Self-Explanation Learning Technique: 12 Prompts and Examples

Self-explanation is the habit of explaining to yourself, while you study, why each sentence or step is true and how it connects to what you already know. Instead of reading "raising the temperature speeds up the reaction" and moving on, you stop and say why: "because faster-moving molecules collide more often, and with more energy."

Students who do this learn more than students who read the same material without explaining it. A 2018 meta-analysis of 64 research reports found that prompting learners to self-explain produced a medium-sized improvement in learning, in subjects from math to computer science.

This guide covers what counts as self-explanation (and what doesn't), the research, worked examples, a list of prompts, the rubber duck method of explaining out loud, and how to fit it into a study session.

What is self-explanation?

Self-explanation is a learning strategy where you generate explanations for yourself as you study: filling in steps the text skipped, connecting new information to what you know, and noticing when something doesn't make sense. The audience is you. Your explanation can be messy, half-spoken, and full of shorthand, as long as it answers "why."

The easiest way to see it is to compare it with paraphrasing:

The paraphrase restates. The self-explanation adds a mechanism, a reason, or a link the text left implicit. That added content is where the learning happens.

Self-explanation vs the Feynman technique vs teaching

These three get mixed up, and they do different jobs:

What is the self-explanation effect? The research

The original study: what good students did differently

In 1989, Michelene Chi and colleagues recorded eight college students thinking aloud as they studied worked examples in a physics textbook chapter on Newton's laws. Afterwards, the students solved problems. The four who solved the most were labeled "Good" and the other four "Poor."

The difference was in how they had studied the examples. Good students explained each line to themselves: why this force is included, where this equation comes from, which principle justifies this step. Poor students mostly read and paraphrased.

Good students also judged their own understanding more accurately. When Poor students commented on their own understanding, 85% of the time they said they understood, even though their later problem solving showed they hadn't. That is the illusion self-explanation breaks: reading a step and following it feels like understanding it.

This study was correlational and tiny, so it couldn't show that explaining caused the better results.

The follow-up: prompting students to self-explain

In 1994, Chi's team tested the cause directly. Fourteen eighth graders were asked to explain to themselves after reading each line of a textbook passage on the human circulatory system. Ten others read the same passage twice without prompts.

The prompted group improved more from pretest to posttest. Within that group, students who generated many explanations understood more, especially on hard questions that required inferring things the text never said directly. Every high explainer ended up with the correct mental model of how blood circulates; many of the unprompted students and low explainers did not.

One honest caveat: the prompted group spent almost twice as long with the material, about an hour more. Some of the benefit may have come from extra time.

Does it hold up? The 2018 meta-analysis

Kiran Bisra and colleagues pooled 69 effect sizes from 64 research reports comparing learners who were prompted to self-explain with learners who weren't. The average benefit was g = 0.55, a medium effect, and it held across math, science, social science, and computer science, and from primary school to adults.

Two details from the analysis matter for how you study:

• Prompts about the content worked better than metacognitive prompts. "Explain why this step works" beat "describe how well you're understanding this."
• It's not only for memorizing. Prompted learners did better at remembering, explaining, and applying what they studied.

A major 2013 review of ten study techniques rated self-explanation as having moderate utility, a step below practice testing and spacing.

Self-explanation examples

Example 1: a worked calculus example

The textbook shows:

d/dx (3x² + 1)⁵ = 5(3x² + 1)⁴ · 6x

A passive reader nods. A self-explaining reader stops at each part:

• "Why the 5 in front and the power drops to 4? That's the power rule applied to the outside function, treating (3x² + 1) as a single block."
• "Why multiply by 6x? Because the inside is changing too. 6x is the derivative of 3x² + 1. That's the chain rule: derivative of the outside, times derivative of the inside."
• "Prediction: for (2x + 7)³ I'd get 3(2x + 7)² · 2."

The last line is the strongest kind of self-explanation: using the principle to predict a new case before you see it.

Example 2: a history textbook paragraph

In 1923, Germany's government printed money to pay its debts, and prices rose so fast that workers rushed to spend their wages the day they were paid.

Self-explanation: "Printing money means more money chasing the same goods, so each mark buys less. Once people expect prices to keep rising, they spend wages immediately, which pushes prices up faster still. That's why they rushed: money held until tomorrow would buy less. Connects to: why central banks today care about inflation expectations."

