How to Study for Exams (What Actually Works)

There is a cruel joke buried in the research on learning: the techniques that feel most productive are the least effective, and the ones that work feel like going backwards. Rereading a chapter is smooth and pleasant and produces a warm sense of mastery. Closing the book and trying to write down what was in it is uncomfortable, slow, and full of gaps — and it is worth several times as much. Students optimise for the feeling and are then astonished by the result.

The mechanism has a name — the fluency illusion. Familiarity gets mistaken for knowledge. After the fourth read, the text is easy to follow, and easy-to-follow is misread as I could reproduce this, which is a completely different capacity and the only one an exam tests. This is why highlighting is so seductive and so useless: it is an activity that produces a decorated page and no retrieval practice whatsoever.

What works, consistently and across subjects, is retrieval. Shut everything and write what you remember, then check. Answer past questions before you feel ready. Explain the topic out loud to a wall, a sibling, or a patient dog, and notice exactly where the sentences stop — that stopping point is the whole point of the exercise. Every act of pulling information out strengthens it far more than putting it in again.

The second lever is spacing. The same six hours split across six days beats six hours in one evening by a wide margin, because a small amount of forgetting between sessions is what forces effortful retrieval, which is what builds durable memory. Cramming does work for about thirty-six hours, which is exactly long enough to pass a test and lose the subject, and it is why so many people finish a degree unable to recall the second year.

Third, interleave. Doing thirty problems of one type teaches you to execute a procedure you have already been told to use; mixing types forces the harder and more useful skill of deciding which approach applies — which is what the exam actually asks. Performance in practice will drop when you mix. Performance in the exam rises. That trade is the whole discipline in miniature.

Two practical notes to finish. Sleep is not a reward at the end of studying but part of the studying, since consolidation happens during it, and the all-nighter trades the mechanism for the appearance of effort. And write the questions, not just the answers — a student who spends an hour predicting what they would ask if they set the paper understands the shape of the subject better than one who spent five hours colouring it in. Difficulty is not a sign the method is failing. In learning, difficulty is very often the method working.