How to Remember New English Words Permanently: The Method That Actually Works

Why do you forget English words so fast, and how do you remember them permanently?

In 1885, the psychologist Hermann Ebbinghaus spent years memorising nonsense syllables and measuring how quickly they slipped away. What he found became the forgetting curve: within an hour, more than half of new information is gone, and within a day, most of the rest has followed. His original experiments are still cited as the foundation of modern memory research, and you can read the Ebbinghaus forgetting curve research in its original form.

This is not a personal failing. It is how human memory is built. Your brain treats a word you have seen once as noise, not signal, and it discards noise aggressively to save energy.

The practical consequence is uncomfortable: the word "ubiquitous" that you looked up on Tuesday will be gone by Friday unless you do something specific to keep it. The question is not whether you will forget, but what you do in the window before you do.

Most learners respond to this by rereading the word list, highlighting it, or writing it out ten times. These feel productive, and that is exactly the problem. They build familiarity, not recall, and familiarity is a poor substitute for memory.

Retrieval practice: the engine of permanent memory

The single most reliable way to remember new English words permanently is to practise pulling them out of your brain rather than pushing them in. This is called retrieval practice, and the evidence for it is unusually strong. Tutoro builds its lessons around exactly this: it asks you to produce a word from memory inside a conversation, not to recognise it in a list.

In a large body of research, students who tested themselves on material remembered far more a week later than students who simply restudied it, even though the restudiers felt more confident at the time. This is known as the testing effect, and it has been replicated across hundreds of studies, summarised in the testing effect research review. The act of struggling to recall a word, succeeding or not, is what tells your brain this item matters and should be kept.

Here is the key shift. Looking at a word and its translation is passive. Covering the translation and trying to produce it from memory is active. The second takes more effort and feels worse, and that discomfort is precisely the signal that learning is happening.

There is a useful rule of thumb from the research: the harder the retrieval, the bigger the memory benefit, as long as you eventually get the answer. A review that is too easy, like recognising a word in a multiple-choice list, does almost nothing compared to producing it yourself.

This is why flashcards work better when you answer before flipping them, and why a blank page is a better study tool than a highlighted page. To remember English words permanently, you need to test yourself, not just expose yourself.

One way to make this automatic is to let the testing happen inside conversation. When a word you learned two days ago resurfaces in a chat and you have to recall it to answer, you are doing retrieval practice without scheduling it yourself. An AI English tutor that tracks your weak words and brings them back inside a lesson turns this principle into a default rather than a chore.

Spacing: why cramming never sticks

Retrieval practice tells you how to review. Spacing tells you when. The two findings work together, and separating them is the most common mistake learners make.

Cramming works for tomorrow's test. You can hold a list of forty words in your head for twenty-four hours by repeating them intensely the night before. But that memory decays almost as fast as it formed, because the brain interprets massed repetition as a short-term demand, not a long-term one.

Spacing means leaving gaps between reviews: see the word today, test yourself tomorrow, then again in three days, then a week, then two weeks. Each successful recall after a longer gap resets the forgetting curve at a higher point, so the memory gets flatter and more durable with every cycle.

A practical schedule looks like this. Day one: learn the word and use it in a sentence. Day two: recall it from a blank prompt. Day four: recall it again, this time in a new context. Day eight: produce it in conversation or writing. By day eight, the word is no longer being "reviewed"; it is being used.

The reason most vocabulary apps fail is that they make you decide when to review. You forget, you lapse, and the word quietly disappears. The system that remembers for you is the one that keeps the word alive. Weak vocabulary resurfacing on its own, days later, is spacing done properly.

Use the word or lose it

Retrieval and spacing get a word into long-term memory. Using it in a sentence you actually mean is what welds it there. A word that only ever lives on a flashcard stays a translation; a word you have used to describe your own life becomes part of your language.

There is a measurable difference between recognising a word and producing it. Recognition is passive and arrives early. Production, pulling the word out and fitting it into a sentence under time pressure, is a separate and harder skill, and it is the one that matters for actually speaking English.

This is why the best review prompt is not "what does 'reluctant' mean?" but "tell me about something you were reluctant to do this week." The second forces you to recall the word and use it meaningfully at the same time, which is exactly how native speakers hold vocabulary.

