While much research has been done on games as engaging strategies for assessment and education, little has been done to address the specific human computer interaction questions relating to the impact of player engagement in game experiences. This case study examines a section (284 adults) of a larger study on play preference of over 700 participants who were given four versions of an assessment game. While the content remained the same, the versions varied by their structure (rewarding players for acting either appropriately or inappropriately) and by play perspective aesthetics (direct or indirect character embodiment). The results indicate that player preference was for the aesthetics of indirect embodiment and the goals of negative behavior. While this is perhaps unsurprising to makers of commercial games, this is contrary to typical education assessment game design. The case study also demonstrates a difference in player character embodiment that should prove useful to those determining the perspective of their player environments.