VICARIOUS EMBARRASSMENT OR “FREMDSCHAM”: OVERENDORSEMENT IN FRONTOTEMPORAL DEMENTIA

Abstract

Background:

The experience of embarrassment signals violations in social norms, and impairment in this social emotion may underlie much of the social dysfunction in behavioral variant frontotemporal dementia (bvFTD).

Objective:

To evaluate whether impaired self-awareness of embarrassment may also distinguish patients with bvFTD early in their disease from healthy controls (HCs).

Methods:

We investigated self-reported embarrassment among 18 early bvFTD patients and 23 HCs on a 36-item Embarrassability Scale containing items of situations eliciting embarrassment for themselves (“self-embarrassment”) and embarrassment for others (“vicarious embarrassment”). The groups were also compared on the Social Norms Questionnaire (SNQ). The analyses included correlations of SNQ results (total score, violations or “break” errors, and overendorsement of social rules or “overadhere” errors) with Embarrassabilty Scale scores.

Results:

Surprisingly, the bvFTD patients, compared to the HC subjects, did not differ on total or self-embarrassment scores but had significantly higher vicarious embarrassment scores. Unlike the HCs, the bvFTD participants’ reports of vicarious embarrassment did not differ from their reports of self-embarrassment. The Embarrassability Score further correlated with overadherence to norms on the SNQ.

Conclusions:

In the presence of social dysfunction and emotional blunting, these findings suggest that bvFTD patients rely on their own perspective for a rule-based application of social norms in reporting vicarious embarrassment. The assessment of reports of embarrassment for others may indicate an early and previously unrecognized clinical measure for detecting bvFTD.

INTRODUCTION

Behavioral variant frontotemporal dementia (bvFTD) is predominantly a disorder of socioemotional behavior {¹}. This neurodegenerative disease involves frontal and anterior temporal lobes resulting in disturbances in social interactions and emotional blunting {²}. Investigators have characterized patients with bvFTD as being particularly impaired in self-conscious emotions, which are necessary for the type of feedback that promotes social behavior {^{3, 4}}. These include social emotions such as shame, guilt, pride, and, most prominently, the ability to experience embarrassment.

Embarrassment is emotional discomfort associated with perceived social disapproval for violating a social norm {⁵}. Self-embarrassment requires looking at one’s behavior through the eyes of others, knowing the social norm violation, and self-appraisal for corrective or reparative actions. It is most intense when the disapproval is directed at the self, but embarrassment may also be experienced vicariously when perceiving social disapproval directed at another; this embarrassment-by-proxy is sometimes referred to by its German name of “fremdsham” {⁶}.

BvFTD prominently impairs embarrassment and other self-conscious emotions {⁷}. These patients have decreased emotional reactivity to embarrassing stimuli, such as watching themselves singing(1), decreased self-consciousness {⁴}, and impaired self-awareness from others’ perspective or self-referential processing {^{3, 4, 8–12}}. Despite reactivity to simple happy and sad emotional films and facial emotional responses to auditory startle {^{4, 13}}, patients in early stages of bvFTD display fewer facial signs of embarrassment than controls {⁴}. Finally, bvFTD patients fail to show embarrassment and corresponding autonomic reactivity when recognizing their own errors or mistakes {¹⁴}.

A decreased ability to experience embarrassment may be one of the most basic and early features of bvFTD. These patients are known to have emotional blunting, decreased empathy, and impaired self-referential emotions {^{3, 4, 7, 13, 15}}. They are impaired in experiencing embarrassment on an emotional level, although they can report embarrassment on a cognitive level, i.e., based on what they believe should be felt {¹⁶}. They also lack general emotional awareness, often react inappropriately to social norms, and experience difficulties in taking the perspective of others or mentalization (“Theory of Mind”) {^17–19}. All of these deficits may further impair self-awareness of embarrassment. Consequently, an assessment of self-reports of embarrassment in social situations may be a highly sensitive measure for detecting early bvFTD, a disorder which lacks a definitive clinical test.

