Hi, this is Ray.

A few years ago I had what I now think of as the single most useful conversation of my learning life. I was at a party, trying to explain a concept I'd been studying, I think it was something about how compound interest actually works, including all the boring parts about reinvestment and frequency… to someone who, charitably, did not care. They were polite about it. But I could see in their face that I was losing them somewhere around minute three, and worse, I could feel myself losing me too. The harder I tried to explain, the more I realized I didn't fully understand the thing I'd been confidently studying for weeks.

I went home that night and basically had to relearn the concept from scratch. Because attempting to explain it to another human had revealed, with cruel precision, the exact spots where my understanding was hand-wavy. I had been reading about it. I had been highlighting things. I had been doing practice problems. I felt like I knew it. The party had told me, in 30 seconds of awkward silence, that I didn't actually know it. I'd been studying with the illusion of mastery, and the illusion had survived right up until the moment I tried to USE the knowledge by transferring it to another mind.

This experience is, it turns out, one of the most well-documented phenomena in learning research. Teaching (or even just intending to teach) forces a quality of engagement with material that ordinary studying simply doesn't. The science calls it the protégé effect, and once you understand it, you can deliberately use it to make your studying dramatically more effective. The catch is that almost nobody does, because it feels counterintuitive. Why would teaching OTHER people help YOU learn? Shouldn't it be the reverse?

Today's newsletter is about why teaching is genuinely the most selfish thing you can do for your own learning, plus how to use this even when you don't have an actual student. Let's get into it.

Most collagen supplements deliver on one promise. Carrara delivers on five.

Pique's Carrara Marine Collagen combines Type I marine collagen for visible skin radiance with rare Type II collagen from wild Hokkaido salmon, a first-of-its-kind pairing that supports both surface glow and deeper structural resilience. Add biotin for stronger hair and nails, micronized pearl powder for luminosity, and a rich coconut cream base that makes your daily ritual feel indulgent rather than obligatory.

This isn't a supplement you tolerate. It's one you look forward to. Stir it into your morning coffee or matcha and it works from the inside out to build the foundation beauty products can't reach.

Designed for superior absorption, results come faster than traditional collagen. Not a quick fix. A long-term investment in how you look and feel. 🧜‍♀️✨

Get 15% off for life

Let me start with the data, because the research here is genuinely strong. The protégé effect is the consistent finding that learners who are tasked with teaching material to others learn it better themselves than learners who are simply studying for their own knowledge. This shows up across age groups, subjects, and methodologies. It's one of the more robust findings in education research over the past 20 years.

One of the foundational studies was done with eighth graders learning biology. According to summaries of the research, students were given a computer program called Betty's Brain, with some receiving programs with Teachable Agents… digital characters that could learn based on how the students taught them. The students who taught these Teachable Agents learned faster and more effectively than those who simply studied the material on their own. The intention of teaching Betty pushed them to engage deeply with the content, leading to better retention and understanding. Same material. Same study time. The students who had to teach a digital character learned it more deeply than the students who were just studying. The teaching task changed how they engaged with the content.

But here's where it gets genuinely surprising. You don't even need to actually teach anyone. A 2014 experiment by John Nestojko and colleagues found something almost embarrassing to ordinary studying. According to the researchers, simply telling students they were going to teach the material afterwards, and not even making them actually do it, improved learning… when compared to learners expecting a test, learners expecting to teach recalled more material correctly, organized their recall more effectively, and had better memory for especially important information. The mere EXPECTATION of teaching changed how the brain processed the material. The students never actually had to teach anyone. They just believed they would. That belief alone produced significantly better learning.

This finding has been replicated and refined. A meta-analysis by Keiichi Kobayashi looking across 39 experiments found three consistent patterns. As summarized in research reviews, learning to teach significantly improves learning and retention; simply expecting to teach after a lesson can boost learning, but actually doing it is better; and most importantly, this only really works if you're told that you're going to teach before engaging with the lesson. The third point matters enormously. The effect comes from how you ENGAGE with the material, not from the teaching itself. If you find out after studying that you'll need to teach, the effect mostly disappears. You have to know going in.

