The Mediating Role of Autobiographical Memory in the Relationship Between Narcissism and Rejection Sensitivity

Abstract

Both grandiose and vulnerable narcissistic individuals perceive a threat to the self when they face real or imaginary rejection. The sensitivity to rejection may affect them differently. Thus, this research examines the role of grandiose and vulnerable narcissism in the retrieval of self-threatening memories. The study aims to test the mediating role of autobiographical memory (AM) in the relationship between rejection sensitivity (RS) and two dimensions of narcissism. A total of 369 university students (ages between 18 and 32) participated in the study. The Rejection Sensitivity Questionnaire was applied to the participants, triggering the memory of positive and negative autobiographical rejection. Then, the Autobiographical Memory Characteristics Questionnaire and the Pathological Narcissism Inventory were applied. The results showed that the autobiographical memory characteristics differ depending on the dimension of narcissism when they face rejection. This changes depending on having positive or negative content as well as recalling the sensory details especially emotional characteristics of the AM. The use of AM as a mediator in the relationship between narcissism and rejection sensitivity has provided a viewpoint beyond experimental and relational studies. The results have shed light on the role of narcissism and RS in the processing of autobiographical memories.

Keywords

Narcissism autobiographical memory rejection sensitivity positive-negative memory

Introduction

Studies investigating the relationship between narcissism and rejection sensitivity (RS) have shown that increased sensitivity to rejection affects the recalling of the autobiographical moment in the present time (Besser & Priel, 2009; Rhodewalt & Eddings, 2002; Sommer et al., 2009). Individuals who are high on narcissism tend to exaggerate the perceived threats in a given situation and intensify their sensitivity (Besser & Priel, 2009). Rhodewalt and Eddings (2002) investigated the effects of current self-threatening feedback on the recall of relevant autobiographical memory (AM) on narcissism. Their results indicated that self-aggrandizing memory distortions in response to rejection buffered self-esteem from the effects of rejection in narcissistic men. A metacognition study indicated that patients with narcissistic personality disorders could be helped to understand their lives through contact with the details of their raw experience as shown by autobiographical episodes (Dimaggio & Attinà, 2012). Ritchie et al. (2015) found that the magnitude of the negative affect tends to fade faster than positive affect (Fading Affect Bias: FAB), which is moderated by narcissism on autobiographical memories. The results showed that participants with low narcissism showed a large FAB while those with high narcissism showed a reversed FAB. Additionally, the retrospective recall method revealed that participants with low narcissism showed a large FAB and that the FAB diminished or dissipated as narcissism increased. The relevant literature indicates that narcissistic individuals can react differently to rejection sensitivity. The present research extends such ideas by examining the extent to which AM varies by grandiose (GN) and vulnerable (VN) narcissism in the state of rejection.

Rejection Sensitivity

RS develops a self-protecting reaction to rejection, which may differ by personality traits (Downey & Feldman, 1996). The repeated exposure to neglect and rejection in relationships with sensitive others are considered to feed narcissistic tendencies by threatening self-concept. As known, people are mostly motivated to exhibit a positive self-concept; they go to great lengths to gain positive views of the self-concept and avoid negative views (Sedikides & Gregg, 2008). Narcissists display biased memory while remembering memories (past personal experiences) in response to threatening or enhancing social interaction about self (Hepper et al., 2010; Rhodewalt & Eddings, 2002). AM that is closely related to the concept of self and rejection moments are negative life experiences that may cause the remembering of autobiographical memories in a distorted way in adulthood in narcissistic individuals (Rhodewalt & Morf, 1998). Narcissists give defensive reactions to negative feedback about their need for interpersonal self-concept (Rhodewalt & Morf, 1998). This manner of behavior mostly occurs through defense mechanisms. Individuals' RS under the behavioral responses arising after situation-specific personal memories is assumed to be activated by triggering primitive coping mechanisms (Kernberg, 1975; Kohut, 2013). As known, one of the most important triggers that are associated with the appearance of narcissism in the social field is rejection situations that pose a threat to the self (Besser & Priel, 2010; Emmons, 1987).

Grandiose-vulnerable Narcissism

Narcissistic individuals can be passive and avoidant in the concern of rejection and exclusion in their social relationships (Dickinson & Pincus, 2003). On the other hand, they may display active and aggressive behaviors by losing emotional control and exhibiting a burst of anger in the cases of rejection (Atay, 2009; Kohut, 2013; Pincus et al., 2009). In addition to studies showing the reactions with two different edges, such as displaying extreme anger and avoiding social relationships, these people may behave compromisingly in the case of rejection (Young et al., 2003). The examination of the motivation underlying this behavior reveals a viewpoint, that is, narcissism may have two different dimensions: vulnerable and grandiose dimensions (Pincus et al., 2009).

The conditions associated with narcissism may shed light on the main differences between its two dimensions. Withdrawal in interpersonal relations, social avoidance, and exploitation in interpersonal areas are positively associated with the vulnerable dimension; authoritarianism, interventionism, and intensive interpersonal problems are associated with the grandiose dimension (Pincus et al., 2009). Hypersensitivity to criticism and rejection are observed in the vulnerable dimension due to the scheme of imperfection (Young & Klosko, 1994). On the other hand, according to the theoretical and applied model of schema therapy, angry and aggressive reactions are observed in the grandiose dimension due to the justification scheme (Young & Klosko, 1994). In addition, narcissistic individuals may hide their emotions and experience feelings of anger rather intensely but implicitly due to the fear of rejection in their social relationships (Rohmann et al., 2012; Rose, 2002). The same person may experience both dimensions despite them being two separate ends of narcissism (Pincus & Lukowitsky, 2010). The sensitivity to rejection of individuals with narcissistic tendencies may affect them differently depending on the dimensions of narcissism, and this effect may vary depending on the characteristics of AM, that is, whether it has positive or negative content.

