Communication Modes During Romantic Dissolution: The Impact of Attachment and Intimacy on Initiator Breakup Strategies

A majority of romantic pair-bonds will not remain together. Surprisingly, however, less is known about relationship dissolution compared with other stages of romantic relationships, such as initiation and maintenance. The present study addresses this gap by investigating breakup initiators’ communication strategies as outcomes for the individual’s attachment style and the emotional intimacy of a recent terminated romantic relationship. Participants (N = 174) completed a series of empirically reliable and valid Likert-scale measures to assess both predictors: (a) adult attachment and (b) emotional intimacy. Emotional intimacy was a significant predictor of each of the four breakup communication strategies, and adult attachment style – particularly attachment anxiety was significantly related to specific communication strategies, primarily those related to openness. This researcher concluded that emotional intimacy better predictors of breakup communication strategy compared with attachment style. Limitations and a general discussion or summary of findings are provided, followed by some suggestions for future research.

. Recent studies also consider the role of social networking sites (e.g., Facebook) as new ways of communicating and navigating one's emotional state following a breakup (e.g., "unfriending" someone on Facebook) (Blight et al., 2019). Niemyjska (2019), examining inanimate objects synonymous with a former partner, revealed that adults with real situations of separation have far more opportunities to guide their attachment to objects, resulting in diminished feelings of isolation. As stated above, I test two independent variables to predict which strategies might occur. The first of these, attachment style, occurs at the individual level and will be reviewed next.

Attachment Theory
Attachment is defined as the human inclination to form close bonds with specific others that provide fundamental needs, such as nourishment, security, and comfort (Bowlby, 1977). Attachment theory is a theory of human behavior organized around the claim that humans are motivated to seek proximity from caregivers, beginning in infancy and continuing across the lifespan (Bowlby, 1982;Mikulincer & Shaver, 2007). Ainsworth, Blehar, Waters, and Wall (1978) developed the descriptive anxious, avoidant, and secure classifications of attachment styles in infants. Anxious infants would respond in protest when distressed or separated from the primary caregiver, while avoidant infants did not seek out the primary caregiver when distressed. Secure infants sought the caregiver figure as a secure base when distressed. Both cognitive and affective representations about social life and relationships emerge from these early experiences (Simpson, 1990). Thus, while our first associations with attachment figures are shaped early in life, one's attachment style can continue into adult life (Pizzano, Sherblom, & Umphrey, 2013).

