Practical Considerations in Human-Computer Interaction for e-learning Systems for People with Cognitive and Learning Disabilities

‫‪Practical Considerations in Human-Computer Interaction for e-Learning‬‬ ‫‪Systems for People with Cognitive and Learning Disabilities‬‬ ‫‪Mark P. Wa...
Author: Sabrina Wiggins
1 downloads 0 Views 397KB Size
‫‪Practical Considerations in Human-Computer Interaction for e-Learning‬‬ ‫‪Systems for People with Cognitive and Learning Disabilities‬‬ ‫‪Mark P. Wachowiak, Renata Wachowiak-Smolikova, Grant D. Fryia‬‬ ‫‪Department of Computer Science and Mathematics‬‬ ‫‪Nipissing University‬‬ ‫‪North Bay, ON P1B 8L7 Canada‬‬ ‫‪{markw, renatas}@nipissingu.ca, [email protected]‬‬

‫‪ARABIC ABSTRACT‬‬ ‫‪.‬تابساحلا مظن ةوقو ةءافك يف ًارمتسم ًاروطت كانه ناك ةيبوساحلا ةزهجألا ةعانصو بساحلا جمارب يف مدقتلا ببسب‬ ‫ريثكلا ىلع ةبوعص يأ لكشيال ًةداع ديقعتلا اذه نإف كلذ نم مغرلاب و ‪ً.‬ادقعمو ًاريبك حبصأ مظنألا هذه نم ريثكلا نكلو‬ ‫مهيدل يذلا صاخشألا ‪،‬صخألا ىلعو ‪.‬ةصاخلا تاجايتحالا يوذ نم صاخشألل تابوعصو زجاوح ينبي هنكلو نيمدختسملا نم‬ ‫نمم ةصاخلا تاجايتحالا يوذل تاهجاولا ميمصت ئدابم نإف ‪،‬كلذ ىلإ ةفاضإلاب ‪.‬ملعتلا يف لكاشم مهيدل وأ ةينهذلا ةقاعإلا‬ ‫‪،‬ةقرولا هذه يف ‪.‬ةيدسج ةقاعإ مهيدل نمل تاهجاولا ميمصت ئدابم نع فلتخت ملعتلا ىلع ةردقلا يف ةقاعإو ةينهذ ةقاعإ مهيدل‬ ‫ثاحبألا نيمضتل جهنم ميدقتب موقت ةقرولا هذه نإف ‪،‬كلذ ىلإ ةفاضإلاب ‪.‬اهحيضوتو اهتشقانمو تاقورفلا هذه فصوب موقن‬ ‫كلذو )‪(assistive technology‬ةدعاسملا تاينقتلل هجوم كلذو لوصولا ةلوهس ةيحان نم تنرتنالا ةكبش تاهجاوب ةثيدحلا‬ ‫تايساسأ ىلع ينبم يبيرجت ماظن ريوطت مت دقل ‪.‬ينورتكلالا ميلعتلا ةمظنأل مادختسالا ةلوهس ةيصاخ ريوطت فدهب‬ ‫يوذ صاخ لكشب فدهتسي ماظنلا اذه ‪ (rigorous assistive technology).‬ةمراصلا ةدعاسملا ةينقت ىلع ةينبملا ميمصتلا‬ ‫ةيساسألا تافصلا ريوطت مت ثيحب ملعتلا ىلع ةردقلا يف ةقاعإ مهيدل نممو ةينهذ ةقاعإ مهيدل نمم ةصاخلا تاجايتحالا‬ ‫نيفدهتسملا صاخشألا تابلطتمب يفي ينورتكلالا ملعتلا ةمظنأ تائيب نم عونلا اذه ‪.‬اهديقعت نم ليلقتلاو تاهجاولل‬ ‫‪.‬ةيلبقتسم ثاحبأل ةيادب ةطقن نوكيو‬

‫‪International Journal of Information Studies    Volume 2  Issue 1    January 2010‬‬

