Why Rereading Notes Does Not Work for Maths

What the feeling of understanding actually measures

Following a worked solution requires only that each line makes sense given the line above it. That is a low bar, and a well-written example is designed to clear it effortlessly. What it never asks you to do is decide anything — which method, which rearrangement, whether to substitute now or later, whether to draw the diagram. Those decisions have already been made and printed, and they are the part the examination is testing.

There is a second effect on top of that. The second read is faster and smoother than the first, students correctly notice the improvement, and they attribute it to learning. It is the same mechanism that makes a song you have heard twice feel known: the material has become familiar, which is not the same as being available when you need it from nothing.

The two-minute test

This is worth running now rather than being persuaded of, because the demonstration is more convincing than the argument.

Most students stall somewhere between line two and line four, and the stall is nearly always at the same kind of place: the step where a decision was made rather than a calculation performed. The gap between 'I understood that' thirty seconds ago and a blank page now is the entire point of this page.

• Pick a worked example you have just read and believe you understand.
• Cover it completely — turn the page over, do not leave it under your hand.
• On blank paper, write the whole solution from the question down, every line.
• Uncover it and compare line by line, marking your own version honestly.
• Note where you stopped. That location, not the topic, is what needs work.

What to do with the same twenty minutes

None of these takes longer than rereading. They are the same twenty minutes spent so that something has to come out of your head rather than into it.

The third one matters more than it looks. Reproducing one example teaches you that example; changing the numbers is what converts a memorised solution into a method you own.

• Reproduce the example from blank, then mark your version against the original.
• Do the same question again three days later, cold, without looking at anything first.
• Do a question of the same type with different numbers, ideally from a different book.
• Say the choice out loud: 'I multiplied both sides by 12 because…'. If you cannot finish that sentence, you have found the missing piece.
• Work a mixed set with no topic heading on it, so that choosing the method is part of the task rather than something the page has done for you.

Why maths punishes this harder than other subjects

In history, rereading at least deposits something: dates, names, an argument you can half-repeat. In mathematics the content is a procedure and a set of decisions, and neither is deposited by reading. You finish the session holding a memory of having watched something being done.

Then add the conditions. No notes, unfamiliar numbers, a question deliberately written so that it does not look like the example, and a clock. A student whose entire preparation consisted of solved solutions has practised nothing the paper asks for, which is why the gap between what they expected and what they scored is so wide and feels so unfair.

Rereading is not laziness. It is the most conscientious-feeling activity available to a student — quiet, orderly, uninterrupted, and it produces no evidence of failure. That is exactly why it survives, and why telling students it does not work rarely changes anything until they have run the test above on themselves.