The "why" of the protégé effect is more interesting than just "teaching is good." Several distinct mechanisms stack together to produce the effect.

Deeper processing during encoding. When you're studying with the intention of teaching, you process the material differently. You can't just absorb surface-level facts. You have to identify what's important, organize it logically, and figure out which parts will need extra explanation. As one breakdown of the research put it, when you prepare to teach someone else, you engage in a deeper analysis of the material. To teach effectively, you need to identify important concepts, organise them logically and fill any gaps in your own understanding. This isn't just more attention. It's a qualitatively different kind of cognitive processing. The brain treats the material as something it needs to TRANSMIT, not just RECEIVE. Different goal, different encoding.

Forced confrontation with gaps. This is the experience I had at that party. Teaching, or preparing to teach, ruthlessly exposes the parts of your understanding that don't actually hold together. You can think you understand something while reading. You can complete practice problems while having only vague intuition. But when you have to explain it to someone whose understanding starts from zero, you can't fake it. The handwaves don't survive contact with a confused student. Every gap becomes visible. And visible gaps can be filled. Hidden gaps can't.

Retrieval practice in disguise. A 2018 study by Aloysius Wei Lun Koh and colleagues hypothesized that the protégé effect is actually a form of the testing effect… the well-documented finding that retrieving information from memory strengthens learning more than passively re-studying it. To test this, they had 124 participants study the Doppler effect, then split into groups including ones that taught what they learned without notes, taught with notes, practiced retrieval without teaching, or just did unrelated arithmetic problems. The teaching-without-notes group and the retrieval-without-teaching group both showed strong learning gains, supporting the idea that the protégé effect operates partly through making you retrieve and reorganize information from memory rather than just recognize it on a page.

Better organization for memory. Information that's well-organized is much easier to remember than information that's piled in randomly. Teaching forces you to organize, because disorganized teaching is bad teaching, and your brain knows this even when preparing for a hypothetical student. This organizational pressure produces memory structures that are easier to retrieve later. You're not just memorizing facts; you're building a coherent mental model.

Motivation to actually pay attention. Knowing you'll have to explain something later creates accountability that ordinary studying lacks. As one summary noted, knowing that you'll need to explain something to others creates a sense of responsibility, which can provide the motivation we need to pay closer attention and focus more intently while studying. The social stakes of looking foolish in front of a student (even an imaginary or future one) are weirdly motivating. It's the same effect that makes you actually clean your apartment when you know guests are coming.

Here's the part most people don't realize. Because the effect is largely driven by HOW you study (your engagement, your processing, your intention) rather than by the actual teaching, you can capture most of the benefit even if no student ever materializes. You just need to simulate the conditions.

Before you study something, decide who you're going to "teach" it to. Make it specific. Not "imagine teaching it to a student." Imagine teaching it to your 14-year-old niece, or your brother who's smart but doesn't have your background, or a colleague who's never worked in your field. The specificity matters because it forces you to consider what they would and wouldn't already know, what would confuse them, what would resonate.

Then study the material with the deliberate goal of being able to explain it to that person. Read with that question in mind. Take notes with that audience in mind. When you encounter something you don't quite get, flag it, because your pretend student would definitely catch you on that. The pretend student is your conscience. Treat them like one.

Once you've studied something, give a brief lecture on it. Out loud. To no one. To an empty chair, to your dog, to the void of your shower. The act of producing the explanation verbally (not just thinking it through silently) forces a different and more rigorous level of processing.

This sounds ridiculous. It works embarrassingly well. The reason is that speaking forces a linearity and clarity that internal thinking doesn't require. When you're thinking, you can skim past unclear parts. When you're speaking, you actually have to produce coherent sentences, which means you have to actually decide what the sentences should say. The unclear parts get exposed immediately. You hear yourself struggling. You know what to study more.

Write a short article, blog post, email, or document explaining what you just learned to a hypothetical reader. This is what I do constantly, and it's part of why writing this newsletter has, paradoxically, made me a dramatically better learner over the past few years. Every article forces me to actually understand what I'm writing about well enough to explain it clearly. The articles aren't just outputs of my learning. They're catalysts for it.

