Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Attachment, accommodation, and the outcomes of romantic relationships
(USC Thesis Other)
Attachment, accommodation, and the outcomes of romantic relationships
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
ATTACHMENT, ACCOMMODATION, AND THE OUTCOMES OF ROMANTIC RELATIONSHIPS by Erica Watson-Currie A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (COMMUNICATION) May 2004 Copyright 2004 Erica Watson-Currie R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. UMI Number: 3140569 Copyright 2004 by Watson-Currie, Erica All rights reserved. INFORMATION TO USERS The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. UMI UMI Microform 3140569 Copyright 2004 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. EPIGRAPH Inspiration to stay wi& a dissertation project can sometimes be hard to find. To anyone who bothers to read mine - either as a source or a model - 1 wish yon the very best in your own endeavor. On the w ai of my office, I had a buHetin board solely devoted to pictures, poems, quotations, and stories. Along with the story of “The Rabbit, The Fox, and The Wolf^’’ which every graduaie student knows, and Ann Lamott’s Bird bv Bird, here are the three I found most useful: Dream your own dream and follow your own star. There wouldn’t be a heaven &li of stars if we were all meant to wish on the same one. There will always be dreams grander and humbler than your own. But there will never be a dream exactly like your own, for you are unique and more wondrous than you know! Do your best There are no shortcuts on the way to a dream. So give all that you have to all that you do. And above all things believe in yourself. This is all your dream asks of you, but this is everything. Linda Staten “What does not destroy me, makes me stronger.” - Friedrich Nietzsche “Keep it simple, stupid!” - Author Unknown R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. m ACKNOWLEGEMENTS First aid foremost, I would like to thank my committee members; Dr. Lynn M ilte, Dr. Michael Cocfy, and Dr. M tch Earleywine for their time and guidance. Special acknowledganent also must made to Dr. Sheila Murphy, Dr. Margaret M cLau^in, and Dr. Colleen Keough for their support, encouragement, and inspiration along the way. I would furdier like to single out my advisor, Lynn M iier, for h C T commitment to holding me to the highest standards throughout my research; and Michael Cody for being everything one could wish for in a mentor throughout my doctoral journey. I feel fortunate to have had the opportunity to be in the company of my fellow graduate students, my colleagues and fiiends i^ o in^ired and challenged me iiroughoiit my time at Annenberg. Also, my enduring gratitude to Aime Marie Campian for everything! Personal thanks are also oveidue to the myriad friends and family members who Ve supporteci my quest, even when it seemed an impossible dream to me: To my mother, Patricia W at»n, who always ta u ^ t me to push ahea4 stay conmiitted to my ideals, and triumph over any adversity that got in the way; To my father. Dr. George Watson, who alw ^s believed in me, and never let me forget that I could do this; To Jeri aid Kaflierine Leibson, for their oagoing support and enfluisi^m, as well as for putting me in touch with Dr. Bob Ferguson at Saddleback College, whose generosity in making his students available to participate in my research surveys made a vital difference to my gathering of data. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. w Mere wonfe on paper are insufficient to express my gratitude to my wonderful husband, Victor, for Ms positivity and tmderstanding, and for reminding me to sometimes take my eyes off the goal to look back at how far I’d come in its pursuit. And, finaly, I am forever obliged to Nicholas Charles Cuirie for reminding me what tMs research is truly all about, for Ms laughter and hugs wMch inspire me, and esiwially for showing me how qiiiddy time was passing. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. TABLE OF CONTENTS EPIGRAPH ii ACKNOWLEGEMENTS ii LIST OF TABLES vil LIST OF FIGURES viii ABSTRACT ix CHAPTER 1 : INTRODUCTION 1 Statement of the Problem 2 Purpose and Significance of the Study 3 Preview of Remaiiiiiig Chapters 4 CHAPTER 2: REVIEW OF RELATED LITERATURES AND HYPOTHESES 6 Atiaciuneiit Theory 6 Attachment Styles and Internal Working Models 7 Attachment and Internal Working Models in Adultlioi^ 9 Internal Working M c ^ s and Betevior 11 Accommociation 14 Hypothesis on AttacIuBeiit and Accommodation 17 Relationship Outcomes 22 Love 22 Equity 23 AttaciniieDt and Romantic Outcomes 24 Accommodation and Love 25 Integratiag Attachment, Aaxsminodation, and Ckitoomes 26 Reactive RMS & Relationship Outcomes; The Maliational Model 29 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. VI CHAPTER 3: METHODOLOGY 30 Overview 30 Development of Survey Instmineiit 30 Att^hment 31 Reactive RMS/ELVN Items. 31 Relationship Outcome (Love) 32 Equity 32 Demogrsqjhics 33 Current Romantic Status 33 Relationship Length 33 Culture 34 Cjender 34 Age 34 Sampling Plan 35 CHAPTER 4: RESULTS 37 Preliminary Aiwlyses and Demographics 37 Relatior^p Status 39 Relationship Length 41 Survey Type 42 Hypothesis 1 44 Hypothesis 2 47 Hypothesis 3 48 Hypothesis 4 49 CHAPTER 5: DISCUSSION 52 Review of Findings and Implications 52 Limitatioiis 55 Future Directions 58 Conclusion 60 REFERENCES 62 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. va LIST OF TABLES Table 1: Correlations between Jntemal Working Models and Attachment Styles 39 Table 2; Attachment Style by RelationsMp Status.......................................... ..........40 Table 3: Attachment Styles, Mtemal Working Models, Behavioirs, and Outcomes by Romantic Involvement .......... .......40 Table 4: Attadment S^le, Mtanal Working Models, Accommocktion, and Outcomes by Relationship Length ..... 42 Table 5: Attachment Style, Internal Woridng Models, Accommodation, and Outcomes by Type of Survey ........ ..43 Table 6: Responses to Conflict by Attachment Style and Internal Working Models...46 Table 7: Relationship Outcomes by Attachment Style and Internal Working Models47 Table 8: Relationship Outcomes by Response to Conflict ........... .....48 Table 9: Results of Hierarchical Regressions testing Hypothesized Mediational Model ...... 50 Table 10: Attachmait Style by Romantic Involvement ................................ ..59 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. VMl LIST OF FIGURES Figuie 1: Integration of 4-Categoiy Models ............... ...7 Figure 2: Adult Accommodation and Childhood Attachment Behaviors 14 Figure 3: AccomincKiative ResfMjnses to Conflict ........ 16 Figure 4: Hypotiiesized Effect of Attachment on Accommodation.................... 19 Figure 5: Mediational Model ilustrating how Attachment affects Relationship Outcomes ................................ 29 R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. IX ABSTRACT Synthesizing research findings from Attachment Theory, Relationship Maintenance Strategies, Accommodative Responses to Conflict, and Romantic Love this study Invesli^ed the role communicatioii behaviors play in the way atjacluiien.t styles affect the quality and longevity of romantic partoerships. Students enrolled at a commiinity college completed self-rqprt measures of attachment style, r e f u s e to conflict, and feelings of love within romantic relationships (N=192); additionally, participants (N=126) were also recruited to participate in an identical online version of the survey, in order to increase the diversity of our sample. Results revealed that whoeas secure attachment increase fevorable responses to conflict (i.e., Voice and Loyally), and reduces damaging ones (i.e.. Exit and Neglect); insecure attachment precipitates enactment of behaviors that are less constructive and more destructive to the relationship. Furthermore, the active behaviors function either to enhance or reduce feelings of love (i.e., commitment, intimacy, passion) within romantic partnerships. This pattern of findings was strongest for the Secure and Dismissive profiles, as well as for the underlymg Intemal Working Model of Anxiety about abandonment. Additionally, these findings were moderated by relationship length: the strongest effects being found for those currently involved in serious and committed romantic relationships of less than two years duration. TMs study provides strong evidence in favor of viewii^ relationship maintenance strategies as attachment R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. t5ehaviors mediating the relalionsMp between intemal working models and love witMn romantic relationships. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1 CHAPTER 1: INTRODUCTION Aristotle, Plato, Kieikegaard, Darwin and others have famously observed: "M an is a social animal.” Social relationships often have repercussions on both physical and mental healfti (see review by Kiecolt-Glasser, 1999): Maintaining satisfactory social ties has time and again been demonstrated to predict greater physical and mental health, overall well-being, and longevity; whereas, possessing relatively fewer social ties, and having greater relationship conflict, correlate to negative physical and mental hadth outcomes. The patterns of links between social relationships and health outcomes are especially pronounced for intimate relationships, such as the relattonsMps between parents and their children, and between romantic partners. For example, accounts of warm childhood relationships with parents predict greats' lifelong health; whereas, cold and distant relations predict greater likelihood of illnesses (e.g., heart conditions, ulcere, alcoholism) at midlife (Russek & Schwartz, 1997). And, arguments between husbands and wives affect both spouses’ hormone levels in ways that tend to weaken the body’s immune system (Kiecolt-Glaser, Glaser, Cacioppo, & Malaikey, 1998). Adults' ^ ilily to create and sustain intimate relationships, which may guard against these negative health outcomes, spears to be forged early in life through children's relationships with responsive and sensitive caregivers (Bowlby, 1959; Rholes, Simpson, & Blakely, 1995). These parent-child interactions foster a secure attachment style, in which individuals are comfortable with closeness, trusting of R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. others, and can more readily and effectively coiniminicate to reduce conflict (Simpson, Rholes, & Phillip, 1996). Less optinmin caregiving (e.g., less responsive, more distant) instead produces insecure atlachment styles, characterized by one or more of the following: greater anxiety about relationship; greater tendency to avoid emotional ties; and/or less ad^tlve responses to stress, such as conflict (Collins, 1996; Collins, Cooper, Albino, & Allard, 2002; Mikulincer, 1998). In short, attachment styles effectively predict both positive and neptive future interactions (Feeo^ & Noller, 1991; Langan, 2001; Simon & Baxter, 1993) and outcomes in, for example, romantic relationship (for a review see Crowell, Fraley, & Shaver, 1999). Statement of the Problem Communication strategies may in fact serve to mediate this link between attachment styles and relationship outcomes. Psychology and communication have produced volumes detailing the passive and active strategies that seem to result in either positive or negative relationship outcomes (Baxter & Simon, 1993; Bell, Etely, & Gonzalez, 1987; Dindia & Baxter, 1987; Lundquist, 1996; Maguire, 1997; Montgomeiy, 1993; Rabby, 2001; Rusbult & Buunk, 1993; Waldron, 1991). However, researchers have long discounted the role played by individual differences - such as attachment styles - in pedicting the use of these behavioral strategies (e.g., responses to conflict). For example, Canary and Stafford (1993) considered it desirable to eschew what they perceived to be the ‘“needless proliferation of personality measures’ in explaining maintenance behaviors” (as cited in Yum, 2000, p. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 33). Therefore, the questioH remaiiiis: Do more secure attachmeat styles predict use of communication strategies that are more effective in mainlainiiig romantic relationships, particularly in response to conflict or distress? Does the ability of attachment to predict relationship outcomes stem from its ability to predict behaviors, which in turn predict those outcomes? Purpose and Significance of the Study Attachment provides an effective theoretical approach for studying questions regarding conflict within close relationships, such as romantic partnerships, by enabling assessment of both the sources of relationdiip conflict and individual differences in handUng conflict. Research suggests that the “anxiety over relationships” dimension of attachment is of particular importance here. Individuals high on this dimension report more relationship conflict.... Those who are anxious about their relationships also engage in coercive and distrusting ways of dealing wifli conflict, which are likely to bring about the very outcome they fear most. (Feeney, 1999, p. 374) Examining behavior and relationship outcomes from a perspective, our current study extends both the literature on Attachment Theory and that on Relational Maintenance by determining which aspects of Attachment Styles best predict selection and usage of Reactive Relational Malntenaaoe Strategies (i.e., Rusbuh’s Accommodative Responses to Conflict), thus promoting partners’ feelings of love (i.e., Sternberg’s triad of passion, commitment, and intimacy) and equity within the romantic relationship. Therefore, we would like to suggest that, as such, these maintenance strategies are the adult correlates of childhood attachment R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. behaviors: as, in bolb cases, distress activates the attachment system and behaviors intended to reduce that distress. The ensuing chapters 'lay out the links in a predictive model involving comections among attachment style, the strategies people use to maintain their romantic relationships in the face of conflict, and the ramificatioiis those reactions to conflict have for the relationship. Following this, hypotheses and tests of this proposed model are more explicitly stated. Previous research has already made initial explorations of the systematic ways romantic partners with differmt attachment styles maintain their respective romantic relationships (i.e., in the fece of conflict and other threats), as well as the ability of attachment style differences to predict differences in relationship outcomes (e.g., longevity, satisfaction, commitment). Unfortunately, these two lines of inquiry have seldom been combined, either conceptually or empirically, within a single study. Separate research ventures have explored the linkages between conflict responses and relationship outcomes, and these will be discussed later. Taking the results and implications of these studies together, the present study suggests that there is a potential to derive a theoiy-based model combining these separate elements in relation to one another. Preview of Remaining Chapters The next chapter reviews the pertinent literature in each of the domains relevant to this study (i.e.. Attachment, Relational Maintenance, and Love): (a) R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. explorii^ how Attachmait Theoiys although originally developed through observations and studies of infants and children, is nonetheless applicable to adults’ reMonsbips, particularly romantic reMonsMps; (b) documenting the connections betwem juvenile attachment behaviors and adults’ responses to conflict; (c) ilkstrating the ability of Attachment Styles to predict both response to conflict and relationship outcomes; and (d) examining the connection between responses to conflict and relationship outcomes from an attachment perspective. One Research Question and four Hypotheses have been generated based on these connections. Chapter 3 describes the methodology used in our investigation; Chapter 4 details the results; and Chapter 5 discuss^ the implications the results have for flieory and inteipersonal relationships. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. CHAPTER 2: REVIEW OF RELATED LITERATURES AND HYPOTHESES Below it is argued that attachment style dififercnces sha|» behavior in certain stressful situations (i.e., conflict) and thereby predict relationship outcomes (i.e., love and equity) th io u ^ a comnniJiiicatioii mediator (i.e., reactive relationsMp mainteoance strategies). In essence. Attachment Theory is used to hypothesize explanations as to why people employ dififereiit types of communicatioii strategies in response to distress within their romantic rclationsMps. First, we discuss atlachinent, describing the process by which secure and insecure styles develop, as rooted in Mteinial Woridng Models, and how these models shape behavior in childhood. This knowledge then informs our hypofteses regarding tiie activation of and flie effects of the attachment system in adulfliood. Finally, the elements of our explanatory model arc revealed. Attachment Theoiy In the late 1930s and early 1940s, as John Bowlby observed and analyzed case histories of over forty juvenile delinquents, he began to theorize attachment to one’s caregivers as “an important key to explaining the origin of delinquent character” (1944, p. 20). The fwuliarly affectionless behavior common amongst those who had experienced significant separations from or abandonment by their primary caregivers early in flieir lives led to the inability - or at least unwiilingness - to trust others, marked by: detachment, derogation of interpersonal reMonsMps, and lack of empathy. Attachment Theory emphasizes the importance of warm and dependable parental caregiving in establisMng airrent and future secure emotional ties; whereas, R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. cold or inconsistent parental caregiving consistently results in different insecure attachment styles manifested in distmctlve patterns of behaviore and personality tm ts that become increasingly set as the children matore (Bowlby, 1969). Attachment Stvles and Intemal Working Models Bowlby, Ainsworth, and others originally recognized Just three atlachment styles (one secure group and two insecure groups: Anxious-Ambivalent and Dismissive). However, as interest in and explanations for the cognitive processes underlying these styles of attsKjiment grew, later researchers (e.g., Bartholomew, 1990) began to argue compellingly for models of attachment with four distinct styles. In order to better capture the cognitive stojctures (Intemal Woikiiig Models) imderlkg the styles themselves. Bartholomew’s origiiial model explained the four styles based upon (a), the individual’s view of self, and (b), the individual's view of others - both either positively or negatively valenced. Ergo, the self may be perceived as either worthy or unwortiy of attention, care, and/or concern; and as competent or as less so; while others are perceived to be either trustworthy or untrustworthy; available or unavailable; wiiling/able or unwilling/unable to meet one’s needs; etc. (See Figure 1) H G B R E l SELF POSmVE (Low Anxiety) SELF NEGATIVE (H i^ Anxiety) OTHER POSITIVE (LowAvoidfflice) SECURE: low avoidawe of odwre, low anxiety about ^bandoniiiiait; comfort with intinjacy, autoncai^, and dependence upon or by othm . PREOCCUPIED: ovfa’ -dtefwadence upon oiters, combined w iii m anxious view o f self as anworfliy of anoter’scarefflid/crcQaceni. High d egi« of anxiety alwti: OTHER NEGATIVE (Hyti Avoidance) DISMISSING: low depeadtece C O oiicm, combined with high avoidance. Attachmeittttgeds are lypicaly «knie4 FEARFUL: disfrHst of others, belief iia t self is incapable or unwcrthy of close involvements. Tendeasgr to dHin otihm and/or avoid social situatkxs. Figure 1 : iHtegratioa of 4-Categoiy Models R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. This figure incorporates the thiinklng and tenninology used in. a more recent incamation of the four-category model (Fialey, Waller, & Bieiinaii, 2000), which captures the underlyiiig psychology of the inteimal working models in a somewhat different manner. The tram 's underlyli^ dimensions (e.g., Anxiety and Avoidance) hark back to features of Ainsworth’s coding scales and subsequent discriminant analysis, while still cohering well wifii Bartholomew’s precepts. A relatai article by Fraley and Shaver (2000) makes the connectioiis between fiieir respective dimensions explicit: ,..anxiety, corresponds to anxiety and vigilance concerning rejection and abandonment.... Avoidance^ corresponds to discomfort with closeness and dependency or a reluctance to be intimate with others. Empirically, these dimensions map onto the model of self and model of other dimensions, respectively in Bartholomew’s theoretical model (pp. 142 -143, italics in original) Accordingly, fi»m early infency, parmtal care received sets up expectations for continuation of the same (i.e., level and/or type of care). Over time different styles of attachment emerge (i.e., secure or insecure), rooted in assumptions fostered by the child’s experiences. As the child develops peater cognitive abilities, these conditioned ex^ctetions are refined, gradually creating - and later sustaining and reinforcing - certain coherent intemal representations o f one’s perceived and reflected self, as well as assumptions about others. Thus, these Intemal Working Models (IWMs) are tih e mentel representations of one’s self, one’s attachment figure(s), and one’s environment each individual develops throu^ experience. In fact, as early as R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 1973, Bowlby specified the process by wMcli these models develop and become ingrained: Confidence that an attachment figure is, apart from being accessible, likely to be responsive can be to turn on at least two variables: (a) whether or not the attachment figure is judged to be the sort of person who in general responds to calls for support and protection; (b) whether or not the self is judged to be the sort of person towards whom anyone, and the attachment figure in particular, is likely to respond in a helpful way. (p. 238) Attachment and Intemal Working Models In Adulthood Although attachmmt theory’s origins lie in child developmental psychology, from his earliest writings on the subject, Bowlby consistently staled h as belief that attachment was an ongoing process “from the cradle to the grave” (1954/1973, p. 333). Generally speaking, IWMs are at least somewhat susceptible to modification and revision should the child’s circumstances change. However, as they tend to be confirmed rather than r e d E u t e d by subsequent experiences, they become increasingly resistant to modification as the child matures. A fret Ainsworth attested to in writing of the Reversible and irreversible effects o f maternal deprivation on intellectual development. ... an early experience can set up certain dynamic processes that become consolidated and ingrained and that tend to continue despite the subsequent alteration o f the environmental situation.,..Once consolidated, this defensive operation tends to maintain itself, insulating the child against interaction with an environment that could prove supportive, responsive, and helpful if he would, or could, only be receptive. (1962, p. 46) R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 10 Once established, these models tend to become self-sustaining; shaping perceptions of others’ motivations and/or intentions (e.g., availability, responsiveness); and behavioral reactions based on those perceptions. Ihose reactions, in turn, elicit predictrfjle patterns of responses from others with whom inter»t, thereby supporting, fidfilling, and sustaining those beliefs. In 19§7, Hazan and Shaver provided die initial explorations that helped est^li&h the argument that romantic relationships between adults were true attachment partneisMps. They argued that romantic relationships are emotionally significant and persisting affectional bonds between pffltoers who; 1.) could not easily replace the attachment figure wifii another who othorwise fiilfill the same needs or purposes; 2.) desire to maintain proximity to and/or contact with one another; 3.) find involuntaiy separation from one another distrosing; 4.) seek ^cim ty and comfort from their relationsMp, ideaiy providing one anotilier with a secure base and safe haven as needed (see also Ainsworth, 1989). These romantic relationsWie were, thus, notably similar to the reMonsMps of cMldren to their caregivers, where each of these elements had also been esteblished (see Ainsworth, 1964), Given iiese pmaliels between the cMld-parent attachments and attachments between romantic partners, researehere began hypothesizing effects that might be found for different attachment styles in romantic reMonsMps; such as: the need or desire to bond with one’s romantic partaer; the ways in v ^ c h that bond is maintained, particularly under stress; and the resulting emotions that pffltoere experience within tiieir romantic reMonsMps. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 11 Furthermoie, as Hazam and Shaver specijfie4 adults’ pircess of becoming attadied to their romantic partners is ‘‘ experienced somewhat differently by different iw ple because of variations in their attachment histories” (1987, p. 511). Their research deinonstra,tai that internal working models, developed and refined througtout one’s childhood, foster “different beliefs about the course of romantic love, the availability and trustworthiness o f love partners, and tibeir own love-worthiness” (Kazan & Shaver, 1987, p. 521). Yet, far less attention has focused on the ramifications of these different beliefe, specifically for behavior: e.g.. What are the behavioral correlates in adults of flie attechnaent Iteliavlors utilim i In infancy and cMldhood? Internal Working Models and Behavior These mental rqwesentalions, sometimes alteiiiativeiy referred to as one’s “Representational Models,” exist as patterns or templates to guide behavior in fiiturc encounters. Accordingly one’s mental representations also influence the degree to which an individual may f« l anxious about or wish to avoid social interactiom. As was succinctly stated in a comprehensive review of attechment in children, these Inteinal working models are the mental representations of the attachment figure, the self, and the environment, all of which are largely based on experiences.... The child is thought to rely on these models, for instance, when making decisions about which specific attachment behavior(s) to use in a specific situation with a specific person. (Cassidy, 1999, p. 7) R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 12 What distinguishes a true attachment behavior for children (Le., vocalized calling, raising arms to be picked up, following) fiom other similar acticms a child may perform is that these are behavioral strategies (e.g., proximity maintenance) directed at one’s caiegivea- in order to ensuK the continued availability and responsiveness of said caregiver as a secure base for exploration and/or a safe haven in times of distress. As such, attachment behaviors tend to increase when a child is under stress, when the desire for attachment becomes a priority. The effect working models have on behavior can be seen in an early experimraital procedure called the “Strange Situation,” developed by Ainsworth in 1978, Observations of infants and toddlers temporarily separated from their mothers in an unfamiliar environment revealed clear differences in these cMldten’s reactions when their mothers returned. Diawkg upon knowledge of the ways in which cognitive processes shape behavioral responses, Ainsworth reasoned tiiat tiiese consistent differences in the patterns of behaviors displayed were largely due to the children’s different attachment experiences, which had created distmcfly different patterns of expecMions as to their mothers’ responses, Hiiis, the nidimmtaiy models, which allow an infeat or young child to forecast a caretaker’s likely behaviors in response to Ms or her own actions, over time form the basis for more M ly develc5ped models of the other’s mind (i.e., motivations, desires, intentions), allowing youngsters to fine-tune their own bdiaviors In a manner that may best be desciited as strategic. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 13 Those classified as securely attached, anticipated responsiveness and emotional avmlability fiom their mothers, and so actively sought and received comfort (i.e., being soothed by her with hugs, cuddles, etc.) when their mothers returned. Greater passivity and/or negativity were characteristic reactions among the anxiously-attached youngsters, many of whom were not soothed by any attentioii they received J & o m Iheir mothere. These yom^sters - accustomed to and anticipating rejection - tended to ignore their mothers, approached cautiously, or even aggressed against them (e.g., Mckii^, Hting), These negative reactioiis were fiaictional adaptetiom to their insecure partnerships, in that tiiey served either to guard the child against fiiilher parental rejectton, or vent the child’s negative emotioiis and perhaps even to punish the inattentive caregivers (Bowlby, 1969/1982, p. 338-339). These children’s attachment behaviors were elidted In r^ponse to the stress produced by their mothers’ brief absence. Simpson m d Rholes (1994) point to stress as the trigger for attachment behaviors. From an attachment fwspective, her absence represented the child’s perceived loss of a secure base and/or potential safe haven iiat the behavioral strategy of proximily maintenance - thwarted by researehers in the Strange Situation Experiments - would otherwise ensure. The children’s behaviors demonstrate a remarkable similarity to the strategies adults use to accommodate their romantic partners in periods of stess (e.g., conflict). In fact, althou^ Ainsworth’s experiments were designed to assess only three types of R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 14 attachment, the children’s behaviors may be plotted along functionally equivalent dimensions to fliose foimci in Rusbulf s Accommodation Model. A lth o i^ Rusbulfs model will be described in more detail subsequently, her four reactive styles are briefly described (See Figure 2) below to illustrate their overlap with the children’ s behaviors. Although these models are fairly accurate reflections of the individual’s real life experiences up to that point, their past-as-precedent usage may, In later life, be misapplieci and lead to behaviors that are less fimctional for present cireumstances and/or fiiture encounters. H GKRE2 ACTIVE PASSIVE POSITIVE VOICE in adults: making dissatisfaction known to partner, and discussing possible soluticms. Child expresses need or desire for, and receives, ccmfort (e.g., picked up, held, cuddled). LOYALTY in aduhs: raaHrtainfag feith In partner anti/w the reJation^ip; dm e, waiting f « good times to return. C hid explores little, close to motfao- (e.g., clinging, whining fiir attention). NEGATIVE EXIT in ffiiults: behaving in ways inlemied to hastai llie end of lelationdiip. Child mishehaves, a^essiveiK ss (e.g., biting, hitting, Mcklng) aaid/or ai^er. NEGLECT