Self-Concept in Narcissism: Profile Comparisons of Narcissistic Manifestations on Facets of the Self

Abstract

Objective

Clinical theories suggest that narcissists have a compromised self-concept. However, empirical investigation on attributes of the self that would be impaired in pathological narcissism is limited and inconsistent. The present study aims at detecting distinctive profiles of narcissistic manifestations on facets of the self that have been indicated as relevant in clinical and empirical literature on narcissism.

Method

We measured adaptive and pathological narcissistic traits in a community sample of adults (N = 539). Participants also completed measures of self-uniqueness, self-authenticity, self-consistency, and self-other comparisons on agentic and communal domains.

Results

Results indicate distinctive profiles of adaptive and pathological narcissistic manifestations on these facets of the self. Among the set of distinctive facets for each narcissistic manifestation, however, some showed to have a more prominent role. Adaptive and pathological narcissism were captured mostly by a greater sense and need for uniqueness that was primarily expressed by public exposure. Sense of superiority over others in the agentic domain, however, showed to have an essential role only in adaptive narcissism. Moreover, self-concept in adaptive grandiose narcissism was qualified by high levels of self-authenticity and a consistent sense of self. Self-concept in vulnerable pathological narcissism revealed greatest impairment, especially in facets of high concern regarding others’ reactions and feeling of a tenuous existence.

Conclusions

The study points out that adaptive and pathological manifestations of narcissism can be profiled based on specific facets of self. Theoretical and research implications are discussed.

Introduction

Clinicians describe narcissists as individuals having an inflated but extremely fragile sense of self (Caligor, 2013; Ronningstam, 2009). In line with these observations, the DSM-5 Alternative Model for Personality Disorders (AMPD; APA, 2013) suggests that personality disorders, including narcissistic pathology, are characterized by identity impairments, which involve difficulties in emotion regulation and a compromised sense of self. These assumptions, however, do not seem to consider the heterogeneity of narcissistic manifestations that have been recently stressed in contemporary theories (e.g., Miller et al., 2017; Pincus & Lukowitsky, 2010) and that have been supported empirically (e.g., Miller et al., 2011 Pincus et al., 2009). Moreover, empirical research has under-examined self-concept (and its specific impairments) in pathological narcissism.

The heterogeneous expressions of narcissism

Narcissism is a multifaceted construct, and differences among its expressions need to be taken into account when investigating its psychological and behavioral correlates. Among the field of clinical psychology, Pincus and colleagues (Pincus & Lukowitsky, 2010; Pincus et al., 2009) provided a new conceptualization of pathological narcissism which clearly distinguished adaptive and pathological expressions, and that took into account both vulnerable and grandiose manifestations of pathological narcissism.

According to the authors (Pincus & Lukowitsky, 2010), adaptive narcissism reflects the ability to maintain a stable and positive self-image, as well as to cope with needs for validation and affirmation through adequate self-, field, and emotion regulation processes. In this sense, normal personality functioning always includes adaptive narcissistic tendencies to some extent (e.g., Ronningstam, 2016). Conversely, pathological narcissism involves intense needs for validation and recognition in the context of impaired self-regulating processes (Pincus & Lukowitsky, 2010), as shown by its association with several maladaptive correlates such as unstable self-esteem (Di Pierro et al, 2016), dysfunctional perfectionism (Di Pierro et al., 2020) and aggression (Pincus et al., 2009). In particular, pathological narcissism comprises both grandiose and vulnerable manifestations (Pincus et al., 2009). Grandiose manifestations include a sense of entitlement, fantasies of unlimited power, and exploitative attitudes, whereas feelings of emptiness, envy, hypersensitivity, and an avoidant interpersonal style typically describe vulnerable manifestations (Pincus et al., 2009).

Self-concept in pathological narcissism: theoretical evidence perspectives and empirical

A clear and unambiguous definition of self-concept and its essential features lacks in the theoretical literature on personality pathology. As a consequence, it is difficult to identify univocally which specific aspects of the self would be compromised in people with personality disorders. On the one hand, the term “self-concept” has been used interchangeably by theorists with other words such as sense of self, self-representation, self-image, and self-perception. Nevertheless, these different appellations reflect constructs that only partially correspond with each other. Moreover, there has been significant variability across theories in describing essential aspects of the self that are distinctive of people with personality pathology in general, and specifically of narcissists. It is the case, for instance, of theories underlying major evidence-based treatment models for personality pathology, such as the Transference-focused Psychotherapy, the Mentalization-based psychotherapy, and the Schema-focused therapy.

Kernberg’s object relation theory (Yeomans et al., 2015) states that identity diffusion is one of the key elements of personality disorders laying at the borderline level of personality organization. Identity diffusion involves “the lack of an integrated self-concept and an integrated and stable concept of total objects in relationship with the self” (p. 39, Kernberg, 1985). According to this theoretical perspective, narcissists would suffer from polarized, rigid, and poorly articulated mental representations of themselves and others. In fact, narcissistic self-view is predominantly positive, and dominated by all-good and ideal features at a surface level, and negative, all-bad, and vulnerable features are projected onto others (Diamond & Hersh, 2020; Stern et al., 2017). When exploring narcissists’ self-view at a deeper level of analysis, however, it emerges a fragile, superficial, and a vague sense of self that is particularly sensitive to ego-threats (Caligor et al., 2018). In other words, Kernberg’s object relation perspective points out that grandiose attributes of the self in narcissists cover an inconsistent self and substitute for an authentic sense of self (Kernberg, 1984).

