Ambivalent Facebook users: Anxious attachment style and goal cognition

Abstract

Drawing on attachment theory, this study shows that people with greater anxious attachment use Facebook (FB) to maintain their social relationships but feel ambivalent in the process. Integrating attachment theory and goal cognition theory, this study explicates the psychological process that drives anxious people’s pursuit of social goals though their FB use. Specifically, it identifies two social goals that motivate anxious people’s FB use (to belong and to be well-liked) and explores the psychological process that underlies their goal pursuit. Four specific elements of goal cognition function in the pursuit of these goals: A motive to belong/to be well-liked (directive function) triggers social vigilance/self-vigilance on FB (regulatory function), leading to a fear of being excluded/social jealousy (control function) and then quiescent–agitated/happy–dejected ambivalent feelings (arousal function).

Keywords

Ambivalent anxious attachment attachment theory Facebook goal cognition

Some prior research explores individual differences (Oldmeadow, Quinn, & Kowert, 2013) and identifies goals that drive people’s use of Facebook (FB) (Nadkarni & Hofmann, 2012), but no studies explore the psychological processes that underlie how people with distinct attachment styles pursue social goals on FB. Accordingly, this article seeks to integrate attachment theory and goal cognition literature, proposes a model to explicate the psychological process that drives people with an anxious form of attachment to pursue social goals using FB, and thereby helps consolidate diverse findings in extant literature pertaining to people’s relationship-building approaches and FB behaviors, within a unified, overarching theoretical framework.

To establish a foundation for this approach, the current study defines people with an anxious attachment style as those who exhibit relational ambivalence (Mikulincer, Shaver, Bar-On, & Ein-Dor, 2010). On the one hand, they want to maintain social relationships and approach others for support (Mikulincer et al., 2010), but on the other hand, they fear rejection because they have a strong need for social approval (Feeney, Noller, & Hanrah, 1994; Hazan & Shaver, 1987; Mikulincer & Shaver, 2007). Therefore, people with a greater anxious attachment orientation likely are ambivalent FB users. Even if they rely on FB to maintain their friendships, their ambivalence toward relationships may carry forward and manifest as ambivalence toward relationship building on FB.

Many people use FB to achieve social goals (e.g., Chang, 2015; Nadkarni & Hofmann, 2012). For those with anxious attachment orientations who feel ambivalent about their use of FB, the goal pursuit process they undergo on social media represents an interesting conundrum. Therefore, this research draws on goal cognition theory to clarify what might motivate anxious attachment people to use FB and what their underlying goal pursuit cognition process looks like. Goal cognition theory suggests that goal pursuit processes depend on people’s mental models and process representation, and four functions appear in the intended journey from goal intention to goal attainment: directive, regulatory, control, and arousal (Karoly, 1993, 1999; Karoly & Ruehlman, 1995, 1996). The elements that underlie these functions also vary across different goal pursuits and among people with different orientations, such as anxious attachment.

Accordingly, this study specifies the needs to belong and to be well-liked as two central social goals for anxious people’s use of FB. The proposed model then predicts specific goal cognition functions that should be associated with the pursuits of both goals (Figure 1). First, a motive to belong (directive function) should trigger social vigilance on FB (regulatory function), leading to a fear of exclusion (control function) and ultimately resulting in ambivalent feelings, such that the user feels quiescent and agitated simultaneously (arousal function). Second, another goal cognition process begins with a motive to be well-liked (directive function), which triggers self-vigilance on FB (regulatory function), leading to social jealousy (control function) and ultimately results in ambivalent feelings about being happy and dejected concurrently (arousal function). By focusing on people with an anxious attachment style as a relevant example, this study advances understanding of how FB usage might be driven by social goals and the distinct psychological processes that underlie this goal pursuit across different users.

Figure 1.

Proposed goal cognition process for pursuing goals among anxious people.

Attachment theory

Attachment theory proposes that people develop mental representations of their self and others through their interactions with attachment figures. The mental representations in turn affect their cognition, emotion, and behavior in social relationships (Bowlby, 1969, 1973, 1980; Hazan & Shaver, 1994; Mikulincer & Shaver, 2007). Two orthogonal dimensions of attachment, anxiety and avoidance, account for individual differences in attachment styles or orientations (Brennan, Clark, & Shaver, 1998). The anxiety dimension implies concern that important others are not responsive; the avoidance dimension describes a person’s desire to avoid closeness.

A key tenet of the theory is that the patterns created by parent–child interactions have implications for the quality of a person’s other social attachments, because the behavioral styles shaped by interactions with primary caretakers carry forward (Bowlby, 1988; Mikulincer & Shaver, 2007). Early attachment styles shape people’s mental working models and expectations of subsequent romantic and peer relationships (Furman, Simon, Shaffer, & Bouchey, 2002). Hazan and Shaver (1994) argue that attachment theory thus can serve as a generative theoretical framework for deriving predictions and constructing theories about relationships.

Attachment style and FB uses

Attachment theory has been applied previously to explain FB behaviors. For example, an anxious attachment style, but not an avoidant one, is a positive predictor of relationship initiation on FB (Yaakobi & Goldenberg, 2014), the degree of intimate feelings with others while using FB (Nitzburg & Farber, 2013), and the degree of partner surveillance on FB (Fox, Osborn, & Warber, 2014; Marshall, Bejanyan, Di Castro, & Lee, 2013). For people with anxious attachments, FB offers an important platform for maintaining or initiating friendships. However, the motives that drive anxious attachment people to use FB and the goal-oriented cognition related to this process have not drawn much research attention. Focusing on anxious attachment users, this article integrates goal cognition and attachment literature to propose a theoretical framework that details which social goals drive them to use FB and the goal-related cognition they generate in this pursuit.

