Language access software—now that is a very exciting topic! Seriously, it truly is…and I hope you get as excited about it as I do.
Once kids are able to access a computer through eye gaze, what is it they are going to say? How are they going to say it? THAT is what the language access software is all about, and it plays a HUGE role in successful communication.
 |
Little nuances in fit make all the difference in comfort,
even within the same size! |
Getting a good fit between child and language access can’t be stressed enough. It’s like fitting a shoe…even within the correct size, there are nuances to the fit of a shoe that make it successfully comfortable…or irritatingly painful. It’s got to fit right or it just plain isn’t a good match.
The key to selecting the appropriate software is your student—her learning and internal organizational style, what she wants to say, how complex her communication abilities are (or have the potential to be), even what level you hope her to achieve 20 years from now since what she does today can help lay the foundation.
Another consideration is how many hits it takes to create the message she wants, because even eye gaze is very fatiguing. And we know with Rett syndrome or any other disorder involving severe apraxia, motor planning of any kind is extremely exhausting.
Last week we talked about the devices made by Dynavox, Tobii, and Prentke-Romich. All three manufacturers use programs with different methods of organizing language and retrieving vocabulary. While there is a learning curve with each system, it could be that the language organization on one clearly stands out during the trial period as a better fit for your student.
In addition to the software packages traditionally associated with these devices, there are third-party packages that can be installed on most any of them. These open up even more options to the user, but I’m going to stick to the programs typically used by each of the manufacturers to simplify the discussion. The same evaluation principles apply when looking at third party software, such as WordPower or Picture WordPower or PODD.
So that leads us to language organization. I apologize in advance if I'm covering what you already know; I've learned it's best not to make assumptions (sheepish grin).
Language is stored as letters, words, or phrases.
1) When it is stored as letters, we have an onscreen keyboard, typically with each letter taking a single hit to spell out a word. The number of hits to create a message depends on word and message length.
The benefit of spelling is that it allows an infinite number of novel messages. There are also word prediction and abbreviation expansion options that can cut down keystrokes.
But even with these options, the high number of hits to spell is fatiguing. Spelling also tends to be frustratingly s-l-o-w with eye gaze, so that back-and-forth conversation requires excruciating patience.
2) When language is stored as words, it allows for as much novel message creation as most of us are likely to use in a lifetime, although unique words will have to be added.
Words can be accessed through one-key hits or through key combinations.
Creating complete sentences is still slow, but doable. Most AAC users who opt for words have very "telegraphic speech." That is, they might say "cookie" "please" instead of "Will you please get me a cookie?" After all, the motor demands make words expensive in energy output. Kids can still be very effective in communicating a message that way, though.
Usually a person needs more vocabulary than can be stored on a single page. So pages are created to categorize the words. This is the most familiar language organization strategy.
For example, to reach the word "zebra," you might need to navigate from Nouns (or Things) > Animals > Zoo > more Zoo words > zebra, taking 5 hits to access the word “zebra.” Sometimes going through the organizational tree can take more hits than just typing out the spelling, unfortunately.
A challenge with this system is that memorizing the placement of the buttons can be very difficult. Each page offers a new array of buttons that must be visually scanned, which can be a tricky task for eye gazers, since eye movements are being tracked by the computer. The programmer must work hard to maintain consistent placement of buttons to support motor automaticity.
Let’s talk for a minute about motor automaticity and why it is essential for eye gaze users.
Motor automaticity is when your body moves the way you want without you having to think about it. It is the skill that lets you reach the stick shift or turn signal lever in your car without having to look. Touch typists use this skill as they fly over the keyboard, hitting all the right letters without peeking.
Your body knows where the shifter or turn lever or computer keys will be through repeated practice. Touching these things becomes automatic; you don’t have to look or even think about it. They are always located where your body expects them to be.
People who use eye gaze can—and should—develop motor automaticity with their eye movements. Their eyes come to expect certain keys to appear in the same position each time.
By contrast, scanning with the eyes can cause unintentional mis-hits when a child’s eyes dwell a little too long during the scanning process. The sooner a child can learn motor automaticity, the more accurate his communication becomes.
Single words can also be retrieved through a process called “semantic compaction,” which aids motor automaticity. Each stationary key represents a concept, and the concepts are combined to create a single word. For example, if you hit "apple," which represents food, and "verb," you get the word "eat." Similarly, "apple" + "feelings" = "hungry." A large vocabulary can be accessed in 2-3 hits. The keys are static and their layout can be committed to motor memory.
