Psychological consequences of individual narcissism

Social media addiction

Some research has explored whether narcissism may affect the prevalence or magnitude of addictions.  For example, one study, conducted by Nguyen et al. (2025), examined the degree to which narcissism is associated with addiction to social media.  The participants were members of Generation Z and lived in Ho Chi Minh City. The study revealed that, in this population, narcissism was positively associated with social media addiction. 

Arguably, people who exhibit narcissism are fundamentally motivated to boost their status.  To achieve this goal, these individuals, especially if they exhibit grandiose narcissism, crave opportunities to broadcast their contributions or achievements. Social media offers individuals many opportunities to boost their status and is thus reinforcing to narcissistic people, even to the degree that addictions often evolve.

Other addictions

Narcissism is also associated with other addictions as well. To illustrate

• vulnerable narcissism in young adults predicts problem drinking and gambling in the future; feelings of shame tend to mediate these relationships (Bilevicius et al., 2019),
• narcissism in adults, especially if combined with neuroticism, is positively associated with smartphone addiction (Pearson & Hussain, 2017),
• various dimensions of narcissism, such as narcissism admiration and narcissism rivalry, are positively associated with exercise addiction—partly because exercise may enable these individuals to outperform rivals, enhance their appearance, and feel superior (Zeigler-Hill et al., 2021).

Introduction

For several reasons, narcissism may compromise driver safety.  To illustrate, narcissistic people often believe they should be treated as special.  Consequently, these individuals may not feel as obliged as other people to follow traffic rules and regulations; narcissistic people might, for example, feel they are entitled and able to drive fast.  Likewise, if narcissistic people do not feel they were treated appropriately on the road, they may be especially susceptible to anger and aggression (for similar arguments, see Bushman et al., 2018). 

Many studies have indeed explored whether aggressive or risky driving can be partly ascribed to narcissism.  In general, these studies confirm the assumption that narcissistic motorists tend to drive more aggressively (e.g., Edwards et al., 2013; Mithen et al., 2023) as well as perpetrate other risky driving behaviours as well (Hill, 2016). To illustrate

• narcissistic individuals tend to interpret the behaviours of other drivers, such as motorists who disrupt the flow of traffic while completing a U turn, as intentional and inconsiderate, provoking anger (Lustman et al., 2010),
• some facets of narcissism in particular—such as feelings of entitlement, the belief they are superior, and a tendency to perceive themselves as a leader—are especially likely to predict anger while driving (Hennessey, 2016).

Studies on driving simulators

Researchers have utilised a range of approaches to explore this association between narcissism and driving.  For example, in a study that Bushman et al. (2018) published, 60 university students operated a driving simulator.  First, these participants completed a short version of the Narcissistic Personality Inventory, comprising 16 items, to assess grandiose narcissism.  For example, to answer one item, these individuals needed to decide which of two statements are correct: “I insist upon getting the respect that is due me” or “I usually get the respect I deserve”.  In addition, participants completed a measure of trait aggression—an attribute that was included as a control variable.

Next, participants operated a realistic driving simulator and were instructed to complete a route as rapidly as possible to earn a gift card. The simulated traffic included at least five events that might elicit frustration, such as a traffic jam, a construction zone, a red traffic light, and a car that veered in front of the driver.  The researchers measured actions that could be deemed as aggressive, such as tailgating, speeding in response to specific events, driving on the shoulder of a road or on the other side, overuse of the horn, verbal insults, and aggressive gestures.  Narcissism was positively associated with most of these signals or measures of aggression.

The role of risky decisions

Narcissistic individuals often gravitate to risky behaviours—a tendency that might also increase the prevalence of driving incidents. As evidence of this possibility, in one study that explored the association between narcissism and problem gambling, Lakey et al. (2008) revealed that narcissistic individuals tend to overestimate their confidence, tolerate greater risks, and prioritise immediate rewards over future complications.  Accordingly, narcissistic individuals may overestimate their driving capability, tolerate risks on the road, and prioritise the immediate benefits of speeding over the potential to cause accidents. 

Consistent with these premises, Hill (2016), at West Chester University, confirmed that narcissism does indeed coincide with a range of risky behaviours.  In this study, to measure grandiose narcissism, participants completed the Narcissistic Personality Inventory. Next, these individuals answered a series of questions about the degree to which they gravitate to various risky behaviours.  For example, these participants indicated the degree to which they engage in risky driving behaviours, such as not wearing a seatbelt, driving while intoxicated, or texting while driving.  As the findings revealed, even after controlling demographics such as gender and ethnicity

• narcissism was positively associated with the inclination to perpetrate risky driving behaviours,
• narcissism was also positively related to other risky behaviours, such as heavy alcohol use, marijuana use, and other drug use.

The role of impulsivity

According to Dobrucalı and Özkan (2021), some of the risky or aggressive behaviours that coincide with narcissism could be ascribed to impulsivity.  That is, narcissistic individuals tend to orient attention to the potential benefits, instead of the adverse consequences, of their actions.  Therefore, rather than plan carefully to minimise adverse consequences, narcissistic people may choose risky options with minimal consideration and contemplation, manifesting as impulsive behaviour.  Because of this impulsivity, narcissistic motorists may respond somewhat recklessly and aggressively to frustrations on the road, such as a driver who veers in front of their car—rather than consider their response more prudently and carefully (see also Lustman, 2011).

To assess these arguments, in one study, Dobrucalı and Özkan (2021) administered a survey to 304 individuals online.  The survey included

• the Five-Factor Narcissism Inventory (Glover et al., 2012)—a tool that generates estimates of grandiose or overt narcissism and vulnerable or covert narcissism,
• the Barrat-Impulsiveness Scale (Patton et al., 1995)—an instrument that measures the extent to which individuals act without planning as well as fail to sustain attention on their task,
• the Driving Anger Expression Inventory (Deffenbacher et al., 2002)—a measure that gauges the frequency with which motorists, while driving, are inclined to demonstrate verbal aggression, to demonstrate physical anger, and to use the vehicle to express this anger.

The analyses revealed some interesting patterns.  First, vulnerable narcissism in particular was positively associated with the tendency of drivers to use the vehicle to express their anger, such as drive aggressively when frustrated.  Second, impulsivity, and specifically the inability of some individuals to sustain attention on their task, mediated this association.  Finally, grandiose narcissism amplified these relationships.       

