There are currently two major routes of explanation for age-related deficits in discourse memory. The first, more empirical approach, seeks complex interactions among subject, task, and text characteristics (Hultsch & Dixon, 1984; Meyer, in press; Meyer & Rice, 1983, in press). Meyer and Rice have argued that age differences are heightened when people low in verbal ability are compared, relative to when those high in verbal ability are tested, when narratives rather than expository passages are presented, and when short rather than long texts are studied. In a meta-analysis of the extant literature, we verified that the first two suggested factors – high verbal ability and expository passages – reduced the effect size of age differences in prose recall, but the third did not (Zelinski & Gilewski, in press): Age differences were reliably smaller in studies in which subjects high in verbal ability were tested and in which expository texts were used, but the differences remained highly significant across studies. Thus, the interaction of certain factors with age does account for some variance in discourse recall, but the age differences are not fully explained by this approach.
The second, more theoretically based, class of explanations for age deficits in discourse recall is that there are fundamental differences in how adequately older adults process text information as compared to younger ones. The working-memory capacity deficit model, which is reviewed in the chapters by Hasher and Zacks and by Cohen in this volume, suggests that older adults have insufficient working-memory capacity to process complex relationships among concepts in discourse and that this deficit impairs their language comprehension.