This study evaluated self-perceived embarrassment among patients with bvFTD compared to normal controls. They were administered an Embarrassability Scale containing items equally divided between self and vicarious embarrassment. This was contrasted with their behavior on the Social Norms Questionnaire (SNQ), which measures the breaking or overadherence to social rules as part of the Frontotemporal Lobar Degeneration Module of the National Alzheimer’s Coordinating Center database {²⁰}. The results of this study did not support the initial predication that the patients with bvFTD patients would report significantly less embarrassment; instead, these patients reported higher levels of vicarious embarrassment.

METHODS

Participants

Participants with BvFTD were recruited from the UCLA Behavioral Neurology Program and Clinic where they underwent clinical, neuropsychological, and neuroimaging assessments. Age-matched healthy controls (HCs) were recruited from volunteers in the community. Under approval by the UCLA Institutional Review Board, this study enrolled 41 participants: 18 patients with bvFTD and 23 HCs. Participants and caregivers gave informed consent. Patients with bvFTD met International Consensus Criteria for clinically probable bvFTD {²}. Their clinical diagnoses of bvFTD were supported by predominant frontal and anterior temporal involvement on magnetic resonance imaging and/or fluoro-deoxyglucose positron emission tomography scans of the brain. Exclusion criteria included the presence of complicating medical or psychiatric illnesses or psychoactive medication.

Procedures

The bvFTD patients were evaluated with four social and emotional scales. Two were self-administered to both bvFTD and HC participants: the UCLA Embarrassability Scale modified from Modigliani A, 1968 {²¹}, and the Social Norms Questionnaire (SNQ) {^{22, 23}}. Two other caregiver-administered scales were completed by the cargivers of the patients with bvFTD: the Social Dysfunction Scale (SDS) {²⁴}, and the Scale for Emotional Blunting (SEB) {¹⁵}.

Self-Administered Scales

1. The UCLA version of the Embarrassability Scale, which was modified from A. Modigliani’s original scale {²¹}, is a significantly expanded and modernized instrument for contemporary U.S. usage. It is a self-administered instrument containing 36 items divided into 18 self-embarrassing situations and 18 items depicting situations embarrassing for others (vicarious items). The participants are asked to imagine these situations as vividly as possible. These items ask whether the situation would cause the participant embarrassment, including self-consciousness, awkwardness, discomfort, or a sense of social exposure. They are told that some items may involve feeling embarrassed for oneself, and other items may involve feeling embarrassed for someone else. They are then asked to record their level of embarrassment on a five-point Likert scale, with 1 indicating “I would not feel the least embarrassed: not awkward or uncomfortable at all” to 5 indicating “I would feel strongly embarrassed: extremely self-conscious, awkward, and uncomfortable”. Examples of self-embarrassment items include: “Suppose you go to pay at a restaurant and find that you did not bring your purse or wallet”; “Suppose that, while laughing heartily with a group of friends, you pass gas”. Examples of vicarious embarrassment items include: “Suppose you observe someone go into a bathroom and emerge with toilet paper sticking to his shoes”; “Suppose you are watching an amateur comedy show and one of the performers is unable to make anyone laugh with her jokes.”

2. The Social Norms Questionnaire (SNQ) is a 22-item ‘yes’ or ‘no’ questionnaire given to the participants to detect inappropriate social behavior in hypothetical scenarios {^{22, 23}}. The SNQ is recommended as part of the socioemotional evaluation of patients with bvFTD {²⁰}. Written directions are, “The following is a list of behaviors that a person might engage in. Please decide whether or not it would be socially acceptable and appropriate to do these things in the mainstream culture of the United States and answer yes or no to each. Think about these questions as if they were occurring in front of or with a stranger or acquaintance, NOT a close friend or family member.” For example, “would it be socially acceptable to wear the same shirt twice in two weeks?” A total score is obtained by summing correct items with higher scores indicating greater knowledge of social norms. There are subscales that measure two types of errors: Break errors refer to endorsement of a socially inappropriate behavior (e.g., eating pasta with your fingers) as appropriate; Overadhere errors refer to endorsement of a socially appropriate behavior (e.g., wearing the same shirt twice in two weeks) as inappropriate. Although used successfully to distinguish bvFTD and other populations {^{22, 23}}, the reliability of the SNQ is not established.