Adult attachment -The fundamental need to establish bonds with an attachment figure in infancy generated
interest in the role attachment experiences, both cognitive and affective, play in adult relationships (Cowan et al., 2019;Vladislav & Bucur, 2012). Hazan and Shaver (1987) developed a measure of adult attachment styles in romantic relationships. Using the framework from Ainsworth et al. (1978), they were able to develop a study for adult populations that matched the behaviors of infants and children in previous studies. Participants were considered secure, anxious, or avoidant (Hazan & Shaver, 1987). The secure attachment style indicates that individuals are comfortable being close to others. Avoidant attachment style suggests that a person struggles to become close to and trust in others, while the anxious attachment style features greater worry regarding their relationship partner's availability, responsiveness, and involvement (Mickelson et al., 1997).
Attachment in close relationships -Attachment style is posited as a predictor of interaction patterns in social situations (Collins & Gillath, 2012). Scholars have noted a great deal of similarity in peer relationships during early childhood and romantic partners in adulthood (Fraley & Davis, 1997;Hazan & Zeifman, 1994).
In a sample of college students, 55% specified his or her romantic partner as the primary attachment figure (Pitman & Scharfe, 2010). Recruiting a participant sample that ranged from 16-90, Doherty and Feeney (2004) found that 74% of adult respondents designated a romantic partner as her or his attachment figure.
Further, 96% reported having an attachment figure, and of this percentage, romantic partners were ranked first, followed by mothers, friends, children, siblings, and fathers. In a review from Simpson (1990), romantic couples reporting secure attachment had higher levels of reported trust and interdependence, while avoidant individuals experienced fewer 'intense' (e.g., joyful) and 'mild' (e.g., satisfied) emotions. In an innovative observation study, Fraley and Shaver (1998) had a team of coders observe couples separating at an airport. Their conclusions suggested that couples not flying together "engaged in behaviors functionally similar to those exhibited by Hoffman children separating from their attachment figure" (Fraley & Shaver, 1998, p. 1202. As Gillath, Karantzas, and Fraley (2016) stated, though, attachment is both "theory of love, emotional connection, and psychological well-being" (p. 3), as well as a theory of grief and loss (see Bowlby, 1982). Romantic breakup research has examined attachment as a predictor of behavioral elements in couples dissolving their partnership (Bartell, 2006;Cate, Levin, & Richmond, 2002;Davis et al., 2003). In Segrin and Flora (2016), they provided support that a relationship exists between self-esteem in adulthood and experiences in family of origin. Bowlby (1973) explained that a persons' experience with her/his primary attachment figure constructs a cognitive and affective schema for interpreting social interaction (e.g., trusting another person). According to Mickelson, Kessler, and Shaver (1997), these shape the "beliefs about whether the self is worthy of love and support and also influence the kinds of interactions individuals have" (p. 1092). Communication styles are thought to model attachment orientations in close relationships (Lambert & Hughes, 2010). Collins and Gillath (2012) found a positive correlation between adult attachment style and the selection of disengagement communication messages. Davis et al. (2003) reported the anxious attachment was associated with a greater tendency to seek a new relationship, as well as higher likelihood that the person engages in unwanted pursuit of the former partner (Langhinrichsen-Rohling, 2010). Avoidant attachment style has been associated with higher levels of personal blame for the dissolution (Davis et al., 2003) as well as lower levels of stress following the breakup (Feeney, 2005). However, the role of attachment in romantic breakups, as well as with romantic relationships in general, has received inconsistent support. Simpson (1990) found no significant association between distress following a breakup in anxious or secure attachment styles between both females and males. Only attachment avoidant males had significantly lower levels of distress. A meta-analysis from Le et al. (2010) discovered that reported levels of closeness, inclusion, and dependence were better predictors of a breakup, but found some support for attachment orientation. From here, I turn to intimacy levels as a relational (rather than the individual attachment style) predictor of breaking up, and specifically the communication style during a breakup.

The Role of Intimacy
Intimacy is a difficult term to express in the social sciences (Reis & Shaver, 1988). Intimacy has been defined as an interpersonal state that validates one's self-worth (Sullivan, 1953), a state of interpersonal openness, and the activity and outcomes of self-disclosure and reciprocity between individuals (Collins & Sroufe, 1999). Popovic (2005) noted that intimacy represented interdependence, a merging between individual and relational selves. Individual and relational needs, as well as expectations, are revealed as intimacy is established.
According to Hazan and Shaver (1994), children's perceived intimacy with a parent or primary caregiver is transferred to adulthood.
In a longitudinal study, Collins et al. (1997) found a positive correlation between early childhood attachment with parents and adolescent attachment style in teenage dating relationships, specifically measuring individual orientation regarding intimacy. In adulthood, romantic relationship satisfaction and quality has been positively correlated with self-esteem (Segrin & Flora, 2014). Subsequent research discovered that individual associations regarding romantic relationships are formed early on and have the capacity to transfer into close relationships when they reach adulthood (Cooper et al., 2004;Cui & Fincham, 2010). The role of intimacy in relationship stability and satisfaction is well established (Birnie-Porter & Hunt, 2015; Reis & Patrick, 1996;Segrin & Flora, 2000). For example, disclosure of personal information infers that trust is established, but the partner's reaction is key to building a continually high level of closeness. The level of partner responsiveness Communication Modes During Romantic Dissolution is highly relevant in communication between partners . In the breakup context, Banks et al. (1987) concluded that a high level of intimacy between partners was correlated with direct disengagement communication, while low intimacy responses were associated with an indirect communication style. Moreover, higher intimacy has been associated with the use of positive-tone communication during a breakup, while withdrawal communication is less likely to be used. In this review, I introduced the outcome variable (breakup strategy) and predictors (attachment; intimacy). To examine their association(s), the following research ques- H 2b : Subsequently, low emotional intimacy and high attachment avoidance will be significantly associated with indirect and withdraw avoidant strategies RQ 1 : Are there differences related to sex in communication styles to end a relationship?