Schema Therapy (ST; Young et al., 2003) is a relatively new therapeutic approach that has been implemented in personality disorders patients. ST assumes that personality pathology would result from maladaptive schema modes, namely “self-defeating emotional and cognitive patterns that begin early in our development and repeat throughout life” (Young et al., 2003, p. 7). In particular, childhood experiences of unmet basic needs (i.e., physical, psychological, and emotional needs), paired with negative or traumatic interactions with caregivers, would interfere with development of child’s autonomy, connectedness, worthiness, realistic expectations and limits (Young et al., 2003). Basic needs for love and affection are usually unmet in childhood of narcissists, and they get to know the so-called “conditional love”, that is a “positive attention for imposed (and often high) expectations of performance” (Behary & Dieckmann, 2011, p. 446). Starting from these early experiences, narcissists would develop three adult schema modes: the detached self-stimulator mode, which serves to create a distance from others (and from intimacy with others) in order to cut off any uncomfortable emotion; the self-aggrandizer mode, which involves proneness to feel superior over others, and to the need for appearing and for having a high social status at the expense of an authentic self; and the lonely child mode, which represents the underlying state of narcissism, and mirrors proneness to feel unloved and to be easily hurt by others.

The Mentalization-Based Therapy (MBT; Fonagy et al., 2002) is based on the assumption that patients with personality disorders have prominent deficits in mentalization (i.e., awareness of one's own and of others’ mental states), that result from early experiences with caregivers who do not provide adequate mirroring of child’s internal states. In borderline conditions, early experiences with a caregiver who do not provide sufficient contingent mirroring of the baby’s internal states would promote the development of an “alien self” which reflects “a fault in the construction of the self, whereby the infant is forced to internalize the object’s state of mind as a core part of himself” (p. 320, Fonagy et al., 2002). According to Drozek and Unruh (2020), childhood experiences of narcissists would involve parental overvaluation of baby’s qualities related to self-confidence and strength, while undermirroring manifestations of emotional vulnerability such as insecurity and need for closeness. These early experiences would promote the development of a “narcissistic alien self”, namely a “secondary representation of the caregiver overvaluing the child, which, because it fails to map onto the child’s primary affective states, would lead to a profound sense of emptiness and discontinuity in the self” (p. 181, Drozek & Unruh, 2020). As a consequence, narcissists would lack an authentic base for self-hood (Weinberg, 2006), and they would engage in behavioral and internal self-enhancement strategies, such as feelings and attitudes of superiority over others, in order to restore a sense of continuity and of self-coherence (Drozek & Unruh, 2020).

By exploring convergences in theoretical accounts and empirical findings (for a review, see Bender et al., 2011), the AMPD has summarized aspects of the self-concept that would be univocally linked to personality pathology. According to the AMPD (APA, 2013), people with personality pathology show impairments in the “experience of oneself as unique, with clear boundaries between self and others; and stability and accuracy of self-appraisal” (p. 762). In other words, personality pathology would involve an inconsistent sense of self (i.e., an unstable, unclear, incoherent, and distorted view of oneself) and a compromised sense of self-uniqueness that leads to overidentification with others or, on the contrary, to an excessive need for independence from others.

Empirical evidence (Di Pierro et al., 2020; Sollberger et al., 2012) has confirmed that impairments in these specific facets of the self are typical of PDs patients and may distinguish them from patients with other psychopathological conditions. In the case of narcissistic pathology, the AMPD (APA, 2013) indicates that impairment of the self takes the specific form of exaggerated self-appraisal in terms of inflated and deflated self-evaluations (or oscillations between the two), as well as of excessive reference to others for self-definition and self-esteem regulation. Interestingly, Schalkwijk et al. (2021) have shown that specific identity impairments of narcissistic personality disorder in the AMPD mirror key elements of narcissistic pathology described in contemporary psychodynamic theories (e.g., Caligor, 2013; Meissner, 2008). From an empirical perspective, however, the relationship between these specific impairments and pathological narcissism has not been investigated systematically.

Great empirical attention has been given to the association between narcissism and self-esteem for a long time. In particular, most studies have investigated the associations of narcissistic manifestations with implicit and explicit self-esteem (e.g., Di Pierro et al., 2016; Zeigler‐Hill, 2006) in order to test the mask model (Freud, 1914; Kohut, 1966; Kernberg, 1975). According to the mask model, indeed, narcissists would show a grandiose self (i.e., high explicit self-esteem) that masks a deep-seated insecurity (i.e., low implicit self-esteem). However, empirical studies aiming at capturing discrepancies between explicit and implicit self-esteem levels showed inconsistent findings (Bosson et al., 2008).

Beyond the mask model, empirical studies on impairments of the self-concept related to self-consistency and self-uniqueness in pathological narcissism have been limited in number. Moreover, most of these studies have suffered from limitations that question the validity of findings. In particular, past empirical studies have focused exclusively on grandiose narcissism for a long time, neglecting vulnerable narcissistic manifestations. Moreover, they did not discriminate adequately between adaptive and pathological expressions of narcissism. In fact, most of them were based on the Narcissistic Personality Inventory (NPI; Raskin & Hall, 1979), which has been shown to assess a mix of adaptive and pathological traits of grandiose narcissism (e.g., Cain et al., 2008).

In order to overcome this limitation, newly developed measures of narcissism have been recently elaborated, as in the case of the Pathological Narcissism Inventory (PNI, Pincus et al., 2009) and the Five Factor Narcissism Inventory (FFNI, Glover et al., 2012). Both self-report instruments have shown to provide a valid measure of pathological narcissism, by capturing grandiose and vulnerable manifestations (Crowe et al., 2019).

