The relationships between exercise and affective states: a naturalistic, longitudinal study of recreational runners

Although people generally feel more positive and more energetic in the aftermath of exercise than before, longitudinal research on how exercise relates to within-person fluctuations in affect over the course of everyday life is still relatively limited. One constraint on doing such research is the need to provide participants with accelerometers to objectively record their exercise, and pagers to capture affective reports.

We aimed to develop a methodology for studying affect and exercise using only technology that participants already possess, namely GPS running watches and smartphones. Using this methodology, we aimed to characterize within-individual fluctuations in affective valence and arousal in relation to bouts of exercise, and explore possible moderators of these fluctuations.

We recruited a sample of 38 recreational runners. Participants provided daily affective reports for six weeks using their smartphones. Information on their runs was harvested from their own GPS devices via an online platform for athletes.

Average valence and arousal were higher on days when the person had run than on the next day, and higher the day after a run than on the days after that. Over the course of the day of a run, valence and arousal declined significantly as the time since the run increased. Physically fitter participants had more positive valence overall, and this was particularly true when they had not run recently. There was some evidence of higher-dose (i.e., longer and faster) runs being associated with lower arousal on the next and subsequent days. Gender did not moderate associations between running and valence or arousal.

Our study demonstrated the potential for studying the associations between affect and exercise in a way that is precise, undemanding for participants, and convenient for researchers, using technologies that participants already own and use.