Attachment style and the quantity and quality of FB friendships

People with an anxious attachment style seek companionship (Saferstein, Neimeyer, & Hagans, 2005) and desire closeness with partners (Feeney & Collins, 2001). Likely because they try to achieve security by devoting extra mental energy and physical efforts to keeping others engaged (Hazan & Shaver, 1994), they generally spend more time using FB (Oldmeadow et al., 2013; Yaakobi & Goldenberg, 2014), exhibit greater engagement with FB (Jenkins-Guarnieri, Wright, & Hudiburgh, 2012), and feel more intimate with others while using FB (Nitzburg & Farber, 2013). People with an avoidant attachment style instead fear and avoid social closeness and seek to remain self-reliant (Feeney & Collins, 2001). They hold negative perceptions of friendship, do not regard friendship as important (Feeney et al., 1994), and find less companionship among friends (Saferstein et al., 2005). Because FB is a platform to maintain friendships, people with a greater anxious attachment style should be more likely to use FB and have more friends on the site. In contrast, people with a greater avoidant attachment style should exhibit the opposite patterns.

H1: People’s degree of anxious attachment is positively associated with (a) their time spent on FB and (b) their number of friends on FB, whereas their degree of avoidant attachment is negatively associated with (a) their time spent on FB and (b) their number of friends on FB.

People with an anxious attachment style also perceive more personal conflict with friends (Saferstein et al., 2005), so they find less satisfaction in their friendships (Welch & Houser, 2010). Likely because anxious attachment is negatively associated with perceived support, acceptance, and trust from close friends, it does not lead to friendship intimacy (You & Malley-Morrison, 2000). Even if people with anxious attachment spend more time on FB, they might not perceive better friendship quality or believe that they have more good friends on FB, as indicators of friendship quality (Nangle, Erdley, Newman, Mason, & Carpenter, 2003).

H2: People’s degree of anxious attachment is not associated with the number of perceived good friends on FB, whereas their degree of avoidant attachment is negatively associated with the number of perceived good friends on FB.

These predictions suggest that people with an anxious attachment style rely on FB to build and maintain social relationships, but they do not believe they have good quality friendships. Therefore, they constitute ambivalent FB users, who want to use it but are uncertain about the support they can derive from it. Greater understanding of these ambivalent users might be facilitated through the application of goal cognition theory, because if attachment with caretakers shapes the working models that guide the social interactions of people with anxious attachment styles, they also might guide these users’ goal cognition.

Goal cognition

The use of FB is driven by goals (Chang, 2015; Nadkarni & Hofmann, 2012). Goal pursuit depends on mental models and process representations, which are functional elements of the intended journey from intention to goal attainment or goal cognition (Karoly, 1993). Functional, interactive, coordinated elements guide goal-directed behaviors, pertaining to directive, regulatory, control, and arousal functions (Karoly & Ruehlman, 1995, 1996) (Figure 1).

Directive functional elements

In goal cognition literature, the directive functional element specifies intended outcomes or end states of the goal pursuit (Karoly, 1999; Karoly & Ruehlman, 1995, 1996). Chang (2015) identifies belonging and being well-liked as two important social goals related to the use of FB. The working models of people with anxious attachment thus might determine the degree to which they are motivated by these two key social goals.

To belong

The need to belong is a fundamental human motivation, related to desires to “form and maintain interpersonal bonds” (Baumeister & Leary, 1995, p. 497). This fundamental human motivation in turn can alter psychological orientations and trigger goal-oriented behaviors (Baumeister & Leary, 1995). According to Gangadharbatla (2008), the need to belong is associated with more positive attitudes toward FB and greater willingness to use FB. Because people with an anxious attachment style feel insecure, they attempt to keep others close by and engaged (Hazan & Shaver, 1994). Accordingly, they likely develop stronger belonging-oriented social goals that drive their FB use.

H3a: People with greater anxious attachment have stronger goals to belong.

To be well-liked

People also use FB because they seek popularity and want to be well-liked by others (Chang, 2015). Exploring FB behaviors, Chang (2015) shows that to achieve such social goals, people respond to others (e.g., comment on others’ posts and photos) but also reveal their own thoughts and interests (e.g., post activities and opinions). Because anxious attachment people are occupied with relationships, desire to be loved and accepted by others, and seek constant confirmation of their links to close others or reassurance that they are lovable (Evraire & Dozois, 2014; Hazan & Shaver, 1994; Mikulincer & Shaver, 2007), they should develop a social goal to be well-liked that also drives their FB use.

H3b: People with greater anxious attachment have stronger goals to be well-liked.

Regulatory functional elements

According to goal cognition literature, people constantly evaluate their goal progress; the regulatory function involves assessing the discrepancy between a desired goal and current status, such that it requires constant monitoring of goal progress and comparisons with similar others in terms of this progress (Karoly, 1993, 1999; Karoly & Ruehlman, 1995, 1996). Working models of anxious attachment people dictate their goal cognition by guiding them to engage in greater degrees of vigilance in their goal pursuit. Depending on the focus of this attention, high vigilance could manifest as vigilance toward either the social environment or the self.

Social vigilance

Social vigilance refers to monitoring the social environment; on FB, it reflects the degree to which people pay attention to what others are discussing and posting on FB. Because of their history of being treated with inconsistent responsiveness, people with anxious attachment styles have a hyperactivated system, associated with an obsessive preoccupation with others and hypervigilant attention to relationship partners (Hazan & Shaver, 1994; Mikulincer & Shaver, 2005; Mikulincer, Shaver, & Pereg, 2003). In studies of romantic relationships, prior research shows that an anxious attachment positively, but an avoidant attachment negatively, predicts the surveillance of partners’ FB pages (Fox et al., 2014; Marshall et al., 2013). If they use FB to connect with others, people with greater anxious attachment might exhibit greater social vigilance in their goal pursuit process.

H4a: People with greater anxious attachment exhibit higher levels of social vigilance in their goal pursuit to belong.

Self-vigilance

Self-vigilance entails monitoring how others respond to the self. In an FB use context, self-vigilance is the degree to which people pay attention to how well their posts are responded to and received by friends on the site. Because of their fear of being abandoned and neglected, people with an anxious attachment style exhibit heightened vigilance in detecting cues of rejection, disapproval, or waning interest from others (Fraley, Niedenthal, Marks, Brumbaugh, & Vicary, 2006; Mikulincer et al., 2003; Shaver & Mikulincer, 2002). They also exhibit greater evaluation concerns about how they appear to others (Oldmeadow et al., 2013). Thus, if they use FB with a goal to be well-liked, people with a greater anxious attachment style likely express more self-vigilance in their goal pursuit process.

