Why writing down your reasoning makes a chess lesson stick
Fifty years of memory research say the same thing: what you generate yourself, you keep. What you read off a screen, you lose. Here is why a coach asks before they tell.
Published 2026-09-03
You lose. You open the game review. There is a red mark on move 25, the engine shows you the move you should have played, you nod, you close the tab. Next Tuesday you play the same move in the same kind of position, and you are surprised.
Nothing about that is a character flaw. It is exactly what fifty years of memory research would predict. Reading an answer is one of the weakest things you can do with your brain, and the evidence for that is unusually consistent. What follows is the short version of that evidence, what it says a coach is actually doing when they ask you a question, and what it means for how you should spend the ten minutes after a game.
What you produce, you keep
In 1978, Norman Slamecka and Peter Graf ran five experiments with a simple design. One group read pairs of words. The other group had to generate the second word themselves from a rule and a cue. On every test they gave, recognition, free recall, cued recall, the people who had generated the words remembered them better than the people who had merely read them.1
They called it the generation effect, and it has been replicated hundreds of times since. A meta-analysis of 86 studies and 445 effect sizes puts the average advantage of generating over reading at roughly 0.4 standard deviations, a moderate, reliable effect that shows up across materials and test types.2
Now look at what the engine asks of you. It shows you Qe6. You read it. That is the reading condition. You have generated nothing, so there is nothing for the effect to work on.
Explaining it to yourself is where understanding forms
Michelene Chi and her colleagues watched physics students study worked examples while thinking aloud. The students who went on to solve new problems well were the ones who spontaneously explained each step to themselves as they read, and who noticed when they did not understand something. The students who did poorly read the same examples and said things like “okay, that makes sense.”3
That study was observational. The follow-up was not. Chi's group took eighth-graders reading a text about the circulatory system and prompted half of them to explain every line to themselves. The prompted group learned more, and the gap was widest on the deep questions, the ones that required transferring the idea to a situation the text had not covered.4A 2018 meta-analysis of 64 studies puts the effect of prompting self-explanation at about half a standard deviation.5
The result has been tested in chess too. Novices who had to predict a computer's moves in a king-and-rook endgame and explain why those moves were played developed a better principled understanding, and played the endgame better, than novices who only predicted or only watched.6That is beginners on one endgame, so it is a small study, but it points the same way as everything else.
“Why?” is a learning technique
Asking “why” about a fact you are trying to learn has a name in the literature, elaborative interrogation, and the original demonstration is almost comically simple. Adults were given arbitrary sentences to remember. The ones asked to answer why each sentence might be true remembered them better than the ones who just read them, and the more precise the explanation they produced, the better they remembered.7
A word of honesty here, because it matters for what we claim. In the most-cited review of study techniques, elaborative interrogation and self-explanation are rated “moderate utility,” not “high,” because there is less classroom-scale evidence for them than for practice testing and spaced study.8Rereading, highlighting and summarising sit at the bottom of that same table. Reading an engine line is rereading.
The answer key feels great. That is the problem.
Robert Bjork has spent decades on a strange fact: the conditions that make learning feel easy and fast during practice are often the ones that produce the least durable learning, and the conditions that feel slower, generating, being tested, spacing your study, produce more. He calls them desirable difficulties.9,10
The engine is the easiest possible experience. The answer appears instantly, it is correct, and you feel like you have learned something. In one classic study, students who reread a passage did better than students who tested themselves on it, if you asked them five minutes later. A week later the picture had reversed: the rereaders remembered about 40 percent, the self-testers about 61.11Your sense that the engine line “made sense” is the five-minute measurement.
The question comes before the answer for a reason
Put the findings together and the shape of a good lesson is obvious. Before the coach tells you anything, they ask what you were thinking. You have to generate a reason, which is the generation effect. You have to explain your own move, which is self-explanation. The question is a why question, which is elaborative interrogation. And it is harder and slower than being told, which is the desirable difficulty.
Here is the same moment, from a real game analysed on Chessalyz, told both ways.
19. Bxe3 ?
alternatives: Qe6, Qxe5
Read it. Nod. Close the tab.
