WCAG Doesn't Cover the Way an ADHD User Experiences Your Interface
Run an accessibility audit on most enterprise software and you'll surface the usual violations: insufficient color contrast, missing alt text, form inputs without labels, focus indicators that vanish at exactly the moment a keyboard user needs them. These failures are well-documented and fixable. The 2026 WebAIM Million report catalogues them with depressing regularity. Contrast failures on 80% of pages. Missing alt text on most images. The industry knows what the problems are.
What the industry audits don't surface: whether an ADHD user can actually use the interface you built without losing the task they were trying to do. Whether the onboarding flow you shipped assumes working memory that your autistic users may not have available in the way you designed around. Whether the notification architecture — carefully designed to "keep users engaged" — is effectively making your product unusable for anyone whose attention system doesn't filter interruptions automatically.
WCAG 2.2 added three cognitive accessibility criteria. Three. For a category that affects an estimated 15-20% of the population. The gap between what the standard covers and what neurodivergent users actually need is not a minor oversight. It's a structural blind spot in how the field thinks about access.
What the Standards Don't Tell You
WCAG 2.2's cognitive additions are Minimum Focus Appearance (2.4.11), Focus Not Obscured (2.4.12), and Redundant Entry (3.3.7 — don't make users re-enter the same information). These are useful. They're also specific to a narrow slice of cognitive accessibility. They say nothing about:
- Working memory load across multi-step flows
- Cognitive cost of notifications and interruption design
- Predictability and consistency of interaction patterns
- Explicit vs. implicit affordances for users who process language literally
- Time-blindness and how time-based sessions interact with ADHD executive function
- Sensory sensitivities to motion, animation, color intensity
WCAG doesn't cover these because they're hard to specify as binary pass/fail criteria. "Does this interface respect working memory limits?" is not a question you can answer with a checklist. It requires understanding the specific cognitive profile of the user you're designing for and the specific demands of the task you've designed around. That kind of nuance doesn't fit into compliance frameworks.
The result is that organizations treat WCAG compliance as accessibility. Pass the audit, ship the product, mark accessibility done. The users who fall outside WCAG's frame — which includes most of the neurodivergent population — experience the consequences.
How ADHD Users Actually Navigate Your Interface
ADHD is not a deficit of attention. It's dysregulation of attention — a system that struggles to direct and sustain focus on demand, is highly responsive to novelty and stimulation, and has trouble filtering what's relevant from what's present. For ADHD users, the experience of a notification-heavy, multi-tab, feature-dense application is not "slightly distracting." It's cognitively expensive in a way that isn't true for neurotypical users.
Working memory is the variable most designers don't account for. Working memory holds the current task context — what you were doing, where you were in the flow, what you need to do next — while you navigate. Most interfaces are designed with the implicit assumption that users can hold 3-7 items in working memory without difficulty. For many ADHD users, that number is lower, and the system is more fragile. An unexpected modal dialog, an ambiguous back button, a confirmation step that requires re-reading the original selection — each of these is a working memory tax that can drop the whole context.
Stéphanie Walter's research on neurodiversity and UX identifies specific patterns that fail ADHD users:
Buried primary actions. If the most important thing a user can do on a screen isn't visually dominant, ADHD users spend cognitive resources finding it instead of doing it. They're not being inefficient — they're working without the automatic attentional filtering that makes neurotypical navigation feel effortless.
Progress ambiguity. Multi-step flows without clear progress indicators create anxiety and back-button behavior. ADHD users need to know where they are in a sequence at all times, because the cost of losing their place is higher. They don't recover from task interruption the same way.
Notification saturation. Every notification is an attention hijack. For users with typical executive function, the hijack is minor and recovery is fast. For ADHD users, the interruption pattern can make sustained task completion genuinely difficult — not frustrating, difficult. Notification systems designed to "drive re-engagement" are, for a meaningful portion of users, working directly against their ability to use the product.
What Autism Brings to the UX Problem
Autism is not a single experience. It exists across a spectrum with enormous variation. But there are UX implications that appear consistently in research and in the accounts of autistic designers and users.
Literal processing is one. Autistic users tend to process language more literally than neurotypical users. Metaphorical icon labels ("Home," "Magic," "Explore" as navigation labels for unclear functions), jargon-heavy error messages, and implicit expectations that users understand social norms embedded in the interface — these create friction that literal processing makes expensive.
Predictability is another. Autistic users often rely on consistent patterns as a navigation strategy. When interfaces change unexpectedly — a redesign that moves familiar elements, a feature update that alters established flows, an animation that appears in a place where nothing animated before — the processing cost is real. This doesn't mean interfaces can't change. It means that change needs to be communicated clearly, consistently, and with adequate transition support.
Sensory sensitivities vary widely but show up in interface contexts more than designers expect. High-contrast color combinations that feel assertive and energetic to neurotypical users can be genuinely uncomfortable for users with sensory sensitivities. Auto-playing animations at peripheral vision. Sound on product pages that defaults to on. These aren't minor UX irritants — for a subset of autistic users, they're reasons not to use the product.
Designing Neurodivergent-First Changes More Than You Think
The case for neurodivergent-first design isn't purely altruistic. It's the same argument that drove accessible-first web design in the mid-2010s: the constraints imposed by designing for the hardest case produce better solutions for everyone.
Plain language — writing to a 6th-8th grade reading level with clear, direct sentences — is a cognitive accessibility requirement for dyslexic users. It's also demonstrably better for all users. Error messages that explain what went wrong and what to do next, rather than cryptic codes, are a cognitive accessibility requirement for literally-processing users. They're better for everyone. Reduced motion options, visible focus indicators, consistent navigation patterns, explicit affordances, progress indicators in multi-step flows — every one of these makes the interface better for neurotypical users too.
The accessibility debt that accumulates in design files is partly a failure of tooling and process. But it's also a failure of scope — a definition of accessibility that's been too narrow for too long. WCAG compliance is a floor, not a ceiling. Neurodivergent-first design requires going further: asking not just "can this user access the interface?" but "can this user actually do the thing they came here to do, without the interface working against their cognitive architecture?"
That question has no checklist answer. It requires design judgment, user research with neurodivergent participants — not just neurotypical user panels with a neurodivergent user added as an afterthought — and a product culture willing to own the outcomes rather than the audit.
The 15-20% of your users with some form of neurodivergence are using your product right now. Some of them are succeeding despite the interface, not because of it. That distinction is worth sitting with.
Photo by Moe Magners via Pexels