Accessible research presentations for neurodivergent audiences: what actually works in the room
You've spent months on your data. You've got the poster printed, the slides loaded, the findings polished. And then you watch a colleague in the third row cover their ears, close their eyes, and stop reading your results altogether. Not because your science is bad. Because your presentation is loud.
That moment changed how I build academic talks. Not through some dramatic awakening — I just noticed it happening, repeatedly, across conferences, lab meetings, and one painfully bright poster hall in a hotel basement. The fix wasn't heroic. It was mostly removing things: animations, flashing transitions, a top-left logo on every slide, four competing typefaces.
Making research presentations accessible to neurodivergent audiences isn't a grant requirement you satisfy and move past. It's a design decision that changes who can actually stay in the room with your work long enough to understand it.
Key Takeaways
- Neurodivergence covers a wide range of cognitive profiles — ADHD, autism, dyslexia, dyscalculia, sensory processing differences — and the accommodations overlap far more than presenters expect.
- Sensorially predictable slides beat beautiful slides. Animations, transitions, and background patterns cost you attention you need for your findings.
- Plain-language summaries, live captions, and a posted accessible PDF are three changes that take under an hour and change who can follow your method.
- Data visualization for research needs specific work: no 3D charts, no rainbow color scales, direct labels instead of legends when possible.
- Conference logistics (quiet rooms, session length, lighting) matter as much as slide design, and you can often request accommodations before you present.
- Build one accessible version of your deck, then let it serve as the base for every talk after.
Why research presentations are a harder accessibility problem than most talks
A keynote about marketing trends can afford to be vague. A research presentation cannot. You have tables, confidence intervals, p-values, sample sizes, methodological caveats. The density of information is the point.
That density is exactly what breaks access. Someone processing language more slowly, or filtering out background noise at a measurable cognitive cost, hits a wall at slide four. Not because they're less capable — because the format stacked three demands at once: read this, listen to this, and hold the last three slides in working memory.
What actually makes neurodivergent attention different
Neurodivergence isn't one thing. Autistic researchers may find fluorescent lighting physically unpleasant in a way that's hard to describe to someone who doesn't experience it. Someone with ADHD might follow your entire argument and then lose the thread the moment a slide transition sweeps across the screen. Dyslexia affects how numbers and symbols resolve on a busy background. Dyscalculia doesn't care how important your regression table is.
What these profiles share, practically: cognitive load adds up fast, and unpredictable sensory input drains capacity that would otherwise go to understanding your findings. Every animation is a small tax. Every competing visual element is a small tax. Ten slides in, the budget is gone.
I once sat through a talk where the presenter used a spinning 3D bar chart. I remember the chart. I could not, by the end, have told you what it measured.
Slide design choices that matter more than they look
Sensory predictability
Turn off transitions. All of them. Not because they're ugly but because motion at the edge of vision is disproportionately distracting for many neurodivergent viewers, and it buys you nothing.
A short list of what to strip from a research deck:
- Animation on every bullet (fade on one slide, nothing on the rest)
- Loops, GIFs, or video backgrounds behind data
- Decorative gradients that lower text contrast
- More than two typefaces
- Logos on every slide — put them once, on the title and the final slide
What to add instead: a consistent layout, high contrast, and a clear reading order. I use the same three-slide template for title, method overview, and results — same positions, same colors — for an entire talk. Audience members stop having to re-learn where to look.
The alt text and reading order issue nobody checks
Most presenters add alt text to images and stop there. That's not enough. A screen reader reads a slide in its underlying object order, which is often the order you happened to drag things onto the canvas — not the order you intended anyone to read them.
In PowerPoint, the selection pane lets you reorder objects so the title comes first, then the main point, then supporting detail. I check this on every data slide. It takes about ninety seconds per slide and I've caught genuinely embarrassing orders — a chart labeled before its axis, a footnote read as the headline.
