This evaluation of the perceived usability also ties into the anchoring in real-world simulacra, or what Baudrillard (1994) defines as copies without originals. Based on their responses, the judges transfer understanding of artifacts from their own real life experience and project that onto objects found in the game. This is essentially a transfer of character, meaning that the judges use their repertoire of character (Janlert & Stolterman, 1997) to understand concepts found not only in the real world, but also in the virtual world. The group of judges had mixed experience with Minecraft, which can be seen in the ways they interpret the artifacts they encounter in the generated settlements. Judges with Minecraft experience will include this in their evaluation of the settlements, whereas judges without Minecraft experience will instead apply similar knowledge from other games. By extension, this means that the repertoire of character used by each judge is a mixture of notions from the real world, games in general, and Minecraft in particular.

Finally, the thematic coherence within the settlement is critical to the interpretation and believability of the settlement. Settlements are expected to be in coherence with the environment in which they are situated. This is exemplified by the Situated Adaptivity theme, where judges expect settlements built in a desert to have sandstone detailing and for settlements built in or near temperate woods to make extensive use of logs. Similarly, they expect settlements to be internally consistent, with similar design features to houses. This is further indicative of transferal between real world knowledge and in-game knowledge. However, there is no hard and fast algorithm by which this is judged. Instead, judges seem to work on the fuzzy logic of the buildings “making sense” within the given environment. On a similar note, judges expect the settlement to be laid out in ways that adapt to the existing natural features in ways that “make sense.” Examples of this include roads that run in ways that allow traversal of the space, or utility structures such as farms of windmills being places where they would be in the “real world.”

Although these findings are perhaps not revolutionary in and of themselves, they indicate that generated game world components are still subject to similar, if not the same, evaluative criteria as their hand-crafted siblings. As such, this study provides empirical evidence that the user experience of generated content is something that needs to be considered both as a line of research, but also as a concern for anyone building a generator.