“It’s not you, it’s me”: identity disturbance as the main contributor to interpersonal problems in pathological narcissism

Background Core impairments in self and other functioning typify individuals with personality disorder. While interpersonal dysfunction is a known element of narcissistic disorders, empirical research investigating intrapersonal elements is lacking. The aim of this study was to investigate the internal representations of individuals with grandiose and vulnerable features, as manifested through their attachment styles, and the specific role of identity disturbance in explaining the relationship between pathological narcissism and maladaptive interpersonal functioning. Methods A sample of 270 university students completed the Brief Pathological Narcissism Inventory (B-PNI), the Severity Indices of Personality Problems (SIPP), the Relationship Questionnaire (RQ), and the Inventory of Interpersonal Problems (IIP-32). Results Both vulnerable and grandiose narcissism were positively associated with both fearful and preoccupied attachment, and negatively associated with secure attachment, whilst grandiose narcissism was also positively associated with dismissive attachment. Furthermore, unstable representations of self, poor self-reflective functioning, and low sense of purpose fully mediated the relationship between interpersonal problems and grandiose narcissism while partially mediating the relationship between interpersonal problems and vulnerable narcissism. Conclusions Overall, our findings suggest that for individuals presenting with narcissistic features, capacity for adaptive interpersonal functioning is grounded by deficits in identity integration. Implications of these findings are discussed. Supplementary Information The online version contains supplementary material available at 10.1186/s40479-022-00209-6.

The current paper therefore aims to address this gap by investigating how internal relational representations underlying pathological narcissism -and unstable representations of self, more specifically -may explain the relationship between pathological narcissism and maladaptive interpersonal functioning.

Pathological narcissism: a multifaceted construct
Pathological narcissism is commonly defined by significant self-regulatory deficits, a pervasive sense of entitlement, and maladaptive ways of defending against threats to a positive self-image [72]. It is also multifaceted in nature, with existing empirical literature consistently highlighting the existence of both grandiose and vulnerable phenotypes [28,67,84]. Grandiosity and vulnerability may each be expressed either overtly or covertly, with some individuals primarily displaying one or the other of these facets [66]. There is however growing consensus, that most narcissistic patients present with some degree of fluctuation between grandiose and vulnerable selfesteem states [23,35,66,35,73]. Although it has been noted that grandiose and vulnerable expressions of narcissism may represent different clinical manifestations of the disorder rather than discrete subtypes [44], the two phenotypic expressions of narcissistic pathology have shown both diverging and converging associations with a range of constructs, including interpersonal dysfunction (e.g., [21,52,74,86]). Contemporary clinical conceptualizations of narcissistic pathology [44,51,62,68,72] thus suggest that grandiose and vulnerable expressions of narcissism are separable yet related expressions that manifest in distinct experiential and behavioural patterns (e.g. being self-assured and dominant vs. self-conscious and withdrawn when interacting with others), as reflected in recent empirical research [17]. Despite their seemingly contrasting presentations, both share (overtly or covertly expressed) attitudes of being special and entitled to special treatment [37,51].

