Assistive Technology Research Institute
College Misericordia - Dallas, PA 18612
 
Founded and Sponsored by the Sisters of Mercy of Dallas

 

The Efficacy of Three Head Pointing Devices

 



The Efficacy of Three Head Pointing Devices for a Mouse Emulation Task

Denis Anson, MS, OTR/L, Gretchen Lawler, OTS, Alexis Kissinger, OTS, Megan Timko, OTS, Jason Tuminski, OTS, Brian Drew, OTS

Please address all correspondence to the first author at danson@misericordia.edu .

Abstract

Today, there are three competing technologies to provide head-pointing mouse emulation, as exemplified by five products. Since each of these devices claims to provide similar functionality, it can be difficult for the clinician to decide which is the appropriate device for a client. The current study compares the functional performance of the three input technologies using three commercially available devices, the HeadMaster Plus, Tracker 2000, and Tracer. Each device was used to produce a series of drawings of similar complexity until the participants achieved a stable level of performance. The number of trials required to achieve mastery, the speed of drawing at mastery, and the accuracy of drawings was compared for the devices. In addition, the participants were asked about their subjective experience using the devices. Each of the three technologies was fastest for some participants, but the HeadMaster produced the most consistently fast drawing times. Results indicated that, while performance of the devices was similar, participant preference was driven by comfort more than by performance. The two fastest devices, the HeadMaster and Tracer, both resulted in complaints regarding comfort, while the most comfortable device, the Tracker 2000, was preferred by participants even though it was slightly slower in performance.

Key Words: Computer access, head-pointers, mouse emulation, alternative access, assistive technology, disability

Introduction

Electronic assistive technologies are commonly referred to by telecommunications engineers as electronic curb cuts (Vanderheiden, 1985; Jacobs, 1999). Just as physical curb cuts facilitate moving through the physical world, electronic curb cuts may be thought of as facilitating the journey to independence in the electronic world. "To a population that is often physically as well as socially isolated, computers can offer access to information, social interaction, cultural activities, employment opportunities, and consumer goods" (Kaye, 2000). Many people with conditions, such as spinal cord injury, cerebral palsy, or amyotrophic lateral sclerosis (ALS), may find that, while they are able to think as well as their able-bodied peers, they have difficulty controlling their upper extremities (see Wheeless, 1996). Since the standard input devices for the conventional computer are the mouse and the keyboard, both of which depend on use of the hands, upper extremity function is essential for the operation of a standard computer. Hence, conditions that limit upper extremity functioning may make access to the conventional computer difficult or impossible. Electronic curb cuts, however, may remove the barriers to computer access by allowing the individual's residual abilities to provide control. Individuals with conditions that limit upper extremity control but which spare control of the head may find head pointers to be a viable method of computer access (Anson, 1997, p. 162). Considering the variety of head pointing technologies that are currently available for head-pointing, an objective assessment of the efficacy of these devices seems warranted.

Background

Many current educational, vocational and leisure activities depend on computer access. "Today . computer access is required in many aspects of daily living, is fundamental to employment and school participation, and is important as a leisure pursuit" (Anson, 1997, p. 1). An employee with spinal cord injury who has difficulty manipulating a keyboard may be unable to type a resume for an upcoming job interview. A student with cerebral palsy who cannot write or use the keyboard fluently may find writing a 10-page report on the American Revolutionary War impossible. A father with ALS who is interested in recording his life story for his young children in preparation for death may find a computer increasingly difficult to control. Providing access to these activities for those with functional limitations is a legal and ethical imperative. The American's With Disabilities Act (ADA) was enacted "to provide a clear and comprehensive national mandate for the elimination of discrimination against individuals with disabilities" (Americans with Disabilities Act of 1990, Section 2, 1990) The Individuals with Disabilities Education Act (1997) was enacted to assure that individuals with disabilities would have access to mainstream education "to the maximum extent possible." The Rehabilitation Act Amendments of 1998 was intended "to empower individuals with disabilities to maximize employment, economic self-sufficiency, independence, and inclusion and integration into society." For many individuals with severe disabilities, participation in education, employment, and society can be substantially enhanced through computer technology.

