Countering embarrassment-avoidance by taking an observer's perspective
The fear of embarrassment can have harmful effects in many important consumer domains (e.g. health and financial), especially for high public self-consciousness (PUBSC) consumers. This research examines how adopting the perspective of an observer interacts with trait PUBSC to influence embarrassment-avoidance. Study 1 demonstrates that individuals high in PUBSC (vs. not) are more likely to take an actor’s perspective and to feel personal distress when viewing an ad with an embarrassment appeal. Studies 2–3 show that seeing oneself as an observer is a helpful strategy for combatting embarrassment-avoidance for high PUBSC individuals. This process is effortful and requires cognitive resources. Together, Studies 1–3 demonstrate the power of our theory to explain, predict, and modify embarrassment-avoidance among individuals most likely to anticipate and avoid embarrassment.
KeywordsEmbarrassment-avoidance Empathy Perspective-taking Personal distress Public self-consciousness
As consumers, we often forgo opportunities to help ourselves, and others, in order to avoid embarrassment (e.g. Helweg-Larsen and Collins 1994). Our fear of embarrassment prevents us from admitting we do not know how a product, such as a mortgage or birth control, works. It prevents us from asking advice about what we should do, for example, about our mounting mortgage bills and unplanned pregnancies. In many cases, if we are to help ourselves, and others, we must overcome our fear of embarrassment in social situations (Foss and Crenshaw 1978).
This overwhelming tendency to avoid embarrassment is especially true for people who are high in public self-consciousness (PUBSC). PUBSC is defined as the tendency to be aware of oneself as a social object (Fenigstein et al. 1975). High public self-consciousness (HPUBSC) corresponds to heightened feelings of being the focus of attention (Fenigstein 1984), i.e. being in the social “spotlight”, and of paranoia (von Gemmingen et al. 2003). HPUBSC consumers appear to be more concerned than others are with how they are regarded by others (e.g., Froming et al. 1990; Miller and Cox 1982), and are more prone to embarrassment-avoidance (Lau-Gesk and Drolet 2008).
How can we help individuals, especially high PUBSC individuals, to counter their tendency toward embarrassment-avoidance? We suspect one of the reasons that high PUBSC are particularly prone to embarrassment avoidance is that they may be more likely to take an actor’s (versus an observer’s) perspective in an embarrassing situation. They tend to perceive themselves to be in the social “spotlight” and focus too much on the situation (Fenigstein 1984). If we can get high PUBSC individuals to focus less on the actor’s perspective and take an observer’s perspective, we may be able to help them counter embarrassment avoidance. The goal of the present research is to propose and test a potential strategy for reducing embarrassment-avoidance, taking an observer’s perspective.
The present research builds on three streams of past literature. First, the work of Epley, Gilovich, and Savitsky suggests when focusing too much on the actor’s perspective, people fail to take into account observers’ empathy (Epley et al. 2002). This research reveals that observers tend to make kinder judgments than actors expect when observers notice actors’ potentially embarrassing blunders (Epley et al. 2002; Savitsky et al. 2001).1 This bias in judgment was termed “empathy-neglect.” Focusing too much on the actor’s perspective, people may underestimate the degree to which others have experienced similarly embarrassing predicaments and thus others’ ability to empathize with them.
Second, we draw on literature on the effects of rumination on negative self-thoughts. This research suggests that “stepping out of the self” is an effective strategy for interrupting an undue focus on oneself, which past studies has shown, intensifies negative emotions including depression and anxiety (e.g., Nolen-Hoeksema 1991). On a similar theme, taking an observer’s perspective may be an effective strategy to avoid too much focus on the embarrassing situation. Taking the perspective of an observer shifts the actor’s attention away from the self and to those factors that spur observers to have empathy, tolerance, and kindness.
Third, we draw on the literature on the personality variable public self-consciousness (PUBSC), which is correlated with both the anticipation of embarrassment and the tendency toward embarrassment-avoidance (Edelmann 1985). Based on this literature, we propose that high PUBSC individuals are more likely to engage in embarrassment-avoidance because of their increased tendency to picture themselves in embarrassing situations and feel personally distressed.
The main goal of our research, then, is to investigate a potential strategy for reducing embarrassment-avoidance, taking an observer’s perspective. We propose that the key to reducing embarrassment-avoidance is to induce actors (or those who imagine themselves to be actors) in a situation to see the situation from the perspective of observing the embarrassing situation rather than experiencing it.
We test our proposed strategy in three studies, focusing on perhaps the most challenging consumer segment: individuals who exhibit a high level of chronic public self-consciousness (HPUBSC).2 Our results demonstrate the effectiveness of our strategy for reducing embarrassment-avoidance (and not merely judgments) for high PUBSC individuals. Specifically, we examine constructs proposed to underlie embarrassment-avoidance, and demonstrate the need for sufficient cognitive resources to enable an actor to adopt an observer’s perspective. In so doing, we extend the literature on empathy-neglect, the effects of taking an observer’s perspective, rumination on negative emotions, and public self-consciousness (PUBSC).
Researchers define embarrassment as a commonly-occurring, short-lived negative emotional response that arises from a threat to the public self in the presence of an audience, real or imagined (Miller and Leary 1992). A person experiences embarrassment when he or she publicly violates norms, displays out-of-role behaviors, or is incompetent, in turn triggering unwanted and unflattering evaluations that threaten the person’s public identity (Edelmann 1987; Modigliani 1968; Parrott and Smith 1991; Schlenker 1980). Much past research has focused on the triggers of embarrassment (Keltner and Buswell 1997), how it relates to other self-conscious emotions such as shame (Miller and Tangney 1994), and how it correlates with various individual-difference variables such as cultural background (Singelis and Sharkey 1995).
Other research has shown that people will go to great lengths to avoid embarrassment (Miller 2007) and the accompanying threat to their social image. They may minimize or explain away the embarrassing behavior (Modigliani 1971) or simply avoid potentially-embarrassing situations, for example by purchasing embarrassing products online (vs. in-store) or by buying products aimed at preventing embarrassing situations (e.g. adult diapers) (Bell 2009; Brackett 2004; Lau-Gesk and Drolet 2008; Moore et al. 2006). Accordingly, a strategy for overcoming the fear of embarrassment, and perhaps the emotional distress that accompanies it, would be useful for reducing embarrassment-avoidant behaviors (e.g., not using a condom, hiding sensitive information from a doctor, etc.).