Using both the FFNI and the PNI, Kaufman et al. (2020) have recently found a significant link between pathological narcissism and lack of consistency in the sense of self: individuals high in both grandiose and vulnerable manifestations of pathological narcissism are prone to confusing their own feelings and thoughts with those of others, have difficulties in understanding themselves, feel their existence is tenuous, and show sudden shifts in feelings, opinions, and values. Kaufman et al. (2020) have also investigated the association of pathological narcissistic manifestations with authenticity of the self. The authors based their study on the Wood et al. (2008)’s tripartite conception of authenticity. According to Wood et al. (2008), a sense of authentic self is described by low levels of self-alienation (i.e., the mismatch between the conscious awareness and the actual experience of the self), low levels of accepting external influence (i.e., the tendency to being influenced by others or conform to others’ expectations), and high levels of authentic living (i.e., the proneness to behaving consistently with individual’s psychological states). Results from Kaufman et al. (2020) have shown that both grandiose and vulnerable pathological narcissism, measured through the FFNI and the PNI, were related to high levels of self-alienation and accepting external influence, while only vulnerable narcissism was associated also with poor authentic living. These results, however, need further examination since Kaufman et al. (2020) examined the association of PNI grandiose narcissism with authenticity and inconsistency in the sense of self without taking into account the effect of coexisting traits of vulnerable narcissism. In fact, recent studies (Crowe et al., 2019; Edershile et al., 2019) have demonstrated that the PNI grandiose scale comprises vulnerable themes and that scholars should control for PNI vulnerable narcissism when investigating correlates of PNI grandiose narcissism.

Few studies have investigated the experience of self-uniqueness in pathological narcissism. Theoretical literature (Şimşek & Yalınçetin, 2010) on self-uniqueness suggests discriminating between personal sense and need for uniqueness. Personal sense of uniqueness indicates “personal perceptions of self that are unique to the individual and different from others” (Şimşek & Yalınçetin, 2010, p. 576). Personal Sense of uniqueness is a unidimensional construct mirroring one’s private evaluation of specialness, and it is associated with indicators of well-being (e.g., optimism, resilience, and life satisfaction; Şimşek & Yalınçetin, 2010). Conversely, need for uniqueness mirrors uniqueness seeking, meaning the proneness and desire to be unique and distinct from others (Snyder & Fromkin, 1977). Need for uniqueness may represent a defensive mechanism (Şimşek & Yalınçetin, 2010) and it is linked to correlates such as sensation seeking (Baird, 1981) and low agreeableness (Schumpe et al., 2016). Recently, van Doeselaar et al. (2019) have shown that admiration narcissistic strategies are uniquely associated with a greater sense of uniqueness, while rivalry narcissistic strategies do not. According to the authors, people that engage in narcissistic admiration strategies would feel usually well-differentiated from the others as a way to promote a grandiose self. However, these findings do not clarify whether high levels of personal sense of uniqueness are typical of adaptive or pathological grandiose narcissism. In fact, previous studies demonstrated that admiration strategies are related to both adaptive and pathological manifestations of narcissistic grandiosity, while rivalry strategies are linked to pathological narcissistic vulnerability only (Back et al., 2013; Krizan & Herlache, 2018). Grandiose narcissism has shown to be associated also with a greater need for uniqueness (Emmons, 1984). Recent findings on the use of social media support this association: individuals with high traits of NPI narcissism show high proneness to employ social networks to satisfying needs related to uniqueness, superiority, and admiration (Giordano et al., 2019; McCain & Campbell, 2018). Again, however, most of the studies measured only grandiose narcissism administering self-reports that did not discriminate between adaptive and pathological expressions of grandiose narcissism (e.g., NPI), making it difficult to interpret its results (for a review see McCain & Campbell, 2018)

Finally, another relevant aspect of the self in pathological narcissism is sense of superiority (or inferiority) over others, as suggested by the AMPD. Clinical observations (Caligor et al., 2015) and empirical evidence in the field of social personality psychology (Miller et al., 2011) indicate that feelings of superiority are typical of grandiose narcissists. To the best of our knowledge, however, only a few studies have investigated the association between traits of grandiose narcissism and proneness to feel superior over others (Campbell et al., 2002). By comparing individuals’ self-and other-descriptions in both agentic and communal domains, the authors found that the higher grandiose narcissism, the higher proneness to describe oneself as more agentic than significant others (i.e., romantic partner), but not as more communal. A significant association between grandiose narcissism and sense of superiority over significant others in the agentic domain has been found also by Krizan & Bushman (2011), whereas Freis & Hansen-Brown (2021) demonstrated that people high in narcissistic grandiosity are prone to feel superior in social situations as well. These findings confirm clinical observations on grandiose narcissists and clarify that sense of superiority over others is limited to agentic features of the self. Again, however, all these studies measured grandiose narcissism through the NPI. Therefore, further studies are needed to untangle the role of adaptive and pathological expressions of grandiose narcissism in the sense of superiority over others. Unlike grandiose narcissists, some authors suggest that vulnerable narcissists would feel inferior to others (Pincus & Lukowitsky, 2010; Ronningstam, 2009), and recent empirical findings support indirectly this assumption by showing intense shame experiences in vulnerable narcissists (Di Sarno et al., 2020). Any study, however, has investigated sense of inferiority in pathological vulnerable narcissism by comparing directly self- and other-descriptions in both agentic and communal domains.

The present study

The present study examines the association of pathological narcissism with facets of the self-concept that would be primarily impaired in pathological narcissism, according to the AMPD (APA, 2013) and clinical theoretical literature (e.g., Fonagy et al., 2002; Kernberg, 1985; Young et al., 2003). In particular, we investigated whether and how pathological narcissism relates to measures of consistency in the sense of self, self-authenticity, personal sense of uniqueness, need for uniqueness, and sense of exaggerated self-appraisal both in agentic and communal domains. We considered both adaptive and pathological expressions of grandiose narcissism, as well as both grandiose and vulnerable manifestations of pathological narcissism. By doing so, we aimed at drawing up profiles of aspects related to the self-concept that are uniquely linked to pathological manifestations of narcissism.

