|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 THE EFFECTS OF ATTACHMENT STYLE AND CONTEXTUAL ACTIVATION OF ATTACHMENT SECURITY AND ATTACHMENT INSECURITY ON PARTNER EMPATHY AND DISTRESS RESPONSES TO EPISODES OF ROMANTIC PARTNER DISTRESS By STEPHEN CARROLL FIGLEY A DI SSERTATION P RESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DE GREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010
2 2010 Stephen Carroll Figley
3 To my wife, Joanna Grace the most encouraging person I know. You are a safe haven for count less women to whom you minister; a secure base from which I have been blessed to explore and grow.
4 ACKNOWLEDGME NTS I am grateful for the many people whose help and support made it possible for me t o complete this dissertation. First of all, I would like to thank m y committee chair, Dr. Harry Daniels, who persevered with me with much grace th rough multiple seasons of my pausing to live life befo re reengaging in the process Beyond his crucial instruction and guidance, he invested in me by choosing to study how I am wired, enabling him to strike the perfect balance of patience, challenge, and encou ragement, until a season of life more conducive to the task permitted the final push to completion. I would like also to thank my other committee members: Dr. Ellen Amatea, whose research and dissertation seminars were invaluable in preparing me for the challenge ahead ; Dr Pete r Sherrard, whose classes and teaching style were influential in helping me discover my passion for counseling couples; and Dr. David Mill er, whos e suggestions and encouragement relative to design and data presentation were very helpful And I would like to thank my longtime friend, Dr. Mark Atkinson of the College of Medicine, who out of a sacrificial commitment to my success, offered to join my committee so that he could encourage me to persevere and prod me on to completion. I did survive! I am g rateful also to the 132 couples who set aside the time and endured the emotional distress inherent in the study design to part icipate in this journey I am indebted to the pastors of the ten churches who entrusted me with acces s to recruit participants fr om among their congregants. I want to thank Candy Spires, who has been such a kind and gracious source of assistance through seemingly countless UF registration cycles, and Eric Th ompson, whose patient help with the data facilitated my learning to use SPS S freeing me to explore multiple additional avenues of data analyses on my own. I owe a debt of gratitude to my sec retary, Barb Wehrle, whose
5 reliability, sacrificial spir it, and commitment to me as a friend enabled me to balance the demands of completing this project with the need to sustain my livelihood and remain available to my clients. And I am grateful to the hundreds of client couples whom I have been privileged to counsel and/or teach over the years, for their often unwitting roles in refining my thinking about the processes that I have investigated. F or my family, who launched me on th e path to who I am becoming, I am ever grateful : My grandfather, Thomas Figley, who taught me how to frame a narrative; m y mother, Sharyn Figley, who taught me how to frame an argument; and m y broth er, Greg you left us far too early who taught me how to give freely. For those who are yet with me to celebrate this milestone, I share my gratitude: My father, Bill Figley, who has modeled for me faithfulness, pe rseverance and kindness ; m y remaining siblings Chris, Dawn, Lisa, and Tracy each of whom I deeply admire, and whose roles in my f ormative years facilitated my learning to explore the inner worlds of others rather than merely retreating into my own. Wor ds are in sufficient to express my gratitude to my wife Joanna, apart from whom this dissertation cou ld not have been completed, and to whom it has been dedicated. Her assistance with pr otocol administration, her boundless optimism, and her unswerving fait h in me have been a priceless gift. I am blessed beyond measure to be married to her. And I am so thankful for our children Jared and Sarah, whose depth of ch aracter and heart are amazing. What a privile ge it is to be their Dad! F inally, I reserve my d eepest gratitude for the unparalleled kindness of my Heavenly Father. Apart from His healing of my body when cancer struck ear ly in this process, this day would not have come. Apart from His healing in my inner world, I would have little to offer couples seeking to build a secure base and a safe haven.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Overview ................................ ................................ ................................ ................. 14 Statement of the Problem ................................ ................................ ....................... 23 Purpose of the Study ................................ ................................ .............................. 24 Theoretical Ration ale for the Study ................................ ................................ ......... 25 Significance of the Study ................................ ................................ ........................ 26 Research Questions and Hypotheses ................................ ................................ ..... 27 Limitations of the Study ................................ ................................ ........................... 29 Definition of Key Concepts ................................ ................................ ...................... 30 Organization of the Study ................................ ................................ ....................... 38 2 LITERATURE REVIEW ................................ ................................ .......................... 40 Romantic Relationships as Attachment ................................ ................................ .. 41 Attachment Style Continuity and Change ................................ ............................... 46 Internal Working Models of Attachment ................................ ................................ .. 50 Memory Systems Interplay in Attachment ................................ ............................... 57 Attachment Security and Prosocial Motivation ................................ ........................ 66 Secure Base and Safe ...................... 68 Summary ................................ ................................ ................................ ................ 71 3 METHODOLOGY ................................ ................................ ................................ ... 73 Participants ................................ ................................ ................................ ............. 74 Variables ................................ ................................ ................................ ................. 75 Dependent Variables ................................ ................................ ........................ 75 Independent Variables ................................ ................................ ..................... 75 Intervening Variables ................................ ................................ ........................ 76 Experimental Design ................................ ................................ ............................... 79 Research Questions and Hypotheses ................................ ................................ ..... 80 Instrumentation ................................ ................................ ................................ ....... 82 ................................ ................ 82 Pity Experience Inventories (PEI) ................................ ................................ ..... 86
7 Experienc es in Close Relationships Questionnaire Revised (ECR R) ........... 101 Procedures ................................ ................................ ................................ ........... 107 Recruiting Participants ................................ ................................ ................... 107 Obtaining Informed Consent ................................ ................................ ........... 108 Assigning Confidential Identifiers ................................ ................................ ... 1 09 Assignment to Experimental and Con trol Groups ................................ ........... 109 Scheduling Participation ................................ ................................ ................. 110 Experimental Protocol and Data Gathering ................................ .................... 111 Protocol administrator ................................ ................................ .............. 111 Experimental protocol overview ................................ ............................... 113 Experimental protocol session stages ................................ ..................... 114 Pilot study findings ................................ ................................ ................... 122 Data Analyses ................................ ................................ ................................ ....... 124 4 DATA ANALYSES AND RESULTS ................................ ................................ ...... 137 Description of the Sample ................................ ................................ ..................... 137 Reliability Findings ................................ ................................ ................................ 140 Indices Adapted Version (BEDI AV) ............ 140 Pity Experience Inventories Short Form Revised (PEI SF R) ........................ 141 Experiences in Close Relationship s Questionnaire Revised Adapted Version (ECR R AV) ................................ ................................ ................... 141 Description of the Data ................................ ................................ ......................... 142 Data Collection Activities ................................ ................................ ................ 142 Data Inclusion Decision Rules ................................ ................................ ........ 144 Data Analyses ................................ ................................ ................................ 144 Correlational data analyses ................................ ................................ ...... 145 Experimental data analyses ................................ ................................ ..... 157 Research Hypotheses and Questions ................................ ................................ .. 158 Summary ................................ ................................ ................................ .............. 165 5 DISCUSSION ................................ ................................ ................................ ....... 188 Literature Review ................................ ................................ ................................ .. 188 Methodology ................................ ................................ ................................ ......... 189 Study Participants ................................ ................................ .......................... 189 Study Procedures ................................ ................................ ........................... 190 Results ................................ ................................ ................................ .................. 193 Instrument Reliability ................................ ................................ ...................... 193 Effects of Attachment Style ................................ ................................ ............ 194 Effects of Attachment Context ................................ ................................ ........ 201 Effects of Gender ................................ ................................ ........................... 206 Other Effects ................................ ................................ ................................ .. 207 Relationship Communication History Survey (RCHS) ................................ .... 209 Correlates of distress frequency with desire for spousal comfort (Q#1) ... 210 Correlates of perceived spousal comforting competence ( Q#2) .............. 213 Correlates of spousal distress frequency with desire for comfort (Q#3) ... 214
8 tence (Q#4) ........... 216 Instrument Development ................................ ................................ ................ 217 Implications of the Study ................................ ................................ ....................... 223 Limitations of the Study ................................ ................................ ......................... 229 Suggestions for Future Research ................................ ................................ ......... 233 Summary ................................ ................................ ................................ .............. 234 APPENDIX A INSTITUTIONAL REVIEW BOARD APPROVAL FORM ................................ ...... 239 B LETTER TO PASTORS: PERMISSION TO RECRUIT ................................ ......... 240 C SAMPLE CHURCH RECRUITMENT ANNOUNCMENT ................................ ....... 241 D INFORMED CONSENT FORM ................................ ................................ ............. 243 E PARTICIPANT TASK #1 (ECR R AV) ................................ ................................ .. 245 F PARTICIPANT TASK #2 (DISTRACTER TASK) ................................ .................. 247 G PARTICIPANT TASK #3 (MEMORY SELECTION TASK) ................................ .... 248 H PARTICIPANT TASK #4 (GROUP 1) ................................ ................................ ... 249 I PARTICIPANT TASK #4 (GROUP 2) ................................ ................................ ... 250 J PARTICIPANT TASK #4 (GROUP 3) ................................ ................................ ... 251 K PARTICIPANT TASK #4 (GROUP 4) ................................ ................................ ... 252 L PARTICIPANT TASK #5 (MEMORY REVISITING TASK) ................................ .... 253 M PARTICIPANT TASK #6 (BEDI AV) ................................ ................................ ..... 254 N PARTICIPANT TASK #6 (PEI SF R) ................................ ................................ .... 255 O PARTICIPANT TASK #7 (RCHS) ................................ ................................ ......... 257 P EXPERIME NTAL PROTOCOL ADMINISTRATOR SCRIPT [7 PAGES] .............. 259 LIST OF REFERENCES ................................ ................................ ............................. 266 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 276
9 LIS T OF TABLES Table page 3 1 ............. 125 3 2 istress Indices Item Factor Loadings (1 of 2) ............. 126 3 3 Pity Experience Inventories Factor Loadings [Feelings Items] (1 of 2) ............. 127 3 4 Pi ty Experience Inventories Factor Loadings [Feelings Items] (2 of 2) ............. 128 3 5 Pity Experience Inventories Factor Loadings [Thoughts Items] ........................ 129 3 6 Pity Experience Inventories Factor Loadings [Wishes Items] ........................... 130 3 7 ECR R Normative Summary Statistics ................................ ............................. 131 3 8 Development o f Secure Attachment Prototype Priming Descriptor .................. 132 3 9 Development of Avoidant Attachment Prototype Priming Descriptor ............... 133 3 10 Deve lopment of Anxious Attachment Prototype Priming Descriptor ................ 134 4 1 Length of Current Marriage (Recruited Sample) ................................ ............... 138 4 2 Years of Marr iage 5 Year Increment Breakdown (Utilized Sample) ................. 139 4 3 Length of Current Marriage (Utilized Sample) ................................ .................. 139 4 4 Summary of Stages of Sa mple Reduction for Data Analyses ........................... 140 4 5 Reliability Comparisons of Revised /Adapted Instruments with Originals ......... 142 4 6 Overview of Pa rticipant Data Collection Categories and their Uses ................. 166 4 7 Pearson Correlations (1 of 7) ................................ ................................ ........... 167 4 8 Pearson Correlations (2 of 7) ................................ ................................ ........... 168 4 9 Pearson Correlations (3 of 7) ................................ ................................ ........... 169 4 10 Pearson Correlations (4 of 7) ................................ ................................ ........... 170 4 11 Pearson Correlations (5 of 7) ................................ ................................ ........... 171 4 12 Pearson Correlations (6 of 7) ................................ ................................ ........... 172 4 13 Pearson Correlations (7 of 7) ................................ ................................ ........... 173
10 4 14 Gender Comparisons for RCH Survey Questions ................................ ............ 17 4 4 15 RCH Survey Between Response Correlations: All Participants ........................ 174 4 16 RCH Survey Between Response Correlations: Husbands Only ....................... 175 4 17 RCH Survey Between Response Correlations: Wives Only ............................. 175 4 18 Correlation of RCHS Responses with Variables of Interest (1 of 6) .................. 176 4 19 Correlation of RCHS Responses with Variables of Interest (2 of 6) .................. 177 4 20 Correlation of RCHS Responses with Variables of Interest (3 of 6) .................. 178 4 21 Correlation of RCHS Responses with Variables of Interest (4 of 6) .................. 179 4 22 Correlation of RCHS Responses with Variables of Interest (5 of 6) .................. 180 4 23 Correlation of RCHS Responses with Variables of Interest (6 of 6) .................. 181 4 24 Descriptive Statistics for Attachment Style and Memory Measures .................. 182 4 25 Descriptive Statistics for Dependent Variab le Measures (Part 1) ..................... 183 4 26 Descriptive Statistics for Dependent Variable Measures (Part 2) ..................... 184 4 27 One Way Analyses of Variance (P art 1) ................................ ........................... 185 4 28 One Way Analyses of Variance (Part 2) ................................ ........................... 186 4 29 tests for Multiple Comparisons [Group Differences] .......................... 187 5 1 Distress Item Correlations with Attachment Avoidance [Av] ............................. 236 5 2 Distress Item Correlations with Attachment Anxiety [Ax] ................................ .. 237 5 3 Distress Items Discriminating Attachment Avoidance from Anxiety .................. 238
11 LIST OF FIGURES Figure page 3 1 Relationship between Attachment Security, Avoidance, and Anxiety ................. 77 3 2 Schematic of Experimental Design ................................ ................................ ..... 80 3 3 PEI SF R Instrume nt Development: Rationale and Process ............................ 135 3 4 Experimental Protocol Session Activities and Participant Tasks ...................... 136
12 Abstract of Dissertation Presented to the Gra duate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE EFFECTS OF ATTACHMENT STYLE AND CONTEXTUAL ACTIVATION OF ATTACHMENT SECURITY AND ATTACHMENT INSECURITY ON PAR TNER EMPATHY AND DISTRESS RESPONSES TO EPISODES OF ROMANTIC PARTNER DISTRESS By Stephen Carroll Figley December 2010 Chair: M. Harry Daniels Major: Marriage and Family Counseling The purpose of this study was to investigate the separate and interactiv e effects of attachment style and the priming of secure, avoidant, and anxious attachment mental and distress based emotions, cognitions, and motivations in response to romantic partner distress. Study design was informed by earlier studies investigating the effects of attachment security priming on altruistic response s plight. Published instruments were adapted or modified to assess attachment style and to measure empathy and distress responses to partner distress. The partner in distress stim ulus was achieved by participants imagining a recurrence of an earlier experienced episode of spousal distress accompanied by bids for partner support. Pen and paper responses were used to assess participant attachment style, empathy and distress responses, and vividness of engagement with the target stimulus. Participants were 132 married couples (264 individuals) recruited from ten churches in north central Florida. Couples were assigned to one of four groups : attachment security, avoidance, or anxiety priming conditions, or control. Group
13 assignment w as random, with stratification for length of current marriage. Responses to partner distress were found to be significantly correlated with both at tachment avoidance and attachment anxiety. Gender diff erences were found for the level of engagement measure. With the exception of one distress measure subscale, no group differences were found relative to attachment priming. Participants provided end of protocol assessments of self and other distress frequency and spouse comforting Inclu ded in th e data analyses were instrument development validation findings. Full sample distress item responses were correlated with the two insecure attachment measures Reported results included ranking of the 25 distress items as to their avoidance/anxi ety discriminating properties. Clinical implications were discussed, including psycho educational and skills building approaches suggested by f indings of gender differences in preferred emotion regulation strategies. D esign improvem ents for future resear ch were discussed, including the hypothesis that attachment priming too weak and perhaps too general to equally influence the more history laden complexities of partner in distress responses.
14 CHAPTER 1 INTRODUCTION Overview The roles and rules that define marriage relationships are both diverse and multi faceted. Common to most relationships, however, is the expectation that resources will be sha red and exchanged between the couple for the benefit of both parties to the marital contract. Some of the resources commonly shared and exchanged within a marriage are more tangible, such as property, money, and the division of labor. Others are less tan gible, such as the mutual meeting of sexual needs and the real or perceived social status that the relationship may confer. Among the less tangible resources shared and exchanged between couples is the giving and receiving of comfort and security during t imes of distress. Although not easily quantified, the capacity to share and exchange such resources has been identified by some researchers (e.g., Gottman, 1994; Mikulincer and Shaver, 2007) as among the attributes most pivotal to marital satisfaction and stability. The impulse to derive comfort and security from another when distressed begins with the child to parent relationship. Attachment theory, as a way of viewing the nature of the bond between parent and infant, was first formulated by John Bowlby in the mid twentieth century (1969/1982, 1973, 1980) and was empirically tested by Mary Ainsworth (Ainsworth, Blehar, Waters, & Wall, 1978) and the many whom she mentored over the ensuing three decades. Mary si gnificant contributor to the understanding of attachment relevant cognitive and emotional schema, through which earlier attachment experiences with parents influence both later parenting approaches and romantic attachments. Examining these causal
15 links, M ain and her colleagues offered a refined articulation of attachment theory through development of an instrument for measuring attachment state of mind. This instrument, The Adult Attachment Interview, facilitates examination of internal working models of attachment, through analyses of narrative discourse the interviewee provides about family of origin experiences (Hesse, 1999; Mikulincer & Shaver, 2007). Going beyond earlier theorists (e.g., Freud) who emphasized the primacy of biological needs in shaping the child to teens convinced him that there were more complex factors at play. Through extensive readings and interactions with naturalists, Bowlby became convinced that the felt need for security was primary seeking the safety of a stronger other during times of danger. Through his work with orphaned and parenting deprived children in the aftermath of World War II, Bowlby became persuaded that the need for affect regulation, despite the absence of physical during times of distress (1969/1982, 1973, 1980). Bowlby theorized that a child develops an attachment behavioral system that competence. Without the complementary attachment system, however for protection when environmental dangers exceed individual competence (1969/1982). tical constructs,
16 regulation when seeking comfort and security during distress (e.g., Kobak et al., 1993). Of particular interest was the apparent formation of emotion r egulation strategies when behaviors directed at obtaining adult comfort were blocked or unpredictably rewarded (Cassidy, 1994; Kobak, 1999). In the foundational investigations of this phenomenon, Ainsworth and her colleagues devised an experimental sett ing (dubbed the Strange Situation) intended to induce mild to moderate anxiety in one year olds by introducing a stranger into the room with mother and child, and by briefly and alternately removing the mother and both the mother and the stranger from the evidenced mild levels of distress and smooth re when reunited, a significant minority did not. Of those not happily restored to emotional equilibrium, two opposing st rategies for achieving affect regulation were observed: Some of the infants seemed to exaggerate their attachment needs in an apparent effort to coerce responses of comfort from their mothers, while others seemed to suppress their need for soothing as a me ans of distracting themselves from feelings of distress. In a search for links between parenting style and infant attachment behaviors, Ainsworth and her colleagues correlated these differing infant behavior patterns with their earlier extensive observati ons of first year infant/parent interactions (1978). attachment behaviors (1978), oth ers have examined attachment patterns in later childhood (e.g., Thompson, 1999), during adolescence (e.g., Allen & Land, 1999), and throughout adulthood (e.g., Hazan and Shaver, 1987; Mikulincer & Shaver, 2007).
17 Bringing a sense of coherence to the burgeo ning study of attachment related contribute to attachment style stability across the life span and among differing underlying processes led her to conclude that internal working models of attachment possess both cognitive and affective components, they guide attachment relevant behaviors, they exist outside of consciousness, and they have a propensity for stability (19 99). Such conclusions suggest that efforts to influence maladaptive and change resistant patterns of attachment style continuity are unlikely to be successful unless those efforts target the underlying mechanisms by which such stability is maintained, bot h over time and across multiple relationships. Memory research provides such a window into the formation, refinement, and maintenance of internal working models of attachment. Social scientists investigating memory function propose that the brain utilizes two distinct types of memory stores. These have been labeled episodic (or autobiographical) memory and semantic (or procedural) memory. Episodic memory is that store of what one knows on the basis of personal experiences. This type of memory is defined by one s ability to visualize oneself in the midst of an event remembered. Semantic memory is that summary compilation of all that one has learned from any source. Information learned by a variety of means (not personally lived) is categorized i n semantic memory, along with summary and interpretational data (or decision rules) derived from personal experiences (episodic memory) (Wheeler, 2000; Tulving, 2001).
