Minimizing the Pain and Probability of Rejection

Abstract

Some studies have revealed evidence for proximity seeking following interpersonal rejection, whereas other studies have found evidence of relational distancing (Williams, 2007). This study examined whether both processes could occur simultaneously within face-to-face interactions. Participants were accepted or rejected by one person and then interacted with a new partner for the purposes of an impression-formation task. Conversations were recorded and transcribed. Dyads containing a previously rejected compared to accepted participant exhibited higher levels of linguistic style matching and reciprocated conversational content, suggestive of proximity seeking. However, rejected targets also rated their new partners as less kind and reported lower levels of rapport/liking for their partners, consistent with relational distancing. Partner evaluations were statistically mediated by targets’ expectations of rejection. We suggest that automatic proximity seeking and the appraisal-mediated devaluation of new partners reflect efforts to minimize the potential for and pain of future rejection. Recommendations for future research are discussed.

Keywords

social rejection linguistic style matching avoidance social pain attachment regulation

Few would disagree that rejection hurts. In fact, the affective and somatosensory representations of social rejection overlap with those of physical pain (Eisenberger, Lieberman, & Williams, 2003; Kross, Berman, Mischel, Smith, & Wager, 2011), suggesting that physical pain and social pain share a common set of neural substrates (MacDonald & Leary, 2005). Similar to the way that physical pain motivates people to reduce the probability of painful future events, social pain should lead people to take preventative measures within their social contexts. Despite burgeoning work on this topic (Smart Richman & Leary, 2009; Williams, 2007), however, few studies have investigated directly the strategies that rejected individuals employ to reduce their vulnerability to future rejection and its coincident emotions.

Responses to Social Rejection

Several studies suggest that individuals seek proximity to others following social rejection. For example, rejected individuals are more likely than others to attend to smiling faces (Dewall, Maner, & Rouby, 2009), perceive friendliness in the face of others (Maner, DeWall, Baumeister, & Schaller, 2007), remember social events embedded in others’ diaries (Gardner, Pickett, & Brewer, 2000), bolster the value of the groups to which they belong (Knowles & Gardner, 2008), and report a stronger interest in meeting new people (Maner et al., 2007). Rejection also increases implicit behavioral mimicry (Lakin, Chartrand, & Arkin, 2008) and conformity (Williams, Cheung, & Choi, 2000), both of which have their evolutionary roots in self-protection and have been linked to threatened sense of security (Griskevicius, Goldstein, Mortensen, Cialdini, & Kenrick, 2006).

Other research has found evidence for relational distancing and psychological “numbness” in response to rejection. For example, rejection decreases empathy and pain sensitivity (Dewall & Baumeister, 2006) and increases aggression (Twenge, Baumeister, Tice, & Stucke, 2001). Both analgesia and aggression responses to rejection mimic defensive, adaptive responses to physical pain (MacDonald & Leary, 2005) and have been linked empirically (Bernstein & Claypool, 2012; Twenge et al., 2007) and theoretically (Molden & Maner, 2013; Smart Richman & Leary, 2009) to negative appraisals of others’ regard.

The evidence is thus mixed as to whether rejection motivates affiliative, avoidant, or aggressive behavior within subsequent interactions. This issue is further complicated by the fact that most studies examining interpersonal responses to rejection have used hypothetical targets or pseudo-partners with whom rejected individuals have no direct contact. Real-life instances of rejection occur within dynamic social situations in which immediate escape from others is not possible or even desired. Avoidant or aggressive behaviors may be less likely to occur when the prospect for further rejection is both immediate and salient.

The Present Investigation

We propose that rejected individuals will seek to minimize both the pain and probability of future rejection within new social interactions. We envisioned two means by which this could occur. The first involves drawing closer to partners and emphasizing commonalities, which would decrease the probability of additional rejection. The second involves avoiding closeness with partners and accentuating perceived differences, which would minimize the pain of possible rejection. The possibility that both affiliative and avoidant responses could occur simultaneously has received almost no attention to date (for an exception, see Hess & Pickett, 2010). We further reasoned that these motivations could manifest along a continuum of automatic to controlled processes. For example, efforts to minimize one’s prospects of rejection could take the form of implicit proximity seeking or overt ingratiation strategies. Efforts to minimize the pain of rejection could take the form of implicit distancing or overt derogation of the partner.

One promising method for assessing the implicit affiliation/distancing behaviors within conversations is through the analysis of spontaneous word choices (Gonzales, Hancock, & Pennebaker, 2010; Pennebaker, Francis, & Booth, 2001). Research suggests that a person’s linguistic style (LS; Pennebaker & King, 1999) and the extent to which it matches that of a conversation partner (linguistic style matching [LSM]) is reflective of the ongoing nature of an interaction (Gonzales et al., 2010). LSM in particular is presumed to reflect affiliation goals and, like nonverbal behavioral mimicry, occurs outside of conscious awareness (Niederhoffer & Pennebaker, 2002). Relative differences in LSM could thus reveal whether rejected (compared to accepted) individuals behave in a manner that facilitates or subtly undermines an interaction.

Participants were rejected or accepted by one partner following a brief interaction. They were then introduced to another partner for the purposes of an impression-formation task. Prior to meeting their new partners, targets reported their expectations of, and concerns about, the upcoming interaction. Targets and their partners also evaluated each other along a series of traits following the interaction.

Interactions were recorded, transcribed, and subjected to linguistic analyses. This allowed us to assess targets’ spontaneous expressive behavior during the interaction. Whereas linguistic behaviors and LSM were assumed to result from relatively automatic processes (Ireland & Pennebaker, 2010; Niederhoffer & Pennebaker, 2002; Pennebaker & King, 1999), self-report measures were assumed to reflect more controlled processes (Hadjistavropoulos & Craig, 2002). Both types of data were explored for evidence of proximity seeking and social distancing.

