The Protege Effect: What the Research Shows and How to Learn by Teaching
By AI Anyone · 2025-09-20 · Updated Oct 21, 2026 · 9 min read · Learning Tips
Expecting to teach changes how you study, and actually explaining is what lasts. Here is what the protege effect research found, its limits, a worked teaching example on compound interest, and how to teach when nobody is around.
The protege effect is the finding that people learn material better when they study it in order to teach it, and better still when they actually explain it to someone. Two lab studies show it clearly: students who expected to teach a passage recalled more of it, and students who went on to teach a lesson did better on a test a week later. This article walks through what those studies found, what they do not show, and a simple way to teach something you have just learned, even with nobody around to listen.
About 320 people a month in the US search "protege effect", and another 320 search the accented spelling "protégé effect" (DataForSEO, October 2026). Curiously, more people want a saying about the idea than an explanation of it: "learning by teaching quotes" gets about 210 searches a month against about 110 for "learning by teaching" itself (DataForSEO, October 2026).
What is the protege effect?
The protege effect is the learning benefit you get from teaching, or preparing to teach, someone else. The name comes from "protégé", a person who is guided by a mentor: in this effect, the mentor is the one who learns the most.
The idea is very old. Seneca, the Roman Stoic philosopher, wrote in one of his letters to his friend Lucilius that people learn while they teach. What is newer is the evidence for it, and a clearer picture of which part of teaching does the work.
Protege effect meaning and pronunciation
In plain terms, the protege effect means "teaching it makes you learn it". Protégé is a French word and is usually said "PROH-tuh-zhay" in English. Both spellings, with and without accents, refer to the same idea.
Protege effect psychology: why teaching helps you learn
Teaching seems to help because it forces several useful study habits at once. Researchers have proposed a few overlapping reasons:
- You organize before you explain. A learner can hold a pile of facts. A teacher has to decide what comes first, what the main points are, and how they connect. That structure is easier to remember than the pile.
- You recall without the book. When you explain something out loud, you usually have to pull it from memory. Practicing recall is one of the best studied ways to make learning last; Roediger and Karpicke (2006) found that practicing retrieval of a prose passage led to better recall a week later than restudying it. Read the study.
- Gaps become obvious. Reading, you can nod past a step you do not understand. Explaining, you reach that step and stall. The stall tells you exactly what to go back to.
- Someone is counting on you. Knowing another person will rely on your explanation tends to make people study more carefully. This one is the least directly tested of the four.
None of these needs a classroom. That is the useful part: you can borrow each mechanism alone.
Protege effect research: what the studies found
Two studies separate the effect especially clearly into two parts: expecting to teach, and actually teaching. They are not what most searchers see first. The top Google results for "protege effect" this month include a Psychology Today article, a Guardian "big idea" essay and a Stanford lab page on teachable agents (DataForSEO, October 2026).
Expecting to teach. Nestojko et al. (2014) gave people a passage to study. Some were told they would later teach it to another student; others were told they would be tested. In fact, everyone took the same test and nobody taught. The people who expected to teach recalled more of the passage, especially its main points, and their recall was better organized. See the study. Just changing the goal from "pass a test" to "teach this" changed how people studied.
Actually teaching. Fiorella and Mayer (2013) had students learn a short lesson on the Doppler effect. In the first experiment, students who studied expecting to teach did better on an immediate test than students who studied expecting a test. In the second, some students went further and actually taught the lesson by recording a short video lecture, and they did better on a test a week later than students who only prepared to teach. Read the study.
| Condition | What students did | What the studies found |
|---|---|---|
| Study for a test | Studied normally | The baseline |
| Expect to teach | Studied, believing they would teach later | Better recall soon after, with main points better organized |
| Prepare and actually teach | Studied, then explained it on video without an audience present | The advantage that lasted to a test a week later |
What the research does not show
Read those results with their limits in mind:
- They are lab studies with short materials. A passage or a brief science lesson is not a semester of chemistry.
- "Teaching" was often a recorded explanation, not a live class with questions. That is good news for solo learners, but it means the research says less about the back and forth of real tutoring.
- Preparing to teach is not magic. The lasting benefit in Fiorella and Mayer (2013) came from doing the explaining. Planning a lesson you never give captures only part of it.
For a broader look at how this compares with other techniques, see our earlier piece on how the protege effect can transform your study sessions, and on the science of retrieval practice, which covers the recall side in depth.
How to teach something you just learned: a five step method
To teach something you have just learned, plan a short lesson with three main points, explain it out loud without notes, and then answer the questions a beginner would ask. Here is the method in full.
- Study it as if you will teach it. Before you read, tell yourself who your learner is: a friend, a classmate who missed the lesson, a curious adult with no background. This is the expecting-to-teach step from Nestojko et al. (2014).