18 Attachment theorists propose that semantic memory decision rules are the building blocks of one s internal working model s of attachment. Such decision rules do not dictate attachment behavior, but rather they serve as lenses through which one views attachment relevant situations. And one s lenses bias strategies for implementation of attach ment relevant behaviors. In other words, what one has interpreted about the self and others in earlier attachment relevant events (attempts and outcomes by self and/or other) becomes a guide for current and future behaviors (Tulving, 2001). The problem p osed by working models constructed under suboptimal earlier attachment experiences, however, is that the validity of mental representations formulated from generalizations about self and others may be clouded by the limited knowledge and perspective of the one experiencing those attachment events. Yet the conclusions from these earlier attachment events help to shape a mental model of attachment in ways that guide current and future attachment relevant strategies (Thompson, 1999). This perspective on how i nternal working models of attachment are formulated and subsequently refined presupposes that earlier attachment experiences with parents and other significant caregivers serve to shape later attitudes toward attachment related needs, expression of those n eeds, and strategies for ensuring that those needs are met. Viewed from the vantage of interacting episodic and semantic memory stores, emotion punctuated cognitions are encoded in episodic memory relative to attempts and outcomes of previous attempts to derive security and comfort from a caregiver during times of distress. Interpretational analyses of those events (how to meet needs and/or how to avert threats) are encoded in semantic memory. And subsequent attachment
19 relevant contexts serve to heighten or restrict access to and retrieval of the episodic memory linked semantic memory decision rules that one perceives (whether consciously or unconsciously) as most analogous to the situation at hand (Bowlby, 1973; Main, 1999; Mikulincer & Shaver, 2007). In harmony with this understanding of how internal working models of attachment are formulated, one s predominant strategy for approaching attachment relevant events is shaped by the preponderance and/or magnitude of differing quality attachment experience s encoded in memory It is also shaped by one s interpretations about self and other that are generalized from those experiences, accompanied by default strategies for regulating em otions linked to those events in memory The patterns of cognition, emoti on, and behavior arising out of one s internal working model s of attachment are referred to as one s (global or default) attachment style. Attachment theorists (e.g., Bowlby, 1969/1982, 1973, 1980) distinguish three primary attachment styles: (a) Secure a ttachment; (b) Avoidant (or anxious avoidant or def ended or dismissive) attachment; and (c) Anxious (or ambivalent or anxious ambivalent or preoccupied or resistant) attachm ent. Securely attached individuals operate from an internal working model with a g enerally favorable concept of both self and others involving attachment related needs. That is, they see themselves as generally worthy of receiving comfort and security when expressing the need, and see others as generally possessing the good will necess ary to provide a helpful response to the expression of attachment relevant needs. Avoidantly attached individuals see others as generally unreliable in the meeting of attachment related needs, or responding in ways that feel too intrusive. While seeing t hemselves
20 (or attempting to frame themselves to themselves) in a generally favorable light, they tend to suppress attachment related e motions, choosing to distract themselves from reflecting on attachment related needs rather than relying on others when di stre ssed. By contrast, anxiously attached individuals tend to see themselves as less than able in attachment relevant contexts, and feel a limited ability to self soothe. They see others as generally unpredictable or unreliable sources of comfort during times of distress, yet they perceive others as essential to the meeting of their own needs for affect regulation. Exaggerating emotions of distress in order not to be overlooked by the target of their bids for comfort, they nevertheless exhibit an approac h avoidance conflict because the responses they need cannot be trusted to foll ow from unavailable or dismissive caregivers (Bowlby, 1969/1982, 1973, 1980). eventually expanded upon b oth by him and by those who built upon his work. Presently, attachment theorists think in terms of a much broader range of emotional needs compelling one to seek out an attachment figu re when feeling distressed. Two helpful way s for cons idering this brea dth of needs come through the lens es (Belsky, 1999; Fonagy, 1999). resources are necessary to achieve basic surv ival and safety needs, more and more energies are diverted toward the intangibles of affirming one s worth and significance. It might be assumed, then, that given the environmental/cultural availability of resources (and one s competence in obtaining thos e resources when n eeded), matters occupying
21 loftier distress and the resultant need for security and comfort from an attachment figure. hat there are specific life tasks that one is compelled to master at each stage of individual development. Reasonable mastery of these stages is a necessary prerequisite to the establishment of healthy attachment relationships in early adulthood (Intimacy vs. Isolation). Looking at childhood attachment experiences with Erickson in mind illustrates the breadth of the various sources of distress that a child might be experiencing and the criteria by which the child might assess outcomes to attempts to deriv e security and comfort from another. For example, an avoidantly attached child might suppress the need for security and comfort during distress because of an angry or impatient response by a parent. But such a child might suppress such a need as well if the parent tended to respond in ways that he associated with diminishing his emerging autonomy or competence. In other words, the anticipation of an overdone response that interfered with how the child wanted to view his emerging autonomy might be regarde d as every bit as unsafe (and therefore every bit as undesirable) as a directly hurtful response. Such an internal working model of attachment might prompt the child to suppress attachment related emotions and alter attachment related behaviors. Although evolving with age, need hierarchies and competing life stage goals remain relevant throughout the life span (Belsky, 1999), and Over the past two plus decades, attachment theorists have begun to explore romantic relationships through the lens of attachment relationships (e.g., Hazan & Shaver, 1987, 1990; Feeney, 1999; Shaver & Mikulincer, 2006; Mikulincer & Shaver,
22 2007). Romantic relationships meet the four criteria that Bowlby (1 969/1982, 1973, 1980) set forth to differentiate attachment relationships from other types of relationships: (1) proximity maintenance (who do you like to spend time with or to be near?); (2) safe haven (who do you turn to during times of heightened distre ss?); (3) separation distress (who do you hate to be away from or miss the most during separations?); and (4) secure base (who can you always count on when you need them?). Romantic relationships differ from parent to child attachment relationships in the reciprocal nature of the attachment bond. In conceptualizing couple romantic relationships as attachment relationships, it becomes evident that internal working models of attachment formed during childhood influence how one elicits the meeting of attachme nt needs from one s partner, how one responds to such requests, and how the goals of affect regulation throughout the process shape both approaches to and outcomes from one s partner (Hazan & Shaver, 1987; Hazan & Zeifman, 1999). Ideally, these processes interact in such a way that the distressed partner nor feels compelled to be dismissive in achieving affect regulation (Mikulincer & Shaver, 2007). Requests for empathy from a romantic partner are more likely to be adversely affected, however, when one or more partner operates from an insecure internal working model of attachment. To respond with empathy, avoidant partners must resist their own self protective strategies of emotionally distancing from a distressed partner as a means of regulating their own discordant emotions. By co ntrast, anxiously attached partners may find themselves engulfed in their own feelings of distress, leaving lim ited energies available for providing
23 the comfort elicited by their partners (Mikulincer et al., 2001; Mikulincer & Shaver, 2007; Shaver & Mikulincer, 2007) A better understanding of how couple attachment dynamics are influenced by earlier at tachment me ntal representations can help to specific attachment contexts less likely to be hampered by earlier insecure attachment experiences Statement of the Problem Among the strategies employed by mar riage counselors are those intended to enhance empathy exchange during times of partner distress. The likelihood of receiving empathy from one s relationship partner when distressed is a significant correlate of relationship satisfaction and stability (e. g., Feeney, 1994; Davila & Fincham, 1998; Davila et al., 1999; Marchand, 2004; Ben Ari & Lavee, 2005; Birnbaum, 2007). Couple communication that reflects this type of mutuality is considered by some researchers to be the variable most predictive of relati onship viability (e.g., Gottman, 1994). The propensity for providing supportive responses to others is affected by one s lifetime of attachment relevant experiences and how those experiences have been interpreted (Bowlby, 1980; Mikulincer & Shaver, 2007). Because internal working models of attachment are long in the making, counseling approaches geared toward mediating the effects of insecure attachment styles through the reprocessing of earlier attachment experiences vary in their effectiveness and tend not to be brief. This problem argues for the implementation of briefer therapy approaches designed to minimize the negative effects of insecure working models of attachment on couple interactions. tionally Focused Marital Therapy (Johnson, 1996, Johnson & Whiffen, 2003) utilize attachment informed
24 research to formulate strategies for helping couples create affirming communication contexts. Recent studies have probed even deeper into the internal em otional priming of secure attachment representations affects those emotional processes (Mikulincer et al., 2001; Bartz & Lydon, 2004; Mikulincer et al., 2005). Yet the de sign of these studies limits the generalizability of findings to couples and their relationship contexts. Purpose of the Study Recently, researchers of adult attachment have discovered that the priming of secure attachment memory representations helps to increase one s empathy related 2001; Bartz & Lydon, 2004; Mikulincer et al., 2005). The findings of these studies suggest that the bringing to conscious awaren ess of embedded secure attachment memory representations may help to mediate the effects of more g lobal insecure attachment orientations the studies sampled from populations of couples. Populations sampled in each study were university students or other young adults. And (2) None of the study designs made romantic partner communication exchanges their focus. Each study used the plight of a stranger as the stimulus that elicited empathy and distress respon ses from participants. T he proposed stu dy represents an attempt to build upon the methodologies of these earli er studies while using a sample of adult couples in monogamous relations hip s and the plight of own romantic partner as the empathy and distress eliciting stimulus. The purpose of the proposed study then is to investigate how the separate and interactive
25 along with the primi ng of secure and and distress Theoretical Rationale for the Study The theoretical rationale for the current study builds upon the foundation of attachment theory in general as developed by John Bowlby (1969/1982; 1973; 1980) and investigated by Mary Ainsworth and her colleagues (Ainsworth, Blehar, Waters, & Wall, 197 8). It further builds upon the rationale firs t forwarded by Hazan and Shaver (1987) that adult romantic relationships can be conceptualized as attachment relationships, albeit with reciprocal bonds, rather than the unilateral care giving bond s of parent to child relationships. Even more specifically the theoretical basis for this study builds upon the extensive work of Mary Main (1999), examining how internal working models of attachment shape approach and avoidance behaviors in attachment related contexts. Within and extending the framework of lit erature investigating attachment working models, the theoret ical rationale for the current study is supported by an examination of c ognitive and affective processes in attachment, as well as interpersonal and intrapersonal regulation in c It builds upon investigations that illuminate how the interplay of episodic and sema ntic memory processes shape at tachment relevant behaviors, and how attachment security affects prosocial (altruistic) motivations. And finally, the theoretical rational e for the current study rests upon findings suggesting that the heightening of a sense of felt attachment security increases the current study is to test the applicab ility of these latter fi ndings to contexts of romantic partner distress accompanied by bids for support.
26 Significance of the Study This study may prove valuable for researchers of adult attachment, marriage and family therapists, their client couples, and others who may subsequently benefit from any future presentation of study findings through various media or seminar formats. Researchers of adult attachment may benefit from any advancement this study may bring to the knowledge base of attachment theory, emotion regulation theory, or couples communication theory. These themes have already been explored in a myriad of ways and in various combinations, yet adult attachment literature reveals a paucity of studies where the interface of these areas of inquiry has been explored with couples populations, and where the design of such studies facilitate the investigation of couple relationship focused interactions. Marriage and family therapists may benefit from this study by expanding their theoretical und erstanding of the interaction of client memory embedded attachment representations and how the accessibility of these networked images serves to cue or to disarm the need for self protective emotion regulating responses in the face of partner distress. En hanced insight into the interweaving of these cognitive and affective processes and the interpersonal and intrapersonal processes they motivate may serve to inform counseling strategies that help couples improve their mutual provision of support during tim es of distress. Client couples may benefit from this study by becoming the recipients of counseling approaches that teach them how to co construct here and now contexts in distress evoking interactions in such a way that embedded insecure working mod els of attachment are less likely to interfere with the exchange of empathy. Adult attachment theory informed counseling approaches currently benefit from theoretical findings
27 ped that the findings of the current study might add to that theory base by further informing approaches for teaching couples emoti on regulation strategies that are less likely to interfere with the giving and receiving of empathy. Finally, this study may benefit couples who do not formally seek professional integrated into presentations of couple communication enhancement themes. Or it may benefit them through their enco untering of any future print or electronic media findings. Couples volunteering to participate in this study were invited to attend one of the free one day communications semi nars presented by the principal investigator. These seminars were developed to provide participants with couples attachment theme based lectures and experiences, informed by the design of this study and presented in a user friendly format. Research Ques tions and Hypotheses The following research questions and hypotheses will be examined and tested in the proposed study: A: Does the priming of attachment security or i nsecurity differentially affect the experience of empathy and distress in response to on e s distressed marriage partner? H1: The independent variables of experimental or control condition will have no effect on the dependent variable measures of empathy and distress responses to partner distress. Alternative hypotheses are: a: Attachment s ecurity priming condition will result in higher empathy positive scores (empathy score minus distress score) than will attachment avoidance priming condition, attachment anxiety priming condition, or control condition
28 b: Attachment avoidance priming condi tion will result in lower empathy scores than will attachment anxiety priming condition c: Attachment anxiety priming condition will result in higher distress scores than will attachment avoidance priming condition B: Does gender differentially affect th e experience of empathy and distress in response to one s distressed marriage partner? H2: The independent variable of gender will have no effect on the dependent variable measures of empathy and distress responses to partner distress. Alternative hypoth eses are: a: Females will exhibit higher empathy scores than will males in response to marriage partner distress b: Males will exhibit greater use of distancing strategies than will females in response to marriage partner distress C: How does attachmen t style interact with contextual attachment security or insecurity priming to affect the experience of empathy and personal distress in response to one s distressed marriage partner? H3: The intervening variable of attachment style will have no effect on the dependent variable measures of empathy and distress responses to partner distress. Alternative hypotheses are: a: Higher attachment avoidance scores will predict lower empathy positive scores (empathy score minus distress score) for all three experim ental conditions and control condition in response to marriage partner distress b: Higher attachment anxiety scores will predict lower empathy positive scores (empathy score minus distress score) for all three experimental conditions and control condition in response to marriage partner distress c: Higher attachment avoidance scores will predict greater use of distancing strategies in response to marriage partner distress when controlling for experimental condition and gender D: Does attachment avoidance promote suppression of partner in distress memory vividness as an emotion regulation coping strategy? H4: Attachment avoidance will have no effect on the vividness of partner in distress memories. Alternative hypotheses are:
29 a: Higher attachment avoidan ce scores will predict lower partner in distress memory vividness ratings when controlling for experimental condition and gender b: Attachment avoidance priming condition will result in lower partner in distress memory vividness ratings when controlling f or attachment style and gender c: Males will exhibit lower partner in distress memory vividness ratings when controlling for experimental condition and attachment style Limitations of the Study The design of the current study permits the extension of ear lier specific relationship dynamics. teractions. Among those participants were self selected by volunteering in response to recruitment efforts; (b) participants were recruited from populations identifyi ng themselves with organized Christian churches; (c) participants were recruited from a mid sized urban population in the southeast region of the United States; (d) participants were married, and therefore ostensibly identified themselves as heterosexual c ouples; and (e) the racial composition of the sample was disproportionately Caucasian. milieus are the following: (a) data was gathered through pen and paper measuremen t instruments, rather than through direct observations of behavior; (b) participant self report responses created potential for measurement bias; and (c) dependent variable measurements assessed present responses to imagined recurrence of past episodes
30 rat her than present responses to here and now (outside of memory) experiences. Because of these limitations, care must be given when generalizing the findings of this study to various populations and interactional milieus. Definition of Key Concepts A ltruistic (or pro social) emotional state: A n experiential condition in which one is freed to respond supportively to a distressed other because energie s are not consumed in personal distress related emotion regulation. Anxious (or ambivalent or anxious a mbivalent or preoccupied or resistant) attachment: Refers to an attachment style characterized by a view of the self as inadequate to self comfort, and a view of others as unreliable sources of needed support. Affect regulation strategies include exaggera tion of attachment related emotions to coerce supportive responses from unavailable others. Over identification diminishing ability to respond empathically. A ttachment behavioral system : A complex arr ay of goal directed behaviors designed to ensure the accessibility of security restoring responses from caring others when needed. See further discussion under exploratory system. A ttachment bond : D enotes that aspect of the relationship between two peopl e the case of a healthy parent/child relationship, the child is said to have an attachment bond with the parent, while the parent is said to have a care giving bond w ith the child. In the case of healthy romantic relationships, the attachment bond is reciprocal. Attachment categories : Classifications defined by one s generalized concepts of self as needing and worthy of supportive responses from others when distresse d, and
31 others as reliable sources of such support. These beliefs shape strategies for securing needed support and/or for the regulation of emotions in response to its perceived unavailability. Attachment context : R efers to a here and now setting between two people who share an attachment relationship. Accessibility of secure and/or insecure attachment memory representations interact with attachment context to bias the interpretation of that context thus shaping the manner of expressing and responding to attachment needs. A ttachment figure : S omeone regarded by another as a likely or necessary source (in the case of a parent or caregiver) or a chosen source (in the case of a romantic partner) of access to comfort and security during times of distress. Attachment insecurity : R efers to attachment avoidance, attachment anxiety, or a combination of the two. Attachment insecurity divergent items : P ersonal distress sampling items in the Pity Experience Inventories Short Form Revised that effectively disting uish between attachment avoidance and attachment anxiety motivated responses to a distressed other. Attachment insecurity homogenous items : P ersonal distress sampling items in the Pity Experience Inventories Short Form Revised that effectively distinguis h between attachment security and attachment insecurity motivated responses to a distressed other, but do not effectively distinguish between the two different manifestations of attachment insecurity.
32 Attachment priming (or contextual activation of attach ment representations) : T he process of facilitating the accessibility in memory of a specific category of attachment relevant event s Clinically, this represents efforts to help couples access past attachment memories where expressed attachment needs were affirmatively met thus enhancing the expectation of similar outcomes in the present. A ttachment relationship : O ne in which both members of the relational dyad utilize the other as sources of comfort and security. Attachment relevant (or attachment rel ated) behaviors, experiences, situations, or events : T hose in which the obtaining of comfort and security from a close other is central to what motivates the behavior, defines the experience, or characterizes the situation or event A ttachment strategy : A conscious or unconscious behavioral schematic whose goal is enhancement of the likelihood that one s attachment related needs will be met. Attachment strategies are shaped by interpretation of outcomes of past attempts at getting attachment related need s met, and are subject to interpretational bias. Attachment style (or attachment state of mind) : R efers to one s predominant default strategy for the meeting of attachment related need s Attachment style is usually expressed as one of three categories (s ecure, avoidant, or anxious) with a fourth category (disorganized) representing a combining of avoidant and anxious features. Avoidant (or anxious avoidant or defended or dismissive) attachment: Refers to an attachment style characterized by maintenance o f a view of the self as independent and competent, and a view of others as unreliable or unsafe sources of needed support.
33 Affect regulation strategies include suppression of attachment related emotions to dissuade the self from seeking support from unrel iable or unsafe others. When encountering bids for support from a distressed other, efforts to emotionally self regulate consume available energies diminishing ability to respond empathically. C ompassionate caring factor : D enotes t he Pity Experience In classification of those emotions, cognitions, and motivations that reflect the active, compassionate, pro social aspects of pity. In this aspect of pity, personal distress does not rise to the level of interfering with the giving of empathy. Dis organized attachment : Refers to an attachment style characterized by a combination of high attachment anxiety and high attachment avoidance, resulting in confusing and disjointed strategies in attachment contexts. This category of attachment is thought t o reflect the experience of childhood care giver trauma. Distress (or personal distress) : R efers to that portion of one s experience when exposed to a distressed other that consumes energy an d focus on the task of emotion self regulation diminishing energ y available to the giving of empathy to the distressed other. Distress related emotions, cognitions, and motivations : R e fers to one of two categories of feelings, thoughts, and wishes one has when exposed to a distressed other. This category of experience s may result in the implementation of distancing strategies to limit personal distress and/or helping strategies motivated in part by efforts to lower one s own distress. See also empathy related emotions, cognitions, and motivations.
34 Ease of escape : R ef ers to the magnitude of any obstacles (whether physical or internally suffering. If escape is easy, supportive responses are assumed to be altruistically motivated. If escape is dif ficult, it is less certain whether supportive responses are motivated by empathy or by the goal of negative state relief. E goistic emotional state : A n experiential condition in which one is hindered from responding supportively to a distressed other b ecause energies are preo ccupied with emotion regulation functions in the face of heightened personal distress. E motion regulation (or affect regulation) strategy : A conscious or unconscious behavioral schematic whose goal is the minimizing of unwanted em otions. Avoidant attachment refers in part to a style of affect regulation that involves suppression of attachment related needs and distancing from t he source of distress. Anxious attachment refers in part to a style of affect regulation that involves e xaggeration of attachment related needs to enhance the likelihood of attaining emotion modulation assistance from others. Empathy : R efers to that aspect o f one s experience when exposed to a distressed other that focuses one s energies on active, compassi onate, and pro social responses Empathy related emotions, cognitions, and motivations : R efer s to one of two categories of feelings, thoughts, and wishes one has when exposed to a distressed other. This categ ory of experiences leads to active, compassionate, and pro social See also distress related emotions, cognitions, and motivations.
35 Episodic (or autobiographical) memory : R efers to a memory system c omprised of information obtained through lived experience. Evidence of whether something that is known is part of episodic memory is the ability to visualize oneself as present during the experiential acquisition of that knowledge. See also semantic memo ry. E xploratory behavioral system : R efers to a complex set of goal directed behaviors designed to develop competencies through interaction with one s environment. The exploratory system interacts with the attachment behavioral system to maintain a s tate of homeostasis between the two. When the availability of comfort and security is ensured, the exploratory system is reactivated as one feels empowered to pursue growth enhancing activities. When one s security feels threatened, the attachment system is reactivated and energies are reoriented toward the attachment figure. See also secure base F alse superiority factor : Denotes t he classification of those emotions, cognitions, and motivations that reflect defensive strate gies for oneself. Index of predominant emotional, cognitive, and motivational response : R efers t o a score assigned to participant responses to the Pity Experience Inve ntories Short Form Revised by subtracting the mean distress item response from the mean empathy item response [shortened to PEI PER ] Index of predominant emotional response : R efers to a score assigned to participant responses to Batson Empathy and Dis tress Indices Adapted Version by
36 subtracting the mean distre ss item response from the mean empathy item response [shortened to BEDI PER ] attachment memory representations) : M ental representations with both cognitive and affective components, likely formed out of generalized event representations of attachment mes biased) interpretations of past attempts to gain or retain proximity to caregivers (and later romantic partners), and the attention and memory permitting or limiting access to certain forms of knowledge about the self, the attachment figure, and the relationship between the two. Negative state relief : R efers to the giving of supportive responses to a distressed other motivated by the need to relieve one s own distres s precipitated by observing the seemingly empathic responses to a distressed other are actually responses motivated by the des ire for negative state relief. Attachment theorists would argue that enhanced accessibility of secure attachment memory representations during responses to a distressed other limits personal distress facilitating (altruistic) empathic responses. P artner in distress episode : A remembered event (or the imagined and visualized recurrence of such an event) where one s spouse encountered distress inducing circumstances and wanted supportive responses from her or his marriage partner. P assive identification factor : D enotes t he Pity Experience Inventori classification of those emotions, cognitions, and motivations that reflect the passive,
37 helpless, painful, and succumbing aspects of pity. In this aspect of pity, identification with the sufferer occurs, but the need for self protection from personal distress interferes with giving of empathy. Proximity maintenance : R efers to one of four defining characteristics of an attachment relationship energies directed toward maintaining closeness to the other. See also safe haven, secure base, and separation anxiety. Safe haven : R efers to one of four defining characteristics of an attachment relationship use of another as a fortress of refuge during times of heightened insecurity or distress. See also proximity maintenance, secure base, and separation anxi ety. Secure attachment : Refers to an attachment style characterized by a generally positive view of both self and others, resulting in direct expressions of attachment needs and reliable responses to attachment needs expressed by others. Secure base : R e fers to one of four defining characteristics of an attachment relationship use of another as a safety net as needed as a means of bolstering confidence for growth enhancing exploratory behaviors. See also proximity maintenance, safe haven, and separation anxiety. Self soothing : R efers to the capacity to give comfort to oneself in the context of distressing circumstances. This capacity is thought to be aided by the accessibility of memories of comfort received from earlier attachment figures during times of distress. Semantic associative memory network : D eno tes an internalized information storage and processing system comprised of one s accumulated base of knowledge, including personally acquired knowledge (obtained through lived experience) and
38 impers o nal knowledge (obtained from external sources). Information may be filed in multiple locations proximal to other information exhibiting cognitive and/or affective similarities. Means of organization may facilitate interpretational biases and other entrop ic feedback into the assessment of present attachment relevant events. Semantic (or procedural) memory : R efers to a memory system comprised of all that we know or think we know. This memory system incorporates information obtained from non experientia l sources as well as information distilled from lived experience. See also semantic associative memory network, semantic memory decision rules, and episodic memory. Semantic memory decision rules : I nternalized guidelines that shape behavior in the co ntext of attachment relevant events. Often operating outside of consciousness, these self commands represent the distillation of past attachment related events, including generalizations regarding the self and others. Separation distress : R efers to one o f four defining characteristics of an attachment relationship a heightened sense of loss or anxiety in the unwanted or prolonged absence of another. See also proximity maintenance, safe haven, and secure base. Organization of the Study This research study is presented in five chapt ers. Chapter 1 provides an introduction and an overview of study related theoretical constructs such as parent to child attachment, romantic attachment, the interplay of memory systems, and the role of emotion regulation strateg ies in shaping internal working mo dels of attachment. Chapter 2 provides a review of the related literature, including outcome studies that serve as the theoretical foundation for t he current study. Chapter 3 presents the
39 methodol ogy for the study. Chap ter 4 presents t he results of the study. And Chapter 5 provides a discussion and summary of the results, implications of the findings, limitation s of the study, and suggestions for future research and practice.
40 CHAPTER 2 LITERATURE REVIEW In their conc eptualization of romantic relationships as attachment relationships, Hazan and Shaver (1987) paved the way for a plethora of studies examining the relevance of attachment relationship lationships. Among the dynamics Bowlby identified as indicative of attachment relationships was the use of another as a secure base of support and as a security restoring safe haven during times of distress (1969/1982; 1973; 1980). Marriage and family therapists seeking to help couples enhance secure base and safe haven communication patterns must contend with resist ant to change attachment states of mind formed in earlier attachment contexts, an d often reinforced in later romantic relationships, including the current relationship (Hesse, 1999) Hope for progress in the face of such impediments argues for accurate understanding of partner intra psychic processes inhibiting chang e (Collins, Guichard, Ford, & Feeney, 2004 ), as well as couple interpersonal dynamics frustrating efforts to sustain n ew communication patterns (Mikulincer & Shaver, 2007). This study builds upon a substantial body of research relevant to these clinical concerns and represents an attempt to extend the findings of several studies built upon some of that sa me foundation (M ikulincer et al., 2001; Bartz & Lydon, 2004; Mikulincer et al., 2005). The review of the li terature serving as a theoretical basis for the current study has been organi zed into six sections: The first s ection contains an examination of research framing ro mantic relationships as attachment relationships. The focus of the second section is attachment style continuity and change. Of particular interest in this section is the investigation of the stability of earlier formed internalized attachment schema acr oss developmental stages and attachment figures. The purpose of the third section is to
41 provide an analysis of research suggesting the role that attachment internal working models play s in attachment style continuity and change. The fourth section includ es an examination of memory research, specifically as it relates to the reciprocal influences of attachment working models and memory systems interplay. Of interest in the fifth section are research find ings suggesting a link between attachment secur ity a nd pro social motivation. Of particular interest are findings suggesting the relationship between felt attachment security and the likelihood of behav ing altruistically toward another who is in distress And the sixth section contains a review of the limi ted body of research investigating secure base and safe haven behavi This chapter then concludes with a summary of the literature review. Romantic Relationships as Attachment In a landmark study exploring romantic relatio nships through the lens of attachment constructs, more than 1200 respondents completed a 95 item survey published in the July 26, 1985 edition of the Rocky Mountain News (Hazan & Shaver, 1987). The bulk of the questionnaire was divided into three parts: T he first part the third part, their relationships with each parent and their paren between the one way child to parent attachment bond and the reciprocal experiences nfirmed, providing strong preliminary evidence that processes at play in romantic attachment are parallel to those heavily investigated over the previous decade relative to the infant parent bond.
42 ing study were asked to self identify with one of three attachment style prototype descriptors (secure, avoidant, or anxious) that best described how they tend to feel in romantic relationships. These descriptors were designed to parallel infant classific ations used in existing attachment studies (Ainsworth et al., 1978). Correspondence between distribution of classifications and distribution of infant attachment type in the cumulative attachment literature was roughly equivalent (confi hypothesis): Secure attachment (56% adults vs. 62% infants); avoidant attachment (25% adults vs. 23% infants); and anxious attachment (19% adults vs. 15% infants). identified romantic attachment types a s the independent variable, the authors found consistent links in the direction hypothesized between parent and parent parent descriptors of their family of origin attachment histories (confirming the hypothesis). A one way ANOVA of these response items with attachment style as the independent variable yielded 51 (of 86) Fs that were significant at p < .05. Of these, 37 were significant at p < .01, and 15 were significant at p < .001. outlooks on romantic relationships (hypothesis 2), their working models of self and others (hypothesis 3), and their vulnerability to loneliness (hypothesis 5). In each area of inquiry, the authors found significant correspondence between child parent attachment category characteristics and cognitive, affective, motivational, and behavioral self reports of study participants. In summary, in addition to noting similarity of a ttachment style distribution between adult romantic attachment and infant parent
43 romantic attachment classifications and earlier identified parenting precursors to infant att achment categories, as well as evidence of cognitive and affective working models replicated in a second study utilizing 108 undergraduate students. In the two d ecades subsequent to the Hazan and Shaver (1987) study, researchers have used the lens of attachme nt style to investigate a broad array of In several studies explori ng mate selection preferences, imagined or p otential partners demonstrating secure attachment styles were pre ferred over those who evidenced attachment insecurity. Such finding s were consistent whether preferences were selected from among attachment prototype descriptions (e.g., Latty Mann & Davis, 1996), from attachment characteristics reflected in a series of questions (e.g., Klohnen & Luo, 20 03); or from secure avoidant or anxious attachment manifestations reflected in behavioral vignettes (e.g., Chappell & Davis, 1998). In observational mea resolution discussions, attachment security was consistently associated with problem solving attributes such as partner support and validation (e.g., Kobak & Hazan, 1991), and respect and negotiation (e.g., Wampler, Shi, Nelson, & Kimball, 2003). Conversely, attachment insecurity was consistently linked w ith greater defensiveness (e.g., Babcock, Jacobson, Gottman, & Yerington, 2000), and more confl ict escalation behaviors (e.g., Alexandrov, Cowan, & Cowan, 2005; Campbell, Simpson Boldry, & Kashy, 2005). In their major update of adult attachment research, Mikulincer and Shaver (2007) summarized the findings of 23 studies examining the link between attachment
44 orientation and intimacy. There were seven different attachment m easures used among the studies, with three of the studies utilizing more than one measure. Ten different measures of intimacy were utilized, and five of the studies assessed attachment dimensions (avoidance or anxiety) rather than categorical attachment s tyles or types (secure, avoidant, anxious). Studies varied in design, with five examining only a security vs. insecurity dichotomy, and the remaining eighteen comparing security, avoidance, and anxiety as correlates of intimacy. In all but six of the stu dies, intimacy was positively linked to attachment security and negatively linked to attachment avoidance and attachment anxiety (or to attachment insecurity in general, depending on study design). In five of the remaining six studies, results were mixed: (a) p < .05 for security and avoidance, but not for anxiety (three studies); (b) p < .05 for security and anxiety, but not for avoidance (one study); and (c) p < .05 for avoidance, but not for security or anxiety (one study). In only one of the studies w as no significant relationship found between intimacy and attachment style. These findings provide evidence of a significant link between attachment security and the experience of intimacy in couple relationships. Mikulincer and Shaver (2007) foun d similar support for the connection between attachment security and relationship satisfaction in their review of 97 studies examining this link among romantic couples. In 28 of the 41 studies using a dating couple sample, securely attached participants r eported significantly greater relationship satisfaction than their avoidantly attached or anxiously attached cohorts. In the remaining 13 studies, findings were mixed, with the hypothesized security/satisfaction link affirmed in all but two s tudies and ei ther the avoidance/ or anxiety/satisfaction links (but not both)
45 aff irmed in the others. For the 57 studies utilizing a married couple sample, findings were less definitive for the studies that utilized categorical measures of attachment style (33 studies ) than for the studies that used attachment dimension measures (24 studies). Among studies utilizing a categorical measure, the full range of attachment style/marital satisfaction links were confirmed a little less than half the time (15 studies). No sig nificant differences were found for one gender or one or more attachment categories in eleven of the studies, or for any category of attachment style in the remaining seven studies. In contrast, among the 24 studies using the more precise attachment dimen sion measures (such as those adapted for use in the current study), three fourths revealed significant negative relationships between marital satisfaction and both insecure attachment dimensions (avoidance and anxiety). In the remaining six studies, the n egative anxiety/satisfaction link was confirmed, but the inverse relationship between attachment avoidance and marital satisfaction was not significant at the p < .05 level, suggesting that attachment avoidance may serve more effectively than does attachme nt anxiety to inoculate oneself from feelings of marital dissatisfaction. Taken as a whole, and with particular emphasis on marital satisfaction studies using more precise attachment dimension measures, the link between secure attachment orientation and m arital satisfaction appears rather well established. In summary, cumulative evidence strongly suggests that romantic relationships may be rightly viewed as attachment relationships, in which partners reciprocally use one another as a source of emotion the parent to self regulate during times of heightened distress. Some of these evidences are more direct such as correspondence between earlier parent to child
46 relationship dynamics and later romanti c attachment states of mind Others must be extrapolated from findings linking attachment style with critical components of romantic partner function, such as confli ct management and intimacy. Still other evidences such as the role of attachment style a s a mediator of relationship satisfaction and marital longevity (e.g., Mikulincer & Shaver, 2007) serve to punctuate the argument that attachment processes are a pivotal component of romantic love. Attachment Style Continuity and Change Social scientist s who investigate the influence of early attachment patterns in subsequent developmental stages report a significant level of attachment style stability over time ( e.g., Thompson, 1999; Waters, Merrick, Treboux, Crowell, & Albersheim, 2 000 ; Main, Hesse, & Kaplan, 2005 ). Negative life events may become global vehicles of discontinuity, moving securely attached children to more insecure attachment orientations. Such life events may include parental death, divorce, or prolonged physical or mental illness. T hey may also include life threatening illness of the child or physical or sexual abuse by a parent (Waters et al., 2000). Less extreme changes in the family system impacting parental availability and/or sensitivity may incrementally shift the developing c temperament (Vaughn & Bost, 1999 ; Stevenson Hinde, 2005 navigation of subsequent developmental stages (Fonagy, 1999), may influence earlier formed attachment patterns toward or away from attachment security. Given the host of environmental factors favoring discontinuity, attachment researchers have nevertheless consistently found a significant lin k between narrative recollections of secure child parent experiences and securely attached romantic relationships in later life (Hazan &
47 Shaver, 1987; Feeney & Noller, 1990; Fraley & Shaver, 1999; Shaver, Belsky, & Brennan, 2000). Attachment researchers differ in how they conceptualize attachment style continuity and change processes. Those who adhe re to the revisionist view (e.g., Kagan, 1996 ) argue for greater malleability of attachment working models. In thi s view, mental models formulated i n earlier attachment experiences are revised as s ubsequent attachment relevant events disconfirm their adaptability. Those who espouse the prototype view of attachment continuity and ch ange (e.g., van IJzendoorn, 1995 ) contend that early attachment schema are less malleable and serve as a filter for interpreting subsequent attachment relevant interactions, consequently influencing those interactions in ways that tend to reinforce those mental models Both the revisionist and the prototype perspectives suggest that attachment working models continually update as a function of both time and (multiple and differing) attachment relationships. Where they differ, however, is that revisionist proponents argue for relative parity among earlier and subsequent attachment ex periences, while adherents of the prototype perspective allege the primacy of early attachm ent experiences (Mikulincer & Shaver 2007 ). Fraley (2002), one of the developers of the Experiences in Close Relationships Inventory adapted for use in this study (Fraley, Waller, & Brennan, 2000), devised a mathematical model fo r testing these two competing hypotheses proposed to account for attachment continuity and change. Fraley began by performing a meta analysis of the 27 childhood attachment longitudinal st ud ies appearing in the literature through 1999 (total N = 1,415). Each of these studies represented Strange Situation
48 assessment s of infants at one year of age, correlated with subsequent attachment category or dimension asse ssment s at a later point in li fe Intervals of assessment ranged from one month to eighteen years ; mean test retest correlation was .37 Fraley divided the se stud ies into five groupings (ages 13 months ; and 2 4 6 and 19 years), calculating for each age milestone the average ret est correspondence with the original Strange Situation attachment assessment. Subsequent to these analyses, Fraley (2002) utilized software assisted mathematical modeling to construct formulas through which he could account for the working mod el plastici ty variations suggested by the respective revisionist and prototype informed constructs of attachment stability and change. For the formula representing the revisionist position, Fraley created a first order autoregressive equation using security as the independent variable, time (gap between initial and later assessments of attachment style) as a weighted function of security, and level of working model plasticity (quantified using theory tenets) as an additional source of variance. The formula represen ting position differed only in that it was not autoregressive reflecting attribution of lesser plasticity to attachment working model s, arising from its position that level of attachment security recursively contribute s to subsequent attachment interactions. As with the formula representing the revisionist position, plasticity estimates in the prototype position formula were quantified using theoretical tenets of th e prototype view. In formulas r eflected th e respective views that initial attachment security is differentially distributed among individuals and that working models of attachment are revised and updated in subsequent
49 attachment relevan t experiences. B ut they also reflected the divergent presupposition s of the revisionist and prototype positions about the weight given to early attachment experiences in resisting subsequent working model revisions. Using these two formulas, Fraley (2002) mapped separate regression curves modeling anticipated patterns of change in attachment orientation between age s one and nineteen. Curves generated under the revi sionist model regressed to levels approaching zero ove r the various time gaps of test retest modeling. By c ontrast, curves generated unde r the prototype model roughly approximated the data of t he longitudinal studies. Fraley and Brumbaugh (2004) later extended this examination of the prototype model in a meta analysis of 34 adult attachment longitudinal studie s. Test retest correlations in the adult studies av eraged .54. An overview of longitudinal studies published subsequent to th is meta analysis (Mikulincer & Shaver, 2007) revealed similar test retest correspondence (r = .56 average for 13 studies). As Fr aley had done with the earlier meta analyzed chi ldhood longitudinal studies, Fraley and Brumbaugh (2004) used th e adult study data to graph attachment stability coefficients at various adult age mileposts (ranging from a ges 18 to 52 years). The y then crea ted a scatter plot of the ( grouped ) data on the stability curves predicted by the mathematical model representing the prototype position As with the childhood longitudinal st udies, these stability curves provided a reasonably go od fit with the d ata. Among the implications suggested by Fraley and Brumbaugh (2004) when comparin g the adult study findings with the earlier childhood study findings were these: (a) Strange Situation attachment style assessment at one year of age is predictive at a level of r .39 of attachment style assessments at any subsequent point in time
50 throughout the life span; (b) Attachment style assessments by age 30 are predictive at a level of r .50 of subsequent assessments of attachment ; and (c) That this greater resistance to change over time is in harmon y with the prototype hypothesis, which holds that earlier attachment experiences shape later experiences in ways that tend to reinforce increasingly entrenched mental models of attachment. In their major update of the adult a ttachment literature, Mikulincer and Shaver (2007) shared their leanings toward the prototype position, suggesting that accommodation of later attachment experiences to earlier mental models is analogous to learning a fore ign language or taking up golf aft er years of playing baseball. In the former analogy, the l earner is likely to retain the accent of the native language, and in the latter, vigilance is required to avoid unconscio us regression to longstanding swing patterns. The cumulative evidence indee d seems to suggest that mental models of attachment formed fairly early in adulthood are not beyond influence, yet show a marked tendency to resist modification and/or a tendency to default toward earlier internalized patterns. Internal Working Models of Attachment Marriage and family therapists encountering couples with maladaptive attachme nt relevant interactional problems are more likely to fashion effective interventions if they understand the processes underlying attachment style stability and cha nge. John Bowlby (1969/1982 ) propose d that attachment style continuity was sustained by mental memory representations of previous attempts to obtain security from a stronger other, and the outcomes of those events. Ma r y A insworth devised a means of opera tionalizing of internal working models in her correlation of first year infant/parent interactional data with one year old infant behaviors in the Strange Situation experimental vignette (Ainsworth, Blehar, Waters, & Wall, 1 978). Mary Main
51 ) infant studies to articulate a set of twelve propositions defining the formation, modification, and transmission of internal working models of attachment as vehic les of attachment state of mind stability and change. The following is a summary of those propositions: Internal working models of attachment are mental representations that include both cognitiv e and affective components, and are integral component s of b ehavioral systems Internal working models are most likely formed from generalized representations of attachment relevant events Once formed, internal working mode ls of attachment have an existence outside of consciousness and a propensity for stability. Internal working models of self and other are formed out of interpreted outcomes of attempts to seek or maintain proximity to caregivers. Infants whose attempts to gain proximity to caregivers are consistently acc epted will develop different internal work ing models of attachment relationships than will those whose attempts are consistently blocked or unpredictably accepted. Where the l atter outcomes predominate, infant patterns of attention, behavior, and emotional expression in attachment relevant contex ts will be reorganized, restricted, and redirected. Individual differences in infant Strange Situation behaviors may be interpreted as reflecting ind ividual differences in internal working model s of attachment relative to the parent Changes in a nternal working model of attachment do not require the to the unavailable caregiver. Internal w orking models provide rule system s that serve to direct behavior in attachment relevant experiences and to influence felt appraisal of those experiences. Internal working models of attachment relationships also provide rules for the direction of atten tion and the organization of memory. These rules many operating outside of consciousness, serve to permit or to limit access to certain forms of knowledge about the self, the attachment figure, and the relationship between th e two, and will be reflected in the organization of thought and language associated with attachment relevant experiences and themes.