Self and interpersonal functioning in pathological narcissism
Grandiose narcissism is typically characterized by attention seeking, arrogance, and self-enhancing and exaggerated self-worth. These individuals have been found to be interpersonally domineering, antagonistic and vindictive [14,74,75,84]. They can also be socially charming despite being oblivious of others' needs, interpersonally exploitative, and envious [55]. In contrast, the vulnerable type is characterized by defensiveness, withdrawal, emotional dysregulation, and hypersensitivity to others' evaluations [38]. Gabbard [28] described these individuals as shy and "quietly grandiose, " with an "extreme sensitivity to slight, " which "leads to an assiduous avoidance of the spotlight. " Subsequent research has supported these clinical descriptions, with both types found to share a core of entitled and dismissive tendencies [30,52]. It has been suggested that narcissistic individuals regulate their self-esteem through interpersonal dominance, denial of any form of reliance or dependency on others, and devaluation of others for not meeting expectations or for threatening their sense of superiority [13,29,64]. In this regard, Robbins and Dupont [70] describe a vicious cycle in which the problematic interpersonal behaviour of the narcissistic individual leads to a breakdown of interpersonal relations, followed by the accompanying reinforcement of these same problematic behaviours that are used to reinforce a fragile sense of self. These conflictual relationship patterns have been directly identified in empirical research, examining the moment-to-moment interactions between individuals with pathological narcissism and their partners or family members [18]. Similarly, Kernberg [44] argued that narcissistic individuals hold contradictory views of the self, which vacillate between the clinical expression of grandiose and vulnerable symptoms. Thus, the overtly narcissistic individual most frequently presents with grandiosity and exhibitionism. However, when the needs for admiration and idealized expectations of these individuals are not met, they become depressed, depleted, and feel painfully inferior. Pincus and colleagues have provided empirical backing to this understanding, supporting both the grandiose strivings toward self-enhancement as well as the expressions of vulnerability (e.g. social withdrawal and emotional dysregulation) following the painful disappointment of losses and failures [12,67].
More recently, significant associations between narcissistic features and identity instability have also been established, with narcissism being linked with inflated but unstable representations of self [19,22,65,71]. For example, Di Pierro and colleagues (2018) found identity instability to partially mediate the association between narcissistic traits and emotional empathy, highlighting the crucial role of internal representations of self when assessing severity of narcissistic functioning. This is in line with existing theoretical and clinical contributions that suggest that distorted internal representations of self and others contribute to the problematic attitudes and interpersonal behaviours of narcissistic patients. For example, representations of self as inferior and of others as potentially rejecting or domineering have been theorized to promote interpersonal withdrawal and avoidance [3,43,61], as reflected in recent empirical contributions [25,85]. Similarly, a grandiose but unstable self-representation is believed to nurture exploitative interactions aimed at obtaining admiring responses from others, the lack of which may in turn foster rage and aggression [35,49]. However, despite extensive support for the association between narcissistic pathology and interpersonal dysfunction as well as existing theoretical contributions, the specific mechanisms underlying this association are often overlooked in the empirical literature.

Attachment and narcissistic pathology: internal working models of self and other
A useful framework and research paradigm for investigating the impact of internal representations of self and others on narcissistic interpersonal functioning is attachment theory. Attachment theory postulates that early life experiences have an enduring effect on personality development through the formation of cognitive-affective templates of self and other, termed internal working models [10]. These internal working models are believed to develop through repeated interactions with early caregivers and can be either positive or negative. For example, repeated benevolent experiences are reflected over time in the emergence of secure attachment patterns, characterized by positive models of self and others. On the other hand, inconsistent experiences of care or consistent rejection, neglect, or abuse, are thought to give rise to more problematic internal working models, in which others are viewed as emotionally unavailable or malevolent, and the self as unworthy of attention and care [59].
Internal working models have been operationalized through attachment styles, which reflect the relative security or insecurity of individuals' attachment relationships [7,54]. Insecure attachment has been associated with patterns of maladaptive personality functioning [11,83], and has been identified as a risk factor for the development of narcissistic and borderline personality disorders [6,57,59]. Specifically, insecure attachment is typically classified into patterns of attachment anxiety and attachment avoidance, operationalized as either preoccupied/anxious (negative model of self; positive model of others), fearful/avoidant (negative models of self and others), or dismissive-avoidant (positive model of self; negative model of others) attachment styles. Attachment anxiety reflects fears of rejection and needs for reassurance in close relationships, with a tendency to feel distressed in response to perceived or anticipated separations. Attachment avoidance reflects discomfort with dependency and closeness, with tendencies toward detachment from relationships [11]. As a result, from an attachment perspective, personality dysfunction is understood to consist largely of the repeated activation of insecure attachment patterns, which reflect internal models of the self and others [58].
Despite their potential utility for improving our understanding of narcissistic pathology, relatively few studies have investigated the role of internal representations of self and others in the context of pathological narcissism. Moreover, the study of narcissism using an attachment paradigm has produced somewhat mixed findings. For example, Kaufman and Jauk [40] found no link between attachment and grandiose narcissism, while other research has associated grandiosity with avoidant and dismissive attachment [21,87]. In contrast, vulnerable narcissism appears to be more consistently associated with insecure (anxious or avoidant) attachment, along with a negative working model of self [27,40], or both [11,21,69,80,87].