Most relevant to our construction of such a strategy to reduce embarrassment-avoidance is the psychological research on empathy-neglect, which shows that individuals’ fears of embarrassment and social disapproval are often unfounded (Epley et al. 2002).3 Although actors often feel or expect to feel embarrassed, observers often do not notice embarrassing blunders. And, even when observers do notice, they are generally forgiving. For example, studies have shown that people asked to imagine publicly tripping a security alarm, failing a test, arriving at a party without a gift, or being introduced as someone who bed-wets believe observers will judge them more harshly than observers actually do (Savitsky et al. 2001). Actors, focus too much on their perspective, appear not to take observers’ empathy into account; they exhibit empathy-neglect (Epley et al. 2002). This bias appears to be fairly robust and difficult to counter.
Several processes have been proposed to explain why actors overestimate others’ harshness. The most common explanation appears to be that empathy-neglect is due to a failure on the part of the actor to process information from the perspective of others (Epley et al. 2002). Egocentric, people focus on their blunders and do not consider other factors that influence observers’ judgments (c.f., Epley 2014, pp. 111–115; Savitsky et al. 2001). Accordingly, embarrassment-avoidance may result from too much attention being paid to the self for an embarrassing situation. First, actors may underestimate the frequency with which others also experience embarrassment and thus underestimate others’ potential empathy. Second, actors may engage in “naïve cynicism,” the tendency to believe that others will not be as understanding as they themselves would be (Epley et al. 2002; Kruger and Gilovich 1999). What is needed, then, is a way to disrupt actors’ focus on themselves and augment this focus with an observer’s view of the situation.
Past research on rumination also shows that a heavy focus on oneself can intensify negative emotions because it focuses people on episodic information concerning the specific chain of events and emotions experienced (Kross 2009; Kross et al. 2005). Alternatively, less focus on oneself can weaken negative emotions such as distress. For example, people felt less personal distress when asked to “take a few steps back and move away from their experience” versus “relive the situation as if it is happening.” They also had more construing thoughts (e.g. “I understand why the fight happened; it might have been irrational but I understand his motivation now”) and fewer ruminative thoughts (Kross and Ayduk 2008). Research also suggests that self-distancing buffers people against blood pressure reactivity (Ayduk and Kross 2008) and future negative affect (Ayduk and Kross 2008). In other words, an actor can benefit by taking another perspective, i.e., one of an observer to the situation.
We propose that, similar to the self-regulation of negative emotions, taking an observer’s perspective in an embarrassing situation will make a consumer step out of the situation, focus on the bigger picture (possibly kinder judgments), and reduce embarrassment-avoidance. Taking an observer’s perspective is an implementable and credible intervention. An observer’s perspective can be primed through advertising, interactions with sales people, cues in the environment, et cetera. An observer’s perspective may reduce self-focus and in turn personal distress arising from an embarrassing situation, and thus reduce embarrassment-avoidance.4
Moderating role of public self-consciousness
Not all actors are equivalent in terms of their expectation of 1987 experiencing embarrassment. In particular, a person’s tendency toward public self-consciousness is associated with expectations of being embarrassed and embarrassment-avoidance. PUBSC is defined as the tendency to be aware of oneself as a social object (Fenigstein et al. 1975). High public self-consciousness (HPUBSC) corresponds to heightened feelings of being the focus of attention (Fenigstein 1984), i.e. being in the social “spotlight”, and of paranoia (von Gemmingen et al. 2003). HPUBSC individuals appear to be more concerned than others are with how they are regarded by others (e.g., Miller and Cox 1982), and thus more susceptible to experiencing personal distress due to an embarrassing situation.
Past research confirms that HPUBSC consumers are more prone to embarrassment-avoidance and so less likely to engage in potentially embarrassing behaviors. For example, Lau-Gesk and Drolet (2008) found that ads for a product aimed at preventing embarrassing situations (e.g., gas prevention product) incurred higher purchase intentions of the product among individuals higher in PUBSC. Further, a study by Froming et al. (1990) confirmed that HPUBSC individuals exhibit embarrassment-avoidance in part due to their greater sensitivity to the negative evaluations of others. Specifically, participants were paid to sing “The Star-Spangled Banner” in front of (1) a friend, (2) a stranger with whom they did not expect to interact with in the future, or (3) a stranger with whom they did expect to interact with in the future. The longer participants sang, the more money they earned. Regardless of audience type, participants with higher levels in PUBSC sang for a shorter time compared to participants with lower levels of PUBSC who curtailed their singing only when the audience was composed of strangers with whom they expected to interact with again.
To demonstrate that HPUBSC participants tend to focus on themselves as an actor in an embarrassing situation and exhibit heightened susceptibility to feelings of personal distress, we conducted a pilot study that examined the correlation between PUBSC and the fantasy and personal distress subscales of the Interpersonal Reactivity Index (IRI) (Davis 1980, 1983). One hundred-three participants were recruited from Amazon’s Mechanical Turk (Mage = 35; 53% female) to complete an “Opinion Survey” which contained both the Interpersonal Reactivity Index (Davis 1980, 1983), or the “IRI” as well as a measure of PUBSC (Fenigstein et al. 1975). The IRI contains 28 questions (anchors 1 = does not describe me well; 7 = describe me very well) that measure different aspects of empathy, 7 items of which measure fantasizing tendencies (e.g. “When I am reading an interesting story or novel, I imagine how I would feel if the events in the story were happening to me”) and 7 items measure personal distress (e.g., “In emergency situations, I feel apprehensive and ill-at-ease,” “Being in a tense emotional situation scares me,” and “I sometimes feel helpless when I am in the middle of a very emotional situation”). The PUBSC scale consists of seven items that measure the extent to which participants consider themselves as social objects (Fenigstein et al. 1975). Example items for the PUBSC scale include “I usually worry about making a good impression” and “I am concerned about what other people think about me.” To ensure that participants were unaware of the survey’s purpose, several filler tasks were placed between the PUBSC and IRI scales. In support of the above theorizing, we found a significant positive correlation between PUBSC and Fantasy sub-scale (r = .30, p < .0024) and between PUBSC and Personal Distress sub-scale (r = .31, p < .002).
Taking the perspective of an observer
The present research examines how taking an observer’s perspective reduces self-focus and can counter embarrassment-avoidance among individuals who expect and fear embarrassment, namely HPUBSC individuals. It is important to note that taking an observer’s perspective is not to be confused or equated with the more general, traditional concept of perspective-taking, a term that usually refers to any attempt to overcome one’s own perspective by considering another’s potentially-different perspective (Davis et al. 1996; Gilovich et al. 2000; Nickerson 1999). Traditionally, perspective-taking describes the active attempt to understand an actor’s thoughts with respect to a third object. In other words, perspective-taking refers to taking the perspective of the actor, not taking the perspective of the observer. In the present research, taking the perspective of an observer describes the attempt to imagine oneself as an observer to an (embarrassing) event (not the actor in it), which in turn influences the salience of a now-wider set of information, including information that observers are typically kinder in their evaluations of embarrassed others.
We hypothesize that the effort involved in taking the perspective of an observer requires cognitive resources. Indeed, cognitive load can diminish people’s ability to imagine how others perceive the situation (Davis et al. 1996). Similarly, we propose that people need cognitive resources to step out of themselves and take an observer’s role, and thus a lack of cognitive resources would reduce the effect of an observer’s perspective.
The results of the above-mentioned pilot study suggest that PUBSC is positively associated with fantasizing, which implies that HPUBSC (vs. not) are more likely to put themselves in the perspective of an actor. Study 1, in a correlational manner, tests the hypothesis that:
PUBSC is associated with an increased tendency to take the actor’s perspective in an embarrassing situation; PUBSC is associated with a decreased tendency to take an observer’s perspective in an embarrassing situation.
PUBSC is associated with increased embarrassment and personal distress.
Studies 2 and 3 test the effect of taking an observer’s perspective, in both a health and an advertising context respectively, on embarrassment-avoidance, and multiple measures to illuminate the underlying mechanism of these effects. Specifically, we hypothesize that:
PUBSC is associated with increased embarrassment-avoidance (control condition).
Taking the perspective of the observer causes HPUBSC individuals (but not others) to exhibit less embarrassment-avoidance (Observer’s Perspective condition).
The process of taking the perspective of the observer presumably requires attentional resources and effort to redirect one’s own perspective. Individuals can take an observer’s perspective only if sufficient cognitive resources are available to them. Thus, one key goal of Studies 3 is to tease apart less effortful processing vs. more effortful controlled processing presumably associated with countering it by taking the perspective of an observer.
Cognitive load has been shown to disrupt more consciously-controlled processes, causing people to rely on more automatic processes and a narrowed focus of attention on immediate, local information (e.g., Drolet and Luce 2004; Drolet et al. 2009).5 When cognitive resources are limited, HPUBSC individuals may not be able to take an observer’s perspective to counter empathy-neglect, and HPUBSC individuals will seek to avoid embarrassment even more. In short, load blocks the effect of an observer’s perspective on embarrassment-avoidant behavior. Based on this reasoning, we hypothesize:
Cognitive load moderates the effects of observer’s perspective-taking. Under load (vs. no load), HPUBSC individuals who take the role of an observer will exhibit greater embarrassment-avoidance.
Essentially, under load, individuals will revert to their chronic response patterns. This last hypothesis is tested in Study 3.
In summary, we conducted Studies 1–3 in order to better understand how to reduce undue embarrassment-avoidance among HPUBSC individuals since they are more likely to be negatively affected by an embarrassing situation than are other individuals, namely individuals with medium and low levels of PUBSC. Past research has not much examined the drivers of embarrassment-avoidance among individuals who are not high in PUBSC, instead treating individuals who are high in PUBSC as a clinical population. In the General Discussion, we offer several observations about this group we refer to as “LPUBSC individuals,” though it includes individuals with moderate levels of PUBSC as well as individuals with low levels of PUBSC, and several suggestions for future research with respect to this less well-defined group.
Study 1 is a correlational study to demonstrate an empirical foundation of our intervention: HPUBSC individuals tend to take an actor’s perspective and tend to feel more personal distress (control condition) which in turn makes them more prone to embarrassment-avoidance. We introduced participants to a potential embarrassing situation, and we measured actor’s and observer’s perspectives, personal distress, and embarrassment-avoidant tendencies.
Method and stimuli
One hundred and eighty people from an online panel of students, staff, and local residents of a large public West Coast university participated in Study 1 (M age = 31.0, SD = 9.1; 44.3% female; approximately 7.1% Asian or Asian American, 70.0% Caucasian, 22.9% Hispanic, African American, Pacific Islander, and mixed/other race). Nine people failed to answer all of the questions, leaving a final sample of 171 participants.
Participants examined an ad for Beano, a gas-prevention product. The ad portrayed a situation in which a person accidentally farts in a yoga class while doing the downward facing dog position (see “Appendix”). The ad read:
Rip. Accidentally passing gas in front of classmates is one of the most embarrassing experiences. Guaranteed to linger forever. Try Beano to avoid future embarrassment.
After reading the ad, participants answered several questions regarding the perspective they took when they read the ad. Three questions assessed the extent to which participants took the actor’s perspective (“When you read the ad, to what extent did you imagine yourself being the actor who farted in the scene?” “When you read the ad, to what extent did you put yourself in the scene and imagine this happening to you?” “When you read the ad, to what extent did you think of yourself?”, 1 = A little, 7 = A lot). The three questions were highly correlated and were averaged to create an actor’s perspective index (α = .77). Another question captured the extent to which participants took an observer’s perspective (“When you read the ad, to what extent did you imagine yourself outside of the picture”; 1 = A little; 7 = A lot).
Next, participants completed two items that assessed their personal distress when they read the ad (“When you read the ad, to what extent did you feel personal uneasiness?” “When you read the ad, to what extent did you feel personal discomfort?” 1 = A little; 7 = A lot). The two questions were averaged to create a personal distress index (α = .96). Participants reported their level of embarrassment in response to the ad using a 7-point scale (“When I read the ad, I felt embarrassed”; 1 = Not at all agree; 7 = Agree). Then, participants answered two questions measuring their intentions to purchase the gas-prevention products (“Would you consider buying this gas prevention product?”, “Would you want to purchase this product?”, α = .96) using a 7-point scale (1 = Not at all; 7 = Definitely). Last, participants rated their tendency to pass gas in public (1–7 scale; 1 = Never, 7 = Very often). Responses to this question served as a potential covariate.
At the end of the survey session, participants completed the Fenigstein, Scheier, and Buss (1975) public self-consciousness (PUBSC) sub-scale and supplied demographic information. We conducted covariate analyses for age, ethnicity, and gender. However, no significant differences emerged (ts < 1) in Study 1, or in Studies 2–3.
We followed Aiken et al. (1991) method and mean-centered the level of PUBSC scores (M = 4.43, SD = 1.33). We then conducted a regression analysis (SAS PROC SURVEYREG) in order to examine participants’ responses to the embarrassing situation with respect to the centered level of PUBSC. Although we have made directional predictions, we report the (more conservative) two-tail t-tests.