52 In childhood, it is possible that internal working models can be altered only in response to changes in concrete attachment experiences. After the onset of the developmental stage of formal operations (abo ut age 12), alteration of earlier formed internal working models of attachment may not be dependent on changes in concrete experiences. This potential relates to an individual emerging capacity to ste p outside a relationship system and to reflect about its rules and function. Although showing a strong propensity for stability, i nternal working models of attachment are best conceived not as templates, but as structured processes that serve to permit or limit access to i nformation. [Main, Kaplan, and Cassidy, 1985] In the quarter century since the presentatio n of these twelve propositions, attachmen t researchers have examined in considerable detail each of the tentative conclusions proposed by Main and h er colleagues. Of particular interest in the current study are th e propositions that hint at enlightening a clinical conundrum frequently How to uncover, bring to client consciousness, and find therapeutic leverage to i nfluence, embedded attachment experience relevant during times of distress. The balance of this section contains a brief overview of a series of studies whose findings sugg est how attachment relevant internal working model s permit or limit access to certain forms of knowledge about the self, the attachment figure, and the relationship between the two, contributing to the stability of those m odels. The subsequent section wil l then be devoted to discussion of memory research suggesting how mental representations of attachment relevant experiences are organized in memory, how felt appraisals of those experiences influence behavioral processes in current relationship experiences and whether and how these processes may be accessible to clinical influence.
53 A crucial body of evidence for erning the vehicles of inter an d intra generational transmi ssion of attachment state of mind emerged with the development of the Adult Attachment Interview (AAI). The AAI is a semi structured interview composed of 20 main questions and follow up prompts designed to assess individuals relevant experiences with the ir families of origin. AAI items include inquiries into the following areas of experience and perception: (a) degree of closeness with e ach parent or parent substitute; (b) how episodes of upset were handled; (c) feeling s of rejection or threat that may have been encountered; (d) how one responded to time s of parental illness or separation; (e) any experience s of loss of a family member or other traumatic attachment relevant events ; (f) perceptions of why parents b ehaved toward them as they did; and (g) the perceived effects of these experiences developmental trajectories and adult personalities (Main, 2010). Developed by George, Kaplan, and Main in 1984, with an accompanying system for scoring and classification developed by Main and Goldwyn the AAI has been refined multiple times over the ensuing years yet has never been formally pub lished for general use The time required to score the AAI makes it impractical for most clinical applications: T ime to create a verbatim transcript of the recorded interview and code and score the transcript is estimated at 14 hours per administration Certification to administer the AAI requires extensive formal training, followed by validation of scoring proficie ncy on 30 subsequent administrations of the AAI protocol (Hesse, 1999; Mikulincer & Shaver, 2007).
54 De spite the impracticality of administering the AAI in normal c ouples counseling s of the AA I can inform clinicians efforts to help couples step outside their own attachment contexts to gain insight prerequisite to changing maladaptive communication patterns. Recently, Bakermans Kranenburg and van IJzendoorn (2009) summarized the findings of 206 AAI studies, totaling more than 10,500 participants and representing both genders, various ages, a diversity of socio economic status es e thnicities, and countries of origin, and both general and clinical populations. Distribution of secure, dismissing ( avoidant), and preoccupied (anxious) attachment categories were similar to those found in infant Strange Situation categorizations, across gender, age, country, and culture, with the exception of moderately higher dismissive (avoidant) classifications amon g European interviewees. As expected, clinical populations repr esenting externalizing disorders (such as borderline p ersona lity d isorder) reflected an over representation of preoccupied (anxious) classifications, while those repr esenting internalizing dis orders (such as antisocial personality disorder) reflected a pronounced incidence of dismissing (avoidant) classifications. The AAI findings among those diagnosed with internalizing and externalizing disorders serve to illustrate (albeit in a somewhat exa ggerated form) the attachment internal working model cognitive and affective processes that the Adult Attachment Interview wa s designed to assess. The rationale supporting such a statement relates to the high levels of concordance consistently found betwe en infant 1995; Mikuli ncer & Shaver, 2007), and requires an u attachmen t state of mind likely influences infant attachment
55 Coding of AAI t ra capacity to provide a coherent narrative of childhood experiences in responding to the interviewer s queries. The AAI transcripts of individuals rated securely attached generally reflect the four maxims of cooperati ve, rational discourse proposed by the linguistic philosopher, Grice (1975, 1989; as cited by Hesse, 1999): (a) Quality communication that is truthful and supported by evidence; (b) Quantity communication that is succinct, yet complete; (c) Relation communication that is relevant to the topic at hand; and (d) Manner communication that is clear and orderly. Because the AAI is measuring of mind in regard to attachment, and not articulate ness or veracity per se, the concept of coherence re lates to the capacity to come and go freely from attachment related material in memory while simultaneously maintaining here and now presence with the interviewer and the task at hand. T he three AAI classifications are described below : Secure/Autonomous Transcripts suggest attachment relationship s are both valued and regarded as influential; able to maintain sufficient objectivity to explore thoughts and feelings about attachment memories during the c ourse of the interview; provide supportive evidence fo r positive descriptors of p arents; a sense of balance, allowances for context, perhaps humor, and implicit forgiveness in negative portrayals of parents; evidence of the capacity for meta cognitive monitoring of attachment memories and the language used to describe them. Dismissing (Avoidant) Transcripts suggest attempts to limit the influence of attachment relationships in cognitions, emotions, and behaviors; implicit claims to strength, normality, and/or independence; tend to describe parents in positiv e to highly positive terms yet when prompted to support such descriptions, are unable to provide concrete examples or provide contradictory examples; deny or minimize importance of attachment related phenomena, perhaps to the de gree of showing contempt fo r them. Preoccupied (Anxious) Transcripts suggest an excessive, confused, and non objective preoccupation with specific attachment relationships or attachment relevant experiences in memory; discussions of these experiences evidence neither objectivity n or insight, seeming rather to intensify distress; descriptors of early relationships may alternate between vagueness and angry and conflicted enmeshment in the memories called upon to retrieve. [Hesse, 1999; citing Main & Goldwyn, 1998]
56 In essence th en it might be said that the AAI assesses attachment internal working models by relationship between the two ru les that doubtless serve the goal of emotion regulation relative to attachment relevant experiences and memories. Secure/autonomous individuals are able to maintain AAI interview narrative coherence because cumulative attachment relevant experiences were generally positive and/or sufficiently processed and positively integrated into the life narrative that the interviewee can freely access memories of them while remaining fully present with the interviewer. Dism issive individuals exhibit rules limiting ac cess to attachment relevant content in memory, and thus are unable to weave a coherent interview narrative supportive of generalized descriptions of childhood. Preoccupied interviewees, by contrast, are unable to maintain narrative coherence because enmes hment in unresolved attachment memories they are prompted to re visit interferes with their ability to remain on task ( Mikulincer & Shaver, 2007). In conclusion, it should be stated emphatically that although parent state of mind in regard to attachment is rather highly predictive of child attachment orientation as much as 20 years later (Hesse, 1999), it is far from determinative. Intervening factors including life events and the contribution s of subsequent attachment figures can serve to shift an attachme nt trajectory from security toward insecurity, and vice versa. Attachment researchers (e.g., van IJzendoorn, 1995) refer to remembered family of origin dyna mics and subsequent assessments of attachment.
57 y ). Individuals who have found a pathway to attachment security despite life experiences that seem unlikely to have support ed such an outcome Genetic differences in meta cognitive skills may play a significant role for some in regard to the phenomenon known a 1991; 1996; cited by Hesse, 1999). Although couples counselors obviously cannot influence heritable traits its seems arguable in light o f foregoing discussions that teaching couples with maladaptive attachment patterns to enlist existing meta cognitiv e capacities in the task of reflection about their own attachment dynamics Toward further informing such counseling strategies the next section pertains to the role that memory systems play in su staining maladaptive attachment patterns. Memory Systems Interplay in Attachment Examination of r esearch relating to memory storing, coding, and retrieval functions, brings an illuminating lens to the understanding of processes governing atta chment internal working model stability and change. Although the contribution of memory system s interplay is implicit in any detailed analysis of attachment worki ng models, many such discussions address these dynamics only obliquely, if at all. In my own clinical practice with couples I have found that efforts to make these often unconscious processes explicit can provide an indispe meta cognitions prerequisite to altering their own attachment patterns. It is beyond the scope of this study to provide a comprehensive review of memory res earch bearing upon attachment dynamics Nevertheless, this section contains an ove rview of
58 Memory researchers a re generally in agreement that the brain provides two distin ct yet interacting domains of memory Episodic memory, also known as autobiographical m emory, is regarded as that repository of memories that were generated in a specific learning context, while semantic or procedural memory is that memory store encompassing all that one knows from any source, including interpretive analyses of lived events encoded i n episodic memory (Tulving, 2001 ). To illustrate the distinction between the two memory domains w hen I say that I know that Santiago is the capital of Chile, my source of knowing was a grammar school geography lesson, backed up by a quick internet search to ensure that my memory had not faded and that intervening geopolitical factors had not nullified what I had earlier memorized as a fact. Therefore, this bit of knowledge is part of my semantic memory store, my general repository of knowledge obtained from any source. When I say, on the other hand, that I know that electricit y is nothing to be trifl ed with, my knowing is attributable in part to the outcome of a childhood hide the object game with my siblings, where I unscrewed a light bulb from a lamp and inserted my fingers into the socket, believing that I had discovered the location of the sought after treasure. Because this latter kind of knowing resulted from a lived experience, it is part of my episod ic or autobiographical memory. Further because this particular experience was not wasted on me, analyses of potential outcomes of naive encounte rs with elect ricity were consciously and unconsciously calculated, and decision rules about future behaviors relative to electrical current were filed in my semantic memory punctuated by sensory and affective data associated with the physical pain and subs equent feelings of foolishness that accompanied the remembered episode.
59 Cognitive psychologists offer case studies of brain damaged individuals and neuroimaging data as converging evidence for the episodic/semantic memory distinction (Moulin & Chauvel, 2010). Tulving cited the case study of a man of normal intelligence and functional skill level whose specific area of brain injury left him without the ability to re member any event in which he had ever participated no matter how recent. Yet he functione d as one who had benefited from the lear ning of those experiences (2001 ). Examining such a phenomenon through the lens of neurological and internal working model function, such a person would be thought to have intact those brain structures necessary for the momentary consolidation of an experience in semantic memory, but not those regions of the brain essentia l to retain recollection of experience s previously consolidated into semantic memory decision rules (Moscovitch, Winocur, Ryan, & Nade l, 2008). Neu roimaging studies reflecting brain activity associated with semantic and episodic memory retrieval functions for both norma l and brain injured subjects have both corroborated and helped refine theoretical formulations about the independent and interdepende nt functioning of these two memory stores (e.g., Shrager & Squire, 2008; Brand & Markowitsch, 2008). In contrast to semantic memory stores that seem to be fairly regionalized and fully faceted episodic mem ory function integrates a multiplicity of brain structures spanning both hemispheres (e.g., Shrager & Squire, 2008; Brand & Markowitsch, 2008; Fujii, 2008; Trimble & Cavanna, 2008; Smith, 2008). Epis odic memories are conceived as summary recordings of eve nts that include sen sory, perceptual, conceptual, and affecti ve data processed from working memory. Visual images tend to predominate, but auditory, olfactory, gustatory, and tactile images
60 are stored in episodic memory as w ell. Episodic memories are ret ained in a durable form only if they are linked to au tobiographical knowledge; they are characterized by a perspective, either field or ground; and they are regarded as the mental representations from which concepts are formed (Conway, 2008). T he details of episodic memories may shift over time or be consolidated with other memories (Nadel, Hupbach, Hardt, & Gomez, 2008). Looking more closely at my childhood encounter with electricity, the record of this event in my epis odic memory i s rich in detail: The room where it occurred; the lamp and the end table upon which it rested giving way to looks of concern; the initial startle response of my siblings giving way to expressions of amusement; the cognitive dissonance that I felt when I was abruptly disabused of the notion that I had cleverly divined the location of the hidden object; and s omatic memories of the sensation of serving as ground (in the non gestalt sense of the word) for an electrical field. Yet when I mentally travel back in time to that event, I am uncertain whether the picture of a rustic farm I see hanging over the couch where my parents were sitting is a detail actually encoded from that time, or whether I have c onsolidated that aspect of the memory with others of having seen that picture hang in that location for many years. Similarly, when I relive the event in my mind, I feel certain siblings. There is some likelihood that this aspect of my memory of the event may be natured) sharing with various audiences about my childhood missteps or events of comeuppan ce whether actual or apocryphal.
61 Memory theorists articulate internal working model function by tracing the episodic and semantic memory processes associated with lived experiences. If a lived event is perceived as essentially irrelevant to conceptual au tobiographical knowledge, it is unlikely to be retained in a durable form If it is deemed relevant, however, a record of it is formed in episodic memory, coded (often unconscious) perceptions of its import survival or being (Conway, 2008). In the latter case, analyses of relevant information is performed (again, often at an unconscious level), and decision rules arising from these analyse s are encoded in semantic memory, for the purpose of inf orming behaviors in future situations deemed analogous to the lived event serving as background for those decision rules (Wheeler, 2000 ; Brown & Craik, 2000 ). Encoding occurs with a level of specificity designed to heighten semantic memory accessibility o f relevant decision rules when future events with corresponding contextual features call for application of the learned in formation as a means of guiding adaptive responses (Anderson & Schooler, 2000; Rya n, Hoscheidt, & Nadel, 2008) O ver time, mental re presentations sharing co mmon or overlapping themes beco me linked together in episodic memory, and these are interwoven with an ever evolving system of semantic memory decision rules, promoting certain behavioral options above others in approaches or respon ses to new situations deemed analogous to earlier experiences. And behavioral prescriptions (sometimes rising to the level of imperatives) emanating from this system of affect laden decision rules are mediated (at least in part) by the goal of emotion reg ulation (Allen, Kaut, & Lord, 2008). The application of this rudimentary explanation of memory systems interplay to the
62 counselors. T h e processes by which such working m odels are formed and refined through lived experiences are essentially adaptive In somewhat simplified terms, working models reflecting attachment security promote exploration and growth buoyed by an intuitive belief that support can be competently summo ned and will be reliably provided when the need arises. Working models indica tive of attachment avoidance compel suppression of attachment related needs, as a means of emotion regulation in response to an attachment milieu perceived as harsh or interferin g. And working models reflecting attachment anxiety compel an amplification of emotions surrounding bids for comfort and support, in the hopes of enlisting emotion regulation assistance from a reluctant other. Worki ng models are maladapt ive, however, to the degree that they were formulated upon biased interpretation of lived experiences, to the extent that cognitive and affective coding heightens the accessibility of less than optimally applicable models for prescribing behaviors in current attachment rel evant situations, and to the limit that, as a consequence of these factors, they serve to elicit reciprocal behaviors that recursively reinforce themselves. The same processes that benefit more securely attached individuals by promoting stability of attach ment orientation during subsequent relationships may serve to thwart their more insecur ely attached peers who hope to leave behind painful patterns of relating when emerging from families of origin and/or earlier adult relationship s. egularly called upon to influence such maladaptive patterns can attest to the multiplicity of obstacles conspiring against attachment working model updating. To illustrate why this is so, let us return briefly to my childhood introduction to electrical cu rrent. My present comfort level with performing a fair number of household
63 electrical tasks with appropriate caution is attributable in part to my capacity to interact freely with that memory over subsequent years, and thus to both challenge a nd refine ge neralizations deduced about electrical current in the context of that episode on the basis of subsequent learning experiences and competency acquisitions. Prior to such updating initial generalizations that electrical phenomena were to be avoided at all costs were adaptive in that they abetted my opportunity for survival in the short term. Subsequent refinement of event outcome interpretations contributed to survival in the long term, however, by promoting a set of behavioral guidelines that did not make danger aversion and competency acquisition mutually exclusive goals. There are several factors, however, that make attachment internal working model updating less straightforward. The first and perhaps most obvious factor when comparing attachment inter nal working mode l updating to non attachment competency model formation and revision is that during the formative years one does not have the luxury of avoiding attachment relevant events until competencies catch up. The likelihood of having maintained re lative attachment figure and milieu continuity, along with the belated emergence of information and perspective vital to making more adaptive prior attachment experience generalizations about the self and other s tend to conspire against model updating eff orts for those with highly insecure internal working models of attachment (see earlier discussion in this chapter about contin uity and change processes). A second factor relates to issues of access. Episodic memory does not fully emerge until about the f ourth year of life (Atance, 2008) This seemingly limits metacognitions about early formed attachment working models to more removed from experience after the fact
64 extrapolations of the contexts in which they were formed, as opposed to the richer sensory affective laden reprocessing possible when the original experience is accessible in memory (Suddendorf & Corballis 2008 ). Even given the theoretical accessibility of later insecure attachment working model precursor events, individuals seem to differ wid ely in their neurological capacities for the inter hemispheric memory store navigation prerequisite to maximal attachment working model challenge and revision (Christman & Prop per, 2010), and the time intensive nature of counseling approaches so directed m ake s them often prohibitive. Yet a third factor arguing for insecure attachment worki ng model resistance to change relates to the aff accumulated life span repertoire of attachmen t relevant experiences, encoded at each jun c ture of the interwoven episodic and semantic memory store networks (Allen, Kaut, & Lord, 2008) These affective components of the system can blur objective assessments of current attachment contexts (Main et al., 1985) goal of emotion regulation provides an interpretational bias toward more familiar patterns of response, despite the apparent ineffectiveness of such emotion r egulation strategies to observers with different attachment histories. Paradoxically, this latte r factor serving to obstruct attachment internal working model update may suggest a fulcrum for therapeutic leverage in helping couples construct a relationship specific attachment context more secure than what their more global attachment orientations mig ht predict. The potential for such a window of influence builds upon three theoretical premises: (a) that differing internal working models of attachment are constructed for differing attachment figures (e.g., Main et al., 1985); (b) th at the goal of emoti on
