Narcissus locked in the past: Vulnerable narcissism and the negative views of the past

We examined how the perception of past events might contribute to the understanding of vulnerable narcissism. Across seven samples ( N Grand = 1271), we investigated the association between vulnerable narcissism and individual differences in negative view of the past as well as how both were associated with basic personality traits, intrapersonal (i.e., affect, life satisfaction, and self-esteem) and interpersonal (i.e., anger, and hostility) outcomes, and memory biases of immediate life events and early life traumas. We found that vulnerable narcissism was reliably correlated with a negative view of the past. Additionally, both variables showed similar personality profiles (e.g., high neuroticism) and overlapped in explaining various outcomes, including self-esteem, anger, hostility, recalled traumas, and a negative memory bias.

The strategy of avoiding other people observed in vulnerable narcissism has been widely studied.However, less is known about how vulnerable narcissists perceive their personal experiences.For instance, frequent feelings of shame and anger are often associated with intrusive thoughts (Ghim, Choi, Lim, & Lim, 2015).Although the ruminative thinking may be an important characteristic leading to increased aggression in vulnerable narcissists (Krizan & Johar, 2015), surprisingly little attention has been paid to narcissists' tendency to concentrate on the past.Thus, within the current research we ask how vulnerable narcissists view the past.Are they excessively concentrated on the negative side of their past?What are the foundations and processes underlying the aversive thinking about the past among vulnerable narcissists?We aim to answer these questions in a series of studies examining the association between vulnerable narcissism and the negative view of the past.

How vulnerable narcissists view their past?
The crucial role of the past is a part of many theories of narcissism.The psychoanalytic concepts emphasized the importance of early experiences in childhood as the etiology of narcissistic personality (Freud, 1932;Kernberg, 1975;Kohut, 1971).For instance, Kohut (1971) suggested that inappropriate feedback from parents (both lack of approval or excessive attention) may lead to narcissistic injury and, in turn, to the creation of an unrealistic sense of self.Other psychoanalysts indicated that narcissists experience anger in response to rejection because it opens childhood wounds (Kernberg, 1975).Depending on how strong the sense of self is, it could either express itself in grandiose (i.e., approaching) or vulnerable (i.e., avoiding) ways.
Evidence from social-personality psychology is generally in line with the psychoanalytic concepts emphasizing the role of the negative past in the development of narcissism.Vulnerable narcissists report more adverse childhood experiences (Crawford & Wright, 2007;Kim & Cicchetti, 2010;Rogosch & Cicchetti, 2004), especially emotional abuse and emotional neglect (Nguyen & Shaw, 2020), which may lead to mistrust and increased antagonism in adulthood (Miller et al., 2011).In addition, they present anxious and avoidant attachment styles (Miller et al., 2011;Pistol, 1995), which are often linked to childhood maltreatment (Baer & Martinez, 2006).The focus on the negative past manifests in vulnerable narcissism also by the tendency toward the rumination on anger, that is, recalling past events evoking anger and thinking over the causes and consequences of anger episodes (Sukhodolsky, Golub, & Cromwell, 2001).Such rumination of anger could be viewed as a form of compensation and restitution for early deprivation and humiliation (Bishop & Lane, 2002).It has been suggested that the rumination of anger fueled by shame might be a potential driver of the increased levels of aggressiveness in vulnerable narcissism (Ghim et al., 2015;Krizan & Johar, 2015).
Consistent with the above, those with more vulnerable narcissism tend to have a more negative perspective on their past (i.e. the negative view of the past; Zajenkowski, Witowska, Maciantowicz, & Malesza, 2016).This negative view of the past is often accompanied by a concentration on unpleasant events from the past and a negative interpretation of past events (Zimbardo & Boyd, 1999).Interestingly, the nomological network of the negative view of the past resembles that of vulnerable narcissism in, for example, associations with high neuroticism and low extraversion (Kairys & Liniauskaite, 2015).The negative view of the past correlates also with a wide spectrum of negative emotionality, including anxiety and depression (Stolarski, Matthews, Postek, Zimbardo, & Bitner, 2014;Zimbardo & Boyd, 1999), anger and hostility (Stolarski, Zajenkowski, & Zajenkowska, 2016), low levels of self-esteem (Zimbardo & Boyd, 1999) and life satisfaction (Zhang & Howell, 2011), more severe symptoms (e.g., intrusive memories) of post-traumatic stress disorder after experiencing trauma (i.e., motor vehicle accident; Stolarski & Cyniak-Cieciura, 2016), and frame their time as a child by emphasizing parental alienation, lack of closeness, emotional remoteness, and inadequacy (Kostić, Pejičić, & Chadee, 2017).
