An experimental investigation of the influence of agentic and communal Facebook use on grandiose narcissism

Highlights

• Explored in two experiments the impact of Facebook use on narcissism.

• Expected agentic Facebook use to lead to higher narcissism than communal or control.

• Experiment 1 revealed a marginal effect of Facebook use on narcissism, whereas Experiment 2 did not.

• Research synthesis revealed no reliable effect of Facebook use across experiments.

Abstract

Does activity on a social networking site (e.g., Facebook) increase grandiose narcissism? It was hypothesized that the type of activity (agentic vs. communal) might make a difference. In the first experiment, 88 males were randomly assigned to one of three computer activity conditions (agentic Facebook activity, communal Facebook activity, control computer activity). In the second experiment a larger and more diverse sample (N = 218) engaged in agentic, communal, or control activities either on a computer or not, in order to assess the impact of medium as well as thinking type. In both experiments, self-reported daily Facebook activity was correlated with narcissism. However, the experimental manipulation of agentic vs. communal Facebook use did not significantly increase state narcissism, as indicated by synthesizing the findings of the two experiments. The findings cast doubt on the notion that social networking site use promotes narcissism, though additional research on the topic is warranted.

Introduction

Grandiose narcissism refers to a normally-distributed personality trait that is characterized by inflated beliefs about the self, a sense of entitlement, and a lack of concern about and empathy for others (Pincus & Roche, 2011). Such narcissism is associated with relative psychological health, such as high levels of self-esteem and low levels of depression and anxiety (Sedikides, Rudich, Gregg, Kumashiro, & Rusbult, 2004), but with interpersonal dysfunction in the form of low levels of agreeableness and forgiveness of others, and high levels of “game-like” love styles and hostility, especially in the face of threat (Campbell et al., 2002, Miller et al., 2010). Grandiose narcissism has been effectively differentiated from “vulnerable narcissism” (see Pincus & Roche, 2011) and seems to be on the rise in American society. Indeed, Twenge and Campbell (2009) reported that college students’ levels of grandiose narcissism have increased nearly 30% in the past 30 years and that the lifetime prevalence rate of narcissistic personality disorder is three times higher in Americans in their 20s than it is in Americans over 65, though the former have had 40 or so fewer years in which to be diagnosed. Such a rise in narcissism has catalyzed both scientific and popular theorizing about the origins of the trait. In this paper we focus on environmental factors that are commonly believed to facilitate grandiose narcissism.

Considerations of the environmental origins of narcissism have, until recently, focused solely on parenting behavior. Clinical theorists (e.g., Freud, Kernberg, and Kohut) long suggested that pathological narcissism results from inadequate parenting – parenting that is cold and harsh, overly affectionate, or emotionally manipulative, depending upon the perspective one embraces – and empirical investigations of the origins of grandiose narcissism have supported some, but not all, of these clinical perspectives (see Cramer, 2011, Horton et al., 2006, Watson et al., 1992). For instance, Cramer (2011) found that childhood reports of parenting styles that were characterized by high levels of parental warmth (permissive and authoritative styles) were associated prospectively with high levels of grandiose narcissism as an adult. Indeed, parenting behavior seems to be linked to grandiose narcissism.

Recent theorizing and evidence has also linked the aforementioned rise in dispositional narcissism to an increased emphasis in American culture on the self, uniqueness, and individualistic values. There are several examples of this cultural shift: Song lyrics have become increasingly self-focused, baby names have become increasingly unique, and membership in communal civic and spiritual organizations has been on the decline (DeWall et al., 2011, Twenge et al., 2010, Twenge and Campbell, 2009). Further, as social networking sites have become increasingly popular, with Facebook recently surpassing one billion users, more and more people engage daily in on-line self-promotional activities, such as posting status updates, uploading pictures of themselves, and changing biographical information.

In fact, recent public conversations about the origins of grandiose narcissism have included a healthy dose of indictment of social media. In their engaging popular book, The Narcissism Epidemic: Living in the Age of Entitlement¸ Twenge and Campbell (2009) devote a chapter to social networking as one of the possible “Root Causes of the Epidemic.” In that chapter, the authors make the bold claim that “Web 2.0 and cultural narcissism work as a feedback loop, with narcissistic people seeking out ways to promote themselves on the Web and those same websites encouraging narcissism even among the more humble,” (p. 107). These authors are not alone in espousing the intuitive notion that social networking sites (SNS) like Facebook and Twitter promote narcissism. In a recent interview with Fox News and an accompanying Blog post, psychiatrist Keith Ablow (2013) blamed the rising tide of individual narcissism on a variety of different factors, including internet social networking sites. Quite specifically, he noted that “On Facebook, young people can fool themselves into thinking they have hundreds or thousands of ‘friends.’ They can delete unflattering comments. They can block anyone who disagrees with them or pokes holes in their inflated self-esteem. They can choose to show the world only flattering, sexy or funny photographs of themselves (dozens of albums full, by the way), ‘speak’ in pithy short posts and publicly connect to movie stars and professional athletes and musicians they ‘like.’” According to Ablow and others, such SNS opportunities allow for the “bubble of narcissism” to rise unimpeded (Ablow, 2013).

Interestingly, this sentiment has garnered some empirical support. For example, levels of grandiose narcissism (or components thereof) are higher among Facebook users than among non-Facebook users (Ryan & Xenos, 2011) and are correlated positively with the amount of time spent on Facebook (Buffardi & Campbell, 2008). Such narcissism is also related to a higher frequency of and preference for posting status updates (Ryan and Xenos, 2011, Wang et al., 2012); a greater tendency to post, or preference for posting, attractive pictures of the self (Buffardi and Campbell, 2008, Ryan and Xenos, 2011, Wang et al., 2012); and a greater number of Facebook friends (Buffardi and Campbell, 2008, Gentile et al., 2012, Mehdizadeh, 2010). Overall, notwithstanding some inconsistent results (e.g., Skues, Williams, & Wise, 2012), it seems that there is adequate support for the notion that narcissism is associated with certain elements of social networking site use. Of course, such associations can be interpreted in a variety of ways, including the opposite causal chain: individuals high in narcissism gravitate to social networking sites and to particular activities on those sites more than do individuals lower in narcissism. In short, such correlational investigations do not provide convincing support for the notion that social networking use facilitates narcissism.

The possible influence of social networking use on narcissism seems to be predicated on either or both of two possibilities. First, individuals might represent themselves on social network sites in an idealized form and, through a self-perception/attributional process, come to believe, at least to a greater extent than they did before such self-presentation, that this idealized, “superior self” is real. Second, social networks provide substantial opportunities to receive, or at least imagine, positive interpersonal attention (e.g., positive comments and “likes” for one’s posts and pictures), which may feed a narcissistic belief in one’s uniqueness, superiority, and entitlement.

Unfortunately, the few empirical tests of these assumptions have been equivocal. To start, the evidence suggests that the identities that people convey on their social network sites are relatively accurate portrayals of themselves. Independent observers who rated individuals based only on their social networking sites developed impressions that correlated significantly with participants’ actual, but not ideal, selves (Back et al., 2010). Thus, the notion of an idealized self on Facebook has received little empirical support (but also see Zhao, Grasmuck, & Martin, 2008). On the other hand, individuals do seem to receive primarily positive attention on Facebook. Indeed, Valkenburg, Peter, and Schouten (2006) found that the vast majority of comments that were posted on Facebook and in response to someone else’s posts were favorable. It is true that some feedback, particularly feedback given to posts made by low self-esteem individuals (see Forest & Wood, 2012), is sometimes or even often negative, but the idea that Facebook enables substantial flattering attention has garnered some support.

Experimental investigations of the influence of social networking on narcissism have been similarly ambiguous. Such work suggests that social networking use can affect self-regard; however, the implications for narcissism remain unclear. For instance, Gonzales and Hancock (2011) randomly assigned participants to complete a survey while they worked on their Facebook pages, while sitting at a computer in front of a mirror (objective self-awareness condition), or while sitting at a powered-down computer (control condition) and found that self-esteem levels increased most among those who perused their Facebook accounts. Gentile et al. (2012) performed two experiments in which they assigned participants randomly to either peruse and write about their social networking pages or to peruse an alternative website and then to complete measures of self-esteem and narcissism. The first study observed higher narcissism scores after perusing and writing about MySpace pages, as compared to perusing Google Maps; the second study observed higher self-esteem, but not higher narcissism, after Facebook use. The authors interpreted these findings as reflecting a difference between the two sites in their self-promotional nature, with MySpace offering more such opportunities and being associated more clearly with exhibitionistic self-presentation.

The current work focuses on Facebook use in hopes of providing an additional and somewhat more complex investigation of the possibility that Facebook use can facilitate narcissism. We focused on Facebook, rather than MySpace, because of the relative popularity of this site. As mentioned previously, more than one billion people now use Facebook, making it the most popular social networking site, and in fact, the most popular website, in the world. As such, the possibility that Facebook use can facilitate narcissism has far-reaching implications.

We also differentiated between different types of activities in which one might engage on Facebook. After all, Facebook use (and indeed, all social networking site use) may involve a diverse array of tasks, only some of which seem likely to facilitate narcissism. For instance, each SNS offers a variety of self-promotional activities, such as posting status updates, uploading pictures, and adding or changing biographical information. Such tasks, what we call “agentic” tasks, are those that seem most likely to facilitate narcissism. It is such tasks that correlate most strongly with grandiose narcissism in the correlational work described above (see Wang et al., 2012) and that Gentile et al. (2012) implicate in their discussion of MySpace’s effects on narcissism. Such tasks provide the opportunity for positive attention that is often implicated in theories about social networking’s influence on the self. It is also worth noting that it is on such agentic tasks and traits that narcissists regard themselves quite favorably (Campbell, Rudich, & Sedikides, 2002). On the other hand, social networking use often includes activities that focus on others. These interpersonal, “communal activities” include, among other things, replying to or “liking” others’ updates and posts, reaching out with birthday greetings (a popular use of Facebook), and sending other individuals private messages. Such communal tasks, and the interpersonal interests that they imply, are patently non-narcissistic. Indeed, Wang et al. (2012) found that self-esteem, but not narcissism, predicted responding to other people’s Facebook posts. Further, individuals high in narcissism do not consider communal traits to be self-descriptive (Campbell et al., 2002), and narcissism is either unrelated, or even inversely related to, interest in genuine interpersonal connection (Campbell, 1999). As such, it seems reasonable to think that communal social networking use would undercut, or at least not facilitate, grandiose narcissism.

We investigated the influence of agentic and communal Facebook use in two experiments. A preliminary experiment involved a relatively small sample of male participants who were randomly assigned to one of three computer activity conditions (agentic Facebook activity, communal Facebook activity, control computer activity). The second experiment included a larger, more diverse sample as well as both computer and non-computer conditions of individuals that engaged in agentic, communal, or control activities. We reasoned that any effect of social networking on narcissism would be observed only after agentic activity on the social network site but not after communal activity.