65 regulation selectively heightens accessibility in memory of specific affect laden working models to inform current attachment relevant episode behaviors (e.g., Allen, Kaut, & Lord, 2008); and (c) that more recent episodic memories may better lend themse lves to the metacognitive processes likely to promote attachment working model change than might those that are more remote in time (e.g., Conway, 2008). Taken together, these premises seem to suggest the efficacy of clinical approac hes that more or less circumvent global atta chment orientations in favor of helping couples develop and strengthen partner specific attachment working models. Such approaches utilize counseling session reenactments of recent attachment relevant episodes or current enactments o f such contexts, with primary focus on bringing to conscious awareness self a nd partner emotional processes promoting specific when elic iting support Certain appro aches to couples therapy seem to be predicated upon this perspective (e.g., Johnso n, 1996). Using such approaches, more global insecure working models of attachment become less accessible as more familiar and sustainable patterns of emotionality surround ing partner specific communication contexts render them less applicable and therefore less unconsciously accessibl e to guide current behaviors. Global insecure attachment working models are not left untouched in such approaches, however, but rather are su bject to scrutiny when they intrude upon current attachment context functioning As such the ben efit of immediacy serves to heighten the likelihood that reflection upon such working models and their precursor events will further reduce the ir accessibilit y and that more secure attachment representations in memory, including those associated with the emerging partner specific interactional
66 patterns will become more influential. Subsequent se ctions in this literature review are intended to examine the effe cts of attachment security on altruistic motivations a nd behaviors, how contextual enhancement of such security may serve to offset such outcomes may apply to partner comm unication and support dynamics in particular. Attachment Security and Pros ocial Motivation The debate has long r aged over whether there is indeed such a thing as altruism, or whethe r the motivation to help is merely the pursuit of self in terest in one form or another, such as the seeking of external reward, the avoidance of punishment, or the A shift away from more cynical positions has occurred among some soci al scientists over the past two decades as a result of experiment al designs intended to isolate differing helping motivation vari ables. Among the variables mani pulated in such experiments have been those associated with what has been dubbed of e scape factor Specifically the opportunity to readily remove oneself from witnessing s plight or from internal sa nctions for choosing not to respond with assistance, have been isolated as variables for the purpose of assess ing how muc h of helping is motivated by these more egoistic processes. Despite the convergenc e of multiple factors promoting help giving behaviors, support has been found in these more recent studies for regarding altruistic empathy a s an independent motivator in re (Batson, 2010). In the last decade, attachment researchers have begun to investigate the relationship between attachment security and the likelihood of responding altruistically to others in need. Gillath, Shaver, Mikulinc er, Nitzberg, Erez, and van IJzendoom
67 (2005) examined the relationship between volunteerism and attachment security and insecurity, finding that those who were avoidant in their attachment orientation engag ed in fewer volunteer activities and gave less tim e to those activities than did their securely or anxiously attached peers. In a discussion of their own earlier studies examining the link between gratitude and prosocial behavior, Mikulincer and Shaver (2010) concl uded that a willingness to help others is highly correlated with feelings of gratitude for help g iven, but that such feelings are significantly diminished or otherwise sullied by attachment insecurity In her review of studies examining the relationship betwe en empathy and the propensity to he lp others Fehr (2010) noted limited or mixed evidence su pporting a connection between such predisposition s and age, education, income, or race/ethnicity. However there was evidence of a link between empathic concern and spirituality and/or religious invo lvement ( citing Smith, 2009; Graber & Mitch am, 200 9), an d women consistently evide nced greater tendencies to express empathy and provide emotio nal support than did men (citing Taylor, 2002; Sprecher & Fehr, 2005; Fehr & Zimmerman, 2007; Marks and Song, 200 9). Interestingly, Fehr noted that differences relative to gender (citing Fehr, 2008) and spirituality/religio sity (citing Sprecher & Fehr, partner (2010). Central to the d esign of the current study were five studies conducted by Mikulincer, Gillath, Halevy, Avihou, Avidan, and Eskoli (2001), investigating the effects of attachment security priming on responses to stranger distress. Secure attachment memory representations were induced by a variety of methodologies in the several studies, including the use of a narrative describing help seeking and supportive
68 responses of a family member, a photo of comfort giving, and the subliminal insertion of comfort inspiring words on p administration. The target stimuli of stranger distress in the various studies included the the reading of a story of a peer suffering a severe ph ysical disability, and the more participant specific recollection response inventories were alternated in collecting dependent variable responses. Because of the nature of contrast, the design of the latter studies prompted participants to retrieve emotions, cognitions, and moti vations from episodic memory. Secure attachment priming was found to be significantly correlated with increased empathy and decreased distress responses to stranger distress. Aspects of these studies were duplicated and expanded upon by subsequent studie s (e.g., Bartz & Lydon, 2004; Mikulincer et al., 2005 ), and as couple participants, however, along with its investigation of romantic partner distress as nd distress activating stimuli, argue for differing design criteria for assessing the more complex dynamics associated with this area of inquiry. Secure Base and Safe According to Bowlby (1969/1982; 1973; 1980), i nherent in the infant parent attachment bond were four defining criteria: (a) proximity maintenance: the infant seeks to keep the parent within a distance that can be regulated; (b) separation distress: the infant experiences distress when that goal is imp eded; (c) secure base: the infant utilizes the parent as a base of exploration to which she/he can retreat when needing to
69 be replenished or reassured [Note: Bowlby (1988) credits Ainsworth with introducin g the concept of a secure base ]; and (d) safe haven : the infant regards the parent as a reliable source of protection, including restoration of emotional equilibrium during times ove as attachment spawned an examination of the applicabilit y of these four attachment bond defining Of primary interest to the current study are the secure base and safe haven components of the reciprocal bond shared by romantic couples. These two attachment bond ch aracteristics might be viewed as subsets of intimacy (Hazan & Ziefman, 1999), with confidence in partner availability constituting the secure base, and partner effectiveness in providing validatio n, comfort, and support during times of heightened distress constituting the safe haven Safe haven behaviors ed partner may include the following: (a) signaling openness to validating partner concerns; (c) c ability to providing physical closeness and tenderness; (f) expressing availability to provide task assistance and resources; and (g) signaling continued availability (Collins, Ford, Guichard, Kane, & Feeney, 2010). Many studies have examined base and safe haven competencies as implied subsets of attachment security. Yet few studies have investigated the direct contribution of these attachment bond defining criteria to marital intimacy and satisfaction. In their major update of the state of the adult attachment literature, Mikulincer and Shaver (2007) cited five studies (Fraley & Davis, 1997; Trinke &
70 Ba rtholomew, 1997; Feeney & Hohous, 2001; Feeney, 2004; and Mayseless, 2004) supporting a link between attachment style and use of a romantic partner as a secure base and safe haven. Four of these studies (all but Feeney & Hohaus, 2004) investigated the ear ly adulthood transitioning of secure base and safe haven attachment roles from parents to romantic partners and peers. Although lending credence to the proposition that romantic relationships ought to be conceptualized as attachment relationships (Hazan a nd Shaver, 1987), these studies are limited to young adult (and mostly unmarried) populations, and therefore leave gaps in our understanding of secure base and safe haven phenomena in more seasoned adult romantic relationships of partner willingness to serve as caregiver posed a different set of generalizability limitations to informing secure base and safe haven functions in the more egalitarian mutuality of mar ital relationships. Of greater applicability to the current study, Treboux, Crowell, and Waters (2004) found a rather robust link between secure attachment and observed secure base behaviors in a 6 year longitudinal study of 157 couples. Participants (N=3 14) were assessed three months before marriage and at six year follow up (N=215), using the Adult Attachment Interview (AAI) and the Current Relationship Interview (CRI). Secure base behaviors correlated with AAI coherence at r = .48 and with CRI coherenc e at r = .41 (both significant at p < .01) [Note: A detailed discussion of the concept of narrative coherence and its relationship to attachment security is provided in discussion of the AAI earlier in this chapter]. Several studies have examined couples behaviors in commun ication laboratory settings (see Collins et al., 2010 for a brief
71 summary), but the variables of attachment security or insecurity were not central to the examination of support seeking behaviors and partner responsiveness a midst episodes of distress. marriage partn er distress, the target episodes cognitions, an d motivations we re intended to tap both the secure base and the safe haven features of an attachment bond, particularly as they relate to attachment orientation and context. Given the paucity of existing research isolating these specific area s add to the understanding of how secure base and safe haven competencies contribute to romantic attachment more broadly defined. Summary This chapter has i ncluded the examination of six separat e, yet overlapping and often interwoven areas of investigat and informing its design. It may be concluded that the bond that characterizes normal ns as a reciprocal attachment bond and as such, that attachment research serves as a r formed attachment mental representations were found to exert signific ant influence over subsequently formed attachment relationships, including romantic relationships, yet they main tain some level of malleability. Internal working models of attachment are thought to be the mechanisms mediating attachment continuity and change, and semantic memory decision rules formulated in the context of earlier attachment relevant events seem lik ely to influence interpretation of current attachment related experiences. There indeed seems to be a relationship between attachment security and the inclination to beha ve altruistically toward another in
72 distress, and the results of recent research sugg ests that the priming of secure attachment memory representations may be a supplemental factor to more generalized attachment style in enhancing the availability of more altruistic response motivations. Yet such studies have focused on response to the pli ght of a stranger, leaving open questions about how such factors influence the more reciprocal, durable, and history laden empathy eliciting and response giving pat secure base and safe haven behaviors are an integral to relationship function. The current study is de signed to help inform those questions
73 CHAPTER 3 METHODOLOGY In the two romantic relationships as attachment relationships, numerous stud ies have documented the link between partner secure attachment style and marital satisfaction (e.g., Kobak & Hazan, 1991; Davila, Bradbury, & Fincham, 1998; Marchand, 2004; Birnbaum, 2007). More recently, social scientists investigating adult attachment h ave discovered that the priming of secure attachment memory representations helps to increase ones empathy (Mikulincer, Gillath, Halevy, Avihou, Avidan, & Eshkoli, 2001; Bartz & Lydon, 2004; Mikulincer, Shaver, Gillath, & Nitzberg, 2005). These findings argue for development of strategies for helping couples maximize the accessibility of secure attachment memory representations during marital communications. Existing studies exploring causal links between secure attachment memory accessibility and empathy enhancement have utilized college student samples (Mikulincer et al, 2001; Bartz & Lydon, 2004 [Study 2]; Mikulincer et al., 2005) or other young adult samples (Bartz & Lydon 2004 [Study 1]). The purpose of this study has been to explore the generalizability of earlier findings to populations of adult couples in monogamous relationships, both by sampling directly from such populations and by incorporating couple specific des ign criteria. The purpose of this chapter is to provide a description and discussion of the participants, variables, experimental design, research questions and hypotheses, instrumentation, procedures, and data analyses for this study.
74 Participants The pa rticipants for this study were married couples between the ages of 20 and 59 years, recruited from churches of various denominations in and around Gainesville, Florida. The sample included 132 couples (264 individuals) who volunteered to participate in re sponse to printed recruitment invitations in their weekly church bulletins and (where permitted) announcements at their weekly church worship services. Couples were assigned to one of four groups (three experimental conditions and one control condition) u sing random stratified assignment. The goal of sample stratification was to achieve approximate group parity relative to the length of current marriage variable De facto stratification was achieved relative to the variable of gender by assignment of cou ples to groups as intact pairs. Gainesville is located in Alachua County in north central Florida. Home to the University of Florida and Shands Teaching Hospital, Gainesville is a major educational and healthcare hub for north central Florida, and nearl championships in football (2006 and 2008) and basketball (2006 and 2007), Gainesville offers a variety of cultural and family cent ered activities, and enjoys frequent rankings as one of the best cities in the country to live, work, and play (Gainesville Council for Economic Outreach Website, 2010). A free book and the opportunity to attend a free seminar were offered as inducements to participate in this study. At the close of the experimental protocol, each couple was permitted to choose between two books (Clinton & Sibcy, 2006; or Smalley & Paul, 2006) selected lik ely to engage the interest of the population sampled. In addition to the free book,
75 participating couples were o ffered the opportunity for post study attendance at a free one day marital communications seminar presented by the principal investigator. Vari ables The purpose of this section is to describe and discuss the dependent variables, the independent variables, and the intervening variables of interest to this study. Dependent Variables onse to their related emotions, cognitions, and motivations were operationalized and assessed using an adapted version of the six empathy ex (Batson, Batson, Griffit, Barrientos, Brandt, Sprengelmeyer, & Bayly, 1989) and 13 empathy related items from a 30 item shortened version of the Pity Experience Inventor ies (Florian, Mikulincer & Hirschberger, 2000; Mikulincer et al., 2001) developed for the current study. Personal distress related emotions, cognitions, and motivations were operationalized and assessed using an adapted version of the eight distress relat ed 1989) and 17 distress related items from a 30 item shortened version of the Pity Experience Inventories (Florian, Mikulincer, & Hirschberger, 2000; Mikulincer et al., 2001) developed for the current study. Copies of these instruments can be located in appendices M and N. Independent Variables The independent variables of interest to this study were the manipulated experimental and control conditions and the fixed factor of gender. The three
76 experimental conditions were designed to make accessible internal working models of A more detailed description of the three experimental conditions is provided later in this chapter. A more detailed discussion of internal working models of attachment and the semantic associative memory networks of which they are an integral part is pr ovided in Chapters 1 and 2 The control condition was a neutral condition, designed to avoid any overt priming of specific attachment representations. Gender was an independent variable of interest to the current study because of findings suggesting gend er differences in the physiological experience of emotional distress. Specific to a dependent variable measure used in this study, Florian et al. (2000) reported that men scored lower than women on both the compassionate caring component (empathy) and the passive identification component (one of two distress measures) of the Pity Experience Inventories when remembering and reflecting on a past event when they felt pity for the plight of another. This finding suggests the possibility of a gender bias in th e employment of coping strategies to avoid emotional arousal when confronted with a distressed other. Of interest to the current study is whether such gender differences are observed when the distressed other is one s own marriage partner. Intervening V ariables The principle intervening variable of interest to this study was participant the context of attachment relevant events. Although often presented as a dis crete
77 a point on a continuum. Although such a point may fall within the inner and outer limits of a discrete attachment category, grouping participants by such categor ies alone severely restricts the identification of individual differences. Because one s attachment style is shaped by multiple attachment relationships over time, it represents a composite of converging and competing strategies for interacting in close relationships. Consequently, an optimum method for measuring and reporting individual attachment style uses both an avoidance score and an anxiety score (Brennan, Clark, & Shaver, 1998; Fraley, Waller, & Brennan, 2000). These scores can be understood as points plotted on each of two unidirectional continua, with attachment security at the nexus and attachment avoidance or anxiety increasing with distance from the nexus. F igure 3 1 illustrates how attachment avoidance and anxiety scores reflect level of d ivergence from the construct of attachment security. A = attachment security B = attachment avoidance (Av) C = attachment anxiety (Ax) C Av scores (B) and Ax s cores (C) represent points plotted along their respective continua diverging from attachment security (A). Higher Av and Ax scores represent points increasingly further from attachment securi ty (A) along the represented unidirectional ray (B or C) B A Figure 3 1. Relationship between Attachment Security, Avoidance, and Anxiety
78 Despite the above described limitations of identify ing individuals by discrete attachment categories, the following brief descriptions of these categories serve to inform future discussion of the intervening variable measures: Securely attached persons see themselves and others in a generally favorable lig ht. Because they regard themselves as worthy of support and others as generally of sufficient good will to provide support when needed, securely attached persons tend to be direct in their expression of attachment related needs and generally supportive in their responses to attachment needs expressed by others (Bowlby, 1969 /1982 1973, 1980; Ainsworth, Blehar, Waters, & Wall, 1978; Mikulincer & Shaver, 2007). Insecurely attached persons who are avoidant in their att achment style use defensive processes to maintain a positive image of the self but see others as generally unreliable, unsafe, or overly intrusive in their responses to attachment needs As a result of this orientation, avoidantly attached persons tend to suppress distressing emotions rather t han to risk relying on others for support, and are more likely to withdraw from bids for empathy from others in order to regulate their own feelings of /1982 1973, 1980; Ainsworth, et al., 1978; Mikulincer & Shaver, 2007). In contrast, insecurely attached persons who are anxious in their attachment style see others as more able than themselves, yet nevertheless unreliable sources of comfort and support. Lacking in their ability to self soothe, anxiousl y attached persons find themselves in the unenviable position of needing others to help them regulate their emotions, yet simultaneously perceiving others as untrustworthy to give them what they need. Anxiously attached individuals find it difficult to ac hieve separateness from others
79 at a level necessary to optimally impart empathy, because they tend to experience (Bowlby, 1969/1982, 1973, 1980; Ainsworth, et al., 1978; Mikulincer & Shaver, 2007). Beca use one s attachment style reflects both the predominance and the semantic associative memory networks, exploration of the interplay between this intervening variabl this study. Attachment style was operationalized and assessed by administration of the Experiences in Close Relationships Questionnaire Revised (ECR R) (Fraley, Waller, & Brennan, 2000) as adapted for this study. The ECR R uses two separate 18 item scales to assess generalized tendencies toward avoidance and anxiety, providing scores that differentiate incremental differences along the continua of these two dimensions (see appendix E). Other intervening variables emerged or became apparent during the process of study administration and data analyses. Occurrences unique to a specific administration of the experimental protocol were documented by the session administrator, a nd were giv en appropriate weight in the data inclusion decision rules discussed in Chapter 4 Experimental Design Figure 3 2 on the following page design. Data collected prior to group assignment (Da) was used to veri fy participant eligibility and to facilitate stratification of random assignment of couples to the three experimental groups (G1 thru G3) and the control group (G4). Data collected in Task #1 (Db) related to the intervening variable of attachment style. The four group
80 variations of Task #4 represented the differing experimental and control conditions (E1 thru E3 and C4). Experimental tasks common to all four groups preceded and followed the differing Task #4 conditions: partner in distress episode select ion task (Task #3 / Ea); and partner in distress episode reflection and imagined recurrence task (Task #5 / Eb). Data collected in Task #6 (De) related to the dependent variables (empathy and distress responses to imagined recurrence of partner in distres s episode). Participants were asked also to report vividness of Task #4 prototype reflection (Dc), vividness of memory of partner in distress (Dd), severity of partner distress remembered (Dd), and vividness of imagined recurrence of partner distress (Dd and De). Participant responses to the relationship history survey provided the data collected in Task #7 (Df). Task #7 responses were used for secondary analyses of participant dependent variable response data. Group Task #1 Task #3 Task #4 Task #5 T ask #6 Task #7 G1 D b E a E1 / D c E b / D d D e D f D a G2 D b E a E2 / D c E b / D d D e D f G3 D b E a E3 / D c E b / D d D e D f G4 D b E a C4 / D c E b / D d D e D f Note: G1 G3 = Experimental Groups; G4 = Control Group; E a E b = Shared Experimental Tasks; E1 thru E3 a nd C4 = Differing Experimental Tasks; D a D f = Data Gathering Activities; Task #2 served as a distracter task Figure 3 2. Schematic of Experimental Design Research Questions and Hypotheses The following research questions and hypotheses were examined a nd tested in the study: A: Does the priming of attachment security or insecurity differentially affect the experience of empathy and distress in response to one s distressed marriage partner?
