The Self-Explanation Effect: Why Explaining Beats Rereading
TL;DR
The self-explanation effect is a research-backed learning strategy where you narrate the reasoning behind each step as you study. A 2018 meta-analysis of 69 studies found it produces a moderate-to-large positive impact on learning. It beats rereading because it forces you to connect new ideas to what you already know.
What Is the Self-Explanation Effect?
The self-explanation effect is a learning phenomenon first documented by cognitive scientist Michelene Chi and her colleagues in 1989. When students explain the reasoning behind each step of a worked example out loud, they understand the material more deeply than students who simply read the same example silently.
In other words, the act of saying why something works, in your own words, as you learn it, physically changes how well your brain stores the information.
The strategy is quiet, slow, and unglamorous. It also happens to be one of the most consistently effective study techniques researchers have ever tested.
The 1989 Physics Study That Started It All
Chi and her team gave two groups of students identical physics worked examples. One group was asked to think aloud as they studied, explaining each step to themselves. The other group just read the examples.
Later, both groups were given new physics problems. The self-explaining students solved far more of them correctly, and crucially, they also solved problems that looked very different from the examples they had studied. The passive readers often got stuck on anything that was not a near-copy of what they had seen.
Good students explain the examples to themselves. Poor students do not. The difference is not intelligence. It is behavior you can learn.
Decades of follow-up research in math, biology, computer programming, and reading comprehension have replicated the pattern. A 2018 meta-analysis by Bisra and colleagues pooled 69 studies and found an average effect size of around 0.55, which cognitive scientists consider a moderate-to-large benefit. Few study techniques produce effects that reliable across subjects and age groups.
Why Explaining Out Loud Works
Three mechanisms are doing the work behind the scenes.
1. It Activates Prior Knowledge
When you explain a step, you have to pull related ideas from memory. That retrieval strengthens the connections between what you already know and the new material. Isolated facts become part of a web.
2. It Exposes Gaps
Reading creates an illusion of understanding. Explaining destroys it fast. The moment you try to say why a step is correct, you discover whether you actually know. This is closely related to the illusion of fluency, the trap of mistaking familiarity for mastery.
3. It Repairs Misconceptions in Real Time
If you explain something wrong, you often catch the error yourself a sentence or two later. The internal contradiction is uncomfortable, and that discomfort pushes you to correct the mental model instead of carrying the mistake forward.
Self-Explanation vs. Rereading: The Honest Comparison
Students default to rereading because it feels productive and takes less effort. The data tells a different story.
| Method | Effort Required | Illusion of Mastery | Transfer to New Problems |
|---|---|---|---|
| Rereading notes | Low | High | Weak |
| Highlighting | Low | High | Weak |
| Watching a video again | Low | Very high | Weak |
| Self-explaining a worked example | Moderate | Low | Strong |
The effort is the point. Techniques that feel harder in the moment usually produce more durable learning, a principle cognitive scientists call desirable difficulties.
How to Use Self-Explanation When You Study
You do not need special software or a study partner. You need a worked example or a passage, a quiet room, and a willingness to talk to yourself.
- Pick a worked example. A solved math problem, a sample essay, a step-by-step chemistry derivation, a graded code solution. The method shines when there is a correct chain of reasoning in front of you.
- Read the first step. Only the first. Resist the urge to scan ahead.
- Ask yourself: why is this step here? Say the answer out loud or write it in the margin. Not what happened, but why.
- Connect it to a rule or concept you already know. If you cannot, that is the gap. Flag it and look up the underlying rule before moving on.
- Move to the next step and repeat. By the end, you will have reconstructed the author's reasoning in your own voice.
- Try a similar problem from scratch. The real test is transfer, not recognition.
If you are stuck on how to start, use prompts like: Why did they do this step before that one? What rule justifies this move? What would break if we skipped this step? How is this step similar to the last problem I solved?
When Self-Explanation Works Best (and When It Does Not)
Self-explanation shines when the material has a clear underlying structure, like math, physics, grammar rules, programming, or logical arguments. It is less powerful for pure memorization tasks like vocabulary lists, where techniques like the Leitner flashcard system or the keyword method do more.
Research also finds that self-explanation helps novices more than experts in some domains, because experts have already automated the reasoning. If you are new to a subject, this is your technique.
How It Fits With Other Study Methods
Self-explanation is not a replacement for the other great evidence-based techniques. It is a companion.
- Use it alongside spaced repetition to space out the material over time.
- Pair it with active recall, so you not only explain examples but also test yourself without the page open.
- It is a cousin of the Feynman technique, which asks you to explain a whole topic as if to a child. Self-explanation operates at a smaller unit: one step at a time.
- It differs from elaborative interrogation, which asks why is this fact true? Self-explanation asks why is this step necessary?
How AI Tutors Can Make Self-Explanation Easier
The biggest barrier to self-explanation is honesty. Students explain something confidently, move on, and never find out whether their explanation was actually right. A good AI tutor fixes this.
When you work through a problem with LEAI, you are not handed the answer. The tutor asks you to narrate your reasoning, probes gently when a step is vague, and surfaces the exact rule or concept you leaned on without realizing it. If your explanation is wrong, the tutor does not tell you. It asks another question that leads you back to the mistake, in the same way a patient human tutor would. This is deliberately how the best AI tutors are built, as covered in our piece on why top AI tutors withhold answers.
The result is that self-explanation becomes a dialogue, not a monologue in an empty room, and gaps get caught the day you form them rather than the week before an exam.
A Habit, Not a Hack
Self-explanation is one of those techniques that sounds too simple to matter. Explain the steps. Say why. Keep going. But forty years of research keep landing in the same place: the students who ask themselves why while they study end up understanding more, remembering longer, and transferring their knowledge to new problems. The students who just read, do not.
If you only change one habit this term, change this one.
FAQ
How is self-explanation different from the Feynman technique?
Both ask you to explain in your own words. The Feynman technique has you explain an entire topic as if teaching a beginner. Self-explanation is finer-grained: you explain the reasoning behind each individual step of a worked example as you encounter it. Many students use both, Feynman for overviews, self-explanation for problem-solving practice.
Does talking out loud actually matter, or can I just think the explanations?
Speaking or writing works better than silently thinking, according to the research. Verbalizing forces you to produce a complete sentence, which exposes vagueness that silent thought lets slide. If you are in a library, write your explanations in the margin instead of saying them aloud. The key is externalizing the reasoning.
Is self-explanation useful for subjects like history or English?
Yes, but you adapt the prompt. For history, explain why an event happened, not just what happened. For a literature passage, explain how one sentence sets up the next. The technique transfers to any material with a chain of causes, arguments, or structural choices. It is less useful for pure memorization tasks like vocabulary.
Sources
- Chi, M. T. H., Bassok, M., Lewis, M. W., Reimann, P., & Glaser, R. (1989). Self-explanations: How students study and use examples in learning to solve problems. Cognitive Science, 13(2), 145-182.
- Bisra, K., Liu, Q., Nesbit, J. C., Salimi, F., & Winne, P. H. (2018). Inducing self-explanation: A meta-analysis. Educational Psychology Review, 30, 703-725.
- Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4-58.
- University of Wisconsin-La Crosse Center for Advancing Teaching and Learning: Self-Explanation (practical teaching guide).