You don't need to be right on the first try. If you're unsure whether "expectations" belongs there, mark it and check. Spotting that uncertainty is part of the benefit.

12 self-explanation prompts to use while studying

Keep these on a sticky note or the first page of your notebook. Use one or two per paragraph or step, not all twelve.

For textbook reading

1. Why is this true?
2. How does this connect to the previous paragraph?
3. What does this remind me of that I already know?
4. What would happen if this were not the case?
5. What new information does this sentence add?
6. What's the cause, and what's the effect?

For worked examples and problem solutions

7. Which rule or principle justifies this step?
8. Why did they do this step instead of something else?
9. What is this step trying to achieve toward the goal?
10. What do I expect the next step to be, and why? (Answer before you look.)

For diagrams, graphs, and data

11. What does each part of this diagram do?
12. What would change in this graph if one variable doubled?

Notice these are all about the content. "Do I understand this?" feels useful but is the weakest kind of prompt; your honest answer is usually "I think so," which is exactly the feeling that misled the Poor solvers in Chi's study.

Rubber duck method: explaining out loud to learn

The rubber duck method comes from programming. A developer who is stuck explains their code, line by line, to a rubber duck on the desk. Somewhere in the middle, they usually spot the bug themselves. The idea was popularized by the 1999 software book The Pragmatic Programmer.

For studying, rubber ducking is simply self-explanation said out loud to a listener who can't help you. That constraint is the point: the duck can't nod along, so you have to put every step into words.

Does explaining out loud beat explaining in writing?

Sometimes. A 2016 study by Vincent Hoogerheide and colleagues found that explaining a topic to imagined learners on video improved learning, while writing an explanation for them did not. A 2021 paper by Andreas Lachner and colleagues pooled studies of explaining to a fictitious student and found a small average benefit that was larger for spoken than written explanations. In one of their own writing experiments, students who wrote explanations for themselves did better on transfer questions than students who wrote for an imaginary student.

The practical rule:

• If you're talking, talk to someone: a duck, a pet, an empty chair, or an imagined classmate.
• If you're writing, write for yourself: short why-notes in the margin, not a polished lesson.

How to rubber duck a topic

1. Put the object on your desk and your notes face down.
2. Pick one problem, proof, or textbook section.
3. Explain it from the start, one step at a time, out loud. Say why each step follows.
4. When you hesitate or say "basically," stop. That's the gap. Mark it.
5. Check the source for that spot only, then explain the section again from the top.

Recording yourself on your phone helps. Vague patches that sounded fine in the moment are obvious on playback. If you'd like a second opinion, Active Recalling's Feynman practice transcribes a spoken explanation of a chapter and compares it with your notes.

When self-explanation works best, and its limits

Best for: material with reasons and mechanisms (science, math, economics, history arguments, code) and for studying worked examples.

Limits to know:

• It costs time. Expect a section to take noticeably longer than reading it. That's the trade: slower now, fewer rereads later.
• Wrong explanations are possible. If you know very little about a topic, your explanations can be confidently wrong. Check a few against the source, especially early on.
• It's not for isolated facts. There's no "why" behind a vocabulary word or a drug's brand name. Use flashcards there.
• It doesn't replace review. Explaining today helps you understand; retrieving it again in a few days is what keeps it.

If you want a second check on your thinking, combine it with elaborative interrogation, which turns each fact into a "why" question, and with metacognition habits that test whether your sense of "I get it" is accurate.

How to use self-explanation in a study session

Self-explanation works best while you're learning new material, then hands off to active recall for review.

A 40-minute textbook session:

1. Skim the section (3 min). Headings, figures, bold terms. Know where it's going.
2. Read and self-explain (25 min). After each paragraph, write one why-note in the margin or say it aloud. For worked examples, explain each line before moving on, and predict the next step before you read it.
3. Flag the gaps (5 min). List anything you couldn't explain. Check those in the textbook or lecture slides.
4. Close the book and recall (7 min). Write the section's main ideas from memory, including the reasons. This turns your explanations into retrieval practice.

For problem sets, use the same idea in reverse: after solving a problem, explain to yourself why each step worked and where else the method would apply. It's the habit that set Chi's Good solvers apart, applied to your own work (more in how to study math effectively).