Method

Participants and Procedure

Participants were 182 undergraduates (121 females) enrolled in introductory psychology courses at a university in the Pacific Northwest. Four participants were scheduled for each session. Participants were paired off and escorted to separate rooms. One dyad was chosen randomly to receive the experimental manipulation, while the other provided the interaction partners with whom accepted and rejected participants would later interact. Members of both dyads were informed they would meet with a different individual later for a 5-min impression formation task. First, however, they would complete a “practice” or warm-up session, where they could talk about whatever they wanted for 3 min.

The dyad chosen randomly to receive the manipulation then privately evaluated their partner on 4 items including “How much did you like the other participant?” and “Do you feel you could be friends with this person in the future?” Responses were recorded on 7-point scales (1 = not at all, 7 = very much). The experimenter mentioned casually that, “participants are typically curious to know if they formed similar impressions of one another,” and thus they were permitted to see each other’s ratings. One member was randomly chosen to receive acceptance feedback while the other received rejection feedback. The manipulation was administered by replacing participants’ actual ratings with previously prepared forms reflecting either acceptance (ratings of 6 or 7) or mild rejection (ratings of 3 or 4). The experimenter placed the rating forms face up on the desk where the participant was seated as the experimenter administered the next set of questionnaires (described subsequently). This procedure was skipped for the second pair of participants.

The two dyads then switched partners such that each new pairing contained a person who was just accepted or rejected (“target”) and a person who received no manipulation (“partner”). Targets and their partners were seated in swivel chairs, roughly 5 feet apart, and provided with a list of moderately intimate questions (e.g., “What would be the perfect lifestyle for you?”) from the Relationship Closeness Induction Task (Sedikides, Campbell, Reeder, & Elliot, 1999). The experimenter emphasized that the questions were intended to facilitate conversation and that participants should feel free to deviate from the list. These 5-min interactions were videotaped. Upon completion, participants were separated and asked to report their impressions of one another as well as the interaction. Participants were debriefed.

Dependent Measures

Reactions to the manipulation

Immediately following the manipulation, targets completed 4 items tapping their expectations for the upcoming 5-min interaction (e.g., “How much are you looking forward to the interaction?” Do you think this person will like you?”) and 2 items assessing how concerned they were about the interaction (e.g., “How important is it to you that this person likes you?”). Items were collapsed to provide an overall measure of social expectancies (α = .88) and interpersonal concern (α = .79). A subset of 68 participants also completed measures of positive affect (α = .79) and negative affect (α = .78; Watson, Clark, & Tellegen, 1988).

Trait ratings

Following the second interaction, targets and their partners were asked to rate the extent to which their partner displayed each of 15 positive and 15 negative traits. These ratings were collapsed and subjected to Principal Components Analyses (PCA). Items whose component loadings failed to differ by at least .30 within the initial and secondary PCAs were discarded. The final solution (accounting for 66% of the variance) yielded four components that we labeled: kind (kind, open, patient, understanding, modest, responsive to others’ needs, and tolerant), moody (moody, thoughtless, irrational, complaining, and childish), self-assured (self-assured, sociable, and witty), and critical (critical, arrogant, and fake).

Postinteraction assessment

Participants completed a Deindividuation scale (Rogers & Prentice-Dunn, 1981) and four newly created items tapping the extent to which participants were consciously involved and engaged in the interaction. A PCA yielded four components that we labeled: rapport/liking (6 items, e.g., “Time seemed to pass quickly”; “I liked my partner”), self-consciousness (4 items, e.g., “I was concerned with what my partner was thinking about me”), personal responsibility (2 items, e.g., “I was primarily responsible for what transpired during the conversation”), and focus/concentration (2 items, e.g., “My thoughts seemed concentrated and focused on the moment”).

Verbal behaviors

Conversations were transcribed and analyzed with the Linguistic Inquiry Word Count (LIWC; Pennebaker, Francis, & Booth, 2001). Two sets of linguistic dimensions were identified for analyses. The first set tapped interpersonal and emotional processes associated with rejection and acceptance. Positive and negative emotion words provided linguistic markers of emotional tone (Pennebaker & King, 1999). Laugh symbols were added to our LIWC dictionary as an additional, indirect behavioral marker of emotion. Previous research (Owen et al., 2006) has found that self-report and linguistic markers of affect are generally uncorrelated, suggesting that linguistic emotional content does not reflect the conscious experience of mood. Number of questions asked reflected the amount of attention directed toward the partner.

The second set of dimensions assessed LSM, or coordination in the structure or style of speech, as well as convergence in the topics (“personal concerns”) discussed during the conversation. All major categories of linguistic dimensions reflected in the 2001 dictionary were analyzed, with two exceptions. In lieu of analyzing each of the 19 psychological processes (and to reduce the Type I error rate), we employed an exhaustive set of factor-analytically derived composite variables identified in prior work (Cohn, Mehl, & Pennebaker, 2004; Pennebaker & King, 1999). Psychological immediacy (the inverse of “psychological distancing,” Cohn et al., 2004) reflects the use of a more personal, experiential tone rather than an impersonal, rational-speaking tone. Making distinctions refers the use of more qualifiers and negations (“but,” “maybe,” and “never”) while speaking. Social past reflects a higher number of past-tense references to relationships (friends, coworkers) or social events (talking). Rationalization refers to causation and insight words (“because” and “consider”), and less frequent use of negative emotion words. Emotional positivity refers to the simple difference between the use of positive and negative emotion words. Further, we omitted two personal concerns (metaphysical issues and physical states) due to their very low frequency (i.e., less than 1% of all utterances) within participants’ interactions. Ireland and Pennebaker (2010) provided evidence that LSM and conversational content matching cannot be intentionally bolstered. This increased our confidence that the linguistic patterns under investigation were not subject to conscious awareness or control.

Results

Immediate reactions to the manipulation

As shown in Table 1, rejected targets reported more negative social expectancies and lower concern for the quality of the upcoming interaction. Positive and negative affect scores were strongly inversely correlated (−.50) so the latter was subtracted from the former to provide an overall index of affective responses to the manipulation. Rejected and accepted targets differed significantly in an overall self-reported affect.

Table 1.
Reactions to the Manipulation.