H4b: People with greater anxious attachment exhibit higher levels of self-vigilance in their goal pursuit of being well-liked.

Control functional elements

The control function refers to control-oriented thinking, associated with self-reward or self-criticism (Karoly & Ruehlman, 1995), based on analyses of what has gone well or gone wrong. The purpose of control-oriented thinking is to redirect efforts strategically toward a goal (Karoly & Ruehlman, 1996). Attachment theory also suggests that anxious people develop intensified appraisals of threats, directing their attention to thoughts about their personal weaknesses or failures (Mikulincer & Shaver, 2007), and they tend to be subject to fear and extreme jealousy in their relationships (Collins & Read, 1990; Hazan & Shaver, 1987).

Fear of being excluded

In a social context, such as on FB, fear likely derives from perceptions of being excluded. A fear of being excluded implies that people are worried they will be neglected by or disconnected from their social circle, which is a stressful situation that most people can readily identify (Kerr & Levine, 2008). In an FB use context, this fear implies people worry that, if they do not post, comment, or upload photos, they will be alienated from their friends on FB. People with an anxious attachment style seek validation through excessive closeness in their personal relationships (Griffin & Bartholomew, 1994), but they simultaneously fear and anticipate rejection (Mikulincer & Shaver, 2007; Scharf, Oshri, Eshkol, & Pilowsky, 2014) and worry that their partners cannot be relied on or will not continue to be responsive (Collins & Read, 1990). Therefore, when using FB to connect with others, they likely experience some fear of being excluded in their goal pursuit process.

H5a: People with greater anxious attachment experience more fear of being excluded in their goal pursuit to belong.

Jealousy

People with anxious attachment are sensitive to social cues and tend to interpret them in negative ways (Meyer, Pilkonis, & Beevers, 2004). As a result, they likely believe that they fall behind in their pursuit of being well-liked and may experience social comparison jealousy, defined as “feelings, thoughts, and behaviors that occur when another person enjoys more success and consequently the tangible and intangible advantages, attainments, or possessions of the other person provoke dissatisfaction, resentment, and sometimes behaviors like backbiting” (Bers & Rodin, 1984, p. 767). Compared with people with secure or avoidant attachment styles, people with an anxious attachment style experience more jealousy in their romantic relationships and engage in more jealous acts (Collins & Read, 1990; Hazan & Shaver, 1987). Despite a lack of research into social comparison jealousy, the current study predicts that the sense of insecurity maintained by anxious people may arouse greater jealousy when they assess what goes wrong in their pursuit of the goal of being well-liked. Therefore, the current study predicts that when they use FB to be well-liked by others, anxious people experience social jealousy in their goal pursuit process.

H5b: People with greater anxious attachment experience more social jealousy in their goal pursuit of being well-liked.

Arousal function

The arousal function refers to the emotional sequence of goal pursuit, which can be positive or negative (Karoly & Ruehlman, 1995, 1996). A discrepancy between the desired state and the current state prompts negative arousal; approaching the desired state instead triggers positive arousal. Attachment styles determine people’s affect regulation processes and emotional experiences in relationships (Mikulincer & Shaver, 2005). People with an anxious attachment style oscillate between positive and negative appraisals of their attachment experiences (Mikulincer & Shaver, 2007) and experience extreme emotions in their close relationships, feeling as much pain as joy (Hazan & Shaver, 1987). This ambivalent emotional state likely derives from anxious people’s enhanced desire for support but concomitantly intensified appraisals of threats (Mikulincer & Shaver, 2005).

Quiescent–agitated torn-between feeling

Regulatory focus theory proposes that attaining goals generates different feelings than does failing to approach goals (Brendl & Higgins, 1996). When people achieve a prevention goal, they feel quiescent; if they deviate from this goal though, they feel agitated (Higgins, 1997). The need to belong represents a passive, prevention-oriented social goal, such that people use FB because others are using it, and they do not want to be excluded socially (Chang, 2015). If people with an anxious attachment style achieve the goal to belong on FB, which is a prevention goal, they thus should feel quiescent. However, if they fail to achieve this prevention-oriented goal to belong, they should feel agitated. If they achieve this goal sometimes and fail to achieve it other times, they experience as many quiescent feelings as agitated feelings. In other words, they experience quiescent–agitated ambivalent feelings in their pursuit of the goal to belong.

H6a: People with greater anxious attachment experience more quiescent–agitated ambivalent feelings.

Happy–dejected torn-between feelings

According to regulatory focus theory, when people approach a promotion goal, they feel happy, and when they deviate from this goal, they feel dejected (Higgins, 1997). Being well-liked is an active, promotion-oriented social goal, and people accordingly actively build relationships on FB to enhance their self-worth and feel liked by others (Chang, 2015). If people with an anxious attachment style achieve the promotion-oriented goal of being well-liked on FB, they feel happy; if they fail to achieve this goal, they feel dejected. If they can sometimes achieve the goal and sometimes not, they experience as many happy feelings as dejected feelings, so they experience happy–dejected ambivalent feelings in their pursuit of the goal of being well-liked.

H6b: People with greater anxious attachment experience more happy–dejected ambivalent feelings.

Karoly (1999) suggests that in goal pursuit processes, people constantly compare their progress against the progress of others, to evaluate their performance (regulatory component). Such evaluations serve as feedback for the control and arousal components. For example, when the comparison reveals a large discrepancy, people engage in self-criticism (control-related thinking), which determines the degree to which they become agitated (arousal-related response). Therefore, a causal process is implied, and goal pursuit components may emerge as serial processes:

H7a: The relationship between anxious attachment and quiescent–agitated ambivalent feelings is mediated by a motive to belong, social vigilance on FB, and a fear of being excluded.

H7b: The relationship between anxious attachment and happy–dejected ambivalent feelings is mediated by a motive to be well-liked, self-vigilance on FB, and social jealousy.

