The Learning Pyramid Is a Myth: What Works Instead

The night before your exam, you read the chapter again. You highlighted the key sentences, made margin notes in the gaps, then read the whole thing a second time for good measure. You went to bed feeling confident. You'd seen the material at least three times.

You barely passed.

If that sounds familiar, it's not because you're undisciplined. You were disciplined. You put in the time. The only thing wrong was the method — and nobody told you, because the wrong method feels exactly like it's working.

Somewhere between school and now, you probably encountered a diagram called the Learning Pyramid. It makes a clean, memorable claim: you retain roughly 5% of what you hear in a lecture, 10% of what you read, 20% from audiovisuals, 30% from demonstrations, 50% from group discussion, 75% from practice, and 90% from teaching others.

These numbers feel intuitively right. We've all experienced talking through an idea and suddenly getting it in a way passive reading never produced. So the pyramid gets pasted on classroom walls, cited in corporate training seminars, shared enthusiastically on LinkedIn.

There's just one problem. Nobody can trace where those numbers came from.

The pyramid is frequently attributed to Edgar Dale, an educator who introduced a "Cone of Experience" in his audiovisual-teaching textbooks starting in the 1940s. But his cone contained no percentages whatsoever. The specific retention rates were attached to it later by unknown sources, and researchers who have tried to trace them, including educational technology scholar Michael Molenda, have found no study that validates them.

The pyramid works well enough as a vague heuristic — active beats passive, roughly speaking. As a detailed guide for how to study? It's built entirely on sand. And if you've been unconsciously designing your learning around it, the good news is that switching methods doesn't require more time. It requires a different kind of effort.

Here's what the actual research says.

The Study Methods That Failed You

In 2013, psychologist John Dunlosky and his colleagues published a landmark review in Psychological Science in the Public Interest (full review). They examined 10 commonly used study techniques and rated each one for effectiveness: high, moderate, or low.

The results were uncomfortable reading for anyone who'd spent years in libraries.

Re-reading: low utility. When you see familiar material again, your brain interprets the resulting sense of recognition as understanding. It isn't. Recognition and recall are entirely different cognitive skills, and tests ask for the latter.

Highlighting and underlining: low utility. Passive marking feels industrious but doesn't force your brain to process anything deeply enough for durable storage.

Summarizing: low utility as well, with the review noting it can help only when people already have strong skills for identifying what matters in a text.

These three techniques are what most people default to. They're the habits handed down in school, modelled on YouTube study channels, passed between study partners. If you've relied on them for years, the research isn't calling you lazy. It's saying you were handed a broken toolbox and told it was the best one available. The toolbox isn't neutral, either — it actively produces the feeling of preparation without the substance.

Method	Utility (Dunlosky 2013)	What the experience feels like
Rereading	Low	Smooth, effortless, familiar
Highlighting / underlining	Low	Productive, thorough
Summarizing	Low	Comprehensive
Elaborative interrogation	Moderate	Slow, effortful
Practice testing	High	Uncomfortable, uncertain
Distributed practice	High	Requires deliberate planning

The Testing Effect: The Most Counterintuitive Finding in Learning Science

Here's the experiment that changed how researchers think about memory.

In 2006, Henry Roediger III and Jeffrey Karpicke at Washington University in St. Louis published a deceptively simple study (original study). Students studied prose passages; some restudied the material, while others took recall tests on it instead.

Immediately afterwards, restudying looked better. But on a final test two days and one week later, the students who had been tested remembered substantially more than those who had simply reread. Not on a trick question. On basic recall of what they'd studied.

The mechanism is called the testing effect, or retrieval practice. Every time you actively attempt to recall information from memory — even when you get it wrong, even when you fumble and guess — you strengthen your memory of that piece of knowledge. The act of trying to remember is what builds the memory. Seeing the material again does not.

A useful analogy: rereading is like watching someone else lift weights. Testing yourself is actually lifting.

This changes how you should approach every book you read, every course you take. If you finish a chapter and immediately re-read it "to make it stick," you're generating the feeling of competence without the fact of it. If instead you close the book, write down everything you can recall, then check what you missed — that's the act that makes information durable.

