tzippy@dasys1.uucp (Tzipporah BenAvraham) (05/31/90)
Index Number: 8528 Here is yet another file.. movement impairment Movement Impairment Focus Area There is a wide range of causes for movement impairments. These different causes result in different types and degrees of impairment. Head injury, spinal cord injury, loss of or dysfunctional limbs, cerebral palsy, poliomyelitis, muscular dystrophy, multiple sclerosis, amyotrophic lateral sclerosis, Parkinson's disease, and severe arthritis are some of the causes. In general, an individual's impairment could be characterized by one or more of the following categories: Reduced control: the individual's motor control or range of movements has been reduced. Arthritis, poliomyelitis, and mid-level spinal cord injuries commonly result in this type of impairment. Restricted control: the individual has normal motor control but only over part of their body. High spinal cord injuries (normal control of head but no control below the neck) and loss of limbs could cause this type of impairment. Interference with control: other uncontrolled movements interfere with the individual's motor control. Cerebral Palsy and Parkinsonism are two causes of this type of impairment. Weakness: the individual has fine motor control but has no strength or endurance. The individual may also be unable to move major limbs easily or at all. Multiple sclerosis, muscular dystrophy, amyotrophic lateral sclerosis, and poliomyelitis are causes. For individuals with movement disabilities, it is the input mechanisms (keyboards, mice, touchpads, etc.) that present the greatest problems. Other controls (such as on/off switches, contrast knob on monitor, buttons to eject floppy diskettes) may also prove difficult. Many disabled individuals can use a standard keyboard but have difficulty with multiple-key operations. Others are unable to use the standard keyboard, and must rely on special mechanisms for connecting alternate input systems to the computer. Many such specialized +interfaces+ have been developed for the popular computers, as well as adaptations to facilitate handling of disks, adjustment of screen control, etc. However, advancing computer technologies are creating new and more difficult problems. Individuals who are able to use standard or adapted keyboards are often unable to use newer mouse-, touchpad-, or touchscreen-based systems. Furthermore, attachment points (both physical and software- based) that have been used to connect these specialized interfaces in the past are not necessarily present in the newer generations of computers. In fact, for various technical reasons unrelated to the disability market, many new computer designs have carefully restricted or blocked the connection points which are currently used by disability access aids. Thus, existing solution strategies no longer work, and new adaptations have not yet been developed. Computer companies are now aware of the problem, but the technical reasons for the changes remain. Steps that need to be taken include: @@@ Better design of the standard input devices to allow their use by a wider variety of individuals; @@@ Identification and development of connection strategies to allow the use of special adaptive input systems with the newer generation computers; and @@@ Development of control strategies to allow the multidimensional and graded control by more severely involved individuals, which is needed to simulate newer input systems (e.g., mouse, touchscreen). Finally, many of the existing access strategies are so slow that their users are unable to participate meaningfully in regular education or employment settings. More efficient interface techniques are needed to allow physically impaired individuals to function on a competitive level. Program plan The major program thrusts for the Movement Impairment Area are: 1) Quantification and comparison testing of alternate input techniques and devices. Since input devices such as the mouse have become standard on many computers, a number of alternative input devices for users with physical impairments have been developed. However, no evaluative methodologies or data are currently available for use in research or clinical analysis. Quantification methods such as keys pressed per minute become less meaningful for graphics-based operating systems. 2) Development of extensions for standard computers and operating systems to allow access by motor impaired users. Important features providing access for many users with movement impairments can often be implemented in standard products. This program thrust seeks to develop demonstration versions of such features, and to encourage their implementation by computer companies. 