The Mediating Role of Autobiographical Memory

According to the Self Memory System Model, AM organizes itself toward goal-orientedness (Conway & Jobson, 2012; Conway & Pleydell-Pearce, 2000). The working self functions together with the autobiographical database and the conceptual self, in keeping with the effort to maintain both the self-harmony and the originality of the memory (Conway & Jobson, 2012). The moments conflicting with self-harmony can make access to memories more difficult (Conway, 2005). AM is affected by mental disorders; for example, AM characteristics are impaired in depression, especially in terms of positive and neutral cues (Young et al., 2012). Regarding personality disorders, autobiographical memory studies have focused on depression and its relationship to borderline personality disorder (Kremers et al., 2004; Startup et al., 2001). Depending on the different levels of narcissism, there are studies showing the distorting effect of feedback about the self on memories (Rhodewalt & Eddings, 2002). However, the number of studies on narcissism is insufficient in this area. In addition, memory characteristics may vary depending on positive or negative AM events, especially in narcissistic individuals (D’Argembeau et al., 2003; Ritchie et al., 2015) During the recall of the AM, the motivation to preserve and develop self-value, which is known to be especially high in narcissistic individuals, directs the remembering of the event (Walker et al., 2009). In other words, individuals in the grandiose dimension of narcissism are expected to act in a self-focused and self-centered manner to keep “feeling good about themselves.” Moreover, even if these individuals' sensitivity to rejection is as high as those with the vulnerable dimension, they may tend to over-emphasize positive events by minimizing or neglecting negative events. On the other hand, Kohut (1968) argues that narcissism is a personality structure with cognitive and behavioral reflections that extends beyond inflated self-confidence. Vulnerable narcissistic individuals tend to be rebellious, deceptive, and manipulative when their sense of superiority or entitlement is threatened, even if they do not exhibit a clear grandiose. They tend to exhibit confrontation, violence, and self-destruction, especially when self-esteem is threatened (Besser & Zeigler-Hill, 2010; Vaillancourt, 2013). This is consistent with the view that individuals who scored high in the grandiose dimension of narcissism claim that their high self-confidence is often associated with a number of psychological benefits (Sedikides et al., 2004). On the other hand, individuals with the vulnerable dimension have a tendency to recall negative memories when remembering autobiographical events due to their sensitivity to self-threatening events. The vulnerable ones are expected to remember highly negative memories compared to those who are high in the grandiose dimension. In addition, despite the inflated sense of self-confidence, vulnerable narcissistic individuals might perform poorly in emotion regulation (Ronningstam, 2009). This is consistent with the finding that an increase in the level of vulnerable narcissism is associated with high stress, anxiety, depression as well as reduced happiness and life satisfaction (Twenge, 2014). Narcissistic individuals may recall positive memories better to feel good about themselves. On the other hand, narcissistic individuals are thought to keep negative memories more in mind to minimize these negative events (Ritchie et al., 2015).

From this point of view, the vulnerable dimension will have higher negative recall rates compared to positive memories in keeping with the motivation to avoid a self-worth–threatening situation while the grandiose narcissistic individuals tend to prefer the positive memory when remembering their autobiographical memories.

Aim of the Study

This study derives from the idea that this can be explained by unveiling the differences between the GN and VN. The answers to two basic research questions were sought to test the RS with different dimensions of narcissism through AM characteristics. The research questions were whether AM has a mediating role in the relationship between RS and narcissism and how this role is affected by the memory’s positivity and negativity.

Methods

Participants

A total of 382 participants whose ages ranged between 18 and 35 years participated in the study. The students have been studying at public universities in three different big cities from three different regions. Participants voluntarily attended the study in the 2019–2020 fall semester, and course credit was given as compensation. Thirteen participants from the sample were excluded from the study because of having psychiatric or neurological diagnoses and regular use of medication because of their diagnoses. Finally, 369 students (M = 20.88, SD = 3.10) comprising 272 women aged between 18 and 32 years (M = 20.66, SD = 2.49) and 87 men aged between 18 and 35 years (M = 21.63, SD = 4.48) were included in the study. Since ten participants did not provide information about gender, they were not included in the gender comparisons but included in the total score calculation. Of the participants, 18 (5%) were high school graduates, 339 (92%) were university students, and 12 (3%) were graduate students.

Instruments

Pathological Narcissism Inventory (PNI): It is a 6-point Likert-type (ranging from 0: “not at all like me” to 5: “very much like me”) measurement scale consisting of 52 items that examine the vulnerable and grandiose dimensions of narcissism (Pincus et al., 2009). Şen and Barışkın (2019) adapted the scale to Turkish, revealing a factor structure comprising two dimensions (grandiose and vulnerable) and five sub-dimensions. Accordingly, in the PNI, the “Grandiose Self” (GS) and the “Grandiose Fantasy” (GF) (with negative loading) factors represent the grandiosity dimension of narcissism while the factors of “Recognition Expectations” (REX), “Vulnerable Self” (VS), “Approval Seeking” (AS), and “Self Sacrificing” (SS) represent the vulnerability dimension of narcissism. The Cronbach’s alpha reliability value of PNE was .93, the test–retest reliability coefficient was .91, and the explained variance was 50.24%. In this study, the Cronbach’s alpha reliability value of the total scale was determined to be .92 while the explained variance was 51.19%.

Autobiographical Memory Characteristics Questionnaire (AMCQ): It was developed by Boyacioglu and Akfirat (2015). The AMCQ is a universal, 7-point Likert-type scale consisting of 63 items with 16 items that should be reversed and are valid in different cultures (Rubin et al., 2007). The AMCQ reliably measures 14 different AM characteristics, which are vividness, belief in accuracy, place details, sensory details, accessibility, sharing, observer perspective, field perspective, narrative coherence, recollection, emotional valence, emotional intensity, emotional distancing, and preoccupation with emotions. The AMCQ has so far been tested for self-defining memories, romantic relationships, and childhood memories. The items loaded onto the expected factors with high factor loadings ranging from .36 to .93, with 14 factors accounting for 72% of the total variance (Boyacioglu & Akfirat, 2015). In this study, the Cronbach’s alpha reliability value of the total scale varied from .44 to .91 and explained variance was 70.02%.

Rejection Sensitivity Questionnaire (RSQ): The scale was developed by Downey and Feldman (1996). It measures the RS with its main dimensions, which are the expectation of acceptance and rejection anxiety. Köse et al. (2017) carried out the standardization of the scale in Turkish. The scale consists of 18 items and has a single factor structure. The participants respond to 18 different rejection situations on a 6-point Likert-type rating scale. The Cronbach’s alpha value of the scale was found to be .85 and the explained variance was 30.82%. In this study, the Cronbach’s alpha reliability value for the total scale was .84 and explained variance was 34.51%.

