If you can follow a solution when you see it but freeze when the exam changes one number, you do not have a memory problem. You have a problem-solving transfer problem.
This guide is for math, physics, chemistry, engineering, accounting, and economics students who need to study for problem solving exams where questions look unfamiliar. The short answer: stop trying to memorize finished solutions. Build a repeatable process for identifying problem types, choosing methods, practicing without hints, and reviewing errors before exam day.
You will learn a 5-part study system that works especially well for calculation-heavy and scenario-based exams: sort problem types, choose methods deliberately, redo problems without solutions, keep an error log, and finish with timed mixed practice.
Memorizing solutions feels productive because the page looks familiar. The problem is that exams rarely ask you to reproduce the same page. They test whether you can recognize what kind of problem you are facing, select a method, adapt it, and check whether the answer makes sense.
Cognitive science has a name for this: transfer. Research on practice testing and retrieval practice shows that pulling information from memory improves later performance more than simply reviewing it. A widely cited review by John Dunlosky and colleagues rated practice testing as one of the most useful learning techniques for students. Read the review.
For problem-solving subjects, that means your study method should create small moments where you must decide what to do next. If your notes always show the next line, your brain never practices the decision that the exam will demand.
Before doing more questions, make a map of the problem types your exam can include. This is not the same as listing chapters. A chapter title like “integration” is too broad; a useful problem type is more specific, like “choose substitution,” “use integration by parts,” or “interpret area under a curve.”
Spend 20 to 30 minutes scanning lecture slides, tutorials, past papers, homework, and marking schemes. Create a simple table with 3 columns: problem type, clues, and first move. Keep it rough. The goal is not pretty notes; the goal is faster recognition.
For example, a physics student might write: “Projectile motion with two phases; clues: launch angle, time of flight, horizontal range; first move: split velocity into x and y components; common trap: using final vertical velocity too early.”
💡 If you use Snitchnotes, turn each problem type into a short quiz card: clue on the front, first move on the back. That trains recognition before calculation.
Most students practice by reading a worked solution, nodding, then trying a nearly identical problem. That can help at the start, but it breaks down when the exam changes the surface details. A better method is to pause before each solution step and ask: “Why this method, not another one?”
Worked examples are useful because they reduce unnecessary mental load while you learn a new skill. But cognitive load theory also explains why students need to move from examples to independent problem solving as expertise grows. If examples never fade, you stay dependent on them. Learn more from this overview of cognitive load theory.
Use the “method choice” drill for 10 to 15 problems per topic. You do not need to finish every calculation at first. Instead, write the method you would use, the reason, and the first 2 lines of setup. This trains the exam skill that matters most: picking the right route under uncertainty.
After you understand a worked example, do not file it away. Redo it from a blank page 24 to 48 hours later. If you can only solve it while the original solution is visible, you have recognition, not retrieval.
In classic experiments, Henry Roediger and Jeffrey Karpicke found that students who practiced retrieval remembered more later than students who repeatedly studied the same material. See the retrieval practice paper.
For calculation-heavy exams, the strongest version is “closed-solution redo.” Put the solution away, set a timer for 8 to 12 minutes, and solve from scratch. If you get stuck, write the exact question you cannot answer instead of immediately peeking. That stuck point becomes your next mini-lesson.
A good weekly target is 30 to 50 independent problems, depending on subject difficulty and exam length. If you have less time, choose fewer problems and redo them more honestly. Ten blank-page attempts beat 40 passive solution reviews.
An error log is not a punishment list. It is a study plan generator. Every wrong answer should tell you what to practice next, otherwise you are just collecting frustration.
Use a 4-category error log: concept errors, setup errors, algebra or calculation errors, and timing errors. This connects well with exam-wrapper research, where students improve by analyzing how they prepared and why they lost marks. Vanderbilt University has a practical guide to exam wrappers.
At the end of each session, pick the top 2 recurring error types and make tomorrow’s first 20 minutes about those. This matters because “study harder” is too vague. “Do 6 setup-only questions on equilibrium diagrams” is specific enough to improve.
Topic-by-topic practice is useful early because it helps you learn one method at a time. But real exams are mixed. If you only practice after the chapter heading tells you the method, you remove the hardest part of the exam: deciding what kind of problem it is.
Interleaving, or mixing related problem types, can improve category learning because it forces comparison between methods. It may feel slower during practice, but that difficulty is part of the benefit. See this review of interleaving and category learning.
Start mixed practice once you can solve basic problems from each topic with notes nearby. In the final 7 days, aim for at least 3 timed sets that combine old topics, new topics, easy marks, and stretch questions. Review each timed set the same day while your thinking is still fresh.
The best sign is not that the notes feel familiar. The best sign is that your first move gets faster and your explanations get shorter. You should be able to look at a problem and say, in 1 or 2 sentences, what type it is and why your chosen method fits.
Track 3 numbers during your last week: accuracy percentage, average minutes per problem, and repeat-error count. For example, if your accuracy rises from 52% to 72%, your average time drops from 11 minutes to 7 minutes, and sign errors fall from 8 to 2 per set, your process is improving even if it still feels difficult.
If you want a faster way to build quizzes from your notes, upload your lecture notes to Snitchnotes and generate active recall questions for each problem type. Use the AI output as a starting point, then add your own mistakes from past papers.
If this topic fits your exam, also read the worked example study method, the exam error log guide, self-explanation for STEM problem solving, and how to make your own practice tests.
Study for problem-solving exams by practicing method selection, not just final answers. Sort problems by type, write clues and first moves, redo solved examples from a blank page, keep an error log, and finish with mixed timed practice before the exam.
Memorizing solutions is risky if it replaces understanding. It can help you learn common patterns at the beginning, but exams usually change details. You need to know why each step works and how to choose a method when the problem looks different.
There is no perfect number, but many students do well with 30 to 50 honest independent attempts per week in a hard problem-solving course. Quality matters more than volume: blank-page attempts, error review, and mixed practice are more valuable than passive solution reading.
Pause for 2 minutes and write what you know, what type of problem it might be, and the first move you would try. If you still cannot continue, check only the next hint or first solution step. Then close the solution and finish independently.
The best way to study for problem-solving exams is to practice the decisions the exam actually tests. Do not spend all your time memorizing perfect solutions. Build a system that helps you identify problem types, choose methods, solve from a blank page, learn from mistakes, and perform under time pressure.
Start today with one small change: take 5 problems you already reviewed, hide the solutions, and redo them on blank paper. Then log exactly where you got stuck. That single loop will teach you more than another hour of rereading.
Snitchnotes can help you turn lecture notes, worked examples, and past-paper mistakes into active recall prompts, quizzes, and study guides, so your next session starts with practice instead of guesswork.
Notes, quiz, podcasts, flashcards et chat — en un seul upload.
Essaie ta première note gratuitement