Survey

Participants and procedures

College students at three universities in Taiwan—National Chengchi University (NCCU), National Chiao Tung University (NCTU), and National Chung Cheng University (CCU)—received solicitations via e-mail, containing a request to participate in an online survey in return for the chance to win a gift drawing. NCCU, NCTU and CCU are located in the northern, central, and southern regions of Taiwan, respectively. When participants clicked on the provided link, it took them to an online survey powered by SurveyMonkey^®. The survey remained accessible for 4 weeks,¹ during which students in the schools’ e-mail databases received three reminder e-mails. The survey contained various questions about Internet use and took less than 10 min to complete. In total, 1,053 respondents completed the survey, distributed as follows: 650 from NCCU (63.69% women), 243 from NCTU (45.27%), and 160 from CCU (61.62%). The same IP address could be used to answer the survey only once, to avoid repeated responses. The average age of the respondents was 21.94 years (SD = 3.67, ranging from 18 years to 61 years), and 40.90% of them were men.

Measurements

The question items—except as noted for the time spent online and on FB and the number of friends and good friends on FB—used 5-point Likert-type scales. When scales were adopted from extant literature, a translation and back-translation procedure suggested by Brislin (1987) served to create the measures in Chinese.

Time spent on the Internet and on FB

Time spent online may determine time spent on FB, so it is important to include it as a covariate in the analyses. The questionnaire items tapped the time respondents spent on the Internet and FB, over both weekdays and weekends, in the past month. Respondents indicated both days spent on the Internet and FB (“How many days a week do you use the internet/FB during the weekdays/weekends?”), as well as the specific amount of time spent (“On a typical weekday/weekend when you use internet/FB, how much time do you spend on it?”). The average minutes per day were calculated from these responses. Time spent on FB is defined as active usage, not staying logged in to FB without using it.

Number of friends and good friends on FB

Participants indicated the number of friends they had on FB, as well as the number of good friends they had on FB. Good friends were defined as those who understand and support users and to whom they can pour out their hearts.

Attachment style

The participants rated items derived from Hazan and Shaver’s (1987) descriptors of three attachment styles: secure attachment scale (e.g., “I find it relatively easy to get close to others,” Cronbach’s α = .70), avoidant attachment scale (e.g., “I find it difficult to allow myself to depend on others,” Cronbach’s α = .81), and anxious attachment scale (e.g., “I find others are reluctant to get as close as I would like,” Cronbach’s α = .73). Similar to Simpson, Rholes, and Nelligan’s (1992) approach, the responses to the secure items were reversed and averaged with avoidant attachment items to form an avoidant attachment rating (Cronbach’s α = .81), because secure and avoidant attachment ratings correlate negatively and significantly (Pearson’s r = −.45, p < .01). The resulting two dimensions, anxious attachment and avoidant attachment, are not significantly correlated (Pearson’s r = −.02, p = .60), confirming that they are orthogonal constructs.

Goal to belong

Leary, Kelly, Cottrell, and Schreindorfer’s (2005) scale provides the assessment of participants’ motivation to belong, with 10 items (e.g., “I have a strong need to belong,” Cronbach’s α = .80).

Goal to be well-liked

This study adopted Santor, Messervey, and Kusumakar’s (2000) scale to assess people’s motivation to be well-liked with 12 items (e.g., “It is important that people think that I am popular,” Cronbach’s α = .91).

Social vigilance

Respondents rated the attention they paid to others on FB, to assess their social vigilance on FB, using 3 items developed for the purpose of this study (e.g., “I pay attention to what my friends talk about on Facebook,” Cronbach’s α = .86).

Self-vigilance

Respondents rated the attention they paid to how well they are received on FB, to assess their self-vigilance on FB. The scale features 3 items developed for the purpose of this study (e.g., “I pay attention to how many people ‘like’ my comments,” Cronbach’s α = .92)

Fear of being excluded

Respondents rated the degree to which they were worried about being left out if they did not use FB, using 4 items developed for the purpose of this study (e.g., “I’m afraid that if I don’t post things or leave comments on Facebook, people would distance themselves from me,” Cronbach’s α = .96)

Social jealousy

Respondents rated the degree to which they are jealous of others on FB, using 4 items developed for the purpose of this study (e.g., “I feel jealous of those of my friends who get lots of ‘likes’ on their comments,” Cronbach’s α = .97).

Quiescent–agitated ambivalence

Participants rated the degree to which they feel quiescent if they can connect with others on FB, using 3 items developed for this study (e.g., “I feel more relaxed when I can get on Facebook and find out what others are talking about,” Cronbach’s α = .92). They also rated the degree to which they feel agitated if they fail to connect with others on FB, using 3 further items developed for this study (e.g., “If I don’t know what others are talking about because I haven’t used Facebook in a while, I’d feel uneasy,” Cronbach’s α = .95). To assess the degree to which both feelings are strong, Griffin’s formula, (P + N)/2–/P–N/ (Thompson, Zanna, & Griffin, 1995), captures the extent of ambivalence (i.e., feeling torn between being quiescent and being agitated). In this formula, P represents positive feelings (e.g., quiescent), and N represents negative feelings (e.g., agitated). The first element of the formula, (P + N)/2, represents the intensity of both feelings, and the second element, /P–N/, indicates the similarity between positive and negative feelings. Greater scores indicate stronger experiences of both positive and negative feelings, which in this case suggest ambivalent feelings between being quiescent and being agitated.

Happy–dejected ambivalence

Participants rated the degree to which they feel happy if they are well received on FB, using 3 items developed for this study (e.g., “I feel happy when I see other people ‘like’ my comments,” Cronbach’s α = .94). They also rated the degree to which they feel dejected if they fail to be well received on FB, using another 3 items developed for this study (e.g., “If people don’t ‘like’ my posts on Facebook, I’d feel somewhat down,” Cronbach’s α = .92). In this application of Griffin’s formula (Thompson et al., 1995), the positive feelings (P) indicate being happy, and the negative feelings (N) refer to being dejected. Again, higher scores indicate more ambivalent feelings between being happy and being dejected.

