Self-affirmation theory and Facebook benefits PDF

474694 ality and Social Psychology BulletinToma and Hancock PSPXXX10.1177/0146167212474694Person...

474694 ality and Social Psychology BulletinToma and Hancock PSPXXX10.1177/0146167212474694Person Personality and Social Self-Affirmation Underlies Facebook Use Psychology Bulletin 39(3) 321–331 ­ © 2013 by the Society for Personality and Social Psychology, Inc Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0146167212474694 Catalina L. Toma1 and Jeffrey T. Hancock2 http://pspb.sagepub.com Abstract Social network sites, such as Facebook, have acquired an unprecedented following, yet it is unknown what makes them so attractive to users. Here we propose that these sites’ popularity can be understood through the fulfillment of ego needs. We use self-affirmation theory to hypothesize why and when people spend time on their online profiles. Study 1 shows that Facebook profiles are self-affirming in the sense of satisfying users’ need for self-worth and self-integrity. Study 2 shows that Facebook users gravitate toward their online profiles after receiving a blow to the ego, in an unconscious effort to repair their perceptions of self-worth. In addition to illuminating some of the psychological factors that underlie Facebook use, the results provide an important extension to self-affirmation theory by clarifying how self-affirmation operates in people’s everyday environments. Keywords self-affirmation, emotional well-being, social network sites, Facebook Received October 17, 2011; revision accepted October 18, 2012 Social network sites (SNSs) have taken the world by storm. ego needs regarding desired self-images. In turn, these ego Facebook, for instance, counts more than one billion active needs motivate SNS use. We focus our analyses on Facebook, users who spend nearly an hour online every day (Facebook currently the world’s most popular SNS and one that encap- Statistics, 2012). Even older adults are tuning in, with almost sulates the quintessential features of SNSs (boyd & half of Americans above the age of 50 currently registered as Ellison, 2007). SNS users—a 100% increase from the previous year (Pew Examining the self-affirming qualities of Facebook also Research Center, 2010). A fundamental question, then, is offers a unique opportunity to extend self-affirmation theory. what makes these sites so universally appealing. Why do This theory has been tested extensively in laboratory set- people gravitate toward them in such large numbers and with tings, using contrived self-affirmation activities, such as such dedication? ranking one’s values and writing essays assigned by an The media and public opinion are ripe with speculations experimenter. Such methods of self-affirmation are unlikely on this issue, most of them pessimistic: SNSs are viewed to be adopted spontaneously in people’s own environments as a convenient tool for procrastinating, gossiping, relieving and everyday lives. How do individuals self-affirm on their boredom, or expressing narcissistic drives (see also Buffardi own? As Sherman and Cohen (2006) speculated, after suf- & Campbell, 2008). Academic research has begun examin- fering a blow to the ego, ing subscribers’ self-reported motivations for SNS use and revealed that relational needs, such as keeping tabs on one’s perhaps the individual returns home and browses the social network and maintaining relationships, are frequently Internet, checking for information on a coming elec- cited (see Wilson, Gosling, & Graham, 2012, for a review). tion (affirming their political identity), or examines Yet more basic, and less consciously available, ego needs the scores from last night’s game (affirming a valued may also provide a compelling account of why and when social identity). In such situations, people may think people gravitate toward these sites. that they are procrastinating, but this procrastination In the present article we argue that self-affirmation theory (Steele, 1988) can serve as a cohesive theoretical narrative 1 University of Wisconsin–Madison, USA 2 for understanding important aspects of SNSs’ appeal. We Cornell University, Ithaca, NY, USA propose that SNSs that allow users to (a) craft self-presentations Corresponding Author: that reveal core aspects of their self-concept, such as social Catalina L. Toma, University of Wisconsin–Madison, 6144 Vilas Hall, affiliations and treasured characteristics, and (b) highlight social 821 University Ave., Madison, WI 53706, USA connections with friends and family, satisfy fundamental Email: [email protected] 322 Personality and Social Psychology Bulletin 39(3) may serve an important integrity-reparative function. attitudes and preferences (e.g., politics, religion, favorite (p. 64) music), and, most importantly, social connectedness with friends and family. In fact, the latter has been identified as Here we investigate the possibility that Facebook profiles the most potent source of self-affirmation (Sherman & constitute such an ecologically valid outlet of self-affirmation, Cohen, 2006). It is important to note that social norms around and in so doing we address theoretical and practical facets of Facebook profile self-presentation dictate high levels of dis- self-affirmation theory. closure about these domains (Tufekci, 2008; Young & Quan- Haase, 2009). In one study, 82% of Facebook users revealed highly personal information, such as birth date, political and Facebook and Ego Needs sexual orientation, and the name of their relationship partner The primary premise of self-affirmation theory is that people (Gross & Acquisti, 2005). In another study, large numbers of have a fundamental need to see themselves as valuable, wor- users reported a belief that the disclosure of such private thy, and good. This need for a positive self-image is an information is essential in making SNSs useful (Tufekci, 2008). important motivator of behavior. People routinely dismiss, Second, self-affirming profiles must offer a positive and distort, or avoid information that threatens their self-worth. desirable self-presentation. On Facebook, this should be the Conversely, they value, cultivate, and gravitate toward infor- case for several reasons. The presence of a large audience mation that reinforces it. who can scrutinize the profile should strongly motivate users One strategy for satisfying the fundamental need for self- to put their best foot forward (Leary & Kowalski, 1990). worth is self-affirmation, defined as the process of bringing Moreover, technological affordances such as the ability to to awareness essential aspects of the self-concept, such as select and edit statements (i.e., editability), and take unlim- values, meaningful relationships, and cherished personal ited time to compose them (i.e., asynchronicity), enable users characteristics (Sherman & Cohen, 2006; Steele, 1988). to accomplish this goal (Walther, 2007). Indeed, research According to self-affirmation theory, people are generally suggests that these technological affordances allow online motivated to seek such information in the environment, and communicators to craft optimized, highly desirable self-pre- this need becomes particularly salient after an ego threat, as sentations (Toma, Hancock, & Ellison, 2008). While self- people unconsciously attempt to repair their sense of self- generated