Narcissism may also increase the incidence of problem gambling.  Indeed, many studies have explored the reasons that narcissism may culminate in gambling addiction and other problems around gambling.  For example, as Pace et al. (2021) revealed, because of their obsession with status, some narcissistic individuals experience the profound motivation to outperform or surpass other people, called hyper-competitiveness.  Consequently, these individuals may tolerate extraordinary risks to seek this goal—such as gamble huge amounts of money to earn the cash they need to purchase expensive goods.   

The role of perceived luck

The impact of narcissism may depend on the motivations of individuals to gamble.  Over recent decades, researchers have delineated three main pathways or motivations that culminate in problem gambling (e.g., Blaszczynski & Nower, 2002; Nower et al., 2022). 

• First, some people gravitate to gambling initially to fulfill social or recreational needs but, in response to conditioning and distorted cognitions, experience an increasingly more intense motivation to gamble more frequently.
• Second, other people gravitate to gambling primarily to circumvent unpleasant mood states—mood states that often emanate from a blend of adverse life events and impaired coping skills.
• Third, other people gravitate to gambling because of heightened impulsivity or a tendency to prioritise rewards over costs—inclinations that may coincide with antisocial personality, substance abuse, or attention deficits.

However, cognitive biases or distortions tend to underpin or amplify all three pathways—the behaviourally conditioned, emotionally vulnerable, and antisocial impulsive people (Blaszczynski & Nower, 2002).  For example, some individuals perceive themselves as lucky or fortunate.  They feel they are likely to thrive during games of change. This perception can augment the incidence of problem gambling (e.g., Wohl et al., 2005).  Arguably, narcissist individuals, who perceive themselves as special and maintain distorted beliefs, may be more inclined to perceive themselves as lucky and thus could be more susceptible to problem gambling. 

To explore this premise, Lewi and Bonnaire (2024) conducted a study of 88 individuals who, at least weekly, participate in strategic gambling.  Strategic gambling refers to circumstances in which the decisions of individuals benefit from some knowledge, skills, or experience (Larkey et al., 1997), such as poker or sports betting rather than slot machines or lotteries.  These individuals completed three sets of measures:

• First, all participants completed the French variant of the Problem Gambling Severity Index (Ferris & Wynne, 2001)—nine items that assess the frequency with which individuals report behaviours that are emblematic of problem gambling.
• Second, to assess grandiose narcissism, these individuals completed the Narcissistic Personality Inventory.
• Finally, participants completed the Personal Luck Usage Scale (Wohl et al., 2011) to gauge the degree to which they can utilise their personal luck to succeed in gambling.  A typical item is “When I am gambling, I feel like I can use my own luck to increase my chances of winning”.

As mediation analyses revealed, narcissism was positively associated with problem gambling, and perceived luck partly, but not fully, mediated this relationship.  Thus, the belief they are special and lucky is one reason, but not the sole reason, that narcissistic individuals are more susceptible to problem gambling. 

The role of other cognitive biases

Studies have also explored other biases and distortions that could explain the association between narcissism and gambling.  To illustrate, in one of the three studies that Lakey et al. (2008) reported, 351 undergraduate students completed a questionnaire that included the Narcissistic Personality Inventory and a measure of self-esteem and the Diagnostic Interview for Gambling Severity (Winters et al., 2002).  In addition, this questionnaire included two measures that assess biases that may affect gambling, such as overconfidence: the Georgia Gambling Task and the Iowa Gambling Task.  To complete the Georgia Gambling Task (Goodie, 2003),

• participants answer a set of trivia questions in which they need to choose one of several options
• for each question, participants indicate the probability they answered correctly,
• after they answer these questions, participants are asked whether they would like to bet on whether each of their choices was correct; the bets are calibrated to be fair provided their confidence matches the proportion of correct answers.

In addition, participants completed the Iowa Gambling Task (Bechara et al., 1994)—a task, usually administered online, that assesses the tendency of some people to prioritise the possibility of rewards over the potential of larger costs.  Typically, four decks of cards are presented, face down, labelled A, B, C, and D, each comprising 40 cards.   Participants begin with a specific amount of money, such as $1000.  This amount is specified on the screen, sometimes above the decks.

The participants then select a card from any of the four decks.  Each card can generate a reward, a penalty, or both.  After a card is selected, this reward and penalty appears on the screen for several seconds.  To illustrate, one card might generate a reward of $100, a penalty of $150, and thus a net loss of $50.  Their overall amount would then shift from $1000 to $950. Participants then repeat this procedure many times. 

Unbeknownst to participants, the probability of a reward and penalty as well as the distribution of these rewards and penalties varies across the decks.  That is,

• two of the decks, over time, are more likely to incur losses than are the other two decks,
• specifically, the two disadvantageous decks, on average, might lose $25,
• the two advantageous decks, on average might gain $25. 

Nevertheless, the losses that each deck incurs vary across trials and, therefore, is hard to decipher at first.  To illustrate

• for one of the advantageous decks, each card generates a gain of $50, and one in two cards incurs a loss of $50. 
• for the other advantageous deck, each card generates a gain of $50, and one in ten cards incurs a loss of $250
• for one of the disadvantageous decks, each card generates a gain of $100, and one in ten cards incurs a loss of $1250
• finally, for the other disadvantageous decks, each card generates a gain of $100, and one in every two cards incurs losses that range from of $150 to $350, although variations to these schemes can be included. 

Typically, at the outset, most participants will choose the disadvantageous decks, because the gains are higher.  Over time, usually within the first 40 trials, participants learn to choose the advantageous decks.  Nevertheless, some participants do not learn to choose the advantageous decks—a tendency that implies these individuals are unduly insensitive to possible losses or costs.

In the study that Lakey et al. (2008) conducted, narcissistic personality was positively associated with gambling severity, overconfidence on the Georgia Gambling Task, the inclination to accept bets on the Georgia Gambling Task, and insensitivity to loss in the Iowa Gambling Task.  Furthermore, overconfidence, bet acceptance, and insensitivity to loss partly mediated the association between narcissistic personality and gambling severity.  As these results imply,

• narcissistic people may overestimate their likely performance on tasks,
• narcissistic people may be unduly overlooking the costs of their options and instead orient their attention only to potential rewards,
• these tendencies of narcissistic people may culminate in problem gambling.