Procedure

Before the onset of the study, ethical approval was obtained from the committee. Data were collected under the supervision of researchers. Participants were asked to sign a written informed consent form and fill the socio-demographic information form including questions about gender, age, educational status, and psychiatric diagnosis. The purpose of the study was explained to the participants and the data collection procedure took about 30–35 minutes.

The scale sequence of the research is designed for purpose-made and included three stages that the participants followed individually. In the first step, individuals were presented with the RSQ, aiming to trigger the RS for the later recall of AM. In the second step, participants were asked to write a positive and negative memory while reminding them that the event should be the first thing that comes to their minds and they remember it prominently. We proceeded to the third step when the participants sufficiently remembered the event (Figures 1 and 2). In the third step, the participants were asked to choose one of the positive or negative memories that they had written about and to answer the questions on the back page by focusing on the AM that was evoked by the memories. Here, the AMCQ was presented to the participants while the memories were still vivid in their minds. At the last step, the PNI was presented and the measurement part of the research was completed.

Figure 1.

Theoretical model tested after positive memory-based modification. Narcissism: Pathological Narcissism Inventory (Grandiose and Vulnerable dimension); Rejection Sensitivity: Rejection Sensitivity Scale (rejection anxiety and expectation acceptance dimension); Autobiographic memory: Autobiographic Memory Properties Scale (belief in accuracy, sensory details, sharing, accessibility, narrative coherence, recollection, emotional valence, emotional intensity, emotional distancing, preoccupation with emotions and accessibility dimension).

Figure 2.

Theoretical model tested after negative memory-based modification. Narcissism: Pathological Narcissism Inventory (Grandiose and Vulnerable dimension); Rejection Sensitivity: Rejection Sensitivity Scale (rejection anxiety and expectation acceptance dimension); Autobiographic memory: Autobiographic Memory Properties Scale (belief in accuracy, sensory details, sharing, accessibility, narrative coherence, recollection, emotional valence, emotional intensity, emotional distancing, preoccupation with emotions and accessibility dimension).

This simple set of procedures was intended to trigger individuals' rejection sensitivities and recall their AM. For the purpose of the study, the recalled AM should be related to the RS. The recalling of the AM, which reflects the sensitivity of rejection, is an important component for the functionality of other stages. The Statistical Package for the Social Sciences (SPSS V. 23) and Structural Equation Modeling (AMOS V. 22) programs were used for analyses.

Results

The missing value, normality, and multiple linearity assumptions were examined. Firstly, in line with the recommendations of Tabachnick and Fidell (2013), there was no participant with multiple sequential markings (those with a SD value of less than .3) between items and with a loss value of more than 5%. Secondly, the skewness and kurtosis values of the variables were calculated. The data conformed to the assumption that the kurtosis and skewness values should be between −1.5 and +1.5 for social sciences (Tabachnick & Fidell, 2013). The multiple linearity assumption was not examined since the number of independent variables was no more than one.

The correlation coefficients between the basic variables were calculated using the Pearson correlation for the path analysis in the next step. As seen in Table 1, the grandiose and vulnerable narcissism dimensions were negatively correlated (r = −.45, p < .01). Vulnerable narcissism showed a positive relationship with rejection anxiety (r = .35, p < .01) and a negative relationship with acceptance expectation (r = −.41, p < .01); on the contrary, grandiosity narcissism had a negative relationship with rejection anxiety (r = −.34, p < .01) and acceptance expectation (r = −.25, p < .01). The examination of the relationship between vulnerable narcissism and autobiographical memory characteristics revealed that vulnerable narcissism and grandiose narcissism were related to autobiographical memory characteristics (Range: from r = −.12 to .30, p < .01; Range: from r = −.11 to .11, p < .05, respectively).

Table 1.

Correlation Matrix of Narcissism Sub-Dimensions, Rejection Sensitivity, and Autobiographical Memory Features.

	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
1. Grandiose	—	−.45**	−.34**	−.25**	−.08	.02	−.08	−.03	−.03	−.11*	−.12*	−.06	.11*	−.21**	−.02	−.07	−.06	.09
2. Vulnerable			−.41**	.35**	.10	−.12*	−.04	.27**	−.02	.06	.26**	.01	−.16**	.22**	−.12*	.12*	.22*	.30**
3. Expectation of acceptance				.82**	.15**	.06	.13*	.13*	−.03	.16**	.08	.06	.01	.25**	−.09	.13*	.08	.08
4. Rejection anxiety					.06	−.03	.07	.10	−.05	,14**	.08	.01	−.03	.21**	−.14**	.11*	.18**	.19**
5. Vividness						.53**	.31**	.52**	.46**	.25**	.05	.29**	.44**	.55**	.09	.47**	−.12*	.07
6. Belief in accuracy							.46**	.19**	.36**	.08	−.16**	.18**	.55**	.25**	.06	.18**	−.14**	−.28**
7. Place detail								.07	.17**	.08	−.01	.19**	.25**	.25**	.06	.18**	−.04	.34**
8. Sensory detail									.25**	.14**	.12*	.18**	.24**	.56**	.05	.50**	.04	.34**
9. Accessibility										.24**	−.11*	.06	.47**	.30**	.01	.43**	−.19**	−.04
10. Sharing											−.02	.13*	.13*	.32**	.13*	.30**	−.27**	.02
11. Observer perspective												−.31**	−.21**	.13**	−.15**	.06	.22*	.29**
12. Filled perspective													.16**	.33**	.05	.16	−.01	.06
13. Narrative coherence														21**	.12*	.24**	−.28*	.29**
14. Recollection															.07	.64**	.02	.36**
15. Emotional valence																.09	−42**	−.31**
16. Emotional intensity																	−.08	.35**
17. Emotional distancing																		.46
18. Preoccupation with emotions

Note: *p < .05; **p < .01; ***p < .00.

The scores of the participants were obtained using the PNI scale and with respect to the GN and VN sub-dimensions. The t-test was used to reveal the positive or negative memory preferences in terms of the dimensions of narcissism. The results of the paired sample t-test showed that those who scored high on vulnerable narcissism (M = 91.83) were more likely to choose negative memories (t = 35.50, p = .001) than those who scored high on grandiose narcissism (M = 29.50). The increase in the vulnerable narcissism scores indicated preference for negative memories (M = 96.27) compared to positive memories (M = 89.80) (t = 1.96, p = .05). On the other hand, the increase in the grandiose narcissism scores did not indicate a significant difference in terms of memory preferences (t = 0.76, p = .44). Furthermore, the increase in the total PNI score had no significant effect on memory preference (t = 1.75, p = .08).