‫ ‪60‬‬

Practical Considerations in Human-Computer Interaction for e-Learning Systems for People with Cognitive and Learning Disabilities Mark P. Wachowiak, Renata Wachowiak-Smolikova, Grant D. Fryia Department of Computer Science and Mathematics Nipissing University North Bay, ON P1B 8L7 Canada {markw, renatas}@nipissingu.ca, [email protected] ABSTRACT: Due to advances in both hardware and software, computer systems have exhibited continuous improvement in efficiency and power. However, many of these systems are also becoming larger and increasingly more complex. Although such complexity usually poses no difficulties for many users, it often creates barriers for individuals with disabilities. People with cognitive and learning disabilities are particularly affected by computer system complexity. Furthermore, the issues in designing interfaces for computer systems for these individuals are different from the design considerations of interfaces for people with physical disabilities. In this paper, these differences are described, elaborated, and illustrated. The paper also presents an approach to incorporate recent research into web interface accessibility for assistive technology, with the goal of improving the usability of an e-Learning system. A prototype system based on rigorous assistive technology design principles is developed, targeted specifically to people with cognitive and learning disabilities, wherein essential features are enhanced, while complexities are minimized. This type of e-Learning environment can meet the needs of its intended users and provide a suitable initial point for further research and development. Keywords: Human-computer interaction, Assistive technology, E-Learning Received: 21 June 2009, Revised 18 July 2009, Accepted 28 July 2009 1. Introduction How well and how efficiently a computer system performs are obvious and essential quality indicators. It is expected that such systems exhibit error-free computation with high speed. However, robustness and efficiency, important as they are, are not the sole quality metrics. A system will not serve its purpose unless its intended users can exploit its functionality. The recognition of the need for people to use computer resources in as straightforward a manner as possible has led to the development of the multidisciplinary field of human computer interaction (HCI), an important and active research area involving computer scientists, social scientists, cognitive scientists, ergonomics and human factors specialists, designers and artists, educators, and many others. Clearly, HCI has a direct impact on users and how they perform their work (Maxwell, 2000). Of particular importance are issues of usability, which is the degree to which the design of a particular user interface takes into account the human psychology and physiology of its users. Designing for usability was first recognized in the 1960s and 1970s, when the focus concentrated on body movements (e.g. clicking an icon with a mouse). During this time, “userfriendliness”, human factors, and human-machine coupling also became part of HCI. Cognitive science began to make contributions to HCI in the 1970s and 1980s, when HCI shifted focus from the physical body to the mind. From the 1990s to the present, HCI has become more concerned with communication between people with the rise of ubiquitous computing, as reflected in the growth of networking and the Internet. It was during this period that the social sciences, such as anthropology and sociology, began to contribute to HCI. The focus is now on relationships, and less on modeling the user (Harper, Rodden, Rogers, & Sellen, 2008). In all this development in HCI, however, issues of usability first addressed in the early years of computing retain paramount importance for people with disabilities. HCI therefore has particular importance in assistive technology. Features that pose no difficulty for most users may be difficult or impossible to use for people with disabilities. Such disabilities include visual and aural impairments, cognitive and learning disabilities, and many others. In fact, HCI considerations are particularly important for people with cognitive and learning disabilities (CLDs). These people, and those with disabilities in general, can greatly benefit from web-based access to public services (e-Government) and professional activities (tele-commuting/ working) (Abascal & Civit, 2002). There is also considerable interest in HCI in the education community, especially from International Journal of Information Studies    Volume 2  Issue 1    January 2010

61

those who work with individuals with disabilities. One specific computerized system that is very useful for such people is ­e-Learning (electronic learning), in which instruction is mediated through computer technology (Abascal & Civit, 2002). Used in conjunction with traditional instruction, e-Learning allows students to take specialized courses using Internet technology without having to travel possibly long distances to specialized schools. One of the primary goals of e-Learning is to improve educational access for all individuals. Unfortunately, such systems usually include many features and options that may be too complex for individuals with CLDs. This paper details the creation process of a prototype e-Learning system for people with CLDs, with the goal of alleviating over-complexity by reducing the number of features presented to the user at any one time, and by retaining and enhancing crucial features without negatively affecting overall functionality. Recent investigations support an emphasis on simplicity and on matching features to learners’ needs, and on eliminating unnecessary features (Catarci, De Giovanni, Gabrielli, Kimani, & Mirabella, 2008). Research into web accessibility guidelines (Friedman & Bryen, 2007) are incorporated and adapted to the e-Learning system. The e-Learning interface proposed in this paper meets 15 of the top 22 recommendations, which were considered as the most appropriate for this specific application. Particular emphasis is given to usability and to easing cognitive strain, which are affected by interface complexity, breadth, and depth. In previous work, the authors discussed these web accessibility guidelines in detail, and proposed a web-based ­e-Learning interface incorporating them (Fryia, Wachowiak-Smolikova, & Wachowiak, 2009). The current work focuses on the human-computer interaction components involved in the design of the prototype system, expands and enhances the interfaces previously presented, and supports design decisions through a survey of users of assistive technology. Specifically, the goal is to employ rigorous accessibility guidelines to develop an interface, or “shell”, that can be used in conjunction with many (if not all) existing e-Learning systems. The target group of users comprises individuals who require extra help accessing and organizing information, such as people with learning disabilities, high functioning cognitive impairments, and Asperger’s Syndrome. This target group consists of, but is not limited to, students in junior high to high school, ages 12 to 18. Guidelines such as those described in this paper can also be applied to the design of interfaces for other web-based applications. The approach presented here separates the application level from the interface level. Such separation is preferable to adapting existing systems in a “patchwork” manner (Abascal, 2002). 2. Human-Computer Interaction and Assistive Technology 2.1 The “Abilities” Continuum and Universal Design There is usually not a clear distinction between computer users who have “disabilities” and those who do not. In fact, computer users exhibit a continuum of abilities, and some positions on this continuum require special interface considerations. For instance, as people age, changes in vision necessitating the use of bifocals may make displays with varying font sizes difficult to read, and hearing loss may make subtle audio cues more difficult to recognize (Bergman & Johnson, 2009). It can even be said that vis-à-vis computer technology, all users have some type of “disability”, and that there is really no such thing as an “average user” (Bergman & Johnson, 2009; McMillan, 1992). One way to address this “abilities continuum” is the concept of “universal design”, wherein interfaces are designed for the broadest range of people possible, and not for the fictitious “average user”. Examples from everyday life include automatic doors and remote control thermostats (Bergman & Johnson, 2009). Although universal design does not aim for complete accessibility, by redefining the concept of a “user”, greater accessibility can be provided to more people with a greater range of abilities without significant redesign efforts (Bergman & Johnson, 2009; Vanderheiden, 1992). 2.2 Human-Computer Interaction in Assistive Technologies In computerized systems, interface features that pose no difficulty for people without disabilities may be nearly or completely impossible for people with impairments to use. A classic example is the move towards fully graphical interfaces, which had previously prevented visually impaired users from using the computer because text readers no longer worked with graphical representations (Friedman & Bryen, 2007). Recently, much progress in accessibility for the visually ­impaired has been made. With the use of screen readers – applications that interpret on-screen text for presentation to other output devices, such as textto-speech converters – these users have access to applications that employ graphical interfaces (­ Abascal & Nicolle, 2005). Universal design entails creating interaction mechanisms to be employed by all possible users. Systems would be created without unnecessary barriers, such as graphics, which cannot be represented as text, for the benefit of visually impaired users. The system requires a built-in screen reader that would eliminate the need for additional software. Inclusive design would

