How to Maintain Knowledge After an Exam

The exam is over. The pages still look familiar. That familiarity is a weak forecast of what you will be able to produce next month.

You will not keep the whole course. Split it into three piles. Drop what you will not need. Keep a lookup pile for what you only need to find. Maintain a small set you must still produce — a prerequisite, or a procedure you will actually perform — by saying or solving it with the page closed, checking it, and coming back later.

That is how to maintain knowledge after an exam. Saving the notebook does not do it.

If the exam is still ahead and you have not chosen a set, start with how to study large amounts of information. If you were deciding whether to pack study before a date that had not arrived, that choice is spaced repetition versus cramming. This page starts after the date.

Why the grade is a weak forecast

A strong result often means the material was available on a short delay. Massed practice can look solid on the afternoon of the exam. The same ideas can be hard to produce once a few weeks pass and nothing asks for them.

While you reread, the next line confirms the last. The heading looks known. A page that feels familiar after you reread it is a weak sign you will still produce it later. In one set of experiments with prose passages, extra rereading looked better minutes later, and prior retrieval looked better a week later. Restudying also made people more confident. Confidence while the binder is open is a weak forecast of next month.

The packed review did its job when the job was the exam. The exam is finished. What is left is how long you still need a few ideas.

Split the course into three piles

Once the deadline is gone, the reason to open every page goes with it. People then abandon the course, or they try to keep all of it and quit. Triage once, from the notes you already have.

Keep the maintain pile small. Put a procedure you will perform there, and a prerequisite the next course will treat as already yours. A detail that existed so you could recognize it on the test can be dropped or left as a lookup.

Introductory statistics, after the final

Say the course was one semester of introductory statistics. Next term assumes you can read a simple result and choose a basic comparison. It will not ask you to reproduce who invented a test.

Drop

• Who proposed which test, and in what year
• The lab handout's click-path through the software
• A derivation the exam asked once, if later work hands you the result

Look up

• A short table of which test fits which data layout
• Software commands you will reopen, not recite
• Formulas, if you will be allowed to open them

Maintain. Three targets, not the chapter list.

• With the notes closed, say what a confidence interval is claiming in the sense your course used, and what makes one wider
• Sketch a simple scatter from memory and say what the slope means in the data, and what it does not mean
• Make the choice you will need next term: is this a comparison of group means, or a description of a relationship? Produce the choice and the reason, then check the note

If you must produce a formula with nothing open, that formula moves into the maintain pile. If you may look it up, it stays a lookup. The biographical slide does not move. Maintaining the formula sheet and dropping the slope keeps the wrong pile.

Notes you saved are not knowledge you can produce

The binder, the highlighted PDF, and the folder of slides are still worth keeping. They are where you check a miss. They are not the same as being able to produce the idea.

Open the confidence-interval page and you can follow it. Close it and you may have only the heading. For a maintained item, extra rereading is usually a weaker plan than trying to produce the idea and checking it. Another evening in the highlights is not the maintenance pass.

The maintenance pass

Run this on one item from the maintain pile. Not on the course.

1. Produce with the source closed. Speak the idea, write it, redraw the figure, or rework the problem. A page, a worked problem, or a diagram can be the thing you reproduce. The note does not have to become a set of flashcards. A large set of isolated facts may fit a flashcard tool. This maintained set can stay as notes. The format choice is notes versus flashcards.
2. Check. Reopen a source you trust and see what you missed.
3. Correct. Fix the miss on the note itself. A wrong sentence you leave in place is what you meet next time.
4. Wait. Stop. Do not spend the same evening walking the binder again.
5. Retrieve again. On a later day, start with the source closed. Do not open the page from the top.

Take the slope from the statistics course. Cover the note. Sketch the scatter and write what the slope is saying. Open the page. If you had the direction and missed the units, fix that sentence on the note. Stop. On a later day, sketch it again before you read.