bvFTD Caregiver-Administered Scales

1. The Socioemotional Dysfunction Scale (SDS) is a 40-item informant-based rating scale previously used to characterize social behavioral disturbances in bvFTD {²⁴}. The SDS is completed by a spouse, family member, caregiver, or other informant who knows the patient well. The scale originates from the Social Competency Questionnaire (SCQ), a measure of adaptive social behaviors, with modifications for use with patients with bvFTD. Written instructions are, “For each item, base ratings on a comparison with typical behavior before disease symptoms emerged.” Informants rate items regarding the participant’s social behavior on a 5-point Likert scale (1-to-5) as follows: 1 = Very Inaccurate; 2 = Somewhat Inaccurate; 3 = Neither Accurate, Nor Inaccurate; 4 = Somewhat Accurate; 5 = Very Accurate. For example, “Makes inappropriate comments to others.” The 40-items are summed, yielding a total raw score with higher scores suggestive of greater social dysfunction. Prior work has shown high internal consistency reliability for the SDS (Cronbach’s α = 0.977) {²⁴}.

2. Caregivers complete the Scale of Emotional Blunting (SEB), an instrument used successfully to evaluate patients with bvFTD {²⁵}. This scale queries domains such as absence of pleasure-seeking behavior (Behavior), affective blunting (Affect), and cognitive blunting (Thought). Each behavioral symptom is rated by indicating “condition absent” (zero points), “slightly present or doubtful” (one point), and “clearly present” (two points). The behavior subscale items include symptoms akin to reclusivity or avoiding social contact. The affect domain involves behaviors such as lacking warmth or empathy. The thought subscale includes items of lacking plans, ambition, desires, or drive. The inter-rater reliability coefficient for this instrument has been strong (α=0.91) in prior testing {¹⁵}.

Disease Severity

Disease severity of the participants with bvFTD was assessed with disease duration, the Mini-Mental State Examination (MMSE) {²⁶}, the Montreal Cognitive Assessment (MoCA) {²⁷}, and the Functional Assessment Questionnaire (FAQ) {²⁸}.

Statistical Analysis

Statistical analysis was conducted using SPSS 25.0 software. Demographic descriptive statistics were generated for each group and compared using Chi-square and t-test for categorical and continuous variables, respectively. Testing for distribution of data with the Shapiro-Wilk test rejected the null hypothesis that the data for the two major scales were normally distributed. Mann Whitney U tests were subsequently used for comparison of means between the two groups. The Wilcoxon signed-rank test was used for within group evaluation of embarrassability subscale scores. Across both groups, Spearman Ranked Order correlations were computed to examine the relation between the two self-administered behavioral scales.

RESULTS

Participant characteristics

There were no significant group differences on demographic variables of age, sex, or years of education (See Table 1). The bvFTD patients were, on average, about four years into their disease from first symptoms and had MMSE and MoCA scores in the mild range, despite moderate functional impairment on the FAQ. Their level of cognitive impairment did not preclude understanding and completion of the self-administered scales.

Table 1:

Demographics and Clinical Measures for Participants with Behavioral Variant Frontotemporal Dementia (bvFTD) and Healthy Controls (HCs).

	bvFTD n=18		HCs N=23		p-value
	Mean	SD	Mean	SD
Age	61.25	10.1	56.33	8.16	n.s.
Sex (M,F)	(9,9)		(11,12)		n.s.
Education	15.92	2.43	16.09	1.78	n.s.
Disease duration (years)	4.13	3.72
Mini Mental State Examination (MMSE)	23.67	4.65
Montreal Cognitive Assessment (MoCA)	17.33	6.49
Functional Assessment Questionnaire (FAQ) - Total	19.3	6.50
Socioemotional Dysfunction Scale (SDS)	142.21	31.22
Scale for Emotional Blunting (SEB)	12.25	9.26

Abbreviations: SD = standard deviation. Cutoff scores: FAQ > 9 is abnormal (range 0–30); SDS > 105 indicates social dysfunction (range 40–200); SEB >105 indicates emotional dysfunction in bvFTD (range 40–200).