3) Development of alternate input techniques for severely motor impaired individuals. In addition to the research and design issues addressed above, the Trace Center is involved in development of designs for special access hardware and software for individuals with severe motor impairments. The center's efforts in the Motor Impairment Area are also directed toward a number of the projects included in the the Cross-Impairment Focus Area, including the development of design guidelines for computers, consumer electronic equipment, and standard and special software; technical support to manufacturers; and the development of interconnection standards. Development of an Instrument for Studying Efficiency of Standard and Alternative Input Devices Project Team: Robert G. Radwin, Ph.D; Yi Hu, B.S.; Mei Li Lin,B.S. Background Pointing-based input devices such as the mouse have become standard components of major new computer systems over the past several years, necessitating the development of alternatives for people with movement impairments. In order to provide quantitative data by which particular alternative input devices or strategies can be evaluated, a microcomputer-based research instrument has been developed. The next stage in this process has been to determine effective evaluative and normative procedures for the instrument. Once the instrument has been tested and its capabilities to effectively measure pointing performance determined, it will be used for headpointing comparison and other studies. Approach The objective of this investigation was to determine if a Fitts' Law discrete target acquisition task could be used for evaluating and comparing alternative computer input devices. Performances using a conventional mouse and an ultrasonic head-pointing device intended for individuals lacking normal movement ability were studied using a Fitts' Law paradigm. The discrete movement task consisted of moving a cursor, starting at the center of a computer display screen to a target located at radial distances of 24.4 mm and 110.9 mm, and at angles of 0!, 45!, 90!, 135!, 180!, 225!, 270!, and 315! with respect to the horizontal. The circular targets were 2.7 mm, 8.1 mm, and 24.2 mm in diameter. These conditions provided Fitts indexes of difficulty of 1.0 bits, 2.6 bits, 3.2 bits, 4.2 bits, 4.8 bits, and 6.4 bits. Performance measures for the pointing task included movement time, cursor path distance, and root- mean-square cursor displacement from the straight-line path between the movement origin and the target. Progress Ten normal subjects performed this task using both devices. Average movement time was 306 ms greater (63%) using the headpointer than when using the mouse. The effect of direction on movement time using the mouse was relatively small compared to the headpointer, which was lowest at 90! and 270!, corresponding to head extension and head flexion, respectively. Average path distance and root-mean-square displacement was lowest at the off-diagonal directions (0!, 90!, 180!, and 270!). Fitts' Law adequately described movement behavior for both types of pointing devices. The instrument was also tested with one subject with cerebral palsy. In this test, the task was performed both with and without thoracic support. The instrument was able to measure significant differences between both the disabled and able-bodied subjects, as well as between the supported and unsupported condition for the cerebral palsied subject. Marked improvements in performance were observed in the subject with cerebral palsy after lateral trunk support was provided. This demonstrated that the task would be useful as an evaluative instrument for selection and comparison of alternative pointing devices for individuals with movement impairments, as well as for evaluating modifications in the workplace for such individuals. A paper on the design and testing of the instrument has been submitted for publication. The instrument is currently being used for headpointing comparison studies at the Trace Center. Headpointing Study Series Project Team: Robert G. Radwin, Ph.D; Gregg C. Vanderheiden, Ph.D.; Mei Li Lin, B.S. Background One of the most powerful new techniques for accessing computers is the use of optical headpointers. These headpointers can be a replacement for the mouthstick for persons with spinal cord injuries, and, as a new input technique for some individuals with cerebral palsy or neuromuscular conditions, can provide rapid, low-fatigue interface methods for computers. Because of their relatively simple hardware designs, these headpointers can also be efficiently and economically manufactured. As a result, a number of different headpointing systems have come out or are under development at this time. Each of these headpointing techniques uses a different technology, has different functioning characteristics, and requires different skills on the part of the user. Several of these techniques, such as the Long-Range Optical Pointer and the LightTalker Headpointer, are actually head pointing devices. Others, such as the HeadMaster, use an ultrasonic technique which provides a function very close to headpointing, but with a moderate amount of slippage and hysteresis. Still others, like the FreeWheel from Pointer Systems, Inc., actually use head positioning rather than headpointing. To date, there have been no cross-comparisons of these various techniques. In addition, many individuals who can use a headpointer can also use other