Horizon CDT Research Highlights

Research Highlights

Encouraging Collaboration Between Strangers in Online Games

  Kyle Arch (2013 cohort)   www.kylear.ch


Developing and selling computer and video games is a major industry. According to a report by the Entertainment Software Association [1] over $22bn was spent on the games industry in the United States in 2014. They also estimate that 42% of Americans regularly play more than 3 hours of games a week, and that 51% of US households contain a dedicated gaming console.

This large industry presents unique teamwork situations. Players from across the world join together with complete strangers with the common goal of winning whatever game they are currently playing. The group that a player is placed in may not even be determined by ability; technical factors, like network latency, are commonly prioritised in matchmaking algorithms to promote a smoother gaming experience. This means that players, may end up trying to collaborate with others in the absence of even a common understanding of the game's underlying rules and mechanics.

Regardless of the means for grouping players, the bottom line is that it is unlikely that a player entering a game with no pre-organised team will know the other players. The only information that one player will have about the other is their identifier, nickname, or handle, any in-game player statistics that might be presented (e.g. some persistent rank or scoring), or perhaps any incidental information the player divulges during their play session. On top of this, there is no formal training for everyone to rely on, and for the duration of a game this almost random group of players would, ideally, work together to whatever their common goal is.

This is set in contrast the the established knowledge and research on the subject of the function of various forms of organised virtual teams. Organisational research has studied "classic" virtual teams (VTs), HCI and CSCW has researched more general Collaborative Virtual Environments (CVEs), and organised groups of players in online games have been examined, but with little research examining the common situation of players that do not know each other in so-called "pickup groups" (PUGs).

Work So Far

The first question that was examined via two similar studies was in what ways does teamwork and collaboration in PUGs differ from that in otherwise organised VTs, whether in an organisational or gaming environment. This was initially explored in the first study (Figure 1). A similar, followup study specifically targetted the decision-making processes of players in World of Warcraft dungeons, and this time used screen capture software to more closely examine the events in-game and interactions between players (Figure 2).

A participant from the first study playing <em>League of Legends</em>

A participant from the first study playing League of Legends

A participant from the second study playing <em>World of Warcraft</em>

A participant from the second study playing World of Warcraft

The major finding of these studies was that players appear to enter games with preconceived strategies or ideas of 'what should be done', but fail to communicate this to the rest of their team. Consequently, a common situation finds players coming up against conflicting or incompatible strategies to their own, which at the least causes inconvenience and at the worst acts as impassable obstacles to the team's success.

As a result, future work will examine ways in which strategic communication can be encouraged between players in PUGs, and whether this has any effect on the enjoyment of players or their perceptions of the game. It would be unwise for any intervention to be made at significant expense to entertainment, after all.

Supervision Team

  • Martin Flintham (Computer Science)
  • Harshada Patel (Human Factors)
  • Michael Brown (Horizon)


  1. The Entertainment Software Association. (2015). 2015 Essential Facts about the Computer and Video Game Industry.

This author is supported by the Horizon Centre for Doctoral Training at the University of Nottingham (RCUK Grant No. EP/G037574/1) and by the RCUK’s Horizon Digital Economy Research Institute (RCUK Grant No. EP/G065802/1).