There are long held beliefs that psychosocial stress contributes to diseases of the heart. One of the key mechanisms through which stress is posited to affect CVD is through its adverse impact on health behaviors. Poor health behaviors, such as sedentary lifestyle, account for up to 40% of the risk of incident CVD. While prior approaches to understanding the impact of stress on health behaviors have been informative, significant gaps remain in our understanding of the pathways linking stress exposure and stress perceptions to CVD-relevant health behaviors. Our ability to assess complex, potentially bi-directional relationships between stress exposure, stress perceptions and health behaviors has been limited by dated theoretical, measurement, and statistical models. New research approaches, and innovative and integrative designs are needed to elucidate how psychosocial stress impacts the bi-directional processes by which stress becomes coupled with, and influenced by, behaviors relevant to cardiovascular health. Recent methodological, technological, and statistical advances now enable us to examine these complex mechanistic pathways. Coupling ecological momentary assessment with smartphone technology and activity tracking, we propose to repeatedly monitor exposure to psychosocial stressors and a range of stress appraisals and perceptions in real time, and in ecologically valid settings such as home and work. This methodology will enable us to ascertain the effects of both momentary stressor exposure and stress perception on whether an individual engages in usual physical activity, and whether engaging in usual physical activity, in turn, influences the subsequent perception of stress. We will conduct a single cohort study with 60 intermittently exercising, otherwise healthy adults, collecting observational data for 6 months. Then, for each person we will build a personalized model of stress perception and usual physical activity, with an additional focus on the sources of perceived stress. We will subsequently provide this information on person- specific associations between stress and usual physical activity to 30 of the 60 participants selected in a randomized manner, and will continue to observe all participants for 6 additional months. This design will allow us to explore the novel hypothesis that personalized, within-subject models of the associational network between stress exposure/ perception and usual physical activity better predict usual physical activity than do our traditional nomothetic, between-subject models. Further, it will tell us if an individual’s knowledge of their personalized within-subject model improves their usual physical activity. We will also explore if moderators, such as environmental and personal resources and vulnerabilities, exist for the association between stress perception/exposure and usual physical activity.