pointing systems. A comparative evaluation between headpointing techniques and other approaches is therefore also required. Approach Because of the many dimensions and skills involved in headpointing, a simple comparison of the techniques would not accurately assess their variability, nor would it be clinically sufficient. A series of studies is therefore planned. These will be based upon the research instrument already developed and tested, including extensions to be added to it as needed. The first test will be the Fitts' Tapping Task. In this task, the individuals will use the headpointers to acquire targets of varying size and in varying locations on the test screen. This will be followed by a Selection Task, where the individuals must both acquire the target and then provide some sort of a confirmation signal. This will be either the activation of a switch (e.g., a puff switch) or the use of a delay confirmation. With some of the headpointing techniques, it is possible to vary the gain (the amount that the cursor moves on the screen for a given amount of head movement). A study is being conducted on the effects of input device gain characteristics. Here, a head motion based computer input device will be used in order to determine optimal gain characteristics for individuals. Of interest here will be the ability to measure the differences in performance and whether the optimal gain varies significantly from subject to subject and between device technologies. Progress Study design is currently under way for the first of the headpointing studies. Subjects will be run during the next two years. Modifications to the research instrument will be made as necessary to accommodate future research needs. Development of Extensions for Standard Computers and Operating Systems to Allow Access by Motor Impaired Users Project Team: Charles C. Lee, M.S.; Gregg C. Vanderheiden, Ph.D.; Mark Novak, M.S. Background The most effective technique for providing access to computers for persons with disabilities is to have the computer designed in such a way that it is already accessible when manufactured. When accessibility features can be built in directly to the design of the computer, the cost for these modifications drops to zero or close to zero, and their availability to persons with disabilities is universal. Of particular interest is the need for individuals with disabilities to access computers in libraries and other settings where they would have difficulty modifying the computer to make it accessible if it were not already so. Not all adaptations for persons with disabilities can be done in this fashion. However, the Trace Center has been able to demonstrate that a wide variety of these strategies can be incorporated directly into the computer design. There are four common adaptations for people with motor impairments which can reasonably be incorporated into operating systems or simple operating system utilities: 1) "Sticky Keys": One-finger operation of +modifier+ keys such as Shift and Control, allowing users to type them before the key to be modified, rather than at the same time. 2) "Key Repeat Control": Control over auto-repeat of keys. The user should be able to control the rate at which the key repeats, as well as the duration for which a key must be pressed before the repeat begins. 3) "Mouse Keys": Ability to use keys on the keyboard to accomplish all mouse functions. These functions include mouse movements and pressing of mouse buttons. The standard numeric or cursor keypads would typically be used. 4) "Slow Keys": Ability to increase the amount of time a key must be pressed in order for a keystroke to be registered. Approach The purpose of this program is to develop simulations, demonstrations, or actual functioning extensions to operating systems which can be used to demonstrate to computer and operating system manufacturers how their systems could be modified to make them more accessible. At the present time there are four major operating systems or operating system extensions that need to be addressed in order to cover most of the microcomputer market: 1) PC-DOS/MS-DOS: Versions of these systems are used by IBM Personal Computers and compatibles. 2) Windows (an extension for MS-DOS): This system is being employed by an increasing number of application programs. 3) Macintosh operating system: System used by all Apple Macintosh computers. 