in adufts: Ignoring partaer and/or conffict in hqpe that issues will be resolved cm tbeir own. Child ignores cattegiver, swoids MtiatiBg cxmtact, veers away instead o f m sroachim Figure 2: Adult Accommodatloitt and CMldhood Attacimeiit Behavions Accommodation Attachmait behaviors are elicited in those situatioiis that the child deems ttaeatoiing or otherwise distressing (Bowlby, 1969/1982,1979), By analogy, we would expect them to be eMcited when adults feel threatened or under stess (Simpson, Rholes, Orina, & Grich, 2002), as previous research has esteblished that attachment differences in perceptions and behaviois tend to be manifested more strongly during R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 15 conflict, when the relationship is threatened (Feeney, 1998). Most people find conflict with their relationdiip partners distressing, particularly to the degree that such conflict is seen as a threat to the relationship, Rusbulf s Accommodation Model of reactions to conflict, therefore, provides us with a particuiffl-ly well-suited set of comimjiiicatioii strategies for consideration as attachment behaviors for adults, as these: (a) are elicited under conditions of stress; (b) fijnction to reduce tihat stress for their user, at least in lie short term; and (c) have predictable and well-documented probable reiatioiisMp outeomes which can be viewed, from an attachment perspective, in tem s of increasing or decreeing the likeliiood of a relationship partner’s contfnued responsiveness and availability. Additionally, they cohere well with assumptions underiying Fraley, Waller, & Brennan’s (2000) model of attachment, as w el as sharing parallels with at least some of the deaaiptioiis of children’s reactions within Ainsworth’s Stange Situation experiments, as represented in Figure 2. Also, these have at times been grouped with Relational Maintenance Strategies (Yurn, 2000) as the reactive counterpart The importance of this designation is found in tie assumptions unda-lying relational mamteuauace: Just as infants and children employ strategic behaviors in order to: (a) ensure the availability and responsiveness of their caregivere in times of distress, and/or (b) m kioice their own distress in their caregiver(s) absence; Relational Mamtenance Strategies are intended to: (a) Increase the likelihood one’s romantic partner will be around to provide a secure or safe haven as needed, and/or (b) alleviate distress produced by threats to the relationship. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 16 Joseph Ayres conducted the seminal research into relational maintenance in 1983, investigaling use of strategies such as Avoidance (aMn to Rusbulf s Neglect) and Directness (aMn to Voice). He found that: “reM omhip intent makes a considerable diflference when one’s intent is to maktein rather than end a relationship” (Ayres, 1983, p. 66), obtatnlng MgMy significant dfflFerences (p <.001) between the ^ipes of strategies participants selected. The four accxjimnodative ractioiis to conflict are arrayed along two inteisecting dimensions: 1) Constructive vs. Destructive - serving a positive, !.e., relalioesWp-tenhaiiciiig, versus a negative function; and, 2) Active vs. Passive - exemplifying effortful behaviors to resolve or alleviate the source o f the conflict versus passive acceptance o f status quo. Once plotted out graphically, iiese divide possible reactions to dissatisfaction or conflict into four distinct categories. (See Figure 3) FIGURES ACTIVE PASSIVE CONSTRUCTIVE VOICE: encompasses verbaHy tffinging iqi the cause o f one’s dissatis&dioo, M tM ag discussion LOYALTY: desiKMistrated by taking no overt actitm to resolve issues, bat hoping and/or fmiyiBg fee situaticai will diange; or otherwise mamtaining failh in one’s partner. o f sources o f conffict, m d talktag wilii one’s fwrtner sbout bow best to resolve any probtens. DESTRUCTIVE EXIT; comprised o f b^vional taAics such as: leaving, coasidaring lesving, <r •ist'^aEiiijg to leave tilie relationdiip. Exit may also be extended to cover other socii overtly negative behaviors one could engage in to fermg about or h a ^ n tbe end o f a relatio»ship. NEGLBCT: dbarKterized by actions SBch as avoiding one’s partner, leaving the situatkM until one’s partner cx»ls down or fee conflict Mows over. N ^ te c tm ^ tA e th e fonn o f reflising to discuss issues, pretmding they don’t exist,« at least minfmizmg irecognition of mnflict or problaHS. Figore 3: AcccHBnnodative Responses to Conflict R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 17 This quadratic model of Reactive Relational Mmntenance Strategies has therefore been selected to supply the behavioral component for the current study. Hypothesis on Attachment and Accommodation Taking an attachment approach to the study of interj^tsonal reMonships (e.g., Collins, 1996; Mikulincer, 1998), researchers have already confirmed that securely attached individuals maice more positive attributions for partners’ ambiguous behaviors than do their insecurely attached counterparts, who tend to make more negative attributions (for a review see Crowell, Fraley, & Shaver, 1999). Even earlier, in 1993, Rusbult and Buunk observed that: .. .strong commitment... promotes a variety of relationship maintenance behaviors such as adaptive social comparison and perceived relationship superiority, derogation of attractive and threatening alternatives, effective management of jealousy and extra relationship involvements, willingness to sacrifice for the good of a relationship, and tendencies to accommodate rather than retaliate when a partner behaves poorly, (p. 180) Reading the results of these studies together provides a more comprehensive explanation: That is, due to their relatively more favorable Intemal Working Models of themselves and others (Bartholomew, 1990) and/or their lower rates of mixiety and avoidance (Brennan, Claik, & Shaver, 1998), securely attached individuals tend to attribute their partners’ negative actions to situational pressures rather than to their partners themselves (i.e., to a transient state rather than a chronic trait), and therefore tend to react constructively in order to compensate for and/or counter their partners’ negativity. Thus, security of attachment predicts both perceptions of one’s partner. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 18 and subsequent behavioral reactions. In 1985, Donovan and lessor linked insecurity to less ftmctional behavior within romantic pffltaerships. “Secure individuals, experiencing responsive caregiving, leam to acknowledge distress and to seek support from others. Avoidant individuals leam to avoid acknowledging or expressing distress so as to minimize conflict with insensitive caregivers” (Feeney, Noller, & Roberte, 2000, p. 193). This succinct statement, clearly illustrates how the dimensions underlying attachment styles and responses to conflict cohere so well with one another. Secure attachment encodes expectations that others will respond helpfully if made aware of distress; whereas less secure attachments are the end result o f consistently finding this not to be true. Therefore, we anticipate that a positive view of self, or less anxiety (i.e., fear of rgection and/or abandonment) over their relationships, will predict positive or constructive reactions to transgressions by others (i.e., voice and/or loyalty) and/or lower negative or destructive reactions (i.e., exit and neglect); whereas, a negative model of self, characterized by greater overall anxiety regarding relationships, will instead foster more negative reactions and/or reduced tendency towarf positive reactions. A negative model of others, characterized by greater avoidance of reMonsliips due to discomfort with closeness, will similgffly yield more negative and less positive reactions to conflict and/or discontentment with one’s partner’s actions; whereas, a positive model of others, and a genaally lower inclination toward avoidance, will result in greater likelihood of using positive strategies, as fostered by R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 19 their (secure) inherent belief that others can and will respond appropriately If given the opportunity to do so. According to Kpp and Harmon (1987, p. 650) this is due to their “physiological coding that the universe is benign and need-satisfying, that is, homeostatic disruptions will be set right." Hypothesis#!. Secure attachment will predict more functional reactions to conflict; whereas, insecure attachmeEt will predict less functional types of responses. (See Figure 4.) SELF POSITIVE (LOW ANXIETY) SELF NEGATIVE (HIGH ANXIETY) OTHER POSITIVE LOW AVOIDANCE SECURE - W aofe! to maintain reMoDshipCs) & believes in own ability to do so & value of lelationships wifli others. Positive correMioii to use of VOICE, possibly LOYALTY Inversefy related to use of NEGLECT, possibly EXIT PREOCCUPIED - StfMig desire to maintain relationsMp(s), l» t laste M h in own desirability as reMonship paitaer and/or own interpeiBM ial abilities. Positive correlatioa to use of VOICE, possMy LOYALTY Inverse^ related to use o f EXIT OTHER NEGATIVE HIGH AVOIDANCE DISMISSING - No stim g desire to maintaia relati(3oAip(s); but, no doubt in own interpersonal abilities. Positive OKreMoQ to use of EXIT, and NEGLECT Inviasely related to use of LOYALTY, and VOICE FEARFUL - Lacks d«ire for interpersonal reMonsliips. Tends to avoid social ties. Lacks belief in own desirability aaad/or interpersonal abilities. Positive correlation to use of NEGLECT, and EXIT Inversely related to use of VOICE Figure 4: Hypotiiiesirad ElTect o f Attacbnent on Accoimnodation In this model, we see that inclination toward avoidance is anticipated to positively predict tfie constructive/destructive dimension of response to conffict. Thus, those who « relatively higher in avoidance will tead to withdraw in response to interpersoMl difficulties: either neglecting partners during times of relational R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 20 stress or exiting the relationsMp entirely. They should also report less active engagement (i.e., voice), and/or passive support (ie., loyalty). Whereas, hiose individuals with less avoidant teiwJaicies will react more constractiveiy: micing their concerns and/or remaining loyal to their partners. They should also report less of a tendency towards either Exit or Neglect. Furthemore, anxiety w il also predict wheiier one rei^onds to conflict actively or passively. Those with less anxiety aibout being abandoned by tihteir partners will report using more positive approaches to defuse the situation: either voicing their discontentment and seeking solutions or remaining loyal to their partner’s in anticipation of a retum to a more favorable sWe. Whereas, greate anxiety about being abandoned w il tend to increase activity of response, either increasing use of Voice (for those with low avoidance), or Exit (for those with higher avoidance sc»res. Also, those who are most likely to use Voice should be least likely to Neglect their partners, as these are the most diamefrically opposed responses to conflict; conveisely, those whose prefared strategy is Neglect will tend to avoid Voicing their discontentment aad/or concerns. The above predictions arc siipporte4 albeit somewhat tangentialiy, by evidence in a review article on Attachment in Adolescence by Allen and Land (1999) revealing that within parent-cMld dyads, securely attached teens ofx:nly discussed problems with their parents so as to malnfaifi their close relationships; whereas Insecurely attached teens tended to avoid discussions of these topics and to disengage fliemselves from the R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 1 relationsMp. Similariy, Kobak & Sceery (1988) reported that dismissing adolescents ten&d to r e ^ to stress&l situations by pushing away those closest to them; and^ those with the most insm ire attachment profiles were also consistently ranked as hostile and lacking in social skills, even by their close ftlends. In fact, from infancy onwads, trends in directiom consistent witii support for the hypothesized reMonsMp between attachment and response to conflict can Mieady be seen. Infants who had confident expectations concerning their caregivers actively sought comfort and used the parents as a safe haven.... Avoidant infents, who had reason to expect rejection from their caregivers, modified their attachment behavior by avoiding the caregivers, effectively reducing anticipated conflict or rejection.... Ambivalent/resistant infants, who had reason to be uncertain about their mothers' response, showed angry resistant or passive behavior that served to increase their proximity to caregivers. (Kobak, 1999, p. 32) We may now begin to turn our attention to the next leg of our model: looking at how these four distinct behavioral strategies affect the nature and quality of the romantic relationMiip, thereby predicting markedly different relationsMp outcomes. As we are specifically focused on romantic relationsMps for fliis study, the outcome most relevant to our study will be self-refwrted feelings of love toward one’s partner. Sternberg’s (1988) Triangular Theory ofLove provides an excellent model, as it segmCTts love into attachment-relevant component parts: Commitment, Intimacy, and R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 22 Relationship Outcomes Much of the existmg literature has concerned itself with deteiiniiiiiig which maintenance strategies best predict different relationship outcomes. Despite the feet that most of these indings are based upon correlational data, and thus causation cannot truly be determined, relationsMp satisfaction, equity, commitmmt, longevity, and such are genaally termed the relatiomkip outcomes o f RMS usage, a lth o i^ the researchers are often quick to note that the evidence cannot support any contention tiiat the RMS usage actually causes or precipitates these outcomes rather flian the reverse possibility. The widespread view is that the two are recursively linked, yet the designation of tiiese as “outcomes” persists and will be used throiJ^iout this paper as well, despite this acknowledgement that it is at least somewhat of a misnomer. Sternberg’s Love Triangles allow examination of romantic partners’ feelings of different aspects of love: 1.) Commitment is related to caring, stability, and contiiKiaiice of the relationsMp. It carries with it certaia relevant perceptions, such as that the relationsMp will be ongoing and that one’s partner will continue to be responsive and available; 2.) Intinracy refers to feelings of trust, coimectioiii, and/or bondedness with one’s romantic partner. As such it taps into attachment relevant constructs such as: warmth, closeness, and comfort; 3.) Passion encompasses desire for physical contact, excitement, and the preference for one’s partner over olhers. Accordingly, It is germane to atfechmeiat in that the attachment figure is the one R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 23 prefeired above all others, not replaceable by others, and fiom whom separation is emotionally upsetting. Just as attachment tiieorists distkguish between attachment and affectioiial bonds, so too does Sternberg’s theoiy specify distinctiom between loving and liking: The fornier is ctesified as a sttonger fonn of aifectfon and attechment to one person; whereas, the latter is considered as a less intense emotion, a type of fcmdness that may be extended to many others In one’s life at die same time (Sternberg, 1988). Eguity In order to be consistent witii much of the literature in relationsMp maintoiaiice (e.g., Canary & Staffor4 1992), equity ratings were also obtaked and