Consistent with previous findings showing that adaptive narcissism relates to psychological wellbeing (e.g., Rosenthal & Hooley, 2010), we hypothesize that adaptive traits of narcissism would be linked to a strong sense of self, namely a unique, authentic, and consistent sense of self (Campbell et al., 2007). Also, we expect that adaptive narcissism would not be significantly linked to need for uniqueness and to sense of superiority over others. Conversely, based on previous findings (Kaufman et al., 2020) and clinical observations (e. g., Ronningstam, 2005), we expect to find that pathological narcissism would be linked to sense of superiority over others in the agentic domain, greater sense and need for uniqueness, and low consistency and authenticity of the self. In line with van Doeselaar et al. (2019), we expect that vulnerable narcissism would not be significantly associated with personal sense of uniqueness. Moreover, unlike pathological grandiose narcissism, we expect people high in pathological vulnerable narcissism to feel inferior to significant others. Albeit no studies have specifically investigated comparisons between self-and other-descriptions in vulnerable narcissists, we based our expectation on existing findings showing that individuals high in pathological vulnerable narcissism are prone to experiencing intense shame (Di Sarno et al., 2020), are low in self-esteem (Di Pierro et al., 2016), and have feelings of helplessness (Pincus et al., 2009).

Method

Participants and Procedure

The study involved a community sample of 539 Italian adults (412 females), with a mean age of 24.64 (SD = 5.103, age-range 18 – 59).

Around 44.9% (N = 242) of participants were full-time students, 24.7% (N = 133) were working-students, 24.9% of participants declared to be employed (N = 134), and 5.5% of participants were unemployed (N = 30). About half of the participants had a high-school level of education or below (N = 267; 49.6%), and the other half reported to have a university degree or higher (50.4%, N = 272).

Participants were invited to participate in this study through announcements on social networks. All participants signed an informed consent and did not receive any incentive (e.g., money or credits) to participate in the study. Exclusion criteria for all participants were as follows: 1) being less than 18 years old; 2) not being Italian. The study was approved by the Ethical Committee in charge (prot. n. RM-2020-283), and all procedures followed were in accordance with the Helsinki Declaration of 1975 and its later amendments.

Measures

Pathological narcissism was assessed by the Pathological Narcissism Inventory (PNI; Pincus et al., 2009). The PNI is a 52-item self-report measure of grandiose and vulnerable pathological narcissism. Items are scored on a 6-point likert scale (0 = not at all like me, 5 = very much like me). Both the scales showed good internal consistency (GN: α = .82; VN: α = .93) In the present study, we administered the Italian version of the PNI (Fossati et al., 2015). Adaptive narcissism was assessed by the Narcissistic Personality Inventory (NPI; Raskin & Terry, 1988), which consists of 40 forced-choice items. For the purpose of the study, we only considered the NPI Leadership/Authority scale (α = .73), since it captures adaptive traits of narcissism (Ackerman et a., 2011). In the present study, we administered the Italian version of the NPI (Fossati et al., 2008).

Need for uniqueness is assessed by the Need for Uniqueness scale (NFU; Snyder & Fromkin, 1977). The NFU consists of 32 items rated on a 5-point likert scale (1 = strongest disagreement, 5 = strongest agreement). The NFU assesses 3 scales: Lack of concern regarding others’ reactions (NFU_L), Desire not to always follow rules (NFU_D), and Willingness to publicly defend one’s beliefs (NFU_W). All the scales showed good internal consistency in our sample (range α = .71 ‒ .78).¹

Personal sense of uniqueness was assessed by the Personal Sense of Uniqueness (PSU; Şimşek & Yalınçetin, 2010). The PSU consists of 5 items scored on a 5-point scale (1 = not at all true, 5 = very true). The PSU showed acceptable internal consistency in our sample (α = .73).¹

Self-Others Comparison. Participants were asked to describe themselves according to some adjectives reflecting both agentic and communal attributes (Abele et al., 2008), using a 9-point likert-scale (0 = not at all, 8 = extremely). They were also asked to describe their significant others according to the same adjectives, and using the same likert-scale. Agentic attributes included three adjectives with positive valence (assertive, determined, striving) and three adjectives with negative valence (gullible, shy, vulnerable); communal attributes included three adjectives with positive valence (helpful, sympathetic, understanding) and three adjectives with negative valence (detached, dogmatic, egoistic). A measure of Agentic Self-description (α = .68) was obtained by averaging self-ratings of the three adjectives with positive valence and the three adjectives with negative valence (reversed scores); the same procedure was implemented for obtaining Communal Self-description (α = .65), as well as Agentic (α = .67) and Communal descriptions (α = .71) of significant others. Delta scores between self- and others-descriptions in both agentic (Self-Other Agency) and communal domains (Self-Other Communion) were computed to obtain measures of self-others comparison. Positive scores in Self-Others Agency and Self-Others Communion indicate people describe themselves respectively as more agentic and communal than significant others; negative scores in Self-Others Agency and Self-Others Communion indicate people describe themselves respectively as less agentic and communal than significant others.

Authenticity of the Self was assessed by the Authenticity Scale (AS; Wood et al., 2008), which consists of 12 items measured on a 7-point scale (1 = does not describe me at all, 7 = describes me very well). The AS includes three scales: Self-alienation (AS_SA), Authentic living scale (AS_AL), Accepting external influence scale (AS_AEI). The three scales showed good internal consistency in our sample (range α = .73 ‒ .83). In the present study, we administered the Italian version of the AS (Casale et al., 2018).

Consistency in the sense of Self is measured by the Sense of Self Scale (SOSS; Flurry & Ickes, 2007; Culwell, 2008), which consists of 16 items measured on a 5-point scale (1 = strongly disagree, 5 = strongly agree). The SOSS includes four scales: Lack of understanding of oneself (SOSS_L); Sudden shifts in feelings, opinions, and values (SOSS_S); Tendency to confuse one’s feelings, thoughts, and perspectives with those of others (SOSS_C); Feeling of a tenuous existence (SOSS_T). The four scales showed good internal consistency in our study (range α = .62 ‒ .80).¹

Statistical Analyses

Statistical analyses were conducted using SPSS 25.0 (IBM Corp., Armonk, NY, USA).

