What’s in a name? New research finds video game usernames reveal real-world personality traits and age-related patterns
Psychologists at the University of York have shown that the usernames of online gamers can carry meaningful information about the players behind them. By analysing half a million anonymised records from a hugely popular multiplayer game, researchers found clear links between the words and numbers people choose for their usernames and how they behave inside the game.
The study, conducted by Professor Alex Wade and PhD student Athanasios Kokkinakis as part of the EPSRC-funded Intelligent Games and Game Intelligence (IGGI) project, examined anonymised data from League of Legends, a multiplayer online battle arena (MOBA) played by tens of millions worldwide. The dataset included usernames, in-game behavioural metrics and simple social feedback indicators — specifically the post-match “Honour” and “Report” feedback players can give to one another.
The researchers used these anonymised indicators of social interaction to measure the valence of player-to-player behaviour: that is, whether interactions tended to be positive (cooperative, helpful, respectful) or negative (abusive, disruptive, antisocial). They then compared those measures with two features of usernames: the presence of profanity or other antisocial language and the inclusion of numerical sequences that often reflect age.

Key findings from the analysis include:
- Players whose usernames included profanity or other antisocial expressions were more likely to behave antisocially in-game. Such accounts received more negative social feedback and reports from other players.
- Usernames containing positive or neutral language correlated with more constructive in-game behaviour, including rapid learning, cooperation and leadership tendencies cited by teammates.
- Numerical sequences embedded in usernames often provided a reasonable proxy for age. When numbers appeared in names, they frequently reflected age-related patterns: younger players displayed proportionally more negative behaviour while older players tended to be more positive and cooperative.
Professor Alex Wade commented that video games constitute a rich, population-level source of psychological data: “Video games can provide a wealth of useful population-level information on developmental, cognitive and psychological processes. We found that people who have anti-social names tend to behave in an anti-social way within the game. Younger people behave poorly and older people less so. This data is like a window on individual players’ personalities so we believe that we might be able to use video games as a way of testing people’s personalities.”
Athanasios Kokkinakis added: “We think this is just the tip of the iceberg — these massive datasets offer an unprecedented tool for studying human psychology across the globe.” The authors note that, while the results are promising, such large-scale observational data are best seen as complementary to traditional psychological assessment rather than as a replacement.
The study demonstrates how anonymised behavioural traces from online games can reveal consistent patterns tied to real-world characteristics. Usernames are an unobtrusive signal chosen by players outside the direct competitive context of play, yet they correlate with measurable social outcomes during matches. The authors suggest that carefully anonymised game-derived statistics could be used to investigate developmental trends, personality traits and broad social dynamics across diverse populations.
Potential applications include improved understanding of community dynamics within games, designing better moderation systems, and developing research tools that exploit naturally occurring behavioural data. The researchers also highlight ethical considerations: analysis must respect player privacy, remain transparent, and avoid stigmatizing individuals based on limited behavioural signals.
Funding: The research was funded by the Engineering and Physical Sciences Research Council.
Source: University of York. Image credit: Riot Games.
Abstract (concise)
What’s in a name? Ages and names predict the valence of social interactions in a massive online game
Multi-player online battle arena games create complex social environments that generate large quantities of behavioural data. This study asked whether usernames, chosen by players outside the game, predict in-game social behaviour. Analysing a large anonymised dataset from League of Legends, the team found two username features that correlate with social valence: numerical sequences often indicative of player age, and the presence of strongly antisocial words. Both factors were linked to the positivity or negativity of player-to-player interactions. The findings indicate that anonymised statistics from online games may offer a valuable tool for studying psychological traits across global populations.
Reference: Kokkinakis, A. V., Jeff Lin, Davin Pavlas, and Alex R. Wade. “What’s in a name? Ages and names predict the valence of social interactions in a massive online game.” Computers in Human Behavior. Published online November 12, 2015. DOI: 10.1016/j.chb.2015.09.034.