For a genuinely practical guide to this, the official Section 508 resources and the W3C Web Accessibility Initiative are more useful than most accessibility blog posts, and they cover the document and PPTX specifics in more depth than I can here.
Plain language summaries and live captions
Here's the change with the biggest return for the least effort: a one-page plain-language summary of your talk, handed out or posted before you start.
Not a simplified version of your science. A plain-language version of your argument: what question you asked, how you tried to answer it, and what you found. Two hundred words. It lets someone with slower processing speed, or someone who missed your opening because the room was noisy, follow along with the spoken version instead of racing to catch up.
Live captions do similar work. Zoom, Teams, and most conference platforms now offer automatic captioning, and while the accuracy on technical vocabulary is mediocre, it's usually better than nothing — particularly if you supply a glossary of your key terms to the captioner or the ASR tool in advance.
Should you record and post the talk?
Yes, if you can. Asynchronous access is one of the most underrated accommodations for neurodivergent audiences. A recorded talk with a searchable transcript lets someone watch at their own pace, pause between slides, revisit the methods section three times without feeling watched.
If recording isn't on the table, at minimum post the accessible deck and the plain-language summary. A PDF is not the same thing as access, but it's a start.
Visualizing research data for diverse brains
This is where research presentations diverge from general accessibility advice. Your charts are not decoration. They carry the argument.
Rules I've settled on, from trial and documented error:
| Practice | Why it helps | Cost to you |
|---|---|---|
| Direct labels on lines and bars | Removes the legend back-and-forth for every viewer | A few minutes per chart |
| Sequential or diverging color scales, not rainbow | Rainbow scales are unreadable for many color vision profiles and noisier for everyone | One setting change |
| No 3D charts, ever | Perspective distortion misrepresents magnitude | Zero |
| One idea per chart | Reduces working memory demand | More slides |
| Confidence intervals shown as bands, not error bars alone | Bands read faster and survive glancing attention | Minor rework |
One habit I'd push harder on: say the number out loud when it appears. "Treatment effect of eleven percent, confidence interval from four to eighteen." Many neurodivergent audience members will remember the spoken figure better than the chart, and everyone in the room benefits from the redundancy.
Conference environments and real-time accommodations
You control more of the room than you think. Even in a session you don't organize, you can request: dimmed house lights in the presentation area, a consistent microphone level, no audience Q&A without a queue (so people who need processing time get it), and a stated break before questions start.
Ask the organizers, before the session, whether a quiet room is available near the presentation space. This is increasingly common at larger conferences and it costs you one email. If you're presenting a poster, request a spot away from the entrance and the coffee station — poster halls are acoustically brutal.
If you're organizing your own session, build in the things you can't retrofit: a written Q&A option alongside verbal questions, a fifteen-minute sensory pause halfway through a long block, and a posted deck link in the program so no one is taking photos of slides with their phone.
Should you use the APA poster format, and does it matter for accessibility?
The APA-style poster is a standard layout: title banner, three columns, methods-results-discussion flow. What matters isn't the format itself but whether you use its structure to reduce cognitive load. A three-column poster gives viewers a predictable path. A single dense block doesn't. If you're using the APA template, keep the column structure and increase your body text size to at least 24pt — the standard minimum assumes 20/20 vision and typical distance read, and it fails for a good portion of neurodivergent viewers.
Building your accessible deck once
I stopped treating accessibility as a per-talk project about two years ago. I built one base template with proper reading order, high contrast, direct-labeled chart styles, and a slide layout that maps cleanly to a plain-language summary. Every new talk now starts from that template, and I spend my time on the science instead of on remediation.
The tradeoff is real: my slides look less flashy than they used to. A colleague once told me they were "a bit plain." They probably are. But the last time I presented, someone came up afterward and told me they'd been able to follow the entire methods section for the first time at that conference, because the layout didn't move and the numbers were said out loud.
That's the whole point. Not the deck. The person still in the room at slide twelve.
