What are the cons of groupthink

Decision-making processes

Another cause of incorrect conflict resolution is the misconduct of a group. This misconduct can be evoked, among other things, by the subjective feeling of invulnerability, by the “moral correctness” of the group and by the one-sided, stereotypical perception of the opponent. The first model description of this misconduct was published by Janis (1972, 1982) and has since been known as the Groupthink model.

According to Janis and Mann (1977), so-called hot cognitions. These are triggered by stress caused by the need to find a quick solution to the problem. The hot cognitions reduce the search for alternatives and induce to ignore warning messages and references to problems of the chosen solution. The group's attention is only drawn to the positive aspects of an action and the proposed solutions are only discussed and thought through in fragments. Groupthink is defined by Janis (1972) as follows:

I use the term "groupthink" as a quick and easy way to refer to a mode of thinking that people engage in when they are deeply involved in a cohesive in-group, when the members' strivings for unanimity override their motivation to realistically appraise alternative Courses of action ... Group think refers to a deterioration of mental efficiency, reality testing, and moral judgment that results from in-group pressures.“(Janis, 1972, p. 9).

For Groupthink to occur, three conditions must be met. (1) The group must show a high level of group cohesion, (2) there must be a very serious ("provocative") situation (e.g. a serious situation exists when the group as a whole has been attacked or insulted, or the previous ones Decision norms are not sufficient to successfully cope with a problem). In addition, the behavior of the members of the group must be very similar, as they come from the same social milieu. (3) As a third condition, the will to find a better solution or “the best solution” must be articulated. The self-esteem of the group members at this point is very low, as they fear they can fail and the perception of the whole problem is very difficult, perhaps even impossible. This triggers internal stress. External stress is evoked by outside threats and fears of loss (see Figure 6).

The symptoms of groupthink are manifold: the illusion of being invulnerable, rationalizations, ignoring clear warnings and negative feedback, belief in the moral authority of the group, stereotyping of opposing outgroup members, exerting pressure on those who think differently, self-censorship and installing by "mind-guards" who pay attention to the positive group climate.

The effects of groupthink can be very dramatic. The group now restricts its discussion of problem solving to just a few alternatives. After an action is set in motion, it is followed through to the end. Warnings that the actions are incorrect are ignored. At the same time, the search for better alternatives is abandoned. The group believes in their decisions and believes that after a few failures the "tide will turn for the better". The mishaps and negative reactions are attributed to inaccurate execution of the decision. The decision in and of itself is felt to be the right one and vehemently defended against critics (Janis, 1982).

Janis (1972, 1982, 1989) also gives methods how group think can be avoided. The group leader should encourage members to express their thoughts and criticism aloud rather than suppressing them. Furthermore, he should only bring his own suggestions into the group discussion at the end in order to avoid influencing the members in a certain direction. The group members should also be given the opportunity to talk about group decisions with colleagues who are friends and who are not involved in the group discussion. The reactions of these bystanders should be discussed in the group. Experts should also express their opinion in front of the group in order to obtain the most objective information possible. Not only the advantages, but also the disadvantages of each alternative must be discussed sufficiently. After a decision has been made, the group should meet again to clarify the resulting activities and to reflect on the positive and negative events that have occurred since then. In this second meeting, the rejected alternatives should also be discussed again.












Group cohesion

Structural, organizational weaknesses: insults,

Leadership weakness, lack of norms, homogeneity of group members Provocative situation:

Stress from external threats with little hope of finding a better solution to the problem than the group leader

Temporarily low self-esteem due to mistakes already made, difficulties in solving current problems and moral dilemmas.









Tendency to only consider what is applicable






Illusion of invulnerability, belief in the moral authority of the group, collective rationalization, stereotyping of the outgroup, self-censorship, pressure on deviants, self-appointed "memory policemen"

















One-sided selection of alternatives, advantages and disadvantages are weighted differently, after the decision the unselected alternatives are no longer considered, one-sided selection of information, errors in the creation of contingency information







Low likelihood of positive problem solving

Fig. 6. The groupthink model by Janis and Mann (1977).

Janis (1982) tested his model in several crises in US foreign policy. The Bay of Pigs invasion crisis and the escalation of the Vietnam War served as examples of groupthink, the Marshal's Plan, the invasion of North Korea and the Cuba crisis served as examples of how conflict resolution can be successfully discussed in a group. Tetlock (1979) also analyzed these five crises and was able to replicate Janis' results. Smith (1984) used the hostage-taking in Tehran in 1979 and the unsuccessful attempt at liberation to show that American foreign policy also made a “classic” wrong decision in the late 1970s. Esser and Lindoerfer (1989) conducted a quantitative case study and analyzed the Challenger disaster of 1986 using the Report of the Presidential Commission on the Space Shuttle Challenger Accident.The results impressively show that a lot of mistakes have happened in the last few hours. However, Esser and Lindoerfer note critically: “It is possible that the many instances of decision-making defects would be consistent with any theory of poor decision making. We have no comparison group with which to evaluate this alternative ”(Esser & Lindoerfer, 1989, p. 175).

Flowers (1977) tested the groupthink phenomenon with an experiment on group cohesion and leadership behavior. Students had to work in teams to make a decision. In one test condition the subjects knew each other, while in another they were strangers. Before the attempt, the group leaders were instructed either to assert their own opinion or to bring about a joint decision. The behavior of the guide was decisive for the outcome of the problem-solving. If the subjects could decide together, more solution variants were produced and the underlying information better considered than if the group leader made the decision alone. However, Flower's experiment failed to confirm the critical hypothesis that group cohesion plays an important role in decision-making.