Descriptive analyses on all the variables under investigation were conducted. In order to detect profile comparisons of narcissistic manifestations on facets of the self, we conducted correlation analyses of adaptive and pathological narcissistic traits (both grandiose and vulnerable ones) with facets of the self, and tested whether these correlations significantly differed from one another (Meng et al., 1992). Based on results from correlation profile comparisons, we carried out multiple regressions to investigate unique associations of those facets of the self that showed to be significantly and specifically correlated with adaptive and pathological narcissistic traits (both grandiose and vulnerable ones). We implemented Bonferroni correction (i.e., a p-value equal to or less than .001 was set) in all the statistical analyses because of the number of significance tests conducted. Associations of PNI grandiose narcissism with facets of the self were computed controlling for PNI vulnerable narcissism both in correlation and regression analyses, as suggested in recent studies (Crowe et al., 2019; Edershile et al., 2019).

Results

Correlation analyses among narcissistic traits reveal that pathological vulnerable narcissism correlates significantly with pathological grandiose narcissism (r = .58, p < .001), but not with adaptive narcissism (r = .02, p = .62), and that pathological grandiose narcissism positively correlates with adaptive narcissism (r = .44, p < .001), after controlling for PNI vulnerable narcissism.

Means and standard deviations of narcissistic traits and facets of the self are in table 1.

Table 1.

Means and Standard Deviations of narcissistic traits and facets of the self

	M	SD	Min	Max
Pathological Grandiose Narcissism	2.55	0.58	0.79	4.29
Pathological Vulnerable Narcissism	2.32	0.70	0.42	4.41
Adaptive Narcissism	3.44	2.49	0.00	11.00
PSU	3.28	0.78	1.00	5.00
NFU_L	3.29	0.54	1.73	4.67
NFU_W	3.20	0.74	1.17	5.00
NFU_D	3.04	0.58	1.55	5.00
AS_SA	3.01	1.50	1.00	7.00
AS_AEI	3.08	1.30	1.00	7.00
AS_AL	5.66	0.92	1.00	7.00
SOSS_L	2.64	0.90	1.00	5.00
SOSS_S	2.54	0.86	1.00	5.00
SOSS_C	2.24	0.80	1.00	4.75
SOSS_T	2.95	1.02	1.00	5.00
Self-Others Agency	-0.47	1.34	-5.17	4.67
Self-Others Communion	-0.01	1.10	-5.33	3.83

N= 539; PSU= Personal Sense of Uniqueness; NFU_L= Lack of concern regarding others’ reactions; NFU_W= Willingness to publicly defend one’s beliefs; NFU_D= Desire not to always follow rules; AS_SA = Self-alienation scale; AS_AEI= Accepting external influence scale; AS_AL= Authentic living scale; SOSS_L= Lack of understanding of oneself; SOSS_S= Sudden shifts in feelings, opinions, and values; SOSS_C= Tendency to confuse one’s feelings, thoughts, and perspectives with those of others; SOSS_T= Feeling of a tenuous existence; Self-Others Agency = Delta score between self- and others-representations in the agentic domain; Self-Others Communion = Delta score between self- and others-representations in the communal domain.

Bivariate correlations of both adaptive and pathological traits of narcissism with facets of the self, and their statistical comparisons, are in table 2.

Table 2.

Correlations among narcissistic traits and facets of the self

		Pathological narcissism		Comparisons among correlations
	Adaptive Narcissism	Grandiosea	Vulnerable	Z Diff. r1 vs r2	Z Diff. r1 vs r3	Z Diff. r2 vs r3
PSU	.44	.32	.01	2.91	7.46	7.96
NFU_L	.40	.27	-.41	3.07	13.66	17.13
NFU_W	.48	.32	-.15	3.93	10.81	11.89
NFU_D	.19	.15	.06	.89	2.18	2.29
AS_AL	.13	.05	-.29	1.76	7.01	8.65
AS_AEI	-.33	-.12	.39	-4.75	-12.10	-13.00
AS_SA	-.20	-.09	.42	-2.44	-10.47	-13.10
SOSS_L	-.27	-.11	.41	-3.58	-11.44	-13.30
SOSS_S	-.05	.01	.34	-1.31	-6.58	-8.49
SOSS_C	-.28	-.14	.36	-3.15	-10.71	-12.69
SOSS_T	-.22	-.10	.54	-2.89	-13.15	-16.48
Self-Others Agency	.40	.25	-.14	3.53	9.13	9.85
Self-Others Communion	-.11	.03	.01	-3.06	-1.99	.50

N = 539; PSU= Personal Sense of Uniqueness; NFU_L= Lack of concern regarding others’ reactions; NFU_W= Willingness to publicly defend one’s beliefs; NFU_D= Desire not to always follow rules; AS_SA = Self-alienation scale; AS_AEI= Accepting external influence scale; AS_AL= Authentic living scale; SOSS_L= Lack of understanding of oneself; SOSS_S= Sudden shifts in feelings, opinions, and values; SOSS_C= Tendency to confuse one’s feelings, thoughts, and perspectives with those of others; SOSS_T= Feeling of a tenuous existence.; Self-Others Agency = Delta score between self- and others-representations in the agentic domain; Self-Others Communion = Delta score between self- and others-representations in the communal domain. Z Diff. r1 vs r2 = Z score difference between correlations of adaptive narcissism and those of pathological grandiose narcissism with facets of the Self; Z Diff. r1 vs r3 = Z score difference between correlations of adaptive narcissism and those of pathological vulnerable narcissism with facets of the Self; Z Diff. r2 vs r3 = Z score difference between correlations of pathological grandiose narcissism and those of pathological vulnerable narcissism with facets of the Self.

^a Correlations between pathological grandiose narcissism and facets of the self are controlled for vulnerable narcissism Bold correlations and z-scores are significant at p < .001 (Bonferroni correction).

Profile of Adaptive Narcissism on facets of the self

Correlations showed that the higher adaptive narcissism the greater personal sense of uniqueness, need for uniqueness, authentic living and Self-others agency. Also, adaptive narcissism was negatively associated with self-consistency scales (with the exception of SSOS_S), Self-others communion, and self-authenticity scales of accepting external influence and self-alienation. All these correlations significantly differed from those of pathological vulnerable narcissism (all p_s < .001), with the exception of correlations with Self-others communion (p = .047) and NFU_D (p = .029). Conversely, results showed that only correlations of adaptive narcissism with facets of NFU_W, AS_AEI, SOSS_L, and Self-others agency significantly differed from those of pathological grandiose narcissism (all p _s < .001).