Hypothesis H1a suggests that PUBSC is positively related to the tendency to take the actor’s perspective and negatively related to the tendency to see oneself as an observer in embarrassing situations. We ran a regression with the actor’s perspective as the dependent variable and (continuous) PUBSC as an independent variable. The analysis revealed a main effect of PUBSC (F(1, 169) = 3.90, p = .05). The results became more significant if we included the tendency to pass gas in public as a covariate (F(1, 168) = 4.93, p < .03). We reported the results below without the covariate. HPUBSC individuals were more likely to take an actor’s perspective (β = .19, t(169) = 1.97, p = .05) and less likely to take an observer’s perspective (β = − .39, t(169) = − 2.40, p < .02). These results are consistent with H1a.
Consistent with H1b, a regression with the personal distress index as the dependent variable and PUBSC as an independent variable revealed a main effect of PUBSC (F(1, 169) = 5.34, p < .03) on personal distress such that HPUBSC individuals felt more personal distress when they read the ad of the embarrassing situation (β = .40, t(169) = 2.31, p < .03). We also found a main effect of embarrassment. HPUBSC (vs. LPUBSC) individuals felt more embarrassment when reading the ad (β = .35, t(169) = 1.96, p < .05).
In this specific consumer context, higher intentions to purchase embarrassment-prevention products indicate greater embarrassment-avoidance. A regression with purchase intentions as the dependent variable and PUBSC as the independent variable revealed a main effect of PUBSC (F(1, 169) = 8.16, p < .005). HPUBSC individuals were more likely to purchase embarrassment-prevention products in order to forestall embarrassment (β = .31, t(169) = 2.86, p < .005).
Personal distress as mediator
We tested whether personal distress mediated the effect of PUBSC on embarrassment-avoidance (purchase intentions) using Hayes’ mediation model (2013, Model 4). Preacher and Hayes’s (2008) SAS macro with 10,000 bootstrapped samples revealed indirect-only mediation (Zhao et al. 2010). We ran a regression with purchase intentions as the dependent variable and both PUBSC and personal distress as independent variables. Estimation revealed that, controlling for PUBSC, personal distress was a significant predictor of embarrassment-avoidance (β = .31; t(168) = 4.43, p < .0001); controlling for personal distress, the direct effect of PUBSC on embarrassment-avoidance became marginally-significant (β = .196; t(168) = 1.83, p = .07). The indirect path (β = .12, 95% CI [.04, .23]) had a 95% confidence interval that did not include 0. Therefore, we conclude that personal distress mediated the effect of PUBSC on embarrassment-avoidance. HPUBSC individuals’ tendency toward embarrassment-avoidance appears driven, at least in part, by higher personal distress experienced by HPUBSC individuals in an embarrassing situation.
Study 1, along with the pilot study, provides an empirical foundation for our proposed intervention: HPUBSC individuals are more likely to imagine themselves to be in an embarrassing situation, experience greater personal distress at its prospect, and then are more likely to seek to avoid embarrassment. These results are supportive of H1a and H1b.
Studies 2–3 build on these results by investigating the effects of taking an observer’s perspective on embarrassment-avoidance. Specifically, studies 2–3 test whether a cue to take the perspective of an observer decreases embarrassment-avoidance among HPUBSC individuals. Since HPUBSC individuals’ higher embarrassment-avoidance may be driven partly by greater personal distress or egocentric attention, an observer’s perspective may reduce embarrassment-avoidance by reducing personal distress. Study 3 also provides insights as to the effortful nature of processing information from the standpoint of an observer and provides direct evidence as to process via thought protocols.
Study 2 tests the hypothesis (H2) that a cue to take the perspective of an observer will have a helpful effect on HPUBSC individuals. Again, taking the perspective of an observer should make one’s own perspective as an observer salient. By stepping out of oneself and viewing the embarrassing situation from an observer’s perspective, HPUBSC individuals should realize that others will not judge them as harshly and so HPUBSC individuals should experience less personal distress. In brief, HPUBSC individuals should display less embarrassment-avoidance when they are encouraged to focus on their role as observers rather than actors (H2). Because individuals who are not high in PUBSC are less likely to feel embarrassment in the first place, we hypothesize that taking an observer’s perspective will have no effect on them.
Method and stimuli
One hundred and seven students at a large West Coast university participated in Study 2 for a $20 payment (M age = 19.9, SD = 1.5; 74.0% females). The ad read:
We are looking for volunteers to talk to researchers about how to improve communication lines between doctors and their patients about sensitive healthcare issues. For example, people who should get tested for HIV or need treatment for genital herpes frequently opt not to seek help from doctors, family members and friends. You will be asked several personal questions during the interview. The goal of this research is to identify ways to make it easier for people to talk more openly about such sensitive topics and thereby feel more comfortable seeking treatment. You will be paid $50 in exchange for your time and insights (approximately 1 h).
After participants examined the volunteer ad, they answered one of two versions of the survey corresponding to the ad. The two different versions of the survey asked the same set of questions but in a different order. Past research has shown that different ordering of the same set of questions can differentially affect responses (Sudman et al. 1996).
One version of the survey served as a control condition. Participants in the control condition were first asked about their intentions to volunteer (1–7, not at all/definitely volunteer). Here, per H2a, we expected to find that higher levels of PUBSC correlate with lower intentions to volunteer, a manifestation of embarrassment-avoidance. Participants then indicated how they expected to feel during the interview (1 = not at all, 7 = very comfortable, nervous, embarrassed, confident, calm), how they expected the researchers to react towards them as volunteers (1 = not at all, 7 = very positive, favorable, good; α = .97), and how they would react towards other volunteers in the study (1 = not at all, 7 = very positive, favorable, good; α = .98).
The second version represented the observer’s perspective condition. Participants in this condition first indicated how they would react towards the volunteers if they were the researchers. They then indicated (in order) their intentions to volunteer, how they expected to feel during the interview, and how they expected the researchers to react towards them as volunteers. In sum, participants answered the four questions in a different order.
At the end of the survey session, participants completed Fenigstein et al.’s (1975) public self-consciousness scale (7 items). Participants then answered demographic questions. We suspected that low-income individuals might volunteer for the interview for $50 even if they would feel embarrassed, so we included income as a covariate in the analyses reported for this study.6 Below we report the two-tail t tests with an alpha level of 0.05.
In this study, we used volunteer intentions as our main dependent variable. Higher volunteering intentions to disclose embarrassing information correspond to less embarrassment-avoidance. Note, different from the dependent variables used in studies 1 and 3 (i.e., purchase intentions for embarrassment-prevention products), the dependent variable used in Study 2 (i.e., volunteer intentions) is negatively correlated with the underlying construct of embarrassment-avoidance. We conducted a regression analysis (SAS PROC SURVEYREG), exploring participants’ responses to the embarrassing volunteer opportunity from the mean-centered level of PUBSC (M = 3.76, SD = .54) and question-order condition (control vs. observer perspective). Analysis revealed a highly significant two-way interaction effect of PUBSC and question-order condition (F(1, 102) = 4.23, p < .04). The two main effects were not significant (ps > 0.25). Estimated means for volunteer intentions are provided below using + 1SD and − 1SD to depict the nature of any interactions that emerged. Figure 1 depicts these results using this criterion.