examined. However, it should be note4 that this outcome is most frequently seen in explorations of Proactive Relational Maintenance, and theoretical considerations only really make sense in the context of anxiety about abandonment, as wil! be discussed later. In short, equity measures the degree to which participants believe the benefits they receive from the relationsMp are comparable to their partners’; as well as relative faimess regarding the balance of inputs and outputs. Averaging the items from this scale provides an estimate as to overall degree of perceived feimess, e.g., 1.) does one benefit from Ihe relationsMp as much as one’s partner does; and 2.) considering different aspects of the reMoesMp, do you get out of your relationsMp what you put it? R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 4 Attachment and Romantic Outcomes Since Hazan & Shaver’s classificatton of romantic relationships as adult attachment relationships, it has repeatedly been demonstrated that secure individuals tend to create and sustain the most satisfectory and lastiiig romantic prtoersMps (e.g., Feeney, Noller, & Cailan, 1994). Furthermore, Attachment Styles also predict ratings of love-relevant constructs such as commitment, satisfaction, interdependence, and trust (e.g., Keelan, Dion, & Dion, 1994; Pistole, Clark, & Tubbs, 1995; Simpson, 1990). In fact, a recent study by Collins, Cooper, Albino, and Allard (2002, p. 966) provided “the jS rst prospective evidence that avoidant attachment places individuals at risk for adverse relationship outcomes,” a lth o i^ this has long been accepted as true: ...it is an essential condition of subsequent satisfactory interpersonal relations for a child to have formed an attachment to someone—his mother or a substitute for her. On the basis of all of our clinical knowledge of the relationship between early parent-child relations and later intimacies, it seems desirable for a baby to have a secure attachment to his mother rather than an insecure one. (Ainsworth, 1954/1972, p. 196) We therefore predict that security of attachment will positively predict love for one’s partner. As higher avoidance reflects the disinclination to become intimate with other people, and a tendency to shun or be wary of emotional commitments, it is predicted to correlate to lower reported feelings of love and/or higher likelihood of ended relationships. This prediction is consistent with findings from fdor research studies (e.g., Kirkpatrick & Davis, 1994; Mikulincer & Erev, 1991). The latter of which found that more highly avoidant individuals desired less commitment, intimacy, R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 25 and passion in even an ideal love relationsMp, as compared to secure m anxious participants. Additionally, Collins et al. (2002, p. 982) demonstrated that ‘Respondents ■ w h o were higher in avoidance at baseMne viewed their felationsMp as more poorly fimctioning overall; they were lower in satisfaction, intimacy, and shaisd disclosure. They also reported less effective problem-solving comiramication and greater conflict” (Note it is this latter point thM directiy leads into the ensuing section.) Also, inasmuch as anxiety is indicative of negative view of self (Bartholomew, 1991) and/or fear of abandonmait (Bfennan et al., 1998), those who are more M ^ y aixioiis should therefore perceive greater benefits to being in the romantic relationship; that is, either their perception of self as unworthy of any benefits received or gratitude and/or relief that their partner hasn’t yet abandoned them, diould enhance their equity ratings relative to less anxious participants. Hypothesis #2a. Secure attachment ■ w ill predict more positive love feelings (i.e., commitment, intimacy, passion); whereas, inseaire attachment w il predict less love within the reMonsMp. Hypothesis #2b. Greater anxiety will predict greater feelings ofbenefittedness ■ w ithin a romantic relationsMp. Accommodation and Love A tendency to behave negatively in response to conflict or the perceived transgressions of one’s partner will lead to reduced relationship satisfection (e.g., Mien, Markman, & Lindahl, 1989; Noller, Feeney, Bonnell, & Cailan, 1994), and R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. otherwise tend to destabilize a relationship (e.g., Kamey & Bradbury, 1997) or lead to an overall decline in relationsMp quality, wMch may be measured in tire relative amounts of love m d equity reported by participants. WMle it is undoubtedly true tiiat tMs relationsMp is in many ways recursive, in that great® commitment to one’s partner and perceived benefits from the relationsMp will tend to foster greater inclination to maintain that relationsMp (see Rusbult’s Mvestment Model, 1980,1983, for elaboration on tins connection), our hypotheses have been drawn from tenets of attachmeait theory. Thus, we assume that attachment motivated attributions and subsequent accommodative behaviors as reactions to conflict exist a priori to relationsMp outcomes and, therefore, are more usefiilly envisioned as the predictors rather than as consequents of these outcomes. Hypothesis #3. Accommodative behaviors will positively predict relationsMp outcomes, such that negative behaviors (i.e., exit, neglect) will be associated with less reported love feelings (e.g., commitment, intimacy, passion); whereas more positive behaviors (i.e., voice, loyalty) will jwedict more love in a romantic relationsMp. Mtegatlng Attachmeat Accommodation, and Outcomes [W]orking models o f self and others may have a direct influence on relationsMp outcomes by coloring social perception, shaping affective response patterns, and directing interpersonal behavior. In tins way, early working models exert their influence on future relationsMps by predisposing individuals to think, feel, and behave in ways that either facilitate or interfere witii relationsMp ftinctioning,... Insecure adults engage in less adaptive interpersonal behavior across a variety of domains. (Collins et al., 2002, p. 970) R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 7 TMs longitudinal study clearly illustrated that at least the avoidant participants went on to develop less satisfying romantic relationships due in part to their tendency to in. fewer pro-relationsMp behaviors, wMch was dimcfly linked to poor relationship quM ity for both themselves and their partned’ (Collins et al., 2002, p. 995). Hiey, thus, concluded that: .. .avoidant men and women experienced interpersonal difficulties across a variety of domains including intimacy, communication, and trust.. ..Because these patterns emerged in prospective analyses, they are consistent with the idea that avoidant adolescents entered their relationsMps with maladaptive behavioral tendencies that then contributed to low satisfaction for both themselves and their partnere. (Collins et al., 2002, p. 998) Internal Working Models shape an individuaTs actions and reactions, as w el as on one’s perc^tions regarding the actions and reactions of others. The bdiavioral adaptations iiese mtwiels inspire tend to be fenctional for Ae situations and conditions that led to their development. However, they can prompt less functional behaviors under different - future - circumstances. TMs is particularly true of insecure attechment “[T]he re^oa that the equipment proves so unsuccessM is that it is being required to operate within an enviroimiait to wMch it is not ad ^ted ... .(M y within its environmeiit of adaptedness can it be expected that a system will work efficiently” (Bowlby, 1969/1982, p. 46-47). Hius, over subsequent iteiatiom (i.e., interactions with one’s relationsMp partner), the anxiety and/or avoidance characteristic of inseaire attachment styles - according to the tenets o f Attachment Theory - generate distmctly different actions, then perceptions of the actions of others, then reactions, and so on. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 2 8 These comistmt distortions produce distinctly different behavioral responses, over time resulting in maricedly different relationship outcomes. Thus, alliM u^ the reciprocily of the linkages between indicators of relationship quality (“Relationship Outcomes”) is iinquestioiie4 our model proceeds from the standpoint tiiat it is these behaviors that are primarily predictive of and/or precipitate the outcomes. That is, as attachment styles develop from infancy onward and tend to resist change, it is cleffl-ly fair to classify individuals by attachment styles that precede both tfieir behavior and their commitment within a romantic relationship. The latter two constructs (i.e., behavior and commitment) are somewhat more recursively tied: each mutualy shaping and reinforcing the other (see Figure 5, next page). However, as one’s attachment style, intemal working models, and ^sociated behavicaal tendencies are kgraiiied prior to the establishment of a felationsMp w iii anotfier person (i,e., establisMng a romantic necessarily entails at least some degree of action and the desire to do so), the predictive direction of the factors being examined in the current study consider one’s tendency to engage in the relevant behaviors (i.e., Exit, Voice, Loyalty, Neglect) as primarily antecedent to such outcomes as critical relationship outcomes as commitment, passion and intimacy (Sternberg, 1988). This position is consistent with Collins’ preliminary model (Collins et al., 2002) as described previously. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 29 K g sfe e ,iM S -& R g M o n i|lE .Q a c o ffie sL lllg .^ ^ ^ As Ihe preceding discourse illustiates Attachment Theoiy provides a firoework (see Figure 5) for exploring and better understanding the process by Relational Maintenance Strategies predict a variety of possible outcomes in adults’ romantic leMonsMps. RELATIONSHIP OUTCOMES ATTACHMENT: WORKING MODELS REACTIVE RELATIONAL MAINTENANCE STRATEGIES Figure 5: Mediadoual Model illustrating how Attachment affects Relationship Outcomes As described previously, different attachment styles are associated with different patterns of beliefs: one’s intemal working models. Although Bartholomew (1990) originally explicated these as one’s beliefs about self vs. one’s beliefs about others, Brennan et al. (1998) followed by Fraley & Shaver (2000) make a strong case for the principal components of workkg models instead being anxiety and avoidance. Either assessment, however, is consistent with the piedictlom made in our model. Hypothesis #4. In a hierarchical regression equation, following the procedure laid out by Baron m d Kenny (1986), whereas Attachinent will initiafiy significantly predict love scores, it will cease to do so when Accommodative Behaviors enter the equation. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 30 CHAPTER 3: METHODOLOGY Overview Ttiree hundred and eighteen participants (33% males and 67% fanales) compldied^ a survey using measuies designed to assess their attachment styles (i.e. Bartholomew’s RQ and Fraley & Shaver’s ECR-R), responses to conflict (i.e., Rusbult’s Accommodation Scaled aad feelings toward their romantic pntner (i.e., Stemb«g’s Love Triangles). Demographics and other kfomiatioii were also collected via this instrument Botii online (n=155) and pen-and>paper (n=206) versiom of fliis measure were conducted. Participants’ answera were analyzed using SPSS software. The basic design wjb correlational, assessing whetier and how accurately H I, H2, and H3 were supported; witih MeraicMcal regression analysis used to test our model (H4). In short, we assessed the relationship between participants’ attachment style and/or mental models, their behavior and their feelings within their current or most recent romantic relationships. Development o f Survey Instrument As proven measures alrea<fy exist for the elements of our model, no new items were - or needed to b e - created for our survey. Thus, it was more of an assemblage than a wholly unique or original creation. Items were culled from already created and tested measures related to our hypothesized model, as will be described below. * In all, thiee hundred and six^-one pec^ie filled out at least some portion o f tb s surwey, however, a total o f forty-three sttrveys did not meet om minimom ieqnireinent for kdusion (^e., rule for iodusioa v w that at l«ist seventy-five percent o f the questions needed to be filled out This eikmnated Aose surveys for which insufSrient sections were completed to dlow specified txHnpaasons.) R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 31 Attj^hment Two sqw ate measures of attaclaneiit were used for ttiis study; 1.) ECR-R: Experiences in Close Relationships - Revised (Fraley, Waller, & Brennan, 2000) a 36- item self-report inventory, utilizing a seven-point Likert scale, ’ SJ^ch assesses attechnieiit style based upon one’s intemal working models. The 18 Anxiety items measure aspects of Anxiety about abandonment; while another 18 items measure Avoidance of interpersonal involvements, thereby allowing for hypothesis generation and testkg based upon these underlying components.^ Certain items fiom each scale were reverse coded, as per instnictions. Then internal consistency was computed to assess reliability: Cronbach's coefficient alpha was .93 for Anxiety items, and .94 for Avoidance. 2.) Bartholomew & Horowitz’ (1991) RelationsMp Questionnaire (RQ) was also used as a very brief assessment of a partidpaiif s geiMal attachment style, to enable comparisons and ensure conveigeiit validity. The RQ assesses attachment style by asking how one generally feels within romantic relationsMps, and rating the extent to wMch the participant agrees with descriptions of each relationsMp style individually. Alpha vali» for RQ was .69. Reactive RMS/ELVN Items. Rusbulf s 16-itan Accommodation Scale for Responses to Conffict was used m an effective self-report assessment of partfcipants’ penchfflit for choosing each of 2 Foi f te csifteot study, parOdpaots weie asked to answer these items thinking exclusively about their cnrsent or most tEcent romantic p artneish^. This required that a few o f the items be reworded dighdy in order to into tbe specific rebtioiiship o f interest for oar study. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 32 the possible four responses to conflict and/or dissatisfection within their relationships: Exit, Voice, Loyalty, and Neglect. Responses were arrayed along a 9-point Likert- type scale. Alpha values for items on this scale were: Exit .71, Voice .78, Loyalty .76, Neglect .74. Average scores for each response wcjre then computed and used for subsequent analyses. Sternberg’s Triangular Love Scale was selected to measure participants’ feelings regarding coininiliiieiit, intimacy, and romance or passion within Iheir romantic relationships, as these items tap into the construct of love and do tend to predict relationship longevity. The scale, or components fliereof, has also been used in many of the other studies used as models informing this the current project. Thus, the measure is not only reliable, with an overall alpha of .98 for all 45 items (with individual alphas of: .95 for Passion, .95 for Intimacy, and .96 for Commitment), it is also consistent with prior analyses wifliin the crarent area of interest Additionally, it coheres well with components of Rusbult’s Investment Model. ia a ix An Equity scale was also utilized, as this has been found to be an effective measure consistently used in studies of Proactive Relationship Maintenance (e.g. Canary & Stafford, 1992; Canary & Stafford, 2001; Messman, Canary & Hause, 2000; Vogl-Bauer, Kalbfleisch, & Beatty, 1999). Participants’ feelings of over- or under- benefittedness in relation to their partners provides a basis for estimating flieir R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 33 satisfaction with the relationship and comparison level for alternatives. Alpha reliability was .80 for this scale. Demographics Potential moderators of the proposed relationships would include the following: ClMMRoiBintLC,M t e By definition, romantic partnerships that function as attachment relationships should be cmisidered serious and exclusive by the participants. Casual dating relationships wifl not necessarily produce the same effect on behavior, as partners do not rely on each other to the same degree. In the current study, almost two-thirds of the participants (n=199) reported themselves to be currently in serious romantic relationships. M atipnshipl^gtii As even initially unstable partnerships may tend to stabilize over time, volatility is difficult, even exhausting to maintain, also trust and liking tend to grow over successive interactions, even in non-romantic pairings. Hazan and Zeifinan (1994), for example, found significant qualitative differences between adolescent relationships of less than 2 years duration and greater than 2 years duration. Among those we surveyed, sixty percent (n=l 19) had hem in their current romantic relatiomdtips for less than two years, forty-percent of participants (n=80) were in serious and committed romantic relationships at or over this 2-year mark. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 34 Ciiiture This also exerts a shaping-force on personality and perceptions, which could tend to modify the impact of attachment styles. Yum (2000), for example, foimd some significant differences attributable to coUectivistic-individualistic societal orientations. However, we did not actively recruit based upon any ethnic, racial, or cultural considerations. Our sample was insufficiently balanced with regard to ethnic diversity for this type o f analysis. Gender Numerous studies have consistently fouiM i differences in patterns of attachment, with more women found in the Preoccupied category whereas more men fall into the Dismissive-Avoidant group (e.g., Collins et al, 2002). However, due to the gender Imbalaiice in our particiilar sample, we were similariy unable to subdivide for analyses based on this criterion. Age This fector would tend to correlate with levels of maturity and relationship experience. The proposed Attachment Model of Relational Maintenance was tested for confirmation on a sample composed primarily of undergraduate college students. Most college students range in age from 18 to 24, which we determined to be a i ideal age range for our study. During adolescence a child's attachment to his parents changes. Other adults come to assume an importance equal to or greater than that of R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 35 the parents, and sexual attectioii to age-mates begins to extend the picture. As a result individual variation, already great, becomes even greater. At one extreme are adolescents who cut themselves off from parents; at the other end are those who remain intensely attached and are unable or imwilling to direct their attachment behaviour to others; between the extremes lie the great majority of adolescents whose attachment to parents persists but whose ties to others are of much importance also. For most individuals the bond to parents continues into adult life and affects behaviour in countless ways. (Bowlby 1969/1982, p. 207) Therefore, the participants we surveyed had mostly likely completed their adolescent transference o f primary attachments from parents to peers - particularly to their romantic jxutners (Bowlby 1969/1982). Thus, this age group is ideally suited for inquiiy into the behaviors used to maintain those attachment relationshijK. However, due to this, we did not have sufficient divarsity in this regard to attonpt analysis: Fully 70% of our participants were 18 to 24 years of age, only shortly over 10% were over 30 years of age. Also, age tends to be confounded, to a substantial degree, with relationship longevity - as older participants have had more time to forge romantic relationships - and with seriousness of the relationship - as older participants are more likely to be married tia i are the youngest participanite. SaffiplingPlgfi A total of 361 individuals participated in the current study, however, only 318 of these submitted useable data. Snowball Sampling was used to recruit participants to visit the survey website. Advertisements for the survey were distributed via e-mail to student listservs asking recipients to participate and/or forward the information on to R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 36 others. This resulted in 204 “hits” on the website, with 155 online surveys completed. Additionally, the investigator was invited to conduct a pencil-and-paper version of the same survey in three introductory psychology classes at Saddleback College, which returned a total o f206 surveys. As an incentive to encourage participation, or at least visits to the online survey website, recruitment solicitations offered a chance to win one of four gift certificates from Amazon.com, one In the amount of $100 and 3 in the amount of $50. However, survey completion was not required for entry into the drawing. The first page of the survey was ftie Informatioii Sheet advising potential participants of their rights jmd die intentions of the study. Saddleback College participants were also enrolled in the drawing, and received a printout of the Information Sheet. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 37 CHAPTER 4: RESULTS Wherein we explore our date, providing the relevant breakdowns and reviewing the statistical support for our hypotheses, the discussion and implications of which are reserved for the subsequent chapter. Prelimmaiy Analyses and Demographics Although no willing participaiit was discouraged from filling out the survey, entitled “Attachment and Behavior within Romantic Rektioiisliips,” many of the questions asked about, and only really made sense for, those who were in a romantic reMonsMp at the time they were surveyed. P eA ^s in large part dee to tMs reason, Aere were a large number of pairticipaiits who left entire sections blank (n=43), and were dropped fi»m analysis. Therefore, in some sections, as notaJ, data aaoalysis will proceed on data obtaiiied from Aose paiticipaiits who boA: 1) completed at least seventy-five percent of Ae survey (n=318); an4 2) reported Aemselves as “currently in a serious romantic relationship” (n=199), as casual romantic relationsMps do not meet Ae standards for being true attechment relatioiisMps (Ainsworth, 19S9; Hazan & Shaver, 1987). A A o u ^ , at oAer times, specific comparisons between single and romantically involved psrticlpfflits are relevant and warranted. To briefly and broadly describe ow sample in demographic terms, females (66.6%) outnumbered males (33.4%) by a ratio of two-to-one; 3.8% of our participants were Afiicaii American, 1.9% Americaaa Indian, 13.8% Asian/Pacific Islander, 9% ffispanic/Latino, 69,4% WHte/Caucasian, Ae remainder selected Other or MuMethnic, R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 38 As anticipated, most of our participants (70.7%) were between the ages of 18 and 24 years old; 89.1% were under 30; the oldest reported participmt was 69 yeais of age. Almost two-titiirds (ii=199) described themselves as cuirently in serious romantic relationships. As for mmtal status, 89% reported they had never been married, 9% had been married once or were curreatly married, 2.5% twice, the rest declined to state. Furthermore, 19.6% were not In any sort of romantic relationsMp; 14% were in a casual/non-exclusive relationsMp; 40.2% were in serious/exclusive datmg relationships; 7.3% lived with their romantic partner; 2% were engaged but living separately; 3.7% were engaged and living with their fiancd; 13% were married Finaliy, it was a predominantly heterosexual sample (93.7%), w th only 2.6% classifying themselves as homosexual, 1.3% bisexual, and 2.3% uncertaia As indicated by thdr responses to Bartholomew’s RQ, 41% of participaiits were Swure, with 24% FearM, 16% Preoccupied, and 18% Dismissive, percentages generally consistent with previous research findings (e.g., Bartholomew & Horowitz, 1991). Furthermore, in accordance wifli previous findings (e.g., Breraian et al, 1998) of association between our attachaieat measuires (Barfliolomew’s RQ and Fraley & Shaver’s ECR-R): Secure Attachment was significantly negatively related to bolh Anxiety and Avoidance (See Table 1, next page); both Fearful and Dismissive Attachment Sfyles were si^ficantly positive correlates of Avoidance (ie,, discomfort with closeness, intimacy, dq»ndence upon others); while both Fearfii! and R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 39 PFCOccupied Attachment positively conrelated with Anxiety (i.e., fear or preoccupatioii with abandonment). Table 1: Coirelations between Internal Working Models and Attachment S^les TABLE 1 Secure FearM Preoccupied Dismissive Anxiety -.253*** .389*** _344*** -.028 Avoidance -.331*** .249*** .036 .220*** p<.001*** The lack of significant correlations - Preoccupied with Avoidance, and Dismissive with Anxiety - fits with previous research, as the Preoccupied style is chamrterized by excessive desire to bond or merge with attachment figure(s); and the Dismissive style evinces a denial, repression, or lack of anxiety about relationsMps (Brennan et al., 1998). Thus, our pattern of results mirrors previous findings. The following preliminaiy analyses were also conducted on our data in order to explore factors that may potentially influence the relationsMps we have anticipated in our hypothesized model. RelationsMp Status In accordance with predictions, we found that being in a serious and exclusive romantic relationship was associated with being more secure (i.e., more strongly agreed that description fit them) and/or less insecure (more strongly disagreed that those descriptions fit them). Furthermore, those who classified themselves into the more insecure styles were far less likely to be in serious and exclusive romantic relationsMps than their more securely attached counterparts (See Table 2, next page). R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 0 Table 2; Attachment Style fay Relationship Status TABLE 2 Not Currently CiHTratlyin Romantically Involved Serious RelationsMp Secure (M=127) 31(24.4%) 96(75.6%) Feaifiil(N=72) 34(47.2%) 38(52.7%) Preoccupied (N=49) 20(40.8%) 29(59.2%) iWsmissive (N=55) 27 (49.1%) 28 (50.9%) Theipe were also reported differences in strategy usage, with those not currently in relatioiiships reporting significantly greater tendencies to use both negative strategies (i.e., exit and neglect) in response to conflict within Iheir fsevious romantic relationships. (See Table 3) Table 3: Attachment Styles, Intemal Woridng Models, Behaviors, and Outcomes by Romantic TABLES Not Currently Romantically Involved Currently in Serious Romantic Relationship N = 116 N - 199 Mean St Dev. Mem St. Dev. Secure 4.41** 1.93 5.04** 1.90 Fearfiil 4.49** 2.05 3.77** 2.09 Preoccupied 3.74 2.04 3.36 2.14 Dismissive 4.30^ 1.91 3.85* 1.90 Anxiety 3.54*** 1.42^ 2-37*** 1.17^ Avoidmce 3.02*** 120*' 2.02*** 1.02 * - Exit 4.51*** 1.60 3.35*** 1.66 Voice 6.03 1.47 6.17 1.57 Loyalty 5.11 1.55^ 5.05 1.78^ Neglect 4.69*** 1.58 3.66*** 1.54 Ecpiity 4.08 0.72^ 3.99 0.41® ' Passion 4.80*** 1.35*' 5.88*** 1.06 Intimacy 5.31*** 1.18*- 6.39*** 0.78®' Commiteent 4.59*** 1.43*' 6 21*** 0.98® ' Love 4.90*** 122*' 6.16*** 0.88®' p<.05*, p<.01**, p<.O01*** (Note: Saperscffipt notations (*') beside StatKlard Deviattcm Scores indieate t o t equal variauces cannot be assamed for these gncwps; Levene’s Test t o Equality o f Variances signfficant at p<.05.) R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 41 FinaBy, as would be expecte4 those who were cmrmtly in reMonsMps reportai more love for their partoers (passion, intimacy, commitment) flian those who were answering questions based upon ended relationships and/or rclatioiisMps tiey did not consider to be serious or exclusive. Relationship, len gth As the process of becomii^ attached to one’s romantic partner takes time, generally considered to be about 2 years (e.g., Kazan & Zeifinan, 1994; Weiss, 1988), we explored this factor’s influence. In other words, vvhat differences do we see between those whose current serious relationship has been exclusive and ongoing for more than 24 months versus those whose current serious relationship has been exclusive for less than 24 months. We found that for those participants currently in romantic relationships, the median relationship length was 18 months, whereas the mean was 37 months. Close to sixty percent of the seriously romantically involved participants’ relationships were 23 months along or fewer (N=l 19), leaving a substantial number of participants in relationships at or over the 2-year mark (N=80). (See Table 4, next page) Assessing attachment profiles, those in relationships o f less than two years duration rated themselves as both less Secure and higher on all of the insecure descriptors (i.e.. Fearful, Preoccupie4 Dismissive) and higher in Anxiety and Avoidance, than did those in relationships at or over 2 years in length; however, tiiese differences were only significant for Dismissiveness and Anxiety. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 2 Table 4: Attadmient Style, Intemal Workteg Models, AccommodatioH, and Outcomes 1 Relationship Lengfii TABLE 4 <2 years N=119 >2 years N=80 Mean SO M «n SD S « ire 4.91 1.87 5.23 1.94 Fearfld 3.98 2.10 3.44 2.06 Preoccupied 3.59 2.17 3.00 2.05 Dismissive 4.09* 1.96 3.47* 1.75 Anxiety 2.52* 1.18 2.14* 1.11 Avoidmce 2.08 1.02 1.91 1.01 Exit 3.49 h i t 3.15 1.43 Voice 6.31 1.62 5.95 1.48 Loyalty 5.17 1.80 4.89 1.75 Neglect 3.76 1.61 3.52 1.43 Equity 4.00 0.45 3.97 0.33 Passion 5.94 1.04 5.80 1.10 Intiinacy 6.36 0.79 6.44 0.77 Ccmimitmsnt 6.07* 1.03 6.41* 0.87 Love 6.13 0.90 6.22 0.86 Asterisks iirficate titort means in this row aie significantly different irt p<.05* level? (Note; Sufserscript notattons (^) beside Stmdard Eteviation Scores indicate fliat equal varifflices camot be assumed f « these ^ lip s : Levene’s Test for Equality of Variances significant at p<.05.) As for our outeome measures, only commitment showed a significant difference based on relationship length, with those in relationships of greater than 2 years duration reporting significantly greater feelings of commitment toward their partoers than those in shoite relationships; however, fliis difference was not strong enough, to influence the overall love measure. S gvexlte Testing for mean differences between our onliiK and pen-fmd-p^ser surveys revealed some significant differences. The most significant of these was the age difference: our online participants were significantly older than those at the communi^ ^ Note: Fearful mai^iially s^nificant in hypothesized direction at p<.078; Preoccupied mafgimlly significant in hypoAesized direction at R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 3 college who participated in the pen-and-paper version. As would be expected, these older participants also tended to report having been in their cunent relationships for much longer periods of time, although due to the large variation on tins item, this difference (over 10 months) was not statistically significant (See Table 5) Table 5: Attachment Style, Intemal Woridng Models, Accommodation, and Outcomes by Type of Survey TABLES Online N=126 Pen and P ^ » r N=192 Mean SD Mean SD ...............Age................... 25.4*** 5.78 21.9*** 7.28 Current Rel (in months) 41.8 43.1 31.6 58.1 Secure 4.85 1.98 4.76 1.91 Fearfol 3.87 2.1 4.17 2.1 ft'eoccupied 3.56 2.10 3.48 2.11 Dismissive 3.76 1.87 4.18 1.93 Anxiety 2.58* 1.39 2.95* 1.36 Avoidance 220* 1.15 2.51* 121 Exit 3.58 1.72 3.94 1.74 Voice 6.16 1.41 6.07 1.62 Loyalty 5.03 1.71 5.10 1.69 Neglect 3.79* 1.46 4.21* 1.71 Equity 4.14** 0.40 3.95** 0.61 Passion 5.56 122 5.43 1.34 trttimacy 6.15* 1.04^ 5.89* 1.10^ Cammitment 5.78 1.40 5.49 1.42 Love 5.83 1.12 5.60 123 Asterisks indicate that means in this row are significantly different at p<.05* level. (Note; Superscript notatiom (^) beside Standard Deviation Scores indicate that equal vari®ces cannot be assumed for iiese groups: Levene’s Test for Equality of Variances significant at pS05.) Initially, our intent in runnii^ the online sample was based on the perception that there might be a greater proportion of insecure types to be found there than tonong a population o f college-attending students. However, m these analyses demonstrate, that was not found to be true. In fact, we found no significant differences between scores on self-reported attachment styles, using Bartholomew’s RQ; however, the R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 44 diflferences between ECR-R woridng models indicated greater amounts of both Anxiety and Avoidance among those who took the pen-and-p^jer version. To investigate the possibility that these attachment differences were attributable to age differences in the sample, possibly compounded by length of time in ciBTent relationship, we tested for mean differences between groups, using MANCOVA to control for age. S n this analysis, we found that although the main effect of survey Q ?p e was significant on its own for both dependent measures: Anxiety F (2 ,317) = 5.773, p<05; and Avoidance F (2,317) = 5.206, p<.05; the effect for anxiety ceased to be significant with Age as our covariate: F (2,304) Age = 6.818, p<.01; Anxiety = 3.821, p>.05. Furthermore, using MANCOVA to control for length of relationship revealed even more dramatic reductions for both Anxiety F(2,307) RetMioiiship Length = 15.257, p<.001; Anxiety = 3.356, p>.05; and Avoidance F(2, 307) Relationship Length = 15.257, p<.05; Avoidance = 3.442, p>.05. Thus, we feel confident that the online measure has in fact sraved to increase the diversity of our resoltiiiig sample as a whole. We teve, iierefoie, merged the data obtained from these two groups together for all subsequent analyses. Hypothesis 1 We predicted in Hypothesis 1 that secure attachment would predict more functional reactions to conflict; whereas, insecure attachment would predict less functional types of r efu ses. As discussed previously, secure individuals are more likely to be romantically involved at any given time, as they are better able to maintain R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 45 iieir romantic relationships. As insecure attachmeEt is linked to less fimctional behaviors (e.g., Donovan & lessor, 1985), we have provided resulting correlations for both those participants in serious romantic relationships (n=199) and all participants (n=318) to illustrate the comparisons. As anticipated, both attachment styles (i.e., Secure, Fearful, Preoccupied, Dismissive as measured by Bartholomew’s RQ) and intemal working models (i.e„ averages for Anxiety and Avoidance as measured by Fraley and Shaver’s ECR-R) predicted responses to conflict in theory-consistent ways; and these results were generally stronger when the full sample was used - thereby including a greater proportion of insecure participants. Secure attachment predicted si^iificantly lower reported use of neglect; vriiile reported use of voice was in the predicted direction, but fell short of achieving significance (p=.052) for those currently in relationships; lower use of exit was similarly significant overall, but insignificant (p=.068) for just our romantically involved group. Conversely, greater fearfulness predicted significantly greater use of both negative strategies. Greater agreement with the preoccupied description was a highly significant predictor of more reported use of voice; 'Pereas, dismissiveness was inversely predictive for the same strategy. Dismissivaiess also predicted greater reported use of botii negative strategies, although for neglect this difference was not significant (p=.092) for those currently in relationships; and predicted lower reported use ofboth positive strategies overall, and was significant for less use of voice for those currently in relationships. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 46 We found in these correiatioiial analyses (HI), tiiat both t ppeoc»iipieci attachment do Indeed tend to precipitate more enacteent of active, fimctional response to conflict (i.e., voice); althou^ secure attachment is the only negative predictor of the destaictive :^ponses (i.e., exit, neglect); while insecure attachment on the ■ w hole tends to generate behaviors which are less conducive to relationship maintenance (i.e., exit, ne^ect) and inhibit the performance of more pro relationship behaviors (i.e., loyalty). (See Table 6) Table 6: Responses to Conflict by Attachment Style and Internal Working Models TABLE6 EKit Voice Loyalty Neglect InRds (n=199) ALL (N=318) In Reis (n = l» ) ALL (N=318) In Reis (n=199) ALL (N=318) InRek (0=199) ALL (N=318) S eo « -.130 -.142* .139 .191*** .031 .045 -.190** -.151** F«rftd .183* .217*** -.013 -.010 -.112 -.062 .183* .160** fteocoped .001 .055 .176* .157** .059 .055 .019 .m Dismissive .157* .110 -.191** -.175** -.096 -.133* .120 .135* Anxiely .408*** .423*** -.099 -.048 -.063 .094 237*** 308*** A voickD oe .455*** .460*** -334*** ” 344*** -.234*** -.180** 355*** 319*** p<.05*,p<.01** p<.001*** The connections between these concepts is ftu rth er explained by examining correlations with Fraley and Shaver’s ECR-R dimensions: Anxiety predicte significantly greater usage ofboth exit and neglect; while Avoidance predicts both significantly greater exit and neglect scores, along with significantly lower reported use of voice and loydty. One surprising finding was the statistically significant positive correlation between preoccipied attachment and use of voice; which will be discussed in greater detail in Chapter 5. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 7 Hypothesis 2 Hypothesis #2a. Secure attachmeat will predict more positive love feelings; whereas, insecure attachmettt will predict less love within the relattonship. Hvpolhesis #2b. Greater anxiety will predict a heightened perception of benefittedness fiom a romantic reMonsMp. Data analyses also showed support for Hypothesis 2, indicating that Attechment and InteiMl Working Models predict Romantic Outcomes. Greater security was a significant positive predictor of higher reported feeling of passion, intimacy, and commitment; whereas, fearfiilness was a significantly negative predictor ofboth intimacy and commitment for our overall sample. Dismissiveness was also a significant negative predictor of passion, intimacy, and commitaient, as w eM as of underbenefittedness in ow equity measure for our overai sample. (See Table 7) Table 7: Relationship Outcomes by Attachment Style and Internal Working Models TABLE? ........ Equift. Passion Intlimcy Commiteent In Reis ALL In Reis ALL InRek ALL In Reis ALL {N=199> (N=318) (N=199) (N=31S) (N=199) (N ==318) (N=199) (N-318) Secure .065 -.004 .111 .192*** .250*** .313*** .159* .224*** Fearfal .025 .076 .031 -.038 -.106 -.160** -.076 .129* PreoccBpied .129 .083 -.002 -.026 -.027 -.083 -.031 -.034 Dianisslve -.052 -.111* -.193** -.239*** -.159* -.161** -.259*** -.254*** Anxiety -.014 .165»* -.206** -.207*** -.415*** -.426*** -.338*** • 2 8 B * * * Avoidance •MO .039 -.678*** -.690*** -.839*** -.801*** -.747*** -.688*** p^05*,p<.01** iK.001*** The basis for these connections becomes apparent when examining the underlying dynamics with the ECR-R measure: as both anxiety and avoidance averages were highly significant negative predictors for passion, intimacy, and R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 4 8 commitment Finally, anxiety was the only construct to show a significant positive correlation with equity (H2b), an otiierwise disappointing measure, as will be discussed later. Hypothesis 3 As predicted, behavioK were found to predict romantic outcomes in the following ways; Negative behaviors (i.e., exit and neglect) were both significantly negative predictors of all love items (i.e., passion, intimacy, and commitment). Whereas the positive strategies (i.e., voice and loyalty) significantly predicted more favorable relationship outcomes (i.e., higher scores for passion, intimacy, and commitment). (See Table 8) Table 8; Relationship Outcomes by Response to Conflict TAB1E8 ___ _ PassicM Mbmcy inRgls ALL In Reis ALL In Reis ALL In Reis ALL (b=199) (N=318) (ff=199) (N=318) (n=199) (N=318) (0=199) (N=318) E K it -.032 .058 -.360*** -342*** -.407*** -.416*** Voice .051 .014 319*** 315«* .328*** 309*** 2 7 7 * * * .263*** L qy£% -.069 .008 384*** .280*** .245*** .230*** 287*** 279*** -.072 -.014 -.0» -.186*** -.160* -228*** -.103 -.211*** p<.05*, p<.01**, fK.001*** Furthermore, the individual items comprising Sternberg’s Love Scale were also highly correlated with one another (.810 to .853, p<.001); therefore, a single itan “Love” > composed of tiiese three components (i.e., passion, intimacy, commitment) > has been used for model-testing in the final section of our analysis. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 49 Hypothesis 4 Our overall model showed pfomise, receiving good support for both the Secure and Dismissive attachment styles, as well as for the Internal Working Model: Anxiety about abandonment Current involveinent in a serious and committed romantic relationsMp ftmctioned as a moderator of these effects, as demonstrated k Table 8, with the strongest influence demonstr^ed by data from those in reMonsMps of less than two years duration (n=l 19). In this maimer, our hyfwthesized model was generally supported. In assessing our model using Sequential Regression Equations, where Atfswhmait items entered first (i.e.. Secure, Dismissive, and Anxiely in respective models), these significantly predicted the romantic outcome (i.e., love); however, our hypofliesis diat at least some behaviors - specifically the active strategies - mediated this relationship was supported by their subsequent entiy wMch reduced the predictive ability of Attachment, either substantially - mdicating partial mediation - or even to statistically insignificant levels - indicating M l mediation - in the following cases. (See Table 9, next page) Secure attachment was a significantly positive predictor of reported love scores, mediated by significantly lower reported use of exit and significantly greater reported use of voice in response to conflict For those in relationships of less flian two years duration, the resulting model (i.e., security of attachment combined with these active strategies of exit and voice) predicted almost a quarter of the variance in R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 50 reported love, a tremendous improvement over tib ie mere five percent accounted for by security alcme. Tafeief; Resute o f Hierarchical Regressions testing Hypothesized Mediational Model TABLE 9 Models R R S<juare Adjusted RSq. Std. Error Model Sifi. Stamfardized Coefficieate mm.... t Sig Not 1: Secure .217 .047 .039 1.1975 .020 217 2.357 .020 Romaiitically 2: Secure .447 200 .178 1.1073 .000 .125 1.409 .162 Involved -.178 -2.049 .043 .329 3.641 .000 Romaiitically 1: Secure .177 .031 .026 .8750 .013 .177 2.506 .013 Involved 2: Secure .450 .202 .190 .7981 .000 .105 1.613 .108 Exit -.282 -4227 .000 Voice .250 3.745 .000 Romantic 1: Secure .242 .059 .051 .8748 .008 242 2.701 .008 Rel < 2 years 2: Seam .508 258 238 .7836 .000 .134 1.617 .109 Exit -.246 -2.957 .004 Voice .335 4.002 .000 Not hDisnasv .186 .035 .026 12002 .046 -.186 -2.018 .046 Romantically 2:Disnusv .430 .185 .163 1.1126 .000 -.139 -1.588 .115 Involved Exit -.187 -2.140 .035 Voice .311 3.507 .001 Romantically l:Dismsv ,220 .048 .043 .8759 .002 -220 -3.114 .002 Involved 2;Diansv .457 .209 .196 .8027 .000 -.130 -1.962 .051 Exit -279 -4.150 .000 Voice .240 3.549 .000 Romantic l:Dismsv .232 .054 .045 .8843 .012 -.232 -2.553 .012 R ei< 2 years 2:Dlstnsv .505 .255 236 .7912 .000 -.117 -1.393 .166 Exit -.248 -2.935 .004 Voice .338 3.997 .000 Romaottc 1 "Anxiety .216 .047 .039 .8803 .018 -216 -2.398 .018 Rel < 2 years 2:Aicdety .505 255 236 .7848 .000 -.127 -1.500 .136 Exit -224 -2.586 .011 Voice .360 4.348 .000 The inverse connection between Dismissiveness and reported love was mediated by significantly greater rei»rted use of exit and signiflcfflitly lower reported use of voice in response to conflict. Ihese findings were most pronounced for those k R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 51 serious and coimnitted romaiitic relationships of less than two years duiatioE, again increasing the predictive power of the resulting model to well over twenty percent. Finaly, for those in serious romantic rclationsMps of less thffli two years duration, the ability of Anxiety to predict lower reported love scores w a® M ly mediated by its tendency to increase use of exit and reduce use of voice. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 52 CHAPTER 5: DISCUSSION This study sought to explore the complex relationships between and among romantic attachmentes strategic behaviors (in response to conflict), and relationship outcomes. In pursuiiig this objective, our research has provided new evidence as to how it Is fliat the attachment system exerts its influence over the quality and outcomes of romantic relationships, and offers insights regarding some of the mechanisms through which this occurs. TT ie resulte of our correlational analyses and hierarchical regression analyses employing attachment and behavior to predict relationship outcomes were consistent with theory-based expectations. Generally speaking, secure attachment stimulates more functional |wceptions of others and resulliiig behaviors within romantic reMonsMps; whereas. Insecure atfeichment gives rise to less benevolent perceptiom and actions within romantic partaersMps. It is these behaviors that in turn go on to eitho- support or destroy the relationship (see review by Feeney, 1999). Review of Findings m d Implications Our findings demonstrate that only throng consideration of flie impact attachment styles have on behavior - specificafly, the way romantic partners commiinicate or opt not to communicate - can their influence upon reMonsMp quality and outcomes be fully undeistood. Specifically, our results provide a new means of demonstrating how insecure attachment increases enactment of behaviors that are less conducive to the successfiil maintenance of romantic paitneisMps. Over time those R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 53 behaviors predict decreased love between the partners and often contribnite to the destruction of the relationship, thereby reifying the internal working models imderlykg the insecure styles, and thus reinforcing them .