UNDERSTANDING HOW TO IMPROVE THE ACCESSIBILITY OF COMPUTERS THROUGH CURSOR CONTROL STUDIES

ABSTRACT

People with motion-impairments often find it difficult to perform many of the actions required to interact with a computer. This paper presents the results of an on-going series of experiments designed to understand how using force feedback affects interaction for motion-impaired users. Point and click tasks were analyzed using new cursor control measures. The results showed significant improvement in throughput for all users with force-feedback and the cursor control measures were effective in capturing the differences between the conditions.

Keywords

force feedback, motion-impaired users, cursor studies

INTRODUCTION

Computer use often involves interaction with a graphical user interface (GUI), typically using a keyboard and mouse. For people with reduced physical capabilities, this can be difficult, if not impossible [2]. Conditions such as cerebral palsy, muscular dystrophy, and spinal injuries can give rise to many symptoms, such as tremor, spasm, and reduced strength, and can restrict the user’s performance in typical GUI tasks, such as "pointing and clicking".

Previous studies have shown haptic feedback, to be beneficial for able-bodied users in point and click tasks, using both gravity wells [4] and haptic tunnels to targets [1]. Studies have shown that haptic feedback also offers noticeable improvement in point and click tasks for motion-impaired users [2].

UNDERSTANDING CURSOR MOVEMENT

Analyses of "point and click" interactions are common, but tend to focus on gross measures such as movement time and error rate. Although these measures may establish that differences exist, they are limited in explaining why they exist. MacKenzie et al. [3] proposed seven new accuracy measures to evaluate computer pointing devices. The measures are intended to elicit subtle differences through an analysis of the cursor movement along the cursor path:

Supplementary measures of cursor movement include: movement time, effective target width and throughput:

Motion-impaired users are more likely to press the input device buttons prematurely or simply by accident. Consequently, an additional potential measure is:

EXPERIMENTAL METHOD

The study involved five volunteers (Table 1) with a range of motion impairments. For benchmark comparison, data from three able-bodied users (CU1-3) were also collected. The aim was to understand in detail the mechanisms by which gravity wells assist motion-impaired users in selecting targets.

Participants performed a series of multidirectional point and click tasks. 16 circular targets were arranged in an equidistant circular layout, Figure 1. A sequence of trials began when a participant clicked in the target at the top of the circle. The next selection was the target directly opposite and so on clockwise round the targets. Each sequence comprised 33 target selections, and was repeated both with and without haptic assistance, using a Logitech Wingman force-feedback mouse. Haptic assistance was provided by gravity wells around the targets. When the cursor entered the well, a spring force pulled the mouse toward the center of the target until the target was selected.

RESULTS AND CONCLUSIONS

The cursor movements were recorded and analyzed using the cursor measures previously described. Figure 1 shows example cursor paths for (a) an able-bodied user, (b) a motion-impaired user, both without gravity well assistance and (c) the same motion-impaired user with assistance.

A 2 factor repeated measures ANOVA showed that there was a significant difference between the conditions with and without gravity well assistance (F=40.77, df=1,959, p<.001). There were significant decreases in means for missed clicks (MCL), target re-entries (TRE), movement direction changes (MDC), movement time (MT) and effective width (W_e), and an increase in throughput (TP), particularly for the motion-impaired users. The other measures were very similar for both conditions, suggesting that they were incapable of distinguishing between them.

The experiment has shown that gravity well assistance not only produces significant reductions in movement time for this task, but also significant increases in throughput. The cursor measures offer insights as to the mechanisms by which the throughput improves. The inherent circling of targets typical of motion-impaired users, Figure 1(b), is reduced and this is shown by the reductions in TRE and the ability to select a target accurately is shown by the reductions in CL and W_e. Thus the cursor measures are likely to be useful in future experiments.

Figure 1. Cursor traces and the target layout:

(a) CU1 unassisted; (b) PT7 unassisted; (c) PT7 assisted

FUTURE WORK

Although the experiment described is suitable for demonstrating the use of the cursor measures for interpreting cursor movement for motion-impaired users, it is insufficient to provide the sole basis for more accessible computers. Further work is continuing looking at adding gravity to targets in more realistic tasks, such as multiple targets in close proximity to each other. Other methods of haptic support such as tunnels to targets [1] are also being studied. Finally, new measures of cursor control are being developed and evaluated throughout these experiments.

REFERENCES

1. Dennerlein, J.T., Martin, D.B., and Hasser, C. Force-feedback improves performance for steering and combined steering-targeting tasks, in Proceedings of CHI 2000 (The Hague, 2000) 423-429.

2. Keates, S., Langdon, P., Clarkson, P.J., and Robinson, P. Investigating the use of force feedback for motion-impaired users, in Proceedings of the 6th ERCIM Workshop (Florence, Italy, 2000), 207-212.

3. MacKenzie, I.S., Kauppinen, T., and Silfverberg, M. Accuracy measures for evaluating computer pointing devices, in Proceedings of CHI 2001, (Seattle, WA, 2001), 9-15.

4. Oakley, I., McGee, M.R., Brewster, S. and Grey, P. Putting the feel in ‘look and feel’, in Proceedings of CHI 2000 (The Hague, 2000), 415-422.

Copyright © 2002 by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept, ACM Inc., fax +1 (212) 869-0481, or permissions@acm.org