How to Review Notes After Class Without Rereading

Review the notes from a class within about a day of it ending, and split the time in two. Spend the first half filling in what you did not have room to write. Then close the notes and answer questions built from your own headings. The second half is the one that builds memory, and it is the half almost everyone skips.

Most advice stops after the first half, which is why so many post class reviews quietly turn into a tidy up. But these are not two steps of one task. They are two different jobs with two different deadlines, and understanding which is which tells you what to protect when you only have five minutes.

The review after class is really two jobs

Walk out of a lecture and look at what you just wrote. Most of it is a compressed record: half sentences, arrows, an abbreviation you invented four minutes in, a blank line where you fell behind. What makes that record legible at this moment is not the page. It is everything still in your head that never got written down.

That surrounding context is on a much steeper decline than the material itself. The content of a lecture fades along the usual forgetting curve, which is steep but survivable. The specific memory of which example was used to make the third point, or what the arrow between two terms was supposed to mean, is often gone within hours, because it was never encoded as anything worth keeping in the first place. The first job of the review is to move that context onto the page while it still exists. Call it repair.

The second job is different in kind. It asks whether any of the material can be produced with the page shut, which is a question about your memory rather than about your notes. Call it retrieval. Retrieval is what most of the research on effective studying is actually about, and it is reliably the half that gets dropped, because it is the half that feels bad while you do it.

How soon after class should you review your notes?

Within a day for the repair pass. The familiar 24 hour rule is right about that, but it is usually defended with the wrong argument.

The standard justification is the forgetting curve, which does fall most steeply straight after learning. That much is true. It also proves less than it appears to, because the claim it supports is one about the material, not about your notes. If timing were purely a memory question, the best moment for a first review would depend on when the exam is, not on when the class was.

That is close to what the spacing evidence actually shows. In a 2008 study in Psychological Science, Cepeda, Vul, Rohrer, Wixted and Pashler taught a set of facts to more than 1350 people, varied the gap before a single review, and then tested at delays running out to a year. The best gap was not a fixed number of hours. It scaled with the test date: roughly 20 percent of the delay when the test was a few weeks away, falling to about 5 percent when the test was a year out. For an exam six weeks after the class, that points at a first proper review roughly a week later, not the same evening.

So the deadline is not coming from the memory side at all. It is coming from the notes. Repair has a hard same day deadline, because the context that makes repair possible is the thing decaying fastest. Retrieval has a soft deadline set by the exam calendar. The reason to do a short retrieval pass after class anyway is that it is the only reliable way to find out whether the repair worked, and it costs about five minutes while you are already sitting there with the page open.

Why rereading your notes does not count as a review

Because rereading measures how familiar the page looks, and a test measures something else entirely. The cleanest demonstration of the difference comes from a 2006 experiment by Henry Roediger and Jeffrey Karpicke, and its design is unusually close to a post class review. One hundred and twenty undergraduates read a short prose passage, then either read it a second time or sat a free recall test on it with no feedback at all. Everyone took a final test 5 minutes, 2 days, or a week later.

Five minutes out, rereading won: 81 percent of the passage recalled, against 75 percent for the group that had been tested. That is the result every student has already felt, and it is why rereading remains the default. By two days the ordering had inverted, 68 percent for the tested group against 54 percent. By a week it was 56 against 42, and the group that had been tested exactly once recalled as much after a full week as the rereading group managed after two days.

Read that crossover carefully, because it is not a modest effect with a caveat attached. It is the same activity being better and then worse depending only on when you measure, and the moment it looks best is the moment you are deciding whether to keep going. Rereading your notes at six in the evening genuinely will make you feel more prepared than testing yourself would. That feeling is the illusion of competence, and it is at its strongest right after a class, when the lecture is fresh enough that every line on the page reads as obvious.

The two pass review, step by step

Here is the whole procedure for a one hour class. It takes about fifteen minutes and it has a deliberate order, because doing these steps out of sequence is what collapses the review back into rereading.

