Gunslingers-in-arms: three studies on two-player cooperative play in first-person shooters

Some time ago, I congratulated my grad colleague John Velez (Ohio State University) on his recent publication in Cyberpsychology, Behavior and Social Networking (he noted that I was ridiculously on top of publications). A few weeks later, Tobias Greitemeyer the prosocial VG researcher (University of Innsbruck) and his colleagues published a similarly themed article in Computers in Human Behavior. The near-synchronous publications had me kicking to review three articles on cooperative gaming in first person shooters.

Abstract (Ewoldsen et al., 2012; Cyberpsychology, Behavior and Social Networking)

Research on video games has yielded consistent findings that violent video games increase aggression and decrease prosocial behavior. However, these studies typically examined single-player games. Of interest is the effect of cooperative play in a violent video game on subsequent cooperative or competitive behavior. Participants played Halo II (a first-person shooter game) cooperatively or competitively and then completed a modified prisoner’s dilemma task to assess competitive and cooperative behavior. Compared with the competitive play conditions, players in the cooperative condition engaged in more tit-for-tat behaviors—a pattern of behavior that typically precedes cooperative behavior. The social context of game play influenced subsequent behavior more than the content of the game that was played.

Abstract (Greitemeyer et al., 2012; Computers in Human Behavior)

The present research tests the idea that playing a team-player video game in which players work together as teammates and assist each other in achieving a common goal ameliorates the negative effects of violent video game play on cooperative behavior. In fact, two studies revealed that, relative to a single-player mode, playing a cooperative team-player violent video game increased cooperation in a decision dilemma task. Importantly, cooperative behavior generalized across targets in that the decision dilemma was played with a partner who was not the video game play partner. Mediation analyses revealed that cooperative team-play promoted feelings of cohesion, which activated trust norms, which in turn increased cooperative behavior.

Abstract (Schmierbach, 2010; Communication Research)

Although scholars have repeatedly linked video games to aggression, little research has investigated how specific game characteristics might generate such effects. In this study, we consider how game mode—cooperative, competitive, or solo—shapes aggressive cognition. Using experimental data, we find partial support for the idea that cooperative play modes prompt less aggressive cognition. Further analysis of potential mediating variables along with the influence of gender suggests the effect is primarily explained by social learning rather than frustration.

I am aware that J.J. De Simone (University of Wisconsin-Madison) has several co-op gaming articles currently under review.

The studies all examined cooperation and competition, but each had different ideas how cooperative behaviours in gaming are elicited. Ewoldsen et al. investigated reciprocal behaviours strategies as a means for eliciting cooperation, “you help me, I help you”. Greitemeyer and company sought to examine the roles of trust and feelings of cohesion. Schmierbach was more focused on competition than cooperation. He tested for differences in cognition, affect, arousal, gender and throwing two competing theories into a cage. In the one corner is General Aggression Model positing competition rewards violent play whereas on the other side is frustration theory positing that competitive play hinders one’s success, thus frustrating one into aggression.



Ewoldsen et al.: 119 undergraduates from an American university, 96 men and 18 women.

Greitemeyer et al.: 32 undergraduates from a German university, 24 women and 8 men. A second study they conducted consisted of 70 undergraduates from a German university, 46 women and 24 men.

Schmierbach: 102 undergraduates from an American university, 59 women and 43 men.


Ewoldsen et al.:  The 29-item Aggression Questionnaire which assesses for aggressive personality, it is on a 5-point agreement scale.

Greitemeyer et al.: Manipulation checks were used to assess for the videogame’s violent content, cooperativeness, and excitement on a 7-point quantity scale. Study 2 is the same, except they asked the participant about the videogame’s difficulty and liking of it. Participants were also asked how much of a bond they felt towards their partner as a measure of feelings of cohesion. For trust, they asked how much they trusted their partner for the task at hand. The latter two measures used a 5-point quantity scale.

Schmierbach: 6 items that assess for videogame experience using on a 7-point frequency scale. 4 items that assess for violent strategizing using a 7-point agreement scale that would relate to the General Aggression Model. At the opposite side, 2 items that assess for frustration using a 7-point scale.

Outcome variables

Ewoldsen et al.: A game theory dilemma task with 10 trials. Participant gets 4 dimes per trial. If participant gives one or more dimes, the value per dime is doubled for the partner. The same is true if the partner gives one or more dimes to the participants. A participant may end up with 0 to 1.20$ per trial. The tenth trial was not used in the analysis as the participant started hoarding every dime. The tit-for-tat behaviour in this task will be used to quantify cooperative behaviours.

Greitemeyer et al.: A game theory dilemma task with one trial for both studies. Similar to Ewoldsen et al.’s, participants gets four chip each valued at 1 euro for the participant, but is valued at 2 euro if given to their partner. How many chips they gave to their partner is determined as cooperative behaviours. No, the participant does not get anything from their partner nor do they know their partner’s identity.

Schmierbach: a whole bunch of outcome variables. Aggressive cognition is measured by completing six incomplete words. Aggressive affect was measured by asking participants how angry they felt using a 7-point scale. Arousal was also measured by self-report using a 7-point scale. The number of items used was not explicit.

Videogames used

Ewoldsen et al.: Halo 2, using competitive through deathmatch mode, two-player separate campaign mode known as indirect competition, and two player co-op campaign mode. Play time is 15 minutes. A control group did the dime task before playing Halo 2.

Greitemeyer et al.: For study 1, Far Cry’ single-player and two-player co-op mode were used. Tetris was also used as a comparison videogame. For study 2, FlatOut was used for its single-player and co-op modes. Play time is 15 minutes.

Schmierbach: Halo, using single-player campaign mode, two-player co-op campaign mode, competitive through deathmatch mode, the kill score limit was 15 that meant several rounds were played. Play time is 30 minutes.


In all studies, participants were tested in pairs. Ewoldsen et al. had the participants played in separate cubicles and were randomly assigned to one of the three game modes. Greitemeyer et al. had their pair in a room together to one of the three game modes. Schmierbach did something further by testing their abilities prior to the actual experiment and later matched two participants with similar abilities. One limitation is that these studies examined dyadic and live cooperation videogame play. This is quite different from the online and anonymous multiplayer experience in typical shooter matches.


Ewoldsen et al.: a one-way ANOVA analysis revealed that players in co-op (M = 5.93$, SD = 4.03$) used more tit-for-tat strategy than the selfish players in deathmatch (M = 9.39$, SD = 5.54$) or those in the two-player campaign (M = 7.68$, SD = 3.51$). The co-op players were also more cooperative than the control (M = 7.77$, SD = 5.33$) as it seems.

Greitemeyer et al.: Their analyses were complicated due to the nature of the experimental procedure (i.e. two strangers playing together). Players in the co-op condition were more generous (M = 2.83 chips, SD = 0.82) than their single-player brethrens (M = 1.70 chips, SD = 0.48) or Tetris players (M = 2.5 chips, SD = 1.18).

For study 2, those in the co-op mode were more generous (M = 2.83 chips, SD = 0.75) than those in the single-player game (M = 2.33 chips, SD = 1.17). The co-op players felt more cohesion with their partner, their trust was greater as well, but it was statistical not significant. Further meditational analyses revealed that feelings of cohesion and trust accounted significantly between co-op play and cooperative behaviours in a game theory task.

Schmierbach: His analyses were also complicated because of the experimental procedure involving testing two participants at the same time. Participants in the competitive condition had the highest aggressive cognition scores (M = 2.81, SD = 0.32) than the co-op participants (M = 1.08, SD = 0.38) and solo participants (M = 2.08, SD = 0.33).

From there, marginal significances cropped up in his analysis. Aggressive affect was marginal significant (p < .10) where participants in single-player were most angry followed by the co-op group whereas those in the competitive deathmatch were the least angry. But then again, these scores could have been due to chance. Arousal was not affect by game mode, although gender was significant and was not discussed any further.

Both frustration and violent strategizing was marginally affected by game mode where participants in the single-player mode were most frustrated followed by co-op and competitive. The competitive deathmatch group reported the greatest violent strategizing whereas those in co-op were the least.  To test the two competing theories, he entered the frustration and violent strategizing as mediating variables with aggressive cognition as the outcome variable. Results revealed that violent strategizing was the significant predictor whereas frustration was not.


The overall findings from three experimental studies indicated that paired in-person cooperative play lead to an increase of prosocial behaviours and a reduction of aggressive thoughts. The different theoretical approaches were quite interesting enough to compare them in a single experimental procedure. Perhaps tit-for-tat, trust, feelings of cohesion would all fit together which might explain there are less thoughts of aggression.

The significance in Schmierbach’s findings are marginal, nonetheless he wrote paragraphs on that topic. I’d like to address his observations that players in the co-op group were more (statistically insignificantly) angry and frustrated than those in the competitive group. It is perhaps that both players in the pair were matched in skills, probably had higher expectations towards the other or perhaps seeing the incompetence of another, but not recognizing one’s own incompetence might explain the anger and frustration differences.