There are many approaches to providing computer access for individuals with disabilities. The access method for a client must be selected to match the skills and abilities of the individual, and the demands of the specific task being considered, and the environment in which the task is performed (Anson, 2001). These access methods include such simple adaptations as key latches, expanded keyboards, and enlarged monitor screens which may meet the needs of an individual with a mild to moderate disability. For the individual with more severe limitations, greater adaptation of the interface is required. Such extreme interventions include screen readers for the individual who is blind, and the use of head-controlled mouse emulators for the individual with profound paralysis (Anson, 1997).

Head-controlled mouse emulators, or head pointers, are the electronic equivalent of physical head pointers and mouthsticks. While they do not provide the ability to manipulate the physical world provided by a mouthstick, they do provide the ability to perform tasks that are performed by a computer mouse. However, rather than requiring the individual to move a mouse with their hand, head pointers use various tracking methods to translate the movement of the user's head into movement of the mouse cursor on the computer screen.

Head pointers can be used to create two basic types of documents: text and graphics. For a head pointer user, text input via an onscreen keyboard is similar to accessing menu and button controls, since the targets may be relatively large and control demands are limited. Angelo, Deterding and Weisman (1991) compared the three pointing systems widely available at the time of their study for text input. This single-subject design study showed one system to be clearly inferior, but was unable to detect differences between two widely differing devices, and suggested clinical trials as the best way to determine efficacy. Of the three devices tested, only one remains available today. DeVries, Deitz and Anson (1998) compared mouth-stick and head pointer typing for two individuals with high-level disabilities. Because of the restrictions of the two input methods, the computers were fully set-up at the beginning of each trial, so the relative advantages of being able to manipulate the environment were not part of the study. Again, there were no clear advantages detected in either input method.

Graphics input requires much more precise control of the cursor since the user must access the screen at the pixel level to, for example, assure that two line segments meet. Because of higher demands for control of the cursor for graphics, graphics input tasks are more likely to show subtle differences between input technologies. According to Kanny and Anson (1992), "graphics input is apparently more demanding than text input using alternative interface devices, and we may see more similarity of performance with text input." The differences in performance found in graphics input may also affect text generation, but at a level too subtle to detect with a limited and short-term study. While a graphical image may require hundreds of mouse actions, typing a typical page of text may require several thousand mouse actions. Even very small differences in performance, accumulated over thousands of actions, may result in substantial differences in productivity and satisfaction with a device.

A previous study compared two head pointing technologies available at the time with the conventional mouse for graphics input (Kanny & Anson, 1992). A decade later, there are three commercially available head pointing technologies ; infrared (HeadMouse , Tracker 2000 , and Track IR ), ultrasonic (HeadMaster Plus ), and gyroscopic (Tracer ). Only the HeadMaster Plus, in an earlier but essentially identical version, was available in 1992. Since each of the currently available products claim to provide similar functionality, it could be very difficult for a therapist to make a decision regarding which product is the most appropriate for a specific client. Factors that might be considered in such a decision can include cost, cosmetics, comfort, and function. The cost of these input devices varies: the TrackIR AT Package costs $299, the Tracer has been introduced at $795, HeadMaster Plus starts at $995 without accessories, and the Tracker 2000 retails at $1,895. The relative importance of cost and cosmetics must be determined by each individual client, but information related to functionality and comfort can be derived empirically. The potential benefit of functionality data in decision making between head pointing systems motivates this study.

Research Questions

The purpose of this study was to compare the performance of the three head pointing technologies as exemplified by the HeadMaster Plus (ultrasonic), Tracer (solid state gyroscope), and Tracker 2000 (infrared reflection). This study explored the use of these head pointing devices for graphics input to detect relative differences in performance. The research questions were:

  1. Are there measurable differences in the time required to produce a standard drawing with these three head-pointing devices?
  2. Are there differences in the accuracy of drawings produced using these three head pointing devices?
  3. Are there differences in the willingness of individuals to use these three devices that are based on comfort or control?

Method

Design

This study used a single-subject, successive interventions design with six able-bodied participants (Bloom, Fischer & Orme, 1995).

A balanced order design was used to control for possible order-effects of the head pointing devices. Each participant used the devices in a different order such that, over the six participants, each device was used in the same part of the sequence twice. Any difference in performance detected in this study was therefore, assumed to be based on differences in the devices being tested. The individual with marginal head control might find differences based on the weight of the control component of the devices, but this is outside the scope of this study.

Participants

In this study we used a convenience sample of six able-bodied individuals who were 22 to 29 years of age. The sample consisted of 4 men and 2 women. The 6 participants included 4 college students and 2 professionals (both male). None of the participants reported having difficulty in hearing and vision. Each participant had at least 15 degrees of head movement to the right and left, up and down from the neutral head position. Each participant had sufficient cognitive capacity to follow the directions and reproduce spatial relationships of the drawings.

In order to separate the effects of disability (which will vary for each participant) from the functionality of the devices, we chose to use able-bodied participants for this study. Since the target population for all three devices is those with limited upper body control but intact head control, the performance seen in able-bodied participants should be very similar to that seen in individuals for whom these devices are intended.

Equipment

Figure 1. Tracker 2000, in this study, used a reflective dot mounted on a ball-cap. (Click image for larger view)
Tracker 2000

The Tracker 2000 consists of a camera and infrared light source that rests on top of the computer monitor and tracks a reflective 'dot', one quarter inch in diameter, worn on the user's forehead or glasses. The movement of the dot within the field of view of the camera is interpreted by the computer as the movement of the mouse. The primary advantage of reflective technologies is that they are minimally intrusive. For mouse cursor movement, the only equipment that the user must wear is the small dot. Mouse clicking requires either a software dwell clicking tool or the addition of a physical switch that can be operated by the user, in some cases reducing the minimalist philosophy of the technology. For the purpose of this study, the reflective dot was attached to the band of a baseball cap (See Figure 1). This allowed preservation of the dots. The hat was moved from person to person and was fitted to each individual through its adjustable head-band. The Tracker provides directly for mouse cursor movement, but offers mouse clicks only through input switches on the camera unit.

Figure 2. Tracer Headset (Click image for larger view)
Tracer 1 Headse

The Tracer uses a solid-state gyroscope that is contained within a headset modeled after a sports visor (See Figure 2). The gyroscopic sensor of the Tracer, which is based on the GyroMouse, acts as an inertial platform. Where the ultrasonic technology requires a direct line between the user and the computer, and the infrared reflection requires the reflective dot to be in the field of view of the camera, gyroscopic technology allows the user to be at any angle to the computer, and up to 20 feet away without loss or change of control sensitivity. The gyroscope is connected to a rechargeable radio sender that communicates with a base tethered to the computer. When the client moves his/her head, the solid-state gyroscope generates signals that are translated into movements of the mouse cursor. The wireless connection of the Tracer supports two switches for the primary and secondary mouse clicks, but the user must provide the switches and mounting.

Figure 3. Headmaster Headset (Click image for larger view)
Headmaster

The HeadMaster Plus consists of a lightweight headset containing three microphones (See Figure 3), and a control box, which sends ultrasonic sound toward the user. The headset is connected to the control box via a cable. Changes in the signal of the microphones caused by head movements of the client produce movements of the mouse cursor. The headset supports a puff switch which acts as the primary mouse button, and the second mouse switch can be connected to the control box, though it is not supported in the headset. A remote version of the HeadMaster that is not tethered to the computer is available but was not used in this study.

Mouse clicks for each device were produced by a button switch on the table top (e.g. Jelly bean switch ).

The target stimuli for this study consisted of 7 drawings of equal mechanical complexity. Each drawing is composed of 10 geometric shapes (circles, arcs, rectangles, polygons, and rounded rectangles) with unique fill patterns. These shapes were selected to require frequent access of the tools and fills palletes. The equality of mechanical complexity was assured by using the same graphical elements in each drawing, and simply rearranging them for each drawing (See Figures 4 and 5).

Figure 4. Sample drawing for graphics task (Click image for larger view)
Figure 4.
Figure 5. Sample drawing for graphics task
Figure 5

The drawings were produced using Canvas 7 running on a Pentium II/450 MHz computer with 128 MB of memory. The Canvas program was set up to display a letter size page in landscape mode on the seventeen-inch monitor. The program was set up with the "grid" visible, and with "snap to: grid" turned on, with the "snap to increment" set to ½ grid size. These settings assisted in the production of consistent drawings, and are also similar to real-world drawing situations.

Task

Mastery - The participant was considered to have mastered an input system based on his/her ability to produce three consecutive trials with an elapsed time for consecutive drawings within 7% of each other. The 7% standard was based on the experience of prior studies of the first author. In those studies, a 10% standard for mastery arguably showed continued progress when the participants met the standard. With a 5% standard, individual variation in the drawing times made mastery difficult to achieve. A mastery standard of 7% seems to balance these two issues.

Error - The following conditions were considered as a single error for the purposes of this study:

  • A drawing element was not the appropriate size or shape. The "snap to" feature meant that all the errors were at least ½ of a grid increment, assuring that trivial errors were not an issue.
  • A drawing element had an incorrect fill-pattern applied.
  • A drawing element was missing or in the wrong position relative to other elements.
  • A drawing element had the wrong orientation.

It was not considered an error if the entire drawing was displaced on the page. Drawings were scored for accuracy by placing a transparency of the target drawing over the printed target and identifying each misalignment of drawing elements from the target.

Procedure

Each participant in the study was asked to produce a series of drawings using each of the three head pointing devices. The drawings were presented to each participant in a predetermined order (different for each participant) so that the participants could not memorize the drawings. The order of using the interface devices was balanced to control for learning effects.

Prior to the beginning of data collection, each participant was provided with instructions and a demonstration of how to operate the Canvas graphics program by the researcher. This instruction included a demonstration in creating each of the elements of the graphical target drawings in isolation. Participants could practice using the Canvas program to produce the elements of the drawings with the mouse but could not practice producing the target drawings using the head pointers as the number of trials to fluency with the devices was one variable in the study. This practice typically occupied from 30 to 60 minutes. None of the participants had prior experience with Canvas, and they had only minimal experience with drawing programs in general.

Prior to each trial, the computer was setup with the appropriate head pointing device and the Canvas program running in the test configuration.

The participant was seated in a comfortable chair 18 to 36 inches from the monitor as determined by participant preference and introduced to that specific head pointer. Following the demonstration by the researcher, the head pointing device was adjusted to fit the participant and the first stimulus drawing was displayed on the right side of the monitor in the plane of the screen. The participant was instructed to "Copy this drawing as quickly and as accurately as you can. Are you ready? Go." At the word "Go," the tester started a stopwatch, and the participant began copying the target drawing. The tester monitored the participant until the drawing was completed and the "Print" command of the Canvas program was activated. At that time, the tester stopped the timer, and recorded the elapsed time for the drawing session. Each of the participants drew the pictures using one device until he/she had produced three consecutive drawings with elapsed times within 7% of each other, indicating that the user had achieved mastery. After plateau was reached for each device, the researchers provided the participants with a survey which assessed the participants experience with the head-pointing device. This process was repeated for each of the pointing devices in the study.

The participants were allowed to continue drawing trials for up to 1 hour in each session, with a maximum of two sessions in a day, with at least an hour rest break between. This restriction was imposed to control for possible eye and neck fatigue from using the devices.

Measurement Tools

All timing was done using a Lorus Digital Sports Stopwatch, Model R230019 . Times were truncated to whole seconds.

The drawings were assessed for errors by placing transparencies of the target image over the participant's drawing. Variations from the target drawing were identified and counted by two researchers, with a standard of 100% agreement on error count, and the count was recorded on the participants' data sheet. Errors included individual objects out of position relative to the balance of the drawing, objects drawn to the wrong size or orientation, incorrect hatch patterns, or missing objects. It was not considered an error for the entire target drawing to be shifted on the page.

Data Analysis

Mismatches between the printout and transparency were counted as errors. The time to produce each drawings and the number of errors in each drawing were entered into a spreadsheet and analyzed for difference from the prior drawing to establish whether the participant had achieved fluency. When the participant had produced three drawings within 7% elapsed time, the average duration of these drawings was used to establish the celleration line for comparison between input devices. The surveys were examined for trends and themes.

Results

Trials to Achieve Fluency

Initially we felt that, because participants were learning to control the graphics program as well as the head-pointing device when using the first device, the number of trials required to achieve fluency should be assessed only when the device was in second and third position. In fact, the number of trials to achieve fluency did not appear to vary with position. There was a remarkable homogeneity of trials needed to master these head-pointing devices (See Table 1). Across the six participants, the typical number of trials to achieve fluency with any device was between 6 and 8. The typical total time spent using any device before achieving fluency was less than 45 minutes, again, independent of device.

