In the following study, we investigate the epistemological perspectives found in Danish Gamedesign education (Game Education), in a pragmatic and qualitatively oriented double-case study. We find that cases have convergent form and content, and unarticulated epistemological perspectives that align with the latest epistemological ideals of game research. This points to further work and development opportunities within the field, which is still largely unexplored - Especially in the Danish context. The author has previously worked with design models for game development and game design, with the aim of developing teaching material and teaching competence within the field in his own practice. In 2015, the author was associated with the quality assurance department of the Danish game study Progressive Media, and during the period 2016-2020 the author has taught, consulted, and supervised especially Danish lower secondary-schools in game design and game development education. This led the author to map the landscape of Danish gaming education in the project spiluddannelse.dk - And through this the author has experienced how diverse, interesting and unexplored the field seems to be. First we introduce the reader to the Danish game education development tradition, with an explanation of the research object: What is game education - How is it defined, where does it come from - and how can it be handled. For processing and defining the research object, domain theory is used (Huvila, et al., 2014), and previous index work (Kirkegaard 2019). State-of-the-art research in game research design epistemology is highlighted, and it is discussed what forms of knowledge are used in game education, in a span between design theory and practice (Chiapello L., 2019) (Khaled et al). Game Studies recent growing reflection in design research, which is oriented towards the gaming industry's practice, is highlighted as an ideal for gaming education (Kultima, 2018): but challenged by both early and later design-pattern game research (Aarseth, Debus, Björk), as well as engineering an alternative form of knowledge (Vincenti). In a summary of best practice in game education development, DIGRA's game education curriculum development is highlighted, as well as the GameHighED EU project and the Danish game education model HAGI (Lodahl). This clarifies the research object complex, and becomes the basis for formulating thematic focal points and refined theses for further study of the specificity of the research object: Including the span of practical iterative processor and theoretical forms of knowledge, epistemological tensions, and the exploration of game educational institutional practice. Based on this, we select two educational institutes for a double case study. These educational institutions meet the criteria for maximum variation within selected parameters of belonging to large game education clusters at the higher education level. Method for research interview in epistemic cognition is informed by (Li, Sandoval et al) and (Brinkmann, Kvale). Research interviews are conducted with two educators from each institution, as well as a coordinator for each education, to create a space for reflection on didactic practice and epistemological cognition, and give us an insight into the institutions' handling of the epistemological tensions. Game education is thus investigated with a focus on the discrepancy between game design as practice and as a scientific-theoretical understanding of the phenomenon. The subjects' statements are processed with thematic analysis (Braun & Clarke), and challenge our assumptions about epistemological tension and didactic practice. Likewise, Chiapello's epistemological anchoring model is used to discuss teachers' epistemological anchoring. The results are condensed into themes and discussed, and potential further developments within the field are clarified. We recommend chapters 4-5 as well as 7-8 for a quick read through of the key points. Furthermore, it may be of interest to the English reader, that the case 2 interviews C and D are recorded and transcribed in English; and can be found in the “bilag og data” chapter 10.