Shifting research goals.

I’m going to use this post to talk about how my research interests in AAC have changed, and why I think that other people might want to make a similar change. Even if you are unfamiliar with AAC, you might find the process illuminating.

I’m aware a lot of my readers are more ‘general disability’ than AAC focused so the quick Wikipedia definition might be helpful.

Speech-generating devices (SGDs), also known as voice output communication aids, are electronic augmentative and alternative communication (AAC) systems used to supplement or replace speech or writing for individuals with severe speech impairments, enabling them to verbally communicate their needs.[1] SGDs are important for people who have limited means of interacting verbally, as they allow individuals to become active participants in communication interactions.[2]

We’ve had a range of posts(here, here, and here) talking about modifications performed on AAC devices  – or to help with AAC use. The more I look at such ‘hacks’ the more two things become clear.

  • None of these are software solutions; this is the work of solderers, of people twisting wires together, and if stitching together harnesses on sewing machines.
  • If people are hacking at this level,  the fancy technology that researchers like me like to develop is probably unhelpful.

The second point is key.  I’m starting to believe that assitive technology research in the area of AAC (as opposed to social research in AAC, which is what most people do) is aiming at the wrong level.

Since I started working with Lizzie Coles-Kemp it’s been much easier to articulate certain points about the context AAC devices are used in. It’s a context that is complex, mostly unmapped, and rooted in a range of family/organizational relationships.

So while the programmer in me wants to investigate using the iPad’s GPS to automatically generate utterances for AAC users, the researcher in me is starting to think that it would be much much more useful to compare the lifespan of AAC devices that have been ‘hacked’ or at least, extensively modified, with those that are standard factory models. (On the one hand it could be shorter because of this sort of activity, but it could also make the device more relevant for a longer period of time).

The big brother in me to see AAC devices using much more nuanced speech recognition so that Richard can use it.  But the researcher in me wants to know how families cope where the surrounding family  (reasonable) doesn’t have the skills to open up the device.

The computer science academic in me wants to see how artificial intelligence can be used to improve word prediction, but the researcher in me wants to ask to talk to the 10 ‘most successful’ AAC users at a given SEN college or AAC conference, because I’m willing to bet that  nine of the ten of them will have customised their setup with a bit that was put together out of sticky tape and staples.


One problem is that assitive technology  researchers have a tendency to look at what the technology can give, rather than what the context requires. But automatic generation of content isn’t of interest to a user whose main problem is that the screen is too dim to see in sunlight.  In that case black tape and cardboard is more use than any million pound research project. Using highly sophisticated artificial intelligence technology to improve touchscreen access is great,  but what a user needs is to find out that rubbing a candle on it in the morning makes it much easier to clean the dribble off.

It’s worth at this point highlighting something that I put out the other week.  This is the table showing sales of AAC devices in England over the last five years (from the Domesday Dataset)

Screen Shot 2014-07-05 at 10.26.16

And this is the relevant quote:

 A more sobering result to consider for researchers in technical AAC is the popularity of devices that are less obvious targets for customisation and improvement.  The GoTalk and Tech/Speak ranges are solid favourites for a particular section of the market and part of their appeal is that they are relatively ‘non-technical’\footnote{For example, neither device has a LCD screen, instead they have buttons with printed icons} and are much easier for users and staff to get to grips with: this appeal is somewhat in tension with advanced features like automatic generation of content and voice banking.  It is entirely possible that technical research would have more impact if it focuses on making  high-capability devices more acceptable to existing users rather than increasing the already impressive capability of existing devices.

This table largely started the evolution of my own thinking. I love the work I do/have done, but I’m both motivated by stories and ruled by science and the numbers say that maybe I’ve been solving the wrong (interesting) problem.

My feeling is that the technical advances that make the most difference for AAC is the work that is done by hackers in colleges assessment centres, families, and social groups.  I think that the personalisation and the hacking makes far more difference on the ground level than word-prediction will do in a decade.




4 thoughts on “Shifting research goals.

  1. Joe, sadly with the restrictions Medicare is now imposing on SGDs, there is a completely new paradigm to address for SGDs. All the strides the talented persons like yourself have made and continue to conceptualize for the SGD user community are being squashed and reversed by Medicare. Rest assured, insurance companies will follow as they tend to take their lead from CMS for coverage policies.

    The CMS restrictions require SGDs to be a singular tool utilized by a user for conversation or learning purposes only with a Medicare invoked “locked-in” approach. CMS does NOT allow SGDs to be communication devices. They now require them to be limited as a face to face conversation device. Adding the rental program aspect to it, devices won’t be hacked for the first 13 months!!

  2. Come join us at Team Gleason where we are working with various organizations to push SGD technology forward, bringing the costs down to affordable levels for everyone while bringing the quality and customization up. We want to get SGDs to the point where CMS rules are irrelevant and these devices are commodity devices running free software. Hit my email for more information.

Leave a Reply