Attributions about ostracism under uncertainty

Attributing ostracism is relatively easy when observers either know about the sources’ motives or have witnessed the events that resulted in ostracism of the target (Weiner, 2006). In many studies that have investigated possible reasons for ostracism, participants were either informed about the deviant behavior of the target, or had even observed it themselves (Ditrich & Sassenberg, 2016; Gooley, Zadro, Williams, Svetieva, & Gonsalkorale, 2015; Wesselmann, Wirth et al., 2013). In other situations in which observers know the target and the sources well, they might at least have sufficient information to make an attribution based on repeated observations: if a group of usually friendly and highly inclusive people decides to ostracize a known troublemaker - someone who has previously been ostracized by other groups - it might appear to be more reasonable to attribute ostracism to a punitive motive rather than a malicious one (Kelley & Michela, 1980). However, in many real-life situations, observers simply do not have such precise knowledge. For instance, think about watching an unknown group of children who ostracize another child: You may not have enough information about the group of children to fully understand what is going on, or you might not have the time or motivation to get a complete picture of why ostracism has occurred. In other situations, it might be necessary to immediately come to a moral judgment. Imagine starting a new job and observing that one new colleague is shunned by the others: Whereas you may not wish to affiliate with a group of mean bullies, you might also be happy to stay away from a potentially selfish, unfriendly colleague, just as everyone else seems to be doing. In such situations, observers must rely on heuristics and social stereotypes in order to make a quick moral judgment about observed ostracism episodes. Two examples for such heuristics that we investigated in our research were (a) the social dissimilarity within the observed group as well as (b) stereotypes about the facial appearance of the observed group.

The social dissimilarity rule

When asking participants to report ostracism situations they have witnessed, many participants describe a situation in which the ostracized target was “different” from the rest. For instance, in some of the situations targets were of a different ethnicity, poorer than others, had some kind of cognitive or physical disability, or any other cue that separated them from the rest of the group (Rudert et al., 2018). Consistent with this observation, we suggest that participants use a heuristic called the Social Dissimilarity Rule to decide whether ostracism was motivated by a malicious or by a punitive motive. In particular, if observers see a dissimilar person being ostracized by two similar others, observers attribute ostracism to a malicious motive, namely discrimination of the target by the (similar) sources. This is because individuals tend to dislike discrimination and prejudice that is displayed by others (Crandall, Eshleman, & O’Brien, 2002; Sommers & Norton, 2006), and thus perceive ostracism that is apparently due to such motives as highly unfair. Of course, this rule is less likely applied if the dissimilarity represents a strong norm violation in itself, such as if dissimilarity results from the target being a convicted criminal. In these cases, however, uncertainty is likely low, so that observers may not need to resort to a heuristic such as the social dissimilarity rule.

In contrast, if targets are ostracized by sources who are apparently similar to the target, social dissimilarity is not a plausible explanation for ostracism. In these situations, it appears more likely that the reason must be because of some previous misbehavior of the targets for which they are being punished. This attribution is consistent with studies showing that individuals tend to punish and exclude members from their own group, particularly if these members deviate from group norms (Marques & Paez, 1994; Marques, Yzerbyt, & Leyens, 1988; Wesselmann et al., 2014; Wesselmann, Williams, & Wirth, 2014; Wesselmann, Wirth et al., 2013).

In several studies, we demonstrated evidence for the social dissimilarity rule (Rudert et al., 2018). Participants either read vignettes about an alleged experiment in which the sources refused to work with the target in a subsequent task or read the abovementioned chatlog, in which the sources ignored the target’s contributions. No matter whether the target differed because of his/her nationality, his/her ethnic background, the football team s/he supported or even his/ her hairstyle: when a dissimilarity was apparent, observers attributed ostracism to a malicious motive of the sources and devalued the sources for ostracizing. This extends earlier research showing that targets’ own attributions focused on malicious racial motives when the sources and targets differed in race (Goodwin, Williams, & Carter-Sowell, 2010). In contrast, when the target was similar to one or even both of the sources, participants attributed ostracism to a punitive motive and blamed the target instead. Interestingly, the effect of social dissimilarity was neither diminished by limiting the cognitive capacity of the observers, nor by using a dissimilarity dimension that should be perceived as non-essential, namely, the hairstyle of the observed group. These findings suggest that the social dissimilarity rule represents a frugal cue that can possibly be processed easily and quickly by observers.

Stereotypes about the groups' facial appearance

In some situations, though, participants might not even be aware of the group constellation because they do not know who the sources are or what they look like. Instead, observers may only see the target, or more precisely, the target’s appearance. Think again, for instance, of the colleague you observe sitting alone for lunch. You might not know him personally, nor do you know his team or the people he might usually surround himself with, so your attribution of your observation might strongly depend on how the colleague looks like.

Whereas most people agree that “a book should not be judged by its cover,” studies nevertheless frequently find that individuals’ facial appearance affects how other people evaluate them and their actions, as well as how they react towards them (Ballew & Todorov, 2007; Hassin & Trope, 2000). Thus, it appears plausible that the facial appearance of the target as well as the facial appearance of the sources can influence how people judge an ostracism episode.