profile statements should be positive, those worth. Another key proposition of self-affirmation theory is generated by friends should be so as well. Editability enables that after attending to self-affirming information, individu- users to delete any objectionable postings made by friends, als’ tendency to engage in defensive processes, such as dis- and even to terminate friendship connections altogether if missal or distortion, is reduced or eliminated. This is the case they are no longer desired. Consistent with this reasoning, because self-affirmation has already secured individuals’ research shows that messages from friends that are posted on sense of self-worth and self-integrity, rendering these other the Facebook profile are typically warm, supportive, and defense mechanisms unnecessary. Consequently, self-affirmation validating (Sas, Dix, Hart, & Su, 2009). has the salutary effect of making people more open-minded Finally, self-affirming profiles must be accurate. Lying about and secure toward threatening events (Steele, 1988). one’s characteristics can only highlight one’s inadequacy in How does self-affirmation theory apply to Facebook? meeting internal standards for self-worth, and hence the decep- Facebook operates by connecting users with their friends, tive information should be useless for self-affirmation. Facebook family, and acquaintances. Users create detailed profiles profiles are indeed governed by constraints that minimize decep- describing their activities, interests, and values, and then tion. One such constraint is the presence of “friends” who know articulate “friendships” with other users in the system. the self-presenter both online and offline, and therefore may be Communication between “friends” is publicly posted and able to verify the veracity of profile statements (Donath & boyd, archived onto the profile (boyd & Ellison, 2007). Given this 2004; Walther & Parks, 2002). People are significantly less focus on representing the key aspects of self and, most likely to lie when there is a high likelihood of getting caught. importantly, embedding the self in a network of close per- Research on online dating confirms that online self-presenters sonal relationships, it is plausible that Facebook profiles may are more honest if their friends and acquaintances have access to constitute a venue for self-affirmation and that users gravi- their profile (Toma et al., 2008). A related constraint to decep- tate toward them in an unconscious effort to elevate their tion is the fact that some of the profile information is contributed perceptions of self-worth and self-integrity. by friends. This information should be more credible because it Do Facebook profiles meet the criteria for self-affirmation? is less amenable to the self-presenter’s control. Importantly, First, to be self-affirming, profiles must represent the friends’ postings are a testament of friendship that, because of its domains of self on which self-worth is contingent (Crocker public nature, should constitute a reliable indicator of social con- & Wolfe, 2001). Indeed, Facebook self-presentation involves nectivity (Donath & boyd, 2004). Consistent with these claims, displaying a multitude of defining aspects of self: social recent research has shown that Facebook profiles are sufficiently roles (e.g., friend, student), affiliation to networks important accurate to enable reliable personality judgments (Back et al., to the self-concept (e.g., educational institutions), central 2010). Toma and Hancock 323 In sum, this analysis suggests that Facebook profiles have standard for determining whether an activity is self-affirming the potential to confer upon users self-affirmational benefits (McQueen & Klein, 2006; Sherman & Cohen, 2006; Sherman because they encapsulate a flattering, socially connected, & Hartson, 2011). meaningful, and accurate self-portrait. Here, we use this procedure to establish whether brows- ing one’s own Facebook profile is a self-affirming activity. In addition, we directly compare the effect of Facebook pro- Advancing Self-Affirmation Theory file exposure with that of a well-established self-affirming As mentioned earlier, SNSs provide new avenues for repre- activity, to assess the relative strength of the self-affirming senting the self and the domains of self that are critical to qualities of Facebook. self-worth (e.g., relationships, values). Their proliferation provides an opportunity to test self-affirmation theory in a new setting and, in so doing, to extend it on several fronts. Method First, thanks to Facebook’s ubiquity and ease of access, we Participants. Participants were undergraduates at a Northeast- are able to investigate the operations of self-affirmation in ern university who received course extra-credit (N = 98, 68% people’s own environments and everyday lives, and thus women; M age = 19.81). Ten participants were excluded provide an important extension to extant literature that has because they were suspicious (n = 5) or because they were mostly used contrived self-affirmation exercises (McQueen not Facebook users (n = 5), reducing the effective sample & Klein, 2006). Second, existing research on self-affirma- size to N = 88. tion has overwhelmingly investigated the effects of self- affirmation, with significantly less emphasis on its causes Self-Affirmation Manipulation. A 2 (self-affirmation condi- (Harris & Epton, 2010; Sherman & Cohen, 2006). What tion: affirmed vs. non-affirmed) × 2 (self-affirmation type: motivates people to self-affirm on their own, without being Facebook vs. values essay) experimental design was used. prompted to do so by an experimenter? In our efforts to In both Facebook conditions, participants were told explicate users’ attraction toward Facebook, we provide they would take part in a “website evaluation study,” which empirical evidence regarding people’s spontaneous use of involved spending 5 min on a website and then answering self-affirming outlets. questions about it. In the Facebook self-affirmation condi- tion (n = 21), this website was participants’ own Facebook profile. Participants were told they could view any element Overview of Studies of their profile (e.g., photographs, wall, list of friends), but Our previous analysis indicates that Facebook profiles meet could not navigate to someone else’s profile. At the end of the criteria for self-affirmation. We empirically test the self- the study, participants in this condition were asked to temporar- affirming value of Facebook in two studies. Study 1 exam- ily “friend” the experimenter on Facebook, so we could have ines whether Facebook profiles are self-affirming in the access to their profile information. All participants agreed. sense of satisfying users’ fundamental need for self-worth In the Facebook control condition (n = 24), participants and self-integrity. Study 2 asks whether users capitalize on were asked to examine a stranger’s Facebook profile. This Facebook’s self-affirmational qualities by seeking it out for stranger was in fact the previous participant in the Facebook the purpose of ego repair. self-affirmation condition. Thus, participants were yoked such that each participant in the control condition viewed the pro- file of a participant in the Facebook self-affirmation condi- Study 1 tion. Care was taken that participants in the Facebook