The role of shame

Besides cognitive biases and distortions, emotional problems may also explain why narcissism could increase the incidence of problem gambling.   To illustrate

• in contrast to grandiose narcissism, vulnerable narcissism, in which individuals often depict themselves as victims and experience powerful negative emotions in response to criticism or other personal challenges, often coincides with feelings of shame (e.g., Ritter et al., 2014),
• that is, people who experience vulnerable narcissism are motivated to establish their status or importance, but often fail to achieve this goal, culminating in shame,
• because of this shame, these individuals often gravitate to tasks in which they can divorce themselves from the reality of their lives—a sense of escape or diversion that regulates their emotions—such as problem gambling (Schlagintweit et al., 2017).

To verify this possibility, Bilevicius et al. (2019) administered a survey to 497 university students, 210 of whom completed another survey one month later.   Specifically, at the first time, the survey included the Pathological Narcissism Inventory—a measure that can generate a measure of both grandiose narcissism and vulnerable narcissism.  At the second time, the survey included the Rizvi Shame Inventory (Rizvi, 2010), in which participants indicate which of various unpleasant events they experienced, such as broke a promise, and the level of shame these experiences evoked.  Furthermore, at both times, participants completed the Problem Gambling Severity Index as well as measures of life stress and excessive alcohol intake.  As structural equation modelling revealed,

• vulnerable, but not grandiose, narcissism at one time predicted an increase in problem gambling over time,
• as hypothesised, shame partly mediated this association between vulnerable narcissism and problem gambling.

The role of emotional regulation

Whereas shame appears to partly mediate the association between vulnerable narcissism and problem gambling, other emotional tendencies could mediate the association between grandiose narcissism and problem gambling. That is individuals who experience grandiose narcissism inflate their capabilities and contributions. Therefore, these individuals may

• experience powerful emotions when offended—and therefore not be able to regulate their behaviours or concentrate when upset,
• like to perceive themselves as resilient and thus feel guilty about their unpleasant feelings or doubts,
• because of their tendency to disregard unpleasant feelings or doubts, be oblivious to their emotions and therefore not be able to modify their mood states effectively.

Because they cannot regulate their emotions proficiently, people who experience grandiose narcissism may gamble either because they want to shun these feelings or because they cannot regulate their impulses. To validate this possibility, Rogier and Velotti (2018) published a study in which 178 adults completed

• the Pathological Narcissism Inventory to measure both grandiose narcissism and vulnerable narcissism,
• the South Oaks Gambling Screen to gauge the severity of their gambling problems (Lesieur & Blume, 1987),
• and the Difficulties in Emotion Regulation Scale (Gratz & Roemer 2004; for a shorter version, see Bjureberg et al., 2016 ) to assess the capacity of these individuals to regulate unpleasant emotions well—exemplified by items like “When I’m upset, I lose control over my behaviours”.

Both grandiose narcissism and vulnerable narcissism were positively associated with problem gambling.  The inability of some participants to regulate unpleasant emotions, as measured the Difficulties in Emotion Regulation Scale, mediated the association between grandiose narcissism and    problem gambling.

The Dark Triad

Some of these drawbacks of narcissism could instead be ascribed to overlapping traits, such as psychopathy.  Therefore, rather than investigate narcissism alone, some research has explored how the dark triad—narcissism, psychopathy, and Machiavellianism—are related to problem gambling (e.g., Refaie et al., 2024).   For example

• as Trombly and Zeigler-Hill (2017) showed in a study over of 500 undergraduate studies, narcissism, psychopathy, and Machiavellianism all coincide with problem gambling—but only psychopathy remained significant after controlling the other two traits,
• in a study in which participants could choose to gamble with money that another person had received during a previous game (Jones, 2013), people who scored high on narcissism were especially likely to lose bets, whereas people who scored high on psychopathy were especially inclined to gamble—even knowing they are likely to lose,
• indeed, as Onyedire et al. (2021) revealed, the unique features of narcissism—that is, the level of narcissism after controlling psychopathy and Machiavellianism—may be inversely associated with problem gambling in older individuals.

As these findings imply, further research is warranted to ascertain which features of narcissism and psychopathy are more likely to promote gambling problems and the circumstances in which these problems are most likely to unfold.xpress their anger, such as drive aggressively when frustrated.  Second, impulsivity, and specifically the inability of some individuals to sustain attention on their task, mediated this association.  Finally, grandiose narcissism amplified these relationships.       

Introduction to loneliness

Many people experience profound levels of loneliness, defined as feelings of distress that emanate from unfulfilled social needs.  Indeed, loneliness can be regarded as the social equivalent of hunger or thirst—a powerful and aversive state that motivates people to establish and to maintain social relationships.  This state thus evolved to facilitate cooperation and protection, facilitating the survival of our genes (Cacioppo et al., 2006).  Although often adaptive, loneliness may promote a range of undesirable consequences.  For example

• loneliness during childhood increases the risk of cardiovascular health problems in younger adults (Caspi et al., 2006);
• loneliness also tends to coincide with cardiovascular health risks (Hawley et al., 2006) or predict these risks (Thurston & Kubzansky, 2009) later in life, even after controlling levels of depression,
• loneliness tends to predict cognitive decline over time (Tilvis et al., 2004), including Alzheimer’s Disease, again even after controlling depression (Wilson et al., 2007).

Researchers have proposed a range of accounts to explain how loneliness might impair health and cognition.  For example

• when people are lonely, they cannot depend on social circles to guarantee their safety—and hence loneliness tends to elicit undue vigilance and suspicion, sometimes impeding potential friendships and thus exacerbating this loneliness (Cacioppo et al., 2006),
• because lonely individuals may be unduly vigilant, they do not as readily orient attention to their personal goals—often diminishing their tendency to resist temptations and to choose healthy behaviours, such as exercise (Hawkley et al., 2009),
• this vigilance can also disrupt their sleep (Cacioppo et al., 2002), exacerbating the likelihood of cardiovascular, inflammatory, and metabolic diseases (Mullington et al., 2009),
• loneliness tends to coincide with other physiological changes, such as elevated blood pressure, increases in specific hormones, including cortisol, and modified gene transcription (for a review, see Hawkley & Cacioppo, 2010).