Hypothesis tests

The tests for H1 and H2 relied on hierarchical multiple regression analyses, with gender and age entered in the first step, time spent on the Internet in the second step, and the two attachment orientations in the third step. The regression analysis (Model 2a in Table 1) reveals that avoidant attachment is a negative predictor, but anxious attachment is a positive predictor, without reaching significance. Some students access FB on their mobile phones, such that FB is always on, and they check the site as soon as a notification alerts them to new messages. These users may refer to all this time as time spent on FB, such that they would overstate the amount of time they actually spend on FB—despite the explicit instruction that, for the survey, the measure sought to identify time spent checking and posting on the site, not simply being logged in to it. Therefore, this study applies Hoaglin and Iglewicz’s (1987) outlier labeling rule procedures to identify outliers of time spent on FB. Another regression analysis, conducted after excluding 85 outliers who spent more than 540 min daily on FB (Model 2b in Table 1), revealed that the avoidant orientation remained a negative predictor. Anxious orientation emerged as a significant positive predictor. Therefore, H1a was not supported by the full data set but was supported after removing the outliers for time spent on FB.

Table 1.

Regression analysis results for H1a, H1b, and H2.

	Model 1			Model 2a (H1a)			Model 2b (H1a)			Model 3 (H1b)			Model 4 (H2)
Dependent variable	Time spent online			Time spent on FB (full sample)			Time spent on FB (no outliers)			Number of friends			Number of good friends
Predictors	B	SE β	β	B	SE β	β	B	SE β	β	B	SE β	β	B	SE β	β
Step 1 ▵R²	.01			.01			.01*			.01*			.01**
Gender (male = 1; female = 0)	6.43	14.21	.01	−22.11	12.47	−.06	−9.66	6.90	−.05	19.52	22.82	.03	11.81	6.96	.05
Age	3.36	1.90	.05	−2.22	1.67	−.04	−2.05	.90	−.07*	−8.27	3.06	−.08**	−2.59	.93	−.09**
Step 2 ▵R²	NA			.23**			.11**			.04**			.02**
Time online	NA	NA	NA	.43	.02	.48**	.18	.02	.34**	.01	.06	.01	−.02	.02	−.04
Time on FB	NA	NA	NA	NA	NA	NA	NA	NA	NA	.38	.06	.21**	.08	.02	.14**
Step 3 ▵R²	.01			.02**			.02**			.02**			.02**
Anxious attachment	13.43	9.08	.05	12.15	6.92	.05	14.29	4.21	.10**	41.59	14.16	.09**	3.39	4.39	.02
Avoidant attachment	7.92	11.43	.02	−37.97	8.71	−.12**	−16.54	5.23	−.10**	−71.73	17.95	−.12**	−26.03	5.56	−.14**
Total R²		.01			.25**			.14**			.07**			.05**

Note. FB = Facebook; SE = standard error.

*p ≦ .05, **p ≦ .01.

A regression analysis shows that anxious attachment relates positively, but avoidant attachment relates negatively, to the number of friends on FB, in support of H1b (Model 3 in Table 1). In another regression analysis, as predicted, anxious attachment did not predict the number of good friends on FB, but avoidant attachment did so, in a negative direction. These findings confirm H2 (Model 4 in Table 1).

The tests of H3–H7 leverage the PROCESS macro (Model 6; Hayes, 2013), with 10,000 bootstrap resamples; age, gender, time spent online, and time spent on FB serve as covariates. In the first analysis, the independent variable is anxious attachment. In Figure 2, Panel a, anxious attachment significantly predicts the goal to belong, in support of H3a, and social vigilance, in support of H4a. It is positively associated with a fear of being excluded, in line with H5a, and with quiescent–agitated ambivalence, confirming H6a. As expected, the total indirect effect is significant, with a point estimate of .0099 and a 95% confidence interval of [.0045, .0186]. The resulting mediation process contains three mediators, in line with H7a.

Figure 2.

Proposed goal cognition process for pursuing the goal to belong. Panel (a): Anxious attachment as the independent variable. Note. The bootstrapped 95% bias-corrected confidence intervals using 10,000 bootstrap samples are as follows: TB as a single mediator (AA → TB → QA) [−.0573, .0977]; TB and SV as two serial mediators (AA → TB → SV → QA) [.0292, .0718]; and TB, SV, and FE as three serial mediators (AA → TB → SV →FE → QA) [.0045, .0186]. Panel (b): Avoidant attachment as the independent variable. Note. The bootstrapped 95% bias-corrected confidence intervals using 10,000 bootstrap samples are as follows: TB as a single mediator (VA → TB → QA) [−.0598, −.0063]; TB and SV as two serial mediators (VA → TB → SV → QA) [−.0385, −.0136]; and TB, SV, and FE as three serial mediators (VA → TB → SV →FE → QA) [−.0100, −.0024].

As Figure 2, Panel b, indicates, when avoidant attachment is the independent variable, its effect on the goal to belong is negative and significant, but its effects on the two mediators are not significant. Moreover, it lacks any direct effect on quiescent–agitated ambivalence, whereas serial mediation appears, with a point estimate of −.0054 and a 95% confidence interval of [−.0100, −.0024]. In other words, people with an avoidant attachment are less likely to have a goal to belong and, therefore, are less likely to experience quiescent–agitated feelings when they are able or not able to feel as if they belong due to their use of FB.

In Figure 3, Panel a, anxious attachment significantly predicts the goal to be well-liked, in support of H3b, and self-vigilance, in support of H4b. It is positively associated with jealousy, in line with H5b, and with happy–dejected ambivalence, confirming H6b. Consistent with expectations, the total indirect effect again is significant, with a point estimate of .0118 and a 95% confidence interval of [.0073, .0183], such that it suggests another three-mediator mediation process, as predicted by H7b.

Figure 3.

Proposed goal cognition process for pursuing the goal to be well-liked. Panel (a): Anxious attachment as the independent variable. Note. The bootstrapped 95% bias-corrected confidence intervals using 10,000 bootstrap samples are as follows: TL as a single mediator (AA → TL → HA) [.0003, .0664]; TL and SI as two serial mediators (AA → TL → SI → HA) [.0417, .0881]; and TL, SI, and SJ as three serial mediators (AA → TL → SI →SJ → HA) [.0073, .0183]. Panel (b): Avoidant attachment as the independent variable. Note. The bootstrapped 95% bias-corrected confidence intervals using 10,000 bootstrap samples are as follows: TL as a single mediator (VA → TL → HA) [−.0321, −.0025]; TL and SI as two serial mediators (VA → TL → SI → HA) [−.0435, −.0053]; and TL, SI, and SJ as three serial mediators (VA → TL → SI →SJ → HA) [−.0107, −.0013].