62

International Journal of Information Studies    Volume 2  Issue 1    January 2010

also eliminate compatibility problems associated with assistive technology. Inclusive design practitioners believe that by removing barriers on standard commercial systems, the need for assistive technologies would shift focus to alternative input and output methods, and individual expenses for disabled users would decrease, as the need for buying expensive assisted technology would be reduced (Abascal & Nicolle, 2005). There are many challenges that inclusive design must overcome (Abascal & Nicolle, 2005; Abascal & Azevedo, 2007; Lewis, 2007). Many designers do not know about this design approach, or they are unaware that by designing certain features differently, they can create a system that is accessible to a greater percentage of the population (Abascal & Nicolle, 2005; Abascal & Azevedo, 2007). Designers that are unfamiliar with the abilities of people with special needs often inadvertently place barriers on systems (Abascal & Nicolle, 2005; Abascal & Azevedo, 2007; Lewis, 2007). In addition, because there is diversity amongst disabilities, creating a system that is user friendly for everyone would be impossible (Savidis & Stephanidis, 2004). Artificial intelligence has contributed to advances in adaptive HCI. Much work has been done so that systems change and adapt to users (Microsoft Word 2007 is a prime example) (Abascal, 2002; Abascal & Azevedo, 2007; Lewis, 2007). Although the intent is to create a truly inclusive system, for many users with a cognitive disability, such a system may potentially add to their confusion and stress, as re-learning the interface is required (Lewis, 2007). 2.3 Cognitive and Learning Disabilities CLDs include a spectrum of disabilities, including mental retardation, autism, traumatic brain injury, aphasia (language impairments), dyslexia (a disorder primarily presenting as a reading and/or spelling impairment), and high-functioning autism (a type of autism in which intelligence is normal or above normal, but non-verbal communication, recognition of social contexts, and social skills are lacking). Alzheimer’s disease, attention deficit disorder and attention deficit hyperactivity disorder are also considered to be forms of CLDs (Keates et al., 2007). People with CLDs often exhibit deficiencies in attention, memory, perception, and problem-solving (Keates et al., 2007), affecting the manner in which they interact with computers. Cognitive impairments affect “thinking power” and IQ, usually lowering it. By contrast, learning disabilities are impairments which affect a person’s ability to acquire, process, or utilize knowledge at a level appropriate to his/her IQ (Keates et al., 2007). Possessing either or both of these types of disabilities can make accessing Web resources very difficult. In the United States, approximately 20 million people have a form of cognitive disability (Lewis, 2007). In a survey conducted between 1999 and 2004, 5.2% of Americans reported having a mental disability (Keates et al., 2007). Despite these high percentages, many web based systems are not designed with these people’s specific needs in mind. Until recently, computer accessibility for individuals with CLDs has not been as widely studied as for people with physical impairments. Individuals with CLDs exhibit a wide range of different abilities, from being able to live independently and knowing how to read and write, to needing round-the-clock supervision. Designing a system that is accessible for all intended users is very difficult, due to these varying levels of ability (Friedman & Bryen, 2007). To create an accessible system for these people, many issues must be addressed. First, will the interface be “deep” – that is, will there be many layers of menus or screens to go through to access certain applications, or will the interface be a “broad” – that is, most applications can be reached through only a few layers, with many features on each layer? Will the interface be rich with features that may prove helpful in assisting users, or will it include only the necessities (Lewis, 2007; Grynszpan, Martin, & Nadel, 2008)? In selecting a deep interface, each layer would contain only a few choices on each layer, minimizing the risk of overwhelming the user with too many options. However, in this case, the user must learn and possibly memorize many steps to access their intended application, which can be a serious problem for some people with CLDs (Friedman & Bryen, 2007; Lewis, 2007; Keates et al., 2007). By using a broad interface, the user does not need to navigate through many different layers to locate the desired application. However, this may also cause problems if not designed carefully, because with fewer layers, there is a greater chance of creating a cluttered appearance that has been shown to cause distractions (Keates et al., 2007). When designing an interface, it is important to not present excessive options, and to include only the tools necessary for the application. The e-Learning system discussed in this paper employs such a broad interface (Lewis, 2007; Keates et al., 2007; Gregor & Dickinson, 2007). A recent study involving adolescents with high functioning autism was conducted to determine which type of interface supports better performance on a game to teach social skills. It was determined that with richer interfaces, participants lacked the initiative to explore all the available features and performed more poorly than they did with simpler interfaces (Grynszpan et al., 2008). It should be noted that while the goal of incorporating more features into an interface is to increase utility and functionality, with the target group, such an interface was shown to have the opposite effect. International Journal of Information Studies    Volume 2  Issue 1    January 2010