Considering the above, vulnerable narcissism and the negative view of the past may have a similar psychological profile.However, both constructs might slightly differ at the conceptual level as well as the magnitude of associations with specific intra-and interpersonal outcomes.While people who negatively view their past seem to be more withdrawn, passive, and less reward-dependent (Zimbardo & Boyd, 1999), some studies indicate that vulnerable narcissists may display an increased tendency toward approach motivation (e.g., Miles, Smyrnios, Jackson, & Francis, 2019).Additionally, one of the central characteristics of vulnerable narcissism is antagonism, especially the frequent experience of anger and hostile attitude (e.g., Krizan & Johar, 2015;Maciantowicz & Zajenkowski, 2020a), whereas a negative view of the past is associated with well-being (Stolarski et al., 2014;Zhang & Howell, 2011).Correspondingly, both constructs are highly correlated with neuroticism, however, for vulnerable narcissism more important might be neurotic irritability while for negative view of the past the aspect related to anxiety.Examining the similarities and differences between vulnerable narcissism and negative view of the past might broaden the understanding of the former.

The current research
The primary aim of the current research was examining how the perception of past events might contribute to the understanding of vulnerable narcissism.For this purpose, we comprehensively investigated the association between vulnerable narcissism and the negative view of the past.Using data from seven samples, we tested where and to what extent individual differences in negative view of the past explain the variance of vulnerable narcissism in various areas.The current studies and hypotheses can be grouped into three overarching research questions characterized below.A summary of the hypotheses is presented in Table 1.
First, we examine how individual differences in vulnerable narcissism and negative view of the past are related to basic personality characteristics.Generally, we expect that vulnerable narcissism and negative view of the past will be positively correlated with one another (H1a) and that both will be associated positively with neuroticism (H1b) and negatively with extraversion (H1c), agreeableness (H1d), and conscientiousness (H1e; Kairys & Liniauskaite, 2015;Miller et al., 2011Miller et al., , 2018;;Rogoza et al., 2018).Furthermore, both vulnerable narcissism and past negative perceptions are characterized by high anxiety and social withdrawal (Krizan & Herlache, 2018;Zimbardo & Boyd, 1999).Therefore, we expect they will be positively linked to behavioral inhibition system (H1f).Additionally, negative view of the past was unrelated to reward dependence (Zimbardo & Boyd, 1999), while in case of vulnerable narcissism there is no clear evidence; however, some studies show its positive association with approach motivation (H1g; e.g., Miles et al., 2019).Lastly, we expected to find weak or no relationship with grandiose narcissism of both vulnerable narcissism and negative view of the past (H1h; Miller et al., 2011;Rogoza et al., 2018;Zajenkowski, Stolarski, Witowska, Maciantowicz, & Łowicki, 2016).In lower-order personality facets, we expect a difference within neuroticism: vulnerable narcissism will be more strongly associated with the aspect of volatility (H1i), while negative view of the past will correlate more strongly with the aspect of withdrawal (H1j).
Second, we examine how individual differences in vulnerable narcissism and the negative view of the past are correlated to intrapersonal (i.e., affect, life satisfaction, and self-evaluation) and interpersonal (i.e., antagonism) outcomes.With respect to well-being, we analyzed both affective experiences and life satisfaction.In case of affect, we hypothesize (H2a) that they will be associated with a generally negative affect (Maciantowicz & Zajenkowski, 2020a;Miller et al., 2011;Stolarski et al., 2014).More specifically, we expect that vulnerable narcissism and negative view of the past will be linked to high negative affect and low positive affect from the two-dimensional model of mood (Watson, Clark, & Tellegen, 1988).In the three-dimensional model, negative mood is defined as high tense arousal, and low energetic arousal and low hedonic tone (Matthews, Jones, & Chamberlain, 1990).Similarly, we expect (H2b) a negative association with life satisfaction (Miller et al., 2011;Zhang & Howell, 2011).