control To establish whether a certain activity is self-affirming, the condition not be acquainted with the people whose profile self-affirmation literature uses a defensiveness-reducing they were viewing. The yoking procedure ensured that, as a paradigm (McQueen & Klein, 2006; Sherman & Cohen, group, participants in these two conditions examined the 2006). This paradigm is based on the theoretical claim that, exact same profiles. by satisfying the need for self-worth and self-integrity, self- In the values essay self-affirmation condition (n = 22), affirmation diminishes the defensive responses that are natu- participants ranked six values in order of personal impor- rally elicited by ego threats. tance (business, art-music-theater, social life–relationships, Several steps are involved in implementing this paradigm. science–pursuit of knowledge, religion–morality, government– Participants are first subjected to an ego threat. Then they are politics) and then wrote for 5 min about why their highest prompted to either engage in a self-affirming activity or not. ranked value was important to them. In the values essay con- Finally, their defensive responses are measured. If the activ- trol condition (n = 21), participants wrote about why their ity is self-affirming, a decrease in defensive responses will lowest ranked value was important to the average college be observed. This indirect approach, whereby self-worth and student. This value-affirmation procedure is the single most self-integrity are not explicitly measured but are assumed to widely used self-affirmation manipulation (McQueen & increase if reduced defensiveness is observed, is the gold Klein, 2006). 324 Personality and Social Psychology Bulletin 39(3) Procedure. Because self-affirmation occurs nonconsciously, participants were given a cover story about the purpose of the study (adapted from Swann, De La Ronde, & Hixon, 1994). Participants were asked to pilot the viability of a distance-learning Public Speaking course. To this end, they would prepare a short (3-5 min) speech on the legality of abortion, a common topic in such classes, and deliver it through an ostensibly live camera to an evaluator, who would provide written feedback on the speech. Participants’ main task would then be to rate the fairness and usefulness of the feedback: Was it accurate? Was the evaluator able to form a good impression of the participants’ public speaking abilities? Participants were observed through a one-way mir- ror to ensure that they delivered the speech. While waiting for the evaluator to write the speech feed- back, participants were invited to complete an additional study to double their extra-credit points. All participants agreed. This ostensibly unrelated study was in fact the self- affirmation manipulation, and it was run by a different experimenter to enhance credibility. The first experimenter then returned with a sealed enve- lope containing the speech feedback. All participants were given the same generic negative feedback. A manipulation Figure 1. Acceptance of feedback by condition check confirmed that the feedback was perceived as nega- tive, regardless of whether participants were affirmed (M = 3.12, SD = 1.55, on a scale from 1 = not at all positive to 9 = a tion venue (Facebook or values essay) as the between-sub- great deal positive) or not (M = 2.78, SD = 1.46). Participants jects factors. The model fit the data well, F(3, 84) = 8.43, p < then filled out a “confidential” questionnaire about the valid-.001, partial η2 = 0.23, and revealed the expected main effect ity of the feedback. This allowed them to express defensive of self-affirmation, F(1, 84) = 24.51, p <.001, partial η2 = responses to the ego threat. The self-affirmation exercise 0.23, whereby participants who were self-affirmed, regard- was thus completed prior to the ego threat, consistent with less of the venue (M = 5.72, SD = 1.21) were more accepting research demonstrating that timing is critical in effectively of the feedback than non-affirmed participants (M = 4.31, SD reducing defensiveness (Critcher, Dunning, & Armor, 2010). = 1.42). As predicted, the effect of self-affirmation venue Debriefing was done through a funneled procedure, which was not significant, F(1, 84) = 0.41, p =.52, suggesting that identified suspicious participants. participants who spent time on Facebook (M = 5.05, SD = 1.57) reacted to the feedback in the same way as those com- Measures. The dependent measure, completed at the end of pleting the values essay self-affirmation (M = 4.94, SD = the study, was participants’ acceptance of the negative feed- 1.41). Finally, there was no interaction between the self- back received on their speech (adapted from Swann et al., affirmation manipulation and the self-affirmation venue, 1994). Participants rated the feedback on a scale from 1 (not F(1, 84) = 0.45, p =.51, which indicates that self-affirma- at all) to 9 (a lot) across five dimensions: (a) perceived accu- tion operated similarly on both venues (Facebook and val- racy (5 items), (b) evaluator competence (3 items), ues essay; see Figure 1). (c) task diagnosticity (2 items), (d) attribution of perfor- Simple effects tests confirmed that (a) participants who mance to self or external circumstances (3 items), and (e) lik- examined their own Facebook profiles (M = 5.91, SD = 1.10) ing of the evaluator (2 items). Because these measures were were more accepting of the feedback than participants highly correlated, they were averaged into an acceptance of who examined a stranger’s profile (M = 4.31, SD = 1.57), feedback index (α =.88). t(43) = 3.90, p <.001, Cohen’s d = 1.19, and (b) partici- Measures of trait self-esteem (Rosenberg, 1965) and pants who were affirmed on Facebook (M = 5.91, affect (Crocker, Niiya, & Mischkowski, 2008) were also col- SD = 1.10) were equally accepting of the feedback as those lected but are not reported in this study. affirmed through the values essay (M = 5.53, SD = 1.31), t(41) = 1.01, p =.32. Together, these results provide evi- dence that Facebook profile exposure is a self-affirming Results and Discussion activity and that the self-affirmation earned from Facebook An analysis of variance was set up with the self-affirmation does not differ from that earned from the classic values manipulation (affirmed or non-affirmed) and self-affirma- essay manipulation. Toma and Hancock 325 In sum, Study 1 found that participants who had exam- of five “unrelated” studies ostensibly going on in the lab, in ined their Facebook profiles for 5 min were more likely to exchange for an additional extra-credit point. assume responsibility and less likely to blame others when One of these studies was browsing their own Facebook receiving negative feedback on an academic task. This non- profile. The four decoy studies involved an online activity defensive attitude toward an ego threat is the hallmark of designed to be as similar as possible to Facebook profile self-affirmation (Steele, 1988) and provides compelling evi- browsing, but not self-affirming. Pilot testing revealed four dence that Facebook profile exposure increases self-worth activities that did not differ from Facebook profile browsing and self-integrity. Moreover, exposure to one’s own in terms of level of engagement, excitement, or difficulty: Facebook profile had the same effect as engaging