Notice what this implies for your study materials. A word list is a starting point, not a destination. The destination is a sentence, spoken or written, that you generated yourself. Every word you learn should have at least one real sentence attached to it before you consider it learned.

The technique above is hard to run by hand, because it asks you to remember what you are weak at, when you last saw it, and how to use it, all at once. Tutoro removes the bookkeeping: it tracks which words you are weak, medium and strong on, and brings the weak ones back inside a conversation rather than a review screen. If you prefer managing your own deck and your own schedule, a simple flashcard app is enough and you may not need it.

How many words can you realistically learn a day?

Learners love this question because they want a number. The honest answer is that the number matters far less than the system. Ten words a day that survive a month beat fifty words a day that are gone in a week.

That said, there is a useful ceiling. Most research and most experienced teachers converge on roughly ten to twenty new words per day as a sustainable maximum for an adult learner. Beyond that, new words start crowding each other out, and your recall rate collapses.

But the daily count is the wrong metric. The right metric is how many words you can still produce a week from now. If you learn twenty words today and can recall five next week, your real rate is five. If you learn eight and keep seven, your real rate is seven, and you are winning.

So the practical answer: start with five to eight new words a day, and spend your remaining energy on reviewing the words you learned in the previous days. Review is not the boring part of the routine. It is the routine.

Build a weekly routine that runs itself

Knowledge of how memory works is useless without a routine that enforces it. Here is a concrete weekly structure that combines retrieval practice, spacing and real use, and that you can start this week.

Step one: collect. Each day, gather five to eight new words from real material you are reading or listening to, not from a random frequency list. Words met in context are easier to remember because they arrive with a hook already attached.

Step two: make a sentence. For each new word, write or say one sentence about your own life. This is your first retrieval, and it happens on day one.

Step three: test yourself the next day. Cover the definitions and produce the word from your sentence, or produce the sentence from the word. Struggle is the point.

Step four: space the next reviews. Test again on day four and day eight, each time in a slightly different sentence so the word is not tied to a single context.

Step five: use it in real output. By the second week, the word should appear in something you wrote or said for a real purpose, not for study. That is the moment it stops being a vocabulary item and starts being yours.

The weak point in this routine is step three and four, because they require you to remember what to review and when. That is the part most learners drop within a fortnight. Automating it, so weak words come back on their own, is the difference between a routine you run for a week and one you are still running in six months.

What to do when a word keeps coming back

Every learner has a handful of words that refuse to stick, no matter how many times they are reviewed. These are usually words with no emotional hook, no obvious context, or a confusing similarity to another word you already know.

The first thing to accept is that this is normal. The forgetting curve is steeper for some items than others, and a word that keeps slipping is not evidence that your memory is broken.

The fix is to change the encoding, not to repeat the same review harder. If "eventually" will not stick as a translation, attach it to a specific memory: "I eventually found my keys under the sofa." Give it a story, an image, or an emotion. The more distinctive the hook, the harder it is to lose.

You can also lean on production. A word you have said aloud in a real sentence, even an awkward one, is remembered far better than a word you have only read. Pronunciation and memory share a pathway, and speaking a word recruits more of your brain than reading it silently.

Some words also resurface because you keep meeting them only in one form, usually as a recognition exercise. A word you can pick out of a list but cannot produce is not yet yours. The moment you have to type it or say it from nothing, the memory either forms or reveals the gap.

If a word keeps coming back after all of this, it is usually because you have not used it in a sentence you care about. Make one, however trivial, and say it out loud. That single act does more than another ten passive reviews.

The bottom line

To remember new English words permanently, you need three things working together: retrieval practice instead of rereading, spacing instead of cramming, and real use instead of recognition. None of these is complicated, but all of them require consistency, and consistency is where most good intentions die.

The learners who actually keep their vocabulary are the ones who removed the decision-making. They do not ask "should I review today?" because the review happens anyway. Whether that means a paper schedule, a flashcard app, or a tutor that brings weak words back into conversation, the principle is the same: the system remembers so you do not have to.

Start small. Five words a day, tested tomorrow, spaced over the week, used in a sentence you mean. Do that for a month and count how many of your words survive. The number will tell you more than any article can.