An individual-differences approach was used to examine the component processes that predict episodic long-term memory performance. A total of 301 participants ages 20-90 received a 7-hr cognitive battery across 3 days. Key constructs hypothesized to affect long-term memory function were assessed, including multiple measures of working memory and perceptual speed. Latent-construct, structural equation modeling was used to examine the relationship of these measures and age to different types of long-term memory tasks. Speed was a key construct for all 3 types of memory tasks, mediating substantial age-related variance; working memory was a fundamental construct for free and cued recall but not spatial memory. The data suggest that both speed and working memory are fundamental to explaining age-related changes in cognitive aging but that the relative contributions of these constructs vary as a function of the type of memory task. It is well documented that some aspects of memory function differ among adult age groups. Older adults show poorer performance on working memory tasks (Light & Anderson, 1985;

This research examined the role of contextual integration in memories of younger and older adults. In 2 experiments, recall of a target picture to a context picture cue was better when sentences were generated that integrated the picture pair and when the picture pairs were already related to each other. Age differences were smallest when sentences were generated for semantically related pairs. Older adults generated the same type sentences as younger adults, although they generated fewer integrations for unrelated pairs. In a 3rd experiment, younger adults could not differentiate between younger-and older-generated sentences from Experiment 1, and the sentences did not differentially affect recall performance. The results are discussed in terms of age differences in self-initiated processing when using context. Episodic memory performance is in part determined by the availability and utility of contextual cues at the time of recall. One earlier hypothesis suggested that older adults might use contextual information less than younger adults, and this would account for their poorer performance on episodic memory tasks (Burke & Light, 1981; Craik & Simon, 1980). Subsequent research findings on age differences in the effects of context on memory, however, failed to provide clear support for such an hypothesis. Some research has shown that the availability of contextual cues equally benefited the memory performance of younger and older adults (e.g., Cherry & Park, 1993; Earles,

Latent construct structural equation modeling of the relations among age, self-rated health, and speed was conducted with two samples, each containing 372 adults between 18 and 87 years of age. The major results were confirmed in both samples. Self-rated health, sensory-motor speed, perceptual speed, and reaction time speed all decreased as age increased, but health only partially mediated the relationship between age and speed. There were direct effects of age on all types of speed in addition to the indirect effects of age on speed through the self-rated health measures, and there were direct effects of age on perceptual speed in addition to the indirect effects through sensory-motor speed.

No abstract

No abstract