Another important area in which the traits can manifest themselves is related to self-evaluation.Here, we expect (H2c) that people scoring high on vulnerable narcissism and negative view of the past will display a low level of self-esteem (Miller et al., 2011;Rogoza et al., 2018;Zhang, Howell, & Stolarski, 2013;Zimbardo & Boyd, 1999).The last area we consider is correlated with antagonism.Vulnerable narcissism and a negative view of the past are associated with similar manifestations of antagonism, which mainly concentrate on internalizing aggression (Krizan & Johar, 2015;Maciantowicz & Zajenkowski, 2020b;Stolarski et al., 2016).Thus, we expect that both dimensions will be positively correlated with the frequent experience of anger (i.e., aggression triggering emotion; H2d) and hostility (i.e., aggressive interpretation of other people's intentions; H2e).
Third, we explore two potential processes (i.e., memory bias and traumatic experiences) that might be responsible for the heightened tendency to think in an aversive way about the past.For example, people's negative views of the past might stem from the actual traumatic experiences of a person.Thus, we examine the hypothesis that people with high levels of vulnerable narcissism and people with high levels of past negative views report more early traumatic experiences than people low on vulnerable narcissism and low on past negative (H3a).Alternatively, we investigate the possibility that the concentration on the past might be a function of memory bias.We expect that people with highly vulnerable narcissism, as well as people with a tendency to view their past negatively, are more susceptible to recall negative, rather than positive, events (H3b).

Participants and procedure
Sample 1. 1 Via a snowball recruitment of in-person participants, we sampled 233 (123 women, 110 men) undergraduate students between the ages of 18 and 39 (M = 23.62;SD = 3.80) who were tested individually in a laboratory at the University of [blind].A set of questionnaires and cognitive tasks were completed by participants after providing informed consent.A small gift (e.g., a cup ≈ 10€) was offered for participation.The data was collected as part of a larger project (Zajenkowski et al., 2016).
Sample 2. Via a snowball recruitment of in-person participants, we sampled 199 participants (99 women, 100 men) that were Polish undergraduate students between the ages of 18-40 (M = 22.80; SD = 3.78) who were tested individually in a laboratory at the University of [blind].Participants provided informed consent and were offered a cash reward (≈ 10€) for participation.Participants completed a set of questionnaires and cognitive tasks.The data was collected as part of a larger project (Witowska & Zajenkowski, 2019).
Sample 3. Via a snowball recruitment of online participants, we sampled 195 participants (163 women, 32 men) between the ages of 18 and 50 (M = 24.00;SD = 4.83).They were mostly (52%) undergraduate students and professionals (48%).After giving their informed consent, participants completed a set of questionnaires.
Sample 4. Via a snowball recruitment of online participants, we

Table 1
Summary of the expected correlations between vulnerable narcissism, negative view of the past, and external variables and their corresponding hypotheses.
Vulnerable Narcissism

Negative View of the Past
Vulnerable Narcissism/Past Negative (H1a)  1 The studies presented in the manuscript were not pre-registered.The summary of the variables and measures used in each study are presented in Table 2.
sampled 216 (115 women; 91 men; 10 nonresponsive) undergraduate students from various universities in Warsaw between the ages of 18-54 (M = 25.22;SD = 7.09) who were tested individually in a laboratory at the University of [blind].After giving their informed consent, participants filled a set of questionnaires.Participants were offered a small gift (e.g., a cup worth ≈ 10€) in exchange for participation.Sample 5. Via a snowball recruitment of in-person participants, we sampled 267 participants (132 women; 131 men; 4 nonresponsive) between the ages of 18 and 49 (M = 23.31;SD = 4.97) who were tested individually in a laboratory at the University of [blind].Participants were recruited via website announcements and offered cash for participation (≈ 20€).Upon giving informed consent, participants completed a packet of questionnaires and cognitive tasks.The data was collected as part of a larger project (Zajenkowski, Czarna, Szymaniak, & Dufner, 2020).Sample 6.During classes at the University of [blind] and the [blind], we recruited, in person, 84 (44 women, 38 men) undergraduate students who were aged between 19 and 27 (M = 21.16SD = 1.72).Students who gave their consent completed a set of questionnaires during a lecture.
