A new social experiment has found that conflict within a group makes people more likely to support dominant leaders. Highly dominant individuals, who tend to punish others, are endorsed as leaders when the group faces significant conflict, but not when conflict is low. The study was published in Adaptive Human Behavior and Physiology.
Throughout history, strong and dominant leaders have risen to power in difficult times. For example, in the ancient Roman Republic, emergency situations allowed the appointment of temporary dictators. Wars, occupations and other threats to nations have often led to dominant leaders. Although some of these leaders used coercion and force to achieve compliance, many of them had genuine public support.
Scientists are puzzled as to why people accept “strongman” leaders despite their intimidating and coercive tendencies. Early researchers suggested that the people who would favor dominant leaders may have certain characteristics, such as viewing the world as full of conflict and threat, conservative and right-wing beliefs, hierarchy and competition between groups, and aggressiveness. However, recent studies have shown that the preference for dominant leaders can change depending on the context.
Study author Joey T. Cheng and his colleagues hypothesized that dominance functions when a group is threatened or in conflict as a very powerful source of prestige. In such cases, group members are likely to support dominant leaders. On the other hand, if the group is not threatened the support for dominant leaders will decrease. These researchers organized a study in which they focused in particular on the effects of the threat of conflict and disorder within a group on support for dominant leaders and the prestige that results from their dominance.
To conduct the study, the researchers set up a mobile test site on the grounds of the University of Illinois at Urbana-Champaign. Eight research assistants would contact potential participants on the university grounds and, if they agreed to participate, bring them to the mobile testing site. In this way, the 1,026 participants who completed all aspects of the study between May and December 2017 were recruited.
In their experiment, the researchers combined three different levels of punishment applied with 10 different levels of threat faced by the group (in the experiment) to create 30 different experimental conditions. Each participant was randomly allocated to one condition.
In each of the conditions, the participants were observed through a player playing a game. In the game, the first player had the opportunity to take some tokens for himself from the second player (threat). The second player couldn’t do anything about it. There was also a third player who was presented as the leader of the group. He had the opportunity to punish the first player for taking the tokens from the second player (penalty). The third player would do that by sacrificing some of his own tokens so that the first player would lose five times as much.
The experimental conditions differed in terms of how many tokens the first player decided to take from the second player (the threat level, with 10 different numbers of tokens taken, between 6 and 60) and how high the punishment the third player enacts on the first (three levels: low – no signs were erected, moderate – 30 tokens received, and tough – 60 signals received).
After watching the game, participants were asked to report how much they would support the third player as a leader (7 test items taken from a leader endorsement scale) and how much prestige they perceived the third player to have (a 4-item scale developed by the authors). Researchers also found a measure of the perceived dominance of the third player.
The results showed that the participants felt that the third player was more dominant in situations when he gave harsher penalties (ie, decided to take more tokens from the first player). As the punishments increased, the perceived dominance of the punishment became the mean rate. This pattern applied to each of the 10 threat level groups. In these conditions, the third player was also considered to have the highest prestige.
Looking at the level of support received by the third player as a leader, the results showed that the leader received the strongest support from participants to be the leader when the threat facing the group was the highest (ie, when the first player took the highest number of tokens). In contrast, the weakest leader support occurred when the level of group threat was low (ie when player one did not take enough cues from player two) and the perceived dominance of player three was the highest (ie when player three decided to severely punish player one).
“Our findings show that the extent to which dominant individuals excel as leaders depends on the level of threat and conflict facing the group. Strongest leader support is found at high levels of threat faced by both the group and the candidate [the third player] perceived dominance,” the researchers said.
“Conversely, leader support is weakest when threat is low and candidate is high [the third player] dominance, as well as high threat and low dominance candidates. These patterns indicate how, overall, those who are more exposed to threat and conflict are more likely to support a dominant figure as leader; on the contrary, when there is little threat and conflict, a person who is less dominant is better.”
The study sheds light on an important aspect of the psychological dynamics of groups. However, there are also limitations that must be taken into account. In particular, although the threat conditions exceeded the experimental manipulation check, the threat and the entire situation were essentially artificial and minimal. In addition, the study participants were students. Results in other social and age groups and in more natural settings may not have equivalent results.
The paper, “When Persistence Wins Respect: Dominant People Gain Prestige and Leadership by Facilitating Intragroup Conflict Resolution”, written by Joey T. Cheng, Nathan A. Dhaliwal, and Miranda A. Too.