in a well- (a) watching YouTube videos, (b) listening to online music, validated self-affirming activity. These findings indicate that (c) reading online news, and (d) playing online videogames. spending time on Facebook may fulfill important ego needs. Participants did not in fact complete any of these studies. By showcasing a version of self that is attractive, successful, Funneled debriefing was used to identify suspicious and embedded in a network of meaningful relationships, participants. Facebook enhances users’ perceptions of self-worth. Measures Dependent measures. Participants ranked the five online Study 2 activities in the order of personal preference (1—I would Study 1 provides evidence that Facebook profiles constitute MOST prefer to participate in this study; 5—I would LEAST a source of self-affirmation, in the sense of endowing users prefer to participate in this study). Subsequently, they wrote with a sense of overall self-integrity and well-being. But do a brief paragraph about why they chose their most preferred users spontaneously seek out Facebook for the purpose of activity, to assess any conscious awareness of self-affirmational buttressing self-worth, without being instructed to do so by benefits. an experimenter? Do they capitalize on the benefits provided by their Facebook profile in time of psychological need? Covariates. Several measures that may affect activity Study 2 addresses this question. choice were collected. Participants’ age and gender were Self-affirmation theory posits that perceived ego threats assessed in the beginning of the study. Then, participants activate an unconscious motive to restore the adequacy and rated each activity across the following dimensions: (a) per- integrity of the self. This motive leads people to search for ceived effort to complete, (b) level of familiarity, and self-affirming information in the environment (Steele, (c) how interesting, engaging, and positive each was per- 1988). Although central to self-affirmation theory, this ceived. Finally, participants reported whether they were proposition has received little empirical attention (for Facebook users and, if so, how satisfied they were with their exceptions, see Steele, 1975; Stone, Wiegand, Cooper, & profile self-presentation, and how positive, accurate, and Aronson, 1997; Tesser, Crepaz, Collins, Cornell, & Beach, comprehensive they considered their self-presentation to be. 2000). We now test it in the context of affirming the self via Facebook, a ubiquitous social media outlet. In doing so, we attempt to clarify not only whether Facebook profiles can Results and Discussion serve as a venue for self-affirmation, but also whether they The rank of the activities was recoded to reflect whether par- do serve as one. ticipants’ first preference was Facebook or not (i.e., any one of the four decoy activities). After receiving neutral feedback, participants were as likely to choose Facebook profile brows- Method ing as any of the other activities. In this condition, 30.8% of the Participants. Undergraduate students received course extra- participants chose Facebook, a rate not significantly different credit for their participation (N = 86; 66% women; M age = from chance, z(39) = 1.69, p =.10. After receiving negative 19.79, SD = 1.04). Ten participants were excluded because feedback, however, almost twice as many participants (59.6%) they were not Facebook users. Fifteen others were excluded chose Facebook, a rate significantly higher than chance, z(47) = because of suspiciousness. Excluded participants were 6.79, p <.001. A chi-square analysis confirmed that partici- equally dispersed across conditions. pants were more likely to choose Facebook as their first prefer- ence when their ego had been threatened (negative feedback Experimental Design. The cover story, with participants condition) than when it had not (neutral feedback condition), required to engage in a public speaking task, was identical to χ2(1, N = 86) = 7.11, p =.01, Cohen’s w = 0.27 (see Table 1). that used in Study 1. However, participants were randomly A fixed-effects linear model was used to test whether this assigned to receive negative (n = 47) or neutral feedback pattern held when controlling for covariates. The model (n = 39) on their speaking performance. In addition, after contained each activity’s rank as the dependent variable, receiving the feedback, they were not instructed to self- condition (negative vs. neutral feedback) as fixed-effects affirm. Instead, they were given a choice to participate in one predictor, and all the proposed covariates. As before, when 326 Personality and Social Psychology Bulletin 39(3) Table 1. Number and Percentage of Participants Who Chose 1988), is a useful framework for understanding why (i.e., to Facebook Versus Any of the Four Non-affirming Websites in increase perceptions of self-worth) and when (i.e., following Each Condition a threat to one’s desired self-image) people tend to spend time on Facebook. The results provide several theoretical Other advancements to self-affirmation theory and also contribute Facebook activities new insights on the causes and effects of SNS use. Below, Total % Total % both types of contributions are discussed in detail. Neutral feedback condition 12 30.8 27 69.2 Negative feedback condition 28 59.6 19 40.4 Self-Affirmation Theory Revisited A long research tradition in psychology has examined the uniquely human motivation to pursue self-worth (Tesser, participants’ egos were not threatened, they did not exhibit 1988). Chief in this line of research is self-affirmation the- a preference for Facebook. The rank of Facebook (rank = ory (Steele, 1988), that has launched over 20 years of aca- 2.80) was not different from the average rank of the other demic investigation on the importance and benefits activities (rank = 3.09) in the neutral feedback condition, associated with maintaining self-worth and self-integrity. By t.s.(400) = 0.29, p =.29. However, when participants’ egos examining whether SNSs, a recent Internet phenomenon, were threatened, they displayed a preference for spending play a role in self-worth maintenance processes, the present time on Facebook: Facebook’s rank was significantly lower research addressed several lacunae in the extensive self- (rank = 2.42) than the average rank of the other activities affirmation literature, as detailed below. (rank = 3.09), t.s.