The PNI sub-dimensions were evaluated in terms of memory preference to better understand which characteristics of narcissism are decisive in choosing a positive or negative memory. The evaluation of the PNI with respect to its sub-dimensions revealed that the increase in the scores of the vulnerable self and approval seeking sub-dimensions, which are two important sub-dimensions of the vulnerable basic dimension, had a significant effect on memory preference (t = 2.41, p = .05; t = 1.92, p = .05, respectively); the increase in the recognition expectations sub-dimensions, grandiose self, grandiose fantasy, and self-sacrificing subscales did not significantly affect the preference (t = 1.68, p = .09; t = 0.61, p = .55, t = 0.38, p = .70; t = 1.06, p = .30, respectively). Accordingly, the increase in the vulnerable self sub-dimension indicated preference for negative memory (M = 31.66, S = 10.56) compared to positive memory (M = 28.90, S = 10.03); similarly, the increase in the approval seeking sub-dimension indicated preference for negative memory (M = 14.63, S = 3.66) compared to positive memory (M = 13.24, S = 3.03).

The participants were asked to write down both positive and negative memories, but they were allowed to choose one to think about and focus on. Here, 252 (68%) participants preferred positive memories and 131 (32%) participants preferred negative memories. The t-test was used to understand which characteristics of autobiographical memory are decisive in choosing a positive or negative memory. The evaluation of the characteristics of the AM revealed that, albeit overlapping, AM differed by the emotional features of memory preference. Emotional valence, emotional distancing, and preoccupation with emotions significantly affected the preference for positive or negative memory (t = 19.33, p = .00; t = 5.88, p = .01; t = 5.15, p =.01, respectively); the increase in emotional valence indicated preference for positive memory (M = 34.01, S = 9.45) compared to negative memory (M = 14.36, S = 8.15); on the contrary, the increase in emotional distancing and preoccupation with emotions indicated a tendency to choose negative memory (M = 14.93, S = 6.26; M = 19.47, S = 9.56, respectively) compared to positive memory (M = 10.71, S = 6.46; M = 10.50, S = 8.13).

The ANOVA was conducted to examine the differences in rejection anxiety and acceptance expectation at different levels of narcissism (GN and VN). The results showed that, regardless of memory preference, the vulnerable narcissistic individuals experienced more rejection anxiety and acceptance expectation than the grandiose narcissistic individuals (F (1, 319) = 2.72, p = .001; F (1,319) = 2.07, p = 01, respectively). When the data were divided into two as positive and negative memories, it was found that those who were vulnerable in both positive and negative memories experienced more rejection anxiety and acceptance expectation than the grandiose dimension (F (1, 221) = 3.07, p = .001; F (1,319) = 2.07, p = 01, respectively).

The mediator analysis was conducted to examine whether there were significant differences between the grandiose and vulnerable narcissism and RS in the AM characteristics with respect to the type of memory (positive and negative). A correlation matrix was applied to examine the relationship between the sub-dimensions, RS, and AM characteristics. Table 1 shows the findings. According to the correlation table, the four sub-dimensions of the AMCQ, namely, vividness, place details, observer perspective, and field perspective, were not related to the grandiose and vulnerable dimensions of narcissism. Hence, these variables were not included in the model in which the mediating role of AM characteristics in narcissism and RS was examined.

Structural Equation Modeling

The model test method was used to examine the mediating role of AM characteristics in the relationship between narcissism and RS. The analysis coming from theoretical information revealed that the effect of the narcissism variable, which comprises the grandiose and vulnerable dimensions of narcissism, affected RS and the effect was determined mediated by some AM characteristics (p < .01). The related variables were determined following previous analysis and included in the proposed model to carry out the path analysis, which was deemed appropriate for testing the model. Thus, the model was tested separately for those who chose positive and negative memories for the purpose of the research. The models of those who chose the positive memory first and then the negative memory were presented respectively.

The Maximum Likelihood Method, Structural Equation Modeling (SEM), and Path Analysis were conducted to explain the relationship between the variables in the theoretical model using the AMOS 22 program. In the analysis, the extent of the ability of the model to predict the data was determined using fit indices. The Chi-Square (X²) test is frequently used to evaluate data compliance in the Structural Equation Modeling, (Sümer, 2000). A χ²/SD ratio that is lower than 2, which is calculated using equation modeling, an RMSEA value lower than .05, and the GFI, AGFI, NFI, and CFI values above .90 indicate a perfect fit between the model and the data. The model fit determines the fit of the variance–covariance matrix to the structural equation model. The Chi-square, GFI, AGFI, and RMR indices are commonly used to examine model fitness (Tabachnick & Fidell, 2013). After the use of the modification indices, the GFI and AGFI values of both models were found to be acceptably high (.90 and above).

Since the assumptions required for the execution of the structural equation model were met, the path analysis was carried out for the theoretical model that was created for positive memory. The analysis revealed fit indices of X²/SD = 4.89, p < .001, CFI = .78, IFI = .73, NFI = .75, and RMSEA = .09. The values indicated the need for adjustments to the model: The paths of vividness, place details, observer perspective, and field perspective were determined to be not statistically significant and, thus, were removed from the model. In addition, e11–e12 (X² = 27.87, p < .001), e3–e7 (X² = 14.61, p < .001), e9–e12 (X² = 14.17, p < .001), and e7–e12 (X² = 21.46, p < .001) error scores of the same factor matched in line with the modification suggestions. After this process, the fit indices of the model slightly increased, equaling to X²/SD = 3.77, p < .001, CFI = .80, IFI = .78, NFI = .80, RMSEA = .09. However, this fell behind the acceptable point. Thus, the modification suggestions were reexamined and the e6–e7 (X² = 6.36, p < .001), e7–e11 (X² = 6.93, p < .001), and e3–e12 (X² = 9.22, p < .001) error scores in the same factor were matched accordingly. Finally, in line with the modification suggestions, the error points of e9–e11 (X² = 6.36, p < .001) and e5–e11 (X² = 6.93, p 001) were matched. After this procedure, the fit indices of the model reached acceptable levels: X²/SD = 2.03, p < .001, CFI = .94, IFI = .94, NFI = .90, GFI=.91, AGFI=90, RMSEA = .07.

