In this systematic review, we synthesized evidence from 66 ambulatory assessment studies examining within-subject associations between physical activity, sedentary behavior, and affective well-being in daily life. By focusing on intensive longitudinal designs across diverse populations and age groups, the review integrates findings from studies that repeatedly assessed behavior and affect under real-world conditions. A central methodological feature is the use of wearable and mobile technologies—such as accelerometer-based activity trackers and smartphone-based ecological momentary assessment (EMA)—which enable continuous or high-frequency measurement of physical behavior alongside in situ affect assessment. This approach provides unique insights into short-term and momentary processes that are not accessible in laboratory or cross-sectional research.

Overall, the evidence suggests that even short bouts of physical activity tend to be associated with more positive affective states. In particular, associations with higher energy (arousal) and more positive valence appear relatively consistent across different samples and measurement approaches. The accompanying figure provides a graphical synthesis of these findings, illustrating the proportions of positive, negative, and non-significant associations across affective dimensions, time windows, and directions of influence.

Importantly, the visual overview highlights the bidirectional nature of these relationships. Associations are observed both when physical behavior predicts subsequent affect and when affect predicts later physical activity or sedentary behavior. This pattern indicates that affective well-being functions both as an outcome and as an antecedent of physical behavior. In this context, wearable technologies are particularly valuable, as they enable temporally fine-grained analyses of these dynamic feedback loops in real-world settings.

At the same time, the review reveals substantial heterogeneity in findings. The strength and direction of associations vary across populations, affective dimensions, time scales, and methodological approaches, reflected in a considerable proportion of non-significant effects. This heterogeneity partly arises from differences in how wearable-derived behavioral metrics and affective measures are operationalized and synchronized. Accordingly, the findings underscore the importance of contextual and methodological factors and highlight the need for more standardized and harmonized ambulatory assessment protocols in future wearable-based research.