A number of models, methodologies and conceptual frameworks have been proposed on how to design digital educational games aiming at better learning outcomes, while offering a fun experience. Our study combines two different methods, namely Delphi study and electroencephalograms (EEG), in order to provide an integrated digital educational game design model. Firstly, a Delphi survey was conducted in order academic experts in the field of digital educational games give their opinion and propose guidelines for designing digital educational games. Secondly, EEG recordings were used to examine how the game element of feedback for earned and lost points would be best incorporated in the game and how these elements affect players' experience. The results from the Delphi survey showed that the most important features to be included in the games are constructive elements, learning environments suitable for authentic learning, as well as elements which promote fun. Moreover, the experts highlighted that learning objectives should be included in activities in a way that the learners will not realize the actual learning process, but feel that they participate in everyday life experiences and examples. Bloom's taxonomy was also proposed as a way to match the learning elements with the game elements. EEG data combined with empirical data from questionnaires showed that players prefer to have audio feedback when gaining or loosing points during the gameplay rather than visual feedback or no feedback at all and contributed to the development of our digital educational game design guidelines.