The results of Study 2 support our hypotheses. In the control condition, volunteer intentions were significantly lower among HPUBSC individuals (+ 1 SD) versus LPUBSC7 individuals (− 1 SD) (M LPUBSC = 5.78 vs. M HPUBSC = 3.40; t(102) = − 2.32, p < .03), consistent with H2a. According to H2b, HPUBSC (vs. not HPUBSC) individuals will respond differently to a cue to take the perspective of an observer. According to our theory, because an observer’s perspective can make HPUBSC individuals step out of themselves and reduce personal distress, HPUBSC individuals should display less embarrassment-avoidance (i.e., higher volunteer intentions) when taking an observer’s perspective (vs. control). Because individuals who are not high in PUBSC do not easily feel embarrassed in the first place, a cue to take an observer’s perspective may simply be ineffective at reducing embarrassment-avoidance. Consistent with H2b, an analysis across question–order conditions showed that the observer perspective-taking prime increased HPUBSC participants’ intentions to volunteer (M HPUBSC,observer’s perspective = 5.15 vs. M HPUBSC,control = 3.40; t(102) = 1.97, p = .05), making HPBUSC participants less embarrassment-avoidant. When participants took an observer’s perspective, those with higher levels of PUBSC (+ 1 SD) displayed the same level of embarrassment-avoidance as those with lower levels of PUBSC (− 1 SD) (M HPUBSC = 5.15 vs. M LPUBSC = 4.59; t(102) = .94, p = .35). However, taking the perspective of an observer did not affect LPUBSC participants’ volunteer intentions (MLPUBSC, observer’s perspective = 4.59 vs. M LPUBSC,control = 5.78; t(102) = 1.57, p = .12).8
The results of Study 2 support our theorizing. Taking the perspective of an observer appears to be a helpful strategy to combat undue embarrassment-avoidance among HPUBSC individuals but not among individuals who do not embarrass easily (i.e. individuals not high in PUBSC).
Study 3 builds on the results of studies 1–2 in two important ways. First, Study 3 tests our theorizing in another important consumer context, advertising, which generally follows different social rules compared to volunteering. Second, in order to examine our proposed process in more detail, we collected process measures, such as the perceived empathy participants expect from others and participants’ open-ended responses relating to an egocentric focus. Third, we tested cognitive load as a moderating variable in order to demonstrate the cognitive toll of adopting an observer’s perspective. We expected load to interfere with taking the perspective of an observer and to reverse the effect of an observer’s perspective on embarrassment-avoidance.
Study 3 examines the mechanism of the effect of taking an observer’s perspective on countering embarrassment-avoidance. We introduced cognitive load to moderate the effect of observer’s perspective. We expect cognitive load to interfere the effect of observer’s perspective. We also collected open-ended thinking protocols on self-related thoughts.
Participants and procedure
Study 3 participants were two hundred and twenty undergraduate students at a large public West Coast university (M age = 20.0; 75.7% female). All participants received a survey packet with a cover page instructing them to provide feedback on an ad for a real product targeted towards college-aged people like themselves. We created two ads for a real-world flatulence prevention brand (i.e., Beano), with a photo used by Lau-Gesk and Drolet (2008, Study 2) that depicted four college-age individuals at a party sharing a couch. A male is sitting alone, slouching at one end of a sofa. The side of his head is resting on his hand. His head is turned slightly down. At the other end of the sofa, three females are sitting together. One of the females is looking nervously sidelong at the male. The other two females are in animated conversation. The consumers in this photo displayed nonverbal behaviors that signaled embarrassment (Keltner and Buswell 1997). Participants in the control condition read the following ad copy:
Rip. Accidentally passing gas in front of a crush is one of the most embarrassing experiences. Guaranteed to linger forever.
Participants in the observer’s perspective-taking condition read the same ad copy along with an additional sentence:
Others will know what it’s like. Put yourself in their shoes…would you giggle? Would you be horrified? Would you stare?
To ensure that taking an observer’s perspective could successfully make people think others would not judge them harshly, we conducted a pretest with 164 undergraduate students (M age = 24.0; 52% female). Half of participants received the control-ad copy and the other half of participants received the observer’s-perspective ad copy. All participants were asked to read the ad copy and answer two questions about the empathy they would expect from others if they were to pass gas in public (1–7 scale, 1 = not agree at all; 7 = agree): “If I pass gas aloud, people would put themselves in my shoes”; “If I passed gas aloud, people would empathize with me”. The PUBSC scale was embedded among questions at the end of the survey.
Regression analysis found a highly-significant two-way interaction between ad condition and PUBSC (F(1,160) = 3.98, p < .05). Consistent with past research, HPUBSC participants (+ 1 SD) in the control condition expected less empathy from others (β = − .46, t(160) = − 2.30, p < .02). Taking an observer’s perspective caused HPUBSC participants to expect the same level of empathy as participants not high in PUBSC did (-1 SD) (β = .27, t(160) = .87, p < .39). The cue to take an observer’s perspective led HPUBSC participants to expect more empathy, compared to HPUBSC participants in the control condition (m observer’s perspective = 3.79 vs. M control = 3.14, t(160) = − 2.51, p = .01). However, the cue did not affect the amount of empathy participants not high in PUBSC expected (M observer’s perspective = 3.59 vs. M control = 3.48, t(160) = − .41, p = .48). Therefore, the pretest results suggest taking an observer’s perspective could successfully make HPUBSC participants think others have more empathy, providing a foundation for Study 3.
In the main study (Study 3), approximately half of the participants in each ad condition received a cognitive load manipulation before being asked to provide feedback on the print ad they were shown (Ward and Mann 2000). Participants in the load (vs. no load) condition were asked to study 20 words for 2 min and keep them in mind when they answered questions regarding the ad. At the end of the study, they were asked to write down as many of the 20 words they could remember. All participants were exposed to one of the two ads described above and then answered two questions that assessed purchase intentions. Participants then provided open-ended responses; they were instructed to write down any thoughts or feelings that occurred to them while reading the ad. This task was followed by a manipulation check question assessing whether participants viewed the ad as relatively self-related versus other-related, the feelings they experienced while viewing the ad (PANAS scale; Watson et al. 1988), and potential covariates including their own tendency to pass gas in public and the feelings that arise from such situations. Last, participants completed the PUBSC scale and provided demographic information.