(400) = 0.66, p =.008. None of the covariates affected the rank of the online Operationalizing Self-Affirmation. Theoretically, self-affirmation activities, with the exception of how engaging the activities is the process of bringing to awareness defining aspects of were perceived (β = −0.17, p =.003). Participants in the neu- the self-concept, such as values, goals, and treasured per- tral feedback condition were more likely to select engaging sonal attributes. This broad definition encompasses many activities, consistent with the claim that people select engag- activities that could potentially affirm the self. Yet research ing media to relieve boredom (Zillmann, 1988). to date has operationalized self-affirmation quite narrowly. Together, the chi-square and linear model analyses A systematic review of self-affirmation manipulations show that participants tended to gravitate toward their (McQueen & Klein, 2006) shows that by far the most widely Facebook profiles, a self-affirming venue, when their egos used operationalization of self-affirmation is ranking a series were threatened, but not when their egos were intact, regard- of values in order of personal importance and writing a short less of gender, perceptions of the online activities, or the per- essay about the highest ranked value. Other, less frequently ceived quality of their own Facebook profile. In addition, an used operationalizations, include bogus positive feedback on analysis of the open-ended responses showed that only five tests of personality and social skills (e.g., Heine & Lehman, participants (5.81%) seemed consciously aware of self- 1997) and unscrambling sentences that contain self-affirming affirming properties of their chosen activity (e.g., “I chose messages (Stone & Cooper, 2003). Note that these self- Facebook because I love me!”). affirmation manipulations are activities developed by social In sum, Study 2 provides evidence that Facebook users psychologists and administered in lab settings. An important gravitate toward their online profiles after experiencing psy- question, then, is what are the everyday, “natural” equiva- chological distress, in an apparently unconscious effort to lents of the contrived self-affirmation exercises? improve perceptions of self-worth. These data illuminate an The present research shows that Facebook profiles consti- important, and previously unexamined, motive for Facebook tute such an everyday outlet for self-affirmation and is thus use: restoring perceptions of self-worth. Study 2 provides one of the first studies to empirically show how the theoreti- support for the key proposition of self-affirmation theory cal construct of self-affirmation translates into everyday that people are intrinsically motivated to maintain elevated activities. We offer several pieces of evidence in support of perceptions of self-worth and self-integrity and that they the contention that Facebook profiles are self-affirming. seem to do so by unconsciously seeking self-affirming activ- First, we use the defensiveness-reducing paradigm of self- ities in the environment. affirmation effects, the gold standard in the field, to show that Facebook profile browsing reduces defensiveness to ego threats and hence can be assumed to increase perceptions of General Discussion self-worth and self-integrity. Second, we show that the What makes Facebook so attractive and compelling to mil- effects of Facebook profile browsing are similar to those of a lions of users? The present studies reveal that one reason well-validated self-affirmation manipulation (see Napper, may be its self-affirming qualities. The findings reported Harris, & Epton, 2009, for a similar comparison proce- here provide evidence that self-affirmation theory (Steele, dure). Third, our experiments highlight several important Toma and Hancock 327 characteristics of Facebook that are consistent with self- possible that this everyday activity may also fulfill related affirmation theory. There is indication that Facebook is psychological functions. Consider first self-verification the- sought unconsciously (Study 2), in line with claims that con- ory, which claims that individuals seek psychological coher- scious affirmation attempts are ineffective (see Sherman & ence, manifested as a desire to be perceived by others in a Cohen, 2006, for a review). Most importantly, Facebook way that is aligned with self-perceptions (Swann et al., profile browsing soothes ego threats that are unrelated to 2003). Facebook profiles may well represent individuals’ Facebook content (Study 1). Recall that the ego threat efforts to present themselves to others in a manner consistent administered affected participants’ academic abilities, while with self-perceptions. Similarly, symbolic self-completion Facebook profile content has been shown to not contain theory claims that individuals create symbols of attainment information about academics (Rosenbaum, Johnson, Stepman, to present themselves as capable in defining areas (Wicklund & Nuijten, 2010; Toma & Carlson, 2012). This decoupling & Gollwitzer, 1981). Facebook profiles symbolically repre- of the threatened domain of self from the affirmed domain is sent, through photographs and textual statements, social the sine qua non of self-affirmation theory. achievements and treasured characteristics. As a “natural,” An important question is whether Facebook profile everyday activity that is not narrowly tailored by experi- browsing, the everyday activity scrutinized here, is self- menters, it is possible that Facebook profile browsing fulfills affirming or represents a related psychological process for multiple psychological roles, depending on the ego threat regulating feelings of self-worth. Indeed, a multitude of experienced by the participants (i.e., related or unrelated to theories make similar claims about people’s ego needs and Facebook content). Future empirical work is needed to iden- the means they use to achieve them. This literature, dubbed tify any self-verification or symbolic self-completion prop- the “self-zoo” because of its blurry boundaries (Tesser, erties of Facebook. If such properties emerge, an argument 2000), includes self-affirmation theory, self-verification could be made as to the substitutability of these processes theory (Swann, Rentfrow, & Guinn, 2003), symbolic self- and self-affirmation. Within the “self-zoo” literature, Tesser completion theory (Harmon-Jones, Schmeichel, & Harmon- and colleagues (Tesser, 2000; Tesser & Cornell, 1991; Jones, 2009), the self-esteem evaluation (SEM) model Tesser et al., 2000) have proposed that self-esteem regula- (Tesser & Cornell, 1991), implicit self-esteem compensation tion mechanisms may be substitutable for one another (Rudman, Dohn, & Fairchild, 2007), social comparison because they fill the same higher order purpose of protecting (Festinger, 1954), and cognitive dissonance (Festinger, 1957). self-worth. The substitution principle could be bolstered by Our experimental procedure and findings help rule out findings that one activity (i.e., Facebook profile browsing) some of these alternate possibilities. First, as mentioned ear- can serve multiple self-worth restorative functions. Future lier, in Study 1 the threatened self-domain (academics) dif- research is necessary to fully investigate these claims. fered from the boosted domain (Facebook self-presentation), Future work is also necessary to replicate these findings as per self-affirmation theory’s claims that self-affirmation and to clarify how exactly Facebook profile browsing affects is an indirect process. Conversely, self-verification, sym- users. For instance, one way to demonstrate the suitability of bolic self-completion, social comparison, and cognitive dis- self-affirmation theory to Facebook profile self-presentation sonance thoeries posit that individuals repair ego threats by is to investigate whether exposure to these profiles elicits accessing or boosting information related to the same domain other known effects of self-affirmation. Such effects include as the one that had been threatened (Festinger, 1954, 1957; an increase in self-concept clarity (Waksiak & Trope, 2009), Gollwitzer, Wicklund, & Hilton, 1982; Gollwitzer & a boost in ego resources, and the use of higher level mental