The Forgetting Curve in Reverse: How to Make Ideas Stick for Others
23-07-2026 · 12 min read · By Anshul Garg
In 2009, Dr. Anika Patel gave what she considered the best lecture of her career. She was a biochemistry professor at Imperial College London, and she'd spent three weeks preparing a 90-minute masterclass on enzyme kinetics for her second-year students. The slides were beautiful. The explanations were precise. She covered everything the exam required, in order, with examples. Students took notes furiously. Several told her afterward it was "really clear."
Six weeks later, she gave the exam. The average score on the enzyme kinetics section was 34%. Not 34 out of 50 — 34 out of 100. Two-thirds of the material had vanished from her students' brains as though the lecture had never happened.
Dr. Patel was devastated. Then she was curious. She started reading the cognitive science literature — the same research we've explored in essays on the forgetting curve and the spacing effect — and realised something that changed her teaching forever: the problem wasn't her students. The problem was her lecture. She'd designed it to be clear. She should have designed it to be remembered.
Clarity and memorability are not the same thing. They're often opposites. And the science that explains why you forget things is the same science that explains how to make other people remember — if you flip it around.
The Encoding Problem
Everything you've learned about your own forgetting applies, in reverse, to your audience's forgetting. The forgetting curve tells us that 70% of new information is lost within 24 hours. But that's an average — and averages hide the useful variation.
Some things are remembered vividly for decades. Your first kiss. Where you were on September 11th. The time your teacher humiliated you in front of the class in fourth grade. These memories weren't encoded through repetition or study. They were encoded through a combination of factors that cognitive science has mapped precisely — and that most communicators ignore completely.
Your audience's brain doesn't record your message like a transcript. It filters it through an encoding pipeline that keeps roughly 5% and discards the rest. The question isn't whether they'll forget. It's whether the 5% they keep is the 5% that matters.
The Four Encoding Amplifiers
Research by cognitive psychologists Henry Roediger, Mark McDaniel, and Peter Brown — synthesised in their book Make It Stick — identifies four factors that amplify encoding strength. Each one is a lever you can pull:
1. Emotional activation. Information paired with emotion is encoded more deeply than neutral information. The amygdala — the brain's emotional processing centre — modulates the hippocampus, enhancing consolidation for emotionally charged material. This is why stories beat data: the narrative triggers emotion, and the emotion strengthens the memory.
2. Retrieval effort. Information that your audience has to work to recall — through questions, pauses, or challenges — is encoded more durably than information passively received. This is the testing effect applied to communication: make the audience pull the idea out rather than pushing it in.
3. Distinctiveness. Information that breaks a pattern is remembered better than information that continues one. Your brain allocates extra encoding resources to the unexpected — the surprising statistic, the counterintuitive claim, the violation of expectation. The tenth slide in a deck of ten similar slides is invisible. The slide that breaks the pattern is unforgettable.
4. Concrete imagery. Abstract concepts encoded as abstract language are fragile. The same concepts encoded as vivid, specific images are durable. "Inflation reduces purchasing power" is forgettable. "A dollar bill that could buy a loaf of bread in 1990 can now buy half a slice" is sticky. The image gives the hippocampus something physical to attach the concept to.
The Lecture That Redesigned Itself
Dr. Patel didn't abandon her enzyme kinetics lecture. She rebuilt it using the encoding amplifiers — and the results were immediate.
Before: The Clear Lecture
Her original lecture followed the standard format: concept, explanation, example, next concept. Michaelis-Menten kinetics was introduced, the equation was derived, a graph was shown, an example was calculated. Everything was orderly, complete, and correct.
The problem: orderly, complete, and correct are encoding suppressors, not amplifiers. When information arrives in a smooth, expected sequence, the brain processes it fluently — and fluency signals "this is easy, no need to encode deeply." The illusion of competence affects audiences the same way it affects learners. The smoother your presentation feels, the less your audience retains.
After: The Sticky Lecture
The rebuilt lecture opened with a question she didn't answer: "A patient arrives in the ER with a suspected overdose. You need to calculate how quickly their liver will metabolise the drug. If you get the rate wrong by 20%, the patient dies. How do you calculate the rate?"
No explanation. No slide. Just the question, and silence.
For forty-five seconds, the students sat with a problem they couldn't solve — and that discomfort triggered the encoding pipeline. Their brains registered: "I need this information. It's not available. It matters." When the Michaelis-Menten equation arrived three minutes later, it wasn't a formula on a slide. It was the answer to a life-or-death question they'd been holding in their heads.
She broke the lecture into three segments of twenty minutes each, separated by two-minute "retrieval breaks" — ungraded, low-stakes quizzes where students tried to recall what they'd just heard. Not review. Recall. The struggle to pull the information back out — the productive failure that feels like forgetting — is the signal that tells the brain to consolidate.
She replaced one of her three examples with a case where the model fails. Michaelis-Menten kinetics assumes a single substrate and no inhibitors — conditions that rarely hold in real biological systems. Showing the model's failure didn't confuse the students. It made the model memorable — because the failure was distinctive, the expectation was violated, and the brain allocated extra encoding resources to the anomaly.
The rebuilt lecture covered less material in more time. The exam scores rose from 34% to 71%. Not because the students studied more. Because the encoding was stronger. The information arrived through doors the brain keeps open — emotion, effort, surprise, imagery — instead of through the door marked "smooth, complete, and easy to ignore."
The Presentation You Already Forgot
Dr. Patel's problem is your problem. Every presentation you give, every email you write, every meeting you run, every pitch you deliver is competing against the same forgetting curve. And the default communication style in professional environments — clear, comprehensive, sequential — is optimised for the speaker's comfort, not the audience's memory.