Current study
Considering the well-known deficits in interpersonal functioning in narcissistic pathology and the theorized role of unstable representations of self and others, the present study aimed to investigate whether grandiose and vulnerable features of pathological narcissism were associated with specific deficits in identity integration and attachment styles that contribute to interpersonal dysfunction, as well as to test the hypothesis that deficits in identity integration may indirectly explain the relationship between pathological narcissism and associated interpersonal problems. To our knowledge, no other research has investigated the role of identity disturbance as the underlying mechanism responsible for the association between narcissistic features and broader interpersonal dysfunction.

Participants and procedure
Two hundred seventy first-year university students aged between 18 to 25 years (M = 19.48, SD = 1.732) took part in the current study. Participants were provided course credit for their participation. Amongst them, 33 were male (12.22%) , 236 were female (87.41%) , and one declared as non-binary (0.37%) . All participants were English speaking. 103 participants (38.15%) reported having at least one previously diagnosed mental health condition and 147 participants (54.44%) reported having at least one immediate relative with a previously diagnosed mental health condition. Participation was voluntary and consent was obtained after reviewing the study's information found in the participant information sheet. The study was approved by the University's Human Research Ethics Committee.

Brief-Pathological Narcissism Inventory (B-PNI)
The B-PNI [76] is a 28 item self-report questionnaire that assesses seven facets of pathological narcissism: Exploitativeness,Self-sacrificing self-enhancement; Grandiose fantasy; Contingent self-esteem; Hiding the self; Devaluing; and Entitlement rage. The B-PNI requires participants to rate the items based on a 5-point scale, ranging from 0 (not at all like me) to 4 (very much like me), with higher scores reflecting higher levels of narcissism. Within the current study, the B-PNI had high internal reliability (α = .91) . Separately, both narcissistic grandiosity and vulnerability reported high reliabilities of α = .84 , and α = .88 respectively.

Severity Indices of Personality Problems-118 (SIPP-118)
The SIPP-118 [81] is a dimensional self-report measure of the core components of (mal)adaptive personality functioning, and provides indices of the severity of personality pathology. In this study, only the Identity Integration domain was used, consisting of 36 items and covering five distinct facets: Self-respect (e.g. "I often feel that I am not as worthy as other people"); Stable self-image (e.g. "I am often confused about what kind of person I really am"); Self-reflexive functioning (e.g. "I am often not fully aware of my inner feelings"); Enjoyment (e.g. It is hard for me to really enjoy things); and Purposefulness (e.g. I often feel that my life is meaningless). SIPP-118 requires participants to rate the items based on a 4-point scale, ranging from 1 (fully disagree) to 4 (fully agree), with higher scores reflecting greater degrees of identity disturbance. Within the current study, the Identity Integration domain had high internal reliability (α = .93) . Separately, self-respect, stable self-image, self-reflexive functioning, enjoyment, and purposefulness all reported high reliabilities of α = .88 , α = .81 , α = .78, α = .85 , and α = .78 respectively.

Relationship Questionnaire (RQ)
The RQ consists of four short paragraphs, each describing a prototypical attachment pattern as it applies to close relationships in adulthood. The four attachment styles (i.e. secure, fearful, preoccupied, and dismissing) are each represented by a single paragraph, which is rated "Strongly Disagree" (1) to "Strongly Agree" (7). Participants are then asked to rate their agreement with each prototype on a 7-point scale. The highest of the four attachment prototype ratings is then used to classify participants into an attachment category. The RQ is widely used and has demonstrated sound psychometric properties [5,33].

Inventory of Interpersonal Problems (IIP-32)
The 32-item version of the IIP [4] was used to assess eight facets of interpersonal problems: Hard to be sociable, Hard to be assertive; Too aggressive; Too open; Too caring; Hard to be supportive; Hard to be involved; and Too dependent. IIP-32 requires participants to rate the items based on a 5-point scale, ranging from 0 (not at all) to 4 (extremely), with higher scores reflecting higher levels of interpersonal problems. In the current study, the IIP-32 had high internal reliability (α = .89).