Sample 7. Via a snowball recruitment of in-person participants, we recruited 77 (32 women; 45 men) undergraduate students aged between 18 and 35 (M = 23.25;SD = 3.54), who were tested individually in a laboratory at the University of [blind].After completing paper questionnaires, participants were asked to recall a recent memory (see Measures below).A small gift (e.g., a cup worth ≈ 10€) was offered for participation in the study.

Measures
Vulnerable Narcissism.In all studies, we assessed vulnerable narcissism using the Polish translation (Czarna, Dufner, & Clifton, 2014) of the Hypersensitive Narcissism Scale (Hendin & Cheek, 1997).Participants reported their agreement (1 = strongly disagree; 5 = strongly agree) with 10 items (e.g., I often interpret the remarks of others in a personal way) that were summed to serve as indexes of vulnerable narcissism (Cronbach's αs = 0.68-0.80).
Negative View of the Past.In all studies, we measured individual differences in negative view of the past with the Past Negative scale from Polish translation (Kozak & Mażewski, 2007) of the Zimbardo Time Perspective Inventory (Zimbardo & Boyd, 1999).Participants reported their agreement (1 = strongly disagree; 5 = strongly agree) with 10 items (I think about the bad things that have happened to me in the past) that were averaged to capture individual differences in negative view of the past (αs = 0.80-0.88).
Grandiose Narcissism.In all studies except 3, we measured individual differences in grandiose narcissism with the Polish translation (Bazińska & Drat-Ruszczak, 2000) of the Narcissistic Personality Inventory (Raskin & Hall, 1979).Participants reported how much 34 items (e.g., I will be a success) applied to them (1 = does not apply to me; 5 = applies to me).The items were summed to create indexes of grandiose narcissism (αs = 0.91-0.93).In Study 3, we used the reduced version of this scale with 13 items (Gentile et al., 2013) in its Polish form ( Żemojtel-Piotrowska et al., 2018) where participants choose between two items, one that reflects a narcissistic statement (e.g., I insist upon getting the respect that is due me) and the other that does not (e.g., I usually get the respect I deserve).Again, items were summed consistent with the parent-measure and prior research.
Approach/Avoidance Motivation.In Study 2, we assessed individual differences in approach and avoidance motivations with the Polish translation (Muller & Wytykowska, 2005) of the BIS/BAS scale (Carver & White, 1994).Participants indicated how true (1 = very true for me; 5 = very false for me) 24 statements were to them in terms of behavioral inhibition ([7 items; α = 0.81], e.g., Criticism or scolding hurts me quite a bit), reward responsiveness ([5 items; α = 0.55], e.g., When I see an opportunity for something I like I get excited right away), drive ([4 items; α = 0.75], e.g., If I see a chance to get something I want I move on it right away), and fun-seeking ([4 items; α = 0.65], e.g., I will often do things for no other reason than that they might be fun).Items were summed to create indexes of each trait.
Life Satisfaction.We assessed life satisfaction twice.In Study 1, we used the Polish translation (Jankowski, 2015) of the Satisfaction with Life Scale (Diener, Emmons, Larsen, & Griffin, 1985).It consists of five statements (e.g., In most ways my life is close to my ideal) where participants reported how true (1 = very untrue; 7 = very true) that were summed to create a dispositional measure of life satisfaction (α = 0.85).In addition, in Study 2, we measured it with the Polish translation (Byra, 2011) of the Temporal Satisfaction with Life Scale (Pavot et al., 1998).Participants reported their agreement (1 = strongly disagree; 7 = strongly agree) with 15 items (e.g., I will be satisfied with my life in the future) that were summed to measure participants' life satisfaction related to the past (α = 0.80), present (α = 0.86), and future (α = 0.79).