1. Fill the blanks before you read anything. Go to the white space and the half sentences, not to the top of the page. The order matters more than it sounds: start at the beginning and read forward, and you will be rereading inside a minute. Work outward from the gaps instead. Expand every abbreviation you would not recognize in a month, write out the example that went with each claim, and where you left a question mark during class, either resolve it from the slides or promote it to a question for the lecturer.
2. Turn each heading into a question, and write nothing else. This is the step that makes step three possible, and right now, with the page still open, it costs almost nothing. A heading that reads “second messenger cascades” becomes “which three second messenger cascades were covered, and what does each one trigger?” Put the questions in the margin, on the back of the page, or in a Cornell cue column if you use one. Do not write the answers next to them. Whatever sits beside a question is what your future self reads instead of thinking.
3. Close the notes and answer the questions. Out loud or on a blank sheet, both work. What matters is that a full answer leaves your mouth or your pen before the notes open again. Look first and the verdict is worthless, because a seen answer immediately reads as one you were on the point of giving anyway. Four to eight questions is a normal load for one class, and most of them take under a minute. If you find yourself glancing at the notes, put them face down.
4. Flag the misses, and only the misses. Mark the questions you could not answer, or answered vaguely, and leave everything else alone. That short list is the entire output of the review and it is worth more than the notes themselves, because it is the only part that was produced under test conditions. It becomes the opening item for your next session on that subject.

Notice what is not on the list. There is no rewriting the notes into a neat second copy, no color coding, and no reading the page through from the top. Each of those expands to fill the available time and each of them keeps the answer visible throughout, which is exactly the arrangement that buys a satisfying hour and a disappointing exam.

What if the notes are too thin to review?

Then the review has already done something useful, because it told you so while you can still do something about it. A page that cannot be repaired an hour after the class is a page that was never going to be revised from in November, and the only reason most students do not discover this is that they open the notes for the first time weeks later.

The fix belongs upstream, in how the notes get captured in the first place. Notes that survive a repair pass tend to share a few features: white space deliberately left where you fell behind, headings that mark the structure rather than just the topic, and short phrases in your own wording instead of transcribed sentences. Our guide to taking lecture notes during a lecture covers that side of the problem. Treat two or three unrepairable pages in a row as a signal to change how you capture, not as a reason to review harder.

In the short term, a thin page is still worth ten minutes. Reconstruct what you can from the slides, write the three or four questions the class was clearly building toward, and mark the parts you cannot rebuild at all. A note saying that you lost the second half of the derivation is far more useful in six weeks than a page that quietly pretends to be complete.

Where the after class review fits in the rest of the term

It is the first entry in a schedule, not a standalone habit. The point of doing retrieval after class is that it produces a short list of misses, and that list is what the later reviews run on. Without it, revision two weeks later starts from a whole chapter and no information about which parts of it are weak.

From there, the spacing evidence sets the rhythm. Because the ideal gap scales with the test date, a class in week one of a twelve week term does not need daily attention. It needs a handful of retrieval passes at widening intervals, each one starting from the items that failed last time. The relationship between the two ideas is worth understanding properly, and our explainer on active recall versus spaced repetition sets out why those two names describe a single cycle rather than rival methods. If you want a concrete calendar to hang it on, the spaced repetition schedule guide has one.

Put together, the sequence is unremarkable and that is the appeal. Capture in class, repair the same day, retrieve once while you are already there, then retrieve again on a widening gap using the misses as your queue. Nothing in it requires more total study time than what most students already spend rereading.

Generating the questions with GeniusPal

The honest friction in all of this sits in step two. Writing good questions from your own headings is not hard, but it is one more thing to do at the end of a long day, and it is the step that gets cut first. That is where a tool helps.

Feed it the repaired notes, the lecture PDF, or the slide deck, and GeniusPal writes the questions for you in one of three shapes: a quiz, a set of flashcards, or a recall drill. Each shape reproduces what step three demands, which is a prompt on screen and no answer visible until you have given yours. Uploads can be a PDF, a .pptx deck, plain text, Markdown, or a CSV, and the size ceiling is 10 MB. Two generations come free, counted once against the account rather than refilling monthly, which is enough to run the loop on one real lecture and see whether it survives contact with your week.

One limit is worth stating plainly, because it follows from everything above. No tool can do the repair pass for you. The context that makes your notes readable in six weeks is sitting in your head right now and has never been written down anywhere, so there is nothing for software to read. Repair the page yourself while you still can. The question writing is the part worth handing off.