Taking a Breath of the Wild: are geoscientists more effective than non-geoscientists in determining whether video game world landscapes are realistic?

Abstract. From the wilderness of Hyrule, the continent of Tamriel, and the geographies of Middle Earth, players of video games are exposed to wondrous, fantastic, but ultimately fake, landscapes. Given the time people may spend in these worlds compared to the time they spend being trained in geoscience, we wondered whether expert geoscientists would differ from non-geoscientists in whether they judge the landscapes in these video games to be “realistic”. Since video games present a great opportunity for tangential learning, it would be a missed opportunity if it turns out that features obviously fake to geoscientists are perceived as plausible by non-geoscientists. To satisfy our curiosity and answer this question, we conducted a survey where we asked people to judge both photos from real landscapes as well as screenshots from the recent The Legend of Zelda: Breath of the Wild video game on how likely they thought the features in the picture were to exist in the real world. Since game world screenshots are easily identified based on their rendered, pixelated nature, we pre-processed all pictures with an artistic “Van Gogh” filter that removed the rendered nature but retained the dominant landscape features. We found that there is a small but significant difference between geoscientists and non-geoscientists, with geoscientists being slightly better at judging which pictures are from the real world versus from the video game world. While significant, the effect is small enough to conclude that fantastical worlds in video games can be used for tangential learning on geoscientific subjects.