The generation effect: why producing an answer beats reading one

There's a reliable finding in memory research, going back to work by Norman Slamecka and Peter Graf in the late 1970s, called the generation effect: information you generate yourself is remembered better than the same information you're simply given. It shows up in something as small as filling in a missing letter versus reading the whole word, and in something as large as writing your own explanation versus reading someone else's.

Why generating beats reading, even when the content is the same

Generating an answer forces a small act of retrieval and construction - you have to search memory, assemble something, and commit to it, even if what you assemble is wrong. That process leaves a stronger trace than passively receiving the correct answer, because encoding is deeper when you did the mental work rather than had it done for you.

This is closely related to the testing effect, but it's a distinct claim: it's not only that testing yourself is good, it's that the act of producing an answer - correct or not - is doing real work, separate from whatever feedback comes after.

A counterintuitive part: guessing wrong first, then correcting, can beat being told the answer directly

Several studies on "errorful generation" have found that attempting an answer and getting it wrong, followed by seeing the correct answer, produces better retention than simply being shown the correct answer with no attempt at all - as long as the correction actually happens. The guess itself seems to prime memory for the correction that follows. This does not mean wrong answers are good in isolation; it means the attempt is doing something reading alone cannot.

Where this shows up in studying

Try to answer before looking, even when you're fairly sure you don't know it. The attempt matters more than being right on the first try.
Writing your own summary of a concept, in your own words, beats copying or re-reading someone else's summary of the same concept.
This is also the argument against skipping straight to the answer key on practice problems. The struggle before you check is not wasted time - it's the part of the exercise doing the most.