This study was conducted in order to test whether group members attribute more embarrassment (a complex, human emotion) to a fellow ingroup member than a higher- or lower-status outgroup member who is the protagonist in a faux-pas situation. So far, most studies on the so-called infrahumanisation effect have examined whether ingroup members rate emotions indicating human uniqueness as more typical of the ingroup than outgroups. Embarrassment is seen as an emotional expression indicating an apology. Research also indicates that others who are seen to be embarrassed in a faux-pas situation are perceived as more friendly and mature. A different attribution of embarrassment to ingroup and outgroup members in a faux-pas would thus imply that outgroup members are seen as less apologetic, and consequently less mature and friendly than ingroup members.
We also tested whether culpability of the protagonist would affect results. To an audience, particularly an ingroup audience, appeasement (e.g., in the form of embarrassment) should be particularly desirable when the protagonist is seen as culpable.
Design: 3 (ingroup versus high-status outgroup versus low-status outgroup; between-subjects) x 2 (high versus low culpability; within-subjects)
Participants: Overall N = 75; n = 15 participants were excluded because ten participants indicated at the end of the study that they had doubts about the manipulation, four participants indicated contact with the fictitious outgroup, and one participant's answers deviated significantly from the sample (i.e., multivariate outlier)
Results are based on N = 60 students from the University of St Andrews (77% women, age: 18-36 years, M = 21, SD = 3; nationality: 19 different countries, 46% British, 26% other-European; 20% US/Canadian, 8% other; major: 17 different subjects, 32% Psychology, 13% International relations; 55% other)
Participants per condition: 18 ingroup, 21 high-status outgroup, 21 low-status outgroup (no difference between conditions in age, sex, language, university identification, bias: multivariate F(10,104) = 1.66, p = .099)
Ingroup/Outgroups: In the ingroup condition, participants were asked to imagine a student from the University of St Andrews, in both outgroup conditions, participants were asked to imagine a student from University College Edinburgh. A fictitious university, University College Edinburgh, was used instead of a real rival university in order to exclude differences in contact and personal experiences with the outgroup. We chose another Scottish university in order to exclude explanations based on nationality rather than university affiliation. Although naming the fictitious university as University College Edinburgh led to a higher risk of participants knowing that it did not exist due to the proximity of Edinburgh to St Andrews, it also created the advantage that participants may believe the following status manipulation since Edinburgh has a high-status reputation in Scotland.
Outgroup status manipulation: Status of the outgroup relative to the ingroup was manipulated by a fictitious table that ostensibly showed a section of the European Ranking of universities according to a number of criteria commonly used in such rankings (student satisfaction, research quality, entry requirement, employment, teaching quality). In the ingroup condition, participants were told that "The University of St Andrews is famous for its world-leading research and student's satisfaction", with the table showing St Andrews ranked higher at #16 than University College Edinburgh which ranked #20. Similarly, participants in the high-status outgroup condition were told that "University College Edinburgh is famous for its world-leading research and student's satisfaction", with the table showing University College Edinburgh ranked higher at #16 than St Andrews which ranked #20. In the low-status outgroup condition, participants were told that "University College Edinburgh is famous for its community integration and students' satisfaction", with UCE ranked lower at #20 than St Andrews which ranked #16.
Scenarios: 10 faux-pas scenarios (5 high culpability, 5 low culpability) were selected from 20 scenarios after a pretest (N = 26), and were presented to participants in random order.
Example low culpability scenario: "Imagine you are watching [a student from the University of St Andrews/University College Edinburgh] slip and fall on a patch of ice in a public place, dropping a package of groceries in the process." (Culpability: M=1.93, SD=1.17)
Example high culpability scenario: "Imagine you are watching [a student from the University of St Andrews/University College Edinburgh] walk past you. Engrossed in text messaging, the person walks right into a lamp post." (Culpability: M=5.93, SD=1.71)
Measures: After each scenario, participants were asked to rate how the person in the scenario would feel with regard to the following four emotions: amused (positive primary emotion, alpha = .68), angry (negative primary emotion, alpha = .70), hopeful (positive secondary emotion, alpha = .88), and embarrassed (negative secondary emotion, alpha =83). As control variables, participants were also asked to indicate their level of identification with students from the University of St Andrews (alpha = .91), the amount of contact with St Andrews and UCE students, the extent to which they perceive a common identity with St Andrews/UCE students, and their general evaluation of both St Andrews (alpha = .82) and UCE students (alpha = .85).
Manipulation check: Relative status of St Andrews compared with UCE was assessed with two items (r=.60, p<.001).
Manipulation check: As intended, participants in the high-status outgroup condition (M = 4.14, SD = 0.59) gave University College Edinburgh a better rating than participants in the low-status outgroup (M = 2.43, SD = .60) or ingroup condition (M = 2.33, SD = 0.91) on 5-point response scales (F(2,59) = 42.71, p < .001).
High culpability faux-pas situations: An ingroup protagonist (M=4.28, SD=0.49) committing high culpability faux-pas was seen as more embarrassed compared with a high-status outgroup protagonist (M=3.70, SD=0.94, t=2.35, p<.05), with a low-status outgroup protagonist (M=3.89, SD=0.67) in-between (F(2,59)=3.09, p=0.05; see Figure 1 a). No differences between conditions emerged for amusement (F(2,59)=1.01, p=0.372), anger (F(2,59)=0.57, p=0.569), or hope (F(2,59)=0.62, p=0.542) with regard to high culpability faux-pas situations.
Low culpability faux-pas situations: No differences emerged for embarrassment (F(2,59)=0.33, p>0.05) with regard to low culpability faux-pas situations. However, an ingroup protagonist (M=1.85, SD=0.58, t=2.27, p<.05) and high-status outgroup protagonist (M=1.92, SD=0.42, t=3.20, p<.01) committing low culpability faux-pas were seen as significantly more amused than a low-status outgroup protagonist (M=1.44, SD=0.54, (F(2,59)=5.19, p<0.01; see Figure 1b). No differences between conditions emerged for anger (F(2,59)=0.22, p=0.802) or hope (F(2,59)=0.49, p=0.615) with regard to low culpability faux-pas situations.
Figure 1a. Embarrassment in high culpability situations by condition Figure 1b. Amusement in low culpability situations by condition
Embarrassment was attributed more readily to an ingroup member than a (high-status) outgroup member, but only in situations where the protagonist was seen as culpable for the mishap. This finding is in line with research on the infrahumanisation effect, suggesting that more complex, human emotions, such as embarrassment, are seen as more typical for the ingroup than outgroups (see also Study 1: Infrahumanisation in Japan). More interestingly, in this study, participants imagined emotional reactions of ingroup and outgroup members, respectively, in specific faux-pas situations, rather than rate overall typicality of an emotion for a group. Since faux-pas situations are part of everyday life, this finding suggests that ingroup members already expect different levels of embarrassment from a culpable protagonist, depending on whether that person is an ingroup or outgroup member. Since ingroup members also tend to recognise emotions more accurately in ingroup than outgroup members, these two effects may combine in a way that creates a disadvantage for outgroup members. Consequently, outgroup members may be seen as less apologetic, and thus less mature and friendly, in situations were appeasement is particularly important, that is, in situations where the protagonist is seen as culpable for violating a norm. Such biased judgements may then have implications for future intergroup interactions.
Surprisingly, the expected infrahumanisation effect was not found in faux-pas situations where the protagonist was seen as non-culpable. In such situations, the only differences between groups were found in how amused the protagonist was imagined to be. Low-status outgroup protagonists were seen as less amused in a low culpability faux-pas than ingroup and high-status outgroup members. Since the ingroup is also a relatively high-status group, it seems that status was more relevant for judgements of amusement than ingroup/outgroup membership. In contrast to embarrassment, which implies an inferior position within a situation, amusement reflects a high status position because the protagonist is not seen as being overly concerned with his social impression. Since the protagonist was not culpable for the mishap, high-status protagonists were imagined to be less concerned, and therefore more likely to be amused, about the mishap than low-status protagonists. This effect was further investigated in a laboratory study.