The mechanics
How memory works
To understand how to memorise effectively, you first need to understand the path a piece of information takes before it becomes knowledge you can call on in the exam room.
Working memory
Working memory handles information temporarily. Its capacity is limited, it is fragile and highly sensitive to distraction. It is what you use when reading a lesson or listening to a teacher. It gives you the feeling of understanding — but it stores nothing for the long run.
Long-term memory
Long-term memory keeps knowledge for days, months, years. Passing an exam assumes the information has been transferred there. That transfer does not happen through mere exposure: it takes a retrieval effort, a moment when you try to recall the information without having it in front of you.
The forgetting curve
German psychologist Hermann Ebbinghaus measured it as early as 1885: without recall, memory drops exponentially, and the loss is steepest at the very beginning. This is normal — your brain is sorting, and it keeps only what seems useful to it.
70%
of a course is forgotten within 48 hours, without recall
The good news fits in one sentence: every recall pushes forgetting a little further away.
After the 1st recall
a few days
After the 2nd
one to two weeks
After the 3rd
several weeks
It is this progressive lengthening that turns a revision session into lasting knowledge — and it is exactly what spaced repetition organises.
The reality check
What does not work, and why that is reassuring
Three habits dominate student revision. What they have in common is producing a feeling of mastery without producing memory.
Re-reading your notes
Re-reading creates an illusion of knowing. On the second pass everything looks familiar — but recognising a piece of information is not the same as producing it. On exam day, the question does not hand you the text to recognise.
Highlighting
Highlighting helps structure a document visually and gives the impression of working. But the gesture triggers no retrieval effort: nothing is recalled, only marked.
Cramming the night before
Concentrating your learning into one night can be enough for a test the next morning. Retention then collapses very quickly — which becomes a problem as soon as the subject comes back the following semester, in a resit or in a competitive exam.
The major review by Dunlosky and colleagues (2013) assessed ten common learning techniques.
High utility
Practice testing · Distributed practice
Low utility
Re-reading · Highlighting
Validated by research
The principles research validates
Active recall
Active recall means asking yourself a question and trying to answer it without looking at the material. The effect has long been documented under the name testing effect: testing yourself on a piece of content produces markedly better retention than re-reading it for the same amount of study time.
It applies in two complementary stages.
Structure first
A memo card isolates one concept and states it clearly. It is not a question: it is the knowledge base everything else rests on.
Then test yourself
A question card triggers the retrieval effort: multiple choice, true or false, fill in the blank, open question, free recall. The format matters little, what counts is the mental effort of reconstruction.
Spaced repetition
Rather than concentrating revision into a single session, recalls are spread over time with progressively longer intervals. Each reactivation consolidates the memory trace and pushes the next one further out.
The timing of the recall is not incidental: recalled too early, the effort is wasted since the information is still available; recalled too late, it is already gone and has to be learnt again. The right moment sits just before forgetting — hence the value of having it calculated rather than fixed in advance.
Immediate feedback
Getting something wrong is only useful if the correction arrives straight away, and explains why. A mistake corrected immediately turns into learning; a mistake left unanswered settles back in. This is the third principle, and the one most often forgotten by self-directed revision methods.
The verification
Putting it to the test: training under real conditions
The three principles above build memory. Training under real conditions checks that it holds up under pressure.
Limited time
Questions out of order
No notes to hand
It measures your actual ability to produce the answer.
It reveals weak spots while there is still time.
It reduces exam-day anxiety by making the day familiar.
In the product
How Memojin applies the method
Memojin does not add one more method: it runs the ones above, from your own course material. Here is what is available today.
One import, four working materials
PDF
Photos of handwritten notes
Slide decks
Pasted text
YouTube video
Anki deck
A lesson structured into chapters
A printable summary sheet
A set of memo cards
A quiz with answers
Everything stays editable by hand: you review, you adjust, you complete.
Generating from your courses, in detail →Two objects, two roles
The memo card structures, the quiz tests
The memo card
It states a concept as a statement — it is the knowledge base.
The quiz
It attaches one or more questions to that card: multiple choice, single choice, true or false, fill in the blank, open question, with optional image and audio.
The first structures, the second tests. It is the direct translation of the two stages of active recall.
Memo cards and quizzes, in detail →Four ways to work
Review
Go through the memo cards and rate yourself card by card.
Practise
Answer the attached questions, with immediate correction and explanation.
Mixed mode
Alternate review and practice within a single session.
The mock exam
Sit under real conditions, timer included.
Reviewing and practising are unlimited on every plan, including the free one. This is a design rule: the quota applies to content generation, never to learning.
The ways to revise, in detail →The engine
The spaced repetition engine
Every card carries its own schedule, computed from your past answers. Memojin uses FSRS, the most advanced spaced repetition scheduler available today.
For each card it models three quantities, then schedules the recall at the precise moment the third one falls below a threshold.
The card's own difficulty
The stability of your memory of it
The probability that you could still answer it on a given date
Intervals are therefore not fixed multiples applied to everyone: they adjust card by card and person by person.
20 to 30%
fewer reviews
for the same retention — that is what measurements on tools using this scheduler show. The same result, in less time.