Wicklund, 1985; Swann et al., 2003; Swann & Hill, 1982; construals (Schmeichel & Vohs, 2009). Swann & Read, 1981; Trepte, 2005). Second, the form undertaken by some of the related processes is distinct from Spontaneous Self-Affirmations. Existing empirical work on self-affirmation’s, albeit their function is similar (i.e., to self-affirmation has favored effects rather than causes of maintain positive self-evaluations). Specifically, SEM and self-affirmation, and positive, rather than negative, effects social comparison processes represent interpersonal strate- (Sherman & Cohen, 2006). This is understandable given the gies whereby individuals compare themselves with, and help sizable benefits conferred by this simple psychological inter- or hinder, others for the purpose of regulating their own feel- vention. But it is equally important to determine under what ings of self-worth. Implicit self-esteem compensation occurs circumstances self-affirmational benefits materialize. When automatically after one’s self-worth has been threatened and do people pursue self-affirmation? What causes them to seek does not involve engaging in any specific activity. Hence, ego-restorative activities? these related processes are narrowly defined and do not meet An explicitly motivational theory, self-affirmation pro- the criteria for self-affirmation (i.e., reassuring individuals of poses that threats to the ego activate an unconscious need to their overall self-worth and self-integrity, through state- self-affirm—that is, to restore perceptions of the adequacy ments of quintessential values, beliefs, and characteristics). and integrity of the self. As discussed earlier, similar claims While our experimental procedure allows us to conclude are made by cognitive dissonance, self-verification, sym- that Facebook profile browsing is self-affirming, it is bolic self-completion, and SEM theorists. While these other 328 Personality and Social Psychology Bulletin 39(3) theoretical perspectives’ claims have received empirical test- participation (Valenzuela, Park, & Kee, 2009); increased ing, self-affirmation theory’s claims about the pursuance of social self-esteem and emotional well-being (Valkenburg, self-affirmation following an ego threat have been relatively Peter, & Schouten, 2006); and increased college student sidelined in favor of examining effects of self-affirmation. In motivation, affective learning, and positive classroom cli- showing that participants spontaneously gravitated toward mate (Mazer, Murphy, & Simonds, 2007). The self-affirmation Facebook, a self-affirming venue, after having their aca- framework adds a new class of intrapersonal benefits to demic identity threatened (Study 2), the present research extant research—boosting perceptions of self-worth and provides much-needed empirical support for this key theo- self-integrity. Together, this literature pinpoints to consider- retical proposition. able psychological benefits of an interpersonal and intraper- How does spontaneous self-affirmation operate? In his sonal nature that can be derived from SNS use. original conceptualization of the theory, Steele (1988) pro- Consider next the causes, or motivators of SNS use. Prior posed that individuals’ means of self-affirmation will be dic- research in this area has predominantly used a self-report tated by (a) accessibility, or “the degree to which a given methodology to examine why users gravitate toward SNSs. adaptation is accessible in the individual’s perception, mem- In these studies, users report interpersonal motivations, such ory, or imagination” (p. 291), and (b) effectiveness-cost as maintaining social connections, sharing identities, and ratios, or its judicious use of time and resources. For instance, engaging social surveillance (Gross & Acquisti, 2005; assuming that both means of self-affirmation are available, Joinson, 2008; Lampe, Ellison, & Steinfield, 2008). The an individual will prefer doing the dishes to volunteering to present work adds to this literature by exposing motivations organize a charity benefit. For millions of users, Facebook is that may not be consciously available and that stem from both easily accessible and cost-effective, making it a plausi- intrapersonal goals (i.e., preserving self-integrity). In show- ble venue for spontaneous self-affirmation in everyday life. ing that users gravitated toward their Facebook profiles fol- However, future research is needed to examine to what lowing a blow to their ego, this work offers a nuanced view extent Facebook self-affirmation is preferred to other types of Facebook use as the product of situational factors (in this of spontaneous self-affirmations, such as calling a friend or case, having one’s desired self-image challenged through going shopping. negative feedback on an important task) and users’ implicit goals to maintain a positive self-view. Understanding the Psychological Draw of Facebook Practical Implications The self-affirmation framework highlights the profound The finding that SNS profiles can act as an everyday source implications that Facebook profile browsing can have on the of self-affirmation has several practical implications. First self-concept. Whereas conventional wisdom maintains that and most evidently, the experience of engaging with one’s Facebook use is merely a time sink and leads to an assort- profile-based self may confer emotional benefits to the mil- ment of negative consequences, the present findings provide lions of SNS users, in terms of restoring deep-seated notions evidence that it can be a psychologically meaningful activ- of self-integrity. Perhaps online daters who are anxious ity, that supplies a sense of well-being at a deep level. In about being single or recently divorced may find comfort in addition, the self-affirmation frameworks offers insights on the process of composing or reviewing their online profiles, the baffling mass appeal of Facebook, that cuts across gen- as it allows them to reflect on their core values and identity. erational and cultural lines. The extraordinary amount of Perhaps students who are feeling stressed about upcoming time people spend on Facebook may be a reflection of its exams might similarly find solace in their SNS profiles. An ability to satisfy ego needs that are fundamental to the important task for future research is to determine the bound- human condition, yet whose covert operation makes them ary conditions of SNS self-affirmation. How often do SNS less amenable to scrutiny. These ego needs pertain to how users engage in self-affirming activities (i.e., own profile people wish to see themselves—socially attractive and embed- browsing and construction)? If they do engage in these self- ded in a network of meaningful relationships. affirming activities frequently, what are the effects of having The self-affirmation framework is the first to provide a one’s self-worth constantly bolstered? Note that extant cohesive, parsimonious, and psychologically driven expla- research has only examined the effects of solitary self-affir- nation for the appeal of Facebook, and is consistent with an mation inductions; Facebook profile browsing may provide emerging body of research that has identified psychological the opportunity to investigate the effects of repeated self- benefits earned from SNS use. Consider first the effects