Memory strategy use among older adults: Predictors and health correlates in the advanced cognitive training for independent and vital elderly (active) study

Memory encoding strategies for older adult participants in the ACTIVE study are the subject of this dissertation. Strategies are cognitive mechanisms, potentially available to conscious manipulation, that enable individuals to organize information from the environment to facilitate storage and retrieval processes. 'this dissertation is divided into four aims. In the first aim, strategies are defined and a framework for strategy use is proposed to explain strategy acquisition, choices, and knowledge updating processes in older adults. In the second aim, psychometric characteristics of strategy clustering scores are examined and longitudinal trajectories among ACTIVE control participants are estimated to reveal demographic correlates associated with strategy clustering scores. Aim three explores the impact of memory training on strategy use. A review of prior memory training research involving older adults informs about relative training effect sizes of strategy use behaviors compared to conventional memory performance measures. ACTIVE data show that memory training is associated with an immediate increase in the level of both semantic and subjective strategies used to remember a categorical list of words, and subjective and serial strategies used to remember a list of unrelated words. This improvement is maintained through five years and strategy use improvements mediate training-related memory change. Analyses of intrusion errors further reveal memory-trained individuals remember categorically related information qualitatively differently than do control participants. Further, the method of loci is an effortful strategy taught during ACTIVE training and data suggest it is used by 25% of memory-trained individuals, a finding explained by this strategy's inherent novelty for older adults. The third aim's methods and findings demonstrate the potential of public health interventions to benefit health of populations while also informing theory about mechanisms of change. The fourth aim empirically identifies distinct patters of trajectories of strategy clustering scores and uses them to predict health-related outcomes using parallel process general growth mixture techniques. Findings suggest individuals who personalize their strategic approach to memory and who use strategies appropriate for categorical and non-categorical contexts show high health-related quality of life after five years of follow-up and reduced risk of a nursing home admission after ten years of follow-up.