Self-Esteem.In Study 3, we assessed individual differences in selfesteem using the Polish translation of Rosenberg (1965) Self-Esteem Scale.Participants were asked their agreement (1 = strongly disagree; 4 = strongly agree) with 10 items (e.g., I feel that I have a number of good qualities) that were summed (α = 0.89).
Anger.We assessed anger with two different self-report questionnaires.In Study 1, we used the Polish translation (Bąk, 2016) of the trait and state anger subscales of the State-Trait Anger Expression Inventory (Spielberger, 1999).The trait subscale contained 10 statements (e.g., I get angry when I have to wait because of other's mistakes) that were summed (αs = 0.80 to 0.88), while in the state subscale there were 15 items (e.g., I feel angry) that were summed (α = 0.93).Participants rated how much they agreed (1 = strongly disagree; 4 = strongly agree) with each statement.In Studies 4 and 5 we used the Polish translation of the anger subscale of the Aggression Questionnaire (Buss & Perry, 1992).Participants reported the accuracy (1 = very inaccurate; 5 = very accurate) of seven statements (e.g., When frustrated, I let my irritation show) which were summed to create measure of anger (αs = 0.68 to 0.86).
Hostility.In Studies 4 and 5, we measured individual differences in hostility using the hostility subscale of the Aggression Questionnaire (Buss & Perry, 1992).It contains eight items (e.g., I am sometimes eaten up with jealousy), where participants rate their answers on a five-point scale (1 = very inaccurate; 5 = very accurate).The items were summed to create the index of hostility (αs = 0.77 to.82).
Early Life Traumatic Experiences.In Study 6, we assessed individual differences in recalled early life negative experiences.We used Polish version (Sokołowski & Dragan, 2017) of the Early Life Stress Questionnaire (Cohen et al., 2006).Participants reported whether (yes/ no) they experienced any of 19 (e.g., negligence, parental death) adverse events in before the age of 12.The sum of adverse events gives a global index of early life trauma (α = 0.69).
Memory Bias.In Study 7, we assessed biases in memory with a memory recall task.Participants were asked to recall and vividly imagine any memory related to their recent trip or journey.Subsequently, they were instructed to "please write a short story about your recent trip/journey.It could be to anywhere, to a store, another city, or another country.Keep writing until the experimenter says stop." Participants had five minutes to complete the task.The stories provided by the participants were then assessed for valence (1 = very negative; 7 = very positive) by three independent, blind raters (α = 0.96).

Analytic plan
First, for H1a, we calculated effect sizes estimates using meta-analysis of correlation coefficients across seven studies.Next, Hypotheses 1b to 1j were tested using correlation as well as difference in correlations (Steiger's z).Because the Big Five traits and grandiose narcissism were assessed in each sample, we report meta-analytically derived correlations for robustness.Hypotheses 2a -3b were tested using Pearson's correlation, simultaneous linear regression and commonality analyses.

Results
We first examined the correlation between our two main variables: vulnerable narcissism and negative views of the past.As both variables were included in all studies, we conducted a meta-analysis of their relations.Given the variation across the studies (I 2 = 49.1%),we used the random effects model, which revealed a positive and relatively large association between vulnerable narcissism and past negative perspective (Cohen's d = 0.46 [95% CI 0.39,0.53];p < .001)thus, confirming H1a.
Second, we examine the nomological network of vulnerable narcissism and a negative perspective on one's past (Table 3). 3 Vulnerable narcissism and past negative were associated with four traits from the Big Five model: positively with neuroticism and negatively with extraversion, agreeableness, and conscientiousness (H1b to H1e).
Additionally, vulnerable narcissism was more strongly correlated with agreeableness, whereas past negative was more strongly linked with conscientiousness.Both vulnerable narcissism and past negative were positively correlated with BIS (H1f).Additionally, vulnerable narcissism was correlated with BAS Reward (H1g).Grandiose narcissism was weakly associated with both constructs (H1h); however, we found a small positive correlation for vulnerable narcissism and a small negative one for past negative.The analysis of the Big Five facets revealed that vulnerable narcissism and past negative tap slightly different aspects of neuroticism, since the latter showed stronger correlation with withdrawal (H1j), while for volatility we did not find differences in correlations (H1i).