of affirmations. Facebook use. Recent research has uncovered a series of cor- Second, self-affirmation is associated with a cornucopia relations between SNS use, broadly defined, and positive of psychological benefits, such as reducing the gender psychological and social outcomes, such as increased social achievement gap in the sciences and engineering (Miyake capital (Ellison, Steinfield, & Lampe, 2007); increased life et al., 2010), rendering individuals more receptive to anti- satisfaction, social trust, civic engagement, and political smoking messages (Armitage, Harris, Hepton, & Napper, 2008), Toma and Hancock 329 promoting wellness in breast cancer patients (Creswell et al., conference and at the conference on Human Factors in Computing 2007), and reducing self-serving biases that distort social Systems. perception (Sherman & Kim, 2005). Precisely because such sizable benefits ensue from self-affirmation, researchers are Declaration of Conflicting Interests becoming interested in designing self-affirmation interven- The author(s) declared no potential conflicts of interest with respect tions that can be implemented in applied settings (Napper to the research, authorship, and/or publication of this article. et al., 2009; Sherman & Hartson, 2011). As noted in recent research, designing such interventions is a challenging task. Funding Ideally, these interventions should be applicable to a wide The author(s) received no financial support for the research, author- range of potential subjects; should not require individual tai- ship, and/or publication of this article. loring; should offer easy access to a control equivalent, for the purpose of gauging intervention success; should operate References unconsciously; and should be compatible with mass com- Armitage, C. J., Harris, P. R., Hepton, G., & Napper, L. (2008). munication, for the purpose of wide dissemination (Napper Self-affirmation increases acceptance of health-risk informa- et al., 2009). tion among UK adult smokers with low socioeconomic status. The present research points to the intriguing possibility Psychology of Addictive Behaviors, 22, 88-95. that Facebook profiles may constitute an easily implemented, Back, M., Stopfer, J., Vazire, S., Gaddis, S., Schmukle, S., Egloff, wide reaching and subtle self-affirmation intervention in B., & Samuel, D. G. (2010). Facebook profiles reflect actual everyday settings. Indeed, these profiles are nearly univer- personality not self-idealization. Psychological Science, 21, sally applicable, with SNSs’ user base rapidly growing; they 372-374. do not require individual tailoring—in fact, most profiles boyd, d. m., & Ellison, N. B. (2007). Social network sites: Defini- do not even need to be created specifically for the interven- tion, history, and scholarship. Journal of Computer-Mediated tion; and they offer easy control equivalence, as shown in the Communication, 13, 210-230. self-affirmation manipulation detailed here. In addition, Buffardi, L., & Campbell, K. (2008). Narcissism and social net- Facebook self-affirmation appears to operate noncon- working web sites. Personality and Social Psychology Bulletin, sciously, and is part of mass communication. Facebook pro- 34, 1303-1314. files may then be used strategically in applied self-affirmation Creswell, D., Lam, S., Stanton, A. L., Taylor, S. E., Bower, E., & interventions, such as those aimed at decreasing resistance to Sherman, D. K. (2007). Does self-affirmation, cognitive pro- antismoking messages among young adults. cessing, or discovery of meaning explain cancer-related health benefits of expressive writing? Personality and Social Psychol- ogy Bulletin, 33, 238-250. Conclusion Critcher, C. R., Dunning, D., & Armor, D. A. (2010). When self- Everyday life is rife with threats to the ego, ranging from the affirmations reduce defensiveness: Timing is key. Personality trivial (e.g., being treated rudely by a clerk, being ignored by and Social Psychology Bulletin, 36, 947-959. a friend) to the consequential (e.g., failing an exam, getting Crocker, J., Niiya, Y., & Mischkowski, D. (2008). Does writing into an argument with a spouse). These setbacks are both about important values reduce defensiveness? Psychological common and unavoidable, raising the question of how indi- Science, 19, 740-747. viduals manage to maintain a sense of self-worth and avoid Crocker, J., & Wolfe, C. T. (2001). Contingencies of self-worth. being plagued by anxiety and self-doubt. This research Psychological Review, 108, 593-623. shows that, just as setbacks and challenges are pervasive in Donath, J., & boyd, d (2004). Public displays of connection. BT everyday life, so too are opportunities to offset their effects. Technology Journal, 22, 71-82. Facebook, an SNS that is ubiquitously available, has the Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits ability to repair the damage caused by ego threats and it is of Facebook “friends:” Social capital and college students’ use actively sought after by users for the purpose of soothing a of online social network sites. Journal of Computer-Mediated wounded ego. The availability of everyday sources of self- Communication, 12, 1143-1168. affirmation, such as Facebook, appears to be a useful Facebook Statistics. (2012). Retrieved May 1, 2012 from http:// instrument in individuals’ efforts to preserve self-worth and www.facebook.com/press/info.php?statistics self-integrity. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 71-82. Acknowledgments Festinger, L. (1957). A theory of cognitive dissonance. Evanston, The authors are grateful to Dennis Regan for his valuable feedback IL: Row, Peterson. on this project, and to Ashley Downs, Rachel Boochever, Stephanie Gollwitzer, P. M., & Wicklund, R. A. (1985). The pursuit of self- Friedman, Jessica Hwang, Kerrie Lopez, Kate Pascucci, and Thaisa defining goals. In J. Kuril & J. Beckmann (Eds.), Action con- Tirado for their help with data collection. Portions of this project trol: From cognition to behavior (pp. 61-85). New York, NY: were presented at the International Communication Association Springer-Verlag. 330 Personality and Social Psychology Bulletin 39(3) Gollwitzer, P. M., Wicklund, R. A., & Hilton, J. L. (1982). Admis- Sas, C., Dix, A., Hart, J., & Su, R. (2009). Dramaturgical capital- sion of failure and symbolic self-completion: Extending ization of positive emotions: The answer for Facebook success? Lewinian theory. Journal of Personality and Social Psychol- In Proceedings of the British Computer Society Conference on ogy, 43, 358-371. Human Computer-Interaction. Cambridge, UK: ACM Press. Gross, R., & Acquisti, A. (2005). Information revelation and privacy Schmeichel, B. J., & Vohs, K. D. (2009). Self-affirmation and self- in online social networks. In Proceedings of the 12th Annual control: Affirming core values counteracts ego depletion. Jour- Workshop on Privacy in the Electronic Society. New York, NY: nal of Personality and Social Psychology, 96, 770-782. ACM Press. Sherman, D. K., & Cohen, G. L. (2006). The psychology of self- Harmon-Jones, C., Schmeichel, B. J., & Harmon-Jones, E. (2009). defense: Self-affirmation theory. In M. P. Zanna (Ed.), Advances Symbolic self-completion in academia: Evidence from depart- in experimental social psychology (Vol. 38, pp. 183-242). ment web pages and email signatures. European Journal of San Diego, CA: Academic Press. Social Psychology, 39, 311-316. Sherman, D. K., & Hartson, K. (2011). Reconciling self-protection Harris, P. R., & Epton, T. (2010). The impact of self-affirmation on with self-improvement: Self-affirmation theory. In M. Alicke health-related cognition and health behaviour: Issues and pros- & C. Sedikides (Eds.), The handbook of self-enhancement and pects. Social & Personality Psychology Compass, 4, 439-454. self-protection (pp. 128-154). New York, NY: Guilford. Heine, S. J., & Lehman, D. R. (1997). Culture, dissonance, and Sherman, D. K., & Kim, H. S. (2005). Is there an “I” in “team”? self-affirmation. Personality and Social Psychology Bulletin, The role of the self in group-serving judgments. Journal of Per- 23, 389-400. sonality and Social Psychology, 88, 108-120. Joinson, A. N. (2008). Looking at, looking up or keeping up with Steele, C. M. (1975). Name-calling and compliance. Journal of people? Motives and use of Facebook. In Proceedings of the Personality and Social Psychology, 31, 361-369. 