Table 1. Trials to Achieve Master with Three Devices

 

Trials to Mastery

Subject

HeadMaster

Tracer

Tracker

1

6

7

7

2

8

14

8

3

4

9

7

4

8

19

7

5

8

6

8

6

11

5

6

The number of trials to reach fluency was greatest for the Tracer for 3 out of the 6 participants. Interestingly, for two of these three, the Tracer also allowed the fastest input times. The HeadMaster Plus and the Tracker 2000 were similar in the number of trials to achieve fluency for all participants, and were similar to the Tracer for those who did not experience the drift difficulty of our unit.

Speed of Drawing Production

While not as homogeneous as the number of trials required to achieve mastery, there was a great similarity in times to produce the drawings across participants as well. For five of the six participants, the fastest times were between 200 and 220 seconds per drawing at mastery. (See Table 2) These minimum times are interesting because they were spread across all three input devices. However, there were clear preferences in the performance of the three devices, with the fastest device for each participant being some 30% faster than the slowest at mastery.

Table 2. Mean Time to Complete Drawings at Plateau

 

Average Seconds at Mastery

Subject

HeadMaster

Tracer

Tracker

1

318

622

253

2

247

222

260

3

217

185

415

4

197

221

273

5

217

287

196

6

228

330

229

The devices were in a three-way tie for fastest input. Each device proved to be fastest for two participants, The HeadMaster Plus was second fastest for four participants and never the slowest. The fastest input at mastery for the Headmaster was 196 seconds, and the slowest 318 seconds. The Tracer, which exhibited a drift problem in our preproduction unit, produced both the fastest and the slowest times at mastery. One participant was able to produce the drawings in an average of 185 seconds at mastery, while another, plagued by drift, required 622 seconds to complete the standard drawings. The fastest user of the Tracker 2000 was able to produce the drawings in 196 seconds, while the slowest required 415 seconds. Both the Tracer and Tracker 2000 required more than the average across devices for half of the participants, while the HeadMaster required more than the overall average for only one participant.

Accuracy of Drawing Production

There did not appear to be any systematic differences in accuracy using these devices. Of the 22 errors in drawings identified across participants, virtually all were consistent across devices for a given participant, indicating that the errors were caused by the participant's perception of the drawings rather than any specific device's ability to reproduce the drawings. One participant, for example, consistently selected an incorrect fill pattern from the palette on each drawing, regardless of input device. The consistency of this single error suggests that these errors were due to a perceptual difficulty of this participant rather than any cognitive load of the input devices.

Subjective Data

Where the objective measurements failed to show a consistently superior device, there was a clear preference for use by our participants. The Tracker was said to be "jumpy," and very fast by three of the six participants. Overall, it was reported the most comfortable device used in the study. As noted earlier, we had placed the reflective dot of this system on the hat band of a baseball cap, and all six of our research participants considered using the baseball cap to be a very comfortable way to complete the task, and five especially felt that the hat was a comfortable way to don the device. Five out of the six participants reported the Tracker as being the most comfortable device in the study.

The HeadMaster Plus was reported to be the easiest to use by four out of six participants. Five of the six participants commented that the HeadMaster Plus was uncomfortable to wear. Of these five, three complained of the headset becoming uncomfortable because of tightness and pinching over a period of time. Two out of the five complained of neck discomfort because of the increased bending, turning and twisting needed for sufficient arrow control. One research participant found no discomfort with the HeadMaster Plus at all.

Five participants complained of the arrow drifting either up or down the screen when using the Tracer. Two of the six were fastest using the Tracer but, when drifting occurred, the headset needed to be removed for 10 seconds in order to "recalibrate" the device. Three participants reported the Tracer was difficult to use. Four participants reported that the headset was uncomfortable, while one felt this to be very comfortable to wear. Another felt that it was comfortable except that it loosening consistently, and had to be readjusted.

When the participants were asked, should they require a head pointer for computer access, which they would prefer, all six participants preferred the HeadMaster Plus and the Tracker 2000 over the Tracer, and five of the six preferred the Tracker 2000 overall.

