Designing for the Neurodivergent

When Features Become Barriers

When I open most productivity apps, I’m greeted by a barrage of notifications, color-coded priority systems, and gamified achievement badges. These features are meant to motivate, to organize, to help. For many users, they work exactly as intended. For others—those with ADHD, autism, dyslexia, or any number of neurodivergent conditions—these same features create chaos, overwhelm, and ultimately, abandonment of the tool altogether.

The term “neurodivergent” encompasses a wide range of neurological differences, from attention deficit hyperactivity disorder and autism spectrum disorder to dyslexia, dyspraxia, and Tourette syndrome. These aren’t deficits to be fixed but variations in how brains process information, manage attention, and interact with the world. Yet our digital interfaces are largely designed around a narrow conception of the “normal” user—someone who can filter distractions, parse dense text quickly, maintain sustained focus, and navigate complex hierarchies without cognitive overload.

This is not just a failure of empathy. It’s a failure of design thinking itself. When we design only for the statistical average, we create systems that work poorly for everyone. The curb cut effect—named for how sidewalk ramps designed for wheelchair users benefit parents with strollers, travelers with luggage, and delivery workers alike—applies just as powerfully in digital design. Features that serve neurodivergent users make interfaces clearer, more flexible, and more humane for all.

The Cognitive Load Framework

Before addressing specific design patterns, we need to establish what makes an interface difficult for neurodivergent users. The answer lies in cognitive load theory, developed by educational psychologist John Sweller in the 1980s. Cognitive load refers to the amount of working memory resources used during information processing.

Working memory is limited. For neurotypical users, this limitation manifests in subtle ways—difficulty multitasking, forgetting what you were searching for mid-search, losing track of complex instructions. For neurodivergent users, these limitations are often more severe and differently configured. Someone with ADHD might have particular difficulty filtering irrelevant stimuli. Someone with autism might excel at sustained focus but struggle with rapid context switching. Someone with dyslexia might process visual patterns quickly while finding dense text exhausting.

Sweller identified three types of cognitive load. Intrinsic load comes from the inherent complexity of the task itself. Extraneous load comes from poor design—unnecessary complexity that makes the task harder than it needs to be. Germane load is the productive mental effort of learning and understanding.

Good design minimizes extraneous load so users can focus their limited cognitive resources on intrinsic and germane load—on the actual work, not on figuring out how to do the work. This principle applies to all users, but for neurodivergent users, reducing extraneous load often makes the difference between usability and complete inaccessibility.

What Goes Wrong: Common Design Failures

Consider the typical onboarding flow for a project management application. You create an account, and immediately you’re presented with a modal dialog asking you to connect your calendar, another asking permission to send notifications, a third offering to import your contacts, and a tutorial overlay with seven sequential steps explaining features you haven’t needed yet. Each interaction seems small, but together they create what researchers call “interruption overload.”

For someone with ADHD, this sequence is particularly punishing. The executive function challenges that characterize ADHD make it difficult to maintain attention on a primary goal while processing interruptions. Each modal dialog represents a context switch, a decision to be made, another thing to remember. By the time the user reaches the actual interface, they’ve exhausted their available attention and decision-making capacity.

I’ve watched this play out in user testing sessions. A participant with ADHD downloads a task management app specifically designed to help with organization. Within two minutes, they’ve closed it in frustration. The problem wasn’t the participant’s lack of focus—it was the design’s demand for sustained attention across fragmented tasks.

Text-heavy interfaces present different but equally significant barriers. Many neurodivergent users, particularly those with dyslexia or certain autistic traits, find long paragraphs of unbroken text difficult to process. The visual density creates a barrier before comprehension even begins. Yet countless interfaces—from legal disclaimers to product descriptions to settings panels—present walls of text with minimal visual hierarchy, no breaking of content into digestible chunks, and dense, justified paragraphs that make it hard to track from line to line.

Another common failure point is inconsistent navigation. Users develop mental models of how interfaces work, expecting similar actions to produce similar results. When navigation patterns change unpredictably—when a back button sometimes returns to the previous screen and sometimes closes the entire workflow, or when swipe gestures do different things in different contexts—the extraneous cognitive load increases dramatically. Neurodivergent users often rely more heavily on these learned patterns and experience more disruption when patterns break.

Sensory overwhelm represents yet another category of design failure. Interfaces with autoplay videos, animated backgrounds, multiple moving elements, bright flashing components, and constant notification badges create visual noise that neurotypical users learn to filter out. For users with autism or sensory processing differences, this filtering is much harder or impossible. The result is not just distraction but actual physical discomfort.

Principles for Better Design

The failures above share a common thread: they prioritize aesthetic appeal or feature density over cognitive clarity. Designing well for neurodivergent users requires inverting these priorities. Several principles emerge from research and practice.

Reduce default complexity. Interfaces should open to their simplest, most focused state. Progressive disclosure—revealing advanced features only when needed—serves all users but particularly helps those who struggle with visual overwhelm or decision paralysis. The email client Hey demonstrates this principle well. The inbox shows only new mail, with all other organizational features available through clearly labeled, separate views. You don’t have to process folders, labels, and archive options every time you check email.

Provide clear visual hierarchy. Typography, spacing, and color should create obvious relationships between elements. Headings should look like headings. Buttons should look like buttons. Related items should be grouped visually. This isn’t about aesthetics—it’s about reducing the cognitive work of figuring out what’s important and how elements relate to each other.

Notion, for all its complexity, does this well in its basic text editing. Headings are clearly differentiated by size and weight. Lists are obviously lists. The visual structure mirrors the content structure, so you don’t have to decode the interface to understand the information architecture.

Make actions reversible. Fear of making mistakes increases cognitive load. When users can easily undo actions or know that experimentation won’t cause permanent changes, they can focus on learning rather than on anxious prevention. Figma’s version history and branching makes it safe to try things. Gmail’s undo send gives breathing room to reconsider. These features reduce the psychological weight of each action.

Design for customization. Different neurodivergent users need different accommodations. Someone with autism might want reduced motion and high contrast. Someone with ADHD might benefit from gamification features that provide external motivation. Someone with dyslexia might need increased line spacing and specific font choices. Rather than trying to find a single “accessible” design, provide meaningful customization options.

The reading app Matter offers font choices, spacing controls, color themes, and even text-to-speech options. These aren’t hidden in settings—they’re front and center, acknowledging that reading needs vary and that the interface should adapt to the human, not the other way around.

Minimize interruptions and notifications. Default to fewer interruptions, not more. Make notifications opt-in rather than opt-out. Batch related notifications rather than pinging users repeatedly. Respect focus modes and do-not-disturb settings absolutely. The attention economy has trained us to think that more engagement equals better design, but for many neurodivergent users, constant interruption makes the product unusable.

Provide multiple pathways to completion. There should rarely be only one way to accomplish a task. Some users navigate via search, others through visual browsing, others through keyboard shortcuts. Some prefer text instructions, others prefer video tutorials, others learn by experimentation. Good design supports multiple approaches rather than forcing everyone through the same flow.

The Ethics of Inclusive Design

These design principles aren’t just usability improvements—they carry moral weight. When we create digital interfaces that exclude neurodivergent users, we’re not just making bad design choices. We’re restricting access to education, employment, social connection, and civic participation.

Consider job application systems that require applicants to complete lengthy forms in a single session without the ability to save progress, or that use gamified assessments that advantage certain cognitive styles over others, or that require cover letters uploaded in specific formats that screen readers can’t parse. These design choices aren’t neutral. They filter out qualified candidates based on neurological traits that have nothing to do with job performance.

Or consider educational platforms that rely heavily on synchronous video with no transcripts, or that use timed assessments that privilege processing speed over understanding, or that require students to navigate complex interface hierarchies before they can access course content. These choices create barriers to learning that have nothing to do with a student’s capacity to master the material.

The counter-argument often presented is one of resource constraints. Building accessible, neurodivergent-friendly interfaces takes more time and money. Features must be prioritized. The “normal” user represents the largest market segment, so it makes business sense to optimize for that group.

This argument fails on both ethical and practical grounds. Ethically, it treats neurodivergent users as an edge case, an accommodation to be made if resources allow. This position is untenable in a world where digital interfaces mediate access to essential services. We wouldn’t accept “we don’t have the budget to make our building wheelchair accessible” as justification for architectural barriers. The same standard should apply to digital design.

Practically, the argument assumes that designing for neurodivergent users conflicts with designing for neurotypical users. The evidence suggests otherwise. Many features that serve neurodivergent users—clear visual hierarchy, reduced clutter, straightforward navigation, customization options, reversible actions—improve the experience for everyone. The curb cut effect applies.