Third, we considered how vulnerable narcissism and negative views of the past are related to intrapersonal and interpersonal processes (Table 4).Vulnerable narcissism and past negative were generally negatively related to the indicators of well-being (H2a -H2e), except for a null association with positive affect.Additionally, they also correlated negatively with self-esteem and positively with the indicators of antagonism.The associations did not differ in most cases; however, past negative was more strongly correlated with life satisfaction than vulnerable narcissism.
Fourth, given the correlation between vulnerable narcissism and negative view of the past, we also tested simultaneous linear regression and commonality analyses (Table 4).We report the results of the commonality analysis, which allocates the total variance explained by the two predictors into variance unique to each predictor and their common variance.We only interpreted the commonality coefficients of the significant models.The results revealed several patterns.Individual differences in having negative views of one's past was the dominant predictor of well-being (both affect and life satisfaction) and self-esteem (our largest effects at 57%), apart from energetic arousal.In contrast, individual differences in vulnerable narcissism were the dominant predictor of antagonism (anger and hostility).Vulnerable narcissism and
3 Descriptive statistics, and correlations as well as the raw data and the script of statistical analyses are presented as supplementary material available at the OSF project site (https://osf.io/j8r73/?view_only=b50d2e9e584f41989fc7d6f0 b335c0a1).
past negative had least in common in case of well-being, with one exception of negative affect.
Lastly, we consider two potential processes (i.e., memory bias and experienced trauma) linking vulnerable narcissism and a negative view of the past (Table 2).Our evidence is consistent with the hypotheses, suggesting that those characterized by a tendency to remember more negative events in one's past (H3a) and their immediate life (H3b) were both more vulnerably narcissistic and to have a negative view of one's past, the correlations that did not differ across each trait.Specifically, both predictors shared almost half (48.02%) of the explained variance in the memory bias, however, vulnerable narcissism uniquely explained only marginal amount of variance (2.51%), while past negative perspective explained the lion's share (49.46%) of the memory bias variance.While vulnerable narcissism and past negative uniquely explained some of the experienced trauma's variance, they also had much in common (44.33%). 4

Discussion
The idea that past experiences shape individual differences in narcissism is at the heart of various conceptualizations of narcissism, from the early days of psychoanalysis (Freud, 1932;Kernberg, 1975;Kohut, 1971) to contemporary research (Thomaes & Brummelan, 2018).In the current study, we comprehensively examined the relationship between vulnerable narcissism and the tendency to view the past in a negative way.Across seven studies, we found a relatively large association between vulnerable narcissism and negative view of the past.Additionally, these two variables share numerous psychological correlates in terms of personality traits, intra-and interpersonal outcomes, as well as specific memory processes.
We found that vulnerable narcissism and negative view of the past have similar personality profiles.Specifically, they were equally correlated with high levels of neuroticism and behavioral inhibition along with low levels of extraversion.They were also negatively associated with agreeableness and conscientiousness; however, they were slightly different in the magnitude of the correlations.Vulnerable narcissism was more strongly associated with agreeableness, while negative view of the past was more correlated with conscientiousness.These results suggest that antagonism is more characteristic of vulnerable narcissists, especially low empathy and concern for others as indicated by their lower levels of the compassion facet (Zajenkowski & Szymaniak, 2019).Additionally, in contrast to past negative time perspective, vulnerable narcissism was associated with more reward sensitivity as part of the behavioral activation system, which might underlie its positive link with grandiose narcissism.Although negative view of the past and vulnerable narcissism had similar personality profiles, they differed in intrapersonal and interpersonal outcomes.In general, the past negative time perspective was more important for well-being than vulnerable narcissism.This is consistent with prior research showing that past negative perspective is a strong predictor of affective states, life satisfaction, and self-esteem even after controlling for other personality traits (Stolarski & Matthews, 2016;Zhang, Howell, 2011).There were no clear differences between past negative and vulnerable narcissism with respect to their unique contributions to other outcomes.However, we observed their high degree of overlap in explaining the variance of some variables.The most striking were the models with self-esteem and hostility where both predictors explained relatively large part of the variance, while having also much in common.