Practice until you can produce the idea, check it, and do that again on a later day. Course studies have found better exam performance than students' usual study, and some of the advantage has still been visible weeks later. The size of the gain depends on the course. After your exam, the part to use is the later return: one clean success on the night you sort the piles does not finish the item.

If the notes are still a messy capture from lecture, repair them before you trust the check. Running produce-and-check on a note you already have is how to review notes effectively. If you are now inside the next course, the loop while that course is running is how to remember what you study.

Sleep after learning is associated with better consolidation than staying awake for the same period. Sleep does not replace a later retrieval session. An all-night reread that deletes sleep is a poor trade if later memory is the goal.

When to come back

There is no single correct gap. A useful gap depends on how long you still need the material. Come back soon enough that the idea is not already gone. If you can still produce it, lengthen the next gap. If you cannot, correct the miss and wait less before the next attempt.

In research on the length of the gap, a gap tuned to an exam a week away is not automatically the right gap for knowledge you want months later. Both cramming reviews together and waiting so long that the idea is gone can be worse than a middle gap.

Reviews of verbal memory tasks find that the gap that works best tends to grow as the time you care about gets longer. There is no formula. A fixed ladder such as 1–3–7 or 2–3–5–7 is not one.

If you need the idea next year, stopping the day after the exam is the weak plan. Research on knowledge that lasts for years finds that repeated, spaced retrieval matters more than one intense bout. Those studies do not supply a calendar for a finished course.

Students often report that a course feels gone even when the grade was high, and that a short refresher can bring parts of it back faster than the first time. That is not a measurement, and it does not keep a prerequisite usable if you never retrieve it. There is no universal percent forgotten. The forgetting curve names the pattern of decline. It is not a countdown with a number attached. A refresher is a reasonable re-entry to something you dropped. It is a poor plan for the small set you still have to produce.

The paper that comes back

If the paper comes back, use it once. Walk the misses.

Where you were wrong about an item in the maintain pile, correct that part of the note. Where you missed a detail you already dropped, leave it dropped. A miss on a lookup item means the note should be easier to find. It does not promote that item onto a review schedule.

The pass repairs the source you will check against. It is not the maintenance system. A corrected page you never produce with the source closed is still only a saved note.

Common mistakes

Keeping every card from the cram deck. The deck was built so exam-week items would feel familiar. Keeping every card puts the whole course back on a schedule. Keep the few targets in the maintain pile. Leave the rest as notes, or drop them.

Copying 1–3–7 onto a calendar and calling it science. The gap follows how long you still need the idea. Come back before it is gone, then lengthen the gap when you can still produce the idea.

Rereading highlights. Marked lines make the next sentence feel expected. They do not show that you can produce the idea with the page closed.

Waiting for the next lecturer to reteach a prerequisite you need. Some courses open with a review. Retrieve the maintain pile yourself. The next course may already assume you can use it.

Staying up to reread the binder. If you want the idea later, an all-night pass that cuts sleep is a poor trade. Sleep does not stand in for a closed attempt on a later day.

How Yuhimo fits this workflow

Yuhimo is a notes-and-learning app. The unit is a page. Learning is optional, and turning it on does not convert the page into flashcards.

After the three piles, leave dropped material alone and leave lookup notes as ordinary notes you can open. Turn learning on only for a page in the maintain pile. The steps are in turn on learning. After you do that, Start learning is the home-screen button for the session.

The screen says "Try to answer on your own" and "Then open the answers and rate how well you remember." Show answers is the check. Hide answers closes it. You then rate "How well did you remember this?" The labels are Forgot (Couldn't recall), Barely (Recalled with difficulty), Remembered (Recalled correctly), and Easily (Recalled without effort).

Yuhimo does not judge whether your answer was correct. Questions, if the page has any, are prompts, not an exam. The app does not ask for an exam date and does not build a countdown. The next return follows that self-rating. How reviews work is the return. Self-assessment is the rating.