Moreover, neurodivergent users are not a tiny minority. Estimates suggest that 15-20% of the population is neurodivergent in some way. That’s not an edge case. That’s a substantial user base being systematically excluded by design choices that often aren’t even intentional but simply reflect unexamined assumptions about “normal” cognition.

Implementation in Practice

Moving from principles to practice requires both technical changes and organizational shifts. On the technical side, several concrete practices make a difference.

Start with comprehensive user research that includes neurodivergent participants. Don’t treat neurodivergent users as an afterthought or a compliance checkbox. Include them from the beginning of the design process. Their insights about cognitive load, attention management, and information processing will improve the design for everyone.

Build in accessibility and customization from the start rather than retrofitting later. Options for text size, color contrast, motion reduction, and content density should be part of the initial design system, not features added after launch. This isn’t just more efficient—it ensures that accessibility is part of the core experience rather than a separate “accessible mode” that’s clearly an afterthought.

Test with assistive technologies and in varied contexts. Screen readers, voice control, keyboard-only navigation—these aren’t niche use cases. Understanding how your interface works with these tools reveals assumptions you didn’t know you were making about how users interact with your product.

Document cognitive load considerations in your design system. Visual hierarchy guidelines should explain not just how things should look but why—what cognitive work each design choice performs. This makes it easier for teams to maintain consistency and for new designers to understand the reasoning behind decisions.

On the organizational side, inclusive design requires leadership commitment and resource allocation. Accessibility can’t be one person’s job or an optional add-on. It needs to be integrated into the design process, included in success metrics, and prioritized alongside other product goals.

This means training designers and developers in accessible design practices. It means including neurodivergent users in user research as a matter of course. It means reviewing existing products through an accessibility lens and being willing to make changes even when they require significant effort. It means measuring accessibility as part of product quality, not as a separate concern.

Looking Forward

The conversation around neurodivergent-inclusive design is still relatively young. We’re learning what works, what doesn’t, and how to balance different needs when they conflict. But some directions seem clear.

Personalization will become more sophisticated, moving beyond simple settings toggles to AI-driven adaptation. Interfaces might learn from user behavior—noticing when someone consistently disables animations or prefers keyboard navigation—and adjust automatically. The challenge will be making these adaptations transparent and user-controllable rather than opaque and imposed.

Voice interfaces and conversational AI offer new possibilities for users who struggle with traditional GUI navigation. Rather than clicking through menus and forms, users can describe what they want in natural language. The interface handles the complexity of execution. This works particularly well for users with executive function challenges who struggle with multi-step processes.

At the same time, we’ll need to be careful that new interface paradigms don’t simply recreate old exclusions in new forms. Voice interfaces struggle with accented speech and speech differences. AI-driven systems can encode biases in their training data. Gesture-based interfaces assume certain motor capabilities. Innovation in interaction design must be accompanied by thoughtful consideration of who’s included and who’s left out.

The broader cultural shift matters too. As neurodiversity becomes better understood—not as a set of deficits to be remediated but as natural human variation—design practices will evolve to reflect that understanding. Just as good architecture now assumes that buildings should be accessible to wheelchair users from the start rather than retrofitted with ramps as an afterthought, good digital design will assume cognitive diversity as a baseline rather than an edge case.

Conclusion

I began with productivity apps and their tendency to overwhelm rather than assist. The problem isn’t that these apps have features—it’s that they present all features all the time with little regard for cognitive load or attention management. The solution isn’t to remove features but to design interfaces that adapt to different cognitive needs.

This requires humility about our assumptions. When we design, we inevitably encode our own cognitive styles into the interface. We prioritize what feels natural to us and treat as edge cases patterns that differ from our own. Inclusive design demands that we question these assumptions systematically.

It also requires recognizing that good design serves human flourishing. Interfaces aren’t neutral tools—they shape how we work, learn, communicate, and participate in society. When we design them to be accessible only to those who match a narrow cognitive profile, we’re not just making bad products. We’re constructing barriers to human potential.

The work of inclusive design is never finished. As our understanding of neurodivergence deepens, as technologies evolve, as use cases change, we’ll need to continually reassess and improve. But the direction is clear. We design for humans in all their cognitive variety, or we design systems that serve only some humans while excluding others. The choice, ultimately, is ethical as much as technical.

And that choice matters. Because on the other side of every interface is a person trying to work, trying to learn, trying to connect, trying to participate. Our design decisions determine whether that interface facilitates their goals or stands in their way. We should design accordingly.