Semantic compaction tends to require more learning for the adults supporting a child than for the child herself. I'll share my bias--I was highly skeptical at first but forced myself to learn the system and found it to be VERY slick! I'm a believer now. Several summers ago, I had the opportunity to meet an amazing woman with severe CP who can speak at normal conversational rates using a head mouse with this system. Impressive, let me tell you.
3) A third system of vocabulary organization is phrase-based. In this, a person navigates to categories for various situations and hits a key to activate a pre-programmed phrase. This can make for fast conversation, but the person is limited to saying only what is programmed on the device.
Imagine going up to a friend. In one hit, you can ask "How are things going at work?" That's quick access! And sometimes, it meets your needs well.
But imagine this friend has just lost his job... Suddenly, "How are things going at work?" is completely inappropriate. The likelihood that "Sorry to hear about your job situation" is programmed onto the device is pretty slim.
To address this problem, some programs include partial phrases, like sentence starters, or “slot fillers” (fill in the blank). Some are more elegant in their capabilities to do this than others.
The software programs on each of the devices combine options of choosing from letters, words, and phrases, which is a GOOD thing. No one wants to be stuck with just the phrases someone else decides are important! All three include some keyboard for spelling, so I won't mention that again.
Dynavox’s Vmax runs Windows 7 and uses InterAACt. This is a dynamic (meaning "changing" in this context, as opposed to static or stationary) page-based system for organizing single words and phrases. With it, you start with a page of broad categories and narrow down to smaller categories until you find the page with your word/phrase.
A plus to this is that it makes sense to the staff that programs the device. As an aside, the version of InterAACt I used six month ago desperately needed its programming design updated...taking many steps that should have been automated by now. It is possible that the newest version, 1.07, may have some improvements since that time.
Another plus to InterACCt is that it can hold a nearly infinite number of topic pages, which is nice for school.
The downside for the user is that chances for motor automaticity are very small--every page looks different. And every time a new vocabulary word is added, the page changes.
I find that it really doesn't lend itself to composing sentences--too many hits. For single words, it can be okay though. And for phrase-based conversation that will not change often, it works well.
PRC devices run Windows XP and use Unity, which is based on semantic compaction. The core vocabulary is offered in several sizes, with each supporting motor automaticity and few hits to access a huge vocabulary.
It also incorporates a dynamic activity row which changes automatically with the categories as they are selected, including phrase-based conversations. From the "apple" key from the earlier example is hit, the activity row will show food-specific words that would be cumbersome to memorize with key combinations.
It also has sets of pages of one-hit topic vocabulary that can be built for school subjects.
I find Unity easy and quick to program, although you have to fully understand the language structure to locate new vocabulary wisely…or you can foul up the logic if you program into the Core set poorly.
On the downside, the software is less elegant for creating things like quizzes for class. The key sizes are pre-defined and sometimes you would like to have creative license on that.
The other drawback is that staffs tend to perceive Unity as being overwhelming to learn, which is unnecessary, but sadly often the case. PRC in our area offers excellent training that can help overcome that fear. The kids tend to look at it like a super-cool secret code, so I have yet to meet a child resistant to learning it.
Tobii C-12 uses Windows Vista and uses Communicator 4 as a base set. It functions much like InterAACt but with state-of-the-art programming.
Equally important, they have an add-on package (included as part of the device purchase) called Sono Lexis that combines some of the best of the other two navigation systems.
It has rows of one-hit high-frequency core words (such as I, you, want, like, not) that are always accessible. There are stationary category keys that open additional rows of vocabulary above these core rows. It also has a category for phrases.
It provides more motor automaticity than Dynavox's InterAACt but less than Unity on the PRC devices. It's pretty well thought-out, fairly customizable (and I have requests in to the company for features that would make is moreso), and easily programmed.
You can step outside of Sono Lexis to make pages that look however you want, which is nice for classroom activities.
To be honest, I wasn't in love with Sono Lexis at first but the more I play with it, the more smitten I become.
The major complaints I have are that I'd like to be able to customize the core vocabulary and to solve the problem of arranging new vocabulary added to the category rows so that it is logically organized (alphabetizing isn't always the most logical, and inserting a new word in ABC order throws off the motor memory). I’d also love to see a “replace all” feature, but then I want it all!
The new Tobii C-15 will run Windows 7 and the Communicator 4/Sono products.
There you have the basic run-down on the Major Three software programs for language access. Each has strengths and areas that could be improved. They all can be simplified for children who need to start at a basic level. But it is important for you to look at the capabilities of each to get a feel for what each can offer at higher levels.