63

2.4 Web Accessibility Guidelines and Recommendations Because the Internet and World Wide Web are global technologies, web accessibility criteria should also be as global as possible (Abascal & Civit, 2002). The European WC3/WAI WCAG (Web Content Accessibility Guidelines) and the Rehabilitation Act (Section 504) in the United States were among the first standardized set of guidelines (Abascal & Civit, 2002). However, in many cases, these guidelines only address the technical aspects of accessibility (e.g. screen readers for the visually-impaired), and pay less attention to cognitive aspects (Catarci et al., 2008). In addition to these efforts, there has been a vast literature proposing, studying, and assessing web accessibility guidelines (Friedman & Bryen, 2007). In order to ascertain the most pertinent principles of web accessibility specifically for people with cognitive and learning disabilities, a recent study listed and ranked eighty-six separate design guidelines by the frequency of citation by existing web guidelines. A guideline is deemed significant if there is 15% or more agreement amongst the guidelines. Twenty-two recommendations that met these criteria were found (Friedman & Bryen, 2007). These top recommendations included incorporating graphics and icons, using clear, unambiguous text, simple screen layout, consistency, contrasting colors, and large, clear navigation buttons. The second set of recommendations, cited by between 5% and 15% of the literature, include features such as descriptive hyperlinks, minimizing scrolling, limiting the number of fonts, and supporting text browsers (Friedman & Bryen, 2007). The recommendations that were most appropriate for an e-Learning interface were incorporated into the prototype presented here. 3. Methods To further study the effects of interfaces on user perceptions, a survey was conducted to assess the importance of interface design and complexity on usability in computer systems for assistive technology, especially their effects on individuals with CLDs. A popular technology used by many individuals with disabilities is text-to-speech software. The Kurzweil Text-toSpeech Educational System (Cambium Learning) is one effective, widely-used tool (Parr, 2008). The Kurzweil system does not contain unnecessary features: it simply converts text to speech. The survey was not taken to analyze the features of a specific system (in this case, Kurzweil) in any way. The sole purpose was to correlate ease of use (simplicity) with propensity to use the system. Kurzweil was selected because of its popularity among the population group at the survey site. The survey was sent to Nipissing University students with a range of learning disabilities, such as dyslexia and other reading challenges. The questions concerned issues of “look and feel” and ease of use. The Research Ethics Board Committee at the Nipissing University Research Services Department determined that the survey did not require an ethics protocol, and allowed the survey to be distributed. The survey, specifically designed to not be time-consuming, was intended to provide a general overview of users’ opinions. The survey consisted of five questions: (1) How often do you use Kurzweil? A. Daily. B. Weekly. C. Monthly.

D. Less

(2) How easy is using Kurzweil between 1 and 10? (1 not easy at all, 10 very easy) (3) Do you like the look of Kurzweil between 1 and 10? (1 looks unappealing, 10 looks great) (4) When did you start using Kurzweil? A. Elementary school. B. High school. C. University/College (5) Are there any changes you would like to make to Kurzweil? Yes/No If “yes”, please explain. Users generally rated the program as easy to use, with 87.5% rating an 8 (out of 10) or higher. The data were normally distributed, and suggests that the entire user base roughly agrees with this assessment (see Figure 1). Concerning aesthetic appeal, most users also rated the Kurzweil system highly, with 62.6% giving it an 8 (out of 10) or greater. There were, however, a few dissenting opinions, as seen in Figure 2. Based upon these results, a statistically significant model (p = 0.037) suggests that the more an individual finds the program easy to use, the more likely he/she is to use it frequently. Because of the small sample size (N = 16), it is unlikely (p = 0.716) 64

International Journal of Information Studies    Volume 2  Issue 1    January 2010

that this model can be generalized to a larger population. However, this small survey provides empirical support to the claim that feature-rich, yet complex interfaces may discourage their intended users. Further surveys are required to adequately examine this phenomenon. Question 4 asked when students were introduced to assistive technologies. The answers indicated a surprising result that all users did not use text-to-speech systems until they entered University or College, indicating that there is a possibility that students with learning disabilities are not receiving all the necessary support for them to succeed. However, as indicated above, further study is needed to substantiate this observation.

