Objective: To examine the association between physical activity levels and cognitive functioning over 1-6 years of follow-up in a sample of cognitively normal healthy middle-aged and older adults. We hypothesized that higher levels of physical activity would be associated with better cognitive performance at baseline and on subsequent follow-up assessments. Background As Alzheimer9s disease research shifts focus from curative treatments to prevention measures, physical activity has become an area of interest. There is emerging evidence suggesting that increased physical activity may protect against cognitive decline, the primary symptom of Alzheimer9s disease. In the existing literature on physical activity and dementia risk, most of the studies are cross sectional, and use measures that assess general cognitive functioning. It was our aim to strengthen the findings of previous cross-sectional studies by using a large sample and an increased number of sensitive cognitive tests, and also to show that these effects can be seen longitudinally. Design/Methods: One-hundred seventy-three middle-aged and elderly (mean age 60.7 +/- 8 years) cognitively normal participants (Clinical Dementia Rating of 0) completed a self-report measure of physical activity derived from the Nurses9 Health Study Exercise Questionnaire and a comprehensive battery of neuropsychological tests, including measures of processing speed, basic attention, executive functioning, and episodic memory. Cross-sectional data were analyzed with hierarchical linear regression modeling, and longitudinal data were analyzed with linear mixed modeling. Results: In our cross sectional analyses, we found that higher physical activity levels were significantly associated with better performance on tests of processing speed, executive functioning, and verbal episodic memory. However, levels of physical activity had no significant associations with cognitive functioning over 1-6 years of follow-up. Conclusions: Our results suggest that physical activity levels do not predict cognitive trajectories in middle aged and older cognitively normal populations. Supported by: National Institute on Aging grants: P50 AG05681, P01 AG03991, and P01 AG026276; the Charles and Joanne Knight Alzheimer Research Initiative of the Washington University Alzheimer9s Disease Research Center, St. Louis, MO. Disclosure: Dr. Pizzie has nothing to disclose. Dr. Hindman has nothing to disclose. Dr. Roe has nothing to disclose. Dr. Head has nothing to disclose. Dr. Grant has received research support from Pfizer Inc. Dr. Morris has received personal compensation for activities with AstraZeneca Pharmaceuticals, Bristol-Myers Squibb Company, Elan Pharmaceuticals, Genentech, Inc., Eli Lilly & Company, Merck & Co., Inc., Novartis, Otsuka, Pfizer Inc, Schering Plough Corporation, and Wyeth. Dr. Morris has received research support from Janssen Alzheimer Immunotherapy Program, Eli Lilly & Company, and Pfizer Inc. Dr. Hassenstab has nothing to disclose.