Out of Sight, Out of Mind
Aphantasia as a Missing Design Variable. Or, why I can't use Canvas
Current accessibility and User Experience (UX) guidelines focus on a few standard categories of impairment: vision (blindness, low vision, color vision differences); hearing (deafness and hard-of-hearing); motor (fine-motor limitations, keyboard-only use, switch devices); and cognition (attention, memory, reading, language, executive function, autism, ADHD, dyslexia, broad “learning difficulties”).
Mental imagery capacity is almost never considered as a design variable. There is no common practice of asking how much an interface demands internal visualization or spatial recall once the user leaves a page.
What Is Aphantasia?
Aphantasia is an inability to voluntarily call up mental images. I’ve written about my own aphantasia and differences in mental modeling. Research suggests that 2–5% of the population (7-15 million in the U.S.) have some degree of aphantasia. Many of us discover the difference only by happenstance in adulthood, when we come to understand that concepts such as a “mind’s eye” or phrases like “count sheep” describe a literal experience for most people.
Many people with aphantasia report weaker visual working memory, weaker autobiographical memory, and sometimes weaker object and spatial memory. The degree varies, but the pattern is consistent: perception can be fine while mental imagery and recall are weak. In the context of UX, a person with aphantasia may see a website clearly when looking at it, but when the image is gone it is impossible to “picture” it again. Those of us with aphantasia cannot call up a visual picture on demand.
A concrete example: I can read a chart and understand it completely while looking at it. The moment I close the tab, the chart is gone. I cannot summon a mental image of it. I retain the conclusions I drew, but the image itself does not persist in my mind the way it might for someone with typical cognition.
In UX terms, this means that any workflow that depends on internal pictures, spatial maps, or visual memory of the interface creates extra cognitive load, even if the UI is fully visible and technically accessible when in front of the user.
Canvas as a Case Study
Users with typical mental imagery capacity don’t have a problem with platforms that require visual recall. For the rest of us, the user experience is full of friction. Canvas, a learning management system used by many universities and K-12 schools, is the absolute worst. I’ll use it as a case study because it is widely used, highly visual, and marketed as accessible. It passes all the usual accessibility checks and follows common UX patterns, yet it is almost impossible for users with aphantasia.
Like many platforms, Canvas uses a global navigation bar on the left side, then a course-level menu inside each course, then breadcrumbs (the clickable path at the top of a page showing where you are in the site hierarchy) and page content. Users are expected to learn that certain functions “live” in specific places in this layout.
To navigate, it helps to have a mental map: “Grades is the fourth item down in the left course menu” or “To get to quizzes I move from Home to Modules to this link.” Oh, did I remember to hit publish? How do I tell?
Many of us with aphantasia cannot build mental maps. The locations do not attach to an internal picture, so the user is repeatedly reading labels and scanning the page instead of navigating from a remembered layout. The screen is legible, yet the interface never turns into an easy internal map.
Icons as Landmarks
Canvas’s design philosophy assumes that users will quickly internalize icons as landmarks. The global navigation bar can be collapsed to show only icons; course cards use icons for key actions; and three-dot menus (a common icon indicating “more options”) or drag handles appear in many places.
The design assumes that users will quickly internalize these icons as landmarks: “megaphone equals announcements,” “three dots equal more options,” “grid handle equals drag here.” Take a look below — maybe it makes sense to you; it does not to m.
Without the capacity for visual imagery, landmarks do not become mental shortcuts in the same way. I track them with verbal rules instead of pictures, so every instruction that says “click the megaphone” or “open the three-dot menu next to the assignment” leads to more scanning and uncertainty. Collapsed, icon-only navigation is especially costly because it removes the text labels that my brain relies on.
The problem of imagery load extends to the concept of the dashboard. The dashboard and calendar in Canvas are presented as visual overviews where the user is supposed to glance at them and gain a sense of “what is happening” across courses and time. This works while those views are open. The friction appears once I move away. I cannot walk around with a stable internal picture of next week’s calendar or the dashboard layout. If I want to know how busy a week looks, I have to open the calendar again and read it in the moment.
In short, Canvas assumes visual persistence in both UX and pedagogy: “once you have seen your calendar, you know your week.” I never really “know” it in that visual sense; I only know what I can reconstruct from text and repeated checking. Guides rely on screenshots, video walkthroughs, and references to specific visual elements like the purple bar on the left. While I am staring at the screenshot or the tutorial, I can follow along. Once I close it, the instructions that hinge on a remembered image lose a lot of power. I need explicit, verbal paths like “In your course, look at the left menu and click Grades.”
