Two experiments were performed to investigate the effects of prior knowledge on recognition memory in young adults, younger old adults, 76-year-olds, and 85-year-olds. In Experiment 1, we examined episodic recognition of dated and contemporary famous persons presented as faces, names, and faces plus names. In Experiment 2, four types offaces were presented for later recognition: dated familiar, contemporary familiar, old unfamiliar, and young unfamiliar. The results of both experiments showed that young adults performed better with contemporary than with dated famous persons, whereas the reverse was true for all groups of older adults. In addition, the data of Experiment 2 indicated that (l) young adults showed better recognition for young than for old unfamiliar faces, (2) younger old adults performed better with old than with young unfamiliar faces, and (3) the two oldest age groups showed no effect of age of face. These results suggest that the ability to utilize rich semantic knowledge to improve episodic memory is preserved in very old age, although the aging process may be associated with deficits in the ability to utilize prior knowledge to support memory when the underlying representation lacks semantic and contextual features. The overall data pattern was discussed in relation to the notion that, with increasing adult age, there is an increase in the level of cognitive support required to enhance episodic remembering. Recent reviews of the literature on memory and aging suggest that although older adults are typically impaired in performing tasks that draw on basic episodic memory skills, the magnitude of the age-related deficit in episodic remembering varies greatly as a function of multiple external and subject-related factors (Backman, 1989; Backman, Miintyla, & Herlitz, 1990;. Regarding external factors, several investigators have made the observation that age-related differences in memory performance are pronounced in tasks that involve little cognitive support (e.g., free recall of single words), whereas tasks in which the instructions, the retrieval conditions, or the material itself support the learner in initiating appropriate memory operations show reduced or eliminated age differences (e.g., Backman, 1985aBackman, , 1986Backman & Nilsson, 1984Craik, 1983Craik, , 1985Craik, Byrd, & Swanson, 1987;Hultsch & Pentz, 1980). Although the results of a number of studies indicate that older adults require more cognitive support than young adults do to optimize episodic memory functioning, it should be noted that there are exceptions to this pattern.
63There is research indicating a parallel improvement in young and old adults as a function of increasing cognitive support (e.g., Mitchell, Hunt, & Schmitt, 1986; Rabinowitz & Craik, 1986), and there are studies showing that young adults may be more effective than older adults in utilizing some forms of support (e.g., Kliegl, Smith, & Baltes, 1989;Simon, 1979). Thus, the relationship between degree of cognitive support, age, and memory performance is c...