Discussion

Although the results of this study do not provide a clear leader in performance, they do bear clinical relevance for the therapist trying to select an appropriate input device, and provide guidance to the manufacturers of the devices.

The results indicate that, while considered the most comfortable of the head pointing devices, the Tracker 2000 is generally somewhat slower than the Headmaster and Tracer. However, over the short term, the participants were willing to trade some performance for increased comfort, as shown by the clear preference for the Tracker 2000 as the long term input device, even when the dot was applied to the band of a hat rather than directly to the skin. We also noted that, when participants are having difficulty controlling the Tracker 2000, they tended to lean toward the computer. Since this magnifies the movement of the reflective dot in the camera's view, it makes control more difficult. When teaching clients to use this device, it is important to watch for this behavior. (It is also possible that this will not be an issue in individuals for whom the device is intended, since they may lack the trunk control required to lean forward.)

During setup of the devices, the sensitivity (mouse speed) was adjusted so that the movement of the pointer on the screen of the computer matched the researchers' head movements closely. In spite of this, the HeadMaster Plus was found to be uncomfortable due to excessive turning, twisting, and bending of the neck by two out of the six participants. One of these participants used the HeadMaster Plus first in the device sequence and may not have learned the technique to recenter the mouse cursor, which would explain the comment. The other subject who had this complaint used the HeadMaster Plus later in sequence, making the reason for discomfort unclear.

The HeadMaster produced the most consistently fast input times of any device in the study. For each participant, the Headmaster was either the fastest or second fastest device, and when it was second fastest, it was within 10% of the times of the fastest device. Although the HeadMaster is the oldest of the input devices in the study, having been first introduced in 1985, it is apparent that the basic function of the Headmaster has been excellent from the beginning. The primary limitations of the HeadMaster are the comfort and appearance of the headset, and the wire linking the participant to the computer. (The remote option removes this link, but at a cost of $575)

The preproduction Tracer used in this study was frustrating to participants because of the occasional occurrence of drift throughout the trials. Based on communication with the manufacturer about this device, the manufacturer did additional development work to correct this problem, and provided a second unit for testing. While this unit was received too late for inclusion in our data collection, the researchers did test the device extensively, and were not able to produce any drift. They were also unable to detect any loss of performance due to this correction. This suggests that production units should perform at least as well as the study unit at mastery and, since mastery was delayed in two of our participants by the drift, that the time to master the Tracer should be similar to that of the other devices.

In terms of time to produce the drawings, the maximum performance of participants using the HeadMaster and Tracer were very similar, with the Tracker somewhat slower. This suggests that both of these devices are able to respond to user input faster than the user is able to produce input, in most cases. Further development of speed of response is not indicated for either device, but users of each complained of discomfort of the head gear required. Further development in headgear is therefore indicated for both the HeadMaster and the Tracer. Despite the high quality of performance the HeadMaster Plus and the Tracer offer, the Tracker was preferred overall because of the level of comfort it provided. The participants in this study held comfort at a higher value than performance.

Conclusion

In general the results of the study showed that, when performance is similar, the participants preferred comfort over performance. Differences in performance of between 30 and 40% were not great enough on this drawing task to make participants willing to tolerate decreased comfort. In this study, there were no rewards based on performance, so this finding might be different when employment is based on performance.

Because of this, the makers of the HeadMaster Plus and Tracer could justify additional efforts to develop more comfortable headgear for their devices. The current headset of the HeadMaster, which is based on the Shure SM-10a microphone headset, is substantially more comfortable and stable than the original headset introduced in 1985, but, according to our participants, continues to apply too much pressure over too small an area on the sides of the skull. If a headset could be developed that increased the bearing area, or decreased pressure without loss of stability could be developed, the device might be considered to be more comfortable. The "visor" of the Tracer has a small foam pad which presses against the forehead. When the restraining strap is tight enough to provide stability, several participants complained of a red mark on their foreheads that persisted for some time after completion of a trial. Again, a system that increased bearing area or decreases pressure without loss of stability would seem to be desirable for this device. Both of these devices offered somewhat faster input times than the Tracker 2000 for some users, but neither was preferred because of comfort. Boost Technologies, the manufacturer of Tracer, has already invested engineering time in solving the drift problem based on the preliminary findings of this study, and may want to reconsider the sports visor design based on the comments of our participants.