Figure 1. Distribution of ratings of ease of use for Kurzweil Text-to-Speech software, µ = 8.44, σ = 0.727, N = 16

Figure 2. Ratings for aesthetics for Kurzweil Text-to-Speech software, µ = 7.38, σ = 2.187, N = 16

Two common answers were given to Question 5, regarding desired interface changes. The first was that the voices could be improved in order to sound more realistic, and secondly, that file format conversion should be simplified, or eliminated altogether. These changes would increase the usability of the system, as voices that sound more natural are easier to understand and are less intimidating than synthesized voices. The format conversion step, if made more transparent, would reduce the cognitive load on the users, and reduce the time required to learn “housekeeping” tasks. By taking into account the results of the above survey, and by incorporating the results of previous HCI research in designing computer systems for individuals with CLDs (Friedman & Bryen, 2007; Abascal & Nicolle, 2005; Lewis, 2007; Grynszpan et al., 2008; Gregor & Dickinson, 2007), an original prototype of an e-Learning system was developed and implemented. Traditionally, web accessibility has focused on HTML (hypertext markup language) static content. Screen readers analyze HTML tag structures, facilitating accessible environments. However, more recent interfaces incorporating multimedia content are frequently developed using DHTML (dynamic HTML) and Flash (Asakawa, Itoh, Takagi, & Miyashita, 2007). The prototype presented here was initially developed using Axure software (Axure Software Solutions, Inc.), which produces HTML output. An enhanced version of the interface was developed in Adobe Flash (Adobe Systems Inc., San Jose, CA), which is highly flexible, and offers many options required to create an interface for an e-Learning system (Rhodes, 2007). Flash is also a good development environment for interfaces because of its scripting language, Action Script 3, which supports a wide range of functionalities. For instance, different types of interactions, such as mouse events, are easily handled. Flash also incorporates the object oriented paradigm, which helps in bridging the semantic gap between interface specifications and implementation. The e-Learning system was created by adding objects to selected layers and frames. This feature facilitates modifications because each object is independent. They can be changed without affecting other objects. Flash displays the layers in frames, similar to how a motion picture works. Although Flash and DHTML-based content poses some challenges for screen readers (Asakawa et al., 2007), this is not seen as a limitation for the current e-Learning system, which does not employ screen readers. Using these tools, an e-Learning interface specifically tailored to individuals with CLDs was developed. The intent is to create an interface, using guidelines suggested and established in the current assistive technology literature, which does not require extensive learning time, but which can be used with a minimum cognitive strain. Of particular importance were the following guidelines (Harper et al., 2008). Each link is represented as both an icon and with a clearly written caption (Recommendations 1 and 2, as described in Friedman & Bryen, 2007). All pages use the same background and place the title of each page in the same location (Rec. 3); Headings, International Journal of Information Studies    Volume 2  Issue 1    January 2010

65

titles, and appropriate prompts are incorporated (Rec. 4); Large sized sans serif font is used throughout the system (Recs. 6 and 11); Clutter is avoided (Rec. 7); Navigation buttons are clearly labeled, and display as highlighted to indicate that it is being selected (Rec. 12); A color scheme is selected in which text is clearly viewed and icons are clearly distinguished (Rec. 15); Each page outside the home page has a “Back to Home Page Button” in case the user accidentally navigates away from the home page (Rec. 21). The target user group for this system is, but not limited to, students in junior high to high school (ages 12 to 18). The ­e-Learning system is intended for students who require extra assistance in accessing and organizing information. This group includes people with learning disabilities, high functioning cognitive impairments, and Asperger’s Syndrome. Because this is a webbased system, it was developed following guidelines for web accessibility found in the literature, and also from the third author’s experiences in the field of special education (Friedman & Bryen, 2007; Keates et al., 2007). 4. System Description As stated above, the user interface is one of the most important features of a computerized system. In addition to being functional and error-free, the interface must accommodate the user, and must not require excessive time and effort to learn (Quinn & Wild, 1998). To accomplish the goal of usability, the interface was designed to be conceptually simple, yet powerful and effective. The quality of “simplicity” varies from person to person. In the present context, interface simplicity is not a structural property, but rather refers to how the demands of an interface complement the capabilities of a user (Lewis, 2007). Simplicity is especially important for individuals with CLDs (Catarci et al., 2008; Keates et al., 2007). The features and appearance of the interface are designed to minimize memory strain, clutter, distraction, and cognitive strain (Recommendation 7, with 30% agreement amongst guidelines) (Friedman & Bryen, 2007). In addition, the interface incorporates the most frequently-cited design recommendation (75%), concerning the use of graphics, icons, and symbols in addition to text. All text, except text within icons, was displayed with a large-sized sans serif font (Recs. 6 and 11 – 30% and 20%, respectively). The background employed only bright colors to highlight the edges of the page. Dark color schemes were considered to be distracting, and may cause difficulties in clearly viewing text and distinguishing icons (Rec. 15 – 15%). The rest of each page was left as white, facilitating easy recognition of text and icons. Headings, titles, and appropriate prompts were incorporated into the design (Rec. 4 – 50%). The login page, shown in Figure 3, provides very clear sign-in instructions. Instead of prompting for user ID, the interface prompts for the name of the user, because asking for the user’s name does not require additional memorization. Additional security measures could be added, if necessary. The page layout is designed to minimize the effort required to log in by placing the text boxes on a darker background, thereby allowing the user to easily distinguish between the background and the locations in which text is entered. Additionally, the information required from the user is placed in order from top to bottom, and is clearly labeled. To maintain a clutter-free login page, there are no additional options on the screen that could potentially lead to confusion. The navigation buttons are clearly labeled, and “highlight” to indicate that it is being selected (Rec. 12 – 20%). In order to avoid distractions, the page contains no other options. The homepage is shown in Figure 4. The homepage, which is intended to be the interface focal point, features essential tools to assist in taking online courses, links to course assignments, and email access. These items were placed on the homepage to reduce the amount of navigation required. The tools included in the e-Learning system are: a standard commercial word processor, spreadsheet, a presentation program, and an Internet search option. These tools were selected because they are currently used very frequently in education for people with CLDs, and one or all of these tools are usually routinely employed in their classes, as observed from the third author’s experience in working with affected individuals. Each of these tools is identified with an intuitive icon, and a clearly labelled button placed underneath the icon, which reacts to the user by highlighting when the cursor hovers over it (Recs 1 and 2 – 75%, 70%). As with all current Windows-style interfaces, this feature clearly identifies that the cursor is over the button, and when the button is selected it appears to be pushed. This visual feedback is particularly important for users with disabilities because it can be difficult for these individuals to understand why an event occurs without having a concrete action with which to identify it. The assignments link, also identified with an icon and a button, provides efficient and easy access to the user’s assignments. It is clearly labeled for the subject to which it belongs, thereby reducing any confusion in deciding which assignments page to choose. An email inbox was added to the homepage so that the user could immediately see whether new emails have arrived, and if so, who they are from. Such a feature also eliminates the need to navigate away from the page, which is the current focus of interest. This homepage was designed for the user to complete the intended task, rather than having to decide how to perform that task.

66

International Journal of Information Studies    Volume 2  Issue 1    January 2010

Figure 3. Login page

Figure 4. Homepage

Although access to email is another feature taken for granted by most people, it is especially vital for e-Learning. Email is the fastest and most convenient method of communication between students and their instructors. In Figure 5, the e-Learning email page is displayed. It follows the approach used throughout the development of the entire system, which is to minimize complexity. Only the necessities of writing and sending an email are included, resulting in a more useable system (Gregor & Dickinson, 2007). Many people with CLDs experience difficulties with memory and with information organization (Friedman & Bryen, 2007). Therefore, the assignment pages have a multi-step checklist that indicates the completion and submission status of assignments. This feature is shown in Figure 6. Each assignment has an associated check list that allows the user check the appropriate box themselves. To maintain a consistent “look and feel”, and to disallow changes that may cause unnecessary stress, all pages use similar backgrounds and place the title of each page in the same location (Rec. 3 – 60%). This helps to prevent any unnecessary stress on the user (Friedman & Bryen, 2007; Keates et al., 2007). Each page outside of the homepage has a “Back to Home Page” button in case the user accidently navigates away from the homepage, eliminating the burden of attempting to return (Rec. 21 – 15%). This button is located in the same location on each page (Friedman & Bryen, 2007; Gregor & Dickinson, 2007). Throughout the entire system, the common theme is to reduce the amount of clutter on each page, and to facilitate a more intuitive interface. 5. Results It is very difficult to assess the usability and effectiveness of interfaces in general, and it is especially challenging to quantify their effectiveness, intuitiveness, and other impacts on people with CLDs (Friedman & Bryen, 2007; Grynszpan et al., 2008). There is very little published outcomes-oriented research on specific guidelines, as well as on their overall effects (Friedman & Bryen, 2007). Furthermore, due to challenges in experimental protocols, it is very difficult to obtain a sample size of users with sufficient statistical power to make inferences. However, to obtain a preliminary empirical indicator of effectiveness, a student, aged fifteen, currently enrolled in an Ontario (Canada) public school, tested the system. The third author tutors the student in applied mathematics and applied science three times a week for two hours each session. This student possesses a learning disability which affects the speed at which she processes information, and was therefore part of the group to which the system is targeted. In addition to below average processing speed, her organizational skills are poor, and her attention span is very short. The student was asked to perform tasks with the e-Learning system, and was not given any instructions on how to accomplish them. No additional information about the system or its purposes was provided. The student was asked to log into the system and to locate her English assignments, and to subsequently return to the main page and to navigate to send an e-mail. The student accomplished all these tasks without any instruction. She did not need to search through a complex menu, and was therefore able to easily and quickly locate all the appropriate links. She also reported satisfaction with the appearance and aesthetics of the interface. Although detailed, in-depth comparative studies must be designed and performed to assess the