4) Operating System/2 (OS/2): A new operating system developed for the IBM PS/2 line of computers. Progress PC-DOS/MS-DOS Environment: The Trace Center has developed a utility software program called +One Finger,+ to allow one- finger operation of modifier keys and adjustment of auto- repeat rate. Keyboard performance of mouse functions is not included, since the mouse is not a standard input device for MS-DOS. One Finger Version 5.04 has recently been programmed, debugged and released. It is available from the Trace Center Reprint Service for $10.00. This program is also being distributed by IBM, and is available through bulletin board services. There are no copy restrictions, so owners of the software can distribute it freely to anyone who needs it. The program will be revised as necessary to accommodate future changes in PC-DOS and MS-DOS. The program has served as a model for other operating system modifications and utilities. Windows Environment: The One Finger program for MS-DOS will not work with the Windows extension, so similar modifications have been developed for this environment. Since the mouse is integral to Windows, a +Mousekeys+ feature has also been developed. The purpose of this project is to implement demonstration versions of these features as utility software, with the goal of getting them permanently integrated into future versions of Windows. The four key features listed above have all been implemented in Windows Version 2.0. In addition, a +ToggleKeys+ feature has been added which indicates with a beep the status of keys ordinarily indicated with lights. The capability for input/output routines to comply with the General Input Device Emulating Interface Standard is also being explored. The beta version of these features, designed for Windows Version 2.0, is currently being ported to Windows Version 3.0, once again with an eye to encouraging Microsoft to adopt these features as standard. Macintosh Environment: The Trace Center has been working with Apple Computer to assist in their efforts in implementing computer accessibility. These efforts have resulted in the incorporation of several changes and additions to the Macintosh to increase its accessibility by persons with disabilities. These modifications include three of the four key accommodations for users with physical impairments (all but Slow Keys). Since these features are incorporated into the standard operating system software, they are not +add-ons,+ but are available to any user as built-in options. OS/2 Environment: At the present time, the OS/2 operating system is just beginning to be implemented. Trace Center staff are working with IBM and Microsoft Corporation regarding implementation of integral motor access features for this environment. The DOS and Windows implementations described above are being used as models. Publications Lee, C.C. (1989, June). Access to Microsoft Windows 2.0 for users with physical disabilities. In Proceedings of the 12th Annual Conference of the Association for the Advancement of Rehabilitation and Assistive Technology (RESNA). New Orleans, LA. Lee, C. C., & Vanderheiden, G. C. (1988, June). Accessibility of OS/2 for individuals with movement impairments: Strategies for the implementation of 1-Finger, Mousekeys, and Software Keyboard Emulating Interfaces using device drivers and monitors. Proceedings of the International Conference of the Association for the Advancement of Rehabilitation Technology (ICAART). Canada: Montreal. Trace Transparent Access Module (T-TAM) for Apple and IBM Computers Project Team: Charles C. Lee, M.S.; Joseph M. Schauer, B.S.; Gregg C. Vanderheiden, Ph.D.; David P. Kelso, M.S.; Mark Novak, M.S. Background Certain people with physical disabilities cannot operate standard input devices for commercially available computers. Many of these individuals can, however, operate a special communication or computer access aid, using a control system such as an optical headpointer or single switch. The aid in turn can be interfaced to the computer and used as an input device. In the past, Keyboard Emulating Interfaces (KEIs) have been used to provide this access to computers. However, newer models of computer require the use of other standard input devices, in particular the +mouse+ type of pointing device. Thus the computer user must be able to use the mouse (or an equivalent) to operate the computer. This requires a General Input Device Emulating Interface (GIDEI), not just a KEI. Approach The goal of this development effort is to create a commercially viable design for a GIDEI for Apple Macintosh, Apple IIGS and IBM PS/2 computers-all of which use a mouse as a standard input device. This particular GIDEI has been named the Trace Transparent Access Module (T-TAM), since it provides +transparent+ access to standard commercially available computer systems. The T-TAM is a hardware module that translates standard serial ASCII code output from a communication or computer access aid into the keyboard and mouse input signals required by the computer. Serial ASCII code was selected as the form of output from the aid, since it is by far the most common output interface on computer-independent aids. Serial ASCII is also the interface used in existing KEIs. In order to accomplish the function of a key or a mouse movement, the user sends a single ASCII character or string of characters to the T-TAM. According to the definitions set down in the GIDEI standard, the T-TAM converts the ASCII string to the correct keyboard or mouse input signal. Since most aids also provide the capability to store strings of characters under a single selection, the user can program longer command strings (such as moving the mouse 20 pixels or pressing the +Print Screen+ key) to correspond to single selections. This