Statistical analyses
Preliminary and main analyses were conducted using Statistical Package for the Social Sciences (SPSS Version 25.0; [36]) software and Amos . Direct associations between levels of narcissism, attachment styles, identity disturbance, and interpersonal functioning were analysed using bivariate correlations. Mediation and moderated mediation analyses were then conducted to test for the mediating role of identity disturbance in the relationship between narcissism and interpersonal functioning using structural equation modelling (SEM).

Descriptives, correlations and between-group differences
Preliminary analyses were conducted to determine the confounding effects of age, gender, and prior personal and family history of mental illness on the variables of interest. No significant effects of age were found through bivariate correlations. Meanwhile, significant differences between males and females were observed only on the grandiose narcissism variable, with male participants reporting significantly higher scores than female participants (Cohen's d = 0.45). However, this difference was expected considering that males are known to exhibit higher levels of grandiosity compared to females [34]. Hence, gender was not controlled for in the main analyses. Finally, individuals with either prior personal or family history of mental illness scored significantly higher for interpersonal problems and identity integration. Thus, both forms of prior history of mental illness were controlled for in subsequent analyses. Bivariate correlations were conducted to determine the direction and strength of the linear relationships between all of the main variables. As shown in Table 1, grandiose narcissism was positively correlated with fearful, preoccupied, and dismissive attachment, while vulnerable narcissism was positively correlated with fearful and preoccupied attachment and negatively correlated with secure attachment. With the exception of one facet of grandiose narcissism (exploitativeness), pathological narcissism was associated with deficits on all five domains of identity integration. Similarly, with the exception of exploitativeness, pathological narcissism as well as identity disturbance and attachment all correlated positively with interpersonal problems. Hence, all variables were included in the subsequent SEM analyses.
Moreover, One-Way Analyses of Variance (ANO-VAs) were conducted to determine differences in means across the different attachment styles for each   Table 2. Overall, there were significant differences across the groups in terms of identity impairment, interpersonal functioning and vulnerable narcissism, but no significant differences relating to grandiose narcissism. Post-hoc comparisons revealed that individuals with a secure attachment scored significantly lower on vulnerable narcissism, identity integration, and interpersonal problems when compared to their fearful/avoidant and anxious/preoccupied counterparts. These same individuals also scored significantly lower on identity integration and interpersonal problems when compared with those with a dismissive attachment. Similarly, those who endorsed a dismissive attachment style also scored significantly lower on vulnerable narcissism, identity integration, as well as interpersonal problems than those with a fearful/avoidant attachment.

Structural equation modelling
SEM analyses were conducted to investigate the mediating effect of identity integration on the relationship between narcissism and interpersonal problems as illustrated in Fig. 1. In line with existing recommendations, mediation analyses were conducted via bootstrapping, allowing to estimate the confidence intervals of the indirect effects [77]. 5000 bootstrapped samples were used with a 95% confidence interval for each indirect effect. Results of SEM can be found on Table 3.
First, after accounting for the effects of prior personal and family history of mental illness, results showed personal history to have a significant relationship with self-respect ( β = .189, p < .001 ), stable self-image ( β = .119, p < .05 ), and purposefulness ( β = .189, p < .001 ) while family history was significantly related to self-reflexive functioning ( β = .155, p < .01 ). Neither forms of history were associated with interpersonal problems.    The mediation analysis showed both narcissistic grandiosity and vulnerability to have significant indirect effects on interpersonal problems through the various facets of identity integration, namely, stable self-image, self-reflexive functioning, and purposefulness. Goodness of fit indices indicated excellent fit for the overall model: . The direct effects of narcissistic vulnerability on interpersonal problems remained however significant, suggesting a partial mediation.
As we were also interested in the potential role of representations of others (not just of self ), the model was expanded to include the attachment variable. A moderating mediation analysis was therefore conducted with attachment moderating the association between identity integration and interpersonal problems (see Figure S1). Results however revealed a nonsignificant moderating effect while the mediating effect remained significant. Fit indices for the second model

