Psychol Stud (October–December 2011) 56(4):368–372 DOI 10.1007/s12646-011-0095-9
RESEARCH IN PROGRESS
Grandiose Fantasies and Low Self-Esteem: An Experimental Study Giovanni Maria Ruggiero & Guido Veronese & Marco Castiglioni & Sandra Sassaroli
Received: 28 January 2011 / Accepted: 26 April 2011 / Published online: 6 May 2011 # National Academy of Psychology (NAOP) India 2011
Abstract Grandiose states of mind could be an ineffective way to cope with low self-esteem. In fact, it is possible that grandiose fantasies would tend to induce doubts about personal worth and decrease self-esteem. This study explored whether grandiose fantasies help to cope with low self-esteem. The self-reported occurrence and unpleasantness of intrusive thoughts regarding the sense of exclusion, humiliation, and low self-esteem in non-clinical individuals were assessed during self-induced grandiose fantasies and self-induced relaxation. The occurrence and unpleasantness of low self-esteem intrusions were significantly higher during grandiose fantasies, suggesting that grandiose fantasy is ineffective at combating low self-esteem. Keywords Grandiose fantasy . Intrusive thought . Narcissism . Self-esteem
Introduction According to DSM-IV-R (American Psychiatric Association 2000), grandiose fantasies of power, success, and superiority G. M. Ruggiero “Psicoterapia cognitiva e ricerca” Cognitive psychotherapy school and research center, Milano, Italy G. Veronese (*) : M. Castiglioni Dipartimento di Scienze Umane per la Formazione “R. Massa”, Università Milano-Bicocca, P.zza dell’Ateneo Nuovo 1, Milan, Italy e-mail:
[email protected] S. Sassaroli “Studi Cognitivi” Cognitive psychotherapy school and research center, Milano, Italy
are one of the distinguishing features of individuals prone to narcissism and patients suffering from Narcissistic Personality Disorder (NPD) (John and Robins 1994; Kernberg 1975; Kohut 1971). Some scholars have proposed that these grandiose states may conceal a profound feeling of inadequacy, inferiority, and a tendency to feel criticized and humiliated (Emmons 1987; Horowitz 1989; Morrison 1989; Dimaggio et al. 2002). Thus, individuals prone to grandiosity would feel themselves as excluded, despised, and ostracized (Dimaggio et al. 2002; Twenge and Campbell 2003; Kohut 1971). A metaanalysis supports the idea that individuals prone to narcissism would use grandiose fantasies as defense against the feeling of low self-worth underlying their inflated self-image (Campbell and Sedikides 1999). Grandiosity could be a plausible reaction to the feelings of inferiority, fragile self-esteem, vulnerability to criticism and fear of exclusion (Akthar and Thomson 1982; Kernberg 1975; Kohut 1971). Thus, some individuals could cope with criticism by entering grandiose states and conceiving fantasies of power, success, and triumph (Gabriel et al. 1994; John and Robins 1994), would prefer to be admired rather than nurtured by others (Campbell 1999), and may adopt a defensive self-regulatory style to maintain an inflated sense of self-esteem, denying negative experiences and overemphasizing positive ones (Tracy and Robins 2003). From an empirical viewpoint, while some investigations confirmed that narcissistic grandiosity masks a low implicit self-esteem, other investigations failed to replicate this model. In fact, some studies reported an inverse association between narcissistic grandiosity and self-esteem (Rose 2002; Watson et al. 1997; Watson et al. 1996; Soyer et al. 2001), whereas others reported a positive association (Emmons 1984, 1987; Raskin and Terry 1988; Raskin et al. 1991).
Psychol Stud (October–December 2011) 56(4):368–372
There are at least two possible alternative explanations that might shed light on these inconsistent findings: the possibility that there are two subtypes of narcissism, an actually grandiose one and a vulnerable one; and the possibility that narcissists might show a solid, high selfesteem in the performances (e.g., study, work, etc.) and a fragile self-esteem in the affective/relational domain (Crocker et al. 2003). Another possible explanation is that narcissists may demonstrate a self-esteem that is both high and fragile, meaning that their self-esteem is good but easily threatened, requires constant validation and is maintained through self-deception. Thus, it is possible that narcissism correlates with high but unstable self-esteem. In turn, this instability is related to a high emotional reactivity to challenging events that have negative implications for the self (Rhodewalt et al. 1998; Zeigler-Hill 2006). To summarize, there is not a conclusive confirmation of the hypothesis of masked low self-esteem in narcissistic grandiosity. Therefore, the hypothesis needs further empirical research. Ineffective grandiosity may be related to a higher occurrence of intrusive thoughts of low self-worth during grandiose fantasies than in other psychological states. This hypothesis has yet not been systematically explored and little is known about low self-esteem intrusions emerging at consciousness during grandiose fantasies. Nonetheless, we can find something similar in studies focused on obsessive-compulsive disorder. Obsessive individuals involved in suppression of negative thoughts do not manage to eliminate these thoughts, and are actually overwhelmed by intrusions of tentativelysuppressed thoughts (Belloch et al. 2004; Salkovskis and Campbell 1994). This study compared the occurrence and the unpleasantness of intrusions regarding the sense of exclusion, humiliation, and low self-esteem during self-induced grandiose fantasies and self-induced relaxation. Since grandiose fantasies are an ineffective strategy to cope with low self-esteem, it was expected that significantly more intrusions and an increased degree of unpleasantness of intrusions would occur during the grandiose fantasy. It was also expected that narcissistic grandiosity and either the occurrence or the unpleasantness of low self-esteem intrusions would be positively associated.
Method Participants One hundred sixty-six healthy Italian volunteers participated in the study, accurately matched for gender (they
369
were 83 males and 83 females, mean age 27.24± 8.60 years) and age (females had a mean age 27.07± 9.41 years, males had a mean age 27.42±7.76 years). Ninety-two (55.4%) participants were current university students (47 males and 45 females, mean age 22.55± 2.29 years and 22.02±2.42 years, respectively) and 74 (44.6%) were employed full- or part-time during the 6 months prior to the study (36 males and 38 females, mean age 33.78±7.80 years and 33.05±11.02 years, respectively). Forty-three participants were university graduates. Twenty-six individuals were married and 91 were engaged to be married. The participants were randomly assigned to two subgroups that performed the same experiment with a minor procedural difference (see section on procedure). Randomized assignment was performed using EXCEL software. The first subgroup comprised 83 participants (43 males and 40 females; mean age 27.27±8.03 years; 50 university students and 33 employed individuals). The second subgroup comprised 83 participants (40 males and 43 females; mean age 27.23 ± 9.19 years; 42 university students and 41 employed individuals). The participants received no remuneration for their performance. Measures Rosenberg Self-Esteem Scale (RSES; Rosenberg 1965). It assesses global self-esteem and sense of self-worth. It is a 10-item Likert scale. Items are answered on a 4-point scale, ranging from 3 (strongly agree) to 0 (strongly disagree). Narcissistic Personality Inventory (NPI) (Raskin and Hall 1979, 1981). It measures global narcissism and its components. We used the 40-item version (Raskin and Terry 1988). Research demonstrates adequate reliability and validity of the NPI in adults (see Raskin and Terry 1988, pp. 890–893). Procedure The participants were asked to complete the RSES and the NPI before the experiment. At the beginning of the experiment, participants were asked to: 1) remember their worst life event concerning a threat to their selfesteem; 2) conceive a grandiose fantasy showing himself or herself as protagonist; and 3) conceive a relaxing scenario. Eighty-three individuals performed the first phase of the experiment, during which they were asked to: 1) mentally focus on their low self-esteem life event for 1 min; 2) switch to the grandiose fantasy at the command of the researcher, and focus on it for 1 min; and 3) during the
370
1 min grandiose fantasy, raise a hand each time that they remembered the low self-esteem life event (signal of low self-esteem intrusions). The researcher measured the required time frames, and counted the number of intrusions (number of times that the person raised their hand). After a week, these 83 individuals performed the second phase of the experiment, during which the procedure was repeated using a relaxing scenario instead of the grandiose fantasy. The remaining 83 individuals performed exactly the same experiment with the exception of a single detail: they were asked to use the relaxing scenario during the first phase and the grandiose fantasy during the second phase. This procedure was to control for the order of experimental conditions. After completing both phases of the experiment, participants had to report the amount of unpleasantness of the low self-esteem intrusions that appeared either during the grandiose fantasy or during the relaxing scenario, using a 5-point rating scale from “not at all” to “extremely high”.
Results Participants showed an average total score of 14.30 (SD= 2.48) on the NPI and 19.82 (SD=4.47) on the RSES. According to psychometric properties of the two instruments (Raskin and Hall 1979, 1981; Rosenberg 1965), no participant showed a result attributable to a narcissistic personality or a abnormal self-esteem. Reliability based on internal consistency was confirmed with a Cronbach’s alpha coefficient >0.7 for each instrument (0.85 for the NPI and 0.87 for the RSES). As predicted, the number of intrusions and their unpleasantness were significantly higher during the grandiose fantasy than during the relaxing scenario (Table 1). A significant inverse correlation between RSES scores and unpleasantness of intrusions during the grandiose fantasy (Table 2), and a significant direct correlation between NPI score and unpleasantness of intrusions during
Psychol Stud (October–December 2011) 56(4):368–372
the grandiose fantasy when controlling for RSES scores (Table 3) were also found.
Discussion The results suggests that grandiose fantasies possess the power to increase the number of occurrences and the unpleasantness of low self-esteem intrusive thoughts. If confirmed, this finding would provide evidence that grandiosity is an ineffective and even detrimental way to cope with a crisis of low self-esteem. Of course, this result can be used to understand possible aspects of grandiose states, but cannot be considered a confirmation of the model of pathological narcissism of Horowitz (1989), Akhtar and Thomson (1982), and Dimaggio et al. (2002). First of all, the experimental design does not really show the emergence of grandiosity from low self-esteem; it rather follows the opposite direction, because it shows how grandiosity could increase the number of intrusive low self-esteem thoughts. In addition, the use of non-clinical individuals hinders the generalizability of the result and its direct application to the psychopathology of narcissism. In addition, we found that self-esteem correlated positively with the unpleasantness of intrusions during the grandiose fantasy (Table 2) and that narcissism correlated negatively with this same variable when controlling for self-esteem (Table 3). The meaning of these result could be that a higher self-esteem seems to increase the vulnerability of people to sudden and painful intrusive thoughts of low self-esteem, or that a sort of ‘purer’ narcissistic inflated self-esteem (in which components of non narcissistic good self-esteem are statistically controlled using the partial correlation method) would be related to actual feelings of painful, low self-esteem intrusive doubting. However, this interpretation is questionable and needs rigorous confirmation in studies addressed to this specific aim. Thus, the question of low self-esteem in grandiose states remains controversial. According to Semerari
Table 1 Incidence and unpleasantness of low self-esteem intrusions during grandiose fantasy and relaxing scenarios: Descriptive statistics and t-test in the whole sample Variables
Mean
SD
t (165)
Incidence of intrusions during the grandiose fantasy Incidence of intrusions during the relaxing scenario Unpleasantness of intrusions during the grandiose fantasy Unpleasantness of intrusions during the relaxing scenario
2.92 2.23 2.84 2.61
2.10 2.05 .98 1.15
3.659***
***p