81 H1: The independent variables of experimental or control cond ition will have no effect on the dependent variable measures of empathy and distress responses to partner distress. Alternative hypotheses are: a: Attachment security priming condition will result in higher empathy positive scores (empathy score minus di stress score) than will attachment avoidance priming condition, attachment anxiety priming condition, or control condition b: Attachment avoidance priming condition will result in lower empathy scores than will attachment anxiety priming condition c: Att achment anxiety priming condition will result in higher distress scores than will attachment avoidance priming condition B: Does gender differentially affect the experience of empathy and distress in response to one s distressed marriage partner? H2: The independent variable of gender will have no effect on the dependent variable measures of empathy and distress responses to partner distress. Alternative hypotheses are: a: Females will exhibit higher empathy scores than will males in response to marriag e partner distress b: Males will exhibit greater use of distancing strategies than will females in response to marriage partner distress C: How does attachment style interact with contextual attachment security or insecurity priming to affect the experi ence of empathy and personal distress in response to one s distressed marriage partner? H3: The intervening variable of attachment style will have no effect on the dependent variable measures of empathy and distress responses to partner distress. Altern ative hypotheses are: a: Higher attachment avoidance scores will predict lower empathy positive scores (empathy score minus distress score) for all three experimental conditions and control condition in response to marriage partner distress b: Higher att achment anxiety scores will predict lower empathy positive scores (empathy score minus distress score) for all three experimental conditions and control condition in response to marriage partner distress
82 c: Higher attachment avoidance scores will predict g reater use of distancing strategies in response to marriage partner distress when controlling for experimental condition and gender D: Does attachment avoidance promote suppression of partner in distress memory vividness as an emotion regulation coping st rategy? H4: Attachment avoidance will have no effect on the vividness of partner in distress memories. Alternative hypotheses are: a: Higher attachment avoidance scores will predict lower partner in distress memory vividness ratings when controlling for experimental condition and gender b: Attachment avoidance priming condition will result in lower partner in distress memory vividness ratings when controlling for attachment style and gender c: Males will exhibit lower partner in distress memory vividnes s ratings when controlling for experimental condition and attachment styl e Instrumentation The purpose of this section is two fold: (1) To describe and discuss the processes of development and validation for the three measurement instruments modified and/ or Experience Inventories; and the Experiences in Close Relationships Questionnaire. And (2) To describe and discuss the rationale and processes governing modification an d/or adaptation of these instruments for use in the current study. (BEDI) Batson et al., 1989) refer to a list of six empathy related and eight distress related emotion referenced adjectives intended to differentiate between altruistic and egoistic emotional states experienced when exposed to the suffering of another. Batson and his colleagues developed and refined the empathy and distr ess indices through a series of
83 studies intended to further the debate about whether and when helping behavior in Experimenters in these studies used a variety of crea tive designs to expose participants to a distressed other. Affect was assessed by participant self ratings of their emotional states in responses to a series of emotion based adjectives believed to distinguish between empathic concern and personal distres s. Factor analyses of the empathy and distress scales with their individual items confirmed the essentially orthogonal nature of these two indices. Further validation of the empathy/distress distinction was provided by experimental manipulation of partic ipant ease of escape (e.g., Batson Bolen, Cross, & Neuringer Benefiel, 1986) and participant mood (e.g., Batson, et al., 1989), challenging the position of those who suggest that empathy attributed helping behaviors are merely behaviors directed toward th e egoistic goal of negative state relief (e.g., Cialdini, Schaller, Houlihan, Arps, Fultz, & Beaman, 1987) A discussion of how the egoistic versus altruistic helping motivation distinction informs the current study can be found in Chapter 2 In a review of six studies using and validating the emerging empathy and distress indices, Batson, et al. (1987) provided compelling evidence for the inclusion of six empathy related adjectives ( sympathetic, moved, compassionate, tender, warm, and softhearted ) and ei ght distress related adjectives ( alarmed, grieved, upset, worried, disturbed, perturbed, distressed, and troubled ) to Batson et al. (1987) used Varimax rotated principal components fa ctor analyses of the empathy and distress adjectives with their respective indices to validate their emerging empathy/ distress measure. The results of their analyses are displayed in
84 Tables 3 1 and 3 2 at the end of this chapter. Findings revealed that f our of the empathy adjectives loaded higher than .60 in the first two studies ( moved, compassionate, warm, and softhearted ) and all six loaded higher than .60 in the remaining four studies. Similarly, six of the eight distress adjectives loaded higher tha n .60 in each of the first two studies (all but grieved and worried in the first study; all but perturbed and troubled in the second study). In the remaining four studies, all but grieved in the third study (.55) and troubled in the fifth study (.59) refl ected principal factor loadings higher than .60. premise that although there is overlap in the experience of empathy and personal the basic dimensions underlying the empathy and distress indices are largely independent. Consequently, these indices provide a useful tool for assessing egoistically versus altruistically motivated participant responses to the suffering of another. In a set of five studies that the current study is reliability coefficients of .94 (empathy index) and .88 (distress index) for their use of the six empathy related and eight distress related adjectives (Study 1; N=69). Mikulincer and his colleagues also reported roughly congruent effect sizes from their alternate use perience Inventories (Study 2) to assess the effect of attachment In the pilot study experience preparatory to the current study, the six empathy and eight distress adject ives (Batson et al., 1983; Batson et al., 1987; Batson et al., 1989)
85 were presented as single word items. The presentation of these items followed a set of general instructions intended to guide participants in rating the degree to which they experienced each emotion based adjective in the partner in distress episode recalled and reflected upon. While presenting the indices as single word items was apparently adequate for the studies summarized in Table s 3 1 and 3 2 (located at the end of this chapter), p ilot participant feedback highlighted a design distinction of the current study that argued for modifying item presentation. Specifically, in each of the earlier studies, participants were exposed to the present plight of a stranger. By contrast, the cur rent study stimulus was the plight of one s own spouse. Obviously, it would be impractical (if not unethical) to attempt to precipitate within each participant and spouse episodes of distress of a magnitude sufficient to elicit meaningful partner empathy and distress responses. Therefore, the desi gn of the current study represented an attempt to approximate such experience s by induci ng participants to remember a past episode of partner distress and to imagine its recurrence. Remembering a past episode of partner distress and imagining its recurrence poses the problem of intrusion of the memory of past feelings (what was felt when the episode actually occurred) into the memory of presen t feelings (what is being felt as a recurrence of the episode is present ly imagined). This bleeding together of past and present emotional responses is unavoidable, because sufficient engagement in the emotional memo ry of a past episode seems necessary for optimum imagination of its recurrence. Yet to accurately assess the d ifferential effects of present attachment priming variations on resp onses to partner distress, it was focus and reporting of emotions toward those experienced in the imagined present
86 recurrence rather than those felt in the original episode itself. Therefore, the empathy and distress indices (Batson et al., 1983; Batson et al., 1987; Ba tson, et al., 1989) were adapted for the current study by embedding the adjectives in sentences that prompt ed participant focus on the prese nt emotional experience of the imagined episode recurrence. and Distres s Indices Adapted Version (BEDI AV), can be located in appendix M. softhearted myself becoming perturbed substituted for the single word ad jective items of the original indices. Empathy and distress ite ms are intermingled in the BEDI AV, and 7 point Likert response options elicit from respondents the degree to which each emotion is experienced. The inventory and its instructions are readable at a seventh grade level. Scoring was accomplished by summing the Likert response levels (1 7) for items in each index, and then dividing by the number of items (6 empathy items; 8 distress items) to achieve an index average. A single index of predominant emotional response was determined by e from his or her empathy index score (Batson et al., 1983; Batson et al., 1989). Conversion of participant index scores to standard scores with a mean of 50 and a standard deviation of 10 permitte d a merging of participant BEDI AV scores with the indepen dent empathy/distress measures of the Pity Experience Inventories (Florian et al., 2000) (see next subsection). Pity Experience Inventories (PEI) The Pity Experience Inventories (PEI) (Florian, Mikulincer, & Hirschberger, 2000), is a 96 item self report me asure assessing the meaning that the construct of pity holds
87 for respondents (20 items), along with the emotions (28 items), cognitions (15 items), motivations (16 items), and behaviors (17 items) that the experience of pity elicits. The five categories o f items are intermingled within the measure, as are items within each category that are intended to differentiate between the beliefs, emotions, cognitions, motivations, and behaviors that are empathy oriented and those that are distress oriented. Six poin t Likert response options elicit from respondents the degree to which each item descriptor is experienced. Florian et al. (2000) used a phenomenological approach to construction of the Pity Experience Inventories. One hundred and four students (median a ge = 24) from Israeli colleges and universities were asked to freely write about the meaning that the term pity held for them. They were then asked to recall an episode when they experienced the feeling of pity for another, and to describe their emotions, thoughts, wishes, and actual behaviors during that episode. Inter rater agreement was used to extract words or phrases from participant responses that reflected a single unit of content. Units reflecting a similar theme were grouped into labeled categor ies, and those units cited by at least 10 percent of the sample were retained for final coding. The personal meaning of pity task elicited 83 different units, grouped by the authors into 17 categories. The subsequent task responses resulted in the identi fication of 153 differentiated emotions (23 categories), 102 differentiated thoughts (12 categories), 52 differentiated wishes (10 categories), and 86 differentiated behaviors (17 categories). Florian et al. (2000) used the findings of the phenomenological study to create an initial version of the PEI with five separate scales tapping meanings (30 items), emotions (51 items), thoughts (22 items), wishes (28 items), and behaviors (32 items)
88 related to the experience of pity. Construction of these five scale s was performed by selecting the two or three items from each of the above categories that were most frequently cited in the phenomenological study. This initial version of the PEI was then administered to a new pool of 200 Israeli college and university students (median age = 23). Data analyses of participant responses included principal components factor analyses with Varimax rotation for each item with its respective inventory. Items loading lower than .40 with a factor or higher than .40 with more th an one factor were eliminated. These exclusions resulted in a final version of the five PEI inventories containing 20 meanings related items, 28 emotions related items, 15 thoughts related items, 16 wishes related items, and 17 behaviors related items. Fl orian et al. (2000) then administered this version of the PEI to 672 students (median age = 24) from Israeli colleges and universities. Factor analyses with Varimax rotation were performed for each of the items with each of the five scales (personal meani ng of pity, and emotions, thoughts, wishes, and behaviors in t he experience of pity). Analysi s of the pity meaning scale yielded five main factors explaining 55% of the variance. These factors were labeled the cynical superiority position (19%), the bene volent position (16%), the equalitarian legalistic position (8%), the religious position (6%) and the moral commitment position (6%) toward the meaning of pity. Analysis of the PEI emotions inventory yielded five main factors explaining 55% of the varian ce. Sorrow related feelings accounted for almost half of this variance (26%), with the remainder accounted for by feelings of emotional discomfort (12%), vulnerability feelings (7%), sympathy related feelings (5%), and superiority feelings (5%). Three fa ctors accounted for 57% of the PEI thoughts inventory variance, labeled
89 by Florian et al. (2000) as action oriented thoughts (34%), misfortune contemplation (14%) and self focused thoughts (9%). Three factors accounted for 65% of the PEI wishes inventory variance, summarized as care oriented wishes (38%), unrealistic wishes (19%), and distancing wishes (8%). And five factors accounted for 63% of the PEI behaviors inventory variance, identified by the authors as support seeking behaviors (30%), praying (11 %), consoling behaviors (9%), preventive behaviors (7%), and d istancing behaviors (8%) (Note: mismatch between factor sum and total variance is the result of rounding). Looking for a superordinate structure governing the experience of pity, Florian et al. (2000) computed factor scores for each of the PEI scales by averaging items loading above .40 with a given factor, and then performing high order factor analyse s with Varimax rotation on the 16 (above named) factors associated with the experience of pity. These analyses revealed three high order factors accounting for 58% of the total variance. Thirty four percent of the variance was explained by sympathy related feelings, action oriented thoughts, care oriented wishes, support seeking behaviors, consolin g behaviors, and preventive behaviors. This f actor comprised the active, pro social, and compassionate components of pity. Florian and his colleagues labeled these components of the pity experience the compassionate caring factor Seventeen percent of t he variance was comprised of sorrow related emotions, misfortune contemplation, unrealistic wishes, and praying. These components reflect the passive, helpless, painful, and succumbing aspects of the pity experience. These components indicate emotional i dentification with the suffering of others, but they tend to impede active caring behaviors because of the overwhelming nature of the sorrow
90 suffering. This set of constructs was labeled the passive identificat ion experience of pity. A third high order factor labeled false superiority accounted for seven percent of the variance. This factor comprised the emotional discomfort, vulnerability feelings, superiority feelings, self focused thoughts, and distancing w ishes and behaviors of the pity experience. This high order factor reflects the defensive strategies by which people cognitively and behaviorally distance Using Pearson correlati ons Florian and his colleagues discovered no significant association between age and the three high order factors of compassionate caring, passive identification, and false superiority. However, t tests revealed significant differences in the variables o f gender and religiosity. Women scored higher than did men in both the compassionate caring and the passive identification experience of pity, as did those who identified themselves as religious in comparison with those who did not. No significant associ ation was found between gender and religiosity. Florian et al. (2000) examined the construct validity of the PEI scales in three additional studies correlating participant PEI responses with other measures of psychological constructs thought to overla p the three high order PEI factors. Pearson correlation of participant responses to the PEI and the Schwartz Value Survey (Schwartz, 1992) suggest the following: (a) That the compassionate caring factor may involve values of u niversalism and benevolence (b) That the false superiority experience of pity may refle ct values of power, achievement, and enhancing one s self worth. A nd (c) That the passive identification factor may represent a struggle between the self transcending value of identification wit protective
91 needs for security and conservatism. Comparing PEI responses with participant responses to three measures of sense of control (Joe, 1974 [Personal Control scale and Attributio n of Causality scale]; Pearlin & School er, 1978 [Sense of Mastery scale]), the greater the likelihood of experiencing pity in false superiority terms. No significant correlations were found between the s ense of control and the compassionate caring and passive identification factors of the experience of pity. In their third construct validation study, Florian et al. (2000) compared participant PEI responses with their responses to a 15 item attachment styl e measure (Mikulincer, Florian., & Tolmacz, 1990) and a 32 item measure of care giving attitudes and behaviors (Kunce & Shaver, 1994 ). Pearson correlations revealed a significant positive relationship between the compassionate caring pattern of pity and s ecure attachment (r = .30, p < .01), while the false superiority experience of pity was positively associated with insecure attachment (avoidant: r = .36, p < .01; anxious (r = .52, p < .01). T he compassionate caring factor was positively associated wit h giving care as a means of being close with others (r = .31, p < .01), while the passive identification pattern of experiencing pity was associated with compulsive care giving (r = .34, p < .01). This latter finding seems to reinforce earlier referenced interpretations that those experiencing pity in this way may be motivated to help others as a means of relieving their own distress (see Chapter 2 for a further discussion of this theme). Finally, the false superiority pattern of pity was inversely relate d to the sensitivity factor of care giving (r = .30, p < .01).
92 Florian et al. (2000) assessed the internal reliability of the Pity Experience three high order factor s cores and averaging the items corresponding to each factor. Cronbach alpha coefficients for the three factors ranged from .78 to .85 in the first study (N=62), from .76 to .78 in the second study (N=100), and from .70 to .80 in the third study (N=77). Mikulincer et al. (2001) reported Cronbach alpha coefficients of .89 for the empathy related items and .88 for the distress related items in their use of a 24 item shortened version of the PEI to measure the effects of attachment security primi ng on responses to a distressed other (N=60). Mikulincer and his colleagues attributed development of the version used in their study (hereinafter referred to as PEI Short Form or PEI SF) to Hirschberger, a co developer of the original PEI (Florian, Mikul incer, and Hirschberger, 2000), for use in his unpublished doctoral dissertation. SF contained 14 empathy sampling items and 10 distress sampling items (Mikulincer et al., 2001). A reconstruction of the full PEI feelings, thoughts, an d wishes inventories from which Hirschberger drew his 24 PEI SF items can be located in Tables 3 3 thru 3 6 at the end of this chapter. For each inventory item stem sub grouping, a designation of the percentage of inventory variance that is explained by t hat sub grouping is provided. And for each item within each inventory sub loading against its sub grouping construct is noted. Mikulincer et al. (2001) reported that Hirschberger drew his 14 empathy related i caring factor and his 10 distress
93 factor. There is a discrepancy, however, regarding the source of the distress related items, in that both the item categories designate d and the item examples cited superiority factor of the PEI. Compounding the mystery created by this discrepancy is the unavailability of a copy of the PEI SF. Wri tten in Hebrew for studies conducted among Israeli college and university students (Mikulincer et al., 2001), the PEI SF was never been translated into English (personal correspondence with G. Hirschberger, 2007). A new short version of the PEI (her e inaf ter referred to as the PEI Short Form Revised or PEI SF R) was developed for use in the current study. The 30 item PEI SF item PEI SF (Mikulincer, et al., 2001) not only in the number of items, but also in item sele ction criteria and item presentation. Item presentation changes emerged from the same rationale articulated in the earlier example of presentation differences for an it em common to the two instruments is the PEI SF different item selection criteria in develop ing the PEI SF R requires a more extensive discussion. Four considerations shaped the PEI SF R item selection rationale: (a) Only 5 9 of the 96 total PEI items were relevant for use in the development of the PEI SF R. (b ) re ssed other in the current study, contrasted with the
94 certain PEI items as appropriate for use in both instruments (PEI SF and PEI SF R), while revealing certain others items as appropriate for use only in one instrument or the other. (c) Developing a test with good psychometric properties requires attending to such matters as number and type of items selected, strength of relationship between each item and its parent fa ctor, and item face validity to respondents. And (d) The attachment security and insecurity priming design of the current study, contrasted with the attachment security only resulted in certain of the PEI items being of interest to the PEI SF R that were of lesser interest to the PEI SF. A schematic depicting the rationale and process of PEI SF R development can be found in Figure 3 3 at the end of this chapter. The first consideration in understanding the PEI SF R item selection rationale requires a recollection of earlier descriptions of how the original PEI was constructed The 96 item PEI consists of five scales, composed of 20 meanings related items, 28 feelings related items, 15 thoughts related items, 16 wishes related items, and 17 behaviors related items. Because participant beliefs about the meaning of pity are not of interest in the current study, and because the current study design affords no opportunity to observe participant behaviors, o nly the 59 feelings thoughts and wishes related items were relevant for item selection. These 59 items represent the SF was drawn (Mikulincer et al., 2001). The second consideration in understan ding the PEI SF R item selection rationale relates to the use of an intimate other (current study) in contrast to the use of a stranger (studies for which the PEI SF was developed and used) as the person whose plight
95 serves to elicit participant empathy an d distress responses. In the studies using the PEI SF, participants were responding to the plight of a stranger, where issues related to the possibility of ending up has less escapable implications for oneself (such as when the distressed other is one s spouse). Conversely, s marriage partner, but they seem less appropriate if the distressed other is a stranger. SF) fit well in the current current study de sign). A panel of experts was assembled to assist in the determination of which of the PEI feelings thoughts and wishes related items were appropriate for assessing responses to marriage partner distress. The expert panel was composed of three license d mental health professionals experienced in couples counseling. Independent ratings provided by the three members of this panel helped to inform the process by which 45 of the 59 PEI feelings thoughts and wishes related items were determined to refl ect face validity sufficient for their inclusion in the PEI SF R. The final version of the PEI SF R reduced the number of included items to 30, by incorporating additional selection criteria subsequently discussed.