F(1, 88) η² Accepted M (SD) Rejected M (SD)

Social expectancies 24.29** .20 4.95 (.78) 4.02 (.99)

Interpersonal concern 6.07* .07 3.70 (1.45) 2.97 (1.37)

Affect^a 4.32* .06 1.25 (.74) .81 (1.00)

Note. SD = standard deviation. ^a df = 1, 66. Higher means reflect higher positive/lower negative affect.

*p ≤ .05. **p ≤ .01.

Postinteractional assessment

As revealed in Table 2, rejected targets evaluated their new partners as less kind and reported lower levels of rapport and focus/concentration during the interaction than did accepted targets. One might expect these negative self-reports to have reflected negative conversational behavior that their partners would observe and react to, but this did not happen. Although partners interacting with rejected targets reported feeling more self-conscious than those interacting with accepted targets, partners rated rejected targets as being less critical than accepted targets. Apparently the rejected targets were exhibiting interpersonal behavior that was more positive than their self-reports alone would have suggested.

Table 2.
Postinteraction Assessment.

F(1, 87) η² Accepted M (SD) Rejected M (SD)

Targets’ ratings of partners^a

Kind 7.79** .08 5.62 (0.87) 5.00 (1.00)

Moody 1.61 .0 1.54 (0.73) 1.73 (0.69)

Self-assured 3.63 .04 4.85 (1.19) 4.31 (1.03)

Critical 0.11 .00 1.87 (.98) 1.91 (0.62)

Partners’ ratings of targets^a

Kind 0.81 .01 5.48 (1.05) 5.31 (0.84)

Moody 0.02 .00 1.61 (0.74) 1.65 (0.85)

Self-assured 1.14 .01 4.60 (1.31) 4.39 (1.12)

Critical 4.06* .05 1.99 (1.04) 1.67 (0.76)

Targets’ ratings of the interaction^b

Rapport/liking 7.30 .09 4.06 (0.49) 3.68 (0.73)

Self-consciousness 0.02 .00 2.35 (0.82) 2.34 (0.85)

Personal responsibility 0.51 .01 2.62 (0.73) 2.73 (0.74)

Focus/concentration 13.53* .16 4.14 (0.64) 3.51 (0.87)

Partners’ ratings of the interaction^b

Rapport/liking 0.00 .00 3.93 (0.69) 3.92 (0.46)

Self-consciousness 3.95* .05 2.12 (0.73) 2.46 (0.76)

Personal responsibility 1.70 .02 2.54 (0.89) 2.80 (0.81)

Focus/concentration 0.85 .01 4.04 (0.64) 3.89 (0.83)

Note. ^aAnalyses controlled for sex of rater and sex of target. ^b df = 1,74.

p < .05. p < .01. p < .001.

Interpersonal and emotional processes

As shown in Table 3, rejected targets used more anger and anxiety words, evinced fewer laugh symbols, and asked more questions of their partners than did accepted targets.

Table 3.
LIWC Analysis: Target Differences in Language Use.

Definition/Examples F(1, 88) η² Accepted M (SD) Rejected M (SD)

Positive emotions Happy, optimistic .98 .01 3.11 (1.15) 2.87 (1.38)

Negative emotions Enemy, worthless 2.88 .03 1.01 (.66) 1.25 (.68)

Anxiety/fear Afraid, tense 4.09* .04 0.15 (.25) 0.28 (.32)

Anger Hate, kill 4.38* .05 0.13 (.21) 0.23 (.27)

Sadness Cry, sad .13 .00 0.12 (.17) 0.11 (.28)

Laugh symbols Hehe, hehehe 8.58** .09 5.72 (4.79) 3.74 (2.52)

Questions asked Number questions asked 4.58* .05 2.53 (1.15) 3.04 (1.36)

Note. LIWC = Linguistic inquiry word count.

p ≤ .05. p ≤ .01.

LSM and content matching

Table 4 revealed higher levels of matching among rejected compared to accepted dyads on 18 of the 22 linguistic dimensions examined. Because some degree of convergence was found for both accepted and rejected dyads, only two of these comparisons in correlations (for words greater than six letters and psychological immediacy) reached significance. However, a within-study meta-analysis for nonindependent effects (Hayes, 1998) revealed that the combined effect size associated with the difference between rejected and accepted dyads was large and significant (Z′ = 3.50, p < .001).

Table 4.
Language Style Matching (LSM) and Convergence in Conversational Topics.

Definitions/Examples Accepted Dyads n = 46 Rejected Dyads n = 45 Z

I. Standard dimensions

Word count Total word count −.44 −.18 1.34

Words per sentence Words per sentence .20 .48 1.48

Words >6 letters Words over 6 letters .09 .56 2.50

First person I, me, we .52** .06 −2.38*

Second person You −.24 −.27 −.15

Third person He, she, they .33* .41* .43

Negations No, never, not .30* .27 −.15

Assents Yes, OK −.08 .03 .51

Articles A, an, the −.05 .22 1.26

Prepositions On, to, from .04 .19 .70

Numbers One, thirty, million .35* .53 1.04

II. Psychological processes

Immediacy (Zi + Zdiscrep + Zpresent) − (Zarticle + Zsixltr) .18 .59* 2.28*

Making distinctions (Zdiscrep + Zexcl + Ztentat + Znegate) − ZincL .04 .31* 1.29

Social past (Zsocial + Zpast) − (Zpresent + Zposemo) .43 .43 0.00

Rationalization (Zcause + Zinsight) − (Znegemo) .04 .40 1.77

Emotional positivity Zposemo − Znegemo .17 .45 1.44

III. Relativity

Time Hour, day, oclock .19 .31* .62

Space Around, over, up .43 .46 .18

Motion Walk, move, go .45 .58* .85

IV. Personal Concerns

Occupation School, job, goal .43 .67* 1.62

Leisure activity Home, sports, music .36* .60* 1.46

Money/Financial Issues Cash, income, money .40 .63*** 1.46

p < .05. p < .01. p < .001.

