To turn lecture notes into a practice exam, give an AI tool your own notes and the real exam format, and ask for questions that match it. Answer them closed-book and timed, mark your answers against your notes, and list every miss as a weak spot. Restudy those, then repeat with fresh questions every few days until the exam.
Why does practice testing work better than rereading?
Pulling an answer out of your memory strengthens it more than looking at it again. Researchers call this the testing effect, or retrieval practice.
In a classic experiment, Roediger and Karpicke (2006) had university students learn short science passages. One group read a passage four times. Another read it once, then took three recall tests with no feedback. Five minutes later, the rereaders did better (83% recalled versus 71%). A week later the result flipped: 61% for the tested group and 40% for the rereaders, even though the rereaders had read the passage about 14 times and the tested group about 3.
The rereaders also felt more confident they would remember. Rereading feels like learning because the page looks familiar, but recognizing a page is not the same as producing the answer on a blank one.
A major review agrees. Dunlosky and colleagues (2013) rated ten common study techniques in Psychological Science in the Public Interest. Only two earned a high utility rating: practice testing and distributed practice (spreading study out over time). Rereading, highlighting and summarizing were among the five rated low. The Association for Psychological Science noted that both top techniques boosted performance across many kinds of tests, and a later meta-analysis by Adesope and colleagues (2017) found practice tests more beneficial for learning than restudying and every other comparison condition it examined.
How do you turn your lecture notes into a practice exam?
Give an AI tool your own notes and the real exam format, have it write matching questions, then take the test closed-book and mark it against your notes. These steps work with any AI chat tool, or with a study partner and index cards.
- Gather your notes and the exam format. Collect your lecture and reading notes and any review list your instructor posted. Pin down the format: length, question types and counts, open- or closed-book, and which weeks are covered. The syllabus or course page usually says.
- Ask for questions that match the real exam. Give the exact question mix, count and time limit. Ask for every topic in your notes, a spread of difficulty, and a few questions that combine topics. Roediger and Karpicke suggest that practising the kind of retrieval the final test demands is part of why testing helps.
- Answer under exam conditions. Close your notes, set a timer for the real time limit, and write your answers separately. No peeking. If you blank, write what you can and move on.
- Mark against your notes. Only when you finish, check each answer against your notes or textbook. If the AI marks too, have it point to the part of your notes that supports each correct answer. Roediger and Karpicke saw large benefits even without feedback, and note that feedback may help more.
- Turn mistakes into a weak-spot list. List every question you got wrong, guessed or got partly right, grouped by topic, with the reason: you did not know it, mixed it up, or ran out of time. This list, not your full notes, is what you study next.
- Repeat on a spaced schedule. Restudy the weak spots, then take a fresh practice exam with new questions a few days later, rather than three in one night.
Here are my notes for PSYC 210, weeks 1 to 6. My midterm is 75 minutes and closed-book: 30 multiple-choice and 3 short-answer questions. Write a practice exam in exactly that format, covering every topic in proportion to how much space it gets in my notes. Make about a third easy, half medium and the rest hard, including two questions that connect different weeks. Do not show answers until I paste mine.
Here are my answers. Mark each one against my notes only: correct, partly correct or wrong. Quote the line in my notes that supports the right answer, give a total score, then list my weak spots by topic, most serious first.
Here is my weak-spot list: [paste list]. Write 10 new questions only on these topics, worded differently from last time and in my exam's formats. Include two that ask me to explain why a common wrong answer is wrong.
Check before you upload course materials. Your own notes are the safest thing to upload. Slides, handouts and past exams usually belong to your instructor, and the rules vary by school and course. The University at Buffalo gives instructors sample syllabus wording that treats uploading course materials to a tool like ChatGPT without permission as a copyright and academic integrity violation. Queen's University notes that downloading materials for personal study is generally permitted under Canadian fair dealing, but uploading them to a third-party platform is a separate act. At the University of Toronto Mississauga, students are not allowed to use generative AI in a course unless the instructor explicitly permits it. Check your syllabus, and if it says nothing, ask.
Never trust an AI answer key on its own. The University of Toronto Mississauga warns that AI can produce misleading or false information. Mark against your notes or textbook. If they disagree with the AI, trust your notes and textbook, and bring the question to your TA or office hours.
What should your study schedule look like for an exam in 2 to 3 weeks?
Take a full practice exam every few days and fix weak spots in between. In a study of more than 1,350 people, Cepeda and colleagues (2008) found that the best gap between study sessions grows as the test gets further away; for a test one week out, it was roughly 20 to 40% of that week. A York University summary put it plainly: for an exam three weeks away, review several days after first learning the material, and a gap that is too short hurts retention much more than one that is too long.
| When | What to do | Goal |
|---|---|---|
| Day 1 (about 3 weeks out) | Confirm the format. Take Practice Exam 1 closed-book, mark it and write your weak-spot list. | See what you actually know |
| Days 2 to 4 | Restudy weak spots from your notes, with a 10-question quiz on them each day. | Fix the biggest gaps |
| Day 5 or 6 (about 2 weeks out) | Practice Exam 2: new questions, full format, timed. Update your list. | Check what stuck |
| Days 7 to 11 | Short quizzes on weak spots, mixed with topics you already know. | Keep older material fresh |
| Day 12 or 13 (about 1 week out) | Practice Exam 3: a full mock with the real length and time limit. | Rehearse pacing |
| Days 14 to 18 | Work through what Exam 3 exposed. Ask for harder, mixed questions. | Close the last gaps |
| 2 to 3 days before | One short mixed quiz, then one pass through your weak-spot list. | Finish without a last-night cram |
Exam in two weeks? Start at the Day 5 row with Practice Exam 1, and keep two or three days between full tests.
How do you practise different question types?
Practise in the format you will be marked in. Multiple choice asks you to recognize an answer, while short answer, problems and essays make you produce one, so give those formats extra practice.
| Question type | Ask the AI for | How to practise | Common trap |
|---|---|---|---|
| Multiple choice | Four options with believable wrong answers drawn from ideas that are easy to confuse | Cover the options, answer from memory, then choose | Spotting the right option without being able to explain it |
| Short answer | Questions with a length limit and a marking guide listing key points | Write full answers, then check them against the key points | Writing everything you know instead of answering the question |
| Problem sets | New problems like your assignments, with values changed and types mixed together | Solve on paper, timed, with no worked examples open | Practising only problem types you already know how to start |
| Essay questions | Exam-style prompts plus the criteria your instructor marks on | Outline in 15 minutes, or write one in full under time | Memorizing a prepared essay that does not fit the question |
Write three exam-style essay questions from my notes. I will outline an answer to one in 15 minutes. Then compare my outline to my notes and tell me which key concepts, evidence or counterarguments I missed. Do not write the essay for me.
What makes a practice exam worth taking?
A practice exam helps most when it feels like the real thing and you mark it honestly. Before each one, check:
- Notes closed and out of sight
- Timer set to the real time limit
- Same question types and mix as the real exam
- Fresh wording, not questions you have already memorized
- Some questions that ask you to apply or compare ideas, not only define them
- Answer key checked against your notes or textbook
- Several days since your last full practice exam
Where CMPUS fits
You can do all of this with any AI chat tool and careful prompts. CMPUS turns the method into specialized Classmates that share one memory of your courses and ask for what they need first, so you do not have to write the prompts.
Quinn, the Mock Tester, builds practice tests from your own notes, scores your answers and flags weak spots. Leo, the Personal Tutor, explains a weak spot at your level and checks that it landed. Tara, the Task Simplifier, breaks exam prep into steps with a timeline. Classmates are built to quiz and explain, not to hand you finished work. The free Guest plan gives you 5 interactions a day with no sign-up; see pricing or start studying.
Quinn builds practice tests from your own notes, scores your answers and flags weak spots.
Enter the CMPUS 5 free interactions a day. No sign-up, no card.Questions people ask
01Can AI make a practice exam from my lecture notes?
02How many practice exams should I take before a university exam?
03Am I allowed to upload lecture slides into ChatGPT or another AI tool?
04Is practice testing better than rereading my notes?
05What should I do if I do badly on my first practice exam?
06Should practice questions be harder than the real exam?
Sources
- Test-Enhanced Learning: Taking Memory Tests Improves Long-Term Retention (Roediger and Karpicke, 2006) · Psychological Science, Association for Psychological Science (PDF hosted by East Carolina University)
- Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology (Dunlosky et al., 2013) · Psychological Science in the Public Interest (record via Duke University Scholars)
- Which Study Strategies Make the Grade? · Association for Psychological Science
- Rethinking the Use of Tests: A Meta-Analysis of Practice Testing (Adesope, Trevisan and Sundararajan, 2017) · Review of Educational Research (record via Washington State University Research Exchange)
- Spacing Effects in Learning: A Temporal Ridgeline of Optimal Retention (Cepeda et al., 2008) · Psychological Science (record via University of South Florida Digital Commons)
- Researcher finds that cramming damages long-term retention · York University YFile
- AI Guidance for Syllabi and Assignments · University at Buffalo, Office of Curriculum, Assessment and Teaching Transformation
- Open letter to the Queen's community on course materials, AI tools and your course outline · Queen's University, Office of the Provost
- Generative AI and Academic Integrity · University of Toronto Mississauga
CMPUS guides are general advice, not your course's rules. Your syllabus and your instructor decide what is allowed. Spotted something out of date? Tell us.
CMPUS
Alex
Leo
Sofia
Maya
Jordan
Lexi
Quinn
Owen
Tara
Max
Sia
Jade
Noah
Ashley
Nova
Kai