Behavioral assessments

On the caregiver administered scales, the bvFTD group, compared to published norms {^{15, 24}}, had social dysfunction and emotional blunting on the SDS and SEB, respectively (See Table 1).

On the self-administered scales, the bvFTD patients did not differ significantly from HCs on the Embarrassability Scale Total scores (See Table 2). On the subscale measures, there were no group differences in self-embarrassment; however, the bvFTD patients endorsed greater vicarious embarrassment (p<0.05) compared to HCs. Within the bvFTD group, the patients failed to report the significantly greater embarrassability for items involving self vs. others that was present among the HCs (p<0.001).

Table 2:

Embarrassability Scale and Social Norms Questionnaire Results for Participants with Behavioral Variant Frontotemporal Dementia (bvFTD) and Healthy Controls (HCs).

	bvFTD n=18		HC N=23		p-value
	Mean	SD	Mean	SD
Embarrassability Total Score	110.17	32.61	100.0	23.64	n.s.
Self-embarrassment subscore	56.28	17.68	57.13	11.66	n.s.
Vicarious embarrassment subscore	53.89	17.88	42.87	14.86	p=0.032
Social Norms Questionnaire	15.93	2.89	19.86	1.64	p<0.001
Break Error Score	1.53	1.34	0.86	0.89	n.s.
Overadherence Error Score	4.54	2.77	1.27	1.45	p<0.001

Abbeviations: SD = standard deviation. Independent samples with differences tested using the Mann-Whitney U test.

The bvFTD patients had more social norm errors than the HCs on the SNQ (p<0.001). On the subscale measures, there were no group differences in SNQ break errors; however, the bvFTD patients made more overadherence errors on the SNQ (p<0.001) compared to HCs. Correlational analysis showed significant Spearman rho correlations between total Embarrassability Score and SNQ overadherence errors (p<0.05), with a trend to significance between vicarious embarrassment and overadherence errors (p=0.056) (See Table 3).

TABLE 3:

Embarrassability Scale-Social Norms Questionnaire Correlations Across Groups.

	Embarrassability Total Score		Self=Embarrassment subscore		Vicarious Embarrassment subscore
	rho	p-value	rho	p-value	rho	p-value
SNQ Total Score	−0.187	p=0.297	−0.053	p=0.770	−0.21	p=0.236
SNQ Break subscore	−0.199	p=0.266	−0.312	p=0.077	−0.124	p=0.490
SNQ Overadhere subscore	0.359	p=0.04	0.289	p=0.103	0.336	p=0.056

Abbreviations: SNQ = Social Norms Questionnaire; rho = Spearman’s rho. P values given at two-tailed significance levels.

DISCUSSION

A decreased ability to be embarrassed may underlie many of the social behavioral disturbances in bvFTD, and an assessment of their awareness of embarrassment could aid in the early recognition of this disorder. This study investigates self-reports of embarrassment among patients with bvFTD compared to HCs. The bvFTD patients do not differ from HCs on the total Embarrassability Scale or on embarrassing items involving themselves; however, the bvFTD patients, compared to HC subjects, report experiencing significantly greater embarrassment for items involving embarrassing situations for others. In comparison, their caregivers rate the bvFTD patients as having social dysfunction and emotional blunting, and the patients rate themselves worse than HCs in responding to social norms. Finally, the Embarrassment Scale results correlate with overadherence to social norms. Together, these findings suggest that, among bvFTD patients, increased endorsement of vicarious embarrassment may reflect an inability to take another’s perspective resulting in rule-based responses regarding levels of embarrassment.