Mediational analyses

Additional analyses were conducted to determine whether immediate responses to the manipulation (expectancies, concern or affect) mediated the effects of experimental condition on the outcomes reported in Tables 2 and 3. Mediation was tested only when immediate reactions bore significant relationships to both the manipulation and the outcome variable. Table 5 provides the correlations between all possible mediators and outcome variables. Immediate responses were tested as parallel mediators of target outcomes using bootstrapping procedures based on 1,000 samples and a 95% bias-corrected confidence interval (Preacher & Hayes, 2008). Indirect effects (IEs) were considered significant if the confidence interval did not include zero. As revealed in Table 6, social expectancies and interpersonal concern mediated the effects of rejection on targets’ ratings of kindness, and expectancies uniquely mediated the effects of rejection on rapport/liking. No other effects were significant.

Table 5.
Correlations Among Immediate Reactions and Outcomes of the Manipulation.

Immediate Reactions to the Manipulation

Expectancies Concern Affect

Targets’ ratings of partners

Kind .53 .39 .23

Partners’ ratings of targets

Critical .08 .04 −.02

Targets’ ratings of the interaction

Rapport/liking .46** .20 .28*

Focus/concentration .29* .31 .10

Partners’ ratings of the interaction

Self-consciousness −.19 .01 −.12

Language use

Anxiety/fear −.05 −.19 .04

Anger −.29 −.24* −.09

Laugh symbols .11 .13 .05

Questions asked −.98 −.02 −.30*

*p ≤ .05. **p ≤ .01.

Table 6.
Results of Bootstrapping Analyses.

Mediators Tested Indirect Effect SE Lower Limit for 95% CI Upper Limit for 95% CI

Target outcomes

Kind Social expectancies −.370 .134 −.772 −.216

Interpersonal concern −.094 .062 −.285 −.008

Rapport/liking Social expectancies −.244 .115 −.524 −.094

Affect −.030 .052 −.141 .067

Focus/concentration Social expectancies −.055 .114 −.402 .064

Interpersonal concern −.087 .075 −.302 .009

Anger language Social expectancies .040 .031 −.014 .108

Interpersonal concern .016 .015 −.003 .057

Questions asked Affect .169 .132 .010 .568

Note. CI = confidential interval.

Discussion

We began this investigation with the assumption that the pain of rejection would motivate individuals to avoid additional injury to the self. Our findings pointed to a complex pattern of responses that, collectively, appeared to serve the goals of thwarting both the pain of and potential for additional rejection. Evidence for the former was found in targets’ negative ratings of their partners and their interactions. Rejected (compared to accepted) individuals evaluated their new partners as less kind and reported lower levels of rapport with them. We saw no evidence that partners randomly paired with rejected targets actually behaved in ways that were less warm or friendly than partners paired with accepted targets. Rather, targets of a previous rejection appeared to be motivated to evaluate their partners and interactions more negatively, presumably in efforts to weaken the blow of possible criticism by that partner. This interpretation is supported by the fact that targets’ expectations of future rejection (and to a lesser extent their reduced concerns about the upcoming interaction) mediated their subsequent negative evaluations of their partners.

Despite these apparent, preemptive efforts to thwart the pain of rejection, rejected individuals actually behaved in ways that maintained or increased closeness with their partners. Evidence for proximity seeking was found in targets’ linguistic behaviors. Specifically, rejected targets were more likely to match the conversational content of their new partners and to mimic their partners’ LSs. This type of coordination and mimicry is considered to be a defining attribute of rapport and connectedness within social interactions (Bernieri & Rosenthal, 1991; Gonzales et al., 2010; Lakin & Chartrand, 2005).

Two categories of LSM deviated significantly from this pattern. First, word counts among rejected individuals and their partners remained unrelated, whereas word counts among accepted targets and their partners were significantly inversely related. This finding indicates that accepted pairs tended to have a speaker who dominated the conversation while the other listened. Research suggests that complementarity in dominance behaviors is more characteristic of interactions within nonclose relationships than close relationships (Moskowitz, Ho, & Turcotte-Tremblay, 2007).

Second, high levels of linguistic matching were observed within accepted but not rejected dyads on the category of first person pronouns. This category differs from other categories in that matching reflects a tendency to reciprocate a conversation partner’s self-references with references to oneself (e.g., responding to “I” with “I” and “we” with “we”). The lower level of LSM among rejected dyads is reminiscent of prior work showing that rejected individuals tend to avoid self-focus (Twenge, Cantonese, & Baumeister, 2003). It is also consistent with the tendency for rejected targets in the present study to ask more questions of their partners. Collectively, these findings suggest that (1) rejection increases implicit proximity seeking via LSM and reciprocated conversational content and (2) LSM is unlikely to occur when it thwarts rather than strengthens connections with others or refocuses attention on the self.

Note that LSM did not result in higher ratings of interaction quality; rejected individuals reported lower levels of rapport/liking than did accepted individuals, and their partners’ ratings did not differ. Tickle-Degnen and Rosenthal (1987) have argued that self-reports of positivity (e.g., liking) are one of the three independent components of true rapport, which also includes behavioral coordination and mutual attention. Although moderately correlated, these three attributes are driven by independent processes and thus can diverge. Applied to the present study, rejected participants may have been more implicitly engaged in their interactions but defensively motivated to deny that their partners were likeable. Moreover, targets may have exhibited other behaviors—including those that led their partners to feel more self-conscious—that suppressed any partner-reported increases in rapport that would normally come with higher levels of coordination and attention.

Temporal Differences in Attachment Regulation

We entered this investigation unsure of whether our findings would reveal evidence of affiliative or avoidant/aggressive behaviors following rejection and the domains in which these behaviors might occur. Ultimately, most of the evidence for affiliative responses was revealed in spontaneous linguistic behaviors, whereas most of the evidence for distancing appeared in postinteraction self-reports. These findings provide good corroborating support for proximity seeking as a relatively implicit process that occurs quickly and without conscious awareness (Lakin et al., 2008), and relational distancing as a relatively deliberate or higher order mental process that is mediated by conscious appraisals of others’ regard (Bernstein & Claypool, 2012; Twenge et al., 2007).

