From Cognitive Accessibility Task Force

Questions and Answers on the WCAG 2.1 success criteria for people with learning and cognitive disabilities

See our [success criteria proposals for WCAG]

How did you decide on these success criteria. Did you just make them up?

Gosh no. They are based on our [table of user needs ] and how to support them. (Note that this is a working document that includes Working Group notes) That was based on a gap analysis. To build the gap analysis we wrote the following:

  1. A Background research document. This looked at the research about user groups with learning disabilities and cognitive disabilities.
  1. We also wrote issue papers on topics such as security, safety and how they are affected by cognitive disabilities.
  1. Putting them together we found many authoring techniques.
  1. Finally, we built a [table of user needs ] and how to support them. From there we could see what success criteria were needed.

There are 3 Plain Language success criteria. Why?

Each Success criteria has a different scope.

! Success Criteria Suggested Level Scope
Plain language (Success Criteria 30) A Clear and simple language in instructions, labels, navigational elements and

error messages, which require a response to continue.

Plain language (Success Criteria 41) AA

Clear and simple language in headings, error messages and important


Plain language (Success Criteria 42) AAA

Clear and simple language in all content – with exceptions for writing


Note that clear language for all content is an important accessibility principle. However, if the user does not understand words and terms in these critical areas, the whole application or web site often becomes unusable.

A real-life example is a person, with mild dementia, trying to use an application to turn on a heating and air conditioning unit. The menu item for selecting heat or air conditioning is labeled "mode". The user does not know that "mode" refers to heat or to air conditioning, and thus cannot use the whole unit because of this one term. In this real-life example (reported by a task force member), a visitor turned on an air conditioner and did not turn it off when leaving the dwelling. The weather became a bit cooler. The user, who could not turn on the heat because of the language used, developed hypothermia, and needed emergency treatment.

I do not think I have people with learning disabilities using my site. Do I still have to do this?

It is very unlikely that people with learning disabilities do not use my site.

Many learning disabilities only affect one area of the brain. So for example people with dyslexia or autism are often great at programing. People with dyscalculia are often fantastic with words and literature. As learning disabilities affect at least 11% of the population, you probably do have people with learning disabilities in your target audience.

Over half of people over 60 years old will have some sort of will have an age related memory impairment. Often people with these impairments blame themselves. They may say "I am not good with technology so I can not use it" or "I am too old to manage". But actually it is the designer who is making choices that exclude them.

Famous people with learning disabilities include [CEO's] and [other famous people] including inventors and stars.

How many people are affected by the issues of the success criteria for cognitive and learning disabilities?

Over half of people over 60 years old will have some sort of age related memory impairment, including mild cognitive impairment also known as Age-Associate Memory Impairment (AAMI). Learning and memory for new events, and symbols slows so learning new interfaces and terms is less successful. On the other hand most of these users will be as proficient in other fields as they have ever been.

Note that the aging community is between 20 - 30% in most OECD countries and growing. This means that 10-15% of the population will have an age related memory impairment.

Other cognitive disabilities include people with:

  • Language related disabilities
  • Memory related disabilities
  • Focus and attention related disabilities
  • Disabilities that affect executive function and decision making

Together this can effect 11% of school age people

Also sometimes people with anxiety, depression or under stress have the same needs as people with cognitive and learning disabilities. This can effect 17% of the population

How important are the success criteria for cognitive and learning disabilities?

We have only included proposals for success criteria that are essential for some users to be able to access and make use of content.

The success criteria for cognitive and learning disabilities are a lot of work for the author. It that fair?

There may be a learning curve for authors, but that has always been true for accessibility. We will be working on quick tips, testing tools etc, to make it easier.

Also, we are working on semantics for personlisation and promoting other technologies such as the IBM content simplifier, that will make it much easier and faster.

Secondly we only have success criteria that are really important for people living with cognitive and learning disabilities. The amount of work is similar to that necessary to include groups, such as the blind.

Part of the problem is that we have got used to excluding people with cognitive and learning disabilities. Some may feel it is OK if people can not use our systems because they find them confusing. They may blame the user rather then their design. However, just because this is the way it is, does not mean it is the way it should be. The cost of excluding people, is an unfair society.

I have seen some other things that are useful. Why aren't they included?

There are a few reasons why some things may not have been included. They include:

  • It is not testable. All WCAG success criteria need to be testable, so authors can know if they conform or not
  • They need to be very important. All success criteria are necessary to include some people or keep them safe.
  • We

need strong proof that they work. If the research is inconclusive we can not include the content as a success criteria at this point.

If you still think we have forgotten something, please feel free to be in touch.

Why is personalization important?

We need personalization because different user needs can conflict. For example, having familiar terms and symbols is key to being able to use the web. However, what is familiar for one user may be new for another requiring them to learn new symbols. Personalization could include loading a set of symbols that is appropriate for the specific user, ensuring that all users find the icons simple and familiar.

We are making semantics and an architecture to make it as easy as adding a simple attribute to your content. This includes an open source project of example script that a web author can use as an "include ". It can read the user settings in the JSON files and adapt the page to the user needs (example script and example of adaptive page).

But personalization doesn't always work and testing is open ended.

We will include a range of recommended personalization settings and a test suit for testing all the recommended settings. The author will only need to test using a small test suit. If users stay inside the recommended settings the site should work well.

You mention user testing. Isn't that a problem for people with cognitive disabilities

We have an issue paper on user testing for people with cognitive disabilities. That should tell you how and what to do.