Author: Wong, Wing-kai.
Title: Parameters for design of computer-based program for people with mental retardation
Degree: Ph.D.
Year: 2007
Subject: Hong Kong Polytechnic University -- Dissertations
People with mental disabilities -- Rehabilitation
Computers and people with disabilities
Computerized self-help devices for people with disabilities
Department: Department of Rehabilitation Sciences
Pages: xix, 302 leaves : ill. (some col.) ; 30 cm.
Language: English
Abstract: “Universal design” is a fundamental approach to the development of the human-computer interface for modern technology. The principles of “Universal design” have largely improved the user-friendliness and utilization of technology for those in the mainstream population. The drawback however is the compromise of the functionality and even the marketability of these products, especially by the people who have had disabilities. Previous studies have revealed that the people with mental retardation had limitations in using the existing human-computer interfaces, e.g. web browser. Such findings are not surprising as the people with mental retardation are known to have an overall lowering of intellectual abilities. Previous studies have placed its focus on investigating the functional problems encountered by these people and the ways for improving their skills on using these technological products. However, it is a limited research to explore the mechanisms which hinder them from effectively using the technological products and how the interface design could be enhanced for bridging the theory-driven human-computer interaction gap. This research project was aimed to examine the mechanisms behind the phenomenon that the people of mental retardation had problems using the human-computer interface. In particularly, it focused on how the mental deficiencies of these people hindered their competence in operating the interface. It was consisted of three inter-related studies with each informed the research questions asked and methods to be used in the subsequent study. The research question of the study 1 was: what was the performance level of the people with mental retardation on operating a common human-computer interface program? A total of 57 people with mild (n= 30) or moderate (n=27) mental retardation (39 males and 18 females; Mean Age=17.2 years (SD=3.3)) were recruited through convenience sampling to participate in an Internet Explorer (IE) competence test. The competence test was composed of 16 IE tasks (161 subtasks) which were verified by an expert panel. Participants' performances on each task were evaluated against a 4-point scale. The results of the participants' scores on each of the task were analyzed. Results of logistic regression (stepwise forward) procedure indicated that the “general motor function” (B=-3.43, SE=1.59) and “use customized bookmark” (B=-3.54, SE=1.06) were the most predictive of the participants' performance (X2= 38.35, df=2, p<0.001). Hierarchical cluster analysis, decision tree classification and one-way ANOVA further indicated that participants' performance varied which can be clustered into three performance groups. The “general motor function” was useful for differentiating subjects into the low and medium competence groups (F(2,54)= 89.54, p<0.001) whilst the “use customized bookmark” (F(2,54)= 57.02, p<0.001) into the medium and high competence groups. Results of another expert panel review on analyzing the content of the task and subtasks composed of the IE comptence test indicated that comparatively less motor than cognitive abilities were required for performing the IE tasks. Among the list of the cognitive abilities, orientation, visual acuity, and attention had the highest frequency count followed by word recognition and working memory. The research question of the study 2 was: with the performance levels derived from Study 1, what were the core cognitive abilities predictive of these people's performance level? A total of 62 people with mild (n=33) or moderate (n=29) mental retardation (40 males and 22 females; Mean Age: 17.4 years (SD=6.0)) were recruited to establish the cognitive and motor profiles of these people. They were invited to participate in the Internet Explorer (IE) competence test as well as complete 13 tests on attention, visual-spatial, memory, executive functions, frontal lobe functions, word recognition, and sensori-motor functions. The participants' scores on all of these tests were significantly correlated with their overall IE performance score (r=0.36 to 0.71, ps<0.01). Results of stepwise multiple regression analyses indicated that, among all functions, the attention and visual search function measured by the Symbol Digit Modalities Test (R2 = 49.7%), fine motor functions measured by the McCarron Assessment of Neuromuscular Development (R2 = 5.8%), and Chinese word recognition functions measured by the Chinese Characters Test (R2 = 3.2%) were the best three predictors on overall IE performance [total R2 = 0.59, Model F(3, 58) =27.40, p<0.001].
