The way we teach online still hasn’t caught up to how people learn.
Most digital classrooms are paper classrooms in disguise. Learning Labs is a quiet, evidence-based practice for educators who want to teach online the way the brain actually learns. With cognitive load in mind. With real interaction. With care.
“I’m not building another course platform. I’m trying to teach the people who teach. Quietly, slowly, with the research as our floor.”
Six things the research actually says about learning online.
Tap a card once to flip it; tap again for the question. None of this is theory—it has been replicated in classrooms.
Reading slides while listening to a lecture is harder than just listening.
Learning improves when narration replaces, rather than duplicates, on-screen text. Reading and listening to the same words splits attention and overloads working memory.
Re-reading is one of the worst ways to study, but it feels like one of the best.
Pulling information out (a low-stakes quiz, free recall, a quick “what did we just cover?”) beats re-reading on every long-term retention measure tested.
Cameras-on policies don’t improve attention. They tax it.
Self-view is a continuous mirror. Constant facial monitoring drains the same cognitive resources you need for thinking. Optional camera, structured audio, breakout rooms: better.
Spacing the same content over weeks beats teaching it in one block.
Distributed practice (short revisits across days and weeks) outperforms massed practice on long-term tests. Online platforms make this almost free to implement.
Captions help everyone, not just deaf students.
Of students who use captions, most have no hearing impairment. They use them for focus, for second-language support, in noisy spaces. UDL: design the floor, not the exception.
Feedback within 24 hours is worth more than the grade itself.
Hattie’s effect-size work places timely, specific feedback among the most powerful classroom interventions. More than ability grouping. More than homework. Speed matters as much as substance.
Five principles. No tools, no hype.
Tools change every six months. These don’t.
Cognitive load is the budget.
Every screen, every prompt, every transition spends a finite resource. Design lessons that pay for what matters and skip what doesn’t.
Interaction over delivery.
A 20-minute lecture that asks one good question outperforms a 60-minute polished one. Teaching is a conversation, even online.
Difficulty is a feature, not a bug.
Desirable difficulties (spacing, interleaving, recall) feel slower in the moment and stick longer afterwards. Resist the smoothness urge.
Universal Design first.
Caption the video, label the link, structure the heading. Accessibility isn’t accommodation. It’s the floor everybody stands on.
Trust scales slower than software.
The hardest part of teaching online isn’t the LMS. It’s keeping a class of strangers feeling seen by week six.
A 30-second sanity check.
Three quick questions. Tap an option to see what the research actually says. No grading, no judgement.
What’s working in your teaching right now?
Each answer reveals one finding from the literature.
Re-reading creates fluency, which feels like mastery. Retrieval (closing the book and trying to write what you remember) is the cheapest, most underused study method we have.
Identical text plus narration overloads working memory. Use slides for what speech can’t do: visuals, structure, contrast. Strip every word your voice already says.
Attention decays after roughly 10 to 15 minutes. Short teaching segments interrupted by retrieval (a quiz, a “write what stuck”, a paired explanation) outperform long blocks on every measured outcome.
Quietly join the early list.
No mailing-list bombardment. One short note when the first cohort opens. Likely small, likely paid sliding-scale, likely worth the wait.
You’re on the list.
One quiet note when the first cohort opens. Until then, keep teaching well.