Results and discussion
The manipulation check analysis yielded a significant main effect of ad type on the actor index (“When I read the ad, I imagined that I was the person who accidentally passed gas”; 1 = disagree, 7 = agree). Participants exposed to the ad encouraging them to become an observer scored lower on this index than those exposed to the control ad (M observer’s perspective = 3.83 vs. M control = 4.34, t(212) = 1.96, p = .05).
Two independent coders assessed the elaborateness of participants’ thoughts that served as a manipulation check for the influence of load on thinking processes. Decision makers who are under load rely relatively more on automatic processes and locally-provided information rather than stored information (Ward and Mann 2000). Coders were instructed to rate the thoughts along two items using a 7-point scale (1 = not at all, 7 = very) that measured: (1) the degree to which participants provided thoughts elaborating beyond simply describing the ad based on the picture or copy it used; and (2) the degree to which participants provided thoughts indicative of gut reactions to the ad. The latter item was reverse-scored to create an elaboration index (r = .86). Initial inter-rater reliability was high (r = .85). Inconsistencies between the coders were discussed until agreement was reached. Analysis found only a main effect of load consistent with expectation: load participants tended to elaborate less (M load = 2.35 vs. M no load = 3.42, t(212) = 4.47, p < .0001), indicating the manipulation of load was successful.
Study 3 measures the same dependent variable tested in Study 1: purchase intentions for embarrassment-prevention products. Purchase intentions for products aimed at preventing embarrassment correspond to increased embarrassment-avoidance.
We analyzed the effects of mean-centered PUBSC (M = 3.80, SD = .58), ad condition, and load on participants’ purchase intentions. Regression analysis revealed the predicted three-way interaction among PUBSC, ad type, and load on the purchase intention index (r = .75, F(1, 212) = 11.11, p < .001) and a marginally-significant two-way interaction between PUBSC and ad condition (F(1, 212) = 3.62, p < .06). The two-way interaction emerged between PUBSC and load for the control condition (F(1, 212) = − 2.67, p < .008) and the observer’s perspective condition (F(1, 212) = 2.19, p < .03).
Our hypotheses were again confirmed. In the no load control condition, purchase intentions for the gas prevention product were significantly higher among HPUBSC participants (+ 1 SD) participants than among LPUBSC9 participants (− 1 SD) (M LPUBSC = 2.56 vs. M HPUBSC = 4.45; t(212) = 2.40, p < .015). This suggests that HPUBSC individuals do indeed exhibit more embarrassment-avoidance, supporting H2a and replicating Study 2’s findings.
Further, results for the no load, observer’s perspective condition were consistent with H2b. Under no load, taking the perspective of an observer was effective at reducing HPUBSC participants’ purchase intentions (M HPUBSC,observer’s perspective,no load = 2.01 vs. M HPUBSC,control,no load = 4.45; t(212) = 1.08, p < .006). When HPUBSC participants had sufficient cognitive resources to perspective-take, they displayed less embarrassment-avoidance. See Fig. 2. Because individuals who are not high in PUBSC do not easily feel embarrassed, we expected that the cue to take an observer’s perspective would have no effect on their purchase intentions. Unexpectedly, however, we found taking an observer’s perspective increased lower PUBSC participants’ purchase intentions (MLPUBSC,perspective-taking no load = 4.36 vs. M LPUBSC,control,no load = 2.56; t(212) = − 2.44, p < .002), i.e., increased embarrassment-avoidance. We will discuss this finding in detail in the “General discussion”.
In addition, consistent with H3, load reversed the helpful effects of taking the perspective among HPUBSC participants. Among HPUBSC individuals in the observer’s perspective condition, purchase intentions were marginally-higher in the load condition than in the no load condition (M HPUBSC,observer’s perspective,load = 3.50 vs. M HUPBSC,observer’s perspective,no load = 2.01 t(212) = − 1.89, p < .06).
We also found that load reversed the harmful effects of taking the perspective of an observer among participants with lower levels of PUBSC. Among these individuals, purchase intentions were lower in the load condition than in the no load condition (M LPUBSC,observer’s perspective = 2.96 vs. M LUPBSC,observer’s perspective,no load = 4.36, t(212) = 2.04, p < .04). In fact, purchase intentions among both HPUBSC and LPUBSC participants in the load, observer’s perspective condition did not diff in their intentions in the no load, control (non-intervention) condition (M LPUBSC,observer’s perspective,load = 2.96 vs. M LPUBSC,control,no load = 2.56; t(212) = .64, p < .52, and M HPUBSC,observer’s perspective,load = 3.50 vs. M HPUBSC,control,no load = 4.45; t(212) = − 1.30, p < .19). These results for HPUBSC participants under load are consistent with our hypotheses.
We found lower purchase intentions among HPUBSC participants in the load, control condition vs. the no load, control condition (M HPUBSC,control,no load = 4.45 vs. M HPUBSC,control,load = 2.93; t(212) = 2.32, p < .02). Load appears to have blocked cognitively-demanding processes, namely the self-monitoring HPUBSC people usually engage in to compare their behavior vis-a-vis social standards. Consequently, HPUBSC participants in the control condition exhibited less embarrassment-avoidance under load.
To illuminate the process that more self-focus and personal distress lead to more embarrassment-avoidance among HPUBSC participants in the control condition, and that an observer’s perspective reduces self-focus and personal distress, two independent coders classified participants’ open-ended thoughts into three categories (inter-rater reliability = .85): (1) a “self-related” thinking score, i.e. the extent to which participants related the embarrassing situation to self (vs. others), given the embarrassing situation depicted in the ad (1 = other-related/not related to self; 7 = highly related to self); (2) other ad-related thoughts; and (3) irrelevant thoughts. No significant differences emerged for the latter two categories of thoughts (ps > 0.30).
Most pertinent to our theorizing regarding process are the results for the first category. We found a significant three-way interaction among PUBSC, ad condition, and load on participants’ self-related thoughts (F(1, 212) = 6.89, p < .01). There was also a significant main effect of ad condition (F(1,212) = 7.79, p < .006). Further analysis revealed a significant two-way interaction between PUBSC and load in the control condition (F(1,212) = − 2.90, p = .004) but not in the observer’s perspective condition (F(1,212) = 1.11, p = .27). In the no load, control condition, self-related thoughts were significantly higher among HPUBSC (+ 1 SD) versus LPUBSC (− 1 SD) participants (M HPUBSC,noload = 4.01 vs. M LPUBSC,noload = 1.31; t(212) = 2.09, p < .041). This finding is consistent with the notion that HPUBSC participants are more likely to envision oneself in the embarrassing situation. See Fig. 3.
