CURIOSITY → EXPERIMENT → EVIDENCE

Find out how
you learn.

Try two approaches. Measure what you can do without help. Turn a small experiment into a better learning decision.

4 starter experiments · Fair-comparison checklist · Results notebook

Your experiment guide

1. Ask a small question

Replace “What is the best way to learn?” with “Which of these two approaches helps me solve this kind of problem independently?” Name a task you can score.

2. Change one main thing

Keep time, topic difficulty, feedback and testing conditions similar. Comparing ten minutes of one method with an hour of another makes the result hard to interpret.

3. Predict before testing

Record what you expect and why. A prediction makes it easier to notice when your evidence disagrees with your expectations.

4. Test independent performance

Close your notes and AI assistant. Use fresh but comparable questions and a scoring rule you chose in advance. Feeling fluent is a different measurement from solving a task.

5. Revisit and reverse the order

Use a delayed check at the same interval for both approaches. Across repeated trials, reverse the order to explore fatigue and practice effects. Different topics may still differ in difficulty.

6. Make a modest decision

A small personal comparison is a clue, not proof of a universal rule. Record confounders, repeat with a new topic and adjust only what your evidence supports.

This is a personal reflection tool, not a controlled scientific study. Calculations use your inputs; no AI service or automatic grading is connected.

Use Recall & Transfer Lab for independent practice →

01 / MAKE IT TESTABLE

Build your experiment

Start from a template

Loading a template replaces the question and approach fields after confirmation. Your other notes stay intact.

Fair-comparison check Similar total study time Comparable task difficulty Same scoring rule and maximum score Both checks completed without assistance Same delay before later checks

02 / COLLECT YOUR EVIDENCE

Record a paired trial

Each trial compares A and B. Use the same maximum score for both. Leave delayed scores blank until you have checked later; zero means you tested and earned zero.

03 / CHOOSE YOUR NEXT MOVE

Turn results into a decision

Your work saves in this browser. Download a report before starting a new experiment.