Although we used able-bodied participants in this study, we feel confident that the results would be similar for the target population of people with disabilities. These devices are intended to be used by individuals who have lost hand control, but retain head control. Since these individuals have good head control, it should be similar to that of an able-bodied person. In some cases where head-control is marginal, the weight of the headsets of the Tracer and HeadMaster Plus may reduce performance using these devices.

We did not examine the methods of producing mouse clicks in this study. The HeadMaster includes a puff switch on the Headset. The Tracer has wireless connections for two switches, but does not include any way to mount switches, since that will vary by individual. The Tracker 2000 has connections on the camera for two switches, but it is left to the user to find a way to connect the switch to the user. In this study, we used external switches operated by hand by our participants on all three devices. The means of producing clicks might affect the choice of head pointing device in some cases.

Therapists involved in the provision of assistive technology must not only determine the broad category of device that will meet the needs of the client, but must also make specific recommendations for the device that will best meet the needs of the user in the intended working environment. Each client will balance cost and cosmetics differently, but the therapist should be prepared to provide information on the relative functionality of the devices under question. The results of this study should help the therapist to provide that information.

References

Americans with Disabilities Act of 1990, S. 933 (1990) Retrieved from http://www.usdoj.gov/crt/ada/pubs/ada.txt [ May 16, 2002 ]

Angelo, J, Deterding, C. & Weisman, J. (1991). Comparing three head-pointing system susing a single subject design. Assistive Technology, 3, 43-49.

Anson, D. K. (1997). Alternative Computer Access: A Guide to Selection . Philadelphia , F.A. Davis.

Anson, D. K. (2001) Assistive Technology. In Pedretti, L. W. & Early, M. B. (Eds.), Occupational Therapy: Practice Skills for Physical Dysfunction (5 th ed.) (pp. 260). St. Louis : Mosby.

Bloom , M., Fischer , J & Orme , J. (1995). Evaluating practice: guidelines for the accountable professional (4 th ed.). Boston , MA.: Allyn and Bacon

Boost Technology. (1998). The boost tracer. Retrieved from www.boosttechnology.com [ January 31, 2002 ]

Department of Justice. ( July 26, 2001 ). ADA regulations and technical assistance materials. Retrieved from http:/www.usdoj.gov/ert/ada/publicat.htm [ May 16, 2002 ]

DeVries, R. C., Deitz, J. & Anson, D. (1998). A comparison of two computer access systems for functional text entry. American Journal of Occupational Therapy, 52 , 656-665.

Individuals with Disabilities Education Act of 1997 (1997) Retrieved from http://www.ed.gov/offices/OSERS/Policy/IDEA/IDEA.pdf [ May 16, 2002 ]

Jacobs, S. (1999). Section 255 of the telecommunications act of 1996: Fueling the creation of new electronic curbcuts. Retrieved from http://www.accessiblesociety.org/topics/technology/eleccurbcut.htm [ May 15, 2002 ]

Kanny, E. & Anson, D. (1992). A pilot study comparing mouse and mouse-emulating interface devices for graphic input. Assistive technology. p.50-58

Kaye, S. (2000). Computer and Internet use among people with disabilities, Disability Statistics Report (13). Washington , DC : US Department of Education, National Institute of Disability and Rehabilitation Research.

Madentec. (n.d). Solutions for people with disabilities. Retrieved from www.mandentec.com [ January 27, 2002 ]

Prentke Romich Company. (2002). Headmaster plus. Retrieved from www.prentrom.com [ January 27, 2002 ]

The Rehabilitation Act Amendments of 1998 (1998) Retrieved from http://www.tcb.state.tx.us/rehabact.asp [ May 16, 2002 ]

United States Department of Justice. ( July 26, 2001 ). ADA regulations and technical assistance materials. Retrieved from http:/www.usdoj.gov/crt/ada/publicat.htm [ March 4, 2002 ]

Vanderheiden, G. C. (1985) Alternate access to all standard computers for disabled and non-disabled users. In C. Brubaker (Ed.), Proceedings of the Eighth Annual Conference on Rehabilitation Technology (pp.279-281) Washington , DC : RESNA Press.