Consistent with our theorizing, when HPUBSC participants adopted the observer’s perspective and had sufficient cognitive resources, taking an observer’s perspective reduced self-related thinking (M Observer’s perspective,no load = 1.44 vs. M control,no load = 4.01; t(212) = 3.15, p < .002). Taking an observer’s perspective caused no load LPUBSC participants to engage in directionally more, albeit not statistically-significantly more self-related thoughts than they did in the control condition (M observer’s perspective,no load = 2.84 vs. M control,no load = 1.31, t(212) = − 1.59, p < .11). This finding is consistent with the view that individuals who are not high in PUBSC typically do not put themselves in the picture and thus would usually generate fewer self-related thoughts.
Across observer’s perspective conditions, there were no effects of load or PUBSC on self-related thoughts. In comparison to findings with respect to purchase intentions, these findings suggest that, while the content of thoughts can be changed by priming instructions, differences in purchase intentions arise only when ample cognitive resources are available to take an observer’s perspective. The ability to adopt the perspective of an observer is essential. Under load, participants generated fewer self-related thoughts, but embarrassment-avoidance is unaffected by that emotion-as-information.
Consistent with the view that load disrupts cognitive processes associated with HPUBSC, load decreased self-thoughts relative to HPUBSC participants in the no load, control condition (M HPUBSC,control,no load = 4.01 vs. M HPUBSC,control,load = 2.15; t(212) = 2.44, p < .02). In line with the unexpected finding of greater embarrassment-avoidance among LPUBSC individuals in the load, control condition, LPUBSC individuals exhibited more self-related thoughts (M LPUBSC,control,no load = 1.31 vs. M LPUBSC,control,load = 3.21; t(212) = 2.34, p = .02).
In summary, the results of Study 3 support our theorizing about how taking the perspective of an observer reduces embarrassment-avoidance among HPUBSC individuals and how cognitive load moderates these effects; for these individuals, taking an observer’s perspective reduces personal distress in embarrassing situations. Study 3 provides evidence for the proposed process and its consequences.
This research adds to an accumulating body of work on embarrassment and embarrassment-avoidance. Building on past research demonstrating that, in comparison to individuals with medium or low levels of PUBSC, HPUBSC individuals expect more embarrassment and exhibit more embarrassment-avoidant behavior. As a point of departure (Lau-Gesk and Drolet 2008), the present research offers the first comprehensive examination of the relationship between PUBSC and embarrassment-avoidance, and offers an implemental intervention to counter embarrassment-avoidance. As Study 1 demonstrates, HPUBSC individuals tend to imagine themselves as an actor in an embarrassing situation and experience more distress. We demonstrated through mediation analysis that personal distress underlies HPUBSC individuals’ exaggerated embarrassment-avoidance tendency. Further, as shown in studies 2 and 3, HPUBSC individuals are more likely to exhibit embarrassment-avoidance. The present research also introduces an intervention to counter embarrassment-avoidance among consumers. Studies 2 and 3 demonstrate the power of taking the perspective of an observer to an embarrassing situation on the incidence of self-related thoughts, and offer a way to counteract embarrassment-avoidance among HPUBSC individuals.
The present research extends past research on the effects of vicarious embarrassment. Specifically, the literature on vicarious (i.e. empathetic) embarrassment indicates that observers of an embarrassed actor often experience that actor’s emotional distress. They empathize with the actor, and feeling embarrassed they will turn away (Edelmann and McCusker 1986; Miller 1987). This literature implies that if an actor adopts the perspective of an observer, then the actor will still continue to experience embarrassment. However, past research suggests that this vicarious or empathetic embarrassment is qualitatively different from the embarrassment actors experience. For example, Miller (1987) introduced an important distinction when he defined empathetic embarrassment as embarrassment that is felt for another even though one’s own social identity has not been threatened. And, Layton (2016) argues that not only must the observer experience an emotion that closely matches the actor’s (i.e., embarrassment) in order to qualify as an empathetic response but also the observer must realize this emotion is unrelated to his own circumstances. This literature relates to our own theorizing about the lack of personal distress one feels when one adopts the perspective of an observer versus actor. Observers may feel embarrassed but are unlikely to think that their own social standing is at stake. This is a key difference between actors and observers and is related to our hypothesis regarding how observers would evaluate the actor. Our research proposition is not whether observers will experience vicarious embarrassment. Rather, we consider how observers will evaluate an actor regardless of whether they feel embarrassment on behalf of the actor.
It is not entirely clear what a vicarious embarrassment account would predict with respect to observers’ evaluations of embarrassed actors. Past research demonstrates that actors, as opposed to observers, focus on their own behavior and tend to assume the worst in terms of interpersonal evaluations. Regardless of vicariously experienced embarrassment, observers’ judgments or actors are consistently kinder than actors generally expect, perhaps because observers who are not personally threatened they be sympathetic. As Miller (1987) shows, empathetic embarrassment is more likely if observers are susceptible to embarrassment themselves. Such observers tend to view embarrassed actors more kindly. If an actor is able to view the situation as an observer would, the kinder evaluations of the observer may become salient; actors will then realize that observers are generally kinder than they expect. Thus, our main hypothesis is that encouraging an actor to take the viewpoint of the observer will counteract actor’s tendency to anticipate harsher evaluations. In summary, it is important to note that ‘taking the perspective of an observer’ is a fundamentally different tack than ‘trying to predict how observers will evaluate actors’ since actors generally do not predict observers’ evaluations of them very well. Instead, actors need to consider the experience of an observer (vs. their own). Our data is consistent with the hypothesis that when the actor takes an observer’s perspective, he/she distances himself from the situation (e.g., his/her social identity is not threatened) and is able to realize others will not judge him negatively. Future research may want to look into, under what circumstances taking an observer’s perspective can lead to greater embarrassment avoidance instead.
Future research might consider other ways in which individuals’ self-related thoughts can be redirected to dampen HPUBSC individuals’ expectations of overly-harsh observer evaluations. For example, results for the load, control (no intervention) condition imply that load thwarts HPUBSC individuals’ ability to engage in their usual self-monitoring (i.e., fewer self-related thoughts; Study 3) which in turn causes empathy-neglect and embarrassment-avoidance.