The total effect of narcissism and the AM variable on the rejection variable was β = .52 and β = .07 (p < .001). The examination of the sub-dimensions of RS revealed that the total effects on the acceptance expectations were β = .56 and β = .18, respectively; Total effects on rejection anxiety were β = .44 and β = .17, respectively. The examination of the direct effects between the research variables showed that the direct effect of narcissism and AM variable on the rejection variable was β = .51 and β = .08 (p < .001). The examination of the sub-dimensions of RS revealed that the direct effects on the acceptance were β = .47 and β = .03, respectively, while the total effects on rejection anxiety were β= .32 and β = .02, respectively. One of the advantages of the structural equation models is their ability to reveal indirect relationships between variables. According to the findings, the indirect effect of narcissism and the AM variable on the rejection variable was β = .00 and β = .00, their effects on RS were β = .42 and β = .02 while they were β = .44 and β =.67 (p < .01) on the expectation of acceptance, respectively. Table 2 shows the results. The results have revealed the mediating role of the AM characteristics in the effect of narcissism on RS when positive memories were chosen.

Table 2.

Standardized Total, Direct, and Indirect Impact Values Among Positive and Negative Memory-based Variables (β).

	Positive									Negative
	Total			Direct			Indirect			Total			Direct			Indirect
	PNI	AMCQ	RSQ	PNI	AMCQ	RSQ	PNI	AMCQ	RSQ	PNI	AMCQ	RSQ	PNI	AMCQ	RSQ	PNI	AMCQ	RSQ
Autobiographical memory	.34	.00	.00	.32	.00	.00	.00	.00	.00	.34	.00	.00	.32	.00	.00	.00	.00	.00
Rejection	.52	.07	.00	.52	.07	.00	.02	.00	.00	.52	.07	.00	.52	.07	.00	.02	.00	.00
Expectation of acceptance	.56	.18	−.97	.47	.03	−.97	.42	.02	.00	.56	.18	−.97	.47	.03	−.97	.42	.02	.00
Anxiety of rejection	.44	.17	.82	.32	.02	.82	.44	.67	.00	.44	.17	.82	.32	.02	.82	.44	.67	.00
Preoccupation with emotions	.11	.32	.00	.00	.32	.00	.00	.00	.00	.11	.32	.00	.00	.32	.00	.00	.00	.00
Emotional distancing	−.02	−.05	.00	.00	−05	.00	−.11	.00	.00	−.02	−.05	.00	.00	−05	.00	−.11	.00	.00
Emotional intensity	.24	.71	.00	.00	.70	.00	−.02	.00	.00	.24	.71	.00	.00	.70	.00	−.02	.00	.00
Emotional valence	.07	.21	.00	.00	.21	.00	.07	.00	.00	.07	.21	.00	.00	.21	.00	.07	.00	.00
Recollection	.31	.90	.00	.00	.91	.00	.31	.00	.00	.31	.90	.00	.00	.91	.00	.31	.00	.00
Narrative coherence	.12	.33	.00	.00	.33	.00	.11	.00	.00	.12	.33	.00	.00	.33	.00	.11	.00	.00
Accessibility	.11	.32	.00	.00	.32	.00	.11	.00	.00	.11	.32	.00	.00	.32	.00	.11	.00	.00
Sharing	.16	.46	.00	.00	.46	.00	.16	.00	.00	.16	.46	.00	.00	.46	.00	.16	.00	.00
Sensory detail	.23	.66	.00	.00	.66	.00	.23	.00	.00	.23	.66	.00	.00	.66	.00	.23	.00	.00
Belief in accuracy	.10	.27	.00	.00	.27	.00	.09	.00	.00	.10	.27	.00	.00	.27	.00	.09	.00	.00
Grandiose	−.61	.00	.00	−.61	.00	.00	.00	.00	.00	−.61	.00	.00	−.61	.00	.00	.00	.00	.00
Vulnerable	.77	.00	.00	.77	.00	.00	.00	.00	.00	.77	.00	.00	.77	.00	.00	.00	.00	.00

PNI, Pathological Narcissism Inventory; AMCQ, Autobiographical Memory Characteristics Questionnaire; RSQ, Rejection Sensitivity Questionnaire.

Since the assumptions required for the execution of the structural equation model were met, path analysis was carried out for the theoretical model that was created for negative memory. The analysis revealed fit indices of X²/SD = 3.44, p < .001, CFI = .74, IFI = .75, NFI = .67, and RMSEA = .10. The values indicated the need for adjustments to the model: The paths of vividness, place details, observer perspective, and field perspective were not statistically significant, and, thus, were removed from the model. In addition, in line with the modification recommendations, the error scores of e12–e4 (X² = 11.61, p < .001), e7–e12 (X² = 24.88, p < .001), e6–e12 (X² = 12.42, p < .001), and e6–e7 (X² = 11.25, p < .001) were matched. After this procedure, the fit indices of the model slightly increased and reached X²/SD = 2.72, p < .001, CFI = .78, IFI = .81, NFI = .77, and RMSEA = .09. However, its falling behind the acceptable point, the modification suggestions were reexamined and the error scores of e3–e9 (X² = 9.62, p < .001), e4–e9 (X² = 9.37, p < .001), and e3–e12 (X² = 5.87, p < .001) of the same factor were matched accordingly. Finally, the error points of e3–e6 (X² = 9.62, p < .001) and e6-e10 (X²= 9.37) were matched. After this procedure, the fit indices of the model reached acceptable ranges, equaling to X²/Sd= 2.09, p < .001, CFI = .90, IFI = .88, NFI = .89, GFI= .90, AGFI=.90, and RMSEA = .08. Since IFI = .88, NFI = .89 values were lower than .90, the reliability indices were controlled. The goodness-of-fit indices (GFI and AGFI) were found to be at acceptable levels for being equal to or greater than .90 (Iacobucci, 2010).