26 Annual Human Factors in Computing Systems Conference Steele, C. M. (1988). The psychology of self-affirmation: Sustaining (CHI 2008). New York, NY: ACM Press. the integrity of the self. In L. Berkowitz (Ed.), Advances in experi- Lampe, C., Ellison, N., & Steinfield, C. (2008). Changes in use mental social psychology (Vol. 21, pp. 261-302). San Diego, CA: and perception of Facebook. In Proceedings of the Computer- Academic Press. Supported Collaborative Work Conference (CSCW 2008). New Stone, J., & Cooper, J. (2003). The effect of self-attribute relevance York, NY: ACM Press. on how self-esteem moderates attitude change in dissonance Leary, M. R., & Kowalski, R. (1990). Impression management: processes. Journal of Experimental Social Psychology, 39, A literature review and two-component model. Psychological 508-515. Bulletin, 107, 34-47. Stone, J., Wiegand, A. W., Cooper, J., & Aronson, E. (1997). When Mazer, J. P., Murphy, R. E., & Simonds, C. (2007). I’ll see you on exemplification fails: Hypocrisy and the motive for self-integrity. Facebook: The effects of computer-mediated self-disclosure on Journal of Personality and Social Psychology, 72, 55-65. student motivation, affective learning, and classroom climate. Swann, W. B., Jr., De La, Ronde, C., & Hixon, J. G. (1994). Authen- Communication Education, 56, 176-183. ticity and positivity strivings in marriage and courtship. Journal McQueen, A., & Klein, W. (2006). Experimental manipulations of Personality and Social Psychology, 66, 857-869. of self-affirmation: A systematic review. Self and Identity, 5, Swann, W. B., Jr., & Hill, C. A (1982). When our identities are 289-354. mistaken: Reaffirming self-conceptions through social interac- Miyake, A., Kost-Smith, L. K., Finkelstein, N. D., Pollock, S. J., tion. Journal of Personality and Social Psychology, 43, 59-66. Cohen, G. L., & Ito, T. A. (2010). Reducing the gender achieve- Swann, W. B., Jr., & Read, S. J. (1981). Self-verification processes: ment gap in college science: A classroom study of values affir- How we sustain our self-conceptions. Journal of Experimental mation. Science, 26, 1234-1237. Social Psychology, 17, 351-372. Napper, L., Harris, P. R., & Epton, T. (2009). Developing and test- Swann, W. B., Jr., Rentfrow, P. J., & Guinn, J. (2003). Self- ing a self-affirmation manipulation. Self and Identity, 8, 45-62. verification: The search for coherence. In M. Leary & J. Tagney Pew Research Center. (2010). Digital differences. Retrieved from (Eds.), Handbook of self and identity (pp. 367-383). New York, http://pewinternet.org/~/media//Files/Reports/2012/PIP_Digi- NY: Guilford. tal_differences_041312.pdf Tesser, A. (1988). Toward a self-evaluation maintenance model of Rosenbaum, J. E., Johnson, B. K., Stepman, P. A., & Nuijten, K. C. M. social behavior. In L. Berkowitz (Ed.), Advances in experimen- (2010, April). Just being themselves? Goals and strategies for tal social psychology (Vol. 21, pp. 181-227). San Diego, CA: self-presentation on Facebook. Paper presented at the Southern Academic Press. States Communication Association. Memphis, TN. Tesser, A. (2000). On the confluence of self-esteem maintenance Rosenberg, M. (1965). Society and the adolescent self-image. mechanisms. Personality and Social Psychology Review, 4, Princeton, NJ: Princeton University Press. 290-299. Rudman, L. A., Dohn, M. C., & Fairchild, K. (2007). Implicit self- Tesser, A., & Cornell, D. P. (1991). On the confluence of self- esteem compensation: Automatic threat defense. Journal of processes. Journal of Experimental Social Psychology, 27, Personality and Social Psychology, 93, 798-813. 501-526. Toma and Hancock 331 Tesser, A., Crepaz, N., Collins, J. C., Cornell, D., & Beach, S. R. H. Waksiak, C., & Trope, Y. (2009). Cognitive consequences of (2000). Confluence of self defense mechanisms: On integrating affirming the self: The relationship between self-affirmation the self-zoo. Personality and Social Psychology Bulletin, 26, and object construal. Journal of Experimental Social Psy- 1476-1489. chology, 45, 92-932. Toma, C. L., & Carlson, C. L. (2012, May). I’m so much cooler Walther, J. B. (2007). Selective self-presentation in computer- online: An examination of Facebook self-presentation. Paper mediated communication: Hyperpersonal dimensions of presented at the International Communication Association technology, language, and cognition. Computers in Human (ICA) Convention, Phoenix, AZ. Behavior, 23, 2538-2557. Toma, C. L., Hancock, J. T., & Ellison, N. B. (2008). Separating Walther, J. B., & Parks, M. R. (2002). Cues filtered out, cues filtered fact from fiction: Deceptive self-presentation in online dating in: Computer-mediated communication and relationships. In profiles. Personality and Social Psychology Bulletin, 38, 1023- M. L. Knapp & J. A. Daly (Eds.), Handbook of interper- 1036. sonal communication (3rd ed., pp. 529-563). Thousand Trepte, S. (2005). Daily talk as self-realization: An empirical study Oaks, CA: SAGE. on participation in daily talk shows. Media Psychology, 7, 165- Wicklund, R. A., & Gollwitzer, P. M. (1981). Symbolic self-com- 189. pletion, attempted influence, and self-deprecation. Basic and Tufekci, Z. (2008). Can you see me now? Audience and disclosure Applied Social Psychology, 2, 89-114. regulation in online social network sites. Bulletin of Science, Wilson, R. E., Gosling, S. D., & Graham, L. T. (2012). A review Technology & Society, 28, 20-36. of Facebook research in the social sciences. Perspectives on Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital Psychological Science, 7, 203-220. in a social network site? Facebook use, and college students’ Young, A. L., & Quan-Haase, A. (2009). Information revela- life satisfaction, trust, and participation. Journal of Computer- tion and internet privacy concerns on social network sites: Mediated Communication, 14, 875-901. A case study of Facebook. In C&T ’09: Proceedings of the Valkenburg, P. M., Peter, J., & Schouten, A. P. (2006). Friend Fourth International Conference on Communities and Tech- networking sites and their relationship to adolescents’ well- nologies. New York, NY: ACM Press. being and social self-esteem. CyberPsychology & Behavior, Zillmann, D. (1988). Mood management through communication 9, 584-590. choices. American Behavioral Scientist, 31, 327-340.