“You spent a real think on Bxe3. What was your evaluation of the resulting trades, and what other candidate moves were you comparing it against?”
You write: “This seemed to be the simplest reply to prevent the fork on the rook and the queen by exchanging pieces. I get rid of that dangerous attacker, and I could not see an immediate mating sequence that Black is threatening.”
Now the coaching has something to land on: your reason, in your words.
One caveat we take seriously. What you write after the game is not a transcript of what you thought during it. Retrospective reports are reconstructions, and asking someone for reasons can change the reasoning.12That is fine. The written answer is not evidence about the past, it is material for the lesson. It is the thing the coach can correct, and correcting a stated reason is what changes the next game.
- 1.Slamecka, N. J., & Graf, P. (1978). The generation effect: Delineation of a phenomenon. Journal of Experimental Psychology: Human Learning and Memory, 4(6), 592–604. doi.org/10.1037/0278-7393.4.6.592
- 2.Bertsch, S., Pesta, B. J., Wiscott, R., & McDaniel, M. A. (2007). The generation effect: A meta-analytic review. Memory & Cognition, 35(2), 201–210. doi.org/10.3758/BF03193441
- 3.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. doi.org/10.1207/s15516709cog1302_1
- 4.Chi, M. T. H., de Leeuw, N., Chiu, M.-H., & LaVancher, C. (1994). Eliciting self-explanations improves understanding. Cognitive Science, 18(3), 439–477. doi.org/10.1207/s15516709cog1803_3
- 5.Bisra, K., Liu, Q., Nesbit, J. C., Salimi, F., & Winne, P. H. (2018). Inducing self-explanation: A meta-analysis. Educational Psychology Review, 30(3), 703–725. doi.org/10.1007/s10648-018-9434-x
- 6.de Bruin, A. B. H., Rikers, R. M. J. P., & Schmidt, H. G. (2007). The effect of self-explanation and prediction on the development of principled understanding of chess in novices. Contemporary Educational Psychology, 32(2), 188–205. doi.org/10.1016/j.cedpsych.2006.01.001
- 7.Pressley, M., McDaniel, M. A., Turnure, J. E., Wood, E., & Ahmad, M. (1987). Generation and precision of elaboration: Effects on intentional and incidental learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13(2), 291–300. doi.org/10.1037/0278-7393.13.2.291
- 8.Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58. doi.org/10.1177/1529100612453266
- 9.Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe & A. P. Shimamura (Eds.), Metacognition: Knowing about knowing (pp. 185–205). MIT Press. direct.mit.edu/books/edited-volume/3931/chapter/164557/Memory-and-Metamemory-Considerations-in-the
- 10.Bjork, E. L., & Bjork, R. A. (2011). Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning. In M. A. Gernsbacher et al. (Eds.), Psychology and the real world (pp. 56–64). Worth Publishers. bjorklab.psych.ucla.edu/wp-content/uploads/sites/13/2016/07/EBjork_RBjork_2011.pdf
- 11.Roediger, H. L., III, & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. doi.org/10.1111/j.1467-9280.2006.01693.x
- 12.Ericsson, K. A., & Simon, H. A. (1980). Verbal reports as data. Psychological Review, 87(3), 215–251. doi.org/10.1037/0033-295X.87.3.215
Your last game has a question in it.
Connect Chess.com or Lichess, or paste a PGN. Chessalyz finds the moments that decided the game, asks what you were thinking, and coaches from your answer. Five games a month are free.
Try it on your last game- The engine can't teach you. Here is what a question does that +1.3 cannot.
Feedback only works when it lands on a reason. An evaluation is a grade with nothing underneath it. A question makes you commit, and a committed error is the one you remember correcting.
- Make new mistakes: why your flashcards should be your own blunders
Retrieval beats re-reading, spacing beats cramming, and a position you actually lost beats a puzzle you never met. The research behind flashcards built from your own games.
- Learning by thinking: what a fifteen-minute post-mortem does that another game does not
Masters have always analysed their own games. Cognitive science explains why that habit, not raw volume, is what separates the players who improve from the players who just play.