Not all our data were perfectly consistent with this proposed dichotomy, however. For example, rejected participants used a higher proportion of anger and anxiety words, and evinced fewer laugh syllables, while speaking. These findings may have reflected the immediate or reflexive pain of rejection (Williams, 2009), without having any direct bearing on the extent to which targets and their partners were linguistically synchronized. Another source of ambiguity involves the verbal and/or nonverbal behaviors that served as the basis for partners’ ratings. Partners rated rejected compared to accepted targets as less arrogant and critical. The fact that partner impressions were uncorrelated with targets’ expectations of rejection suggests that rejected individuals’ suppression of critical behaviors was a relatively automatic response to the previous rejection rather than a controlled response to the expectation of future rejection. Unfortunately, however, our coding procedures did not allow us to assess every relevant behavior that may have been responsible for partners' impressions or to determine the extent to which they may have resulted from strategic self-presentation goals or relatively automatic connection goals.

These limitations notwithstanding, we note strong parallels between our findings and regulatory responses to threatened security in close relationships. For example, the model of attachment regulation proposed by Mikulincer and Shaver (Mikulincer & Shaver, 2003; Shaver & Mikulincer, 2002) places responses to rejection along a continuum of automatic to controlled processes. Automatic thoughts and behaviors aim to reduce physical or psychological distance from attachment figures and represent unconscious efforts to minimize distress. Controlled thoughts and behaviors result from conscious appraisals of whether others will be responsive to one’s needs. Attachment figures who are perceived as available sources of support are approached as trusted sources of renewed attachment, whereas attachment figures who are perceived as unavailable compound feelings of threat, resulting in “hyperactivation” or “deactivation” of the attachment system (Cassidy & Kobak, 1988; Main, 1990). Both manifestations of attachment insecurity are associated with higher levels of hostility and aggression (Mikulincer & Shaver, 2011; Péloquin, Lafontaine, & Brassard, 2011).

A similar set of processes appears in the risk regulation framework proposed by Murray and colleagues (Murray, Derrick, Leder, & Holmes, 2008). According to this model, threats to felt security automatically activate connectedness goals unless the threat is salient and internal to the relationship (in which case self-protection may be prioritized). The conflict between connection and protection goals is then resolved by an executive (control) system wherein individuals assess the likelihood that partners will be responsive to their needs. Individuals risk greater vulnerability and dependence on partners to the extent that they feel they can trust the partner. Situational or dispositional cues that undermine trust in the partner, by comparison, lead to avoidance processes that manifest in the devaluation of the partner and the relationship.

Thus, one promising set of frameworks for future work that allows for the simultaneous execution of affiliative and avoidant/aggressive responses to rejection is that grounded in attachment theory and risk regulation. Both models place proximity seeking early in the sequence of temporal responses to rejection and highlight the critical role of expectancies in determining controlled responses to rejection. Both would also predict that individuals will seek to derogate and downplay the importance of relationships with those (like strangers) who are viewed as nonviable sources of connection. Though created primarily to account for threats to security within close relationships, these attachment regulation models may be adapted to explain interpersonal responses to rejection within nonclose relationships as well (see also Maxwell, Spielmann, Joel, & MacDonald, 2013; Sommer & Benkendorf, 2009).

Future Directions

Our findings were both consistent and inconsistent with previous work linking rejection to increases in aggression toward novel others (Twenge et al., 2001). Rejected individuals in the present study rated their partners as less kind than did accepted individuals. Simultaneously, partners rated rejected individuals as less critical and arrogant than accepted individuals. These findings call for a more nuanced understanding of aggression—one that gives greater attention to the moderating role of proximity to targets of aggression. Unlike in previous research, accepted and rejected targets interacted face to face with new partners, and the only evidence of “aggression” came in the form of private derogation and relational distancing. This suggests that the rejected can suppress overt displays of aggression when the situation requires that they do so in order to avoid additional rejection.

Furthermore, in line with previous work documenting early stage proximity seeking in response to rejection (Dewall et al., 2009; Gardner et al., 2000; Lakin et al., 2008), our findings revealed evidence of higher LSM and conversational content matching among our rejected compared to accepted participants. However, as Murray, Derrick, Leder, and Holmes (2008) have observed, automatically activated connection goals may occasionally give way to automatic distancing goals in relationships characterized by intense or repeated threat. Specifically, avoidance behaviors stemming from a history of perceived rejection may eventually become automated and hence show stronger activation than connection goals following reminders of threat. By extension, we suspect that the LSM behaviors observed in the present study may be restricted to relationships in which ongoing concerns about rejection are not operating.

Finally, our findings should be interpreted within the context of the rejection manipulation that was employed. Molden, Lucas, Gardner, Dean, and Knowles (2009) showed that “active exclusion” or clear devaluation by others activates a prevention focus, characterized by higher anxiety, regrets over actions taken, and withdrawal from social contact. In contrast, “passive exclusion” or feeling ignored by others activates a promotion focus characterized by increases in sadness and regrets over actions not taken and efforts to increase social contact. Although subtle, our manipulation took the form of active exclusion. A passive exclusion manipulation that merely denied opportunities for connection might have resulted in more proactive social behaviors, including efforts to garner positive impressions (rather than avoid negative impressions) by others. Whereas both active and passive exclusion might be expected to increase automatic (unregulated) proximity seeking, only those who are actively rejected may view novel interaction partners as nonviable sources of reconnection and defensively reduce closeness with them. We view this as a critical area for future research.

Conclusions

Social rejection represents an immediate and painful threat to the belongingness for which humans ordinarily strive (Baumeister & Leary, 1995). Our findings suggest that rejected people exhibit a complex mix of affiliative and distancing behaviors that vary in the extent to which regulatory control is involved. We argue that a fuller understanding of rejection’s outcomes can be achieved by appreciating the multiple means by which people minimize the pain and probability of future rejection.