Profile of Pathological Grandiose Narcissism on facets of the self

Pathological grandiose narcissism significantly correlated only with greater personal sense of uniqueness, need for uniqueness scales, and Self-others agency. All these correlations significantly differed from those of pathological vulnerable narcissism (all p_s < .001), except for NFU_D (p = .022). Moreover, only correlations of pathological grandiose narcissism with willingness to publicly defend one’s beliefs (i.e., need for uniqueness scale) and Self-others agency significantly differed from those of adaptive narcissism (range p_s < .001).

Profile on facets of Pathological of the self Vulnerable Narcissism

Correlations showed that pathological vulnerable narcissism is significantly associated with several facets of the self. Specifically, the higher vulnerable narcissism the lower levels of lack of concern about others’ reactions, willingness to publicly defend one’s beliefs, authentic living, and Self-others agency. Moreover, pathological vulnerable narcissism showed to be negatively associated with accepting external influence, self-alienation, and self-consistency scales. All these correlations significantly differed from those of both adaptive narcissism (all p_s < .001) and pathological grandiose narcissism (all p_s < .001).

Unique associations of facets of the self with narcissistic manifestations

Unique associations of facets of the self that showed to be significantly and distinctively associated with adaptive and pathological narcissistic traits (both grandiose and vulnerable ones), according to correlational comparisons profiles, are in table 3.

Table 3.

Multiple regression among narcissistic traits and facets of the self

		R2	F	β	t	P
Adaptive Narcissism	Model	.38	32.32			< .001
	PSU			.29	7.77	< .001
	NFU_L			.07	1.32	.187
	NFU_W			.27	6.22	< .001
	AS_AL			-.10	-2.58	.010
	AS_AEI			-.03	-.59	.556
	AS_SA			.02	.34	.737
	SOSS_L			-.06	-1.08	.281
	SOSS_C			-.07	-1.31	.191
	SOSS_T			-.03	-.55	.583
	Self-Others Agency			.16	4.02	< .001
Grandiose Narcissism	Model	.45	85.69			< .001
	PSU			.18	5.05	< .001
	NFU_L			.05	1.14	.255
	NFU_W			.16	3.91	< .001
	Self-Others Agency			.08	2.22	.027
	NV			.62	17.46	< .001
Vulnerable Narcissism	Model	.37	31.48			< .001
	NFU_L			-.24	-4.86	< .001
	NFU_W			.08	1.82	.069
	AS_AL			-.06	-1.46	.145
	AS_AEI			.06	1.19	.234
	AS_SA			.12	2.15	.032
	SOSS_L			.02	.27	.785
	SOSS_S			.06	1.25	.210
	SOSS_C			-.08	-1.44	.152
	SOSS_T			.38	8.51	< .001
	Self-Others Agency			.05	1.26	.209

N = 539; PSU= Personal Sense of Uniqueness; NFU_L= Lack of concern regarding others’ reactions; NFU_W= Willingness to publicly defend one’s beliefs; AS_SA = Self-alienation scale; AS_AEI= Accepting external influence scale; AS_AL= Authentic living scale; SOSS_L= Lack of understanding of oneself; SOSS_S= Sudden shifts in feelings, opinions, and values; SOSS_C= Tendency to confuse one’s feelings, thoughts, and perspectives with those of others; SOSS_T= Feeling of a tenuous existence.; Self-Others Agency = Delta score between self- and others-representations in the agentic domain. F values and β scoresare significant at p < .001 (Bonferroni correction).

Personal sense of self, willingness to publicly defend ones’ beliefs, and Self-others agency were uniquely and positively associated with adaptive narcissism. Personal sense of uniqueness and willingness to publicly defend ones’ beliefs were uniquely and positively associated also with pathological grandiose narcissism. Unique associations of feeling of a tenuous existence (i.e., positive association) and lack of concern regarding others’ reactions (i.e., negative association) with pathological vulnerable narcissism were found.

Discussion

The present study demonstrated that adaptive and pathological manifestations of narcissism can be captured by distinctive profiles on facets of the self-concept. However, results also showed that adaptive and pathological manifestations of narcissism, especially those reflecting narcissistic grandiosity, may share some attributes of the self.

One shared attribute of self-concept in individuals with prevailing traits of adaptive and pathological expressions of grandiose narcissism is a greater sense of self-uniqueness. Van Doeselaar et al. (2019) showed that personal sense of self was uniquely associated with narcissistic admiration strategies which, in turn, are linked to both NPI (Back et al., 2013) and PNI grandiose narcissism (Krizan & Herlache, 2018). Our study extends these findings and suggests that having a greater sense of uniqueness represents an essential aspect of narcissistic grandiosity, regardless of its adaptive or pathological nature. In fact, personal sense of uniqueness was positively associated with both adaptive and pathological manifestations of grandiose narcissism. Moreover, this dimension was uniquely related to adaptive and pathological expressions of grandiose narcissism, above and beyond other facets of the self that contributed to profile self-concept in such manifestations. In line with this interpretation and with our expectations, vulnerable narcissism did not relate at all to personal sense of self. Personal sense of uniqueness showed to be associated with indicators of subjective well-being such as life satisfaction and positive strength (Şimşek & Yalınçetin, 2010). Interestingly, both adaptive grandiose narcissism (Rohmann et al., 2019) and pathological grandiose narcissism (Kaufman et al., 2020) have shown positive associations with life satisfaction, while vulnerable narcissists usually repot poor life satisfaction (Kaufman et al., 2020). In this sense, our findings suggest that personal sense of uniqueness could be responsible for high levels of life satisfaction in individuals showing narcissistic grandiosity through adaptive or pathological expressions. Albeit plausible, however, this hypothesis should be tested empirically.