A card's status and last-activity date follow it whichever screen it was worked on: a card seen in a mock exam counts as a card seen.Spaced repetition, in detail →
A deliberate separation
Two engines, two roles
Memojin combines two engines that do not do the same job, and the separation is deliberate.
Generative AI produces and explains
It turns your course into a structured lesson, into cards and quizzes, rephrases a concept that will not stick, builds cards targeted at a gap it has detected. That is Jin's role, and it is central to the product.
The scheduler decides when
Which cards come back today, and how long before each returns again. FSRS is a published algorithm — how it works is documented, its parameters are fitted on real revision data, and its decisions are reproducible: given the same history, it always produces the same decision. No language model takes part in choosing cards or computing intervals.
This is not a retreat from AI, it is what makes the schedule verifiable. A teacher can explain why a card comes back today, an institution can document it, and you know your revision schedule does not depend on a model's mood.
Today's session
MemoBoost, put together for you
Every day, Memojin puts together a session of around twenty cards. The selection follows an order of priority.
Cards to review, from the most forgotten to the least
Cards tied to an exam in the next two weeks that you risk having forgotten on the day
Cards you had not understood last time
Memo cards whose linked question has just been failed
A few new cards, capped in number
In the immediate run-up to an exam — the final forty-eight hours — its whole programme moves to the front.
This order is not arbitrary. It handles what is currently being lost first, then what is at stake in the short term, and it caps new material: opening too many cards today means building up a revision debt for next week.
Within a session, cards from the same lesson stay grouped, the memo card comes before its questions, and no concept is ever worked twice on the same day in two different forms.
Exam mode
Measuring, not planning
You give a date and the content it covers. From there, Memojin does not merely organise a calendar: it measures. The active memory gauge shows the share of the exam's content that is genuinely anchored — it rises with each session, and day after day it answers the only question that really matters: will you be ready?
Concepts that are still fragile are identified and prioritised, recommended sessions are placed in your calendar around your availability, and a timed mock exam checks under real conditions what the gauge is claiming.
A plan tells you where you should be; a measurement tells you where you are.
The first reassures you right up to the day it no longer matches anything — the second stays true, including when the week has gone off the rails.
One design point that matters: the exam does not alter spaced repetition. It prioritises, it makes progress visible and it sets a deadline — the memorisation mechanics themselves stay the same.
Exam mode, in detail →The assistant
Jin, for the concepts that resist
On a concept that will not stick, the assistant explains it differently, gives an example, summarises, or creates extra cards targeted at the gap it has detected.
On a text selection inside a lesson, four actions are directly available.
Explain simply
Give an example
Summarise
Generate a card
Whatever it produces is submitted to you before it enters your revision.
The breaking point
Consistency, which is the real difficulty
Knowing the method is not enough: the breaking point is sticking to it.
Sessions from ten minutes to two hours twenty, set around the slots you actually have
A daily goal you can hold
A streak of consecutive days that only counts if the cards genuinely due have been worked
Revising with others — shared spaces, leaderboards, challenges — adds the most effective lever there is on consistency: other people.
In practice
Your revision plan, in six steps
Turn the course into memo cards
One concept per card, stated clearly.
Create the questions
Each part of the course becomes one or more questions.
Answer without looking
This is the step where memory is built — the one you are tempted to skip.
Correct immediately
Read why the answer was wrong, not only that it was.
Let the algorithm place the recalls
The right moment is computed from your answers, it is not fixed in advance.
Sit a mock exam
Under real conditions, to check what the measurement claims.
For leading a group
For a teacher, a mentor or a training centre
The same principles apply to a group, with one extra benefit: they make visible what would normally only surface once the paper is handed in.
Distribute a teaching space built from your own material, with no re-keying.
Schedule a shared exam: each student receives their preparation plan set around the date.
Spot collective gaps — a concept the whole class fails signals a point of the course to revisit, not twenty students to penalise.
Track individual progress and be alerted to someone falling behind while it is still fixable.
Applied to a group, the method narrows the spread: those falling behind are identified early, and those moving ahead are not held back.
References
Sources
- Ebbinghaus, H. (1885). Über das Gedächtnis—the first experimental measurement of forgetting.
- Roediger, H. L. & Karpicke, J. D. (2006). Test-Enhanced Learning. Psychological Science—the testing effect.
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T. & Rohrer, D. (2006). Distributed Practice in Verbal Recall Tasks. Psychological Bulletin—the spacing effect.
- 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—the review that ranks the ten techniques.
- Bjork, R. A. & Bjork, E. L. (2011). Making Things Hard on Yourself, But in a Good Way—desirable difficulties.
- FSRS — Free Spaced Repetition Scheduler—open-source spaced repetition algorithm.
Going further
The method, feature by feature
Spaced repetition
How Memojin schedules every recall.
Memo cards and quizzes
Active recall in practice.
Exam mode
The preparation plan in three phases.
Calendar and planning
Building consistency in.
Collaborative space
Revising with others.
AI generation
From course to card set, in one import.
Proven progress
What the measurement says about where you are.