One of the assumptions underlying recent physical activity recommendations is that lower doses of activity (i.e. intensity and duration) are more enjoyable for the average person, thus leading to higher involvement and adherence rates. However, the veracity of this hypothesis can be questioned, since little is actually known regarding the association between activity doses and affective responses. The few preliminary attempts at the conceptual delineation of the dose-response relationship, all centred around an 'inverted-U' notion, are reviewed and criticised as lacking empirical foundation. Available meta-analyses, as well as the empirical literature on the role of exercise intensity and duration, are examined. Increased intensity appears to be associated with reduced positivity of affect during and immediately following an exercise bout. Intensity effects appear to be attenuated during recovery. Fitness and training status appear to become significant mediators of the exercise-affect relationship only at high intensities. With intensity being kept constant, different exercise bout durations have not been shown to have a differential impact on pre- to post-exercise affective changes. Recommendations for future research include: (i) a shift from categorical to dimensional conceptualisations and operationalisations of affect; (ii) the examination of psychological theories on the association between activation and affect (e.g. extraversion-introversion, sensation seeking, type A behaviour pattern and related self-evaluative tendencies, reversal theory, optimal stimulation theory, multidimensional activation theory and self-efficacy); (iii) the systematic and theory-based examination of in-task and post-exercise affective responses; (iv) the incorporation of the parameter of fitness and/or activity status in research designs; and (v) the re-evaluation of methods for selecting exercise intensity levels.