Can you use the idea when the example changes?
A small learning exercise: after understanding a worked example, change one condition before trying again. The aim is to see what transfers, rather than count how many pages were finished. This is a proposed practice, not a claim about measured learning gains.
Example for someone learning to evaluate an AI answer:
- Read a fictional note: “The launch date is undecided.” Compare it with an answer that confidently announces Friday. Identify the unsupported claim.
- Close the example and explain the mismatch in one sentence.
- Try a variation: “Friday is the earliest possible launch, pending testing.” Would “We launch Friday” be supported? Explain which condition disappeared.
- Write a corrected version: “Friday is the earliest possible date; testing must finish before it can be confirmed.”
- Return later with a different subject and check whether you still look for missing conditions.
Keep a short record: what you predicted, what you missed, the rule you revised, and the next variation to try. “I was wrong because the condition changed” is more useful than “I am bad at this.”
If the variation is too hard, make it smaller: change the date while keeping the sentence structure, then change the condition. If it is too easy, introduce two plausible sources that disagree.
Finishing the exercise once does not show a lasting skill. The useful question is whether you can explain your reasoning and apply it to a new example without the original answer in view.
What small variation would distinguish understanding from memorizing in a skill you are learning?