Discussion
The present study aimed to investigate the association between pathological narcissism and specific deficits in identity integration and attachment styles that contribute to interpersonal dysfunction, as well as to test the hypothesis that deficits in identity integration may indirectly explain the relationship between pathological narcissism and associated interpersonal problems. Significant associations were found between deficits in identity integration, attachment styles, interpersonal problems, and both grandiose and vulnerable manifestations of pathological narcissism. Notably, different facets of grandiose narcissism were associated with all three types of insecure attachment styles, i.e. preoccupied, fearful, and dismissive. From an attachment perspective, this would suggest that grandiose narcissism involves both (split-off ) positive and negative representations of self, and both (split-off ) positive and dismissing attitudes toward others. Although grandiose narcissism has traditionally been associated with attachment avoidance and dismissal of the need for intimacy and closeness, our results also indicate an association with representations of self as inadequate, fearful, and sensitive to rejection. This suggests that a vulnerable experience of self is also a salient manifestation of narcissistic grandiosity. Despite mixed empirical results supporting these results [9,18,40,42], this association is in line with existing theories of pathological narcissism, such as those advanced by Kernberg [43] and [48,50], which argue that narcissistic grandiosity essentially covers and protects an implicitly negative self-representation and underlying feelings of inferiority and neediness. Subsequent theorists have also noted that narcissistic grandiosity is closely intertwined with external validation [66] despite fears and/or dismissal of dependency on others [15].
Similarly, vulnerable narcissism was mainly associated with both fearful and preoccupied attachment styles, indicating predominantly a negative model of self. Vulnerable narcissism was also negatively related to attachment styles denoting a positive model of self (i.e., dismissive and secure attachment). The association between vulnerability and both fearful and preoccupied attachment styles is in line with findings from Fossati and colleagues (2015) indicating that individuals high in narcissism are simultaneously anxious about maintaining attachments and reluctant to trust others. Moreover, individuals who were identified as having a preoccupied or fearful attachment style reported significantly higher scores on vulnerable narcissism as well as most severe deficits in identity integration and interpersonal functioning. These results are consistent with previous findings [21,27,42,69,78,80,87] and extend the importance of negative internal representations of self in relation to narcissistic functioning.
Moreover, results obtained from the subsequent mediation analysis lent additional support to the crucial role identity disturbance plays in explaining the relationship between pathological narcissism and overall interpersonal problems. In particular, deficits in identity integration partially mediated the relationship between vulnerable narcissism and interpersonal problems, while fully mediating the association between the latter and grandiose narcissism. Interestingly, the same facets of identity integration were found to play a significant role in these associations for both grandiose and vulnerable narcissism, namely Stable selfimage, Self-reflexive functioning, and Purposefulness. These findings suggest that unstable self-representations, poor self-reflective functioning and lack of purpose in life are indirectly responsible for interpersonal difficulties reported in individuals presenting with narcissistic features. This in line with contemporary psychodynamic theories (e.g. [46], where identity diffusion (i.e. an unstable and unintegrated sense of self and of others) is understood as the root of borderline and narcissistic pathology, underlying many of the interpersonal problems that these individuals experience [45]. For example, identity diffusion is fundamentally characterized by problems with self-other boundaries, which may be expressed in fluctuations between extremely idealized and extremely devalued representations of self and others, leading to the typical interpersonal dynamics and difficulties reported in narcissistic patients. Similarly, identity diffusion has been linked to impairments in mentalizing, i.e. difficulties in reflective functioning [26]. Although studies on narcissism and reflective-functioning are sparse, existing studies have found individuals with NPD to have similar mentalizing deficits to those found in BPD (e.g. [20]. Our findings are in line with these results and suggest that deficits in self-reflective functioning may have an indirect impact on interpersonal problems in these individuals. This is also not surprising considering that the capacity to make sense of one's internal experiences is key in our ability to navigate the social world [56]. As for the significant association between purposefulness and interpersonal problems, one explanation for this might be the potential shame and envy experienced by individuals with narcissistic features when facing the reality of failed long-term objectives. Items in the Purposefulness facet also reflected a lack of clear goals in life (e.g. 'One of my problems is that I lack clear goals in , which may also be a by-product of an unstable and incoherent sense of self. Of note, the associations between grandiose narcissism and both identity disturbance and interpersonal problems were negative in the SEM analysis -despite positive initial correlations between these variables. This may be explained by the presence of a suppression effect [82], which occurs when an independent variable contributes significantly to the tested model despite weak initial association with the dependant variable, i.e. despite being a weak predictor on its own. One interpretation of this negative association may be explained as a byproduct of overall healthy personality functioning in participants. That is, by virtue of examining a non-clinical population, the participants in this study were relatively high functioning and as such the grandiosity expression under examination may be relatively adaptive or 'healthy' . Such 'healthy' dimensions of narcissism have been historically theorised [47] as a normal or even essential part of psychological functioning. Such healthy or adaptive narcissism may include the capacity to maintain a positive self-image in the face or setbacks, or as reflecting a realistic sense of self-confidence and self-worth that promotes agency and autonomy -indeed, empirical evidence points to the presence of such a "high functioning/ exhibitionistic narcissist" subtype [74]). However, as the narcissism measure utilised in this research captures an explicitly pathological expression, it is unlikely that the grandiosity captured in this study reflects such adaptive or 'healthy' narcissism. Instead, the negative association between grandiosity and psychopathology may instead reflect the image distorting defensive processes implicit in narcissistic disorders. That is, participants invested in a grandiose self-representation may under-report their degree of psychopathology in an effort to maintain their sense of superiority. Indeed, research has demonstrated that individuals with grandiose narcissism had less consistent and weaker self-reported indices of psychopathology, and instead were positively correlated with adaptive coping, life satisfaction, and image-distorting defense mechanisms [41]. That is, while pathological narcissism confers a defensive, self-protective element -this may in fact be a distortion of actual levels of functioning. Alternatively, such defensive processes may paradoxically actually improve functioning in certain domains (e.g., identity cohesion, occupational functioning), however the true 'adaptiveness' of this process is worth questioning [51] as the resulting narcissistic style often begets significant distress for those in the individual's relational sphere [16,60]. Supporting this notion of both image distortion and negative interpersonal impacts in narcissistic disorders, research has identified discrepancies between self and informant ratings of interpersonal functioning, with individuals with narcissistic features "putting a positive and self-enhancing spin on their personality while being described by peers as domineering, vindictive, and intrusive" [66], p. 435). Some variant of this process is a more likely explanation for the finding of lower reported psychopathology in our sample.
Interestingly, the 'Self-respect' facet of identity integration was found to be non-significant for both grandiose and vulnerable phenotypes. This may appear surprising at first considering that this scale contains items related to self-esteem (e.g. 'I often feel that I am not as worthy as other people') and sensitivity to criticism (e.g. 'criticisms by others can make me feel very uncertain about myself '). However, closer examination of the subscale revealed that only negative views of self were examined (i.e. not fluctuations of self-esteem). It is therefore possible that the facet of 'self-respect' is more suited for the kind of low (and relatively stable) self-esteem issues found in individuals with depression, for example. Similarly, it is also interesting to note that the moderating mediation analysis that included attachment as a moderator of the association between identity integration and interpersonal problems was non-significant. Yet, the mediating effect of identity disturbance remained significant, suggesting that unstable representations of self indirectly impact interpersonal functioning in individuals with narcissistic features, regardless of their attachment style.

