Go to previous article
Go to next article
Return to 2005 Table of Contents
Presenter(s)
Barry Romich, P.E.
AAC Institute
1022 Heyl Road
Wooster, OH 44691-9786
Tel: 330-262-1984 x211
Fax: 330-263-4829
Email: bromich@aacinstitute.org
Katya Hill, Ph.D., CCC-SLP
Center for Assistive Technology Education and Research (CATER)
102 Compton Hall
Edinboro University of Pennsylvania
Edinboro, PA 16444-0001
Tel: 814-732-2431
Fax: 814-732-2184
Email: khill@edinboro.edu
Blaise W. Liffick, Ph.D.
Department of Computer Science
Millersville University
Millersville, PA 17551
Tel: 717.872.3536
Fax: 717.872.3149
Email: bliffick@millersville.edu
Background
For many people with disabilities, assistive technology (AT) can be helpful in achieving their educational, vocational, and personal goals. For some, the human interface with AT can be direct selection, such as using a keyboard or otherwise pointing to targets and making selections directly using various means. Such methods generally result in the highest performance.
However, other users of AT cannot use direct selection or cannot, for various reasons, use it well. For some of these people the use of one or more single switches is required. Single switches can take many forms and can be activated in many ways. The most common form is the mechanical button with an internal microswitch. Another form is the P-Switch, based on a piezoelectric element that can be mounted to detect slight movement, such as raising the eyebrow. The piezoelectric element is connected to a small interface box with a relay output. Yet another example would be biological sensors that detect intentional signals such as EMG (electro-myographic), EEG (electro-encephelographic), or EOG (electro-occular). In these cases, the input signal is detected and an output relay is activated.
Single switches can be used to control scanning, in which choices are made available in a time sequence and chosen upon switch activation. Choices can be made from an array by first scanning rows and then scanning elements within the chosen row. Single switches also can be used in codes, such as Morse Code, either in combination or by distinguishing short and long activations.
Evidence-Based Practice
In the delivery of AT professional services, evidence-based practice (EBP) has become the normal expectation. EBP involves three components: 1) evaluation of external evidence, 2) measurement of personal performance, and 3) the knowledge and skills of the service provider. EBP results in the highest performance possible. This is important to people who use AT. For some areas of AT, such as AAC, performance can directly influence the quality of life experienced by the individual.
When a form of single switch is being considered for the human interface, significant issues surround the selection and positioning of the switch and the development of switch skills. In the evaluation of external evidence, one would seek information on the performance that has been reported in the research literature and experienced by others in considering the switch alternatives.
After determining the various switch options to be considered, each must be tried. It is common that the same switch can be used in many different ways. For example, a button switch could be mounted for activation by head movement or by pushing with the foot. Pushing with the foot could be implemented in many different ways. The switch could be mounted to the front, left, right, top, or bottom of the foot. The performance may be quite different among these choices.
In addition to the actual performance, other factors such as ease of use and fatigue must be considered. Initial performance may not be indicative of long term performance after skills are acquired (Jagacinski & Monk 1985). Ease of use at first encounter may be contrary to most effective long term use (Norman 1980; King 1999). Therefore, following a short evaluation, if one chooses a switch or switch site that appears easiest, one may also be choosing the switch or site that will be least effective in the long term. These are areas where the research and information on use by others may be helpful.
For all of these considerations, the measurement of performance using a switch can be valuable. Without performance measurement, it is unlikely that the best performance for an individual will be achieved.
Single Switch Performance Test
Single Switch Performance Test (SSPT) is a clinical tool to facilitate the measurement of performance using a single switch. SSPT is software to allow a PC to be used to practice switch use and collect switch use data. SSPT measures three parameters: 1) time to activate the switch, 2) time to release the switch, and 3) speed of repeated switch activation. The opening screen allows the selection of background information and also the specific test to be administered. Stimulus is provided by the computer and can be visual and/or auditory.
The switch is connected to the computer through a USB mouse. A standard mouse can be modified by adding a mini phone jack in parallel with the left mouse button. The USB mouse purchase and modification can be done locally using instructions available at the web site of the AAC Institute. Alternatively, a modified USB mouse can be purchased through the AAC Institute web site or from other sources.
The Practice Mode is intended to allow the switch user to get a feel for the activation of the switch prior to administering performance tests. Some switches have an audible click that can be heard. However, some individuals may have hearing impairment, some switches have no click, and/or the site (e.g., under the foot) and/or ambient noise may preclude any activation feedback. In Practice Mode audio and visual feedback indicates activation of the switch and is maintained until the switch is released.
The Activation Performance Test is initiated with a switch activation and release. Then, ten tests are performed. For each test, following a random time from the previous switch release, the stimulus is provided and the time from stimulus to switch activation is recorded. When the test is completed, the average time is displayed.
The Release Performance Test is initiated when the switch is activated and maintained. After the switch has been held for a random duration, the stimulus is provided and the time to release the switch is recorded. This is repeated ten times. When the test is completed, the average time is displayed.
The Repetition Performance Test is initiated with a switch activation and release. Following a random delay, the stimulus is applied and maintained while the subject activates and releases the switch five times. The time is recorded. When the test is completed, the time is displayed.
SSPT provides for saving and/or printing the results of the test session. The report includes the name of the subject, the name of the test administrator, the version of SSPT used, and the date and time from the computer clock. This will allow test results to be compared across time.
Limitations
SSPT is intentionally simple in order to keep the support to a minimum and to allow free distribution. SSPT is not comprehensive. For example, it does not allow consideration of the anticipation factor, measure or report errors, or gauge fatigue. More comprehensive collection and analysis of performance data is possible. One approach that can be used with AAC is to collect language samples using LAM (language activity monitoring) and analyze the results using PeRT (Performance Report Tool). This would reflect actual use of the AAC system. Information on LAM and PeRT is available at the web site of the AAC Institute: www.aacinstitute.org.
More comprehensive measurement of switch performance is available in COMPASS software (Koester & McMillan 1997). COMPASS is still in development and testing. When COMPASS is available, information will be available at the web site of the AAC Institute: www.aacinstitute.org.
Applications of SSPT
The primary application for SSPT is as a clinical tool for measuring switch performance. Obvious values are in comparing the performance using different switches and comparing the performance using different control sites and methods. This data results in an objective approach to switch selection and placement.
Another application for SSPT is the ongoing evaluation of switch performance. This data provides feedback and can offer guidance relative to the acquisition of switch skills. Analysis of this data can indicate which methods and techniques are producing the desired outcomes and also can indicate when achievable levels of performance have been reached.
The secondary application for SSPT is research and product performance testing. The common availability of SSPT allows for the common collection of data. Little is presently known about how people with disabilities learn and use switches. Switch developers can use SSPT to test their devices and to help AT practitioners understand the performance that can be achieved.
Getting SSPT
SSPT is available as a free download from the web site of the AAC Institute: www.aacinstitute.org.
References
Jagacinski, R.J., & Monk, D.L. (1985). Fitts' law in two dimensions with hand and head movements. Journal of Motor Behavior, 17, 77-95.
King, T. (1999). Assistive technology: essential human factors. Boston: Allyn and Bacon.
Koester, H., McMillan, W. (1997). Software for Assessing Computer Usage Skills. Proc. of RESNA '97, 354-356.
Norman, D. (1980). The psychology of everyday things. New York: Basic Books Inc.
Go to previous article
Go to next article
Return to 2005 Table of Contents