	F(1, 88)	η²	Accepted M (SD)	Rejected M (SD)
Social expectancies	24.29**	.20	4.95 (.78)	4.02 (.99)
Interpersonal concern	6.07*	.07	3.70 (1.45)	2.97 (1.37)
Affect^a	4.32*	.06	1.25 (.74)	.81 (1.00)

	F(1, 87)	η²	Accepted M (SD)	Rejected M (SD)
Targets’ ratings of partners^a
Kind	7.79**	.08	5.62 (0.87)	5.00 (1.00)
Moody	1.61	.0	1.54 (0.73)	1.73 (0.69)
Self-assured	3.63	.04	4.85 (1.19)	4.31 (1.03)
Critical	0.11	.00	1.87 (.98)	1.91 (0.62)
Partners’ ratings of targets^a
Kind	0.81	.01	5.48 (1.05)	5.31 (0.84)
Moody	0.02	.00	1.61 (0.74)	1.65 (0.85)
Self-assured	1.14	.01	4.60 (1.31)	4.39 (1.12)
Critical	4.06*	.05	1.99 (1.04)	1.67 (0.76)
Targets’ ratings of the interaction^b
Rapport/liking	7.30**	.09	4.06 (0.49)	3.68 (0.73)
Self-consciousness	0.02	.00	2.35 (0.82)	2.34 (0.85)
Personal responsibility	0.51	.01	2.62 (0.73)	2.73 (0.74)
Focus/concentration	13.53***	.16	4.14 (0.64)	3.51 (0.87)
Partners’ ratings of the interaction^b
Rapport/liking	0.00	.00	3.93 (0.69)	3.92 (0.46)
Self-consciousness	3.95*	.05	2.12 (0.73)	2.46 (0.76)
Personal responsibility	1.70	.02	2.54 (0.89)	2.80 (0.81)
Focus/concentration	0.85	.01	4.04 (0.64)	3.89 (0.83)

	Definition/Examples	F(1, 88)	η²	Accepted M (SD)	Rejected M (SD)
Positive emotions	Happy, optimistic	.98	.01	3.11 (1.15)	2.87 (1.38)
Negative emotions	Enemy, worthless	2.88	.03	1.01 (.66)	1.25 (.68)
Anxiety/fear	Afraid, tense	4.09*	.04	0.15 (.25)	0.28 (.32)
Anger	Hate, kill	4.38*	.05	0.13 (.21)	0.23 (.27)
Sadness	Cry, sad	.13	.00	0.12 (.17)	0.11 (.28)
Laugh symbols	Hehe, hehehe	8.58**	.09	5.72 (4.79)	3.74 (2.52)
Questions asked	Number questions asked	4.58*	.05	2.53 (1.15)	3.04 (1.36)

	Definitions/Examples	Accepted Dyads n = 46	Rejected Dyads n = 45	Z
I. Standard dimensions
Word count	Total word count	−.44**	−.18	1.34
Words per sentence	Words per sentence	.20	.48**	1.48
Words >6 letters	Words over 6 letters	.09	.56**	2.50*
First person	I, me, we	.52**	.06	−2.38*
Second person	You	−.24	−.27	−.15
Third person	He, she, they	.33*	.41*	.43
Negations	No, never, not	.30*	.27	−.15
Assents	Yes, OK	−.08	.03	.51
Articles	A, an, the	−.05	.22	1.26
Prepositions	On, to, from	.04	.19	.70
Numbers	One, thirty, million	.35*	.53**	1.04
II. Psychological processes
Immediacy	(Zi + Zdiscrep + Zpresent) − (Zarticle + Zsixltr)	.18	.59***	2.28*
Making distinctions	(Zdiscrep + Zexcl + Ztentat + Znegate) − ZincL	.04	.31*	1.29
Social past	(Zsocial + Zpast) − (Zpresent + Zposemo)	.43**	.43**	0.00
Rationalization	(Zcause + Zinsight) − (Znegemo)	.04	.40**	1.77
Emotional positivity	Zposemo − Znegemo	.17	.45**	1.44
III. Relativity
Time	Hour, day, oclock	.19	.31*	.62
Space	Around, over, up	.43**	.46**	.18
Motion	Walk, move, go	.45**	.58***	.85
IV. Personal Concerns
Occupation	School, job, goal	.43**	.67***	1.62
Leisure activity	Home, sports, music	.36*	.60***	1.46
Money/Financial Issues	Cash, income, money	.40**	.63***	1.46

	Immediate Reactions to the Manipulation
Targets’ ratings of partners
Kind	.53**	.39**	.23
Partners’ ratings of targets
Critical	.08	.04	−.02
Targets’ ratings of the interaction
Rapport/liking	.46**	.20	.28*
Focus/concentration	.29*	.31**	.10
Partners’ ratings of the interaction
Self-consciousness	−.19	.01	−.12
Language use
Anxiety/fear	−.05	−.19	.04
Anger	−.29**	−.24*	−.09
Laugh symbols	.11	.13	.05
Questions asked	−.98	−.02	−.30*

	Mediators Tested	Indirect Effect	SE	Lower Limit for 95% CI	Upper Limit for 95% CI
Target outcomes
Kind	Social expectancies	−.370	.134	−.772	−.216
Interpersonal concern	−.094	.062	−.285	−.008
Rapport/liking	Social expectancies	−.244	.115	−.524	−.094
Affect	−.030	.052	−.141	.067
Focus/concentration	Social expectancies	−.055	.114	−.402	.064
Interpersonal concern	−.087	.075	−.302	.009
Anger language	Social expectancies	.040	.031	−.014	.108
Interpersonal concern	.016	.015	−.003	.057
Questions asked	Affect	.169	.132	.010	.568

Footnotes

Acknowledgment

The authors thank Alissa Czajka, Nicholas Reyna, Ben Barkowski, Alyssa Raymond, and Jennifer Ochoa-Evers for their assistance with data collection and coding.

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 research was supported by a National Institute of Mental Health grant (#MH066828-01) awarded to the first author.