- Write a lesson plan on one index card. Three main points, one example, one picture or diagram, and one question you will ask your learner at the end.
- Teach it out loud, without notes. Five minutes is enough. Speak to a person, a recording, or an empty chair.
- Answer the beginner's questions. Ask yourself the three questions a newcomer would ask: "why?", "what is an example?" and "how is this different from...?" Wherever you stall, mark it.
- Go back only to the marked parts, fix them, and teach the lesson again the next day.
A worked example: teaching compound interest
Here is the method applied to a topic many adults read about once and then forget: compound interest.
The one card lesson plan:
- Main point 1: Interest is what you earn on money you save (or pay on money you borrow), usually as a percentage per year.
- Main point 2: With compound interest, the interest is added to the balance, so next year you earn interest on the interest too.
- Main point 3: The longer the money sits, the bigger the gap between simple and compound interest becomes.
- Example: $1,000 saved at 5% a year, with interest added once a year.
- Question for the learner: "If you took the interest out every year instead of leaving it in, which column would you be in?"
The example, worked:
| Year | Simple interest balance | Compound interest balance |
|---|---|---|
| Start | $1,000.00 | $1,000.00 |
| 1 | $1,050.00 | $1,050.00 |
| 2 | $1,100.00 | $1,102.50 |
| 3 | $1,150.00 | $1,157.63 |
Where a first attempt usually stalls. Teaching this out loud, most people get through points 1 and 2 and then hit the learner's "why?": why is year 2 only $2.50 different? Answering forces the real explanation: in year 2, the compound balance earns 5% on $1,050 instead of on $1,000, and 5% of the extra $50 is $2.50. That small number is the whole idea. Once you can say it, the growing gap in later years makes sense without memorizing a formula.
This is general education about how interest works, not financial advice. For decisions about your own money, a qualified financial professional is the right person to ask.
The protege effect in the classroom
In a classroom, the protege effect shows up whenever students explain material to each other rather than only hearing it. Common versions:
- Peer tutoring: an older or more advanced student explains material to a newer one, and both benefit.
- Split topics: each student in a small group masters one part of a topic, then teaches it to the others.
- Teach-back: after a lesson, students explain the key idea to a partner in a minute or two.
- Student-made explainers: students record a short video or write a one page guide for next year's class, which mirrors the recorded lessons in Fiorella and Mayer (2013).
What is a learning by teaching example?
A simple learning by teaching example is a student who reads a chapter on photosynthesis, then explains it to a younger sibling using a drawing, and in the process discovers she cannot say where the oxygen comes from. She goes back, finds that it comes from water, and explains it again. She now knows that point better than if she had reread the chapter twice.
Who to teach when nobody is around
You do not need a willing student to get most of the benefit, because much of it comes from the explaining itself. Options, from least to most effort:
| Who you teach | How | What it captures |
|---|---|---|
| An empty chair or a desk toy | Explain out loud | Organizing and recall |
| A voice memo | Record a five minute lesson, listen back | Organizing, recall, and hearing your own gaps |
| A friend or family member | Explain, then let them ask questions | All four, including real questions |
| A study group | Each person teaches one part | All four, plus accountability |
| An AI character set up as a beginner | Explain, and ask it to question you like a newcomer | Organizing, recall and questions, any time of day |
If you like the explaining step, our guide to the Feynman technique uses a similar idea: explain the topic in simple words and look for where you get stuck.
Try this: teach today's lesson in 25 minutes
Pick one thing you learned today, at work, in class or from a video, and teach it before you go to bed.
| Step | Time | Output |
|---|---|---|
| Name your learner, then write the one card lesson plan | 5 minutes | "I am teaching a curious adult who has never heard of this", plus three main points, one example, one question |
| Teach it out loud, no notes | 5 minutes | A voice memo or a live explanation |
| Answer the three beginner questions | 10 minutes | Marks where you stalled |
| Fix the marked parts | 5 minutes | Corrections in a different color on the card |
Tomorrow, teach it again without the card.
Two ways AI Anyone can help, if you want a learner on demand:
- Make your own learner. On AI Anyone's character creator, describe a learner in one sentence, such as "a curious adult who has never taken an economics class and keeps asking why", and keep it private. Then start a voice call in your browser with nothing to install, teach it your lesson, and tap to interrupt whenever you want to restart a point. Afterwards the transcript lands in your chat, so you can find the exact moment you stalled.
- Teach the old way, to an old mind. Call the Seneca AI character and explain your lesson to it. It is an AI character, not the philosopher, but it makes a patient listener who will ask what you mean.
The short version
The protege effect is real in the lab: expecting to teach improves how you study, and actually explaining improves what lasts. Plan a three point lesson, teach it out loud without notes, answer a beginner's questions, and fix where you stall. Teach one thing tonight.