Verbatim and inferential memory in young and elderly adults.

Young and elderly adults were tested for inferential and verbatim recall from a list of unrelated sentences. All participants performed an encoding task (pleasantness rating or comprehension task) during input and, at test, attempted to answer a series of who questions. Half of the questions required verbatim recall; the other half could only be answered on the basis of an inference from an input sentence. For both types of question, young adults recalled more than elderly adults. Age differences were larger after pleasantness rating than after the comprehension task for both kinds of recall, suggesting that any benefits of comprehension encoding are not the result of specific attention to inferences but, more generally, result from the meaning integration or unique representation created for each sentence. Analyses of conditionalized recall (i.e., recall given a particular level of similarity between comprehension encoding and experimenter's subsequent inferential cues) showed that old adults were less likely than young adults to use inferential recall cues highly similar to their own comprehension responses.