Method Sample
Participants (n = 174) were college students recruited from an introductory communication studies course at a public university in the Midwest region of the United States. Each participant reported their age (the requirement was that all participants must be eighteen years of age or older) (M age = 19.79; SD = 2.96) as well as sex (100 females, 75 males; 1 non-binary). All participants were the agent (i.e., leaver) of their romantic breakup. Participants also reported the length of their romantic relationships in months (M months = 12.62; SD = 9.84). Individuals who had been in a romantic relationship at some point in the previous 12 months and experienced a romantic breakup continued the survey (M months = 6.26; SD = 3.51).

Procedure
To test the predictions and answer research question one, a survey method was chosen. Participants clicked on a website link to begin the survey. The survey was created using Qualtrics, a survey tool for research. This survey began with an overview of the research goals, and indicated their rights as participants (i.e., a consent form). After providing consent to participate, the survey will begin with four questions regarding their romantic relationship history. Disengagement strategies were classified following prior work conducted by Baxter (1982), Cody (1982), and Sprecher et al. (2010). Categories were (1) direct, (2) withdraw depart, (3) positive tone, and (4) indirect. After completing the disengagement strategies items, participants completed two additional measures. The first measure assessed attachment style, while the second assessed emotional Hoffman intimacy. Participants were awarded extra credit points for completing the survey. The final section asked for participant demographics, and the entire survey took approximately ten to fifteen minutes.

Major Measures Breakup Communication Strategies
In this study, breakup strategies are the dependent variable (DV), which I test for in accordance with individual attachment styles and relational intimacy. In the current study, I adopted items from Sprecher et al. (2010).
These items were an updated version of an earlier measure from Baxter (1982) The initial study using these items found acceptable reliability ranging from .63 to .82 in their first study, and .75 to .91 in the second (Sprecher et al., 2010).

Adult Attachment
Adult attachment is an individual level independent variable. The experiences in close relationships-revised (ECR-R) questionnaire (Fraley, Waller, & Brennan, 2000) will be used to assess adult attachment style. The ECR-R is a 36-item measure of adult attachment, in which the first 18 items assess anxiety (e.g., I worry a lot about my relationship), while items 19 to 36 assess avoidance (e.g., I feel comfortable sharing my private thoughts and feelings) attachment style. Items were arranged on a Likert scale from 1 (strongly disagree) to 7 (strongly agree). Items 1 to 18 assessed attachment anxiety, while items 19 to 36 measured attachment avoidance. Lower scores will signify a secure attachment style. In research from Sibley and Liu (2004), strong reliability was established through Cronbach α ratings of .94 (anxiety) and .93 (avoidance). The current research reported a .89 α score altogether.

Emotional Intimacy
Emotional intimacy is weighed in relation to attachment style and the communication strategy decisions reported by participants. To measure intimacy, I used the personal assessment of intimacy in relationships (PAIR) scale (Schaefer & Olson, 1981). The PAIR measure is a 36-item scale with five intimate factors: emotional, social, sexual, intellectual, and recreational. For this study, the six items from the emotional intimacy subscale were adopted (e.g., my partner listens to me when I need to talk to someone; I can state my feelings without him/her getting defense). Items were arranged on a Likert scale ranging from 1 (does not describe/my relationship at all) to 5 (describes me/my relationship very well), with higher scores attributed to greater emotional intimacy. Gable et al., (2004) reported high reliability using the PAIR measure in their study of cross-sex married couples (α = .82 for women; α = .86 for men). In the present study, the measure's internal consistency reached an acceptable level (α = .78).