Loneliness and pathological narcissism

Narcissism can significantly affect relationships and, therefore, may influence the likelihood of loneliness. To explore this possibility, Kealy et al. (2022) conducted a study of 120 young adults, aged between 18 and 25.  These participants completed an online survey that included

• the Super Brief Pathological Narcissism Inventory (Schoenleber et al., 2015) in which six items measure grandiose narcissism, such as “I often fantasize about performing heroic deeds”, and six items measure vulnerable narcissism, such as “When people don’t notice me, I start to feel bad about myself”,
• the UCLA Loneliness Scale (Hughes et al. 2004), comprising three items, such as “How often do you feel isolated from others?”,
• the Temporal Satisfaction with Life Scale (Pavot et al., 1998), comprising five items, such as “My current life is ideal for me”,
• as well as a tool that gauges the five main personality traits: extraversion, neuroticism, agreeableness, conscientiousness, and openness to experience

As the analyses revealed, both grandiose narcissism and vulnerable narcissism were positively associated with loneliness, and loneliness was inversely related to satisfaction with life.  However, after controlling personality, only vulnerable narcissism, but not grandiose narcissism, was positively associated with loneliness. 

According to Kealy et al. (2022), if individuals exhibit vulnerable narcissism, they are not as likely to establish friendships.  For example, these individuals might often seem unduly sensitive to rejection or criticism, denigrate other people, and seem entitled.  These behaviours may repel other individuals.  As their friendships dwindle, they may acknowledge their social needs feel unfulfilled, manifesting as loneliness.  

Alternatively, as Rogoza et al. (2022) proposed, to prevent the distress that rejection can elicit, individuals who experience vulnerable narcissism may deliberately isolate themselves from other people. They may, for example, concoct reasons to justify this behaviour, such as convince themselves they do not need to depend on anyone else.  

Loneliness and specific facets of narcissism

Other studies have attempted to clarify how narcissism might shape loneliness.  To achieve this goal, MacDonald and Schermer (2023) explored which specific facets of narcissism are related to loneliness.  In this study, almost 700 university students completed

• the UCLA Loneliness Scale (Hughes et al. 2004), comprising three items, such as the frequency with which the participants “feel isolated from others”,
• the Five Factor Narcissism Inventory—a measure that generates 15 facets of narcissism, such as reactive anger, shame, need for admiration, grandiose fantasies, and entitlement (for sample items, see this webpage).

The researchers discovered that loneliness was positively associated with several facets of narcissism, including

• reactive anger, such as rage in response to comments they perceive as offensive,
• shame, such as feelings of humiliation in response to failures, and
• need for admiration.

These three facets often coincide with neuroticism—or a susceptibility to strong negative emotions.  Arguably, because of their narcissism, these individuals value status.  However, because of their neuroticism, these individuals attempt to shun circumstances that may challenge this status.  Therefore, they will tend to isolate themselves, often culminating in loneliness. 

Nevertheless, loneliness was inversely associated with some other facets of narcissism, including

• authoritativeness, such as the inclination to assume leadership roles,
• grandiose fantasies, such as daydreams about fame,
• indifference or the degree to which people are unconcerned about the opinions that other people express.

Two of these facets, authoritativeness and grandiose fantasies, tend to coincide with extraversion.  This extraversion may facilitate social relationships and thus diminish loneliness.  Alternatively, all three facets may be indicative of a tendency in some individuals to distort their perception of themselves. Individuals who report these traits may thus inflate their relationships, diminishing the likelihood they will acknowledge their loneliness.

Conclusions around the association between narcissism and loneliness

In short, as previous research implies, some facets or variants of narcissism coincide with loneliness, whereas other facets or variants of narcissism may be unrelated to loneliness or may even diminish loneliness.  Specifically, individuals who report vulnerable narcissism crave status but are especially susceptible to unpleasant emotions.  Therefore, either because they often display irritation or because they want to protect themselves from rejection or criticism, people who exhibit vulnerable narcissism may not develop satisfying relationships, culminating in loneliness.

In contrast, individuals who exhibit grandiose narcissism often like to inflate their capabilities, contributions, and importance.  To achieve this goal, some of these individuals often initiate behaviours that are intended to attract respect and admiration, such as boast about their achievements and qualities, called narcissistic admiration.  This pursuit of admiration motivates these individuals to establish social relationships.  Indeed, their confidence may even attract some friends.  Although they may irritate people, these individuals feel as if they have established relationships, diminishing loneliness.

Rogoza et al. (2018) uncovered findings that explicitly corroborate this argument. In their research, over 300 Polish adults completed an online survey that included

• the Hypersensitive Narcissism Scale to measure vulnerable narcissism, epitomised by items like “epitomised by items like “I show others how special I am”,
• the Narcissistic Admiration and Rivalry Questionnaire to gauge narcissistic admiration, epitomised by items like “I show others how special I am”, and
• the Los Angeles Loneliness Scale (Russell et al., 1978), exemplified by items like “There is no one I can turn to”.

As hypothesised, vulnerable narcissism was positively associated with loneliness, whereas narcissistic admiration was negatively associated with loneliness.    

Commencing in the 1970s, but proliferating over the next two decades, were a series of studies that explored the association between narcissism and self-esteem.  The upshot of these studies was that people who exhibit narcissism may experience a fragile or unstable self-esteem.  However, the relationship between narcissism and self-esteem is nuanced and multifaceted.

A pioneering study

Zeigler-Hill (2006) conduced one pioneering study that illustrates this perspective. In this study, 129 undergraduate students completed a series of instruments. First, participants completed a range of measures and tasks that were designed to assess self-esteem.  For example, one of these measures, the Rosenberg Self-Esteem Scale (Rosenberg, 1965), gauges self-esteem explicitly.  That is, this scale includes questions that obviously revolve around self-esteem, such as  

• I am able to do things as well as most other people,
• On the whole, I am satisfied with myself,
• All in all, I am inclined to feel that I am a failure (reverse-scored).

The other tasks assess self-esteem implicitly or obliquely.  That is, these tasks are designed to measure the degree to which people associate themselves with positive attitudes or qualities—but conceal this intention somehow.  Specifically,

• participants completed the Implicit Association Test (Greenwald & Farnham, 2000)—a task that purportedly assesses the degree to which people associate words that represent themselves, such as I or me, with words that represent pleasant qualities,
• participants indicated the extent to which they like each letter—because people who like their initials are assumed to associate themselves with positive qualities (Nuttin, 1985), called the name-letter effect,
• completed the Implicit Self-Evaluation Survey (Pelham & Hetts, 1999)—a task that assesses the degree to which people consider pleasant or unpleasant words after they reflect upon themselves.

To clarify the Implicit Self-Evaluation Survey, participants first indicated the degree to which a statement corresponds to their behaviour or character, such as “I am very sensitive to my inner thoughts and feelings”.  This procedure is designed to prime thoughts about themselves.  Next, participants completed word fragments, such as -ICE or -AD.  They needed to complete each fragment three times in a row, such as -ICE, -ICE, -ICE.  If participants tend to complete the first word fragment with a positive word, such as NICE instead of LICE, researchers assume these participants associate themselves with favourable qualities.  That is, their implicit self-esteem is high.

After finishing the tasks that were designed to assess self-esteem, participants then

• completed the Narcissistic Personality Inventory to assess grandiose narcissism
• repeated a modified version of the Rosenberg Self-Esteem Scale twice a day over two weeks—to gauge the stability of self-esteem over time (Kernis et al., 1989; Kernis et al., 1992; Kernis et al., 1993).

As the findings revealed, narcissistic people tended to report a high explicit self-esteem but a low implicit self-esteem. That is, a low implicit self-esteem was positively associated with narcissism but only if explicit self-esteem was high.  According to the authors, these results imply that perhaps

• narcissistic people, perhaps unconsciously, may associate themselves with unfavourable qualities: In essence, on some level, they perceive themselves as unworthy, manifesting as low implicit self-esteem,
• to compensate, narcissistic people like to inflate their capabilities, achievements, and significance and thus strive to perceive themselves as worthy, manifesting as a high implicit self-esteem,
• yet, this attempt to perceive themselves as worthy is susceptible to reminders of their insignificance, often culminating in a fragile self-esteem in which these individuals become enraged when criticised. 

The mask hypothesis

This distinction between explicit self-esteem and implicit self-esteem—and the associations between these attributes and narcissism—can be traced to discussions that first surfaced during the 1970s (Kernberg, 1975, Kohut, 1977). During this time, scholars often assumed that grandiose narcissists tend to perceive themselves unfavourably on some fundamental level but depict themselves as confident and proud to mask this sense of inferiority. This premise, called the mask hypothesis, resonates with the subsequent assumption that narcissism should coincide with a low implicit self-esteem and a high explicit self-esteem.  Yet, the studies that have investigated this hypothesis generated mixed results.  To illustrate

• in addition to Zeigler-Hill (2006), Jordan et al. (2003) also revealed that narcissism coincides with a high explicit self-esteem, as gauged by the Rosenberg Self-Esteem Scale, and a low implicit self-esteem, as gauged by an Implicit Association Test,
• however, other studies have challenged this model; indeed, a meta-analysis, published by Bosson et al. (2008), that included unpublished studies, revealed no consistent pattern between explicit self-esteem, implicit self-esteem, and narcissism.

The reliability and validity of implicit measures of self-esteem

To explain these mixed results, Bosson et al. (2008) proposed several explanations.  First, these authors questioned whether the measures of implicit self-esteem are reliable and valid.  Consistent with these concerns, these researchers cited evidence to indicate that distinct measures of implicit self-esteem, such as the implicit association test and the name letter effect, are not highly correlated.

Challenges to the assumption that self-esteem is unidimensional

Bosson et al. (2008) proposed a second account to explain the mixed findings around the relationship between narcissism and self-esteem.  Specifically, implicit self-esteem may not be unidimensional but may instead comprise at least two, and perhaps more, distinct facets.  For example, as Sakellaropoulo and Baldwin (2006) suggested 

• narcissistic people may implicitly associate themselves with a favourable appearance,
• but narcissistic people may implicitly associate themselves with unfavourable social relationships.

That is, narcissistic individuals may, perhaps unconsciously, perceive themselves as attractive but not liked.  As evidence of this possibility, in one study, 40 participants first completed a visualisation task.  Specifically, these participants visualised a time in which they had impressed someone—to prime a sense of narcissism—or a time in which they felt accepted—to prime a sense of acceptance.  Next, these individuals

• indicated the extent to which they perceived each letter of the alphabet, including their initials, as attractive,
• indicated the extent to which they perceived each letter of the alphabet, including their initials, as likeable,
• completed the Narcissistic Personality Inventory to assess grandiose narcissism.

In the condition in which participants visualised a time that primes a sense of narcissism, grandiose narcissism was highest when these individuals perceived their initials as attractive but not likeable.  This finding is consistent with the premise that narcissistic people may associate themselves with an attractive physical appearance but not a likeable character.  

These results are telling, because they challenge the notion that narcissism coincides with a high explicit self-esteem but a low implicit self-esteem. Instead, and in contrast to the masked hypothesis, narcissistic people may perceive themselves favourably on a sense of agency, such as dominance or capability, but not communal attributes, such as inclusion or compassion, regardless of whether the measure is explicit or implicit.

Variations across the facets of narcissism

Finally, the masked hypothesis might apply to some facets of narcissism rather than other facets of narcissism.  To illustrate, the masked hypothesis might apply to vulnerable narcissists but not grandiose narcissists.  Nevertheless, contrary to this premise, Brown and Brunell (2017) showed that vulnerable narcissism, as measured by the Hypersensitive Narcissism Scale, was unrelated to implicit self-esteem, as measured by the implicit association test.

Variants of a fragile self-esteem

Rather than a high explicit self-esteem and a low implicit self-esteem, Bosson et al. (2008) suggested that narcissists might demonstrate other manifestations of a fragile self-esteem. Specifically, Kernis (2003) delineated four measures or protocols that researchers can utilise to identify a fragile self-esteem. 

First, Crocker (2002) distinguished a contingent self-esteem from a true self esteem. That is, the self-esteem of some individuals is largely contingent upon whether they fulfill a set of arbitrary standards: academic prowess, social approval, physical attraction, and power, for example (Crocker & Wolfe, 2001).   The problem is that

• individuals who feel unduly motivated to fulfill these standards will often pursue activities that do not resonate with the core, enduring values, compromising their engagement and wellbeing (Deci & Ryan, 1985; Deci & Ryan, 1995),
• if they do not fulfill these standards, these individuals will perceive themselves as unworthy and, thus, may be especially sensitive to criticism or failure,
• in contrast, if individuals pursue activities that are compatible with their core, enduring values, they experience a high self-esteem regardless of whether they fulfill these standards, called a true self-esteem.

Second, either because of these contingencies, or because of other causes, the self-esteem of some individuals varies dramatically across time.  In contrast, the self-esteem of other individuals is not as erratic and more stable instead. If self-esteem is unstable, individuals are more sensitive to feedback or similar challenges (Kernis, Paradise, et al. 2000), manifesting as a defensive manner (Kernis, Greenier, et al., 1997). 

Third, to boost their self-esteem, some individuals may bias their attention, memory, and appraisals towards their achievements or strengths, called self-enhancement biases.  Accordingly, their perception of themselves might diverge from their reality.  Because of this divergence, these individuals may often receive feedback, such as criticism, that conflicts with their perception themselves—often culminating in rage, shame, or other defensive responses.  

Interestingly, in nations in which inequality in income is especially pronounced, as measured by the Gini index, these biases are magnified.  That is, in unequal nations, people often feel more compelled to compete with each other and scale the corporate hierarchy, often prompting these self-enhancement biases (Loughnan et al., 2011). 

Finally, some individuals report a high explicit self-esteem but demonstrate a low implicit self-esteem (e.g., Kernis, Abend, et al., 2005).  These individuals tend to be defensive in response to adverse feedback (e.g., Bosson et al., 2003).  Consequently, this disparity between an explicit self-esteem and implicit self-esteem may reveal a fragile self-esteem.

Variants of a fragile self-esteem and narcissism

Conceivably, narcissism might coincide with any of these measures or definitions of fragile self-esteem, such as contingent self-esteem, unstable self-esteem, or self-enhancement.  And some research has shown that at least some facets of narcissism do indeed coincide with a fragile self-esteem.

In one illuminating study, Rogoza et al. (2018) administered an online survey to 314 participants from Poland, aged between 16 and 35.  To gauge narcissism, participants completed the

• Narcissistic Admiration and Rivalry Questionnaire—a measure that differentiated two narcissistic strategies: narcissistic admiration in which individuals strive to inflate their achievements or contributions and narcissistic rivalry in which individuals strive to denigrate other people,
• Hypersensitive Narcissism Scale to assess vulnerable narcissism—the tendency of some people not only to dismiss criticism but to exhibit contempt or rage towards anyone who delivers this criticism as well as to depict themselves as victims of injustice.

To measure self-esteem and fragility of self-esteem, participants completed the Rosenberg Self-Esteem Scale and the Contingent Self-Esteem Scale.  A typical item in the Contingent Self-Esteem Scale is “My overall feelings about myself are heavily influenced by how much other people like and accept me”.  The self-esteem of participants was labelled as fragile if they reported a  

• a low explicit self-esteem and
• a high contingent self-esteem, in which their self-esteem is contingent upon the fulfillment of various standards.

If the participants reported vulnerable narcissism or narcissistic rivalry, they were more likely to demonstrate a fragile self-esteem. In contrast, if the participants reported narcissistic admiration, they were less likely to demonstrate a fragile self-esteem.  In short, some but not all facets of narcissism seem to coincide with a fragile self-esteem. 

Benefits of self-compassion

Individuals who develop narcissism experience a range of challenges.  Some of these challenges can be ascribed to a decrease in self-compassion—the tendency of individuals to perceive their failings as universal in all humans, to accept their flaws, and to consider their limitations calmly (Neff, 2003a; Neff. 2003b).  As self-compassion diminishes, these individuals experience a range of problems, from health complications to impaired relationships.

Definition of self-compassion

Self-compassion, although emanating from many Eastern philosophies such as Buddhism, materialised in Western science only recently. Neff (2003a, 2003b) promulgated this state, defining self-compassion as a blend of three distinct, but interrelated facets. First, individuals recognise that pain and failure are inevitable, universal features of the human experience.  Second, because of this recognition, these individuals are more inclined to accept and understand all facets of themselves, including their deficiencies, faults, or afflictions.  Third, because of this acceptance, they demonstrate a balanced perspective of themselves and their emotions—neither disregarding nor inflating unpleasant thoughts or feelings, similar to mindfulness.

To measure self-compassion, Neff (2003a) established an instrument that comprises six subscales—a position variant and a negative variant of the three distinct facets.  The subscales include

• common humanity (e.g., “When I feel inadequate in some way, I try to remind myself that feelings of inadequacy are shared by most people.”),
• isolation (e.g., “When I fail at something that’s important to me, I tend to feel alone in my failure”),
• self-kindness (e.g., “When I’m going through a very hard time, I give myself the caring and tenderness I need”),
• self-judgment (e. g., “When I see aspects of myself that I don’t like, I get down on myself”,
• mindfulness (e.g., “When something upsets me, I try to keep my emotions in balance.”), and
• over-identification (e.g., “When I’m feeling down, I tend to obsess and fixate on everything that’s wrong”).

Association between self-compassion and narcissism

Many studies have revealed that narcissism, or at least some variants of narcissism, are inversely associated with self-compassion.  For example

• in a study of over 250 adolescents, individuals who reported vulnerable narcissism, but not grandiose narcissism, tended to report diminished levels of self-compassion (Barry et al., 2015),
• in a study of over 800 college students, the participants who reported elevated levels of narcissism tended to report less self-compassion but greater burnout as a consequence (Barnett & Flores, 2016).

Furthermore, self-compassion can temper some of the detrimental effects of narcissism (e.g., Gu & Hyun, 2021).  To illustrate, Zhang et al. (2024) explored whether self-compassion diminishes the impact of narcissism on antisocial behaviour in sport.  In one study, over 200 football players completed a survey that assessed narcissism, antisocial behaviour—such as “I swore at a teammate”—and self-compassion.  Generally, narcissism was positively associated with antisocial behaviour.  However, when participants indicated they experience self-compassion to a moderate or significant degree, this association between narcissism and antisocial behaviour dissipated.

Presumably, when individuals experience narcissism, they often experience unpleasant feelings, such as shame, in response to criticisms and other challenges.  To override these unpleasant feelings, they may become more inclined to blame or to belittle other individuals, frequently manifesting as antisocial behaviour.  However, if they experience self-compassion, these unpleasant feelings are not as pronounced.  Their shame, for example, may diminish once they recognise these challenges are inevitable, universal, and acceptable.   Accordingly, as several researchers have recommended, activities that foster self-compassion may moderate some of the adverse consequences of narcissism (Gu & Hyun, 2021).

Implications of these decreases in self-compassion

Some of the challenges and problems that often coincide with narcissism could perhaps be ascribed to self-compassion.  That is, various facets of narcissism, especially vulnerable narcissism, seem to be inversely associated with self-compassion (e.g., Barry et al., 2015).  Consequently, these individuals may not enjoy the benefits of self-compassion. 

Many studies have substantiated the benefits of self-compassion.  Here are some examples:

• Self-compassion is positively associated with optimism, life satisfaction, and a feeling of social connection (Neff, 2003a; Neff, Rude, & Kirkpatrick, 2007), all of which can promote wellbeing.
• Individuals who experience self-compassion can also extend this understanding of their own challenges to appreciate the problems that other people experience, manifesting as empathy.  To demonstrate, in one study (Welp & Brown, 2013), after completing a measure of self-compassion, participants read about a man who was either concentrating intently but accidentally hit a pothole or was distracted while speaking on the mobile and hit a pothole.  Self-compassion was positively associated with empathy and willingness to help, but only if the person had been concentrating intently and, therefore, was not at fault. 
• Consistent with this finding, as Neff and Beretvas (2013) showed, if individuals report elevated levels of self-compassion, they are more likely, according to their partners, to be caring, affectionate, and gentle as well as discuss problems, respect other opinions, and encourage autonomy. 
• People who experience self-compassion are not as likely to feel ashamed of their frailties or distress in response to health concerns.  Consequently, they respond more proactively to health problems.  For example, as Terry et al. (2013) showed, people who experience self-compassion are more inclined to observe the recommendations of their doctor and cope more effectively with diagnosed illnesses.
• Self-compassion does not, as perhaps assumed, foster complacency.  Instead, people who experience self-compassion are often more inspired to improve themselves.  For example, in a series of studies, self-compassion encouraged participants to practice more extensively on task (Breines & Chen, 2012).  In one study, participants completed a difficult vocabulary task.  They were granted time to practice before completing the test again. Next, to induce self-compassion, some participants imagined they were talking to themselves about this failure from a compassionate and understanding perspective.  Relative to various control conditions, self-compassion inspired participants to dedicate more time to practice. 
• Finally, self-compassion also improves motivation.  For example, if people experience self-compassion, they feel more inspired to learn and to initiate more adaptive coping strategies after academic failure (Neff, Hseih, & Dejitthirat, 2005).

In short, people who are narcissistic experience diminished self-compassion, and this diminished self-compassion may compromise wellbeing, empathy, relationships, and motivation to learn or change.  That is, if individuals experience self-compassion, they feel consoled—as if receiving comfort from a supportive relative or friend—after failures or problems unfold.  The unpleasant emotions these challenges elicit abate rapidly.  Individuals can thus manage their problems constructively (for a detailed account and evidence, see Inwood & Ferrari, 2018).    

To thrive at work and in life, people need to develop their knowledge and skills.  To achieve this goal, individuals must recognise and then address their limitations.  This capacity to recognise their limitations, called monitoring, and then to formulate plans to redress these shortfalls, called control, depends on a skill called metacognition.  Metacognition refers to the extent to which people are aware of their thoughts and understand how to utilise and regulate these thoughts.   Although John Flavell first used this term in 1976, discussions about meta-cognition can even be traced to Aristotle.  Research suggests that meta-cognition is impaired in narcissistic individuals—and, indeed, many of the adverse consequences of narcissism can be ascribed to this deficit.

To illustrate the notion of metacognition, people tend to demonstrate metacognition when they

• recognise they do not concentrate effectively when learning biology rather than chemistry
• feel they need to read more about a topic, because their understanding now is limited,
• decide how they will learn and memorise content more effectively,
• evaluate the past strategies they have utilised to learn and memorise information, and so forth.

This metacognitive ability does vary across individuals.  According to some researchers, people may demonstrate metacognitive ability on some tasks but not on other tasks (e.g., Kelemen et al., 2000; Moore & Healy, 2008).  For example, people may accurately evaluate their performance and progress in sports but not at work.  Other researchers assume that, to a large extent, the cognitive operations or apparatus that underpins meta-cognitive ability can be applied in all settings (e.g., Carpenter et al., 2019).

The association between grandiose narcissism and metacognition

Individuals who exhibit grandiose narcissism strive to inflate their capabilities and significance.  To achieve this goal, they may not be as likely to activate cognitive operations that monitor their performance accurately.  Consequently, grandiose narcissism may be inversely associated with metacognition.

Consistent with this possibility, studies have revealed that grandiose narcissism seems to compromise metacognition.  For example, Meisel et al. (2016) explored the association between grandiose narcissism and overconfidence in American and Chinese adults.  In the first study, participants from both nations completed the Narcissistic Personality Inventory, a measure of grandiose narcissism, comprising items like “I think I am a special person”.  In addition, these participants completed the Georgia Gambling Task (Goodie, 2003). To complete this task

• for each item, participants need to specify in which of two regions, such as Kentucky and Illinois, is the population larger,
• indicate their confidence in a percentage, ranging from 50 to 100,
• midway during this task, on each item, participants could accept a bet—in which they gain 100 points if they are correct and lose between 101 and 9000 points if incorrect, depending on their level of confidence,
• if they do not accept the bet, their neither gain nor lose points—a strategy that is less risky.

In both nations, participants who reported elevated levels of narcissism were more likely to exhibit overconfidence.  That is, their average confidence on each item exceeded the percentage of correct answers.  However, narcissism was not significantly related to the tendency of individuals to accept bets and thus not affect risk.

The difference between grandiose narcissism and vulnerable narcissism on metacognition

Most studies that have examined whether metacognition is impaired in narcissistic people have explored only grandiose narcissism.  In contrast, Littrell et al. (2024), in a study of 208 adults from North America, assessed whether vulnerable narcissism as well as grandiose narcissism is related to metacognition.  To measure narcissism, participants completed the Five Factor Narcissism Inventory (Sherman et al., 2015)—an instrument that generates measures of agentic narcissism, antagonistic narcissism, and neurotic narcissism.  Grandiose narcissism comprises a blend of agentic narcissism and antagonistic narcissism, whereas vulnerable narcissism comprises a blend of neurotic narcissism and antagonistic narcissism.

Next, participants completed three cognitive tasks: a memory task, in which they needed to memorise 30 pairs of words, a numerical reasoning task, and a vocabulary task.  Before the memory task, participants also indicated the likelihood they will memorise the items correctly, on a scale from 0 to 100%.  In contrast, after each item on the other tasks, participants indicated the likelihood they answered the question correctly.   The researchers computed two indices:

• a measure of overconfidence—the degree to which the average confidence on the items exceeded the percentage of correct responses,
• a discrimination index—representing whether they were more confident on the items they answered correctly. 

Overall, grandiose narcissism was positively associated with overconfidence and inversely associated with the discrimination index or tendency to be more confidence on the items they answered correctly.  This relationship persisted after controlling antagonistic narcissism, implying this impaired metacognition can be ascribed to agentic narcissism.

In contrast, vulnerable narcissism was not significantly association with overconfidence or the discrimination index.  However, after controlling antagonistic narcissism, neurotic narcissism was inversely associated with overconfidence. 

Cognitive reflection—another feature of metacognition—is also more dependent on grandiose narcissism than vulnerable narcissism.  To illustrate, in a study that Littrell et al. (2020) conducted at the University of Waterloo in Canada, 100 adults, recruited from Amazon Mechanical Turk, completed a survey that included the following instruments, presented in a random order:

• the version of the Narcissism Personality Inventory that comprises 13 items, such as “I find it easy to manipulate people” and measures grandiose narcissism (Gentile et al., 2013),
• the Hypersensitive Narcissism Scale to measure vulnerable narcissism, typified by items like “My feelings are easily hurt by ridicule or the slighting remarks of others”,
• the cognitive reflection test—a test that measures the tendency of individuals to contemplate and revise their answers, comprising items like “A bat and a ball cost $1.10 in total. The bat costs $1.00 more than the ball. What is the cost of this ball?”,
• the Self-Reflection and Insight Scale (Grant et al., 2002)

The Self-Reflection and Insight Scale measures two facets of metacognition.  First, this instrument gauges metacognitive reflection—or the degree to which individuals believe they contemplate and evaluate their thoughts, feelings and behaviours.  A sample item is “I frequently take time to reflect on my thoughts”.  Second, this instrument gauges metacognitive insight—or the extent to which individuals feel they understand their thoughts, feelings, and behaviours.  A typical item is “Thinking about my thoughts makes me more confused” (reverse scored).

As hypothesised, responses to the measures of metacognition differed between grandiose narcissism and vulnerable narcissism.  That is, grandiose narcissism was negatively associated with performance on the cognitive reflection test but not significantly associated with metacognitive reflection or metacognitive insight.  That is, if people exhibit grandiose narcissism, they do not contemplate or correct their answers frequently but do not seem to be cognisant of this shortfall.  In contrast, vulnerable narcissism was not significantly associated with performance on the cognitive reflection test.  However, a trend was identified in which vulnerable narcissism was positively related to metacognitive reflection (p = 0.06)

One complication that may bias or skew the research on narcissism revolves around response styles.  That is, people who exhibit narcissism like to inflate their capabilities, achievements, power, and importance—although they may acknowledge some attributes that may be perceived as undesirable in social settings (Jones & Brunell, 2014), such as a disinclination to help another person.  Accordingly, narcissistic individuals may not answer questions on surveys accurately.  They may exaggerate their intelligence or status, for example, as well as conceal their failures.   

Nevertheless, some research has challenged this assumption.  Sleep et al. (2017), for example, revealed that narcissistic individuals do not necessarily conceal their faults or challenges in inconsequential surveys.  Indeed, if people exhibit vulnerable narcissism, they seem more inclined than other participants to acknowledge these faults or challenges. 

Specifically, to investigate this matter, the researchers examined three samples of participants—comprising 482 undergraduate students, 703 male prison inmates, and 228 psychology students, respectively.  The samples completed various measures of narcissism, including

• the Narcissistic Personality Inventory,
• the Five-Factor Narcissism Inventory,
• 17 relevant items from the SCID-II for Personality Disorders (First et al., 1997),
• the Minnesota Multiphasic Personality Inventory or MMPI–2 (Ben-Porath & Tellegen, 2011), and
• the Personality Assessment Inventory (Morey, 1991)

To assess the response style of participants—that is, the degree to which these individuals may have inflated their attributes or concealed their flaws—the researchers utilised

• the Uncommon Virtuous subscale of the MMPI-2—a metric that gauges the degree to which individuals tend to conceal rather than acknowledge common but minor personal flaws,
• the Adjustment Validity subscale of the MMPI-2—a metric that gauges the extent to which individuals tend to deny the experience of common psychological problems (for evidence of validity, see Sellbom & Bagby, 2008),
• the Positive Impression Management index of the Personality Assessment Inventory—defined as the tendency to deny minor personal faults to generate a positive impression,
• the Defensiveness index of the Personality Assessment Inventory—an index that represents inconsistencies in responses.

In general, the various measures of narcissism were not strongly associated with the various measures of response style.  More specifically

• one measure of vulnerable narcissism, derived from the Five-Factor Narcissism Inventory, was negatively associated with the Positive Impression Management index and Defensiveness index—implying that people who exhibit vulnerable narcissism are especially inclined to acknowledge personal faults or challenges,
• measures of grandiose narcissism were not significantly associated with these indices,
• in the prison inmates, leadership and authority was positively associated with the inclinations to conceal minor personal flaws or psychological problems, but entitlement and the willingness to exploit people was negatively associated with these inclinations. 

One possible complication or explanation is that people who are unwilling to concede their flaws may also be unwilling to report the signs of narcissism.  Therefore, the measures of narcissism may not be accurate enough to detect response biases.