The total effect of narcissism and the AM variable on the rejection variable was β = .58 and β = .11, respectively (p < .001). However, the examination of the sub-dimensions of RS revealed that the total effects on the acceptance expectations were β = .53 and β = .10, respectively. The total effects on rejection anxiety were β = .47 and β = .09, respectively. The examination of the direct effects between the research variables revealed that the direct effect of narcissism and the AM variable on the rejection sensitivity variable was β = .56 and β = .11 (p < .001), respectively. The examination of the sub-dimensions of RS showed that the direct effects on the acceptance expectation were β = .52 and β = .03, respectively; the total effects on rejection anxiety were β = .42 and β = .02, respectively. The indirect effect of narcissism and the AM variable on the rejection variable was β = .02 and β = .00, respectively, while their indirect effects on RS were β = .49 and β = .96 and on acceptance expectation were β = .47 and β = .86 (p < .01), respectively. Table 2 shows the results. The results revealed the mediating role of the AM characteristics in the effect of narcissism on RS when negative memories were chosen.

In conclusion, both positive and negative memories had a mediating role in the relationship between narcissism and RS in both models.

Discussion

With the extension of the studies on narcissism, the examination of GN and VN with interdisciplinary studies has become necessary. Today, narcissism is examined by joint studies of personality as well as psychiatry, clinical psychology, and cognitive psychology (Ritchie et al., 2015). This study elaborately examines narcissism and RS in interpersonal relations with the mediating role of AM reflecting the cognitive characteristics of memories. The results revealed that the memories played a mediator role in the RS of narcissistic individuals. The interaction of the RS that triggers autobiographical memories in the narcissism dimensions was examined with respect to the preference of participants for positive or negative memories. In this context, different narcissism dimensions were evaluated using the PNI and the sub-dimensions differing in terms of AM characteristics.

Positive memories were preferred more than negative memories, regardless of the sub-dimensions. Studies have shown that the emotional experience of the AM characteristics has led to the forgetting of the negative memories faster than the positive ones (Walker et al., 2003), but this effect does not occur in narcissistic individuals (Ritchie et al., 2015). This is consistent with human’s avoidance of pain and orientation toward pleasure. The sample of the study was made up of healthy individuals. The comparison of the GN and VN scores revealed that the preference for negative memories increased with increasing vulnerable narcissism scores. The increase in the scores with increasing vulnerable narcissism scores, which is sensitive to self-threatening events, is consistent with the literature indicating the display of self-defense behaviors (Baumeister & Vohs, 2001; Besser & Zeigler-Hill, 2010; Vaillancourt, 2013).

The tendency to prefer negative memories significantly increased as the scores on the vulnerable self and self-affirmation dimensions, which are two important sub-dimensions of the vulnerable dimension, increased. It was also observed that the expectations for being noticed, grandiose self, grandiose fantasies, and the increase in the self-sacrifice subscales did not have a significant effect on memory preference. The need for self-affirmation and the vulnerable self-structure of the vulnerable group seemingly led to the increased recalling of negative memories. The vulnerability and recollection variables were positively related to repeatedly experiencing the memory; the grandiosity variable was negatively related to the recollection variable, which is consistent with the literature (Ritchie et al., 2015; Sedikides et al., 2004). Correspondingly, while those who scored high on the grandiose dimension kept their self-value by not remembering their negative memories, the people who scored high on the vulnerable dimension remembered their negative memories in need of self-confirmation in social relations as a form of acceptance or avoidance of rejection. Rejection anxiety in narcissistic individuals was positively associated with the vulnerable dimension and negatively associated with the grandiose dimension. The expression of anxiety in individuals with a high level of vulnerable narcissism is observed more due to anxiety’s certain gains such as acceptance through self-sacrifice (Twenge, 2014). Autobiographical memories triggered by RS may create a feeling of re-living and may be one of the reasons for the emergence of a strong defense mechanism. The observer perspective was also positively associated with the vulnerable dimension. How the memory is evaluated through someone else’s eyes is an indicator of its importance to vulnerable narcissistic individuals and may be one indicator of their increased sensitivity to the self-threat coming from social relations. On the other hand, the observer perspective was negatively correlated with the grandiose dimension, which is consistent with the literature (Besser & Priel, 2010). The GN seemingly did not prefer to see the memory through an outside eye. As known, narcissistic individuals’ past is far from accurate and realistic mirroring and, thus, the evaluation of the events from the viewpoint of others is important, especially for vulnerable ones, while the GN seems to refuse others' perspective by developing a reaction (Kernberg, 1975; Kohut, 2013). Depending on the need for mirroring, acceptance-approval, social visibility, and recognition of others in terms of both positive and negative memories differ with respect to the grandiose and vulnerable dimensions (Pincus, 2013).

Unlike the grandiose dimension, RS was correlated with sensory details, preoccupation with emotions, belief in accuracy, emotional valance, emotional intensity, and emotional distancing in individuals with high scores on the vulnerable dimension. People with vulnerable narcissistic personality traits who are afraid of being rejected in interpersonal relationships experience their emotions intensely but implicitly (Rohmann et al., 2012; Rose, 2002). Therefore, vulnerable participants might have remembered the sensory details of the past better than the grandiose participants. The VN participants, who are more sensitive to self-threatening situations, seem to remember the emotional aspect of autobiographical memories more elaborately than the GN participants. In addition, vulnerable narcissistic people show more emotion regulation problems (Ronningstam, 2009) and high stress, anxiety, and depression (Twenge, 2014). These problems are related to emotional distancing, emotional intensity, preoccupation with emotions, and emotional valance dimensions, unlike the case of GN (Pincus, 2013). This supports a hierarchical model suggesting a dimensional approach, which is defined as two separate structures comprising vulnerable and grandiose narcissism.

According to the Structural Equation Model, AM has a mediator role in the relationship between narcissism and RS. The results showed that the model was acceptable. The separation of the data into positive and negative memories revealed the mediating role of AM in the relationship between narcissism and RS in both cases, showing that the effect of narcissism on RS was regulated by both positive and negative memories. The concept of working self in the AM literature functions in collaboration with the autobiographical database and the conceptual self to protect both the harmony and originality of the memory (Conway & Jobson, 2012). The conflicting memories with the harmony of the self-make access to the original memory difficult (Conway, 2005). This indicates that the self-related characteristics of memory may be important rather than having positive or negative content. In summary, the participants’ description of the memory as positive or negative was related to the emotional characteristics they attribute to the memory, which was significant both in the grandiose and vulnerable dimensions. However, when the memory was recalled, its other contextual features, especially emotional features, were reconstructed to protect and maintain the self-value.