Furthermore, future research may want to examine the group of individuals who are low, versus not-high or medium PUBSC individuals, in greater detail. We hypothesized that taking an observer’s perspective would have no effect on what we termed ‘LPUBSC’ individuals because these individuals are less easily embarrassed in the first place. However, we found that taking an observer’s perspective can increase embarrassment-avoidance among LPUBSC (vs. HPUBSC) individuals (directional in Study 2 and significant in Study 3). It may be that lower PUBSC individuals tend not to perceive themselves as being in the spotlight, but that taking an observer’s perspective switches the spotlight on and grows embarrassment-avoidance. This is consistent with Froming et al.’s (1990) finding LPUBSC people behave similarly to HPUBSC people when under a spotlight.
For lower PUBSC participants in the control conditions, the load manipulation appears to have increased self-related thoughts and embarrassment-avoidance. Although a firmer explanation for this unexpected finding awaits further empirical experiment, lower PUBSC participants in Study 3 responded as though the load manipulation instructions caused them to focus more on social actions than they would have otherwise. Alternatively, it may be that lower PUBSC participants tend to engage in self-protective thinking that is blocked by working-memory load. The underlying cognitive processes for individuals who are not high in PUBSC have not been studied previously. Rather, researchers have focused on HPUBSC individuals as a clinical population and not individuals with low or medium levels of PUBSC.
Importantly, load’s effects in Study 3 on self-related thoughts and purchase intentions demonstrate that taking an observer’s perspective is an effortful endeavor. In this way, the present research adds to the growing body of research that shows how cognitive load can exacerbate individuals’ reliance on information stored in memory and their chronic tendencies. Under load, taking the perspective of the observer lead to fewer self-related thoughts, in particular fewer thoughts relating the embarrassing situation to themselves. Future research might benefit from explicitly adopting a dual processing view and exploring the automatic vs. effortful tendencies associated with PUBSC levels in detail (Evans 2008).
Our results have significant implications for marketers given the frequency with which embarrassment-avoidance forms the basis for attempts to motivate consumers to buy a wide variety of products from laundry detergents (“ring around the collar”), to dishwashing liquid (unsightly spots on dishes), and even to cars (avoiding the embarrassment of an unfavorable evaluation by the neighbors). Beyond this, however, our research is relevant to those situations in which marketers want to inoculate consumers against a fear of embarrassment and encourage them to take actions they might otherwise avoid. These situations frequently occur. Examples include getting an embarrassing but potentially life-saving medical test, asking a technician “dumb” questions that will increase customer satisfaction with a purchase, or adopting an innovative and socially-visible but potentially-risky product that might invite public ridicule. Our research shows that devising strategies that will successfully universally reduce embarrassment-avoidance is more complicated than devising strategies to increase embarrassment-avoidance since consumers will react differently to persuasion tactics depending on their level of PUBSC and the amount of available cognitive resources.
A second judgment bias, the ‘spotlight’ effect, refers to actors’ tendency to overestimate the degree to which observers notice their embarrassing blunders; see research by Epley and colleagues. We do not investigate the spotlight effect here. Instead, we focus on the segment of consumers who chronically feel as though they are under observation (i.e., in the spotlight).
We use the LPUBSC designation as a shortcut for saying low to medium PUBSC consumers—or not high PUBSC—consumers. LPUBSC is not the focus of the paper because they do not have embarrassment avoidance behavior in the first place.
A reduction in embarrassment-avoidance might also be achieved by reducing the degree to which individuals feel that others will notice their behavior, i.e., that they are in the ‘spotlight’ (Gilovich et al. 2000). In this paper, however, we focus only on empathy-neglect.
We are aware of the literature on vicarious (i.e. empathetic) embarrassment, which reveals that observers of an actor committing a social blunder often experience that actor’s emotional distress (Edelmann and McCusker 1986; Miller 1987). We focus on somewhat different aspects of embarrassing situations. Specifically, whereas vicarious embarrassment focuses on what an observer may feel while witnessing an actor commit a social blunder, we focus on how the observer will evaluate the actor. In brief, we believe that our work represents a novel extension of the vicarious embarrassment literature.
Unawareness, unintentionality, uncontrollability, and high efficiency are four underlying qualities of automaticity (see Bargh 1994). Evidence of one of these qualities is said to indicate automaticity.
The effect of including income as a covariate did not change the significant level of the two-way interaction: (F(1, 102) = 3.80, p < .05), or main effect (ps > 0.05).
Again, we use the LPUBSC designation as a shortcut for saying low to medium PUBSC consumers—or not high PUBSC—consumers.
The results change slightly if income is not included as a covariate (intentions: M HPUBSC,observer’s perspective = 4.91 vs. M HPUBSC,control = 3.39; t(102) = 1.75, p = .08; intentions: M LPUBSC,observer’s perspective = 4.76 vs. M LPUBSC,control = 6.08; t(102) = 1.64, p = .11).
Again, we use the LPUBSC designation as a shortcut for saying low to medium PUBSC consumers—or not high PUBSC—consumers.
The authors would like to thank Loraine Lau-Gesk for contributions to an earlier version of this research.
- Aiken, L. S., West, S. G., & Reno, R. R. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage.Google Scholar
- Bargh, J. A. (1994). The four horsemen of automaticity: Awareness, efficiency, intention, and control in social cognition. In R. S. Wyer Jr. & T. K. Srull (Ed.), Handbook of social cognition (2nd edn., pp. 1–40). Hillsdale, NJ: Erlbaum.Google Scholar
- Davis, M. H. (1980). A multidimensional approach to individual differences in empathy. JSAS Catalog of Selected Documents in Psychology, 10, 85.Google Scholar
- Edelmann, R. J. (1987). The psychology of embarrassment. New York: Wiley.Google Scholar
- Epley, N. (2014). Mindwise: Why we misunderstand what others think, believe, feel, and want. New York: Vintage.Google Scholar
- Layton, J. (2016). The odd empathy of vicarious embarrassment. https://science.howstuffworks.com. Accessed 23 Feb 2016.
- Miller, R. S. (2007). Is embarrassment a blessing or a curse. In: J. L. Tracy, R. W. Robins, & J. P. Tangney (Eds.), The self-conscious emotions: Theory and research (pp. 245–262).Google Scholar
- Miller, R. S., & Leary, M. R. (1992). Social sources and interactive functions of emotion: The case of embarrassment. In M. S. Clark (Ed.), Review of personality and social psychology, Vol. 14. Emotion and social behavior (pp. 202–221). Thousand Oaks, CA: Sage Publications Inc.Google Scholar
- Schlenker, B. R. (1980). Impression management. Grove, CA: Brooks/Cole Publishing Company.Google Scholar
- Sudman, S., Bradburn, N. M., & Schwarz, N. (1996). Thinking about answers: The application of cognitive processes to survey methodology. New York: Jossey-Bass.Google Scholar