96 The third consideration in understandin g the PEI SF R item selection rationale relates to the need to be attentive to established principles of test construction ( e.g., Sax, 1989). In keeping with these principles, the number of items selected for each factor was proportionate to the percentag e that factor contributed to the combined criterion, items were selected from among those available (13 of the 15 relevant compassionate caring items; 10 of the 16 relevant passive identification items; and 7 of the 14 relevant false superiority items) that exhibited a combination of high face validity and high factor loadings. The fourth consideration in understanding the PEI SF R item selection rationale relates to the need to differentiate participant distress responses reflecting attachment avoidance from responses reflecting attachment anxiety. This distinction was of less importance to Mikulincer et al. (2001), when comparing the effects of attachment security priming to the effects of positive affect priming and control condition t han it was to the current study comparisons of the affects of secure avoidant and anxious attachment priming with control. In the current study, it was desirable to explore which distres s responses signal ed the employment of avoidant attachment strategies for coping with the plight of one s spouse and which ones signal ed the employment of anxious attachment strategies a distinction that could have been made only weakly at best by using th e PEI SF. Therefore, an additional item selection priority of the PEI SF R was the inclusion of multiple items that correlate highly with one of the two insecure attachment constructs (attachment avoidance or attachment anxiety) but do not correlat e highly with the opposing construct. To understand this latter objective of
97 instrument development, it is important first to remember that both attachment avoidance and attachment anxiety are descriptive constructs reflecting an anxious state of mind (see definitions in Chapter 1 ). These states of mind differ, however in the strategies they motivate for ameliorating the distress experienced. Whereas avoidantly attached persons t end to use suppression to regulate their emotions of anxiety, anxiously att ached persons tend to experienc e enmeshment alternated with protective distancing in response to distress (see description of attachment categories earlier in this chapter). This distinction creates a psychometric problem in the development of an instrume nt designed both to differentiate attachment security from attachment insecurity and to differentiate the two manifestations of attachment insecurity (avoidance and anxiety) from one another. This psychometric problem can be illustrated by imagining one individual with an upper quartile attachment avoidance score and a lower quartile attachment anxiety score, compared to a second individual with an upper quartile attachment anxiety score and a lower quartile attachment avoidance score. Because these two individuals can be characterized as having equivalent levels of attachment insecurity, they are likely to rate PEI items similarly that assess attachment insecurity characteristics common to both attachment avoidance and attachment anxiety. However, becau se of these types or styles of attachment insecurity, they tend to implement differing strategies for responding to (self and other ) distress, and thus are likely to respond divergently to PEI items that dist inguish attachment avoidance from attachment anxiety The first individual (high avoidance, low anxiety) would likely designate high er numbered Likert responses to PEI items useful for
98 assessing attachment insecurity in general, as well as to those items intended to distinguish attachment avoidance from attachment anxiety. By contrast, however, such an individual would likely select low er numbered Likert responses to items designed to di fferentiate attachment anxiety from attachmen t avoidance. Consequent ly if this distress scores were calculated by averaging responses to all PEI distress items, the resultant score would be diluted by low ratings designated on anxiety specific items resulting in an under reporting of actual l evels of distress. Similarly, low scores of the second individual (high anxiety, low avoidance) on avoidance specific PEI items woul statistically masking the level of distress actually experienced. T he original PEI authors provided no data suggesting the degree to which distress related items were attachment insecurity homogeneous or attachment insecurity divergent. Item face validity in concert with theoretical distinctions between the two attachmen t insecurity constructs provided some hints. Yet the distinctions are complex because both avoidant and anxious attachment are characterized by heightened attachment anxiety, but they diverge in the strategies they promote to mediate that anxiety. An ex amination of such distinctions, then, would require post usefulness for differentiating attachment avoidance or anxiety from attachment insecurity in general could be calculated in the following manner: (a) Correlate R avoidance and anxiety scores. (b) Calculate the absolute value of the difference between the avoidance and anxiety correlation coefficients for eac h of the 17 PEI items. (c) Rank order the resultant
99 (absolute value difference) correlation coefficients from highest to lowest. And (d) Label each item avoidance differentiating or anxiety differentiating, based on whether that item correlates more strong ly (whether directly or inversely) with avoidance or anxiety. The means to distinguish attachment insecurity homogenous items from attachment insecurity divergent items could provide a useful adjunct for examining any differences found between absence of such group differences, exploring this distinction would still prove beneficial to the informing of future research making use of the PEI SF R to assess attachment related emotions, cognitions, and motivations. To increase the likelihood that attachment insecurity divergent items were available for such exploration it seemed best to sample more generously from the false superiority factor of the distress index than did the PEI SF author. Flor factor sub groupings made this assumption seem plausible. In addition, certain of the items on their face (particularly the multiple distancing items) argued for this course of action. This was bo rn out in the pilot study, where the mean of each distancing item was significantly below the overall mean distress score suggesting that these items may be less attachment insecurity homogenous than other of the distress related items. And finally, the f ith attachment anxiety (r = .52; p < .01) than with attachment avoidance ( r = .36; p < .01) suggested that this factor must differentially assess these two attachment insecurity constructs (Florian, et al., 2000). Following each of the considerations earlier discussed, the 30 items selected for the PEI SF R included 13 empathy related items from the compassionate caring factor,
100 10 distress related items from the passive identification factor, and 7 dis tress related items from the false superiority factor. The discrepant sizes of the three indices represent ed an attempt to strike a balance between high factor loaded items appropriate for measuring responses to one s marital partner in distress, and the selection of more items from sub gr oupings account ing for a greater proportion of an A copy of the PEI SF R can be located in appendix N. The reader may comp are the PEI SF R with Tables 3 3 thru 3 6 at the end of this chapter, to ide parent factor and factor sub grouping, as well as its factor loading coefficient. Following the lead of the PEI SF author ( Mikulincer et al., 2001), item construction included embedding the core concept of the item selected within a phra se appropriate to the treatment context (see discussions earlier in this section). Items s ampling empathy and distress are intermingled in the measure, and Likert response options elicit from participants the degree to which each emotional, cognitive, or motivational descriptor is experienced. A 7 category response option to facilitate merging scores with the earlier described measure of empathy and distress (see previous subsection). The invento ry and its instructions are readable at a seventh grade level. Scoring for each of the indices was accomplished by summing the Likert response levels (1 7) for all items in an index and then dividing by the number of items in that index. As with the BED I AV (see previous subsection), a single index of predominant score from his or her empathy index score. Conversion of participant index scores to
101 standard scores with a me an of 50 and a standard deviation of 10 permitted a merging of participant PEI SF R scores with their BEDI AV scores (see previous subsection). Experiences in Cl ose Relationships Questionnaire Revised (ECR R) The Experiences in Close Relationships Questi onnaire Revised (ECR R) (Fraley, Waller, & Brennan, 2000) is a 36 item self report measure of romantic attachment developed by revising the Experiences in Close Relationships Questionnaire (ECR) (Brennan, Clark, & Shaver, 1998). The ECR R uses two separat e 18 item scales to assess attachment avoidance and anxiety, and 7 point Likert response options to elicit from respondents the degree to which each of the attachment facet sampling items describes them. The inventory and its instructions are readable at a seventh grade level. Scoring is accomplished by summing the Likert response levels (1 7) for the 22 direct keyed items and (7 1) for the 14 reverse keyed items, and then averaging the scores for each scale. In revising the ECR R, Fraley et al. (2000) us ed the principles of item response theory (IRT) to analyze the pool of attachment facet assessment items from which the original ECR items were drawn. Their goal was to ensure that questionnaire items they selected optimally discriminated among individual s with contiguous trait levels across full trait range continua. Based on their IRT analyses of the original item pool, Fraley and his colleagues replaced 44% (16 of 36) of the original ECR items. Because development of the ECR R (Fraley et al., 2000) was built upon the research foundation of the original ECR (Brennan et al., 1998), we will begin with an examination of the development of the original ECR. In developing the original Experiences in Close Relationships Questionnaire (ECR), Brennan et al. (199 8) conducted a thorough search of the attachment literature,
102 including examination of the 14 self report inventories of attachment available at the time of their study. From that review, Brennan and her colleagues created a pool of 482 items designed to a ssess 60 identified attachment related constructs. Elimination of redundancy reduced the number of items to 323, distributed among the 60 attachment construct subscales. Some item wording was revised to emphasize romantic relationships over attachment r elationships in general. Brennan, et al. (1998) administered five attachment self classification measures (Hazan & Shaver, 1987, 1990; Bartholomew & Horowitz, 1991; Sperling Berman, & Fagen, 1992; Latty Mann & Davis, 1996) to 1,086 undergraduates (median age = 18) at the University of Texas at Austin. These measures were designed to elicit a composite attachment style dimensional score, falling within the various attachment categories of secure, avoidant, and anxious [with a fourth category, fearful, added by Bartholomew & Horowitz, 1991]). The purpose of this step of the ECR development was to establish a participant attachment style baseline to which later attachment facet sampling item responses could be compared. After individual comparison baselines w ere established, participants responded to the 323 attachment facet items by rating the degree to which each statement described them, using a 7 point Likert scale. Factor analyses of the 60 attachment construct subscales produced two largely indepen dent higher order factors corresponding to the avoidance and anxiety dimensions central to all attachment research. Those measures that include an additional classification of attachment insecurity (often labeled disorganized or fearful ) are describing an attachment style that combines both avoidant and anxious features. Correlations among the 60 subscales (on a 60 x 60 matrix) were highly patterned: 62%
103 were either >.50 or <.20, suggesting multiple independent factors with significant overlap. The corre lation between the two higher order factors (avoidance and anxiety) was very low (r = .12), suggesting that the dimensions underlying avoidance and anxiety are essentially orthogonal. Together, the two higher order factors accounted for 62.8% of the varia nce in the 60 subscales. Brennan, et al. (1998) determined factor loadings of the 60 attachment construct subscales with the two higher order factors. Subscales correlating most highly with the first factor (avoidance) were avoidance of intimacy (r = .91) discomfort with closeness (r = .90), and self reliance (r = .88). Subscales correlating most highly with the second factor (anxiety) were preoccupation (r = .86), jealousy/fear of abandonment (r = .85), and fear of rejection (r = .83). Brennan and her colleagues then constructed two 18 item scales (from the complete pool of 323 items), using the 18 items with the highest absolute value correlations with each of the two higher order factors of avoidance and anxiety. Like their parent factors, the two s cales were virtually uncorrelated (r = .11), and each was highly correlated with its parent factor (r = .95 in both cases). in Close Relationships Questionnaire (ECR), along wit h three of the 14 attachment measures used by Brennan and her colleagues in the development of the ECR (Collins & Read, 1990 [Adult Attachment Scale]; Griffin & Bartholomew, 1994 [Relationship Styles Questionnaire]; and Simpson, 1990 [unnamed attachment qu estionnaire]). Fraley and his colleagues evaluated these four adult attachment measures through the lens of item response theory.
104 Item response theory (IRT) refers to a diverse family of models designed to reflect s item response and the latent trait that the item response options are intended to measure. An important objective of IRT is the creation of models that depict the relationship between the underlying trait an item is designed to measure and the probabili ty of item endorsement. Two important parameters guide the creation of IRT informed models for the measurement of psychological constructs: One incrementally different levels of the trait being measured. And the other relates to an of the trait continuum (Fraley et al., 2000). Using these two IRT based criteria, Fraley and his coll eagues evaluated the four selected attachment measures. Finding the ECR (Brennan et al., 1998) to be clearly superior to the other three, they concentrated their efforts on improving this instrument. Fraley et al. (2000) used a clustering algorithm to regroup the 323 original ECR were computed by averaging item loadings. The authors established discriminant validity of their new item groupings through Varimax rota ted principal component factor analyses. The 18 avoidance scale items and the 18 anxiety scale items selected for the new ECR R were those items loading most highly on the one factor while loading below .25 on the opposing factor. This resulted in a repl acement of 11 of the original ECR avoidance scale items and five of the original ECR anxiety scale items. The ECR R improves on the psychometric properties of the original ECR, while enhancing the e of wording that places a
105 greater emphasis on romantic relationships. Internal consistency of the measure is scale (e.g., Sibley & Liu, 2004). Table 3 7 located at the end of this chapter displays norms established for the ECR R using a sample of more than 22,000 people. Overall mean score for attachment avoidance was 2.93, with the mean for women (2.95) slightly higher than the mean for men (2.88). The overall mean fo r attachment anxiety was 3.64, with women (3.64) and men (3.64) scoring equally. Married respondents were slightly lower in attachment avoidance (2.87 versus 2.94 for singles) and attachment anxiety (3.64 compared with 3.71 for singles). Women were more highly represented in the sample (78% to 22%) than were men, as were singles (85%) compared to those married (15%). Normative data showed attachment avoidance progressively increasing with age throughout adult hood (from 2.90 at age 20 building to 3.18 at age 60), with a reverse trend for attachment anxiety (3.67 at age 20 declining to 3.23 at age 60). Yet the average age reported (24) indicates an underrepresentation of older adults in the normative sample data (Fraley, R. C., 2010) The ECR R usually appears as separate listings of the 18 attachment avoidance items and the 18 attachment anxiety items (e.g., Mikulincer & Shaver, 2007). Fraley recommended joining and randomizing the avoidance and anxiety items for each R. He als o suggested modifying the wording of items to reflect the breadth or specificity of close relationships targeted by the particular research (2010). Following these recommendations, both item chronology changes and item wording changes were made to adapt t he ECR R for use in the current study. The two
106 18 item scales were joined and randomized, similar items remaining in close proximity were subsequently moved, and first page two line items were exchanged with similar second page one line items so that half of the items could be presented on each page (while accommodating first page instructions). No wording changes were necessary for the four items that sampled generalized attachment tendencies. The 22 original ECR y omantic sampling items. For example, the original ECR u those minimally necessary to retain the sense of an item after the earlier specified changes had been made. A copy of the ECR R Adapted Version (ECR R AV) used for this st udy can be located in appendix E. Like the (non randomized and romantic partner generalized) ECR R developed by Fraley et al. (2000), the ECR R AV developed for the current study uses two separate 18 item scales to assess attachment avoidance and anxiety, and 7 point Likert response options to elicit from respondents the degree to which each of the attachment facet sampling items describes them. The inventory and its instructions are readable at a seventh grade level. Scoring is accomplished by summing th e Likert response levels
107 (1 7) for the 22 direct keyed items and (7 1) for the 14 reverse keyed items, and then averaging the scores for each scale. Procedures Permission to conduct this study was obtained from the University of Florida Institutional Revie w Board (UFIRB). The UFIRB approval form can be found in appendix A. The purpose of this section is to describe and discuss the various procedures followed to conduct this study: (1) recruiting prospective participants; (2) obtaining informed consent; (3 ) assigning confidential identifiers; (4) assignment to experimental and control groups; (5) scheduling participation; and (6) experimental protocol and data gathering. Recruiting Participants Participant recruitment began with a letter to the senior pasto rs of selected churches, requesting the opportunity to present the study to their respective congregants (appendix B). Subsequent to the letter of request, the principal investigator met face to face with each pastor (or designated associate) to answer qu estions and relieve any apprehensions. When access to their congregants was granted, a recruiting visit was scheduled and associated details were coordinated. One weekend recruiting effort was conducted for each of the ten participating churches. Six of the ten churches offered multiple primary worship services, ranging from two to five services each. The opportunity to participate in the study was presented at 17 of the 21 distinct gatherings of congregants. Four of the gatherings were excluded becaus e most congregants attending these more traditional services exceeded the upper age limit for study participation.
108 In seven of the ten churches, the principle investigator was permitted to announce the opportunity from the pulpit. In two of the churc hes attention to the opportunity, and in one church no auditory announcement was made. weekly bulletin. The prototype for the printed ann ouncement can be found in appendix C. The first section of the announcement was used for each of the churches. The announcement and was included in the written announcement at those churches where Couples interested in study participation were directed in both verbal and printed announcements to meet the principal investigator after the worship service at a des ignated location (e.g., church foyer). Obtaining Informed Consent Couples contemplating study participation gathered at the designated location subsequent to each worship service recruited. Questions were asked and answered. Those choosing to parti cipate read and completed the informed consent document (appendix D). A few whose spouses were not in attendance that day completed the document with the understanding that both spouses would need to have read and signed the consent form prior to particip ation. Couple s were offered blank copies of the consent form for their own records. The consent form included a section below the signature lines for couples to designate their ages, the date their marriage began, and their e mail and telephone contact i nformation. Reporting of age was necessary to verify study age parameter participation eligibility. Date of marriage was necessary to inform group assignment (see next subsection). Contact information was necessary to
109 facilitate later scheduling of ex perimental protocol session and subsequent reward seminar participation. Assigning Confidential Identifiers Couples completing the informed c onsent document were assigned identification numbers both to protect their identities and to facilitate t racking functions. These numbers were assigned by the principal investigator upon retreat to a private setting after the securing of informed consent. A master list matching participant names to assigned identifiers was compiled, and the list was maintai ned in a secure location by the principal investigator. The assigned identifier was the exclusive participant identification used by those accessing study data. The assigned alpha numeric identifier was in the form of ## ## H ## or ## ## W ##. The firs t pair of digits (01 thru 10) designate d the church from which the couple was recruited. The second pair of digits (01 and above) represent ed the number assigned to each couple participating from that church. This number was determined by arranging each d husband or wife. And the final pair of digits (00 and above) represent ed the length of curr ent marriage T his number represented the (rounded) whole numbe r of years that the couple had been married to each other on the date of testing protocol participation. Assignment to Experimental and Control Groups Random group assignment was stratified by years of marriage by arranging each couple identifiers on the basis of the final two digits, beginning with the smallest and ending with the largest, and then distributing participant couples (as a paired unit) alternately to each of the four groups. For example, the most recently
110 married couple on the first church recruitment roster was assigned to the first group, and the next most recently married couple was assigned to the second group, and so forth. The newest married couple on the second church recruitment roster was assigned to the next group in rotation after the assignment of the longest married couple on the fir st church recruitment roster. This pattern of alternating group assignment was maintained for each of the first ei ght recruited church ro sters. The rosters of the ninth and tenth churches were use d for rebalancing of the groups, with the goals of equivalency in number of couples and parity in mean and median length of marriage. The initial rebalancing effort involv ed using newly recruited couples from the ninth church to replace twelve earlier recruited couples who had dropped out of the study. To accomplish this task while retaining group length of marriage parity, the newest wed couple on the roster of the ninth church was assigned to the same group as the newest wed dropping out couple had been assigned. The second newest wed ninth church couple was assigned to the same group as had been the second newest wed drop ping out couple, and so forth. Group assignment for the remaining ninth church couples (beyond the twelve earlier assigned) and all the tenth church couples was guided by the need to replace subsequent drop out and no show couples and to maintain the aforementioned goals of group balance. Scheduling Pa rticipation Where permitted and feasible, one or more testing opportunities were provided for couples on their own church campus soon after the recruiting visit. When an on site opportunity could not be offered, couples wer e invited to join participants a t a neighboring church testing session or to participate at one of the many testing sessions office group room. T en total testing
111 sessions were provided on seven church campuses Fifty eight couples part icipated in the experimental protocol on their own church campuses, joined by an additional five couples participating at church campuses other than their own. Sixty nine couples participated in an experimental protocol session offered at the principal in counseling office group room. Couple participation scheduling was facilitated by e mail, with occasional follow up by phone. To organize this effort, twenty e mail banks were established a pre test and post test bank for each of the ten churc hes. Participant e mail addresses were entered test banks after their completion of informed consent documents. Periodic reminders of upcoming testing opportunities were sent to all pre test bank addresses until respon ses were received scheduling a testing session or requesting release from the study. Participant e mail addresses were moved to their test banks after their completion of the testing protocol. Post test e mail banks were used to facilitate registration for the couple communication seminars offered in appreciation of study participation. Experimental Protocol and Data Gathering The purpose of this section is to describe and discuss the following: (a) selection criteria and p reparedness training for the surrogate who administered the experimental protocol; (b) the various stages of the experimental protocol administration, including pre and post administration activities; and (c) the pilot study experience that informed refin ement of the experimental protocol. Protocol administrator The experimental protocol was administered by a surrogate whose credentials include d a graduate degree in church min istry with a special emphasis on interpersonal
112 background in church ministry enhanced pastoral comfort level necessary to gain cooperation for couple recruitment and on site study enhanced volunteer participant comfort and cooperation. Because some of the prospective participant couples knew the principal investigator through earlier speaking engagements or in the context of his longstanding local practice as a marriage counselor, the use of a surrogate to administer t he experimental protocol served to protect the integrity of the study from inadvertent influence or the perception of same. Use of the same surrogate for all 42 administrations of the experimental protocol ensured uniformity of experience for group partic ipants. The surrogate underwent a six phased training regimen preparatory to her administration of the experimental protocol: (1) First, the surrogate experienced an earlier version of the experimental protocol as a participant herself (Group 2 protocol). version of the protocol to 20 pilot study participants in the graduate Marriage and Family s erving as the recorder in the post session dialogue between the principal investigator and the pilot study participants. (3) Next, the principal investigator and the surrogate dialogued about protocol design changes informed by the pilot study experience. (4) Next, the surrogate familiarized herself with the final version of the experimental protocol. The experimental protocol is summarized in Figure 3 3 located at the end of this chapter It is discussed in detail later in this section, and copies of i ts various participant tasks and related documents can be found in appendices E through P (5) Next, the surrogate