Consistent with previous studies showing that NPI (Emmons, 1984; Pincus et al., 2009) and PNI grandiose narcissism (Di Sarno et al., 2020; Pincus et al., 2009) were linked to interpersonal dominance and low levels of shame and social anxiety, we found that both adaptive and pathological expressions of grandiose narcissism were positively associated with proneness to defend ones’ own beliefs openly. Moreover, this facet of the self was uniquely associated with both narcissistic manifestations, above and beyond other facets of the self that were distinctive of these manifestations. In other words, searching for public exposure would be one of the essential features of self-concept of individuals with high levels of narcissistic grandiosity. It is of note, however, that individuals reported higher proneness to defend publicly their beliefs when showing adaptive manifestations of grandiose narcissism than pathological ones. These findings can be explained by the fact that, even though both pathological and adaptive grandiose narcissism share assertive and dominant interpersonal tendencies (Jakšić et al., 2014; Pincus et al., 2009), pathological grandiosity also includes covert manifestations that can be a source of shame feelings (Di Sarno et al., 2020). Conversely, pathological vulnerable narcissism showed to be negatively (and weakly) associated with this facet of the self. This result is consistent with findings on proneness to social avoidance (Pincus et al., 2009) in vulnerable narcissists.

Since pathological grandiose narcissists have entitled attitudes (Thomas et al., 2012), we expected that traits of pathological grandiose narcissism would make people prone to express great need for uniqueness through other strategies as well. For instance, we expected a strong link between pathological grandiose narcissism and desire not to follow the rules. Contrary to our expectations, however, neither pathological nor adaptive manifestations of grandiose narcissism were significantly related with this facet of the self. Pathological vulnerable narcissism was not associated with desire not to follow the rules as well. According to Schumpe et al. (2016), the desire not to always follow rules subscale “represents the willingness to take risks or seek stimulation” (p. 235) and describes individuals who are prone to break the rules and to behave riskily to feel unique. Martínez-López et al. (2019) have recently found that narcissistic patients do not differ from controls in levels of novelty seeking. In this sense, both our results and those of Martínez-López et al. (2019) suggest that trying to feel unique and different from others through novelty-seeking strategies is not typical of individuals with narcissistic tendencies at all.

Clinical observations (Caligor et al., 2015) and empirical findings (Miller et al., 2011) indicate that grandiose narcissists feel superior to others, while vulnerable narcissists would be prone to feelings of inferiority (Pincus & Lukowitsky, 2010; Ronningstam, 2009). Empirical research on this topic, however, is limited. The present study suggests that self-concept in grandiose narcissism, regardless of its adaptive or pathological nature, is better expressed by themes of dominance, assertiveness, and performance, rather than of connection and relatedness. Both adaptive and pathological grandiose narcissism showed to correlate significantly with proneness to describe oneself as more agentic than significant others. These findings are in line with recent studies (Mielke et al., 2020; Seidman et al., 2020) showing that agentic self-perception and agentic self-enhancement are specifically associated with narcissistic admiration strategies (Back et al., 2013) which, in turn, are strongly linked to both NPI narcissism (Back et al., 2013) and PNI grandiose narcissism (Krizan & Herlache, 2018). Our study clarified, however, that sense of superiority in the agentic domain has a prominent role in defining self-concept of people showing narcissistic grandiosity only when considering its adaptive manifestations. In fact, sense of superiority over others in the agentic domain showed to be uniquely associated with adaptive narcissism above and beyond other facets of the self that describe this narcissistic manifestation. Conversely, sense of superiority in the agentic domain was positively linked to pathological grandiose narcissism, but not uniquely associated with it when considering the effects of other facets of the self. In order words, grandiose narcissists tend to feel superior over others in the agentic domain, but this sense of superiority can be considered an essential attribute of the self only in adaptive narcissists. When looking at individuals high in pathological vulnerable narcissism, deflated self-representations can be found. More specifically, the study shows that pathological vulnerable narcissism was associated with proneness to describing oneself as less agentic than significant others. This result is in line with Schalkwijk et al. (2021), suggesting that vulnerability would be expressed by an overtly devalued sense of self. This result is also consistent with findings on recurrent feelings of inferiority (Pincus & Lukowitsky, 2010) and shame (Di Sarno et al., 2020) in vulnerable narcissists, and clarifies that these feelings of inferiority might depend primarily on agentic attributes, rather than on communal ones. Taken together, our findings suggest that only agentic attributes, but not communal ones, are relevant for narcissistic self-appraisal.

Furthermore, the study shows that individuals high in narcissistic grandiosity, regardless of its adaptive or pathological nature, are not concerned by others’ reactions. This result is in line with clinical observations showing that grandiose narcissists are disinterested in others (Caligor, 2013) and with psychodynamic theories suggesting that grandiose narcissism is linked to instrumental use of others (Schalkwijk et al., 2021). Moreover, studies on both NPI narcissism (e.g. Dickinson & Pincus, 2003) and PNI grandiose narcissism (Pincus et al., 2009) showed positive associations with social boldness and lack of sensitivity to criticism. Unlike narcissistic grandiosity, pathological vulnerable narcissism was associated to greater concern regarding others’ reactions. This result confirms previous empirical evidence on the link between narcissistic vulnerability and interpersonal hypersensitivity (Hendin & Cheek, 1997; Miller et al., 2011). Accordingly, our study demonstrates the prominent role of this facet in defining self-concept of vulnerable narcissists. In fact, greater concern about others’ reaction was uniquely related to pathological vulnerable narcissism, above and beyond other facets of the self that were distinctive of this narcissistic manifestation

