Discontinuity Effect

Discontinuity Effect Definition

The interindividual-intergroup discontinuity effect is the tendency in some settings for relations between groups to be more competitive, or less cooperative, than relations between individuals. Why is this effect referred to as a discontinuity rather than just as a difference? Unpublished research has demonstrated that variation in the number of people in an interacting pair from one-on-one to two-on-two to three-on-three to four-on-four, and so on, has found a large difference between one-on-one and two-on-two, a smaller difference between two-on-two and three-on-three, and little change thereafter; that is, there is a discontinuity between one-on one relations and two-on-two (intergroup) relations. Research has documented the discontinuity effect in both nonlaboratory and laboratory contexts.

Nonlaboratory Evidence for Discontinuity Effect

Discontinuity EffectThe nonlaboratory research has had participants record on small diaries instances of back-and-forth social interaction that fell into one of five categories: (1) one-on-one (participant interacting with another individual), (2) within-group (participant within a group interacting with other group members), (3) one-on-group (participant interacting with a group), (4) group-on-one (participant in a group interacting with an individual), (5) group-on-group (participant within a group interacting with another group). After classifying the social interaction, the participants then evaluated the interaction as cooperative or competitive. Data collected over a number of days indicated that interactions of types 1 and 2 were less competitive, or more cooperative, than interactions of types 3, 4, and 5. More specifically, there was a discontinuity effect, a difference between interactions of type 1 (one-on-one) and type 5 (group-on-group).

Laboratory Evidence for Discontinuity Effect

Most of the laboratory research has structured the interaction with the use of a matrix game referred to as the Prisoner’s Dilemma Game (PDG). Within the PDG each of two players, A and B, has two choices, X and Y, yielding a total of four possible choice pairings. Typically a choice is made in isolation without knowledge of the other player’s choice. Each choice pairing has a different combination of payoffs or outcomes. These payoffs or outcomes can be illustrated with U.S. dollars. If both A and B choose X, they both receive a moderate payoff, say $3.00. If on the other hand, one player chooses Y, while the other player chooses X, the player choosing Y, whether A or B, may receive $4.00 and the other player may receive only $1.00. Finally, if both players choose Y, they may both receive $2.00. The 2 X 2 matrix of four choice pairings thus presents a dilemma. Either A or B can increase outcomes by choosing Y, but if both A and B are guided by self-interest, they will receive lower outcomes than could have been obtained by mutual X-choices. The X-choice is a cooperative choice, and the Y-choice is sometimes referred to as a competitive choice and sometimes as a defecting choice. If the Y choice is guided by greed, or an interest in increasing outcomes, the competitive label is appropriate. On the other hand, if the Y-choice is guided by fear, or an interest in minimizing the reduction in outcomes resulting from the other player’s Y choice, the defecting label is appropriate.

Common examples of PDG-like situations relate to being honest versus cheating, over-fishing, and pollution of the air and water. In general the PDG models situations in which individual selfishness can lead to collective detriment. Laboratory research has demonstrated that when individuals communicate prior to each trial, they tend to be fairly cooperative. Sometimes the communication has involved face-to-face meeting, sometimes the exchange of notes, and sometimes talking through an intercom. On the other hand, groups who are required to reach consensus regarding the X or Y choice on each trial generally have been found to be less cooperative, or more competitive. Typically the communication between groups has involved the meeting of group representatives but, as with individuals, sometimes has involved the exchange of notes or talking through an intercom.

Three Questions Regarding the Discontinuity Effect

Three questions have been asked regarding the discontinuity effect. First, what are the mechanisms responsible for the effect? Second, what is the generality of the effect across different situations? Third, what are possible ways of reducing the effect by making groups less competitive? These three questions will be considered in turn.

Possible Mechanisms Producing the Effect

Comparison of intergroup relations with interindividual relations uses interindividual relations as a comparison control to identify the distinctive group mechanisms that may lead to the discontinuity effect. To date, evidence for five different mechanisms has been obtained. Each of these possible mechanisms can be formulated as a hypothesis. First, the schema-based distrust, or fear, hypothesis suggests that there is greater distrust in intergroup than in interindividual interactions because the actual or anticipated interaction with a group activates learned beliefs and expectations that groups are competitive, deceitful, and aggressive. Second, the social-support-for-shared-self-interest, or greed, hypothesis suggests that, unlike separate individuals, group members can obtain active support for a competitive choice. Third, the identifiability hypothesis proposes that the group context provides a shield of anonymity, allowing group members to avoid personal responsibility for a selfish-competitive choice. Fourth, the ingroup-favoring-norm hypothesis suggests that membership in a group implies normative pressure to act so as to benefit the ingroup. Fifth and finally, the altruistic-rationalization hypothesis proposes that group members can rationalize their self-benefiting competitiveness as flowing from a concern for benefiting fellow group members.

Generality of the Effect

Research on the generality question has followed either an atheoretical or a theoretical approach. Research following the atheoretical approach has found, for example, that the discontinuity effect occurs not only in the United States but also in Europe and Japan, and that the effect does not change significantly when the values in the matrix vary from those that have frequently been used to values that are increased by a factor of 10 (e.g., $0.66 vs. $6.60 for the highest possible outcome on 1 of 10 trials). Research following the theoretical approach has looked at the correlation between the outcomes for the two players across the four cells of the 2 X 2 matrix. With the PDG, the correlation is negative but can vary as a mathematical function of the ratio of the difference between column means to the difference between row means for the column player (or the ratio of the difference between row means to the difference between column means for the row player). Research has found that as the correlation becomes more negative (and higher outcomes for one player are increasingly associated with lower outcomes for the other player), intergroup competitiveness and the discontinuity effect increases. This result is consistent with the theoretical assumption that as the correlation becomes more negative, the implication of the ingroup-favoring norm becomes increasingly obvious.

Reduction of intergroup Competitiveness

Finally, research on possible ways of reducing intergroup competitiveness has provided evidence that one possible approach is to encourage group members to think beyond the immediate situation to the long-term consequences of their behavior.


  1. Insko, C. A., Kirchner, J. L., Pinter, B., Efaw, J., & Wildschut, T. (2005). Interindividual-intergroup discontinuity as a function of trust and categorization: The paradox of expected cooperation. Journal of Personality and Social Psychology, 88, 365-385.
  2. Wildschut, T., Pinter, B., Vevea, J. L., Insko, C. A., & Schopler, J. (2003). Beyond the group mind: A quantitative review of the interindividual-intergroup discontinuity effect. Psychological Bulletin, 129, 698-722.