International Journal of Information Studies    Volume 2  Issue 1    January 2010

67

quality of this e-Learning approach, this simple test provides some empirical evidence that by reducing the complexity of an interface, an individual with a CLD can generally take the initiative, through exploration, to learn to use it, and to navigate through it without undue cognitive strain, as has been observed in previous studies (Grynszpan et al., 2008). 6. Discussion The third author of this study has had much experience working with individuals with disabilities in various settings. He has complemented his tutoring with computer technology, and has first-hand experience with a wide range of assistive technologies, as well as with different methods of instructing people with various CLDs. The inspiration for the system presented in this paper arose from his experience in working with a teenager with a cognitive disability who used an online course system. The site was considered to be accessible, as it met many of the web guidelines. However, excessive time was expended in navigation, attempting to locate options to complete the assigned task. Once the task was found, more time was spent in determining which assignment was completed. Assignment submission via email was extremely challenging, because files frequently had to be converted into another format, and subsequently attached. When left unsupervised, this teenager became distracted and began to use extraneous features (such as chat). Even with supervision, very little work was accomplished, and some courses were not finished. This empirical evidence underscores the importance of determining essential features in an interface, and of reducing complexity, as documented in the literature (Friedman & Bryen, 2007; Lewis, 2007; Keates et al., 2007; Grynszpan et al., 2008; Gregor et al., 2007). By modifying the presentation to accommodate full functionality, and by facilitating attention-focusing (e.g., by removing or limiting access to potential distractions, such as online chats), a more useful interface was created for people with CLDs. The development of the e-Learning interface presented here, while not specifically achieving “universal” access, nevertheless incorporates elements of universal design in that unnecessary barriers to accessibility were removed (Abascal, 2002). The advantages to reducing interface complexity are well documented in the HCI and assistive technology literature (Friedman & Bryen, 2007; Lewis, 2007; Keates et al., 2007; Grynszpan et al., 2008; Gregor et al., 2007). By reducing the number of features available at any given time, and by rearranging the presentation to accommodate full functionality, a more useful interface was created. That rich interfaces often result in confusion and frustration by people with CLDs, primarily because of the efforts required to focus attention, has been supported by other investigators (Grynszpan et al., 2008; Quinn & Wild, 1998).

Figure 5. E-mail page

Figure 6. Assignments page

By contrast, Figure 7 above shows an example of how, without considering accessibility guidelines, a broad e-Learning system interface may appear. This interface, while powerful and providing full functionality, has a cluttered appearance. All features have links to them on the home page for easy use (and indeed, this breadth may be beneficial to many users). However, the features are spread out over the home page, implicitly forcing users to carefully search through all the options on the page to locate the feature they require. Additionally, to not further increase clutter, the links use small text with no image representation. However, the links are not easily distinguished. An individual can choose from 18 features in this example. The labels of the available features are not clearly delineated; by necessity, they only use one word to describe their function. A user with a CLD may find it difficult to understand what the feature does, and to find desired features. Providing these users with more features can have a negative effect on their performance (Grynszpan et al., 2007). The system shown here also eliminated many unnecessary features, such as chat, which may cause distractions to many individuals, but which are still available by navigating away from the homepage.

68

International Journal of Information Studies    Volume 2  Issue 1    January 2010

Figure 7. Broad e-Learning interface 7. Conclusions and Future Work Computers have greatly enhanced the lives of many individuals with disabilities. Unfortunately, with the lower priority ­currently placed on the development of accessibile Web-based systems for individuals with CLDs, accessibility inequality may potentially increase. Consequently, more importance needs to be placed on the development and implementation of guidelines for creating Web systems for these individuals. Furthermore, although universal design is potentially promising for addressing accessibility issues in general, more research is required to integrate universal design into HCI for people with CLDs. By taking an approach to interface design that meets the needs of individuals with CLDs, and by incorporating results of previous accessibility studies, a prototype web-based e-learning system was developed to provide accessibility for a wide range of users, especially those with CLDs. The approach described in this paper includes essential functionality, while not burdening these users with features that could cause unnecessary stress and frustration, thereby negatively affecting system usability and learning. In future work, improvements will include implementing a method for direct submission of assignments that does not require document attachment, or changing any formats to submit them. This feature would eliminate unnecessary steps, and facilitate a much more continuous flow. As discussed earlier, the survey results presented here suggest that eliminating file conversion steps would prove very beneficial. Additional potential enhancements include developing integrated word processor, spreadsheet, and presentation software, which may further reduce complexity, increase overall usability, and reduce the need for additional software. This paper demonstrates how a fully functional user interface can be created for people with cognitive and learning disabilities by eliminating barriers that are often not considered during systems development. Making e-Learning more accessible to people with cognitive and learning disabilities provides them with an additional learning outlet. The expected outcome of these increased educational opportunities enhances the potential for affected individuals to make their disability less of an obstacle. Acknowledgments The authors thank the staff at the Department of Disability Services at Nipissing University for assisting in distributing the survey, and students who responded to the survey. Thanks are also given to David Hunt for providing assistance with SPSS statistical software, and to Shallen Giroux for assistance with proofreading. References [1] Maxwell, K. (2000). Human-computer interface design issues. The Biomedical Engineering Handbook, Second ed., Ed. J. Bronzino, 153-2 – 153-8. [2] Harper, R., Rodden, T., Rogers, Y., Sellen, A. (Eds.) (2008). Being human. Human-computer interaction in the year 2020. Microsoft Research Ltd. [3] Abascal, J., Civit, A. (2002). Opportunities and risks of the information and communication technologies for users with special needs. Proc. of IEEE International Conference on Systems, Man and Cybernetics, 264-269. International Journal of Information Studies    Volume 2  Issue 1    January 2010

69

[4] Catarci, T., De Giovanni, L., Gabrielli, S., Kimani, S., Mirabella, V. (2008). Scaffolding the design of accessible eLearning content: a user-centered approach and cognitive perspective. Cognitive Processing, 9 (3) 209–216. [5] Friedman, M., Bryen, D. (2007). Web accessibility design recommendations for people with cognitive disabilities. Technology and Disability, 19 (4) 205–212. [6] Fryia, G., Wachowiak-Smolikova, R., Wachowiak, M. (2009). Web accessibility in the development of an e-Learning system for individuals with cognitive and learning disabilities. Proc. of IEEE Networked Digital Technologies Confer­ ence (NDT 2009) 166-171. [7] Abascal, J. (2002). Human-computer interaction in assistive technology. From patchwork to universal design. Proc. of IEEE International Conference on Systems, Man and Cybernetics, 1-6. [8] Bergman E., Johnson, E. (2009). Towards accessible human-computer interaction. Sun Corp. Web Document. http:// www.sun.com/accessibility/docs/access_hci.jsp#3.1. [9] McMillan, W. W. (1992). Computing for users with special needs and models of computer-human Interaction. Proc. of Conference on Human Factors in Computing Systems, 143-148. [10] Vanderheiden, G. C. (1992) Making software more accessible for people with disabilities: Release 1.2. Trace Research and Development Center, Madison, Wisconsin. [11] Abascal, J., Nicolle, C. (2005). Moving towards inclusive design guidelines for socially and ethically aware HCI. Interacting with Computers, 17 (5) 484-505. [12] Abascal, J., Azevedo, L. (2007). Fundamentals of inclusive HCI design. LNCS 4554, 3–9. [13] Lewis, C. (2007). Simplicity in cognitive assistive technology: a framework and agenda for research. Universal Access Information Society, 5 (4) 351–361. [14] Savidis, A., Stephanidis, C. (2004). Unified user interface development: the software engineering of universally accessible interactions. Universal Access Information Society, 3 (3-4) 165-193. [15] Keates, S., Adams, R., Bodine, C., et al. (2007). Cognitive and learning difficulties and how they affect access to IT systems. Universal Access Information Society, 5 (4) 329–339. [16] Grynszpan, O., Martin, J., Nadel, J. (2008). Multimedia interfaces for users with high functioning autism: an empiri­cal investigation. International Journal in Human-Computer Studies, 66 (8) 628–639. [17] Gregor, P., Dickinson, A. (2007). Cognitive difficulties and access to information systems: an interaction design perspective. Universal Access Information Society, 5 (4) 393–400. [18] Parr, M. (2008). More than words: Text-to-speech technology as a matter of self-efficacy, self-advocacy, and choice. Dissertation, McGill University. [19] Asakawa, C., Itoh, T., Takagi, H., Miyashita, H. (2007). Accessibility evaluation for multimedia content. LNCS 4556, 11-19. [20] Rhodes, G. (2007). Flash professional 8 game development. Boston, Massachusetts: Charles River Media. [21] Quinn, C., Wild, M. (1998). Supporting cognitive design: lessons from human-computer interaction and computermediated learning. Education and Information Technologies, 3 (3-4) 175-185.

70

International Journal of Information Studies    Volume 2  Issue 1    January 2010

Suggest Documents