The possibility that there might be common processes underlying vulnerable narcissism and a negative view of the past was further explored in two studies.We found that people with high levels of vulnerable narcissism and those with especially negative views of their past reported more traumas; both variables explained 34% of the variance in early traumatic experiences with almost equal unique and relatively large, common contribution.The results are consistent with previous research showing that vulnerable narcissists recall more negative experiences from their childhood (Crawford & Wright, 2007;  4 Although not included in our hypotheses, we also assessed how vulnerable narcissism predicts experienced traumas over grandiose narcissism.Vulnerable narcissism (β = 0.30, p < .01)and grandiose narcissism (β = 0.37, p < .01)predicted more early life traumas.Interestingly, they explained different variance of the traumas, having only 5.98% in common.Kim & Cicchetti, 2010;Rogosch & Cicchetti, 2004;Nguyen & Shaw, 2020), and display maladaptive attachment styles (Miller et al., 2011).Likewise, the negative view of the past has been linked to a more negative relationship with parents (Kostić at al., 2017) and more severe symptoms after a traumatic event (Stolarski & Cyniak-Cieciura, 2016).However, no study has tested the direct association between negative view of the past and the recalled traumas until now.
Although the questionnaire we used to measure recalled traumas asked about relatively objective events from one's childhood (e.g., parents' divorce), it might have captured also, at least in some cases (e. g., negligence, emotional abuse), the subjective interpretation of the past.Indeed, the concept of past negative perspective implies that the source of an aversive attitude toward the past might be real traumatic experience, or it can be a consequence of negative reconstruction of the past events (Zimbardo & Boyd, 1999).Thus, the question is to what extent vulnerable narcissism and past negative are associated with a memory bias of focusing on negative events.In the last of our studies, we asked participants to recall an event (without indicating its emotional content) from their recent past.People scoring high on vulnerable narcissism and people with higher scores on negative view of the past spontaneously recalled more negative memories than people with low levels of these traits.
Our study was the first to directly examine people's tendency to remember more negative events of one's immediate life in relation to past negative time perspective as well as vulnerable narcissism.This finding contributes to the literature in several ways.First, it supports the concept that people with a highly past negative perspective tend to retrieve from their memory events that are aversive and negative (Zimbardo & Boyd, 1999).Second, past negative and vulnerable narcissism largely overlapped in explaining memory bias (almost 50%) and vulnerable narcissism did not add much (2.5%)beyond the common variance.Thus, it is possible that the process underlying memory bias in vulnerable narcissists might be like the one described among those with a highly past negative perspective.It has been suggested that there might be specific processes of autobiographical memory among people oriented on the negative past (Matthews & Stolarski, 2015).The theory of autobiographical memory suggests two functions in the forms of adaptive correspondence (i.e., accurate records of experience) and selfcoherence (i.e., the need to be consistent with self-concept; Conway, Singer, & Tagini, 2004).These two functions are in tension, because selfcoherence distorts experiences and reconstructs them to fit the selfschema.A negative view of the past might be associated with the availability or accessibility of memories consistent with the pessimistic self-concept (Matthews & Stolarski, 2015).Likewise, vulnerable narcissists might have easier access to negative memories.An illustrative example might be one of the stories rated as highly negative, written by a participant with a relatively high level of vulnerable narcissism (i.e., the top quartile): Recently, I went to the city to earn some money.While going, I was thinking: what is for all this, nothing makes sense, the same thing over and over again, getting up early, slaving etc., no prospect for a better life, I feel like vomiting on all this; the others are resting, they lead a nice life, and here only gray, boring everyday life.
On average, such negative stories were more available for vulnerable narcissists.However, the interaction between self-concept and memory might be dynamic and reciprocal, as the retrieval of negative events may influence the person's sense of who they are currently (Conway & Pleydell-Price, 2000).Thus, the vulnerable narcissists' current selfconcept reconstructs past memories and the past reconstructs their self (Conway & Pleydell-Price, 2000).
