2003 Conference Proceedings

Go to previous article 
Go to next article 
Return to 2003 Table of Contents 


Russell Thomas Cross
AAC Product Manager
Prentke Romich Company
1022 Heyl Road, Wooster, OH 44691
Email: rtc@prentrom.com

The SpringBoard(TM) communication aid is digitized voice output device that uses a touch-screen display. To address the needs of as wide a population as possible, the device features input via;

  1. Single switch
  2. Dual switches
  3. Optical head-pointing
  4. Direct selection

The device also ships with a basic access testing program, which allows a clinician to do a quick evaluation of how well a client can select keys. Essentially, it starts by providing a screen with 4 large keys, one of which has a target image - a rabbit. The client has to select the rabbit using any of the 4 access methods listed above.

On selecting the target, the screen changes so that the rabbit now appears in a new position. Again, the client has to select this to move on. At one stage, the keyboard size changes to 8 smaller locations, then 15 even smaller ones, finally to 36 relatively small keys.

If the client can follow the target all the way through to the final location with 32 keys, then the clinician can then select a key that configures the device to provide a pre-designed vocabulary based on 36 keys. Similarly, if the client succeeds at 15 locations but fails at 36, the clinician can hit a button to change to a 15-key program.

However, if the client cannot even succeed at the 4-location level, then this indicates that some switch training may be necessary. This presentation seeks to provide some such suggestions, the templates for which can be loaded into the SpringBoard.


Arguably the single most important phase of providing a client with switch access is that of evaluating where to position a switch - or indeed switches. Input from OT and PT professionals is critical because they are typically best able to systematical evaluate a range of physical sites that a client might be able to control - the "Activation Site."

The notion of 'control' means that the area of the body that the client uses has to be, as much as is possible, under the voluntary control of the individual. If the site is prone to spasmodic, uncontrolled actions, then it would be a poor choice for an activation site.

The activation site may also determine the switch type to be used. For example, suppose the client can use the movement of the large toe on the right foot voluntarily, this has an impact on the switch choice. Clearly a large switch is not practical, and for such a small movement, a highly responsive switch may be preferable. Contrast this with a client who can operate a switch by a gross movement of the whole foot in an outwards direction. For this person, a larger switch would be better to make sure there is a good area for the foot to hit, and it would need to be very robust, so possible less responsive.

Once the issues of Activation Site and Switch Type are solved, then the next step is some kind of Switch Training.


Although the typical aim of teaching switch use in AAC is to be able to access a communication device, there are three basic stages to consider before reaching that level:

  1. Cause and Effect
  2. Timing
  3. Choosing

The second two may overlap somewhat, but the first stage has to be dealt with before either of the others can be addressed.


Quite simply, if a switch is to be effective, the person operating the switch has to know that an action with the switch causes something else to happen. If the client does not understand the relationship between a cause and its effect, then using a switch to operate some kind of communication device is going to fail. Establishing cause and effect is the first step towards becoming a proficient switch user.

The SpringBoard can be configured to allow a client to hit a switch and cause a response from the device, be that a sound, a song, or an instruction. The session will demonstrate how to set-up and use multiple cause-and-effect activities that can be access by a single switch selection.


For single-switch access, there is typically a need to be able to scan and make a selection within a specific time period. So the client is not only having to physically select a switch, but also work within a particular window.

For clients able to access two separate switches, it is possible to overcome some of these timing issues by using a mode of step scanning whereby one switch causes a marker on a device to move between locations, and the second switch makes an activation. Having a "Move switch" and a "Select switch" operating independently takes the time element out of the equation. It increases the number of activations needed to access an item, but eliminates the stress or inaccuracies caused by time-based selection.

The SpringBoard offers both single-switch and 2-switch step scan options. A number of training options can be demonstrated using both these methods.


Being able to select a switch on command is one skill; being able to then make significant choices is another. For example, it is possible to set up a system whereby a client uses a switch to choose between 4 items, but the items need to be chosen such that real choices are involved, not bogus choices. An example of a bogus choice scenario is to have "I want M&Ms," "I want an Oreo cookie," and "I want a chip" as options, when in fact the client could hit any of them to get a positive reward!

Making choices via a switch is clearly of critical importance when using a communication system. The SpringBoard default vocabularies depend on a client being able to make choices in order to be communicative. In this session, a number of suggested activities to develop choices will be demonstrated.


The basic SpringBoard software and hardware design allows for the development of switch assessment and switch training packages. In this presentation, a number of such packages will be shown, illustrating the power of the device as an evaluation tool as well as a communication aid.

Go to previous article 
Go to next article 
Return to 2003 Table of Contents 
Return to Table of Proceedings

Reprinted with author(s) permission. Author(s) retain copyright.