As Figure 3, Panel b, shows, when avoidant attachment is the independent variable, its effect on the goal to be well-liked is negative and significant, but its effects on the two mediators and the dependent variable, happy–dejected ambivalence, are not. Serial mediation appears with a point estimate of −.0052 and a 95% confidence interval of [−.0107, −.0013]. The findings suggest that people with avoidant attachment are less likely to have a goal to be well-liked, so they are less likely to experience happy–dejected feelings, whether they are able or not able to be well-liked on FB.

Discussion

Findings and contributions

This article contributes to extant literature in several important ways. First, it integrates goal cognition and attachment theory to explicate how anxious attachment people use FB to pursue their social goals, as well as to illustrate their goal-oriented thinking in that process. Second, the proposed theoretical framework provides an overarching conceptualization to help make sense of diverse findings in extant literature, including the positive effect of anxious attachment on FB jealousy and FB surveillance in close relationships (Fox et al., 2014; Marshall et al., 2013). Third, the integrated framework tested in this study refers to an FB use context and friendship, but it also offers the potential to explore relationships in other contexts (e.g., face-to-face) that involve other important partners (e.g., romantic partners, families).

In detail, by integrating attachment theory and goal cognition literature, this study explicates the psychological process that drives anxious people’s pursuit of social goals though their FB use. Prior FB research has explored individual differences and identified goals that drive FB use (Nadkarni & Hofmann, 2012), but no studies explore the psychological processes underlying this goal pursuit. Some people use FB to interact with others but do not derive the expected support from such interactions; in accordance with attachment theory, this study identifies ambivalent users as those with an anxious attachment style. It further illustrates how goal cognition literature can inform and depict unique goal pursuit processes.

Documented differences in time spent on FB and friendship quantity and quality reflect the different social orientations of those with varying attachment orientations. Anxious users maintain friendships through FB, and their anxious attachment relates positively to the time spent and number of friends on FB. However, anxious orientation is not associated with better perceived friendship quality, as indicated by the number of good friends on FB. An avoidant attachment is negatively and significantly associated with time spent on FB, number of FB friends, and number of good friends. These patterns are consistent with research that explores social relationships offline (Saferstein et al., 2005), suggesting that attachment theory offers a useful theoretical framework for exploring online social relationships.

A key assumption of goal-related psychology is that people’s behaviors are driven by motivationally salient ends (Karoly, 1993). As this study shows, anxious attachment people’s use of FB is driven by two social goals: to belong and to be well-liked. The serial mediation analyses demonstrate that anxious attachment activates the goal to belong, which triggers social vigilance, fears of being excluded, and ultimately quiescent–agitated ambivalent feelings. This anxious attachment style also activates the goal to be well-liked, leading to self-vigilance, then social jealousy, and finally happy–dejected ambivalent feelings. In contrast, people with a greater avoidant attachment are less likely to pursue these two social goals, and they do not engage in regulatory- and control-oriented thinking or experience goal-related arousal.

The findings echo Mikulincer, Shaver, Bar-On, and Ein-Dor’s (2010) suggestion that people with an anxious orientation feel ambivalent toward relationships. For example, anxiously attached people express ambivalence toward closeness and hold simultaneously positive and negative views of their partners. In the current study, people with a stronger anxious orientation experience more positive emotion outcomes (quiescent or happy) when achieving their social goals, as well as more negative emotion outcomes (agitated or dejected) when failing to achieve them. In other words, their FB use provides both positive and negative emotional experiences.

Research directions

This article focuses on two social goals of a unique group of people—those with an anxious attachment style who use FB but feel ambivalent about it—and portrays the psychological process underlying their goal pursuit. Because goal cognition likely differs for people with distinct characteristics, further research should explore the goals that motivate people who exhibit other personalities or orientations to use FB, as well as their goal cognition in this process.

Goal cognition involves self- and affect regulation processes, as demonstrated by the regulatory and arousal functional components (Karoly, 1993, 1999; Karoly & Ruehlman, 1995, 1996). Observing FB behaviors, this study identifies social vigilance and self-vigilance as regulatory functional components. However, other important mechanisms also might be involved when people monitor and regulate their goal pursuits. In light of the documented characteristics of FB users, this study explicitly sought to examine ambivalent emotional experiences during the affect regulation processes. However, other emotional experiences also may be worthy of investigation when considering different types of FB users or different goals.

Although this study builds on goal cognition theory (Karoly, 1993, 1999; Karoly & Ruehlman, 1995, 1996), the elements underlying the four components (directive, regulatory, control, and arousal) vary for different goal pursuits or for people with different orientations. The focal elements reflect these variations. Yet Karoly and Ruehlman (1995) also identify some common elements for each function: self-efficacy and attribution for the directive function; social monitoring and comparison for the regulatory function; planning, self-reward, and self-criticism for the control function; and positive and negative emotions for the arousal function. Further research should test whether these elements offer more effective explanations of anxious attachment users’ thought processes. Moreover, this study tests the goal cognition process, from directive- to regulatory-oriented thinking, and then to control-oriented and finally to arousal-oriented thinking. However, goal cognition may follow other orders in different contexts or for different people, or it might be interactive. Continued research should explore these possibilities.

This study does not explore how goal cognition guides FB activities. People can adopt various activities to maintain or enhance their FB friendships. For example, Chang (2015) identifies responding behaviors, such as commenting on others’ posts and photos, and self-revealing behaviors, such as posting one’s own opinions, feelings, and activities. People with different self-concepts engage in different FB behaviors or activities in pursuit of their social goals (Chang, 2015). Further research might explore which precise activities people adopt if they monitor their goal process and find themselves falling behind their expectations.