Collectively, our findings revealed the importance of individual differences in the negative view of the past to understand individual differences in vulnerable narcissism.We found that the two constructs overlapped especially in the case of antagonism (i.e., hostility) and low self-esteem.It is likely that the concentration on adverse experiences from one's past drives malevolence and negative self-evaluation among vulnerable narcissists.Such a possibility has already been suggested for anger rumination which may lead to aggressiveness in vulnerable narcissists (Ghim et al., 2015;Krizan & Johar, 2015).Our findings further support this view by showing that vulnerable narcissists tend to recall negative events from their immediate life as well as their childhood.However, we examined each outcome in a separate study, thus, more research including all important variables (i.e., memory bias, recalled traumas, hostility etc.) is needed to fully understand the role of concentration of the negative past in vulnerable narcissism.

Limitations and conclusions
Our research provided new insights into how vulnerable narcissism is related to the negative view of one's past, it was nonetheless, limited.First, our studies were correlational and examined each association between vulnerable narcissism, negative view of the past, and various outcomes in independent samples.Thus, we were not able to test models including several variables of interest or draw causal conclusions.Future studies should use an experimental procedure to examine whether, for instance, biases in memory lead to more hostile interpersonal style among vulnerable narcissists.
Second, our effect sizes were relatively large in most cases (Gignac & Szodorai, 2016).We observed large correlations with some of the personality factors (e.g., neuroticism and behavioral inhibition system) as well as intra-(e.g., self-esteem) and interpersonal outcomes (e.g., hostility).Negative view of the past and vulnerable narcissism jointly explained a relatively large amount of variance (>30%) in self-esteem, hostility, and traumas.However, there were smaller correlations with affect, which may suggest that our main predictors are more related to dispositional factors rather than momentary states.Third, while we had generally good internal consistencies for our measures (i.e., αs' ≥ 0.70; Nunnally, 1978), some of our scales had only fair internal consistency estimates (i.e., α's ≈ 0.60).However, our findings were consistent with prior research and most of them were replicated across the current studies.Fourth, we did not include any mechanism for identifying invalid responding, while implementing such procedure is recommended, especially for online surveys (like our Studies 3 and 4; Al-Salom & Miller, 2019).Fifth, in the current research only some of the variables (e.g., personality traits) were systematically assessed across all samples.For other variables, we had data from one or two samples and were not able to report meaningful meta-analytic effects for these variables.Lastly, it needs to be acknowledged that both our main variables, vulnerable narcissism, and negative view of the past have been found to highly overlap with neuroticism (Kairys & Liniauskaite, 2015;Miller et al., 2018).However, they also share some specific variance, beyond neuroticism (Zajenkowski et al., 2016).Moreover, while neuroticism is highly associated with general affect (e.g., negative mood, tension), vulnerable narcissism predicts specific states (e.g., anger) over neuroticism (Maciantowicz & Zajenkowski, 2020b).Similar findings pertain to negative view of the past, which predicts well-being over neuroticism (Zhang & Howell, 2011).Nonetheless, future studies could examine more specific content of memories, rather than just simple categorization on negative/positive valence as we did in Study 7.This would reveal whether vulnerable narcissists just concentrate on generally negative events, or events that are also damaging to their sense of self.
In conclusion, the current research revealed that the tendency to concentrate on negative events from the past is a crucial aspect of vulnerable narcissism.This tendency might stem from a bias in memory processes and may influence one's perception of themselves and others.Furthermore, a negative view of the past might explain various characteristics of vulnerable narcissists such as low self-esteem or increased levels of anger and hostility.

Table 2
Number of participants, variables, and measures used in each study.

Table 4
Correlations between vulnerable narcissism and negative past to well-being, self-evaluation, and antagonism.Note.VN = Vulnerable narcissism; NP = Negative view of the past; AQ = Aggression Questionnaire; PN = Past Negative; STAXI = State Trait Anger Expression Inventory; VN = Vulnerable narcissism.All regression models were significant at p's < 0.001, besides the model with positive affect and tense arousal (p > .05).For traumas, we report Poisson regression.*p < .05,**p < .01.