While they may not have been developed as such, online games actually provide an excellent space for large-scale psychological study. They are highly-controlled, richly saturated with sociological interactions, and can even be home to complex in-game economic systems and measures of aggression like war and theft.
One of the most popular online games to boast such complex social elements is EVE Online, a massive online multiplayer game set in a fictitious universe of space exploration, resource mining and crafting, and warring and treaties. It is the subject of a recent study published in PLOS ONE, whose authors sought to explore how real-world geographical, political and economic environments correlated with in-game behavior.
In the study, online data from user behavior, which is meticulously recorded on game servers, was collected from December 2011 to December 2016.
The authors defined three measures of in-game activity to be applied to each player. First, social activity was defined as the number of users the player tagged as a friend vs. an enemy. Belligerence was defined as the number of times a player instigated vs. defended against an attack. Finally, productivity was measured by combining the actual production of in-game items, and extraction of materials from asteroids and destroyed ships.
Additionally, the authors measured real-world social and economic activity using a variety of published indices, including the Global Peace Index, Consumer Price Index, and Unemployment Rate. This was done for each country in which at least 100 players were present.
The authors then aggregated this information based on country, creating an in-game profile and real-world profile for each of those included in the study.
The results of the study reveal some interesting, sometimes counterintuitive correlations between real-world player environments and in-game behavior.
First, individuals living in countries with more violence (as measured by the Global Peace and Global Terrorism indices) were, on average, less aggressive against other players.
Next, higher unemployment rates and weaker currencies are both associated with more efficient and money-conscious in-game trading behavior. These players also tended to spend more time earning in-game money, rather than purchasing it via their subscription.
Likewise, players from countries with higher inflation tended to buy more than sell in-game. According to the authors, this suggests “that real-world inflation experiences in the country-of-origin are transmitted into in-game inflation expectations.”
While this conclusion seems fairly straightforward, other claims by the authors bear greater scrutiny. For example, regarding the first finding (which relates more in-country violence with less in-game aggression), the authors offer the explanation that individuals in these countries may abstain from aggression against others “as a safety valve for inherent aggression, making the real world a safer place.”
However, this presumes that players in these countries are as inherently aggressive as their country profiles suggest, which may not be the case: it could be that those who play online games do so because they are naturally more pacifistic and are looking for an escape from the violence around them, or that aggression is a less sound tact from an economic perspective, as countries higher in violence also tended to be less affluent.
Finally, it was particularly interesting to see how in-game behavior varied between countries, some of which were very similar (France and the United Kingdom) or highly dissimilar (Canada and the Ukraine).
The paper’s size, robust analytics, and detailed theoretical framework make it an important one as we attempt to understand how real-world experiences are reflected by behavior in virtual environments.
The study, “On the connection between real-world circumstances and online player behaviour: The case of EVE Online”, was authored by Andres M. Belaza, Jan Ryckebusch, Koen Schoors, Luis E. C. Rocha, and Benjamin Vandermarliere.
