The effects of generation on item and order retention in immediate and delayed recall

Recent research has shown that generating words from fragments, relative to simply reading them, inhibits processing of order information. Nairne, Riegler, and Serra (1991) showed that this reduction in processing of order information leads to deficits in recall performance. In three experiments, we generally replicate Nairne et al.'s results and demonstrate that the deficit in recall for the generated items is dependent on the easy distractor task and the relatively short (30-sec) retention interval they used. When a difficult distractor task was used, generating produced a deficit in amount of order information processed, but actually facilitated recall when recall was delayed 80 sec. The results are consistent with the hypothesis that generating words inhibits order processing, but they do not support the contention that the reduction in order processing is responsible for the deficit in recall that is sometimes observed for the generated items. The importance of the item-order distinction in explaining the generation effect, as well as the role of the item-order distinction in the long-term-memory arena, is questioned.