The Slide Deck Problem
Mara Gonzalez, a product manager at a SaaS company in Austin, reviewed her team's quarterly slide decks and found a pattern: every deck had 30-40 slides, each making one point with a chart and a bullet summary. The decks were clear. They were thorough. In post-meeting surveys, attendees rated them as "informative" and "well-structured."
When Mara asked attendees, one week later, to name the three most important findings from the deck, the average respondent could recall zero complete findings. They remembered "something about churn" or "I think there was a pricing slide." Thirty slides, an hour of everyone's time, near-total evaporation.
Mara redesigned the next quarter's deck. Thirty slides became twelve. Each of the three key findings got its own "act" — opened by a surprising data point, supported by one vivid customer story, and closed with a single-sentence takeaway. She removed the executive summary slide at the beginning (it anchors the audience on your conclusion before they've felt the problem) and replaced it with a question.
Recall at one week: attendees could name, on average, 2.4 of the 3 findings. Same data. Different encoding.
The Email Nobody Reads
The same principles apply to writing. A 500-word email with five paragraphs of context, three action items buried in paragraph four, and a polite closing is designed for the writer's completeness, not the reader's memory.
The forgetting-curve-aware email: subject line is the action item. First sentence is the single most important thing. Supporting context follows, briefly. The ask is repeated at the end. The structure mirrors how the brain encodes — important first (primacy effect), important last (recency effect), context in the middle where it's least likely to be retained but most likely to be skimmed if needed.
The Stickiness Toolkit
The science converges on a set of practical techniques that work across every communication format — presentations, writing, teaching, conversation.
Open With a Gap, Not a Summary
The brain encodes information that resolves an open question more deeply than information that arrives without a question preceding it. This is the curiosity gap — the distance between what you know and what you want to know. George Loewenstein's research at Carnegie Mellon shows that curiosity functions as an information appetite: it creates a state of mild discomfort that motivates the brain to seek and encode the resolution.
Most presentations open by closing the gap: "Today I'll show you that X is true." The brain says: "Got it. X is true. I can relax now." The subsequent evidence is processed at low intensity because the conclusion is already known.
The reverse: "There's a pattern in our data that none of our existing models can explain. I've spent six weeks trying to figure it out. I think I have the answer — but I need you to see the data first." The gap is open. The brain leans in. Everything that follows is encoded more deeply because the brain is searching for the resolution.
Use the Rule of One
Research on working memory capacity — from George Miller's original "magical number seven" to Nelson Cowan's more recent estimate of three to four items — consistently shows that audiences cannot hold more than a handful of ideas in working memory simultaneously.
One presentation, one core idea. Not three. Not five. One. Everything else in the presentation exists to support, illustrate, or make memorable that single idea. If you can't state the one idea in a single sentence, you don't have a presentation. You have a data dump.
Dr. Patel's rebuilt lecture had one idea: "Enzyme kinetics determines whether a patient lives or dies." Everything else — the equation, the examples, the failure case — orbited that single emotional anchor.
End With an Image, Not a Summary
Summary slides ("In conclusion, we covered A, B, and C") are the worst possible ending for retention. The audience already heard A, B, and C. Repeating them in list form adds no encoding strength. It's the communicator's equivalent of cramming — a fluent repetition that feels productive and produces nothing durable.
Instead, close with a single vivid image or scene that embodies the core idea. Dr. Patel's enzyme kinetics lecture ended with: "The next time you're in a hospital, and you see an IV drip rate calculated on the whiteboard, that number came from the equation you learned today. If it's wrong by 20%, the patient next to you is in danger. That's why this matters."
One image. One emotional connection. One thing the brain will carry out of the room.
The Ethical Dimension
There's a tension in everything above. Making ideas stick is a communication superpower. It's also a persuasion weapon. The same techniques that help a teacher encode life-saving biochemistry can help a demagogue encode dangerous propaganda. Emotional activation, curiosity gaps, distinctive imagery — these are neutral tools that amplify whatever message they're attached to.
The forgetting curve doesn't distinguish between true and false, helpful and harmful. It just encodes what's vivid and discards what's flat. A well-told lie will be remembered longer than a poorly communicated truth. A manipulative story will outlast a genuine statistic. This is why narratives beat data — for better and for worse.
The ethical communicator's responsibility is to use encoding amplifiers for ideas that deserve to be remembered — and to recognise when they're being used against you. When a message is sticking in your mind with unusual tenacity, ask: is this staying because it's true, or because it was delivered through the doors the brain can't close?
Dr. Patel still teaches enzyme kinetics at Imperial. Her exam scores have stayed above 70% for six consecutive years — double the 34% she started with. She didn't change the material. She didn't change the students. She didn't add tutoring sessions or extra credit or review packets.
She changed the delivery. She stopped designing lectures for clarity and started designing them for memory. She opened gaps before filling them. She forced retrieval before allowing review. She broke patterns before establishing them. She replaced abstractions with images and summaries with scenes.
The forgetting curve is real. Within 24 hours, your audience will lose 70% of what you said. That number hasn't changed since Ebbinghaus measured it in 1885 and it won't change tomorrow.
But the 30% that survives isn't random. It's selected — by the encoding pipeline, by the filters your audience's brain applies to every piece of information it encounters. You can't change the pipeline. You can change what you put through it.
The question isn't whether your audience will forget. It's whether the thing they remember is the thing that matters. And that question — unlike the forgetting curve itself — is entirely in your hands.