Courtright (1978) conducted another study with students to test the hypothesis that group cohesion is an important factor in the groupthink phenomenon. The subjects had to work in teams on the question "What are the best methods for recruiting students for the university in Iowa?" Half of the subjects were told that the members in their group did not match (low cohesion), the other half was announced that they would find the group members sympathetic. The test groups were further subdivided by telling them in the first test condition that enough time would be available and that they should try to reason as well as possible. Furthermore, they should bring about the best solution. In the second test condition, they were told that time was limited and that they should cooperate in order to come to a solution in time. In a third condition it was only mentioned that there was a time limit, but no “proposal” for cooperation was made. It was found that only in the high cohesion condition with time limit was there a significantly more frequent agreement between the group members than in the other two test conditions. This result confirmed the groupthink theory. However, no further matches could be found.

Leana (1985) was also able to partially demonstrate the groupthink phenomenon. The groups, which were led by group leaders who behaved directly and wanted to get their own opinion, produced fewer alternative solutions than the other groups in which the group leaders behaved in a participatory manner. Contrary to Janis' theory, the members of the non-cohesive group exercised more self-censorship than the cohesive group.

When threatened, information processing in groups decreases and the stress reported by group members increases (Gladstein & Reilly, 1985). However, the group members experience no restriction in control. Gladstein and Reilly argue that the degree of difficulty of the situation and the group norms act as moderator variables in the relationship between threat and experienced limitation of control.

Callaway, Marriott, and Esser (1985) reported that groups made up of dominant people, expressed more disagreement and agreement, had more influence over others, experienced less state anxiety, took longer to make decisions, and found better solutions.

Moorhead and Montanari (1986) pointed out that no one had reviewed the entire groupthink model. "The most comprehensive empirical test to date (Courtright, 1978) included measures of only two of the seven antecedent conditions, none of the symptoms, three of the seven decision making defects, and outcomes measured on a subjective productivity rating instrument"(Moorhead and Montanari, 1986, p. 405). In their study, students had to play a competitive management game in groups. All relationships of the Groupthink model could be confirmed, but the direction was opposite! For example, low group cohesion resulted in greater self-censorship and fewer alternative solutions.

Some studies have shown that the groupthink phenomenon can be reduced to a few variables (Schafer & Crichlow, 1996; Tetlock, Peterson, McGuire, Chang & Feld, 1992). A further analysis of the available literature by Aldag and Riggs-Fuller shows that the groupthink phenomenon can be integrated into a general problem-solving model presented at the end of this chapter.

Aldag and Fuller (1993, p. 538f), who provide an overview of previous research on the groupthink phenomenon, comment very critically on the empirical study designs: "Experimental studies of groupthink have considered only a small portion of the model, often without a cohesive group and in situations inconsistent with Janis’s (1971, 1972, 1982, 1989) antecedents. Furthermore, they have relied exclusively on student samples dealing with hypothetical or simulated decisions, with potential resultant problems for external validity. Military strategists, managers, politicians, or other "real-world" decision makers have never been used. In the laboratory, many real-world group characteristics, including ongoing power relationships and political maneuvers, have been necessarily ignored. Although student samples in laboratory settings may be valuable to address many issues relating to group problem solving, their use to examine groupthink is problematic". Posner-Weber (1987) is equally critical of groupthink research: “In the study of groupthink, each experimenter has tended to start from scratch, either arbitrarily or intuitively deciding what aspects of the groupthink phenomenon will be studied ... Janis's theory is largely based on intuition; he is well informed, but his statements are not clearly confirmed ”(Posner-Weber, 1987, p. 124).

Aldag and Riggs-Fuller (1993) developed a new, more general problem-solving model based on the previous research results (see Fig. 7). In their model, the procedural nature of problem solving plays an important role. "Problem solving may be viewed as a multistage process that includes problem identification, alternative generation, alternative evaluation and choice, decision implementation, and decision control"(Aldag & Riggs-Fuller, 1993, p. 541). The defects of the groupthink phenomenon can be better explained by this model. They occur mainly due to difficulties in the problem identification phase. The mistake that the risks of the chosen alternative are not given enough attention is made in the evaluation phase, the incomplete search for alternatives in the phase of generating alternatives. Another point of their criticism and suggestions for improvement are the imprecise definitions. The problem must be precisely defined, the quality of the alternatives ascertained, the source of the solution, the group rules, the control-dependent variables and the time at which the solution is found must be determined in order to be able to make concrete predictions about group behavior. Furthermore, it must be taken into account that at different points in time (phases) of the conflict resolution model, other basic conditions must be present in order to be able to solve the problem in the group in a really good quality. Aldag and Riggs-Fuller also doubt that high group cohesion always has negative effects. They cite the study by Shaw (1981), who was able to show that high group cohesion leads to increased communication between members, greater satisfaction with group membership, less tension between members and better task performance.





Fig. 7. The general problem-solving model by Aldag and Riggs-Fuller (1993).


The advantage of the Aldag and Riggs-Fuller (1993) model over the Janis (1982) groupthink model is that it is largely based on empirical findings. Nevertheless, it must be noted that it turned out to be very complex. An important point that has received little attention so far is the exact timing of problem solving. The two authors explicitly point out that this is an important research hypothesis of their model. The general problem-solving model contains the groupthink model by Janis (1982), although some hypotheses had to be reworked based on the previous findings. For further investigations it is therefore more suitable than the old one, empirically, e.g. T. unconfirmed groupthink model.