When forming impressions of others, two dimensions are of especially high importance: perceptions of warmth and perceptions of competence (Fiske, Cuddy, Glick, & Xu, 2002). Perceptions of warmth are related to whether the evaluated person appears to have benevolent or hostile intentions. Individuals perceived as warm are seen as good-natured, trustworthy, tolerant, friendly, and sincere. In contrast, perceptions of competence are related to whether an evaluated person appears to have the capacity to fulfill his or her intentions. A person perceived as competent is seen as capable, skillful, intelligent, and confident (Cuddy, Fiske, & Glick, 2008).

In three studies, we presented participants with facial portraits of allegedly ostracized persons (Rudert, Reutner, Greifeneder, & Walker, 2017). The portraits were manipulated with a statistical face model (Walker & Vetter, 2016) on the personality dimensions “warmth” and “competence,” so that the depicted persons looked either warm-and-competent, warm-and-incompetent, cold-and-competent, or cold-and-incompetent (see Figure 9.2 for an illustration). Participants’ task was to decide for each portrait within four seconds how acceptable it would be to exclude this person. Results indicated that participants’ moral judgment was influenced by facial appearance. In particular, participants felt that it was least acceptable to exclude a person that appeared to be warm-and-incom-petent, and felt that it was most acceptable to exclude a cold-and-incompetent looking individual. A follow-up study revealed that this effect is mainly driven

low competence

--------->

high competence

Face Database: http://mirellawalker.com/face-database/

Face Modelling : Walker & Vetter (2016)

FIGURE 9.2 Exemplary face manipulated in warmth and competence. Faces were manipulated on both dimensions, resulting in either a low/low, low/high, high/low or high/high combination (Walker & Vetter, 2016).The presented face is used for illustrative purposes only and was not used in the reported studies

by the degree of disgust that is evoked by the respective face. Whereas cold-and-incompetent looking faces evoked stronger feelings of disgust in observers, warm-and-incompetent looking faces evoked comparatively less disgust, and as a result, exclusion of these persons appeared to be especially (in) acceptable.

In addition, we were also interested in whether the facial appearance of the sources would further influence results. In another study, the faces of the sources were therefore manipulated, too. The abovementioned pattern of the target faces replicated best when the sources looked cold-and-incompetent. This finding further strengthens our notion that observers, if they have no additional clue about why ostracism occurred, tend to picture sources of ostracism as cold-and-incompetent and consequently, tend to disapprove of ostracism. Taken together, the results demonstrate that even minimal cues such as individuals’ facial appearance can affect observers’ moral judgment of ostracism episodes.

Misattributions of observed ostracism

Misattributions of ostracism are more likely to occur when there is less diagnostic information that observers have about the situation. If some of the heuristic and stereotypical cues reviewed above are not informative or are misleading, the result may be misattribution and biased judgments, in the sense that observers assume a motive for ostracism that does not correspond to the actual motivation of the sources. For instance, it has been repeatedly shown that facial characteristics lack objective validity, and that including facial information in one’s judgment lowers judgmental accuracy (Hassin & Trope, 2000; Olivola & Todorov, 2010). If people nevertheless use facial cues, two kinds of biases may occur: On the one hand, observers could misattribute ostracism to a punitive motive and assume that the target must have done something wrong, even though this might not be the case, and as a consequence, fail to protect an innocent target. On the other hand, observers might misattribute ostracism to a malicious motive, and as a consequence, support a target who has violated group norms, while at the same time blaming the sources who merely wanted to protect themselves or uphold social norms. Both constellations may prove particularly problematic in the online world, when observing cyberbullying on social networks or ostracism situations in the media (e.g., reality TV, reports about current political affairs). Since the audience often cannot obtain further information, or might not be motivated to do so, the result might be an unjustified public shaming of either the target or the sources.

A similar problem arises if there is minimal social consensus about the respective norms of the situation. For instance, sources might ostracize a target out of a punitive motive, but an observer does not share the respective norm and therefore perceives the punitive intent as unjustified. Imagine a person who witnesses a smoking target being ostracized by non-smoking sources. Depending on how much the observer endorses non-smoking norms, s/he might either perceive ostracism as justified because the smoker was hurting the health of the sources, or as an intolerant and mean act from the sources. In the latter case, observers might realize that the sources are punishing the target for a norm violation, but because the observers do not endorse the respective norm, they might not agree with the target being punished.

Finally, misattributing ostracism as role-consistent can also cause severe problems: In some situations, individuals are ostracized unintentionally by others, because the sources feel that the target is not important enough to warrant their attention, or because the sources simply forget to include the target (oblivious ostracism; Williams, 1997). For example, a new member of a group might not be invited to a meeting because s/he is not yet in the respective mailing list, or she was not present when the meeting was arranged and others forgot to inform her. These situations can be highly hurtful for targets, and the situation may become worse if observers notice the situation and derive from the others’ behavior that exclusion of the target is consistent with the norms or roles of the situation. In the example above, other group members may assume that there is a reason why the person was not invited to the meeting and, as a consequence, not invite her either.