Generating beats reading
Shown that information people produce themselves is remembered better than the same information read. This is a related mechanism, not a test of the slip box.
The numbers and sources
- Effect
- In the first five experiments, words participants generated were remembered better than the same words read, on recognition, free recall and cued recall [4]. A meta-analysis of 445 effect sizes from 86 studies put the advantage at .40 of a standard deviation [5]. A later meta-analysis of 126 articles, 310 experiments and 1,653 estimates found the size of the effect depends on how constrained the generating is [6].
- Population
- Mostly laboratory participants generating words or short items, such as producing a synonym [4][5][6].
- Caveats
- These studies tested single words and short items over short delays. Writing a card in your own words involves generating, but no study here tested a card index or links between notes.