Social emotions such as embarrassment are distinct from the basic emotions {⁵}; in particular they particularly require the ability to see oneself and others from the perspective of others {^29–31}. Self-embarrassment involves self-appraisal of adherence to social norms as seen by others, prompting corrective or reparative actions when necessary {^{5, 32–42}}. However, embarrassment can also be a vicarious experience when perceiving the embarrassing actions or mistakes of others {^{6, 43}}. On observing someone else violating social norms with its potential for social disapproval {⁴⁴}, people imagine themselves from the other’s perspective. Vicarious embarrassment, or fremdscham, is distinct from emotional contagion because it does not require observing the victims emotional reaction or even the presence of third parties {⁴³}. Both forms of embarrassment depend on taking the perspective of others, an aspect of Theory of Mind, and assessing violations of social norm rules.

Self-conscious emotions such as embarrassment involve the frontal, temporal and limbic areas affected by bvFTD {^{36, 41, 42, 45–48}}. Frontopolar, ventromedial frontal, and basal forebrain regions are involved in prosocial sentiments such as embarrassment {⁴⁷}. In an embarrassing task where bvFTD patients watch themselves singing karaoke, there is decreased physiological and behavioral reactivity associated with smaller right pregenual anterior cingulate cortex (ACC) gray matter volumes {⁴⁹}. Situations that trigger vicarious embarrassment involve brain areas for pain, either physical or social, such as the ACC and the left anterior insulae {^{43, 50}}. These are all areas affected early in bvFTD.

Given the lack of self-conscious emotions and actual feelings of embarrassment in bvFTD, it is quite notable that these patients report more vicarious embarrassment than HCs. There appear to be several reasons for this. First, patients with bvFTD have deficits in mentalization or Theory of Mind, the basic yardstick for taking the perspective of others {¹⁸}. Hence, their perspective is from their own point of view. Second, the claim of increased embarrassment is associated with overadherence errors to social norms. Increased overadherence errors in bvFTD correlate with difficulty recognizing the changing context of a rule {²²}, and bvFTD facilitates rule-based and utilitarian judgments based on previously learned social rules {^{51, 52}}. Consequently, even in reporting self-embarrassment, these patient s may be responding less to true social discomfort and more to what they perceive as an expected level of embarrassment. They may be applying the same predetermined rule for level of self-embarrassment per social norm violation to the reporting of embarrassment for others.

There are a number of alternative, but less plausible, explanations for these results. The bvFTD patients could have a truly increased sense of vicarious embarrassment. This interpretation, however, goes against the vast body of research on the socioemotional impairments of bvFTD {^{3, 4, 7, 13, 15}}, including the basic diagnostic criteria of loss of empathy or sympathy {²}. A second consideration is that on vicarious items, the bvFTD patients report the perceived embarrassment or social discomfort of others rather than their own. This seems very unlikely, given the clear instructions and “vicarious” reports as well as the impaired mentalization and perspective taking of bvFTD) {^17–19}. Another consideration is that this study cannot entirely exclude an effect on the results from the limited, established reliability and other psychometric aspects of the Embarrassability Scale and other scales.

There are potential limitations to this study that can be addressed in future investigations. First, there is the consideration of the number of patients, especially in the bvFTD group. Nevertheless, the number of participants was sufficient to disclose group differences on some scales, but not others. Second, it would have been beneficial to include a direct measure of mentalization or Theory of Mind. Third, it would have also added greatly to have had autonomic measures for assessing a physiological reaction of embarrassment. Fourth, the modified Embarrassability Scale has not been broadly validated in a large population. Finally, further investigation can profit from other control groups, such as those with Alzheimer’s disease.

In conclusion, the results show that bvFTD patients, despite known impairment in experiencing embarrassment, report significantly more vicarious embarrassment compared to normal controls, and this vicarious embarrassment appears associated with overadherence to social norms. Together, these results suggest a self-centered, rule-based reasoning to questions of embarrassment for others. Given their social behavioral changes, and the relative insensitivity of traditional neuropsychological measures for early bvFTD, a self-rating scale of vicarious embarrassment could be very helpful in the initial recognition of patients with this disease.