Objective: Novel study designs using within-subject methodology and frequent and prospective measurements are required to unravel direction of causality and dynamic processes of behavior over time. The current study examined the effects of physical activity on affective state. A primary and within-study replication sample was derived from twin pairs. Methods: Female twins (n = 504) participated in an experience sampling method study at baseline. Positive and negative affective changes were examined before and following daily life increases in physical activity. Neuroticism was measured at baseline and depressive symptoms were assessed at baseline and at each of four follow-up assessments. Diagnoses, derived by Structured Clinical Interview for Diagnostic and Statistical Manual for Mental Health-IV axis I disorders, (A. P. A., 1994) were obtained at baseline. Results: A significant increase in positive affect (PA) following the moment of increase in physical activity was replicated across both samples up to 180 min after physical activity. There was no effect of physical activity on negative affect (NA). Across the two samples, a history of fulfilling diagnostic criteria for depression at least once moderated the effect of physical activity on PA, in that the effect was lost more rapidly. Conclusions: The study supports a causal effect of physical activity on PA. However, people with past experience of clinical depression may benefit less from the PA-inducing effect of physical activity. These findings have implications for the use of physical exercise in clinical practice.