Lately I’ve been thinking a lot about language learning, especially about why some methods seem to work so much better than others even when the amount of effort is roughly the same.
There is no shortage of methods now. Apps, textbooks, tutors, immersion, Anki, shadowing, AI tools. Everyone seems to have a system, and every system comes with its own logic.
For a long time I thought the real problem was choosing the right method. But recently I’ve started to feel that the more important question is not which method is best in general, but which kind of learning actually fits me. The search for an effective way to learn Japanese has slowly become tied up with a different kind of question, which is what sort of mind I actually have, what holds my attention, what slips away, what stays.
That sounds more philosophical than I intended. But it led me somewhere surprisingly concrete.
The Dota Observation
Some time ago I started playing Dota 2 again before bed, just casually. And I noticed something strange.
After I turned the game off and lay down, the characters kept replaying in my mind. Their abilities, their animations, fragments of team fights, little scenes from the match. I was not deliberately thinking about the game. It was more like the game was still running somewhere beneath conscious thought.
That made me wonder whether the same thing could happen with language.
It turns out there is a well-known phenomenon behind this. Harvard psychiatrist Robert Stickgold ran a study in which participants played Tetris for hours and were then woken repeatedly during the first hour of sleep. Nine out of twelve reported seeing Tetris shapes spinning and falling as they drifted off. What I found especially striking was that even amnesiac patients, who had no conscious memory of playing the game, still reported the same imagery. Their brains were replaying the experience even though they could not consciously recall it.
This is called the Tetris Effect. The brain continues processing what it has been immersed in, even below the level of awareness.
The moment I read about it, something clicked for me. It gave a name to the exact thing I had noticed after playing Dota, and it made me start wondering whether some forms of language exposure might matter more than I had assumed, especially at night.
Sleep as Continuation
What changed my thinking even more was the role of sleep.
During slow-wave sleep, the hippocampus, which temporarily holds recent memories, replays the day’s experiences and helps transfer them into longer-term storage in the neocortex. During REM sleep, the process seems to become more integrative, as newer memories are connected with older networks of knowledge. I am simplifying, of course, but the basic picture is clear enough: sleep is not an interruption of learning. It is part of learning.
Once I started thinking about it that way, the period before sleep began to look very different to me.
The last half hour of the day is usually treated as dead time. Scroll a little, watch something random, let the mind blur out. But if the brain is going to keep working on recent experience during sleep, then that window may not be empty at all. It may be the point at which the mind is quietly deciding what to carry forward into the night.
I used to spend that time scrolling in my native languages without thinking much about it. Now it feels much less neutral than it once did.
Why Some Things Stay
What interests me most is that not everything seems equally likely to replay.
Dota replays itself in my mind almost effortlessly. Grammar drills usually do not.
The difference, at least as it appears to me, is emotional engagement. The things that return at night tend to be vivid, absorbing, and charged with some kind of feeling. They do not feel like obligations. They feel alive.
This is one reason I have started to see immersion differently. A textbook may be useful, but it often remains stuck at the level of deliberate effort. Something I am genuinely absorbed in seems to pass through another channel entirely. It does not disappear when the session ends. It lingers. It follows me into bed. It returns in fragments later.
I do not mean this as a universal claim about how everyone learns best. I only mean that, for me, this distinction has become harder to ignore. The material that seems to stay with me is usually not the material I tried hardest to memorize. It is the material that felt vivid enough to keep echoing after I was done.
That has made me suspect that enjoyment is not just a pleasant extra. In some cases it may be part of the mechanism itself.
The Pull of the Unfinished
There is another idea that has been on my mind alongside this one.
In the 1920s, the psychologist Bluma Zeigarnik observed that waiters could remember unpaid orders well, but once the bill was settled, the details seemed to vanish. From that came the idea that unfinished tasks remain mentally active in a way completed tasks do not.
The stronger claims around this effect seem to be more contested now than the popular version suggests. A recent meta-analysis makes the memory advantage look less straightforward than people often assume. Still, even if the classic claim is overstated, I find the underlying intuition hard to dismiss. Unfinished things seem to hold a certain psychological tension. They remain open. They continue to ask for resolution.
That feels relevant to language learning too.
Some of the moments that stay with me most are not the ones I neatly finished and understood. They are the ones I left slightly unresolved. A grammar point I almost understood. An episode stopped in the middle. A phrase whose meaning I could feel without being able to explain it fully.
I cannot prove that the brain works on these things overnight in the dramatic way people sometimes suggest. But subjectively, unresolved material often feels more alive to me than resolved material. It keeps its grip.
What I Have Started Doing
Without really planning to, I have started shaping my evenings around these thoughts.
At night I watch anime in Japanese, not because I can justify it as an efficient study technique, but because I am actually interested in it. I usually stop before the episode is fully over, often in the middle of a scene or at a moment of tension. After that, I try not to flood my mind with other languages. I might read something light in Japanese, or I might just lie in the dark for a while.
In the morning, before I reach for my phone, I sometimes stay still for a minute and notice what remains. Occasionally it is a word. Sometimes it is a line of dialogue. Sometimes it is only a faint atmosphere from the night before. Sometimes there is nothing I can clearly catch. But even that small act of noticing has changed the way I think about what happened while I was asleep.
When I do notice something, I write it down in Obsidian. Not because it is always useful in any practical sense, but because I want a record of those moments when something seems to have surfaced on its own.
And when I return to a half-understood grammar point the next morning, there are times when it feels unexpectedly easier, as if some hidden part of the work has already been done.
What This Has Made Me Think
I do not think this means conscious study is useless. I still use deliberate tools. I still look things up. I still value explanation, repetition, and structure. I am not trying to romanticize unconscious learning.
But I have started to suspect that I spent too long imagining language learning as something that had to be forced from the front of the mind all the time. The more interesting possibility is that some of the process works best when conscious effort and unconscious processing are allowed to cooperate.
A child acquires language without knowing anything about study methods. An adult learner obviously faces a different situation, but the underlying brain is still doing what brains do. It responds to vivid input. It responds to repetition. It responds to emotional salience. And it keeps working during sleep.
The arrival of AI has changed many parts of language learning. Now it is possible to get instant explanations, tailored examples, corrections, drills, conversation, almost anything on demand. I use those tools and I am glad they exist. But none of them replace the more basic fact that the brain still has to absorb, replay, and reorganize what it is given.
Lately I have been wondering whether the most important part of studying is sometimes simply choosing the right material to place in front of the mind, then giving it enough quiet for the rest of the process to happen.
I am not presenting this as a method. It is really just a description of what I have been noticing.
The brain seems to keep studying after I stop.
And more and more, I think the real question is what kind of experience I want to leave it with before the night begins.
In the end, I want to say that the process of finding the optimal method for myself is also a journey of self-discovery.