Investigating the reinstatement effect in recognition memory.

According to many influential theories of memory, successful retrieval depends on the degree to which the processes used when attempting to retrieve information overlap with the processes that were used when learning that information. This can be illustrated using the generation effect. If participants in a memory experiment are asked to read words (eg, TABLE) and generate others from anagrams (eg, HIRAC = chair) they are more likely to remember the words they generated. This advantage is enhanced if participants have to generate the words again when their memory is tested. However, this "reinstatement effect" doesn't occur with all tasks. For example, memory for words that were read during the learning stage is not enhanced if they are read again at the test stage. The aim of this research is to establish the circumstances under which the reinstatement effect occurs. For example, it is possible that the effect only occurs with tasks that require effortful processing, such as generating from anagrams, and not with tasks that are relatively automatic, such as reading. The research will also investigate the duration of the reinstatement effect and how closely the learning and test processes must overlap in order for the effect to occur.