I sat down with Russell Van Brocklen, a dyslexia researcher based in New York whose work sits at an interesting crossroads between education, storytelling, and why so many Hollywood films fall apart if you think about them for more than five minutes. Russell is state funded and works primarily with dyslexic students, often highly capable ones who have been written off early. What he does is develop practical methods to help them get past the barriers that dyslexia puts in the way. What quickly became clear is that he also has very strong views on how stories are built, and why so many of them fail.
Russell wanted to explore these two points:
1. Worldbuilding for readability –how to keep complexity while still keeping the reader with you.
2. The “logic stress-test” — a practical way to catch plot holes and broken rules before the audience does.
Early on, Russell talked me through a case study involving dyslexic high school students who were bright but badly underserved. Using a very deliberate and demanding curriculum, he took students who were struggling with basic reading and writing and, within a year, had them producing work that held up against graduate level analytical writing standards. The results were impressive not just academically but economically as well.
That idea carried straight into our discussion of speculative fiction. Russell has little patience for scripts that are rushed into production and it shows. His description of modern blockbusters riddled with what he called “Grand Canyon chasms of plot holes” was delivered with equal parts irritation and amusement. I found myself laughing because he’s not wrong. What matters though is that he doesn’t just complain. He offers tools and frameworks that force writers to think through their stories properly rather than hoping spectacle and CGI will cover weak foundations.
A big part of his approach centres on universal themes. Not vague ones, but specific and clearly defined ideas that sit underneath the entire story. Using Star Wars: A New Hope, he challenged me to describe the core theme in a single word. My answer, “battle”, was far too broad, and he was absolutely right to push back. I must admit I was a bit embarrassed but the show goes on, and on it did! Russell’s method involves repeatedly asking why until you reach something precise enough to guide every major decision in the story. It’s tiring work and occasionally uncomfortable, but the end result is clarity. You know what the story is actually about rather than what you hope it might be.
From there, we moved into world-building. Russell argues that worlds should be built from character outward, not assembled as backdrops. In his model, the protagonist’s core desire defines both their personal arc and the shape of the world around them. Luke Skywalker wanting to leave the farm is a simple example on the surface, but Russell keeps drilling into the why behind it. Loneliness, a need for belonging, frustration, restlessness. Each layer informs the logic of the world. Without that grounding, he argues, you are not world-building at all.
He applies the same rigour to antagonists. A strong villain is not simply evil for the sake of it. The antagonist represents a concept that directly opposes the protagonist’s goal. The story works because something meaningful is pushing back at every stage. That opposition creates a structural integrity that holds up under scrutiny and prevents the kinds of narrative leaks that viewers instinctively notice even if they cannot sometimes articulate them. You know, there’s something a little off sometimes.
We also spent a fair amount of time talking about artificial intelligence. Russell uses AI extensively, which will no doubt bother some people. His position is refreshingly practical. AI handles the intellectual grunt work. Definitions, alternatives, logic checks, narrowing ideas, testing assumptions. That frees him up to do the thinking that actually matters. He talked through how he uses tools like Opus 4.5 and ChatGPT to process large amounts of material generated during his “why” exercises and turn it into something usable. The point, he was very clear, is that AI serves the writer. It does not replace them.
At one point he made the remark that if you are not good enough to improve on what AI produces, you may be in the wrong line of work. Yikes.
We finished by talking about the influences behind his thinking. Books like *The Craft of Research* and writing texts focused on structure and high concept played a big role, alongside his own lived experience. Russell is dyslexic himself and did not read fluently until law school. He understands creative frustration from the inside. For him, storytelling is not just an art form. It is a way of navigating complexity without getting lost in it.
I came away from the conversation thinking a lot about accessibility, creativity, and how technology fits into both. I’m a fan of AI as a tool myself. Russell’s methods may not appeal to everyone, but the results speak for themselves. If you are struggling with theme, world coherence, or the sense that your story keeps slipping out of your hands, there is a lot here that applies just as much to science fiction as it does to any other form of writing.
The Memo that was mentioned!
Full HTML transcription (complete):
https://www.alexanderjarvis.com/some-thoughts-on-our-business/
Full PDF (complete, ~29 pages):
https://sriramk.com/memos/katzenberg.pdf
Context tying this memo back to Eisner’s Paramount “high concept” evaluation philosophy:
https://gointothestory.blcklst.com/the-katzenberg-memo-part-3-b603f08a91ce
Eisner’s own autobiography (searchable PDF) where he discusses “high concept” origins and the Paramount approach:
https://www.worldradiohistory.com/BOOKSHELF-ARH/Biography/Work-in-Progress-Eisner-1998.pdf
Russell’s website:
https://dyslexiaclasses.com/
Forum discussion: https://richardsonsrubicon.com/community/season-5-speculative-fiction-where-worlds-meet/is-ai-a-cheat-code-for-better-worldbuilding/#post-80
And the method, distilled by AI:
Russell’s method treats story construction as a structural problem first and a creative one second. The aim is coherence, not inspiration.
1. Lock the theme
- Everything starts with a single governing theme.
- It must be specific enough to constrain the story.
- Broad ideas like “conflict” or “battle” are not usable until narrowed.
2. Use AI to offload analysis
- AI is used to speed up thinking, not replace it.
- Generate definitions, variations, and alternatives.
- Explore options quickly, then discard most of them.
- Treat all outputs as provisional.
3. Build character through opposition
- Define what the protagonist wants and keep asking why until the real motivation surfaces.
- Construct the antagonist as a direct conceptual blocker, not just a villain.
- Conflict must come from opposing goals, not coincidence.
4. Collapse the story down
- Overgenerate arcs, motivations, and plot paths. A lot of work!
- Cut aggressively until only what serves the theme remains.
5. Stress-test the logic
- Re-run the story to expose contradictions and weak causality.
- Fix problems at the structural level, not with patches.
6. Keep authorship human
- AI does the grunt work.
- Humans make the decisions.
The approach comes from Russell’s experience as a dyslexic thinker, where structure reduces friction. The goal is simple: a story that holds up under scrutiny.