* ^ One surprising ftnding was the statistically significant i»sitive coirelatioii between preoccupied attachment and me of voice. Judging from the related ECR-R elements, this would not appear tied to their characteristically heightened anxiety levels (as this correlation was in the opposite direction, although statistically insignificant); therefore, it is most likely due to the inverse relationship between voice and avoidance. Other research (e.g., Kobak & Sceery, 1998) has classified preoccupied attachment as cliaracteri2»d by heightened awareness of threats to the relationship and greater demands on partner’s availability. Hence it does indeed follow that one who was hypervigilant to a partner’s perceived transgressions, combined with a characteristic desire to bond completely or merge with that partner, would be active in voicing concerns and pursuing solutions. It is also possible that although this is generally considered as a positive reactive strategy, it could become less so if overused, or be less fevorably received by some of the other insecure attachment styles (i.e., Dismissives), thereby reducing - instead of enhancing - quality and/or longevity of the romantic relationship over time. Indeed, Kobak and Sceery demonstrated that preoccupied individuals bad an elevated awareness and expression of negative feelings. Other researchers (e.g., Brennan & Shaver, 1995; Collins & * A point similar to that made by Coffins and Read in 1994, which described how working modds shape cc^nitioas. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 54 Rea4 1990; Kirkpatrick & Davis, 1994) have found certain types of pairings to be more likely than ofters (i.e., secures tend to pair with other secures, whereas preoccupieds often partner with dismissives). Further research is needed, preferably with both members of the romantic partnersMps, to detemiine how each strategy is perceived by each partner, and its effect on their relationship over time. Another unmticipated discovery was the degree to which behavioral effects were enhanced for those in serious romantic relationships of less than two years duration. According to prior studies (e.g., Kazan & Zeifinan, 1994; Weiss, 1988) these would be the romantic partnerships that have not yet had time to become full attachment relation^ps. This would tend to bolster our arguments in favor of viewing relationship maintenance strategies as adult correlates of childhood attachment behaviors, in that both tend to decline in frequency two to three years into the relationship (Berscheid, 1983). On the whole, our results were encouraging and supported the thinking behind our proposed model. Although we were unable to completely and unambiguously support the hypothesized mode! in its entirety, this study has indeed demonstrated the potential that such a mode! may eventually be derived. This model is inqx>rtant for those who wish to achieve - or help others achieve - more favorable relationship outcomes. By refiaming maintenance strategies as attachment behaviors, we can begin to see how the negative and potentially destructive behavioral strategies are emotions, and behaviois toward others based on atfachment-rekted w k tb a s in selective attention and lecaE R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 55 exerted not maliciously against one’s partner or with the intent to destroy the relationship, but rather as learned strategies for self-preservation. Alttiough dysfunctional for the current attachment relationship, the ways in which the behaviors a J B fe c t fliat relatioissMp w il tend to cohere with, and iiiis mnforce, expectations consistent with the attachment style precipitating them (the behaviors). Limitations First, although our measures were assembled jfrom the previously tested work of others, we respectfiilly suggest that some refinements to thes® me^ures stil be required. Among those we surveyed, mostly college students whom we have no reason to believe were in any way particularly deficient, we found that Bartholomew’s RQ created some confusion among too many of the participants such that several items were skipped leading to m issing data. Although we did refine the instructions and typographic layout through a series of informal pilots, we were unable to dispel all confusion in this regard. Also, some of our analyses were hampered by conceptual overlaps between scales, particularly the Exit items in the Accommodation Scale and Avoidance items in die ECR-R. Although it is no doubt tme that Attachment is well- measured by asking questions regarding both beliefs and actions, in practice this creates difficulties, as Exit - and even Neglect - become almost synonymous with Avoidance. In the final stages of our analysis (H4), we found that our Regressions using the ECR-R Attachment items to explore IWMs directly, were unable to circumvent the tautology. Thus, we were forced to rely on the older measure (RQ) and R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 56 conduct analyses based on the styles themselves, although the ECR is in many respects a generally better measure of attachment overall (e.g., see Brenriaa et al., 1998 for a review). Second, as demonstrated by our data comparisons, there are notoble diJBferences in the responses ftom those who are currently in romantic relationships and those answering based upon their memories of a prior relationship. It is impossible to determine conclusively whether these differences stem more from: 1.) a hindsight bias causing ended romances to be remembered more negatively; or, 2.) similarly, a recmcy-effect which would also create similar findings. Whereas, those in current relationships are likely to answer questions as to their behaviors in a recent conflict that did not cause the break-up of their relationship, those in ended relationships are more likely to recall die last conflict that caused the end of the relationship, thereby increasing the number of Exit responses among the latter group. Additionally, 3.) those who are generally more secure are able to create and sustain more lasting partnersMps, and thus more likely to be in a romantic relationsMp at any given time. Thus, by eliminating those who are not currently in romantic reMonsMps, we unintentionally and unavoidably ensure that the insecure types included into the study are less insecure tten those who were excluded. Finally, 4.) Biological factors may play a role: as love stimulates the release of endorphins, it may be that those who are currently romantically involved tend to perceive things more positively than do their single counterparts. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 57 Tibinis we have conducted all analyses and displayed results using twO“ tai!ed tests of our date, a more stringeiit standard than necessarily required as our theoretically based a priori hypotheses were directional. We have adopted this more conservative standard in accordance with guidelines laid out by Abelson (1995) in Statistics as principled m'gumenii as our model was exploratory, thus, we did not fee! it could “be convincingly argued tiiat an outcome in the wroi^ tail [was] meaningless and might as well be dismissed as a chance occmreiice” (p. 58). However, researchers following the course we have plotted in the current study may well be justified in considering either a one-tailed, or Abelson’s suggested “Lopsided Test” (p. 59), in attempting to replicate and improve on our findings, Fourtii, date was gathered over a period of three montte, during which time each participaiit was surveyed only once. It would vastly improve the power and generalizabilily of this model if dMa were to be collected over a longer period of time in a longitudiiial m annCT, surveying the same p a rlic i^ ts repeatedly to capture relevant changes (e.g., in their tehaviors, beliefe, attitedes) over the course of their relationsMps. This would enable paired comparisons as well as, potentially, increasing tie num te of opportunities for more insecure participants to be included, as iiey would have more opportunities to Ire k a romantic relationship and thus have their data included in the analyses. Finally, surveying botili members of each couple would enable another set of useM comparisons, and serve as a perception-check (e.g., perhaps “Participant X” R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 5 8 believes she is using Loyalty, but her partnef, “Y,” feels Neglected). Additionally, we could explore the extent to which attachment style of one’s partner also influences strategy selection (e.g., It would be difficult for a Secure to employ Voice with a Dismissive partner who has already Exited). Future Directions As well as correcting for these limitations, as described in the previous section, future studies interested in refining and/or elaborating on a model such as the one we have proposed would do well to focus attention on both the intent^ of strategy usage, as well as the perceived effectiveness of each strategy. Also, sampling methods should be employed in order to balance the number of attachment styles. We recommend oversampling the insecure-types, and/or utilizing a stratified sampling procedure in order to help ensure that there is good comparability of styles, among those who are currently in serious and committed romantic relationships, selected to participate. Although our initial sample size looked promising, the actual numbers used in analyzing our model dropped by one-third once those not currently involved in serious romantic relationships were eliminated. This also did not cause equivalent relictions across attachment styles; As those who are insecurely attached are less likely to pursue and/or be able to maintain romantic * Initially, Shis stucfy intended to do She fomwr by including a measure assessii^ tntrapersonal Cotamiinication Motives (ie„ pleasure, taclusion, control, relaxation, affection, escape); however, fliis was dropped early on as it only added to die error term and did not enhance our ability to predict or explain our Sndings. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 5 9 reMonsMps, their numbers were more drasticaliy cnrtoiled by tihis restriction. (See Table 10) Table 10: Attachment Style by Romantic Involvement TABLE 10 Single lotnairtically Involved Total Pereentageoflype currently in rels. Secure 31 96 127 75.6% Fearfiil 34 38 72 52.7% fteoccupied 20 29 49 59.2% Msmissive 27 28 55 50.9% (missing) 5 8 Furthemiore, it seems intuitively likely fliat - given these differences - it would be the least extreme of those classified as among the insecure ^?pes who would be more likely to be in a romantic relationship at any given time: that is, the most secure among die fearfiils, the least avoidant of the dismissives, the least anxious preoccupieds. If these attachment categories could be balanced to a greater degree, it may still be po^ible to acMeve greater support for our hypothetical model across a i categories. Additional behaviors should be considered in future studies as well. Ideally, we would eventuaiy tike to see Proactive Maintenance Strategies, such as positivity, opomess, assurances, social networks, and sharing tasks (Stafford & Cazmy, 1991) integated into iMs model. Finally, in the original incarnation o f this study, we had intended to allow participants to self-select the type of reMonsMp about wMch they answered questions regarding their use of accommodation tactics and relationsMp outcomes, based upon whom ihey identified as their current primary attachment figure. As interesting as it R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 6 0 may eventually be to include relationship type as an additional layer of analysis; it rapidly became exceedingly clear that doing so, at such m early stage of model” building, introduced too much '‘ noise” and reduced control and comparability to an unacceptable degree for the purposes of this cunrent study. Additionally p-oblematic in this regard were studies indicating very different choices for pritnaiy attachment figures are often reported by tiliose exhibiting diflferent atechment styles (e.g., Fraley & Davis, 1997; Freeman & Brown, 2001). Future research, however, is advised to inquire about responses to conflict in other attachment relationsMps (e.g., best Mends) in Older to assess the model’s generalizabUity. Conclusion This study has advanced a fledgling model of the way in which attachment ex«ts its impact on adult romantic relatiomMps by incorporating a subset of relational maiiitenance strategies (i.e., accommodation processes) as attachment behaviors. The conceptualization of accominodation processes advanced in this study has recast them from independent predictors of relationsMp outcomes, to mediators depmdoit upon Individual differences in parceptlon and Intent based upon attachmeiit styles. As siKh, we have refttttned at least the reactive mamtenance stetegies m attaciimeiit behaviors in order to gain insight into motivations for their use. IMs study is an attempt to make sense of myriad consistent findings across separate domaiiis, notably attachment theory and relationsMp maliiteiiiaiice studies. As such, it provides a fresh approach to and undearstanding of Me ways in wMch early R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 61 cMldhocKi experiences afiFect snbsequmt relationship ftmctionmg. It is hoped that this study will inspire future explorations of this type; thus, bridging the between underlying psychological factors motivating behaviors, piarticularly communication behaviors, and the effects those behaviors produce. Either, on its own, is only one-half of a richer and more interesting story, because understanding the reasons people communicate as they do, enhances our understanding of the process by vriiich relationsMps evolve and are maintained or destroyed. The ability to depict this process in a more fiilly detailed manner than heretofore available is valuable both for those interested in practical implications, such as relationship counseling, as well as for those interested in the process of communication itself. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 6 2 REFERENCES Abelson, R. P. (1995). Statistics as principled arsaiment Hillsdale, N J,: L. Erlbamn Associates. Amsworth, M. D. (1962). Reversible and irreversible effects o f maternal deprivation on inteilectaa! development. New York: Child Welfare League of America. Ainsworth, M. D. S. (1954/1972), Variables influencing the development of attachment. In C. S. Lavatelli & F. Stendler (Eds.), Readings in child behavior and development (3rd ed., pp. 193-201). Oxford, England; Harcourt Brace Jovanovich. Ainsworth, M. D. S, (1964). Patterns of attachment behavior shown bv the infant in interaction with his mother. Merrill-Palmer Quarterly, 10,51-58. Ainsworth, M. D. S., Blehar, M.C., Waters, E. & Wall, S. (1978). Patterns of Attachment: A psychological studv of the strange situation. Hillsdale, NJ: Erlbaum. Ainsworth, M. D, S. (1989). Attachments beyond infancy. American Psvchologist 44,709-716. Allen, J. P., & Land, D. (1999). Attachment in adolescence. In J. Cassidy, P. R. Shaver & M. Main (Eds.), Handbook of Attachment (pp. 319-335): Guilford Press. Ayres, J. (1983). Strategies to maintain relationships: Their identification and perceived usage. Commujnication Quarterly. 1,63-67. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psvchologv. 51, 1173-1182. Bartholomew, K. (1990). Avoidance of intimacy; An attachment perspective. Journal of Social and Personal Relationships. 7,147-178. Bartholomew, K., & Horowitz, L. M. (1991), Attachment styles amoi^ young adults: A test o f a four category model. Journal of Personality and Social Psvchologv. 61,226-244. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 63 Baxter, L. A., & Simon, E. P. (1993). Relationship maintenance stetegies and dialectical contradictions in personal relationships. Journal of Social and Personal Relationships. 10(2), 225-242. Bell, R. R., Daly, I A., & Gonzalez, M.C. (1987). Affinity-maintenance in marriage and its relationship to women’s marital satisfaction. Journal of Marriage and the FamMv. 49,445-454. Berscheid, E. (1983). Emotion. In E. B. H. Kelley, A. Christenson, J. H. Harvey, T. Huston, G. Levinger, E. McClintock, L. Peplau, & P. Peterson (Ed.), Close relationships (pp. 110-168). San Francisco: W. H. Freeman. Bowlby, J. (1944). Forty-four juvenile thieves: Their characters and home life. Internatjonal.Joumal o f Psycho-Analysis, 25,19-53. Bowlby, J. (1954/1973). The child's tie to his mother: Attachment behaviour. In C. S. Lavatelli & F, Stendler (Eds.), Readings.in.CMId Behavior aid Developmeiit (3rd ed., pp. 317-333). New York: Harcourt Brace Jovanovich, Inc. Bowlby, J. (1969/1972). Attachment and loss. Vol. 1 : Attachment (2nd ed.). New York: Basic Books. Bowlby, J. (1973). Attachment and loss. Vol. 2: Separation: Anxiety & Aneer. New York: Basic Books. Bowlby, J. (1979). IM..mai^.mdfaeaMng;.sf.a^ ^ London: Tavistock. Bowlby, J. (1980). Attachment and loss. Vol. 3: Loss: Sadness & Depression. New York: Basic Books. Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-Report Measurement of Adult Attachment: An Integrative Overview, In J. A. Simpson & W. S, Rholes (Eds.), Attachment theory and close relationships (pp. 46-76). New York: Guilford Press. Brennan, K. A. S., P. R. (1995). Dimensions of adult attachment, affect regulation, and romantic relationship functioning. Personality and Social Psychology Bulletin, 21(3), 267-283. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 6 4 Bretherton, I., & Munhoiland, K. A. (1999). Internal Working Models in Attachment Relationships: A Construct Revisited. In P. R. S. Jude Cassidy, and Mary Main (Ed.), Handbook of Attachment (pp. 89-111): Guilford Press. Canary, D. J., & Hause, K. S. (1993). Is there any reason to research sex differences in communication? Communication Quarterly. 4,129-145. Canary, D. J., & Stafford, L. (1992). Relational maintenance strategies and equity in marriage. Communication Monographs. 59(3), 243-267. Canary, D. J., & Stafford, L. (1993). Preservation of relational characteristics: Maintenance strategies, equity, and locus of control. In P. J. Kalbfleisch (Ed.), Interpersonal cominiinication; Eyolving interpersonal relationships. Communication (pp. 237-259). Hillsdale, NJ, England: Lawrence Erlbaum Associates, Inc. Canary, D. J., & Stafford, L. (2001). Equity in the preservation of personal relationships. In J. Harvey & A. Wenzel (Eds.), Close romantic relationships: Maintenance and enhemcement (pp. 133-151). Mahwah, NJ, US: Lawrence Erlbaum Associates. Canary, D. J., Stafford, L., Hause, K. S., & Wallace, L. A. (1993). An inductive analysis of relational maintenance strategies: Comparisons among lovers, relatives, friends, and others. Communication Research Reports. 10(1), 5-14. Cassidy, J. (1999). The nature of the child's ties. In P. R. S. Jude Cassidy, and Mary Main (Ed.), Handbook of Attachmeat (pp. 3-20): Guilford Press. Collins, N, L., & Read, S.J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58(4), 644-663. Collins, N. L. (1996). Working models of attachment: Implications for explanation, emotion, and behavior. Journal of Fefsonalitv and Social Psychology. 71, 810-832. Collins, N. L., Cooper, M. L,, Albino, A., & Allard, L. (2002). Psychosocial vulnerability from adolescence to adulthood: A prospective study of attachment style differences in relationship functioning and partner choice. |gm |a|,ofPg§oii#txani^^^ 70(6), 965-1008. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 65 Collins, N. L., & Read, S. J. (1994). Cognitive representations of attachment: The structure and function o f working models. In K. Bartholomew & D. Perlman (Eds.), Advances in Personal Relationships (Vol. 5: Adult Attachment Relationships, pp. 53-90). Avon: Bookcraft Ltd. Crowell, J. A., Fraley, R. C., & Shaver, P. R. (1999). Measurement of Individual Differences in Adolescent aid Adult Attachment. In J. Cassidy, P. R. Shaver & M. Main (Eds.), Handbook of Attachment (pp. 434-465): Guilford Press. Dindia, K., & Baxter, L. (1987). Strategies for maintaining and repairing marital relationships. Journal o f Social and Personal Relationships. 4,143-158. Donovan, J. E., & lessor, R. (1985). Structure of problem behavior in adolescence and young adulthood. Journal of.Consulting and Clioical...PsvchoiQgy, 53, 890-904. Feeney, J. A. (1998). Adult Attachment and Relationship-Centered Anxiety: Responses to Physical and Emotional Distancing. In J. A. S. W. S. Rholes (Ed,), A iachmen jl l eorym d,q s§ e . R eM ^ (pp. 189-218), New York: Guilford. Feeney, J. A. (1999). Adult Romantic Attachment and Couple Relationships. In J. Cassicfy, P. R. Shaver & M. Main (Eds.), Handbook of Attachment (pp. 355- 357). New York: Guilford Press. Feeney, J. A., & Noller, P. (1991). Attachment style and verbal descriptions of romantic partners. Journal of Social and Personal ReMonsMps, 8,187-215. Feeney, J, A., Noller, P., & Callan, V. J. (1994). Attachment style, communication and satisfaction in the early years of marriage. In K. Bartholomew & D. Perlman (Eds.), Attachment processes in adulthood (pp. 269-308). London, England: Jessica Kingsley Publishers. Feeney, J. A., Noller, P., & Roberts, N. (2000). Attachment and Close Relationships. In C. Hendrick & S. S. Hendrick (Eds.), Close relationships: A sourcebook (1 ed., pp. 185-202). Thousand Oaks: Sage Publications, Inc. Fraley, R. C., & Shaver, P.R. (2000). Adult romantic attachment: Theoretical developments, emerging controvemies, and unanswered questions. Review of General Psvcholoav. 4.132-154. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 6 6 Fraley, R. C., Waller, N. G., & Breimaii, K. A. (2000). Experiences in Close Relationships Revised (ECR-R). Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000, March 22,2003). An Item Response Theory Analysis of Self-Report Measures of Adult Attachment. Retrieved July 29,2003, from http://tigger.uic.edu/%7Efraley/measures/ecix.htm Freeman, H., & Brown, B. B. (2001). Primary attachment to parents and peers during adolescence: Differences by attachment style. Journal ofYouth and Adolescence. 30,653-674. Hazan, €., & Zeifrnan, D. (1994). Sex and the psychological tether. Advances in PersoBglM aiopAlBs, 5 ,151-177. Hazan, C., & Shaver, P. R. (1987). Romantic Love Conceptualized as an Attachment Process. Journal of Personality and Social Psychology. 52(3), 511 - 524. Hazan, C., & Shaver, P. R. (1994). Attachment as an organizational framework for research on close relationships. Psychological Inquiry. 5(1), 1-22, Hazan, C,, & Zeifrnsm, D. (1999). Pair Bonds as Attachments: Evaluating the Evidence. In J. Cassidy, P. R Shaver & M. Main (Eds.), Handbook of Attachment (pp. 336-354): Guilford Press. Julien, D., Markman, H. J., & Lindahl, K. M. (1989). A comparison of a global and a microanalytic coding system: Implications for future trends in studying interactions. Behavioral Assessments, 11,81-100. Kamey, B. R , & Bradbury, T.N. (1997). Neuroticism, marital interaction, and the trajectory of marital satisfactioii. Journal of Personality and Social Psychology. 72(5), 1075-1092. Keelan, J. P. R , Dion, K. L., & Dion, K. K. (1994). Attachment style and heterosexual relationshps among yound adults: A short-term panel study. 11,201-214. Kiecolt-Glaser, J. K., Glaser, R , Cacioppo, J. T., & Malarkey, W. B. (1998). Marital stress: Immunologic, neuroendocrine, and autonomic correlates. Annals of the New York Academy o f Sciences. 840,656-663. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 67 Kiecolt-Glaser, J. K. (1999). Stress, personal relationships, and immune fonction: Health implications. Brain, Behavior, and Immunity. 13,61-72. Kirkpatrick, L. A. D., K. E. (1994). Attachment style, gender, and relationship stability: A longitudinal analysis. Journal of Personality and Social Psychology. 66(3), 502-512. Kobak, R. (1999). The emotional dynamics of disruptions in attachment relationships: Implications for theory, research, and clinical intervention. In J. Cassidy, P. R. Shaver & M. Main (Eds.), Handbook of Attachment (pp. 21- 43): Guilford Press. Kobak, R. R., & Sceeiy, A. (1988). Attachment in late adolescence: Working models, affect regulation, and representation of self and others. Child Development 59,135-146. Langan, E. J. (2001). A friend like you: Attachment and maintenance strategies in young adult friendships. DAI, 62(08A), 222. Messman, S. J., Canary, D. J., & Hause, K. S. (2000). Motives to remain platonic, equity, and the use of maintenance strategies in opposite-sex friendships. Journal of Social and Personal.Relationships, 17(1), 67-94. Mikulincer, M., & Erev, 1 . (1991). Attachment style and the structure of romantic love. British Journal of Social Psychology. 30,273-291. Mikulincer, M. (1998). Attachment woridng models and the sense of trust: An exploration of interaction goals and affect regulation. Jouanal of Personality and Social Psychology. 74(5), 1209-1224. Montgomery, B. M. (1993). Relationship maintenance versus relationship change: A dialectical dilemma. Journal of Social and Personal Relationships. 10(2), 205-223. Noller, P., Feeney, J. A., Bonnell, D., & Callan, V.J. (1994), A longitudinal study of conflict in early marriage. Journal of Social and Personal Relationships. 11, 233-252, Pipp, S., & Harmon, R. J. (1987). Attachment as regulation: A commentary. Child Development. 58, WS-652. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. Pistole, M. C„ Clark, E. M., & Tubbs, A. L. (1995). Love relationships: Attachment style and the investment model. Journal of Mental Health Counseling, 17, 199-209. Rabby, M. K. (2001). Commitment and relational maintenance in online and offline romantic relationships. DAI, 62(06A), 191. Rholes, W. S., Simpson, J. A., & Blakely, B. S. (1995). Adult attachment styles and mothers' relationships with their young children. Personal Relationships. 2, 35-54. Rusbult, C. E. (1980). Commitment and Satisfaction in Romantic Associations: A Test of the Investment Model. Journal o f Experimental Social Psychology. 16,172-186. Rusbult, C. E. (1983). A longitudinal test of the investment model: The development (and deterioration) of satisfaction and commitment in heterosexual involvements. Journal of Personality and Social Psychology. 45,101-117. Rusbult, C. E. (1987). Responses to dissatisfaction in close relationships: The exit- voice-loyalty-neglect model. In D. Perlman & S. W. Duck (Eds.), Intimate relationships: Development dynamics, and deterioration (pp. 209-237). Newbury Park, CA: Sage. Rusbult, C. E., Johnson, D. J., & Morrow, G. D. (1986). Predicting satisfaction and commitment in adult romantic involvements: An assessment of the generalizability of the investment model. Social Psychological Quarterly. 49, 81-89. Russek, L. G., & Schwartz, G. E. (1997). Feelings of parental caring caa predict health status in mid-life: A 35-year follow-up of the Harvard Mastery of Stress study. Joumal of Beliavioral h^edicsne. 20,1-13* Simon, E. P., & Baxter, L. A. (1993). Attachment-style differences in relationship maintenance strategies. Western Journal of Communication. 57(4), 416-430. Simpson, J. A. (1990). Influence o f attachment styles on romantic relationships. Jpuinal of Person^ity.and, SM iti.PsYcMo.gy. 59,971-980. Simpson, J. A., & Rholes, W.S. (1994). Stress and Secure Base Relationships in Adulthood. In D. Perlman & K. Bartholomew (Eds.), Attachment in Close Relationships. London: Kingsley. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission. 69 Simpson, J. A., Rholes, W. S., & Phillips, D. (1996). Conflict in close relationships: An attachment perspective. Journal of Personality and Social Psychology, 71(5), 899-914. Simpson, J. S., Rholes, W. S„ Orifla, M. M., & Grich, J. (2002). Working models of attachment, support giving, and support seeking in a stressful situation. Personality and Social Psychology Bulletin. 28,598-608. Stafford, L., & Canary, D. J. (1991). Maintenance strategies and romantic relationship type, gender and relational characteristics. Joumal of Social aid PersonalMtiOMMB& 8(2), 217-242. Stembei^, R. J. (1986). A triangular theory of love. Psychological Review. 93,119- 135. Stemberg, R. J. (1988). The Triangle of Love. New York: Basic Books, Inc. Tabachnick, B. G., & Fidell, L. S. (1996). Using multiyariate statistics (3rd ed,). New York, NY: HarperCollins College Publishers. Vogl-Bauer, S., Kalbfleisch, P. J., & Beatty, M. (1999). Perceived equity, satisfaction, and relational maintenance strategies in parent-adolescent dyads. J oqmM. of Youth..and Adgjegt ^ ce, 28(1), 27-49. Waldron, V. R. (1991). Achieving communication goals in superior/subordinate relationships: The multi-fiaictionalily of upward maintenance tactics. Communication Monographs. 58(3), 289-306. Weiss, R. S. (1988). Loss and recovery. Joumal of Social Issues. 44,37-52. Yum, Y. O, (2000). Cross-cultural comparisons of links among relational maintenance behaviors, exchange factors, and individual characteristics in close relationships. DM, 61(04A), 145. R eproduced with perm ission of the copyright owner. Further reproduction prohibited without perm ission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Displaced aggression as a function of target power
PDF
Associations and mechanisms among attention deficit hyperactivity disorder symptoms, cognitive functioning, and drinking habits
PDF
Control, health perceptions, and information seeking as correlates of risky behavior in the context of AIDS
PDF
An evaluation of a brief HIV /AIDS prevention intervention using normative feedback to promote risk reduction among sexually active college students
PDF
Developing business communication skills: Leveraging stage versus global processes of change in skills improvement approaches
PDF
Behavioral and neural investigations of perceptual affect
PDF
Evidence for spontaneous situational inferences
PDF
Childhood videotaped neuromotor and social precursors of schizophrenia: A prospective investigation
PDF
A two-dimensional model of cognitive empathy: An empirical study
PDF
Consumer choice under budget constraint: Why consumers overspend
PDF
Factors that influence students' investment of mental effort in academic tasks: A validation and exploratory study
PDF
Experimental evaluation of a new intervention designed to promote dementia caregivers' acceptance and empathic responding towards care -recipients
PDF
Collective efficacy, anxiety, creativity/innovation and work performance at the team level
PDF
Examining the differences in family functioning and meaning in life between Reform, Conservative, and Orthodox Jewish families
PDF
Board involvement in monitoring and strategy making: Antecedents and consequences
PDF
Dopamine related temporal processing errors in prenatally stressed rats
PDF
Hyperactive symptoms, cognitive functioning, and drinking habits
PDF
Is there more to discrete prepulses than meets the eye?
PDF
English phoneme and word recognition by nonnative English speakers as a function of spectral resolution and English experience
PDF
Corporate ethical orientation and crisis management before and after September 11, 2001
Asset Metadata
Creator
Watson-Currie, Erica
(author)
Core Title
Attachment, accommodation, and the outcomes of romantic relationships
School
Graduate School
Degree
Doctor of Philosophy
Degree Program
Communication
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,psychology, cognitive,psychology, social,speech communication
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Miller, Lynn Carol (
committee chair
), Cody, Michael J. (
committee member
), Earleywine, Mitchell (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-530885
Unique identifier
UC11340827
Identifier
3140569.pdf (filename),usctheses-c16-530885 (legacy record id)
Legacy Identifier
3140569.pdf
Dmrecord
530885
Document Type
Dissertation
Rights
Watson-Currie, Erica
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
psychology, cognitive
psychology, social
speech communication