need to "unlearn" when it comes to beginning AAC.

    • jane


      Hi Vanessa,

      I think this is one of the inherent problems with PECS. I would probably start modelling a more comprehensive system where the vocabulary stays in a consistent place while you continue to support PECS. Then, as the individual starts to use the new system expressively, I would fade PECS out. I say this not only because of the difficulty of organising the PEC cards but also because PECS tends to focus on requesting and communication needs to be about so much more. A system where the vocabulary is in a consistent place also allows the individual to use their motor planning when communicating. They don't have to think about where "more" is, they just reach for it. It's a bit like driving your car - you don't have to think where the brake is, you just move onto it. Hope that helps.

  3. Pingback: Comprehensive Literacy Instruction: Meeting the Instructional Needs of ALL Students in our Classrooms | Jane Farrall Consulting

  4. Micaiah


    This has been a very informative article for me. My son is only 2 but had a very large stroke affecting his language. We have trialed and are getting an AAC device. I can set it up however I choose and have only been told about the start simple method. I already found myself adding more complex communication options to the device because it was too simple. I do not have the support of the Speech Therapists in this area. They have no exposure to AAC and barely understand PECS so I am learning on my own. I have been using photographs because I was told symbols would be too difficult for him to understand at his age. If I understood the article correctly that method is outdated? Does this apply with a toddler as well? How complex should his vocabulary be at 2?

    • jane


      HI Micaiah, my apologies for my slow response.

      Your understanding is correct - that belief is outdated. For a 2 year old, or for someone beginning AAC at any age, you would want enough vocabulary that you can model it throughout the day.

      Have a look at the core vocabulary user in Proloquo2Go or at PODD for good examples of a good vocabulary. There is a video here which may be helpful https://www.youtube.com/watch?v=s_mc6AWQOM0. There are also a range of videos and resources at http://www.assistiveware.com/product/proloquo2go.

      I hope that is helpful


  5. Nikolaus F Englen


    I think it is crucial that we give the student sufficient vocabulary and I really like the idea of more descriptors. However, I have had two students with significant cognitive challenges where we had tried a plethora of high tech and low tech. Finally, we set up a shelf with representational objects and it worked!!! Also, we made sure these students had access to preferred activities/objects across the classroom. I have also had other students where that failed right away. This is where our individualized Educational Plan comes in handy. It varies from one kid to another. We have to be open minded and flexible!

    • jane


      Hi Nikolaus, thanks for your comment. I have to agree and disagree with you! I completely agree we have to be open minded and flexible - there are so many different factors involved! However, we know from the National Joint Committee for the Communication Needs of Persons with Severe Disabilities literature review, that we need to be implementing AAC for at least 6 months before we see enough change to determine if it has been successful for not. I don't know how you manage to fit in a plethora of options when you need to do a trial as long as that - but that would definitely be my experience as well. In fact, we often need to use a language system for 12 to 18 months before an individual becomes expressive with it. I know I can setup a system which will work faster in some cases (and as part of the participation model I might do that for my system for today while I work on my system for tomorrow) - but I would never say that those systems "work" as they give each individual such a limited range of communication. I have also found that object based systems are very limited as well - and frequently end up only being used for requesting. How have you got around that? When you say that "it worked" what do you mean by that? Communication is all day, every day - and that's what we need to aim for.

Leave a comment

Your email address will not be published. Required fields are marked *