Moreover, it is not clear whether people with a stronger anxious attachment style suffer lower social efficacy on FB, which might cause ambivalent experiences with FB. On the one hand, an anxious attachment orientation is negatively associated with social efficacy (Mallinckrodt & Wei, 2005; Wei, Russell, & Zalalik, 2005). On the other hand, Oldmeadow, Quinn, and Kowert (2013) argue that FB offers an alternative platform to maintain friendships for people who feel incapable of interacting in the real world. That is, anxious people may have greater social efficacy on FB than in reality. These two competing perspectives deserve empirical exploration.

More reliable subjective measures of the time spent on FB also need to be developed. Online behaviors often are complicated, such that behavioral scales that rely on subjective ratings may fail to reflect reality; the perception–reality discrepancy also likely varies across different users. For example, some users turn on notification functions and respond to or at least check for new messages as soon as they receive the notification. Others, and especially college students, engage in multitasking that includes FB; 79% of respondents in one study used FB while doing work and study-related tasks (Kononova & Yuan, 2017). In this sense, it is difficult to distinguish the time spent on FB from time they are simply logged in to it. Additional research should undertake scale development efforts to measure FB use more accurately.

Limitations

The cross-sectional nature of the data cannot establish causal effects conclusively. To prevent participants from answering the survey multiple times, this study also limited each IP address to one response, which might exclude people who go online using public gateway routers. In addition, there are many ways to capture attachment orientation (see Ravitz, Maunder, Hunter, Sthankiya, & Lancee, 2010). This study used items from Hazan and Shaver’s (1987) descriptors of different attachment styles as measurement items. As reported, the reliability of these scales is acceptable, but continued research might test other well-established measures with good reliability (Graham & Unterschute, 2015). Finally, without relevant scales available in prior research, some scales were developed for the purpose of this study, and they may be low in validity. Despite these limitations, the proposed theoretical framework helps explain goal cognition for anxious people who rely on FB to maintain their friendships.

Footnotes

Authors’ Note

Chingching Chang is also a distinguished research fellow at Academia Sinica, Taipei, Taiwan.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Ministry of Science and Technology (grant number 105-2410-H-004-105-SS3).

Open research statement

This research was not pre-registered. The data and materials used in the research are available upon request by emailing shenc@nccu.edu.tw.

Note

References

Baumeister

R. F.

Leary

M. R.

(1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497–529.

Bers

S. A.

Rodin

(1984). Social-comparison jealousy: A developmental and motivational study. Journal of Personality and Social Psychology, 47, 766–799.

Bowlby

(1969). Attachment and loss: Vol. 1. Attachment. New York, NY: Basic Books.

Bowlby

(1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. New York, NY: Basic Books.

Bowlby

(1980). Attachment and loss: Vol. 3. Loss. New York, NY: Basic Books.

Bowlby

(1988). A secure base: Parent-child attachment and healthy human development. New York, NY: Basic Books.

Brendl

C. M.

Higgins

E. T.

(1996). Principles of judging valence: What makes events positive or negative? In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 28, pp. 95–160). New York, NY: Academic Press.

Brennan

K. A.

Clark

C. L.

Shaver

P. R.

(1998). Self-report measurement of adult attachment: An integrative overview. In Simpson

J. A.

Rholes

W. S.

(Eds.), Attachment theory and close relationships (pp. 46–76). New York, NY: Guilford Press.

Brislin

R. W.

(1987). The wording of translation of research instruments. In Lonner

W. J.

Berry Field

J. W.

(Eds.), Methods in cross cultural research (pp. 137–164). Beverly Hills, CA: Sage.

10.

Chang

(2015). Self-construal and Facebook activities: Exploring differences in social interaction orientation. Computers in Human Behavior, 53, 91–101.

11.

Collins

N. L.

Read

S. J.

(1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58, 644–663.

12.

Evraire

L. E.

Dozois

D. J.

(2014). If it be love indeed tell me how much: Early core beliefs associated with excessive reassurance seeking in depression. Canadian Journal of Behavioural Science, 46, 1–8.

13.

Feeney

B. C.

Collins

N. L.

(2001). Predictors of caregiving in adult intimate relationships: An attachment theoretical perspective. Journal of Personality and Social Psychology, 80, 972–994.

14.

Feeney

J. A.

Noller

Hanrah

(1994). Assessing adult attachment. In Sperling

M. B.

Berman

W. H.

(Eds.), Attachment in adults: Clinical and developmental perspectives (pp. 128–152). New York, NY: Guilford Press.

15.

Fox

Osborn

J. L.

Warber

K. M.

(2014). Relational dialectics and social networking sites: The role of Facebook in romantic relationship escalation, maintenance, conflict, and dissolution. Computers in Human Behavior, 35, 527–534.

16.

Fraley

C. R.

Niedenthal

P. M.

Marks

Brumbaugh

Vicary

(2006). Adult attachment and the perception of emotional expressions: Probing the hyperactivating strategies underlying anxious attachment. Journal of Personality, 74, 1163–1190.

17.

Furman

Simon

V. A.

Shaffer

Bouchey

H. A.

(2002). Adolescents’ working models and styles for relationships with parents, friends, and romantic partners. Child Development, 73, 241–255.

18.

Gangadharbatla

(2008). Facebook me: Collective self-esteem, need to belong, and internet self-efficacy as predictors of the iGeneration’s attitudes toward social networking sites. Journal of Interactive Advertising, 8, 5–15.

19.

Graham

J. M.

Unterschute

M. S.

(2015). A reliability generalization meta-analysis of self-report measures of adult attachment. Journal of Personality Assessment, 97, 31–41.

20.

Griffin

D. W.

Bartholomew

(1994). Models of the self and other: Fundamental dimensions underlying measures of adult attachment. Journal of Personality and Social Psychology, 67, 430–445.

21.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

22.

Hazan

Shaver

(1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511–524.

23.

Hazan

Shaver

(1994). Attachment as an organizational framework for research on close relationships. Psychological Inquiry, 5, 1–22.

24.

Higgins

E. T.

(1997). Beyond pleasure and pain. American Psychologist, 52, 1280–1300.

25.

Hoaglin

D. C.

Iglewicz

(1987). Fine tuning some resistant rules for outlier labeling. Journal of American Statistical Association, 82, 1147–1149.

26.