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. doi:10.1037/0033-2909.117.3.497

Bernieri

F. J.

Rosenthal

(1991). Interpersonal coordination: Behavioral matching and interactional synchrony. In Feldman

Rime

(Eds.), Fundamentals of nonverbal behavior (pp. 401–432). New York, NY: Cambridge University Press.

Bernstein

M. J.

Claypool

H. M.

(2012). Social exclusion and pain sensitivity: Why exclusion sometimes hurts and sometimes numbs. Personality and Social Psychology Bulletin, 38, 185–196. doi:10.1177/0146167211422449

Cassidy

Kobak

R. R.

(1988). Avoidance and its relationship with other defensive processes. In Belsky

Nezworski

(Eds.), Clinical implications of attachment (pp. 300–323). Hillsdale, NJ: Erlbaum.

Cohn

M. A.

Mehl

M. R.

Pennebaker

J. W.

(2004). Linguistic markers of psychological change surrounding September 11, 2001. Psychological Science, 15, 697–683. doi:10.1037/0022-3514.91.1.1

DeWall

C. N.

Baumeister

R. F.

(2006). Alone but feeling no pain: Effects of social exclusion on physical pain tolerance and pain threshold, affective forecasting, and interpersonal empathy. Journal of Personality and Social Psychology, 91, 1–15. doi:10.1037/0022-3514.91.1.1

DeWall

C. N.

Maner

J. K.

Rouby

D. A.

(2009). Social exclusion and early-stage interpersonal perception: Selective attention to signs of acceptance. Journal of Personality and Social Psychology, 96, 729–741. doi:10.1097/PSY.0b013e3182464dd1

Eisenberger

N. I.

Lieberman

M. D.

Williams

K. D.

(2003). Does rejection hurt? An fMRI study of social exclusion. Science, 302, 290–292. doi:10.1126/science.1089134

Gardner

W. L.

Pickett

C. L.

Brewer

M. B.

(2000). Social exclusion and selective memory: How the need to belong influences memory for social events. Personality and Social Psychology Bulletin, 26, 486–496. doi:10.1177/0146167200266007

10.

Gonzales

A. L.

Hancock

J. T.

Pennebaker

J. W.

(2010). Language style matching as a predictor of social dynamics in small groups. Communication Research, 37, 3–19. doi:10.1177/0093650209351468

11.

Griskevicius

Goldstein

N. J.

Mortensen

C. R.

Cialdini

R. B.

Kenrick

D. T.

(2006). Going along versus going alone: When fundamental motives facilitate strategic (non)conformity. Journal of Personality and Social Psychology, 91, 281–294. doi:10.1037/0022-3514.91.2.281

12.

Hadjistavropoulos

Craig

D. D.

(2002). A theoretical framework for understanding self-report and observational measures of pain: A communications model. Behavioral Research and Therapy, 40, 551–570. doi:10.1016/S0005-7967(01)00072-9

13.

Hayes

A. F.

(1998). Within-study meta-analysis: Pooling the significance of doubly nonindependent (nonoverlapping) correlations. Psychological Methods, 3, 32–45. doi:https://dx-doi-org.web.bisu.edu.cn/10.1037/1082-989X.3.1.32

14.

Hess

Y. D.

Pickett

C. L.

(2010). Social rejection and self-versus other-awareness. Journal of Experimental Social Psychology, 46, 453–456. doi:10.1016/j.jesp.2009.12.004

15.

Ireland

M. E.

Pennebaker

J. W.

(2010). Language style matching in writing: Synchrony in essays, correspondence, and poetry. Journal of Personality and Social Psychology, 99, 549–571. doi:10.1037/a0020386

16.

Knowles

M. L.

Gardner

W. L.

(2008). Benefits of membership: The activation and amplification of group identities in response to social rejection. Personality and Social Psychology Bulletin, 34, 1200–1213. doi:10.1177/0146167208320062

17.

Kross

Berman

M. G.

Mischel

Smith

E. E.

Wager

T. D.

(2011). Social rejection shares somatosensory representations with physical pain. Proceedings of the National Academy of Sciences (PNAS) 108, 6270–6275. doi:10.1073/pnas.1102693108

18.

Lakin

J. L.

Chartrand

T. L.

(2005). Exclusion and nonconscious behavioral mimicry. In Williams

K. D.

Forgas

J. P.

Von Hippel

(Eds.), The social outcast: Ostracism, social exclusion, rejection, and bullying (pp. 279–295). New York, NY: Psychology Press.

19.

Lakin

J. L.

Chartrand

T. L.

Arkin

R. M.

(2008). I am too just like you: Nonconscious mimicry as an automatic behavioral response to social exclusion. Psychological Science, 19, 816–822. doi:10.1111/j.1467-9280.2008.02162.x

20.

MacDonald

Leary

M. R.

(2005). Why does social exclusion hurt? The relationship between social and physical pain. Psychological Bulletin, 131, 202–223. doi:10.1037/0033-2909.131.2.202

21.

Main

(1990). Cross-cultural studies of attachment organization: Recent studies, changing methodologies, and the concept of conditional strategies. Human Development, 33, 48–61. doi:10.1159/000276502

22.

Maner

J. K.

DeWall

C. N.

Baumeister

R. F.

Schaller

(2007). Does social exclusion motivate interpersonal reconnection? Resolving the 'porcupine problem. Journal of Personality and Social Psychology, 92, 42–55. doi:10.1037/0022-3514.92.1.42

23.

Maxwell

J. A.

Spielmann

S. S.

Joel

MacDonald

(2013). Attachment theory as a framework for understanding responses to social exclusion. Social and Personality Psychology Compass, 7, 444–456. doi:10.1111/spc3.12037

24.

Mikulincer

Shaver

P. R.

(2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 35, pp. 53–152). San Diego, CA: Elsevier Academic Press. doi:10.1016/S0065-2601(03)01002-5

25.

Mikulincer

Shaver

P. R.

(2011). Attachment, anger, and aggression. In Shaver

P. R.

Mikulincer

(Eds.), Human aggression and violence: Causes, manifestations, and consequences (pp. 241–257). Washington, DC: American Psychological Association.