Clinical implications
The findings presented in this study have important clinical implications. First, they suggest that achieving a sense of self-coherence for individuals with pathological narcissism may be a prerequisite for improving their interpersonal functioning. It is therefore not surprising that Transference Focused Psychotherapy (TFP) and Mentalization-Based Treatment (MBT) -two well-established treatments for personality disorders -both view identity disturbance as the focal point of treatment. Consistent with Kernberg's theory, TFP attempts to achieve a coherent sense of self by helping the patient integrate split-off (polarized) representations of self and others, and by helping the patient learn to reflect on emotional states that were previously not understood (and were thus acted upon). In turn, MBT typically involves helping the patient mentalize about alternative views of self and others as a way of integrating extreme self and other representations, which ultimately leads to a stronger sense of self-coherence [24]. Second, our findings highlight the importance of considering specific interpersonal patterns and problems when working with individuals who present with narcissistic features. Although clinicians frequently recognize their patients' interpersonal problems, they may overlook the potential narcissistic function of these behaviours [63]. Considering the interpersonal nature of psychotherapy, narcissistic interpersonal problems are inevitably brought directly into the treatment relationship (in the form of transference and countertransference [79]), and offer an opportunity to understand the underlying self-related difficulties. Finally, these results are consistent with the recent re-formulation of personality disorder in diagnostic systems (e.g., DSM-5, ICD-11, PDM-2) which propose a general level of personality impairment according to core self and other functional domains amongst more specific features. In this vein, our findings identify associations between such core impairments (operationalised as identity integration and interpersonal functioning with features that are narcissism specific as indexed by the B-PNI (e.g., grandiosity, vulnerability, entitlement).

