The “feel better” effect of exercise has been well established, but the specific influence of exercise duration on affect has not been systematically studied from a multi-level measurement approach. Such an approach offers the opportunity to assess psychophysiological responses that relate to psychological state. One relevant response is the change in frontal brain processes indexed by anterior EEG asymmetry, which is related to approach-withdrawal orientation and affective state [Davidson, R.J., 1993. Cerebral asymmetry and emotion: conceptual and methodological conundrums. Cogn. Emot. 7, 138; Davidson, R.J., 1998. Anterior electrophysiological asymmetries, emotion, and depression: Conceptual and methodological conundrums. Psychophysiology 35(5), 607–614.]. To examine the relationship between exercise duration and affective response, as well as the role of frontal brain processes in this relationship, female undergraduate students ( N = 16, VO 2 max = 35.93 ml kg − 1 min − 1 , aged 19–23 yrs) were assessed for frontal EEG and self-reported affective responses, as measured by the Profile of Mood States (POMS), following rest and three different durations of exercise (15, 30 and 45 min) performed at a standardized intensity (i.e., just below the ventilatory threshold (VT)). Psychological vigor and frontal EEG asymmetry following exercise of 30 min were elevated compared to that observed following rest and other exercise durations (i.e., 15, 45 min). The results support a dose–response relationship between exercise duration and affect characterized by an inverted-U. Furthermore, the covariance analysis, conducted to assess the role of cortical activation at the homologous sites in the post-exercise affective response, suggests that the enhanced vigor observed after 30 min of exercise results from the reduction of withdrawal-oriented processes rather than the facilitation of approach-oriented processes.