26.

Molden

D. C.

Lucas

G. M.

Gardner

W. L.

Dean

Knowles

M. L.

(2009). Motivations for prevention or promotion following social exclusion: Being rejected versus being ignored. Journal of Personality and Social Psychology, 96, 415–431. doi:10.1037/a0012958

27.

Molden

D. C.

Maner

J. K.

(2013). How and when social exclusion motivates social reconnection. In Dewall

C. N.

(Ed.), The Oxford handbook of social exclusion (pp. 121–131). New York, NY: Oxford University Press.

28.

Moskowitz

D. S.

M. R.

Turcotte-Tremblay

(2007). Contextual influences on interpersonal complementarity. Personality and Social Psychology Bulletin, 33, 1051–1063. doi:10.1177/0146167207303024

29.

Murray

S. L.

Derrick

J. L.

Leder

Holmes

J. G.

(2008). Balancing connectedness and self-protection goals in close relationships: A levels-of-processing perspective on risk regulation. Journal of Personality and Social Psychology, 94, 429–459. doi:10.1037/0022-3514.94.3.429

30.

Niederhoffer

K. G.

Pennebaker

J. W.

(2002). Linguistic style matching in social interaction. Journal of Language and Social Psychology, 21, 337–360. doi:10.1177/026192702237953

31.

Owen

J. E.

Giese-Davis

Cordova

Kronenwetter

Golant

Spiegel

(2006). Self-report and linguistic indicators of emotional expression in narratives as predictors of adjustment to cancer. Journal of Behavioral Medicine, 29, 335–345. doi:10.1007/s10865-006-9061-8

32.

Péloquin

Lafontaine

Brassard

(2011). A dyadic approach to the study of romantic attachment, dyadic empathy, and psychological partner aggression. Journal of Social and Personal Relationships, 28, 915–942. doi:10.1177/0265407510397988

33.

Pennebaker

J. W.

Francis

M. E.

Booth

R. J.

(2001). Linguistic inquiry and word count: LIWC2001. Mahwah, NJ: Erlbaum Publishers.

34.

Pennebaker

J. W.

King

L. A.

(1999). Linguistic styles: Language use as an individual difference. Journal of Personality and Social Psychology, 77, 1296–1312. doi:10.1037/0022-3514.77.6.1296

35.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891. doi:10.3758/ BRM.40.3.879

36.

Rogers

R. W.

Prentice-Dunn

(1981). Deindividuation and anger-mediated interracial aggression: Unmasking regressive racism. Journal of Personality and Social Psychology, 41, 63–73. doi:10.1037/0022-3514.41.1.63

37.

Sedikides

Campbell

W. K.

Reeder

G. D.

Elliot

A. J.

(1999). The relationship closeness induction task. Representative Research in Social Psychology, 23, 1–4.

38.

Shaver

P. R.

Mikulincer

(2002). Attachment-related psychodynamics. Attachment and Human Development, 4, 133–161. doi:10.1080/14616730210154171

39.

Smart Richman

Leary

M. R.

(2009). Reactions to discrimination, stigmatization, ostracism, and other forms of interpersonal rejection: A multimotive model. Psychological Review, 116, 365–383. doi:10.1037/a0015250

40.

Sommer

K. L.

Benkendorf

(2009). Looking before leaping: The role of social expectancies in attachment regulation following interpersonal rejection. In Harris

M. J.

(Ed.), Bullying, rejection, and peer victimization: A social cognitive neuroscience perspective (pp. 225–250). New York, NY: Springer Publishing Company.

41.

Tickle-Degnen

Rosenthal

(1987). Group rapport and nonverbal behavior. In Clyde

(Ed.) Group processes and intergroup relations. Review of personality and social psychology (Vol. 9., pp. 113–136). Thousand Oaks, CA: Sage.

42.

Twenge

J. M.

Baumeister

R. F.

Tice

D. M.

Stucke

T. S.

(2001). If you can’t join them, beat them: Effects of social exclusion on aggressive behavior. Journal of Personality and Social Psychology, 81, 1058–1069. doi:10.1037/0022-3514.81.6.1058

43.

Twenge

J. M.

Cantonese

K. R.

Baumeister

R. F.

(2003). Social exclusion and the deconstructed state: Time perception, meaninglessness, lethargy, lack of emotion, and self-awareness. Journal of Personality and Social Psychology, 85, 409–423. doi:10.1037/0022-3514.85.3.409

44.

Twenge

J. M.

Zhang

Catanese

K. R.

Dolan-Pascoe

Lyche

L. F.

Baumeister

R. F.

(2007). Replenishing connectedness: Reminders of social activity reduce aggression after social exclusion. British Journal of Social Psychology, 46, 205–224. doi:10.1348/014466605X90793

45.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. doi:10.1037/0022-3514.54.6.1063

46.

Williams

K. D.

(2007). Ostracism. Annual Review of Psychology, 58, 425–452. doi:10.1146/annurev.psych.58.110405.085641

47.

Williams

K. D.

(2009). Ostracism: A temporal need-threat model. In Zanna

(Ed.), Advances in experimental social psychology (Vol. 41, pp. 279–314). New York, NY: Academic Press.

48.

Williams

K. D.

Cheung

C. K. T.

Choi

(2000). Cyberostracism: Effects of being ignored over the Internet. Journal of Personality and Social Psychology, 79, 748–762. doi:10.1037/0022-3514.79.5.748

	Immediate Reactions to the Manipulation
	Expectancies	Concern	Affect
Targets’ ratings of partners
Kind	.53**	.39**	.23
Partners’ ratings of targets
Critical	.08	.04	−.02
Targets’ ratings of the interaction
Rapport/liking	.46**	.20	.28*
Focus/concentration	.29*	.31**	.10
Partners’ ratings of the interaction
Self-consciousness	−.19	.01	−.12
Language use
Anxiety/fear	−.05	−.19	.04
Anger	−.29**	−.24*	−.09
Laugh symbols	.11	.13	.05
Questions asked	−.98	−.02	−.30*