You don't have to publish anything. The mere act of writing for a hypothetical reader produces the protégé effect. Sit down and write a 500-word explanation of what you've been studying. Make it readable to someone who doesn't have your background. The first draft will be terrible. You'll discover, in writing it, what you don't actually understand. Go back and study those parts. Rewrite. The thing you can finally write clearly is the thing you can finally claim to know.

If you have a friend or two who indulge your tendency to send them paragraphs about things they didn't ask about, this is a great use of the protégé effect. Take something you've been studying and send a short explanation to a friend in a text or group chat. The audience is real but low-stakes. The pressure to be coherent is real but not overwhelming. The protégé effect kicks in fully because there's a genuine person on the other end.

I have one friend who has, over the years, been the unwitting recipient of approximately 400 mini-lectures from me on everything from cognitive science to bread chemistry to whatever I happened to be reading that week. He's a saint. I've also, in those lectures, accidentally learned more than I would have learned in any of the source material alone. The audience makes the difference. Even if the audience is just one tired friend who replies "lol" thirty minutes later.

If you can find an actual person who wants to learn what you've been learning, you've found gold. Tutoring a friend's kid in something you're studying. Explaining a concept to a coworker who's curious. Running a small study group. Posting answers to questions in online communities where you've started developing some expertise.

The protégé effect with a real, engaged student is the gold-standard version. They ask questions you didn't anticipate. They catch holes in your understanding you couldn't have caught yourself. They force you to update your explanations until they actually land. It's the most efficient feedback loop on your own understanding that exists. If you have access to even one student, use them. Often, the "student" gets a lot from it too. Mutually beneficial. Find your protégé.

Here's why the protégé effect is so important. It's not just one technique among many. It's a specific instance of a broader principle that runs through almost all of effective learning: ACTIVE production produces better learning than PASSIVE consumption. Every time you take information and have to do something with it (retrieve it, explain it, reorganize it, apply it, teach it) you encode it more deeply than if you just read or watched.

This is the same principle behind active recall, behind the Feynman technique, behind practice testing, behind the writing-it-down effect. They're all manifestations of one underlying truth. The brain remembers what it has to produce, much more than what it merely receives. Teaching is just one of the highest-production-load activities you can engage in. Which is why it produces some of the strongest learning effects of any common activity.

As one analysis of the protégé effect put it, the most selfish thing you can do for your own education is help someone else understand. Every time you explain a concept, whether to a study group, a blog audience, or anyone else, you organize information more carefully, identify gaps in your understanding, and create stronger memory traces. The framing is exactly right. The selfishness isn't in spite of the helping. The selfishness IS the helping. Both you and the person you're teaching benefit. Mutual gain, not zero-sum. Rare in life. Worth using when it shows up.

Here's what I want you to take from all this. There's a deeply embedded cultural assumption that teaching is something you do AFTER you've learned something. You learn it, then you teach it. The research suggests something quite different: teaching is one of the best ways to actually do the learning in the first place. The order can be inverted. The teaching IS the learning, when you approach it correctly.

If you've been studying something for a while and feeling like the material isn't fully sticking, I'd suggest that the missing piece might be teaching it… or at least preparing to teach it. Pick a hypothetical student. Write an explainer. Give the empty-chair lecture. Find a friend who'll humor you. The protégé effect is one of the highest-leverage learning techniques available, and almost nobody uses it deliberately, because we've been told that teaching is what happens at the end of learning rather than what makes learning happen in the first place.

You don't have to wait until you're an expert to start teaching. In fact, the opposite is true. The teaching is what makes you an expert. Start now. Find someone… real or imagined… who needs to understand what you're studying. Explain it to them. Watch what happens to your own understanding. The shift is dramatic. The technique costs nothing. The benefits compound over years.

Frodo took Sam along not just for company. Sam asked questions that made Frodo articulate what he was actually trying to do. Without the explaining, the mission might not have been as clear even to Frodo. The teaching makes the learning. Find your Sam.

Keep learning (by teaching),

Ray