Wheeless, C. R. (1996). Wheeless Textbook of Orthopaedics. Retreived from http://www.medmedia.com/o14/108.htm [ May 16, 2002 ]


Boost Technology. (1998). The boost tracer. Retrieved from www.boosttechnology.com [January 31, 2002]

Department of Justice. (July 26, 2001). ADA regulations and technical assistance materials. Retrieved from http:/www.usdoj.gov/ert/ada/publicat.htm [May 16, 2002]

DeVries, R. C., Deitz, J. & Anson, D. (1998). A comparison of two computer access systems for functional text entry. American Journal of Occupational Therapy, 52 , 656-665.

Individuals with Disabilities Education Act of 1997 (1997) Retrieved from http://www.ed.gov/offices/OSERS/Policy/IDEA/IDEA.pdf [May 16, 2002]

Jacobs, S. (1999). Section 255 of the telecommunications act of 1996: Fueling the creation of new electronic curbcuts. Retrieved from http://www.accessiblesociety.org/topics/technology/eleccurbcut.htm [May 15, 2002]

Kanny, E. & Anson, D. (1992). A pilot study comparing mouse and mouse-emulating interface devices for graphic input. Assistive technology. p.50-58

Kaye, S. (2000). Computer and Internet use among people with disabilities, Disability Statistics Report (13). Washington, DC: US Department of Education, National Institute of Disability and Rehabilitation Research.

Madentec. (n.d). Solutions for people with disabilities. Retrieved from www.mandentec.com [January 27, 2002]

Prentke Romich Company. (2002). Headmaster plus. Retrieved from www.prentrom.com [January 27, 2002]

The Rehabilitation Act Amendments of 1998 (1998) Retrieved from http://www.tcb.state.tx.us/rehabact.asp [May 16, 2002]

United States Department of Justice. (July 26, 2001). ADA regulations and technical assistance materials. Retrieved from http:/www.usdoj.gov/crt/ada/publicat.htm [March 4, 2002]

Vanderheiden, G. C. (1985) Alternate access to all standard computers for disabled and non-disabled users. In C. Brubaker (Ed.), Proceedings of the Eighth Annual Conference on Rehabilitation Technology (pp.279-281) Washington, DC: RESNA Press.

Wheeless, C. R. (1996). Wheeless Textbook of Orthopaedics. Retreived from http://www.medmedia.com/o14/108.htm [May 16, 2002]

Please send all correspondence regarding this article to Denis Anson. Phone: 570-674-6413, Fax: 570-674-3052, email: danson@misericordia.edu

This research was performed in partial completion of the educational requirements of all but the first author.

Origin Instruments Corporation. 854 Greenview Drive, Grand Prairie, TX 75050 Fax: 972-606-8741 support@orin.com

Madentec. 9935-29A Avenue, Edmonton, Alberta, Canada T9N 1A9. Phone: 780-450-8926 Fax: 780-988-6182 www.madentec.com

Eye Control Technologies. 33872 Eastgate Circle SE, Corvallis, OR 97333 Fax: 541-753-6689 sales@naturalpoint.com

Prentke Romich Company. 1022 Heyl Road, Wooster, OH 44691. Fax: 330-263-4829 www.prentrom.com

Boost Technology. 1601 Ocean Avenue, San Francisco, CA 94112 info@boosttechnology.com

The TrackIR came into existence during the course of this study, therefore we do not have any performance data on this device.

As this paper was being prepared, Tracker 1 was introduced at $695. Additional information can be obtained through the manufacturer.

Able Net Inc. 1081 Tenth Avenue S.E. Minneapolis, MN 55414. Phone: 1-800-322-0956 www.ablenetinc.com

Deneba Software, 1150 NW 72 nd Avenue, Miami, FL 33126, Executive Tower 1, Penthouse Floor, Phone: (305) 273-9069

During the data analysis phase, we discovered that the Tracer being used in this study (a preproduction unit) had an intermittent drift problem that made fine control very difficult. The manufacturer provided a second unit in which the drift had been corrected, which allows much better control, but this was too late for inclusion in the study. Without the drift, the Tracer control seems to match that of the HeadMaster. In applying the results of this study to clinical practice, it should be recognized that the production models of the Tracer should be at least as good as found here, and probably significantly better.