The March-April 2002 interactions Whiteboard column caught my eye. Focusing on usability terminology, the article declares that a product’s intuitiveness depends on its users’ previous life experience and learning. "Aha!" I thought. "This sounds like mental models."
A user’s mental model reflects her understanding of how something functions and her expectations of behavior. It determines what is intuitive to that user.
As a developing country, South Africa asks us a key question: "How do we cope with major differences in what is intuitive?" This wonderfully diverse, rainbow nation is characterized by having many user groups with widely differing technology mental models. Users who have had less exposure to technologyand therefore have weaker technology mental modelsare often "previously disadvantaged" South Africans. In the country’s pre-democratic period, these people generally had very poor education and no access to educational technologies.
Building the New South Africa has had the side effect of creating sensitivity to any hint of discrimination. I sometimes get the impression that the desire to be "politically correct" overrides the actual needs and desires of the individuals concernedand so the differences in opportunity persist.
But how do we compensate for this lack of opportunity? We may be tempted to say, "Well, let’s customize their interfaces to make them more comfortable with the technology," thinking that if we match it to their mental models the technology will seem more familiar. However, this may not be in their best interests; in fact, it limits their options. We would do better to develop fast-track technology training that helps such people build "mainstream" mental models.
I hear well-known usability experts say that design needs to bridge the gap between the need for intrinsic skills and the capability of the technology. However, people have an amazing ability to learn and adapt where it is really warranted, and design should consider very carefully which differences to accommodate and which to eliminate through training (always, of course, using good design principles).
Should the Business Community Customize its Applications?
From my experience with introducing browser-based applications into work environments, I have found that previous experience with a Web browser and with Microsoft Windows significantly affects users’ ease of learning and comfort of use. The more experienced users, who have confidence in their technology mental models, succeed at exploring new features and teaching themselves the new technology. These differences in learning and comfort tend to align with cultural groups. This makes it important to exercise sensitivity in introducing new technologies.
In fact, less-skilled users do not seem to want the interface customized to fit their particular backgrounds. They want help and training to reach a higher level of understanding of the standard technology, so that they can move around more easily in the job market. Some of the less-skilled users do favor customization.
What about customizing language and icons in interfaces for different cultural groups? One can easily imagine "African versions" of Microsoft software with beautifully stylized themes, icons, and sounds.
In neighboring Botswana, cultural groups tend to favor the standard Windows interface, in English, rather than a localized version in their home language. This idea enjoys particularly strong support among business users, whose Windows technology skills are more entrenched. We cannot escape the fact that the Windows icons and interface style form an essential part of how these people learn to use technology. And the first technological environment that users experience is where they will continue to feel most comfortable.
Can Developing Countries Customize in a Broader Context?
In the broadest sense, the digital divide separates developing and developed countries. Even though most of the South African business community resembles a developed country in its use of technology, the nation as a whole is a developing country, with exposure to technology varying widely between the haves and the have-nots.
This variation is even more pronounced in school situations, where children who still live in mud huts or shanty towns and whose parents may not be able to read or write, go to schools in which more and more of them have the opportunity to use computer labs.
To avoid upsetting anyone, South Africa has adopted no fewer than 11 official languages, with English as its business language. Educators and politicians strongly contend that computer material, including technology training, should be translated into the 11 official languages at the school level. I question whether this is the best investment of limited resources. With more than 1,000 indigenous languages, Africa does not lend itself to this level of customization.
In Botswana, even when users support the idea of a localized interface, they tend to disagree on which language to use, even though the country has only one official local language (seTswana) that could provide an alternative business language. The use of seTswana is generally accepted only by those who speak it as their first language.
Such a lack of agreement poses a cultural issue of its own. African cultures place great importance on consensus, and achieving agreement on any topic can require hours of deliberation. Many good concepts never come to fruition, and a technology may never be accepted if it has insufficient consultation and agreement before implementation.
In South Africa, we find that customizing language and other superficial interface characteristics do not play the most critical role in making technology intuitive to all South African cultural groups. Deeper and more complex cultural differences, such as how people collect and structure information, carry far more importance and should command more attention, especially for Web sites.
What Approach Should We Take?
How should we help people from different cultural groups experience a certain technology as intuitive? Should we design for the lowest possible level of complexity? I say no, we should use more imagination. If mental models affect what is intuitive, surely training and structured exposure to technology can help users adapt their mental models. An increased knowledge of technology should give users more comfort with new interfaces, which become intuitive within the task context. Experienced drivers, for example, find it natural and automatic to brake quickly when they need to; learners do not. Similarly, driving a familiar route is easier than driving an unfamiliar route.
Several South African cultural groups have managed to quickly reach a minimum acceptable level of technology literacy in English. We see this happen through exposure to technology in the corporate world and in educational environments, especially universities. We also see a rapid convergence in understanding technology among the different cultural groups at the university level.
Does it Matter What Technology Users See First?
If the intuitiveness of a new interface depends so heavily on technology exposure, we should carefully consider the specific technology to which users have their first exposure. This issue underlies some of the concern about Microsoft’s proposed donation of software to South African schools and universities. I can’t help wondering if this donation is an ill-disguised attempt to breed a standardized (or globalized?) generation of new users! Online articles and columns in IT Web voice this concern by using emotionally charged titles such as "The Battle for the Hearts and Minds of Computer Users Heats Up!" and "Buying Mind Space."
A country as small as South Africa would benefit from a standardized work environment. Corporations depend on it. For this standard we may as well adopt Microsoft’s, since it has already become the desktop standard in most organizations. Don’t think of this as compromising cultural values or as allowing software developers to impose on us the values of the "developed world." See it instead as a change necessary for participating in the global market and an important requirement for achieving usability and productivity.
Do Developing Countries Have Different Usability Rules?
Let us ask instead, "Do developing countries apply usability principles differently?" To this question I see a clear answer: "Yes, in some cases they do." Significant challenges and limited resources make it difficult to bring large sections of the third-world population into the global economy. We must provide a fast track to build technology skills, with training rather than customization bridging some of this gap (always, of course, using good design principles).
What Should Designers Do About the Gap in Mental Models?
Take the following steps in designing for any country, including your own:
- Look at the situation from a broad perspective. Even if you are not directly involved in providing the training and the exposure to technology, do what you can to make it happen.
- Understand the general level of previous exposure to technology of a specific user group.
- Build on previous exposure to take advantage of the mental models that users already have.
- Identify the gap between existing mental models and the knowledge and understanding required to use the technology for a specific task.
- Decide how to bridge that gap. How much design intervention does the bridge require; how much training and exposure to technology?
These steps should go a long way toward helping people in developing countries to "catch up" in the electronic global race.
How Do We Balance Customization and Training?
Consider customizing for diversity, but remember this: When a group of users from different cultural backgrounds and technology exposure perform the same task, learning plays a far greater role than design in making their technology intuitive.
Elizabeth Buie, Senior Principal Engineer,
Computer Sciences Corporation
15245 Shady Grove Road, Rockville, MD, 20850,
301-921-3326, fax: 301-921-2069, email@example.com
©2003 ACM 1072-5220/03/0300 $5.00
Permission to make digital or hard copies of all or part 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. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2003 ACM, Inc.