Limitations and Future Directions

Despite the strong arguments made by this study, the self-report-based data collection was among its main limitations. Social desirability or distortional responses could have acted as confounding variables in narcissistic participants. This poses a major problem in personality studies. Future studies should control social desirability or include it in research design. The testing of narcissistic participants using tasks that are created in the laboratory environment may increase the validity of the findings. Another limitation of the study is its inclusion of college students only. Care must be exercised about the generalizability of the findings due to the limited age range and non-clinical nature of the sample. The sample size was another important limitation. The X²/SD fit index increases depending on the sample size and, thus, it may produce biased results (Iacobucci, 2010), which, was questionably among the decreasing fit indices depending on the increase in the sample size.

This study makes a significant contribution to the literature by evaluating narcissism both as a categorical and dimensional variable in the same study. The use of AM as a mediator in the relationship between narcissism and RS has carried this study beyond experimental and relational studies. The study is a pioneer will for its examination of narcissism and RS through autobiographical memories. Its results can guide psychiatrists and psychologists in the differentiation of access to AM when evaluating and treating narcissistic individuals. In this respect, it will provide a framework for our understanding of the biases and cognitive distortions that are experienced when accessing AM memories during psychotherapy.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Gamze Sen

Gün Pakyürek

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.

Informed Consent

Informed consent was obtained from all individual participants included in the study.

Author Biographies

Gamze Sen is an assistant professor at Manisa Celal Bayar University, Department of Clinical Psychology since 2020. Her doctoral thesis is about narcissistic patterns and responses to distressing interpersonal experiences: based on cognitive interpersonal theory. The areas of her interest are narcissistic personality, self-determination theory, and psychotherapy processes.

Gün Pakyürek is an assistant professor at Aydın Adnan Menderes University, Department of Psychology since 2017. His doctoral thesis is about the effects of smoking and smoking cues on prospective memory performance. His research interests focus on prospective memory, autobiographical memory, and addiction.

References

Atay

(2009). Narsistik Kişilik Envanteri’nin Türkçe’ye standardizasyonu. Gazi Üniversitesi İktisadi Ve İdari Bilimler Fakültesi Dergisi, 11, 181–196.

Baumeister

R. F.

Vohs

K. D.

(2001). Narcissism as addiction to esteem. Psychological Inquiry, 12(4), 206–210. https://www-jstor-org.web.bisu.edu.cn/stable/1449473

Besser

Priel

(2009). Emotional responses to a romantic partner's imaginary rejection: The roles of attachment anxiety, covert narcissism, and self‐evaluation. Journal of Personality, 77(1), 287–325. https://doi.org/10.1111/j.1467-6494.2008.00546.x

Besser

Priel

(2010). Grandiose narcissism versus vulnerable narcissism in threatening situations: Emotional reactions to achievement failure and interpersonal rejection. Journal of Social and Clinical Psychology, 29(8), 874–902. https://doi.org/10.1521/jscp.2010.29.8.874

Besser

Zeigler-Hill

(2010). The influence of pathological narcissism on emotional and motivational responses to negative events: The roles of visibility and concern about humiliation. Journal of Research in Personality, 44(4), 520–534. https://doi.org/10.1016/j.jrp.2010.06.006

Boyacioglu

Akfirat

(2015). Development and psychometric properties of a new measure for memory phenomenology: The Autobiographical Memory characteristics questionnaire. Memory, 23(7), 1070–1092. https://doi.org/10.1080/09658211.2014.953960

Conway

M. A.

(2005). Memory and the self. Journal of Memory and Language, 53(4), 594–628. https://doi.org/10.1016/j.jml.2005.08.005

Conway

M. A.

Jobson

(2012). On the nature of Autobiographical memory. In Berntsen

Rubin

(Eds.). Understanding Autobiographical memory: Theories and Approaches (pp. 54–69). Cambridge, UK: Cambridge University Press.

Conway

M. A.

Pleydell-Pearce

C. W.

(2000). The construction of autobiographical memories in the self-memory system. Psychological Review, 107, 261–288. https://doi.org/10.1037/0033-295x.107.2.261

10.

D'Argembeau

Comblain

Van der Linden

(2003). Phenomenal characteristics of autobiographical memories for positive, negative, and neutral events. Applied Cognitive Psychology: The Official Journal of the Society for Applied Research in Memory and Cognition, 17(3), 281–294. https://doi.org/10.1002/acp.856

11.

Dickinson

K. A.

Pincus

A. L.

(2003). Interpersonal analysis of grandiose and vulnerable narcissism. Journal of Personality Disorders, 17(3), 188–207. https://doi.org/10.1521/pedi.17.3.188.22146

12.

Dimaggio

Attinà

(2012). Metacognitive interpersonal therapy for narcissistic personality disorder and associated perfectionism. Journal of Clinical Psychology, 68(8), 922–934. https://doi.org/10.1002/jclp.21896

13.

Downey

Feldman

S. I.

(1996). Implications of rejection sensitivity for intimate relationships. Journal of Personality and Social Psychology, 70(6), 1327–1343. https://doi.org/10.1037//0022-3514.70.6.1327

14.

Emmons

R. A.

(1987). Narcissism: Theory and measurement. Journal of Personality and Social Psychology, 52(1), 11–17. https://doi.org/10.1037//0022-3514.52.1.11

15.

Hepper

E. G.

Gramzow

R. H.

Sedikides

(2010). Individual differences in self‐enhancement and self-protection strategies: An integrative analysis. Journal of Personality, 78(2), 781–814. https://doi.org/10.1111/j.1467-6494.2010.00633.x

16.

Iacobucci

(2010). Structural equations modeling: Fit indices, sample size, and advanced topics. Journal of Consumer Psychology, 20(1), 90–98. https://doi.org/10.1016/j.jcps.2009.09.003

17.

Kernberg

O. F.

(1975). Borderline conditions and pathological narcissism. New York: Jason Aronson.

18.

Kohut

(1968). The psychoanalytic treatment of narcissistic personality disorders: Outline of a systematic approach. The Psychoanalytic Study of the Child, 23(1), 86–113. https://doi.org/10.1080/00797308.1968.11822951

19.

Kohut

(2013). The analysis of the self: A systematic approach to the psychoanalytic treatment of narcissistic personality disorders. University of Chicago Press.