Limitations
While consistent with results from previous studies, findings from the current study should be considered in light of certain limitations. First, both expressions of narcissism were measured by the B-PNI. Although scores of both expressions of narcissism can be derived from the B-PNI's subscales, some have questioned the PNI's capacity to adequately capture grandiose narcissism [51]. Second, the current study was limited by its cross-sectional nature. Although the use of SEM in cross-sectional studies has been found to have sound psychometric value (see [8], measurements of narcissism were limited to traitlevel conceptualizations [1]. Considering that grandiose and vulnerable manifestations of narcissism involve complex dynamic processes and are likely to wax and wane depending on social situations, this complexity might not have been captured by the current study's methodological design. Third, all measures used in the current study were self-reported instruments and social desirability was not controlled for. Fourth, males were underrepresented in the current study as our sample was predominately composed of female participants (87%). Although the observed gender differences found in this study are consistent with the current literature, with males scoring exclusively higher on narcissistic grandiosity than females [34], recent empirical work on gender differences in pathological narcissism suggests gender disparities in both narcissistic expression and behaviour (for a review, see [31]. Considering that recent work has also highlighted that gender differences may arise in the endorsement of items pertaining to pathological narcissism (i.e., the items contained within instruments measuring pathological narcissism may have different meaning for the two genders, and more generally reflecting the male gender expression than that of females and feminine qualities, [32,39], an equally-distributed sample across both genders as well as the inclusion of different measures of pathological narcissism would allow for a more statistically sound and clinically relevant investigation of potential gender differences. This was however beyond the scope of the current study. Finally, in extension to the previous limitation, while the current sample was drawn from a population of undergraduates it is important to note that the narcissism measure used for analysis has an explicit focus on psychopathology. That is, this study was not exploring so called 'adaptive' or 'normal' narcissism, but instead identified associations between variations in pathological narcissism and underlying issues of identity integration. It is possible that alternate narcissism constructs may be less related to such features, but as such these findings provide a more clinically relevant perspective, and future studies should consider replicating the current findings within a clinical population to support their generalizability.

Conclusion
Overall, our findings suggest that for individuals presenting with narcissistic features, capacity for self and interpersonal functioning are related constructs. That is, while grandiosity provides a façade of stable self-functioning and vulnerability presents a more desolate and fragile self-experience, both are marked by impaired interpersonal functioning as related to a broad underlying disorder of self-identity integration. In this way, grandiose and vulnerable functioning are linked, reflecting a common core of impairment, and perhaps best viewed as reflecting two sides of the same coin [53]. This view is consistent with contemporary diagnostic systems of personality, which account for core impairments in functioning but appreciate the diversity of presentation in personality 'traits' or 'styles' . Ultimately, these findings stress the importance of accounting for intrapersonal factors in the assessment and treatment of narcissism, as opposed to or in conjunction with the perhaps more observable and apparent features relating to the discordant style of interacting with others.