Accessibility practice often treats screenshots and video demos as universally helpful. For people with weak mental imagery, they are useful only in the exact moment, then they are gone. This gap is almost never mentioned.
Don’t even ask me about trying to comprehend what I can see versus what students should see. This is an epic problem even for people with normal visual capacity.
Tools like Modules, Gradebook, and SpeedGrader are visually dense, involving many columns, inline icons, hover states (visual changes that appear when a cursor hovers over an element), and drag-and-drop interactions. Canvas does provide keyboard-based alternatives in some places, which addresses motor and input concerns. The cognitive load remains. When I reorganize modules or columns, I need to anticipate where things will appear, remember what I just changed, and keep a mental structure of the course in mind. With weak imagery, that structure never stabilizes nicely. I end up re-reading labels and re-checking views to confirm where I am.
Canvas claims to make it easy to embed images, videos, and diagrams in pages and modules. Many professors need to implement tasks where students are supposed to read or view something, then respond after navigating away. This pattern assumes that the learner can carry a visual trace of what they just saw. For someone with aphantasia, understanding may be strong while looking at the material, then drop sharply once it is out of sight. When the task requires recalling what a diagram looked like without having it visible, the student with aphantasia is working at a disadvantage even though their comprehension while viewing was fine. Nothing in the platform or the usual guidance warns about this hidden demand on mental imagery.
Accessibility Standards Ignore Mental Imagery Gaps
Accessibility standards focus on whether users can perceive, operate, and understand content given certain impairments or differences. They check color contrast, keyboard access, semantic structure (the underlying organization of a webpage that assistive technologies use), focus order (the sequence in which elements receive keyboard focus), text alternatives, and similar features.
Mental imagery capacity does not fit cleanly into the existing categories. It is estimated between 7-10 million people in the U.S. alone have aphantasia, and can see, hear, and use input devices. Most of us can read fluently and do not have a formal learning disability diagnosis. I suspect there are millions more out there who do not know they have challenges in visual imagery recall. They don’t realize the source of the friction they experience.
In short, no design checklist asks:
How much does this workflow assume that the user can hold a picture of the screen in their head?
How much do these instructions rely on remembered visual layouts?
How often does this system require recall of images or spatial arrangements after the user navigates away?
A Call for Recognizing Aphantasia
Treating mental imagery capacity as a design category would lead to practical shifts. For Canvas and tools like it, this would include:
More reliance on persistent text labels instead of icon-only modes, especially for critical navigation.
Documentation that always includes a verbal path alongside screenshots, using the exact menu names visible in the interface.
Options to favor list views and text-based summaries over purely visual dashboards when presenting schedules and workloads.
Pedagogical guidance inside the LMS that calls out imagery-heavy tasks and suggests alternatives, such as keeping diagrams on screen during questions or providing textual descriptions that can be referenced.
UX research that explicitly recruits users with aphantasia and asks them about imagery load, rather than assuming the only relevant cognitive variables are reading speed or attention.
Recognizing aphantasia means rethinking what counts as a real user variable. As long as accessibility standards ignore mental imagery capacity, millions of users will struggle with interfaces that are assumed to be fully accessible. This is an invitation, a call, to treat “imagery load” as a first-class design concern. Once that category exists, the adjustments become obvious.
Until I realized I had aphantasia this past year I never understood why I hated and felt overwhelmed by visual instructions or video tutorials. My brain literally shuts down and frustration skyrockets. But give me a good old list, maybe with some images I can reference while I’m doing the work, and I’m good to go. I feel a little less broken these days and I really like the idea of using mental imagery as a design principle. It also begs the question of what other kinds of differences we’re not even aware of that affect the way humans move and think about the social/physical/intellectual worlds.
This really landed for me. I kept nodding along, thinking, yes, this, exactly. I also have aphantasia, and you named so many things I’ve struggled with for years without quite having the language for them. That feeling of understanding something perfectly while it’s right in front of you, and then it just… disappearing the moment you navigate away—oof. So familiar.
The Canvas examples especially hit home. I’ve always felt that constant re-scanning and re-reading, like I’m working twice as hard just to stay oriented, and I’ve never seen it framed so clearly as an imagery load issue rather than a “me” problem.
I write about aphantasia too, mostly through meditation and lived experience, and this connected deeply with the work I’ve been doing. My book Living Is Easy With Eyes Closed actually grew out of realizing how many systems quietly assume a mind’s eye and how isolating it can feel when you don’t have one.
Thank you for putting words to this so honestly. This felt really validating to read.