Conflict simulation, peacebuilding, and development

psychological characteristics and simulation role selection

Last year, Michael King—a PhD student in the Department of Psychology at McGill University—undertook a series of surveys of students involved in my Brynania civil war simulation. While the main purpose of his work was to look at the psychology of violence, he also found a number of interesting things related to the sorts of roles that participants select (most roles are self-selected in my simulation, rather than assigned), and the psychological profiles of those players. He’s kindly written up a short summary of some of his findings, which I’ve posted below.

* * *

Are people born violent, or are they bred? Is it the person, or the situation? These are the questions that guided our research during Professor Brynen’s Brynania peacebuilding simulation, which took place from March 31st to April 7th 2008.

Two weeks before the simulation started, 78 undergraduate students who were to take part in the simulation were asked to complete a questionnaire containing measures of their attitudes and beliefs concerning justice, fairness, and violence. This data was collected to explore (1) if these psychological measures could predict role selection in the simulation, and (2) if these psychological measures could predict which participants would engage in violence during the simulation.

In the questionnaires, participants completed six psychological scales. First was the Arnett inventory of sensation seeking. Second was the global belief in a just world scale, which measures the extent to which an individual believes that the world is a fair place where people get what they deserve. Third was the social dominance orientation scale, which is a measure of an individual’s preference for hierarchy within any given social system. Fourth was the system justification scale, which measures general beliefs about the legitimacy of the overarching social order. Fifth is the Velicer attitudes toward violence scale, which assesses the favorableness of individuals’ evaluations of violence. Sixth was the moral disengagement scale, which determines how easily people can relinquish moral self-control in order to perpetrate cruel behaviors.

Figure 1. Standardized scores of psychological scales for insurgents and non-insurgents.

Figure 1. Standardized scores of psychological scales for insurgents and non-insurgents.

Counter to my own expectations, none of the psychological measure predicted participants’ involvement in violence. In other words, participants who attacked during the simulation scored no differently on the psychological scales than participants who did not attack. These results suggest that pre-existing attitudes concerning justice and violence might have less of an impact on violent behavior than do situational factors. Of course, more research is needed to verify this claim.  

On the other hand, the psychological measures did predict participants’ choice of simulation role. Compared to other roles, people who chose to be insurgents scored significantly higher on the social dominance orientation scale, and on the attitudes towards violence scale (see the Figure 1 for standardized scores on all scales). Here, results suggest that people self-select for the roles they undertake. Although additional research must be conducted, knowing what type of person is drawn to what type of role might have pedagogical implications for simulations (and potential security implications in the real world!).

Our future research will involve adding the same psychological questionnaire after the simulation. With pre-simulation and post-simulation measures, we might be able to calculate changes in attitudes and beliefs produced by experiencing the simulation.

If anyone would like to discuss having psychological measures included in their simulation, I’d be glad to help!

Michael King

michael.king (at)

4 responses to “psychological characteristics and simulation role selection

  1. Michael King 15/06/2010 at 1:30 pm

    Yes, the difference in moral disengagement is non-significant.

    People who chose to be insurgents were higher on SDO and had more favorable attitudes towards violence. Those two traits were the only real differences. Although the other bars on the graph might indicate higher or lower, they were not statistically different. So, although “insurgents” seemed to also differ on sensation seeking and moral disengagement, they did not, these were only chance variations. I hope that’s somewhat clear.

  2. Lukas Neville 14/06/2010 at 6:56 pm

    Is the difference in moral disengagement n.s.?

    And am I reading the chart right that self-selected ‘insurgents’ have a laxer attitude toward the use of violence, but yet are *less* likely to relinquish moral control?

  3. Michael King 14/06/2010 at 5:54 pm

    Thanks… But rather than support Zimbardo’s results, these findings contradict them. In Zimbardo’s classic prison experiment demonstrated the power of the situation: normal men can become brutally violent when placed in the right situation. Personality traits were thought to play a negligible role in explaining the violent behavior of the guards, as participants were randomly assigned their roles. Indeed, as each person had an equal chance of being assigned either role, the differences that occurred between groups was attributed to the roles, and not to the particularities of the participants.

    This classic experiment has been revisited, however, to review if personality traits may have been discounted to easily. In a follow-up study, Carnahan and McFarland (2007) recruited people using the same advertisement as the one used for Zimbardo’s prison experiment. In addition to the original version, another version that omitted “prison life” as the purpose of the study was used. Important personality differences were found between the types of people who respond to each ad. Compared to other respondents, people who answered the “prison life” ad scored higher on psychological scales of aggression, narcissism, and… (wait for it) … social dominance orientation! Same as here!

  4. Timothy O'Higgins 27/05/2010 at 1:56 pm

    Interesting study Michael. Seems consistent with The Lucifer Effect and Zimbardo’s conclusions about situational versus dispositional factors in violence.

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 )

Google photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s

%d bloggers like this: