Home‎ > ‎


We are currently working on a 3-year project on the social psychology of embarrassment funded by the Leverhulme Trust, and awarded to Dr Anja Eller. The research is conducted at the School of Psychology, University of St Andrews.

Our main focus is the impact of the
audience on embarrassment.

Although we all get embarrassed from time to time, some people or groups may elicit more embarrassment than others. We will try to find out who these people are and, in particular, to which groups they belong.

Are we more or less embarrassed when faced by people of our own nationality or by foreigners? Will they feel embarrassed on our behalf or laugh at us when we are in a predicament? Does contact with other groups and their members (e.g. other cultures) change how embarrassed we get if we violate norms that are not ours? And how do people show to their own social group that they are sorry for behaving in an embarrassing way? Are people more willing to help someone from their own group after a predicament?

If you would like to follow our progress, you can find regularly updated information on this site. More detailed information about each study can be found by following the links. 

To help us with our research, you can also participate!

1. The Ingroup-Outgroup Audience Effect in Faux-Pas Situations

In four different studies we have found that an audience of ingroup members (e.g., people of the same nationality) make us feel more embarrassed in faux-pas situations than an audience of outgroup members, particularly when the outgroup is seen as lower in status. The effect is stronger for group members that identify highly with their own group. We found this effect in an interview study and five field experiments, in Scotland, Norway, Mexico and Japan, with different group memberships (nationality, university affiliation), and in within- and between-group designs. We also found the effect with an implicit measure (LDT task) of embarrassment, which indicates that the effect is not only due to participants believing that it is more socially appropriate to be more embarrassed in front of an ingroup audience but instead suggests a more fundamental phenomenon. For more information (see end of page): Eller, Koschate, & Gilson (2009, poster) and Eller, Koschate, & Gilson (2011).

Examples of embarrassing moments from our interviews with St Andrews students:

“I was at a social gathering, 30 [students] I did not know. And then I said something quite inappropriate. […] It was a joke… a joke about incest. Because I thought it was quite funny. […] I stood up and then got ready to tell the joke. And I felt quite optimistic about the outcome at the time. […] I told the joke and then…silence. Some people looked quite uneasy and puzzled, other people looking down as if they hadn’t heard anything and some people groaned…ooh. I sat back down and then I laughed to myself, it was a funny situation, a little bit funny. And then decided not to talk about what just happened. Let it pass."

"A seminar in history school. And I have been invited along by my supervisor. And all the rest of people there are staff and current PhD students. And they were talking about something and at first I didn’t quite get what was going on. It was quite intense stuff, and then I started to get nervous because of that. I started to feel embarrassed that I wasn’t getting the idea. But then, what really embarrassed me was that I worked out what they were talking about and I had some ideas about it myself. I wanted to say something but I couldn’t make myself say something. I can’t say something in front of all these people."

Conference poster: Eller, A., Koschate, M., & Gilson, K. (2009). Embarrassment: The Ingroup-Outgroup Audience Effect in Faux-Pas Situations. Poster presented at the General Conference of the British Psychological Society, Brighton, 1-3 April.
Article: Eller, A., Koschate, M., & Gilson, K. M. (2011). Embarrassment: The ingroup-outgroup audience effect in faux-pas situations. European Journal of Social Psychology (Special Issue on Social Image), 41, 489-500.

2. Intergroup Contact and Embarrassment

Intergroup contact has been shown to reduce prejudice and intergroup anxiety, and to increase empathy toward members of an outgroup. In two studies, we have found that contact with members of a lower-status outgroup increases embarrassment in faux-pas situations. More and higher quality contact with outgroup members increases the importance of the opinion of that outgroup about us, their evaluations become more important for our self-image. This in turn increases embarrassment, an emotion that results from a perceived negative evaluation of the self by an audience that is of importance to us. Thus, intergroup contact can mitigate the ingroup-outgroup audience effect found in previous studies (s. above). For more information (see end of page): Koschate, Eller, & Gilson (2009, talk).

Conference presentation:
Koschate, M., Eller, A., & Gilson, K. (2009). Blushing for the Outgroup: Intergroup Contact as a Predictor of Embarrassment. Paper presented at the Social Section Conference of the British Psychological Society, Sheffield, 15-17 September.

3. Priming the audience

In two studies we have explored the possibility that an audience that is activated in an unrelated situation may produce the same ingroup-outgroup audience effect that we found with  audiences that are present in the embarrassing moment. In contrast to our previous studies, we also looked at situations that are more/less embarrassing to the ingroup than the outgroup. When participants thought about their ingroup in an unrelated context prior to violating an ingroup norm (i.e. the ingroup was primed), they reported more embarrassment than when they had thought about a low-status outgroup. When a high-status outgroup was primed, however, they reported more embarrassment after violating an outgroup norm than with the ingroup in mind. These results are exciting because they show that groups can influence our emotions even when they are not actually present within the situation but just a vague memory! Our studies are also the first to show that a fairly complex social context (i.e., group membership and group status) can be primed. 
For more information (see end of page): Koschate, & Eller (2010, talk).

Conference presentation: Koschate, M., & Eller, A. (2010/2011). Embarrassment depends on who you have in mind. Paper presented at the Social Section Conference of the British Psychological Society, Winchester, 7-9 September 2010. A similar version of the paper was also presented as part of the symposium "Intergroup emotions: New directions and old challenges" at the General Meeting of the European Association of Social Psychology (EASP), Stockholm, 12-15 July 2011.

4. Infrahumanisation effect: Embarrassment

Studies in various countries have found that groups tend to rate so-called secondary emotions (e.g. guilt, shame, reminiscence - emotions that are thought to be unique to humans) as more typical of their own group than of other groups (infrahumanisation effect; e.g., Leyens, Demoulin, Vaes, Gaunt, & Paladino, 2007, for a review). Since embarrassment is thought to be a particularly human emotion, we tested whether group members believe that embarrassment is unique to their own group, and not typically shown by other groups. In a study with Japanese participants, we found exactly such an infrahumanisation effect with regard to embarrassment. Japanese participants rated embarrassment (along with other positive and negative secondary emotions) as more typical of Japanese people than of Chinese or American people. They also rated positive primary emotions (e.g., joy) as more typical of Americans, and negative primary emotions (e.g., anger) as more typical of Chinese people. In a second study, we went one step further, and tested whether embarrassment is not only thought to be more typical of the ingroup but whether it is also more likely to be attributed to fellow ingroup members than outgroup members in faux-pas situations.  Participants were asked to imagine a fellow ingroup member, or a higher- or lower-status outgroup member, respectively, as the protagonist in a number of faux-pas scenarios. As expected, the protagonist was seen as more embarrassed when it was an ingroup member rather than an outgroup member. However, this effect was only found for scenarios in which the protagonist was culpable. In contrast, when the protagonist was not culpable for the faux-pas, ingroup and high-status outgroup members were rated as more amused than low-status outgroup members. Since embarrassment is understood as an emotional expression indicating an apology, onlookers may want to see embarrassment when the protagonist is culpable, and therefore needs to apologize. Embarrassment also suggests an inferior position within the situation. Amusement, in contrast, indicates a superior position - the protagonist is in a position to laugh at the faux-pas rather than care about his impression on others. Thus, high-status groups may prefer to see a fellow group member as amused in a non-culpable faux-pas situation, that is, in a situation in which the protagonist doesn't need to apologize. The effect of status on ratings of embarrassment and amusement in non-culpable faux-pas situations was further investigated in a laboratory study. Female participants of upper-middle class or lower-middle/working class background saw a women of similar or higher social class open a coke bottle that spilled over (a low culpability faux-pas). They were led to believe that this faux-pas happened at the beginning of a video conferencing session in which they were meant to interact with the protagonist. The heart rate of participants was measured during the incident, and participants reported their own feelings and those of the protagonist shortly after the incident. Participants who thought the protagonist to be of a high social class showed a stronger increase in their heart rate immediately after the incident than participants who thought the protagonist was of relatively low social class, irrespective of their own social background. A high-status protagonist was also rated as more amused by fellow high-status participants compared with lower-status participants.

Conference presentation: Koschate, M., Murase, K., Saryiski, N., Noel, P., & Eller, A. (2011). The infrahumanisation effect in embarrassment: Culpability and status matter. Paper accepted for presentation at the Social Section Conference of the British Psychological Society, Cambridge, 6-9 September.
Miriam Koschate-Reis,
30 Jul 2009 03:54
Miriam Koschate-Reis,
13 Sep 2010 10:05
Miriam Koschate-Reis,
2 Nov 2009 03:44