H 1 and H 2 : Attachment and Intimacy on Communication Strategy
Pearson correlations were run for all predictor variables (attachment anxiety, attachment avoidance, intimacy) on each of the four communication styles, independent (direct-open, avoidant-withdraw, positive-tone, indirect-closed). In summary, these results (Table 1) indicate that attachment anxiety is positively correlated with indirect closed and withdraw avoidance breakup communication strategies. Furthermore, there was a significant negative correlation between attachment anxiety and direct-open communication. Table 1 also shows that attachment avoidance was negatively correlated with direct-open communication. Results also indicated a positive correlation between attachment avoidance and withdraw depart strategies. As per Table 1, the level of emotional intimacy was significantly correlated with each of the four breakup strategies. In the next phase of this study, a series of regression analyses were performed to examine these relationships.

Regression Analyses
Predictor variables included in the multiple linear regression were attachment anxiety and intimacy (see Table   2, Model 1), and attachment avoidance and intimacy (Table 2, Model 2). In the multiple regression, individual-level attachment styles were tested with the relational variable of intimacy level; that is, attachment with intimacy, and avoidant with intimacy, but not attachment styles together. Multiple regression analyses were run in the IBM SPSS version 25 program. Each dependent variable was tested on attachment anxiety and intimacy (Table 2, Model 1), as well as attachment avoidance and intimacy (Table 2, Model 2). To summarize these findings, I will go through each dependent variable in Model 1, then Model 2.

Attachment Avoidance and Emotional Intimacy
The same standardized data for intimacy was used here (see Table 4), but attachment avoidance was assessed instead of attachment anxiety. Intimacy level, β = 0.160, t = 2.11, p = .036, was positively asso-

Sex Differences and Breakup Communication
To address research question two, an independent-samples t-test showed that women (N = 99) and men (N = 75) differed in their likelihood to use direct-communication strategies. While one individual identified with the non-binary option, the limited number of participants from this sex/gender identity cannot be conclusively studied due to the sample size. The discussion section will suggest future studies elaborate on the role of sex and gender, however. As shown in Table 5 Table 6 are in line with previous findings showing lack of significant difference between women and men (Le et al., 2010). Future work should build on this by including additional factors (e.g., conflict styles) and diversifying the composition of their research sample (e.g., age range).

Discussion
The goal of this study was to examine and predict communication strategies used to end a romantic relationship according to the individual (attachment) and relational (intimacy) level variables. Results indicated that relational intimacy is an important factor infkuencing communication strategies to break up with a romantic partner. Those who had more intimate relationships were more likely to use direct-open and positive-tone communication strategies, and less likely to report using avoidant-withdrawal and indirect-closed strategies.
The association between direct-open communication is in line with prior findings (see Sprecher et al., 2014).
While secure attachment style is not measured in the ECR-R scale (Fraley & Shaver, 2000), individuals reporting higher scores on relational intimacy have shown to be more securely attached (Madey & Jilek, 2012;Madey & Rodgers, 2009 Stafford and Canary (1991) discovered that males were using more maintenance strategies than females. Using the Stafford and Canary (1991) typology, Ragsdale (1996) showed that women used positivity, openness confidence, connectivity, and tasks more often than men did, but men and women used the strategies in the same order of occurrence.  (Davis et al., 2003), the current study showed that intimacy, specifically emotional intimacy, was a better predictor of breakup strategy. It is important to understand if attachment style is connected with the communication strategy to end a romantic relationship. Regarding intimacy, measuring this construct beyond emotional closeness will provide more robust findings. Further, interpersonal scholars should study same-sex couples going through the breakup stage; studying multiple sexual orientation groups could further provide an important series of findings for the role of social support levels and coping following the breakup (Goldberg & Allen, 2013;Menees, 1997). Finally, researchers with qualitative and mixed-methods training should collect data from interviews and focus groups about the communication and behavioral patterns that precede a breakup.
Other possible explanation should not be overlooked. For example, the results are potentially explained further by individual initiators' memory faculties. There could be a tendency for participants to revise or modify the narrative of their former relationship a posteriori simply due to the passage of time. While this study examined dissolution episodes less that occurred less than one year prior, the variability of memories and our need for cognitive consistency should be tested as a possible explanation (see MacLeod & Saunders, 2008). The retrospective nature of this study poses a constraint that can be expanded on for future research. Moreover, researchers can regularly update retrospective studies steps of this kind by considering the mediums and relationship norms of communication and social and psychological factors that shape relationship dynamics.