effectively allows the user the most direct access possible to a particular action. The T-TAM provides connections for all of the compatible computers-that is, the same unit can be connected to an Apple Macintosh SE and to an IBM PS/2 Model 50 (though not at the same time). Progress Hardware design for the T-TAM is being completed. Circuit boards have been manufactured for use in prototypes, and several have been assembled for testing and for seeding to interested manufacturers. Final adjustments are currently being made in software. Development of the software is going hand-in-hand with the development of the GIDEI standard, since the device needs to comply with this standard. A beta test version of the manual has been developed. Preliminary testing of the device and manual has been completed. Although the T-TAM was found to work successfully, the manual proved to be too complex for users and clinicians. A new manual is being designed and is scheduled for field testing in January through February of 1990. Three manufacturers have so far expressed interest in the T- TAM; one has been sent a unit for production costing. Development of Improved Headpointing Computer Access System Project Team: Jon R. Gunderson, M.S.; Joseph M. Schauer, B.S.; Gregg C. Vanderheiden, Ph.D. Background In 1981, the Trace Center began development of an optical headpointing input system for computers. The system was designed for use by persons who can easily aim at a target by moving their head, but who cannot effectively or quickly type keys using a finger, headstick or mouthstick. The user would point a small, light-weight optical detector towards an image of a keyboard displayed on a computer monitor. Pointing at the keys on the image would be equivalent to typing them on the standard keyboard. In this way the user could operate standard commercially available software on the computer. The prototype system used one microcomputer and monitor for the input system and another for the application program (the word processor, spreadsheet, etc. that the user wanted to operate). Before transfer to a manufacturer for production and distribution, the Long Range Optical Pointer (LROP) system was modified so that only one computer was required. However, two monitor screens were still required: one to display the keyboard image and one to display the application program. Feedback from users and from the manufacturer indicated that some people could operate the system better if it was implemented on one screen instead of two. Such an implementation would also be necessary if the program was to be used with portable computers, using headpointers other than the LROP. For these reasons, a +One Screen+ version of the original LROP software was developed. Approach The One Screen version of the LROP is compatible with IBM Personal Computers, using standard CGA graphics. Standard software is run on the computer (word processing, spreadsheet, etc.). To type keys, the user points the optical headpointer at the screen. The application program is then temporarily interrupted while an image of the keyboard appears on the screen. The user can type a string of text or other characters on the keyboard screen, and edit this string before sending it to the application program (this feature also was not present on the two-monitor system). From the keyboard screen, the user can: (1) switch to the application program for viewing, without sending the characters currently typed on the keyboard screen; (2) switch to the application program and send the typed characters to the program; or (3) move the pointer off the screen while retaining the keyboard image (in order to rest or think). The One Screen version of the LROP also provides a typing acceleration system. In the ordinary keyboard image, a list of common words appears at the top of the screen. The user can type any of these words by selecting it with the optical pointer. If the desired word is not on the list, the user can enter the initial letters of the desired word. As letters are entered the word list changes, displaying words that begin with the letters typed so far. The user can type any of these words at any time by selecting it from the list with the optical pointer. The word lists are drawn from a 4000-word vocabulary. Progress Three versions of the One Screen software have been created: one driven by the LROP (sold by Words', Inc. of Lancaster California), one for a mouse pointer, and one for the FreeWheel infrared-based pointer (sold by Pointer Systems, Inc. of Burlington, Vermont). Commercial transfer is currently in process. A manual for the One Screen program has been developed, using the structured documentation approach also used in the documentation for the Trine communication system, developed at the Trace Center in 1984-85. Publications Gunderson, J. R. & Vanderheiden, G. C. (1988, June). One screen multiplexed keyboard for transparent access to standard IBM PC software. In Proceedings of the International Conference of the Association for the Advancement of Rehabilitation Technology (ICAART). Canada: Montreal.