The question addressed by Study 3 was: what were the underlying mechanisms hindered these people from being competent on operating the interface of Internet Explorer? To further contain the scope of this study, based on the results obtained from Study 2, we selected only the visual search processes and designed an eye-tracking paradigm for comparing the efficiency on various distances and orientations between the people with mental retardation and their normal counterpart. A total 24 people with mild or moderate grade of mental retardation (16 male and 8 female; Mean Age=19.0 years (SD=5.4)), and 30 people without mental retardation (16 male and 14 female; Mean Age=20.8 years (SD=1.9)) as the reference group were recruited. We hypothesized that the people with mental retardation, with limited attention and working memory ability, would perform the visual search task differently than their normal counterpart. Their performances would be further modulated by the distances (close: 1 visual lobe, medium: 1.5 visual lobes, and far: 2 visual lobes) and orientation (vertical, horizontal and oblique) between the two target stimuli. This search task required the participant to visually search the target square (one) or squares (two) and indicate as quickly as possible the number of target square(s) which they found. The total number of trials was 432 divided into 24 blocks. Their eye movements were tracked by the Eye-gaze Response Interface Computer Aid (ERICA) infrared corneal reflection system at the time of performing the task. Their visual lobe areas and shape characteristics were measured with the Visual Lobe Measurement System (VILOMS). Results indicated that participants with mental retardation had significantly smaller visual lobe areas (t(38)=4.13, p<0.001), and less likely using consistent search strategies (p<0.001) and lower performances (ps<0.05) on the visual search task, than their normal counterpart . The results on testing the Distance x Orientation effects indicated that, the Distance effect on the scanpath duration and overall fixation duration for both two groups were statistically significant (Pillai's Trace: F(2,22)=11.81, p<0.001 and F(2,22)=7.71, p=0.003 respectively for participants with mental retardation; and F(2,28)=43.69, p<0.001 and F(2,28)=33.18, p<0.001 respectively for normal counterparts). Similarly, the Orientation effect on these two variables for both groups was significant (Pillai's Trace: F(2,22)=5.97, p=0.008 and F(2,22)=4.91, p=0.017 respectively for participants with mental retardation; and F(2,28)= 54.15, p<0.001 and F(2,28)= 41.38, p<0.001 respectively for normal counterparts). The interaction effect between Distance and Orientation was not significant for participants with mental retardation (Pillai's Trace: F(4,20)=1.70, p=0.189 and F(4,20)=1.65, p=0.20 respectively). However, this interaction effect was significant for normal counterparts (Pillai's Trace: F(4,26)=4.85, p=0.005 and F(4,26)=4.65, p=0.006 respectively). Post hoc analyses indicated that normal participants had significantly longer scanpath duration and overall fixation duration in far- than close-distance when the targets were arranged in vertical and oblique orientations (ps<0.001), and in medium- than close-distance when the targets were arranged in vertical orientation (p<0.001). In the close-distance condition, oblique and horizontal orientations had significant longer overall fixation duration than their vertical counterpart (ps<0.001). The condition with a far-distance and oblique-orientation had the longest scanpath duration and overall fixation duration among all of nine conditions. The findings suggest that a plausible account of why people with mental retardation, when compared with the normal counterpart, would have problems when using the human-machine interface. As the people with mental retardation have an overall lowering of motor and mental abilities, this study chose to focus on the mechanisms underlying the deficiencies in cognitive abilities. The most limiting factor as reflected from the findings is that the people with mental retardation had a decline in the visual search function. In the model which was adopted in this study, the problem possibly began in the comparatively smaller visual lobe in these people. This would largely limit their visual attentional field. With a smaller visual field, the people with mental retardation would need to shift their attention (like a spot-light) from one to another field more often than their normal counterpart. This couples with the notion that they had a decline visual working memory. These deficiencies are reflected from their lack of search strategies and longer time for searching information from the visual display than the normal participants. The decline in working memory would prevent the people with mental retardation from recalling the locations which had been previously scanned through. The consequence was that these people took longer time to search for the second target stimulus. The longer fixation time also suggested that they took longer time to process the target stimuli which perhaps was also accounted for by their lower spatial working memory. The findings on the significant effects of distances and orientations on influencing participants' search and fixation times offers insights into the possibility of using these as guidelines for design of human-machine interface e.g. menus for the people with mental retardation. These also stimulate further research on the feasibility of an inclusive approach to universal design of modern technology for our population.
Rights: All rights reserved
Access: open access

Files in This Item:
File Description SizeFormat 
b21656721.pdfFor All Users2.93 MBAdobe PDFView/Open


Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show full item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/5399