Have you been told he is ‘not ready’, or that a tablet will stop him talking? These myths delay AAC (augmentative and alternative communication) or lock the person into a board of requests. Here is what the research says, with the nuance that matters.
‘AAC will stop him from talking’
Studies show no harm to speech; they often see more spoken output (Millar, Light and Schlosser, 2006; Romski et al., 2010; Kasari et al., 2014). The nuance: AAC does not guarantee that speech will come. It guarantees that the person does not have to wait to communicate. See does AAC stop speech?
‘He’ll talk eventually, let’s wait’
Waiting has a cost: fewer chances to learn, more frustration, and behaviours that sometimes become the only way to be heard. AAC avoids that delay while speech develops. And when speech does come, it can stay hard for strangers to understand, even in adulthood: the tool keeps a role as a backup (Beukelman and Light, 2020).
‘He’s not ready: there are prerequisites’
Families are still told about these ‘prerequisites’:
- eye contact;
- object permanence (knowing that a hidden object still exists);
- cause and effect;
- sitting still and paying attention;
- matching pictures to objects;
- pointing accurately with a finger;
- a certain age, a certain IQ, a certain diagnosis;
- being able to read before having a keyboard.
None of them is needed to start, and some actually develop through communication. Training them in isolation does not improve communication (Kangas and Lloyd, 1988; Romski and Sevcik, 1988). The US National Joint Committee opposes access criteria based on age, IQ or diagnosis (NJC, 2003). The only condition, Beukelman and Light sum up, is being alive. Pointing is just one access method among others (see access to AAC).
Developmental stages (unintentional signals, deliberate gestures, first symbols, combining words at around 50 words) help to choose goals; they do not decide who is entitled to a tool.
The real prerequisites are around the person:
- partners who use the tool while they talk, wait, and respond to every attempt;
- a tool that is available all the time, not kept in a cupboard so it doesn’t get damaged;
- suitable access: good seating improves accuracy (Costigan and Light, 2010);
- words that matter to the person.
‘He doesn’t understand, there’s no point’
What he can say is not what he knows
A person who cannot spell can only say the words someone has put on their tool: most have fewer than 1,000 concepts, often fewer than 100, whereas a 5-year-old who talks knows more than 2,000 (Bloom and Markson, 1998). Judging what the person produces often means judging their board.
Tests can mislead
A 3-year-old boy who had always lived in hospital ‘failed’ a picture test. Yet he understood ‘suction’, ‘ventilator’ and ‘injection’: his vocabulary was that of his world, not that of the test (Beukelman and Light, 2020). Tests that require speaking or pointing accurately often underestimate understanding.
The least dangerous assumption
Without proof, choose the assumption that will do the least harm if it turns out to be wrong (Donnellan, 1984):
Wrongly assuming she understandsWe will have spoken to her with respect and offered too many words. Low cost.
Wrongly assuming she doesn’t understandWe will have deprived her of conversation, school, reading and choices for years. Huge cost.
Presuming competence does not stop you assessing: you teach, then observe what the person learns, over several days and with several partners, with the tool in hand from the start (see assessment as a team). Day to day, talk to the person, not about them, and respect their age: no childish drawings for a teenager, and address an adult the way they prefer. The Communication Bill of Rights makes these rights: to choose, to refuse, to have their tool at all times, to be treated with dignity.
‘Too young, too old, too disabled’
Babies benefit from AAC. Adults who went decades without a tool make rapid progress as soon as they are given real opportunities to communicate (Holyfield et al., 2017; Drager et al., 2019). There is no age limit and no level of disability below which ‘there’s no point’. For a young child: My child isn’t talking: where to start.
‘Being able to ask is enough’
Requests (eat, drink, toilet) end up filling the boards. A programme centred on them produces passive communicators, especially in autism (Light, Parsons and Drager, 2002). Communicating also means commenting, joking, telling stories, asking questions, refusing. PECS has strong evidence for teaching requesting, little for teaching commenting (Ganz et al., 2012). See PECS or robust AAC? and beyond requests.
‘A core word list is enough’
Core words (‘more’, ‘all done’, ‘want’, ‘help’) are essential for combining words, but not enough. A beginner starts with their own words (names, favourite food, toy) and a few powerful words (Laubscher and Light, 2018). The ‘80% core words, 20% specific vocabulary’ ratio circulates without a study behind it (Beukelman and Light, 2020, on the LAMP booklet): a rough guide, not a law. See which words to put in the tool.
‘Symbols speak for themselves’
What seems obvious to an adult is not obvious to everyone. Children aged 4 to 7 did not recognise the standard symbols for ‘more’, ‘open’ or ‘what’ (Light et al., 2008; McCarthy et al., 2018). Every symbol has to be taught, in context, and personal photos help at the start.
‘Writing the word under the symbol teaches reading’
The printed word mainly helps the partner; it is not enough to learn to read (Erickson et al., 2010). Real reading instruction is needed. Up to 90% of people with complex communication needs reach adulthood without functional literacy, mostly for lack of teaching (Foley and Wolter, 2010). See AAC and literacy.
‘You set up the tool once and for all’
Vocabulary that does not grow ends up at the bottom of a bag: lack of updates is one of the main reasons tools are abandoned (Johnson et al., 2006). You need a tool that is easy to change and someone in charge of it (see the AAC lead).
‘What works in the session will work everywhere’
Exercises in the therapy room transfer poorly to daily life (Schlosser and Lee, 2000). AAC is learnt in real routines (meals, play, shopping), with real partners: parents, brothers and sisters, teachers. No ‘show me the dog’ quizzes: use the tool yourself as you talk. See model, wait, respond.
In summary
| Myth | What the research says |
|---|---|
| AAC stops speech | No harm, often more speech |
| He’ll talk, let’s wait | Waiting has a cost |
| There are prerequisites | None |
| He doesn’t understand | Tests mislead; presuming costs less |
| Too young, too old | No age limit |
| Requests are enough | Risk of a passive communicator |
| Core words are enough | Necessary, not sufficient |
| Symbols are obvious | Every symbol is taught |
| The written word teaches reading | Reading has to be taught |
| Set it up once | Without updates, abandonment |
| The session is enough | Learning happens in real life |
Frequently asked questions
Do you need to read to have a keyboard?
No. The keyboard is part of learning to read: hiding it until later delays reading.
What if the people around don’t believe in the person’s abilities?
Share concrete observations, film the successes and suggest a trial period with modelling. Results often convince more than arguments.
Pictalk Buddy brings together AAC boards, a visual schedule, guided sequences and notebooks, with the same symbols, on Android, iOS, the web and the watch. A pay-what-you-want app, developed by a non-profit association. Discover Buddy
Sources
- Beukelman, D. R., and Light, J. C., 2020, Augmentative and Alternative Communication (5th ed.), Paul H. Brookes.
- Donnellan, A. M., 1984, ‘The criterion of the least dangerous assumption’, Behavioral Disorders, 9(2), 141-150.
- Studies cited from Beukelman and Light (2020): Millar, Light and Schlosser (2006); Romski et al. (2010); Kasari et al. (2014); Kangas and Lloyd (1988); Romski and Sevcik (1988); NJC (2003); Costigan and Light (2010); Bloom and Markson (1998); Holyfield et al. (2017); Drager et al. (2019); Light, Parsons and Drager (2002); Ganz et al. (2012); Laubscher and Light (2018); Light et al. (2008); McCarthy et al. (2018); Erickson et al. (2010); Foley and Wolter (2010); Johnson et al. (2006); Schlosser and Lee (2000).