I have my own way of assessing the worth of a book – not just a so-called work of literature but any sort of book or, for that matter, any piece of music or any so-called work of art. I could say in simple terms that I judge the worth of a book according to the length of time during which the book stays in my mind […] (Murnane 2021, 9)
– Gerald Murnane
Like many people of my age and place, much of the nostalgia that I feel is rooted in virtual worlds – in the worlds of the video games of my youth. The foremost of those was probably World of Warcraft, which encompassed forests, mountains, farmlands, jungles, deserts, plains, swamps, highlands, badlands and more. It has been something like a decade and a half since I last played it, but I still occasionally find myself thinking back fondly, not to the painful day-to-day grind and the constant striving, but to the (far more detached and pure) exploring and absorbing of it.
To me, the landscapes in Warcraft were really compelling, by which I mean that they powerfully evoked interest and longing.[1] Why? What gave them their allure? Remembering as always the nebulous nature of questions like these, and making no claim to completeness or certainty, I will nevertheless try to sketch an answer.
Now, you have every right to protest that this sort of thing is entirely subjective. I have written about this problem before. It is a thorny issue, but even if there is no objective ground for these judgments, there are still tendencies in people’s judgments (people generally value A Song of Ice and Fire over The Sword of Truth, say), and these tendencies are sometimes correlated with objective features of the things being valued (for example, the more valued fantasy novels tend to have fewer evil chickens[2]). So I still assert the right (and think it can be fruitful) to speculate about this.
To answer the titular question, I make some shallow research and come up with eight candidate features: naturalness, openness, colourfulness, beauty, coherence, complexity, legibility and mystery. I score 15 video games on each of these as well as “compellingness” (the outcome variable). I then make a (statistically questionable) regression analysis and find that the features most associated with how compelling a role-playing-game-like’s environments are are naturalness, mystery and complexity (variety/diversity). My confidence in these results is low (25% that the ordering is more or less correct for people on average) but as an exploratory study it seems all right.
The environments of World of Warcraft aren’t that different from natural landscapes. They probably activate many of the same synapses. Maybe research on why humans find natural landscapes beautiful can point us in the right direction?
(I’ll limit this discussion to the visible, though sound, music, story and gameplay are likely important too.)
The first thing that springs to mind is probably the old explanation from evolutionary psychology: roughly speaking, our ancestors found certain landscapes more hospitable or habitable, as a result of which it was adaptive for us to find such landscapes pleasant or beautiful. In the literature, this is called habitat theory, and it seems related to what E. O. Wilson called biophilia (Wilson 1984).
The biophilia theory of natural appreciation posits this. We humans are still evolving, but the vast majority of our evolutionary history took place in nature. As a result, the brain has evolved to reward certain adaptive behaviour, and sometimes this behaviour is associated with certain environments (Ulrich 1993). For instance, it might have been adaptive for us to enjoy being in places that offer ample food, drink and safety (Ulrich 1993). More specifically, maybe we like plant life because it offered food (both in the form of edible plants and grazing animals), lakes and rivers because they provided us with drinking water and open landscapes and high vantage points because they allowed us to spot dangers and avoid getting lost. (This is similar to people’s intense fear and dislike of snakes and spiders, which must have evolved due to their poisonous nature; Ulrich 1993.)
Biophilia seems plausible enough to me, but Roger Ulrich, one of its main proponents, talks about how the African savanna was so good for us, and how it was the scene for much of our evolutionary history, and how we therefore enjoy its “parklike properties such as spatial openness, scattered trees […] and relatively uniform grassy surfaces” (Ulrich 1993). He compares it to inhospitable environments like deserts and rain forests, but personally I find deserts and rain forests quite beautiful, and I’m not too enamoured with savannas. That said, I have never set foot in a desert and I’ve never seen a real rain forest with my own eyes, and Ulrich mentions several studies indicating that people generally do prefer savannas.
Far from all environments in World of Warcraft were benign. Many of them seemed quite dangerous. Steep cliff faces, treacherous swamps, thick snow, inhospitable deserts … But there is another vein of research that is not dissimilar from biophilia. Apparently – and who would have thought – people prefer landscapes that don’t have any traces of humans in them (Hodgson and Thayer 1980; Ode et al. 2009). This property is called naturalness and it seems really salient to me. The poets have always praised natural beauty, “which gives even the most insensitive people at least a fleeting sense of aesthetic pleasure” (Schopenhauer et al. 2010, 225). There is something about treading untrod ground. This will be my first candidate feature.
My second candidate feature is openness, which was mentioned in relation to biophilia. This expresses the difference between being in a small glade in a forest and standing on a summit looking out over a sea of treetops. There is some evidence that, in eye-tracking experiments, people spend more time exploring open landscapes than enclosed ones (Dupont, Antrop, and Eetvelde 2013). And people, always and everywhere, seek out viewpoints, panoramas and observatories for the sceneries they offer.
My third candidate feature is colourfulness, or vibrancy, or saturation. World of Warcraft is pretty colourful – maybe that makes it more compelling? Though my intuition here is that, while colour is important, colourfulness probably isn’t, it is a possibility.
My fourth candidate feature is plain beauty. I’m including this as a sort of control group for the experiment I will run.
Then there is this whole thing, developed by Rachael and Stephen Kaplan in the 1980s, called information processing theory. The idea here is that “environments that provide increased opportunities for gathering or discovering information allow for improved living conditions including heightened safety” (Dosen and Ostwald 2016). A merit of this theory is that it takes seriously the fact that experiencing a landscape is an activity, where we continuously search, move around and notice new things. It rests on four pillars: