Tag Archives: game design

Competition and cooperation in social impact games

Social impact games are social games with serious goals, usually to stimulate thinking about such weighty topics as public policy, health, education, or organizational policies, roles, and performances. These games, being games, are also still about play and rewarding achievements. But what are the best kinds of rewards in a social impact game? Is the idea of competition for points somewhat antithetical to overall purpose of such games?

Bartle’s personality types

Your feelings about competition and cooperation may depend upon the type of gamer you are. In 1996, Richard A. Bartle, described four player types and their interactions in MUD games (multi-user dungeon games, essentially multi-player games involving virtual worlds). The traits of these player types are illustrated below and described further in the Gamification.org wiki.

Bartle’s 1996 paper led to the development of the Bartle Test of Gamer Psychology, a 30-question test designed to characterize the behavior preferences of gamers.

You can take the test at GamerDNA.  It’s unlikely you’ll be 100% one type.  I turn out to be mainly an Explorer (Explorer: 87%, Socializer: 60%, Achiever: 53%, Killer: 0%), at least given my frame of mind today.

How do Bartle’s personality types play in social impact games?

Tending to be the 0% Killer type, I’ve wondered about the  the issue of competition in social impact games. Are the players most likely to achieve the overarching game goals, the least likely to win in games as they are currently designed?

Player values and traits

I don’t believe the issue has been extensively explored but I came across an interesting paper by Sagiv, Sverdlik & Schwarz (2011) (“Sagiv et al”). Sagiv et al.  looked at two social dilemma games where players had to decide whether to give money to another player or to a group pool or to keep it.  In both games, cooperative behavior was designed to be costly and to conflict with competitive action.

Game 1
In Game 1, Player A  had to decide whether to give money to Player B or keep it. If Player A cooperated (giving away money), she ended up with money only if Player B also cooperated (giving away her money). Player contributions would be matched by donations to the players’ favorite social causes. Player A would maximize her personal stash of money if she competed (gave away no money) and Player B cooperated (gave Player A money). In this scenario, Player A’s charity wouldn’t benefit, of course. If players both cooperated (each player giving money to the other player), no one would lose money and both charities would benefit.

Game 2
In Game 2, players were assigned to one of three 40-people groups. Individual players could contribute any amount of money, ranging from zero to five dollars to a group pool. The group with the most money at the end of the game would receive money from all the other group pools. Thus, in Game 2, it was possible to free ride, since an individual could win by not contributing at all, provided the group as a whole engaged in cooperative behavior. Further, if the group lost, the individual suffered no personal loss, since she would not have contributed any money.

Player characteristics
Game players included 46 undergraduate business students enrolled in an introductory psychology course. There were slightly more women than men (22 women; 18 men, 6 did not report). The average age was about 23 years. Participation was anonymous and voluntary, but students were not told the real purpose of the study.

Prior to the games, students filled out questionnaires to assess their values (e.g., the relative importance of achievement, hedonism stimulation, self-direction, universalism, benevolence, tradition, conformity, and security to them) and traits (agreeableness, conscientiousness, neuroticism, extroversion, openness to experience). Values and traits are necessarily interrelated, but they were evaluated as separate defining characteristics of the players in these studies.  In Game 2, some  players were aware of their value scores while other players were not.

In study 1, contributions of money correlated positively with player values reflecting concern and care for others (universalism, benevolence), and negatively with self-enhancement values (power, achievement, and hedonism). There was no correlation between specific traits and contributions.

In Game 2,  as might be expected, more benevolent players contributed more than the power-hungry players. Contribution correlated positively with benevolence and negatively with power values. Awareness of values strengthened the correlation (i.e., if you were aware that you were characterized as benevolent, you’d be more likely to contribute than if you were not).

Study conclusions
As noted by Sagiv et al., “when people interact with others they frequently  have to decide whether to cooperate (often at some cost to themselves) or to compete, possibly at the expense of others.” When confronted with social dilemmas, people may make irrational choices (i.e, non-game winning choices). In studies reported by Sagiv et al., these choices were tied to player values.

A (possible) big picture

Based on these studies, you might expect that Bartle’s socializers and explorers might not necessarily play social impact games to win unless winning is directly tied to game actions that reflect their values.

The research of Sagiv et al. suggests:

  • Helping players to be aware of, and reflect on, their values can enhance desired behaviors
  • Desired behaviors have to be rewarded in ways that appeal to different player types/values for a game to be enjoyable to multiple player types (a challenging task!)
  • If the goal of a social impact game is to encourage “benevolent” social behaviors, status and prestige might not be the appropriate carrots to dangle in front of those who value these behaviors

Of course, the values identified by Sagiv et al., don’t necessarily correlate to Bartle’s personality types and the games described in this research aren’t exactly what we might consider playful games, but these studies make it worthwhile to consider that some of the traditional notions of game-play (e.g., what it means to be a winner) shouldn’t necessarily be applied to social impact games.


Bartle, R. (1996). Hearts, clubs, diamonds, spades: players who suit MUDS.  Journal of MUD Research 1(1). Retrieved from http://www.mud.co.uk/richard/hcds.htm

Sagiv, L., Sverdlik, N., & Schwarz, N. (2011). To compete or to cooperate? Valuesʼ impact on perception and action in social dilemma games. European Journal of Social Psychology, 41(1), 64-77.