One practical implication: after reading this article, don't reread it. Close it and write down every major concept you can recall. That five-minute exercise will do more for your retention than reading the piece three more times.

A caveat worth stating: most of Roediger and Karpicke's work used students over relatively short delays. Results vary with material complexity and time. The direction of the evidence, however, is consistent and well-replicated across many subsequent studies.

Spaced Repetition — Timing the Test

Retrieval practice is powerful on its own. Combine it with the right timing, and you get something much harder to forget.

Hermann Ebbinghaus mapped the forgetting curve in the 1880s: memory decays in a predictable arc, steepest immediately after learning and gradually leveling off. What's less well-known is that this curve is exploitable.

If you test yourself on something just before you'd have forgotten it, you don't merely pause the forgetting — you slow it down and push the next forgetting point further out. The more times this happens at the right intervals, the more durable the memory becomes. The effect compounds.

In practice, this means spacing your reviews instead of cramming. Review a concept 24 hours after first learning it. Then four days later. Then two weeks after that. Each review session takes less time than the last, because you're testing yourself rather than re-reading — and you're arriving at each review while the memory is still intact enough to strengthen.

Apps like Anki automate this entirely. You grade your own recall on each flashcard, and the algorithm calculates the optimal interval before your next review. It sounds trivial until you've used it consistently for a few months and find you can still accurately recall material from early in the year — without any deliberate extra effort.

The review sessions work precisely because they are active, not passive. When a card comes up, resist flipping it immediately. Sit with the discomfort of partial recall for a few seconds. That struggle is the mechanism doing its work — arriving at an answer after effort builds a stronger trace than arriving there instantly. It's the difference between retrieving something and simply recognizing it.

best books to learn faster

Ask "Why?" in the Margin

There's a third technique from Dunlosky's review worth keeping: elaborative interrogation, which the review rated as moderate utility.

Instead of reading a fact and moving on, you stop and ask yourself why it's true. What's the mechanism? How does it connect to something you already know?

Take the claim that cortisol elevation suppresses immune function. Most readers note it and continue. Elaborative interrogation pauses and asks: why? What biological logic would lead a stress hormone to suppress immunity? When you try to answer — even imperfectly — you're forcing your brain to connect new information to existing structures. That connection is what makes knowledge retrievable later. Information filed with context is information you can actually find.

You can practice this at zero cost in any book, with any pen. Write "why?" in the margin next to significant claims. Try to answer before reading on. When you can't, you've found a genuine gap in your understanding — which is the most useful thing you can find while studying.

How to Start Today

You don't need to overhaul everything at once. Here's a practical protocol assembled from the research above:

Read once, then close the book. Write down everything you remember without looking. It will feel like failing. That discomfort is learning happening.

Check what you missed. Re-read only the sections you blanked on. Then do a brief second recall attempt.

Create retrieval prompts. For each major concept, write one question on one side of a card and the answer on the other. Physical flashcard packs work fine. So does Anki, which is free.

Space your reviews. First review the same evening. Second review the following morning. Third review four days out. After that, software handles the scheduling.

Ask "why?" on anything important. Two sentences in your own words, connecting each concept to something you already knew before today.

This takes more time up front than passive rereading. The difference is that this approach actually produces retention, which rereading was never doing.

how to keep your brain sharp science-backed protocol

---

The Learning Pyramid didn't spread because it was accurate. It spread because it was comforting — it made passive activities like watching videos and listening to lectures feel like legitimate learning. The methods that actually work require effort that feels unpleasant in the moment, which is precisely why they get skipped.

Most study advice optimizes for the feeling of learning, not the fact of it. Rereading makes you feel prepared. Testing yourself shows whether you actually are — and that sting is exactly the signal your memory uses to build something durable.

Vanulos runs on one conviction: you can design almost any capability into your life once you understand the mechanism. Learning is one of the clearest demonstrations of this. You're not fighting your memory. You're either working with it or, until now, unknowingly against it.

What's one topic you've been "studying" by rereading that you've never once tested yourself on?