20.

Köse

A. G.

Özen-Çıplak

Ulaşan Özgüle

E. T.

Sümer

(2017). Reddedilme Duyarlılığı Ölçeğinin Türkçeye Uyarlanması. Nesne Psikoloji Dergisi 11(3), 383–403. https://doi.org/10.7816/nesne-05-11-03

21.

Kremers

I. P.

Spinhoven

Van der Does

A. J. W.

(2004). Autobiographical memory in depressed and non‐depressed patients with borderline personality disorder. British Journal of Clinical Psychology, 43(1), 17–29. https://doi.org/10.1348/014466504772812940

22.

Pincus

A. L.

(2013). The Pathological Narcissism Inventory. In Ogrudniczuk

(Ed.), Understanding and treating pathological narcissism (pp. 93–110). Washington, DC: American Psychological Association.

23.

Pincus

A. L.

Ansell

E. B.

Pimentel

C. A.

Cain

N. M.

Wright

A. G.

Levy

K. N.

(2009). Initial construction and validation of the pathological narcissism inventory. Psychological assessment, 21(3), 365. https://doi.org/10.1037/a0016530

24.

Pincus

A. L.

Lukowitsky

M. R.

(2010). Pathological narcissism and narcissistic personality disorder. Annual Review of Clinical Psychology, 6, 421–426. https://doi.org/10.1146/annurev.clinpsy.121208.131215

25.

Rhodewalt

Eddings

S. K.

(2002). Narcissus reflects: Memory distortion in response to ego-relevant feedback among high-and low-narcissistic men. Journal of Research in Personality, 36(2), 97–116. https://doi.org/10.1006/jrpe.2002.2342

26.

Rhodewalt

Morf

C. C.

(1998). On self-aggrandizement and anger: A temporal analysis of narcissism and affective reactions to success and failure. Journal of Personality and Social Psychology, 74(3), 672–685. https://doi.org/10.1037/0022-3514.74.3.672

27.

Ritchie

T. D.

Walker

W. R.

Marsh

Hart

Skowronski

J. J.

(2015). Narcissism distorts the fading affect bias in Autobiographical memory. Applied Cognitive Psychology, 29(1), 104–114. https://doi.org/10.1002/acp.3082

28.

Rohmann

Neumann

Herner

M. J.

Bierhoff

H. W.

(2012). Grandiose and vulnerable narcissism. European Psychologist, 16, 295–302. https://doi.org/10.1027/1016-9040/a000100

29.

Ronningstam

(2009). Narcissistic personality disorder: Facing DSM-V. Psychiatric Annals, 39(3). 111–121. https://doi.org/10.3928/00485713-20090301-09

30.

Rose

(2002). The happy and unhappy faces of narcissism. Personality and IndividualDifferences, 33, 379–392. https://doi.org/10.1016/S0191-8869(01)00162-3

31.

Rubin

D. C.

Schrauf

R. W.

Gulgoz

Naka

(2007). On the cross-cultural variability of component processes in autobiographical remembering: Japan, Turkey, and the U.S.A. Memory, 15, 536–547. https://doi.org/10.1080/09658210701332679

32.

Sedikides

Gregg

A. P.

(2008). Self-enhancement: Food for thought. Perspectives on Psychological Science, 3(2), 102–116. https://doi.org/10.1111/j.1745-6916.2008.00068.x

33.

Sedikides

Rudich

E. A.

Gregg

A. P.

Kumashiro

Rusbult

(2004). Are normal narcissists psychologically healthy?: Self-esteem matters. Journal of Personality and Social Psychology, 87(3), 400. https://doi.org/10.1037/0022-3514.87.3.400

34.

Şen

Barışkın

(2019). The Standardization of the pathological narcissism inventory in the turkish language and testing its validity and reliability. Turkish Journal of Psychiatry, 30(2). 118–129. https://www.doi.org/10.5080/u23274

35.

Sommer

K. L.

Kirkland

K. L.

Newman

S. R.

Estrella

Andreassi

J. L.

(2009). Narcissism and cardiovascular reactivity to rejection imagery. Journal of Applied Social Psychology, 39(5), 1083–1115. https://doi.org/10.1111/j.1559-1816.2009.00473.x

36.

Startup

Heard

Swales

Jones

Williams

J. M. G.

Jones

R. S.

(2001). Autobiographical memory and parasuicide in borderline personality disorder. British Journal of Clinical Psychology, 40(2), 113–120. https://doi.org/10.1348/014466501163535

37.

Sümer

(2000). Yapısal eşitlik modelleri: Temel kavramlar ve örnek uygulamalar. Türk Psikoloji Yazıları, 3, 49–74. https://psycnet.apa.org/record/2006-04302-005

38.

Tabachnick

B. G.

Fidell

L. S.

(2013). Using multivariate statistics. Boston, MA: Pearson Education.

39.

Twenge

J. M.

(2014). Generation me: Why today's young Americans are more confident, assertive, entitled and more miserable than ever before, New York: Free Press.

40.

Vaillancourt

(2013). Students aggress against professors in reaction to receiving poor grades: An effect moderated by student narcissism and self‐esteem. Aggressive Behavior, 39(1), 71–84. https://doi.org/10.1002/ab.21450

41.

Walker

W. R.

Skowronski

J. J.

Thompson

C. P.

(2003). Life is pleasant—and memory helps to keep it that way!. Review of general psychology, 7(2), 203–210.

42.

Walker

W. R.

Skowronski

J. J.

Gibbons

J. A.

Vogl

R. J.

Ritchie

T. D.

(2009). Why people rehearse their memories: Frequency of use and relations to the intensity of emotions associated with autobiographical memories. Memory, 17(7), 760–773. https://doi.org/10.1080/09658210903107846

43.

Young

K. D.

Erickson

Drevets

W. C.

(2012). Match between cue and memory valence during autobiographical memory recall in depression. Psychological Reports, 111(1), 129–148. https://doi.org/10.2466/09.02.15.PR0.111.4.129-148

44.

Young

J. E.

Klosko

J. S.

(1994). Reinventing your life: The breakthough program to end negative behavior and feel great again. New York: Plume.

45.

Young

J. E.

Klosko

J. S.

Weishaar

(2003). Schema therapy: A practitioner’s guide. New York (NY): Guilford Publications.