Another relevant attribute of self-concept in narcissistic manifestations is (in)consistency in the sense of self. In a sense, our findings suggest this facet to be particularly suitable for differentiating profiles of individuals with adaptive and pathological manifestations of narcissism. In fact, the negative correlations we found revealed that individuals high in adaptive grandiose narcissism have a strong and coherent sense of self. Conversely, individuals high in vulnerable narcissism have greater instability of the self, proneness to confound their own feelings, thoughts, and perspectives with those of others, and to feel fragile. Moreover, pathological vulnerable narcissism was linked to greater difficulties in understanding oneself, and feelings of a tenuous existence were found to be central in vulnerable narcissists’ sense of self. In fact, this facet of self-consistency was uniquely related to vulnerable narcissism above and beyond other the other facets of self-concept. Our results confirm past findings showing that pathological vulnerable narcissism relates to a weak sense of self (e.g., Kaufman et al., 2020) and lack of awareness and clarity about subjective emotional experiences (Di Pierro et al., 2017). Moreover, they are consistent with well-established evidence which pointed out that vulnerable individuals suffer from emotional dysregulation (Di Pierro et al., 2017; Di Sarno et al., 2020), have high neuroticism levels (Miller et al., 2011), and suffer from poor self-esteem (Miller et al., 2017; Pincus et al., 2009) and global distress (Kaufman et al, 2020; Miller et al., 2007). Our study also clarifies results from Kaufman et al. (2020), showing that PNI grandiose narcissism is not linked at all to self-inconsistency, when considering coexisting traits of vulnerable narcissism. One possible interpretation for this result is that the grandiose self-structure would compensate for lack of identity integration in patients with narcissistic pathology (Caligor, 2013). However, since pathological traits of grandiose narcissism are linked to strong self-enhancement tendencies (Pincus et al., 2009), and our study is based on self-report measures, we cannot exclude that this inclination might have led to strongly biased responses. Hence, future studies could use a multi-method approach to measure facets of the self (e.g., behavioral and implicit measures) in order to disentangle our results.

Authenticity of the self showed to be particularly problematic in pathological vulnerable narcissism. In fact, individuals high in pathological vulnerable narcissism were prone to experiencing feelings of self-alienation, and conforming to expectations of others. Moreover, they showed difficulties in behaving consistently with their psychological states. Neither of these aspects of self-authenticity, however, showed to be essential aspects of the self in vulnerable narcissists when considering other facets that were significantly and distinctively correlated with this narcissistic manifestation. In other words, our findings suggest that individuals with pathological vulnerable narcissism present lack of self-authenticity, but this attribute does not have a prominent role in the context of other aspects of the self that are characteristics of such individuals. Unlike pathological vulnerable narcissism, lack of authenticity does not significantly contribute to describe sense of self in individuals with adaptive or pathological narcissistic grandiosity. Kaufman et al. (2020) showed that PNI grandiose narcissism was significantly associated with self-alienation and proneness to conform to others’ expectations. Our study, however, demonstrated that coexisting traits of vulnerable narcissism were responsible for these associations.

Our results can be better understood in light of some study’s limitations. The present study involved a larger sample of participants, compared with past studies (Freis & Fujita, 2017; Kaufman et al., 2020; Van Doeselaar et al., 2019), but it was gender unbalanced. Since expressions of narcissism (Wright et al., 2010) and facets of the self (i.e., sense of uniqueness; Lalot et al., 2017) are both influenced by gender, future studies should involve balanced samples for a better consideration of gender differences and effects. Moreover, future studies could use a multi-method approach to measure facets of the self in order to disentangle our results. In fact, we only employed self-report measures, and pathological traits of grandiose narcissism have shown to be linked to strong self-enhancement tendencies (Pincus et al., 2009). In this sense, we cannot exclude that our findings might be biased by this inclination. Finally, we found slight differences in facets of the self when comparing pathological and adaptive grandiose narcissism. These findings may suggest that adaptive or pathological expressions do not affect the way individuals high in narcissistic grandiosity describe themselves. The lack of significant differences, however, might be explained also by the specific facets we considered. In other words, we may have neglected some facets of the self specifically related to pathological grandiosity, but not to adaptive one, as in the case of overconfidence tendencies and biased self-appraisal (Campbell et al., 2004; Gabriel et al., 1994).

Conclusion

The existing literature has not provided a coherent and exhaustive understanding of self-concept in narcissism until now. This has been mainly due to methodological and measurement issues that have not allowed researchers to consider the heterogeneity (and specificity) of narcissistic manifestations. Our study shows that expressions of the self may vary significantly according to individuals’ prevailing manifestations of narcissism. In this sense, our findings do not support the idea that pathological narcissism is linked to a typical “narcissistic self”, namely a grandiose self-structure that gives coherence and stability to individuals’ sense of self, as often suggested by clinicians (Caligor, 2013). We found that adaptive and pathological expressions of narcissism, as well as grandiose and vulnerable manifestations of pathological narcissism, entail different and unique features of self-concept. These findings have some potential clinical implications. According to Huflejt-Łukasik et al. (2015) psychotherapeutic treatments always lead to structural changes in the sense of self of patients (i.e., self-definition). However, patients with a proper narcissistic pathology, or those with strong pathological narcissistic tendencies, may pose great challenges in treatment (Zalman et al., 2019): these patients usually show lack of motivation for treatment, have difficulties in establishing therapeutic alliance, and often interrupt treatment suddenly. Our findings suggest that individuals with prevailing traits of grandiose pathological narcissism and those having prevailing traits of vulnerable pathological narcissism present specific and unique impairments in the sense of self. Therefore, a careful evaluation of aspects qualifying the sense of self in patients with pathological narcissism may help clinicians to have a more detailed picture of their patients’ functioning and to use this information to establish clinical intervention targets. The study has also relevant implications in terms of empirical research. The study suggests indirectly that researchers should evaluate carefully which measurement instruments of narcissism they want to use, according to the aims of their studies, and they should make clear what type of narcissistic expressions (and measure) they are referring to when discussing results. Only by overcoming these biases, an increasingly integrated, detailed, and comprehensive understanding of narcissism and its psychological and behavioral correlates will be possible.