Greitemeyer et al.’s findings for trust and cohesion are quite relevant for online video game groups, both in short-term and long-term. In the short term, the experiment showed higher cooperative behaviours. We could probably see such cooperative behaviours in-game in games like Left 4 Dead, and most recently Journey. Journey is more cooperative and stories I heard of people feeling very grateful and sending thank-you messages towards their partner is indicative of a powerful interpersonal connection between two strangers in a game that strongly encourages cooperation rather than cooperation in a competitive environment. In the long term, trust norms and cohesion are probably sufficient causes for guilds and clans to form where players help each other in-game as well as outside the game via forums or even in-person. At least in the context of this study, I could say that people who play with each other in LAN parties are nice people.

One big limitation from these studies is they weakly apply to online gaming for groups larger than two persons. Starting with the social environment surrounding certain videogame franchises, a non-cooperative trash talking game play is not exactly conducive for team play. Matthew Eastin’s earlier work of large group play (see Eastin, 2007 and Eastin & Griffiths, 2009) found that larger groups subsequently increases aggression, namely verbal aggression, which in turn lead to increased hostility. Strangely, this seems to predict what goes on in Call of Duty, DOTA, among others and from the stories I read in reddit. Although, I might want to attribute to group size as a sufficient cause of verbal aggressiveness, other causes are identified as well, such as machismo, sexism, social norms of competition that would have been inappropriate in other social contexts, team killings, trolling, bad sportsmanship. Of course, these have yet been empirically verified.

Other gaming environments were more welcoming than others. When Team Fortress 2 went free-to-play, one story showed how female gamers were comparatively treated more equally than in other environments. I could probably say the same for the treatment of people of different ages.  Left 4 Dead was successful in eliciting cooperative play through its gameplay direction. In contrast, the Call of Duty franchise emphasize a different approach where they overly empower one person’s ability to play (i.e. one-man army mentality) that made cooperative play unnecessary. Cooperative behaviours do exist in Call of Duty, but to what extent and compared to which other FPS?

Lastly, a common gaming session in FPS is that players play with strangers. Online anonymity, diffusion of responsibility by group size, skill expectations, person-oriented vs. group-oriented play are among others that would make prosocial and cooperative behaviours less likely and coupled with the culture associated with videogame franchise via discourses in online forums and such. I speculate a vicious spiralling cycle of aggressive behaviours in these games as a way of why these kinds of behaviours persist until now. Jenova Chen’s Journey exemplified the potential of collaborative play and weakens the popular notion of competition as a core element. see article about Jenova Chen elaborating further on collaborative play.

Ewoldsen, D. R., Eno, C. A., Okdie, B. M., Velez, J. A., Guadagno, R. E., & DeCoster, J. (2012). Effect of playing violent video games cooperatively or competitively on subsequent cooperative behavior. Cyberpsychology, Behavior, and Social Networking, (pp. 120410090140008+). DOI:10.1089/cyber.2011.0308

Greitemeyer, T., Traut-Mattausch, E., & Osswald, S. (2012). How to ameliorate negative effects of violent video games on cooperation: Play it cooperatively in a team. Computers in Human Behavior, 28 (4), 1465-1470. DOI:/10.1016/j.chb.2012.03.009
Schmierbach, M. (2010). “killing spree”: Exploring the connection between competitive game play and aggressive cognition. Communication Research, 37 (2), 256-274. DOI:10.1177/0093650209356394

11 thoughts on “Gunslingers-in-arms: three studies on two-player cooperative play in first-person shooters

  1. Funnily enough, none of this is actual science. Social “sciences” are not science and I’m not even going to comment on “cyberpsychology.” It’s all statistics based on very small groups of no statistical relevance, with that all of it is basically worthless and irrelevant, and only really a tool for people with useless degrees to try and justify their existence with pseudo-science.

  2. Pingback: Competition, Cooperation, and Play « The Psychology of Video Games

  3. Pingback: Starsheep Interactive Corporation » Forget violence: Do co-op games make us less aggressive?

  4. Pingback: Forget violence: Do co-op games make us less aggressive?

  5. Pingback: Co-op Gameplay May Make People Less Aggressive from Side Mission Gaming News.

  6. Pingback: Co-op Gameplay May Make People Less Aggressive | Gaming R.S.S.

  7. Shane_Finton à dit:

    J’aime beaucoup le site VGResearcher ( ), qui comme son nom l’indique, recense et analyse les études scientifiques sur les jeux vidéo dans tous leurs aspects. Il est à la fois très exhaustif et honnête. Et une étude en particulier, parue cette année, a retenu mon attention : celle sur les effets de la coopération dans les jeux de tir : … -shooters/

    Merci Beaucoup! c`est un plaisir!

  8. Pingback: Forget violence: Do co-op games make us less aggressive?

  9. Pingback: 2012 year in review for videogames research | VG Researcher

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s