Designed for Every Mind, Or Just the Average User? How Technology Is Rethinking Cognitive Accessibility

As someone who knows what it’s like to have a mind that doesn’t always cooperate with a screen full of distractions, the idea of an “average user” has always felt a little questionable.

Every app and website gets built with a specific user in mind, someone who reads at a steady pace, ignores a pop-up without losing their train of thought and remembers what was on the previous screen without going back to check. Nobody has ever met this person, yet a lot of software behaves as though they’re the only one who ever logs in.

Cognitive accessibility is about the gap between that imaginary user and everyone else, including people with ADHD, dyslexia and autism. A product can leave that gap wide open while still meeting some accessibility requirements, which is how you end up with something technically usable but actually quite exhausting.

The Small Design Choices That Eat Your Concentration

W3C, the World Wide Web Consortium that develops web standards, has published guidance on making content usable for people with cognitive and learning disabilities. It describes many of these difficulties as hidden, which goes some way to explaining why the barriers tend to be ordinary design decisions rather than dramatic failures.

Its recommendations read like a round-up of everyone’s silent grievances with the internet: avoid too much content, stop controls moving unexpectedly, use clear words, leave some white space and explain anything that is only implied.

Memory gets similar treatment, with passwords that rely on memory named as a barrier. WCAG – the Web Content Accessibility Guidelines – includes a requirement in its 2.2 version that a login must not depend on a cognitive function test, such as remembering a password or solving a puzzle, unless an alternative authentication method or assistance is available, subject to specific exceptions in the standard.

There’s even a stopwatch; W3C’s guidance warns that a session timing out while someone hunts for their postcode can be a significant barrier. The mental effort involved has a name, cognitive load, which can build with every unnecessary decision on a screen.

Letting The Technology Do The Adapting

Apple’s Assistive Access, a mode for iPhone and iPad that offers a simplified interface for people with cognitive disabilities, is a useful look at what adapting involves. Camera can be pared back to a single mode that takes a photo instead of offering photo, video and other settings. The Home Screen can use a visual grid or a list of items depending on what suits the person using it.

A trusted supporter can help set it up, avoiding the need for someone to navigate the initial setup alone. Apple also lets people who are sensitive to rapid animations turn off the automatic playback of moving images such as GIFs in Messages and Safari through a separate accessibility setting.

Microsoft’s Immersive Reader, a reading tool available in supported apps and environments including Word and Outlook, takes a different route by changing how text is presented. Text size, spacing, font and background colour can all be adjusted, while line focus highlights one, three or five lines at a time. Words can also be read aloud with highlighting as they go.

The choice stays with the reader, who might want the narrow strip of text or the audio or both. Saved progress and clear step-by-step instructions, both recommended in W3C’s guidance, work in the same spirit by taking some of the remembering off the person.

Can AI Make This Any Less of A Chore?

AI is already doing some of this adapting. Apple’s Writing Tools can summarise selected text, pull out key points or turn it into a list or table. They can be used with selected text in supported apps and websites, although Apple Intelligence isn’t available on every compatible device, in every language or in every region.

Each of those starts with the person choosing to press the button, which is a very different thing from a system deciding what someone can cope with after watching a few clicks. Whether anyone should have to explain their cognitive needs to an assistant every time is an open question; options that work without any explanation sidestep it neatly.

Then there’s accuracy. A study coordinated by the European Broadcasting Union and led by the BBC reviewed more than 3 000 answers from ChatGPT, Copilot, Gemini and Perplexity to news questions and found that 45% had at least one significant issue.

It wasn’t a test of summarising a document someone has pasted in, but it does show that an answer can look perfectly tidy and still be wrong.

When The Fix Turns Into Homework

Settings only help if someone can find them. Nobody should have to spend an evening working through menus to make an app bearable, because at that point the accessibility feature has become another job on the to-do list.

W3C’s guidance recommends involving people with cognitive and learning disabilities in research and usability testing. It also recognises that people can have more than one need at once, using age-related forgetfulness alongside a need for higher contrast as an example.

Personalising an interface around someone’s diagnosis or behaviour can involve collecting sensitive information. Stripping an interface back removes options along with the clutter, which suits some people and frustrates others who wanted them. Personalisation earns its place when it takes work off the person using it.

Nobody Logs In As The Average User

The user that software keeps designing for has never logged in to anything, because every real person turns up with a different attention span, reading speed and tolerance for clutter.

Cognitive accessibility was never about building a separate internet for people who think differently. The goal is the ordinary one with clearer wording, fewer obstacles and enough flexibility that nobody has to negotiate with it – getting something done online shouldn’t depend on thinking the way a product’s designers happened to think.