Jenkins-Guarnieri

M. A.

Wright

S. L.

Hudiburgh

L. M.

(2012). The relationships among attachment style, personality traits, interpersonal competency, and Facebook use. Journal of Applied Developmental Psychology, 33, 294–301.

27.

Karoly

(1993). Goal systems: An organizing framework for clinical assessment and treatment planning. Psychological Assessment, 5, 273–280.

28.

Karoly

(1999). A goal systems–self-regulatory perspective on personality, psychopathology, and change. Review of General Psychology, 3, 264–291.

29.

Karoly

Ruehlman

L. S.

(1995). Goal cognition and its clinical implications: Development and preliminary validation of four motivational assessment instruments. Assessment, 2, 113–129.

30.

Karoly

Ruehlman

L. S.

(1996). Motivational implications of pain: Chronicity, psychological distress, and work goal construal in a national sample of adults. Health Psychology, 15, 383–390.

31.

Kerr

N. L.

Levine

J. M.

(2008). The detection of social exclusion: Evolution and beyond. Group Dynamics: Theory, Research, and Practice, 12, 39–52.

32.

Kononova

A. G.

Yuan

(2017). Take a break: Examining college students’ media multitasking activities and motivations during study- or work-related tasks. Journalism and Mass Communication Educator, 72, 183–197.

33.

Leary

M. R.

Kelly

K. M.

Cottrell

C. A.

Schreindorfer

L. S.

(2005). Individual differences in the need to belong: Mapping the nomological network. Working paper, Wake Forest University, Winston-Salem, NC.

34.

Mallinckrodt

Wei

(2005). Attachment, social competencies, social support, and psychological distress. Journal of Counseling Psychology, 52, 358–367.

35.

Marshall

T. C.

Bejanyan

Di Castro

Lee

R. A.

(2013). Attachment styles as predictors of Facebook-related jealousy and surveillance in romantic relationships. Personal Relationships, 20, 1–22.

36.

Meyer

Pilkonis

P. A.

Beevers

C. G.

(2004). What’s in a (neutral) face? Personality disorders, attachment styles, and the appraisal of ambiguous social cues. Journal of Personality Disorders, 18, 320–336.

37.

Mikulincer

Shaver

P. R.

(2005). Attachment theory and emotions in close relationships: Exploring the attachment-related dynamics of emotional reactions to relational events. Personal Relationships, 12, 149–168.

38.

Mikulincer

Shaver

P. R.

(2007). Attachment in adulthood: Structure, dynamics and change. New York, NY: Guildford Press.

39.

Mikulincer

Shaver

P. R.

Bar-On

Ein-Dor

(2010). The pushes and pulls of close relationships: attachment insecurities and relational ambivalence. Journal of Personality and Social Psychology, 98, 450–468.

40.

Mikulincer

Shaver

P. R.

Pereg

(2003). Attachment theory and affect regulation: The dynamics, development, and cognitive consequences of attachment-related strategies. Motivation and Emotion, 27, 77–102.

41.

Nadkarni

Hofmann

S. G.

(2012). Why do people use Facebook? Personality and Individual Differences, 52, 243–249.

42.

Nangle

D. W.

Erdley

C. A.

Newman

J. E.

Mason

C. A.

Carpenter

E. M.

(2003). Popularity, friendship quantity, and friendship quality: Interactive influences on children’s loneliness and depression. Journal of Clinical Child and Adolescent Psychology, 32, 546–555.

43.

Nitzburg

G. C.

Farber

B. A.

(2013). Putting up emotional (Facebook) walls? Attachment status and emerging adults’ experiences of social networking sites. Journal of Clinical Psychology, 69, 1183–1190.

44.

Oldmeadow

J. A.

Quinn

Kowert

(2013). Attachment style, social skills, and Facebook use amongst adults. Computers in Human Behavior, 29, 1142–1149.

45.

Ravitz

Maunder

Hunter

Sthankiya

Lancee

(2010). Adult attachment measures: A 25-year review. Journal of Psychosomatic Research, 69, 419–432.

46.

Saferstein

J. A.

Neimeyer

G. J.

Hagans

C. L.

(2005). Attachment as a predictor of friendship qualities in college youth. Social Behavior and Personality: An International Journal, 33, 767–776.

47.

Santor

D. A.

Messervey

Kusumakar

(2000). Measuring peer pressure, popularity, and conformity in adolescent boys and girls: Predicting school performance, sexual attitudes, and substance abuse. Journal of Youth and Adolescence, 29, 163–182.

48.

Scharf

Oshri

Eshkol

Pilowsky

(2014). Adolescents’ ADHD symptoms and adjustment: The role of attachment and rejection sensitivity. American Journal of Orthopsychiatry, 84, 209–217.

49.

Shaver

P. R.

Mikulincer

(2002). Attachment-related psychodynamics. Attachment and Human Development, 4, 133–161.

50.

Simpson

J. A.

Rholes

W. S.

Nelligan

J. S.

(1992). Support seeking and support giving within couples in an anxiety-provoking situation: The role of attachment styles. Journal of Personality and Social Psychology, 62, 434–446.

51.

Thompson

M. M.

Zanna

M. P.

Griffin

D. W.

(1995). Let’s not be indifferent about (attitudinal) ambivalence. Attitude Strength: Antecedents and Consequences, 4, 361–386.

52.

Wei

Russell

D. W.

Zakalik

R. A.

(2005). Adult attachment, social self-efficacy, self-disclosure, loneliness, and subsequent depression for freshman college students: A longitudinal study. Journal of Counseling Psychology, 52, 602–614.

53.

Welch

R. D.

Houser

M. E.

(2010). Extending the four-category model of adult attachment: An interpersonal model of friendship attachment. Journal of Social and Personal Relationships, 27, 351–366.

54.

Yaakobi

Goldenberg

(2014). Social relationships and information dissemination in virtual social network systems: An attachment theory perspective. Computers in Human Behavior, 38, 127–135.

55.

You

H. S.

Malley-Morrison

(2000). Young adult attachment styles and intimate relationships with close friends: A cross-cultural study of Koreans and Caucasian Americans. Journal of Cross-Cultural Psychology, 31, 528–534.