This report presents preliminary physical control experiments for capturing and modeling the affective state of entertainment - that is, whether people are having "fun" - of users of the innovative Play-ware playground, an interactive physical playground. The goal is to construct, using representative statistics computed from children's physiological hear rate (HR) signals, an estimator of the degree to which games provided by the playground engage the players. For this purpose children's HR signals, and their expressed preferences of how much "fun" particular game variants are, are obtained from experiments using games implemented on the Playware playground. Neuro-evolution techniques combined with feature set selection methods permit the construction of user models that predict reported entertainment preferences given HR features. These models are expressed as artificial neural networks and are demonstrated and evaluated on two Playware games and the preliminary control task requiring physical activity. Results demonstrate that the proposed preliminary control experiment is not an appropriate control for physical activity effects since it may generate HR dynamics rather easy to separate from game-play HR dynamics, and allows one to
distinguish entertaining game-play from exercise purely on the artificial basis of the kind of physical activity taking place. Conclusions derived from this study constitute the basis for the design of more appropriate control experiments and user models in future studies.