113 performed a trial administration of the final version of the experimental protocol on two separate occasions to adult couples who were not p articipants in the study proper. (6) And finally, the surrogate dialogued with the principal investigator about concerns and questions that arose out of her trial administration experiences. The principal investigator was available in the background or in the vicinity at each of the protocol administration sessions. Although his role was primarily one of helping the surrogate to coordinate set up and testing materials organization, his availability served to provide surrogate feedback after the early admi nistration sessions, to run interference with any outside distractions, and to bolster surrogate confidence of assistance in the event of any unexpected occurrences. Experimental protocol overview Each administration of the experimental protocol was pe rformed in a single session of approximately 50 55 minutes duration. Administration times were moderately compressed in sessions with fewer couple participants, because less than the maximum allotted time per section was utilized when it was clear that ev eryone had completed the section. Because multiple experimental or control conditions were represented in most administrations of the protocol, the four differing conditions were constructed with equivalently timed tasks, and the surrogate script guided p articipants through the group differing portions of their packets with general instructions and time prompts. A copy of the full session administration instructions and script can be located in appendix P Appendices E, F, G, L, M, N, and O reflect the participant tasks common to all four groups. Appendices H through K represent the differing conditions of the four groups (appendix H = secu re attachment priming condition [Group 1]; appendix I = av oidant
11 4 attachment priming condition [Group 2]; appendix J = anxious attachment priming condition [Group 3]; and appendix K = neut ral ( control ) condition [Group 4]). A complete discussion of the common and differing experimental and control condition tasks can be found in the following subsection. An overview of the various pre and post session activities and the seven participant tasks can be found in Figur e 3 4 at the end of this chapter. Experimental protocol session stages Beginning the session Testing room chairs were configured to allow ample spacing among participants. Seating assignments were made to limit tactile and eye contact between spouses. Because several of the session tasks required participants to focus on past memories of marriage partner distress, efforts to block spousal immediacy were intended to hinder present spousal int eractions from interfering with engagement in the targeted memories. The session administrator distributed participant packets along with clipboards and pens, welcomed the participants, and provided opening instructi ons. To ensure correct distribution, each participant packet was numeric identifier, but none of sided worksheets of each participant packet were stapled together to avoid inadvertent loss or intermingling of data. Participant task #1. Participants in all four groups comple ted the 36 item Experiences in Close Relationships Questionnaire Revised (ECR R) (Fraley, Waller, & Brennan, 2000), as adapted to this study. The results of the ECR R Adapted Version (ECR R AV) were obtained to assign attachment style avoidance and anxiet y scores for
115 experimental or control conditions. A copy of this participant task can be located in appendix E. Participant task #2. Participants in all four groups complet ed a brief memory exercise distracter task in which they were asked to recall several rather innocuous memories from their past and to rate the certainty of their recollections using a 5 point Likert scale. The purpose of this task was two fold: (a) to di from the questionnaire completed in the previous task, diluting the likelihood that recollections of the ECR R ubsequent tasks as assessments of their ability to recall events with certainty. The intent was that encourage them to push past their resistance to retrieval of unpleasant memories in subsequent tasks. A copy of this participant task can be located in appendix F. Participant task #3 Participants in all four groups were asked to search their memories for a specific episode in which their marriage partner experienced distress and wante d their comfort and support. Participants were asked to observe the following five criteria in episode memory selection: (a) Episode selected should be one in which ss would serve as insufficient stimuli for dependent variable measures. (b) Episode selected should be one in which the participant was not primarily to blame for the cause focused responses at a level likely to dilute the other focused responses central to dependent
116 variable measures. (c) Episode selected should be one whose impact more directly ss over the to his or her own distress. (d) Episode selected should be one tha t did not result in a grand once and for distress. Recollections of triumph surrounding sweeping problem resolution might overshadow participant recollection of partner distress preceding tha t resolution. And (e) Episode selected should be one that the participant could imagine recurring at some point in the future. In other words, the participant could imagine someday experiencing an episode with her or his partner similar to the one rememb ered. Participants were guided with reasoning, examples and prompts, to help them select and commit to a targeted memory before they moved on to the next task. A copy of this participant task can be located in appendix G. Participant task #4. Thi s task involved the manipulation of the independent variables of the priming of attachment security, attachment avoidance, or attachment such, it was the only task provi ding differing experiences for the three experimental groups and the control group. Participants in the three experimental groups were asked to recall and to reflect upon a close relationship that substantially matched the characteristics of a prototype p rovided. These prototypes were adapted from those
117 blending of two sets of prototy pes used by the above authors in parallel studies. Tables 3 8 thru 3 10 located at the end of this chapter, display in parallel columns the two sets of prototypes used by Bartz and Lydon (2004), along with the composite prototypes developed from them for use in this study. Participants were guided by questions, prompts and exercises to help them maximize and sustain their engagement in the memory of a close relationship matching their group point Likert ratings of memo ry vividness provided the first of several measures of memory engagement used in primary data analyses Control group participants experienced a parallel exercise designed to prompt reflection on a casual acquaintance, following guidelines that served to avoid the intentional priming of specific attachment representations. Copies of participant task variations for each of the four groups can be located in appendices H through K. Participant task #5. In this task, participants were asked to return to the m emory retrieved and selected in Task #3. Participants were guided by questions and prompts, to help them maximize their engagement in the memory of the episode earlier selected. Participants were asked to reflect upon what they were feeling, thinking, an d wanting to do as they remember the targeted episode. Then participants were asked to provide 7 point Likert ratings of the vividness of the remembered episode as well as 7 point Likert ratings of the level of distress they believed their partner was exp eriencing in the episode they chose to think about. After providing these ratings, participants were asked to imagine a present recurrence of an episode of partner distress with dynamics similar to the episode earlier remembered. Participants were guided by questions and prompts, to help them
118 maximize their engagement in the imagined recurrence of the earlier episode. Participants were asked to reflect upon what they were feeling, thinking, and wanting to participants were asked to provide 7 ratings provided in the task #4 treatment variations, th ese measures were included to facilitate examination of participant memor y engagement as a contributor to dependent variable response levels Specifically, it was of interest to explore the degree to which weaker episode memory vividness and/or selection of episodes eliciting lower partner distress might explain less robust pa rticipant empathy and/or distress responses. A copy of this participant task can be located in appendix L. Participant task #6. In this task, participants were asked to make a concerted effort to maintain focus on the imagined present recurrence of t he earlier targeted episode of their marriage partner in distress. While maintaining this focus, participants were asked to continue noticing what they were feeling, thinking, and wanting to do. Participants were then asked to report the degree to which they were currently experiencing empathy and distress related emotions, cognitions, and motivations in regard to the event recurrence they were imagining. Participant feedback was provided through 7 Indices Adapted Version (BEDI AV) (Batson et al., 1983; Batson et al., 1987; Batson et al., 1989) and to the Pity Experience Inventories Short Form Revised (PEI SF R) (Florian, Mikulincer & Hirschberger, 2000), developed for this study (see discussions un der the section entitled Instrumentation ). At the bottom of each of the three pages of responses to these 44 items, participants were asked to provide updated 7 point Likert ratings of the
119 vividness of the imagined present recurrence of the remembered epi sode. This feedback mechanism was designed both to remind participants to remain focused on the imagined episode recurrence while responding to t he items and to inform memory engagement variable analyses of participant responses (as described for task #5) A copy of the two instruments used in this participant task can be located in appendices M and N. Participant task #7. In this task, participants were asked to complete a two page questionnaire inquiring about the quality of their marital comm unication as well as the quality of their communication with other partners in any earlier significant adult romantic relationships. In regard to their current marriage, participants were asked to rate the relative frequency of incidence of self and part ner distress, as well as their regard to any earlier partner relationships, participants were asked to report on (up to) three earlier relationships that were most s ignificant, and then beginning with the most distress in supportive ways. For each earlier relationship, participants were asked to designate the length of the r elationship in years, and to specify whether it was a marriage, a cohabitating relationship, or merely an exclusive dating relationship. Seven point Likert responses were used for all ratings questions. Participant responses to this questionnaire were us ed for post hoc analyses of secondary variable s of interest. A copy of this participant task can be located in appendix O. After completion of this task, participants were asked to return worksheet packets to their envelopes of origin.
120 Emotional equilibr ium restoring activity. After all packets were secured in their envelopes, the session administrator acknowledged the possibility that unwanted emotions aroused in the session might linger on for some participants. As a means of helping to dissipate such emotions, the session administrator presented a reading designed to help restore a sense of being cared for and comforted. This reading was a composite of comfort giving Psalms appropriate for participants recruited from church worship services, and inten ded to prime secure attachment representations within session administration script, and can be located in appendix P. After completion of the reading, the session admi nistrator announced that participants who remained concerned about their level of emotional distress we re welcome to remain after the session for additional assistance. Post session support offered and provided included equilibrium restoring dialogue and prayer, as well as help with finding a class or a counselor to assist with concerns brought to greater awareness by study task performance. P articipant couples lingered at the end of several of the experimental protocol sessions, and the session administr ator provided the assistance or support requested Per instructions, the session administrator was intentional in marriage counseling practice. Ending the session. The session administrator collected the participant packet envelopes, thanked participants for their valuable contribution to important research, and requested that couples refrain from discussing session content with any prospective participants who had not yet experienced the protocol. The session administrator then
121 reminded couples to avail themselves of their participation rewards: Selection (that day) of their choice between two marriage communication books, and opportunity to attend either of two one da y marriage communication seminars scheduled during the month following the three m onth testing protocol season The principal investigator was available at the end of each testing session to provide feedback and to answer questions about the free book cho ices and the free reward seminars. After participants were dismissed, the session administrator remained behind to provide any of the aforementioned support activities that were requested. After the session. The session administrator delivered to the pri ncipal investigator the envelopes containing the participant packets, calling to his attention any upon data integrity. Immediately after each session, the principal in vestigator updated the testing log and the appropriate church rosters. Testing logs entries denoted date, time, and location of protocol administration, the number of participant couples from each church, and any noteworthy happenings that might bear upon data integrity. Church rosters were updated to reflect the specific testing session completed by each couple The principal investigator stored the participant packet envelopes in a secure location until they were accessed for data entry. When the enve lopes were opened, the principal investigator consulted the master list of participant names and assigned identifiers to verify correct match between participant packet identifier numbers and envelope names. After matches were confirmed, the envelopes wer e discarded. From that point on, the master list was secured for study integrity verification only, and participants were identified by alpha numeric designation only.
122 Pilot study findings Analyses of the pilot study data revealed patterns somewhat cons istent with the primary directional hypotheses. Nevertheless, several factors demonstrated the limitations of these findings: (1) Group size was insufficient (N=5 for each of the four groups). (2) There was a less than optimum fit between the pilot part icipants and the targeted study participants (e.g., a number of the pilot participants were unmarried, and some reported either no current relationship partner or limited experiences with their current partner). And (3) Experimental design deficiencies bec ame apparent as a result of the pilot study experience (see discussion below). Most significant among the design changes informed by pilot study participant memory of t heir partner in distress (see Participant T ask #3 earlier in this section). These criteria grew out of questions and comments by pilot participants about the partner in distress memory retrieval task. Reflection on these questions following pilot partici pant feedback made it clear that additional partner in distress episode selection criteria were essential to protect against multiple intervening variable dilution of participant dependent variable response data. Later reflection on the pilot study experie nce led to the making of an additional revision in the final experimental protocol design. In the pilot study, the attachment priming task preceded the task of selecting and reflecting upon the partner in distress episode. A review of individual pilot pa rticipant task sheet responses suggested the possibility that selecting the targeted episode of partner distress after attachment priming might introduce an experimental condition specific episode selection bias into that process. To correct this design f law, the original task of selecting and reflecting on
123 the partner in distress episode was divided into two tasks: The selection of the partner in distress memory (Participant Task #3) was placed before the attachment priming experience (Participant Task #4 ). While reflecting on the partner in distress episode and imagining its recurrence (Participant Task #5) remained in its original sequence of following the attachment priming experience. Additional pilot participant feedback was useful in revising the c instructions, in establishing timing parameters for each task, and in shaping the final protocol admi nistration script can be located in appendix P A critical component of pilot participant feedback was an examination of the level of residual distress experienced by participants in the aftermath of the experimental protocol experience. Pilot participant s rated the degree to which their distress levels had increased (post protocol), using a 7 participants, one reported a de crease in distress level, five reported no change, nine for reducing their distress t o non worrisome levels, those reporting higher distress levels described sel f actions they were already implementing and their confidence in the adequacy of those interventions This pilot participant feedback was used to inform emotional equilibrium rest oring activities later made available to participants in the full study.
124 Data Analyses Pearson Correlations were used to examine the relationships between and among the various demographic, intervening, and dependent variables of interest in this study. Pearson correlations and cross correlations of matched of the Relationship Communication History Survey and the multiple variables of intere st. M were used to tests for multiple comparisons were used to confirm ANOVA findings of significance. The eleven and partial correlations. And various post hoc tests were performed to examine other relationships of interest. An alph .05 was used to test the significance of observed diff erences, both for evaluating the hypotheses and for all post hoc comparisons.
125 Table 3 1. (1 of 2) (6 Earlier S tudies ) [Adapted from Batson et al., 1987] [continued in Table 3 2 ] Study # 1 2 3 Factor Loading D E D E D E Distress Ad jectives Alarmed *.75 0.01 *.72 0.49 *.63 0.15 Grieved 0.51 0.49 *.65 0.48 0.55 0.58 Upset *.84 0.39 *.82 0.32 *.74 0.38 Worried 0.40 *.60 *.87 0.18 *.67 0.35 Disturbed *.83 0.35 *.82 0.38 *.76 0.20 Perturbe d *.84 0.17 0.59 0.11 *.76 0.18 Distressed *.62 0.56 *.65 0.48 *.81 0.32 Troubled *.88 0.23 0.58 0.54 *.80 0.22 Empathy Adjectives Sympathetic ----0.58 0.53 0.23 *.74 Moved 0.31 *.75 0.37 *.78 0. 41 *.78 Compassionate 0.25 *.80 0.09 *.82 0.40 *.73 Tender ----*.66 0.32 0.18 *.86 Warm 0.05 *.82 0.23 *.71 0.03 *.80 Softhearted 0.12 *.85 0.14 *.73 0.11 *.80 Note: Study 1 (Coke et al., 1978); Study 2 (Batson et al., 19 79); Study 3 (Coke, 1980); Values noted are factor loadings: D = Distress component; E = Empathy Component; Studies 1 3: Female participants only; Asterisk (*) denotes factor loading above .60
126 Table 3 2. Loadings (1 of 2) (6 Earlier Studies) [Adapted from Batson et al., 1987] [continued from Table 3 1 ] Study # 4 5 6 Factor Loadings D E D E D E Distress Adjectives Alarmed *.72 0.34 *.77 0.11 *.80 0.19 Grieved *.70 0.33 *.68 0.42 *.72 0.30 Upset *.80 0.38 *.87 0.17 *.89 0.28 Worried *.72 0.34 *.78 0.18 *.81 0.39 Di sturbed *.76 0.38 *.89 0.18 *.90 0.24 Perturbed *.69 0.13 *.82 0.02 *.68 0.11 Distressed *.67 0.48 *.87 0.25 *.86 0.28 Troubled *.75 0.33 0.59 0.39 *.87 0.32 Empathy Adjectives Sympathetic 0.29 *.69 0.04 *.84 0.20 *.82 Moved 0.42 *.74 0.31 *.67 0.40 *.72 Compassionate 0.24 *.80 0.14 *.86 0.17 *.90 Tender 0.28 *.78 0.31 *.78 0.36 *.74 Warm 0.19 *.80 0.20 *.68 0.15 *.66 Softhearted 0.17 *.86 0.05 *.83 0.29 *.86 Note : Study 4 (Toi & Batson, 1982); Study 5 (Fultz, 1982); Study 6 (Batson et al., 1980); Values noted are factor loadings: D = Distress component; E = Empathy Component; Study 4: Female participants only; Studies 5 and 6 Both Females and Males; Asterisk (*) d enotes factor loading above .60
127 Table 3 3. Pity Experience Inventories Factor Loadings [Feelings Items] (1 of 2) (adapted from Florian, et al., 2000) [continued in Table 3 4 ] Item Groupings / Items Factor 1 Factor 2 Sorrow related feelings : False superiority factor Sorrow 0.77 Sadness 0.72 Participation in grief 0.65 Anguish 0.58 Hope that suffering will end 0.58 Helplessness 0.55 Emotional pain 0.53 Despair 0.47 Anger at the cause of the suffering 0.45 Emotional discomfort: False superiority factor Guilt feelings 0.73 Shame 0.70 Pangs of conscience 0.67 Discomfort 0.60 Disgust 0.58 Embarrassment 0.57 Explained Variance (%) 26% 12%
128 Table 3 4. Pity Experience Inventories Factor Loadings [Feelings Items] (2 of 2) (adapted from Florian, et al., 2000) [c ontinued from Table 3 3 ] Item Groupings / Items Factor 3 Factor 4 Factor 5 Vulnerability feelings: False superiority factor Feeling you might get hurt 0.83 Concerns over the future 0.77 Fear of suffering a similar misfortune 0.64 An xiety 0.55 Feeling exposed 0.45 Sympathy related feelings: Compassionate caring factor Love 0.81 Closeness and warmth 0.77 Sympathy 0.64 Emotional identification 0.55 Superiority feelings: False superiority factor Sense of superiority 0.85 Haughtiness 0.85 Feelings of power 0.77 Joy over your own good fortune 0.45 Explained Variance (%) 7% 5% 5%
129 Table 3 5. Pity Experience Inventories Factor Loadings [Thoughts Items] (adapted from Florian, et al., 2000) Item Groupings / Items Factor 1 Factor 2 Factor 3 Action oriented thoughts: Compassionate caring factor How to encourage the other 0.81 How to change the situation 0.78 Your obligation to help 0.74 How to cause pleasant feelings 0.71 The need to comfort 0.68 The way in which you behaved 0.65 Misfortune contemplation: Passive identification factor How unfortunate the oth er person is 0.79 0.70 Why did this have to happen 0.68 0.67 Self focused thoughts: False superiority factor The possibil ity of ending up similarly 0.76 0.69 How comparatively fortunate you are 0.63 Similar situations you are familiar with 0.62 Your part in causing this situation 0.51 Explained Variance (%) 34% 14% 9%
130 Table 3 6. Pity Experience Inventories Factor Loadings [Wishes Items] (adapted fr om Florian, et al., 2000) Item Groupings / Items Factor 1 Factor 2 Factor 3 Care oriented wishes: Compassionate caring factor To talk to the other 0.81 To comfort the other 0.81 To get close to the other 0.79 To understand the other 0.78 To encourage and cheer up the other 0.78 To calm the other 0.77 To hug and caress the other 0.73 To get to know the other better 0.70 Unrealistic wishes: Passive identification factor To turn back the clock 0.80 Wanting situation to suddenly go away 0.79 That a miracle will happen 0.72 That the suffering will cease 0.67 Distancing wishes: False superiority factor To run away from the scene 0.85 To distance from the scene 0. 80 To forget the event 0.73 Explained Variance (%) 38% 19% 8%
131 Table 3 7. ECR R Normative Summary Statistics University of Illinois at Urbana Champaign W ebsite (2010)] Avoidance Anxiety Overall (full sample) M = 2.93 SD = 1.18 M = 3.64 SD = 1.33 Male M = 2.88 SD = 1.15 M = 3.64 SD = 1.33 Female M = 2.95 SD = 1.19 M = 3.64 SD = 1.33 Married M = 2.87 SD = 1.27 M = 3.64 SD = 1.33 Si ngle M = 2.94 SD = 1.16 M = 3.71 SD = 1.31 Age 20 M = 2.90 M = 3.67 Age 30 M = 2.97 M = 3.56 Age 40 M = 3.04 M = 3.45 Age 50 M = 3.11 M = 3.34 Age 60 M = 3.18 M = 3.23 Note: Normative data established on a sample exceeding 22 ,000 people ; Average age = 24; SD = 10 ; Female = 78%; Male = 22%; Single = 85%; Married = 15% ; Age values represent predictions based on regression model ; Full sample correlation between avoidance and anxiety scales = .41
132 Table 3 8 Development of Secur e Attachment Prototy pe Priming Descriptor [Blending of Earlier Study Descriptors for Use in Current Study ] Secure attachment prime (Bartz & Lydon, 2004 [Study 1] Secure attachment prime (Bartz & Lydon, 2004 [Study 2] Secure attachment prime (Current Study) [Task #4; Group 1] Please think about a relationship you have had in which you have found that it was relatively easy to get close to the other person and you felt comfortable depending on the other person. In this o ften worry about being abandoned by the other worry about the other person getting too close to you. Please think about a relationship you have had in which you have found that it was easy to be emotionally close to the other person. In this relationship, you felt comfortable depending on the other person and having them depend on you. In this relationship worry about being alone or about the other person not accepting you. In this task, we want you to take a fe w moments to think about a relationship you have had in which you found it relatively easy to be emotionally close to the other person. In this relationship, you felt comfortable depending on this person and having this person depend on you. In this rela worry about being abandoned by this person or about this person wanting to be closer than you wanted her or him to be. In this relationship you being alone or about this person not accepting you.
133 Table 3 9 Development of Avoidant Attachme nt Prototype Priming Descriptor [ Blending of Earlier Study Descriptors for Use in Current Study] Avoidant attachment prime (Bartz & Lydon, 2004 [Study 1] Avoidant attachment prime (Bartz & Lydon, 2004 [Study 2] Avoidant attachment prime (Current Study) [Task #4; Group 2] Please think about a relationship you have had in which you have found that you were somewhat uncomfortable being too close to the other person. In this relationship you found it was diff icult to trust the other person completely and it was difficult to allow yourself to depend on the other person. In this relationship you felt yourself getting nervous when the other person tried to get too close to you and you felt that the other person wanted to be more intimate than you felt comfortable being. Please think about a relationship you have had in which you felt comfortable not being emotionally close to the other person. In this relationship you felt that it was very important to be indepe ndent and self sufficient and you preferred not to depend on the other person or have the other person depend on you. In this task, we want you to take a few moments to think about a relationship you have had in which you felt somewhat uncomfortable being emotionally close to the other person. In this relationship you found it difficult to trust this person completely, or you felt it was very important to be independent and self sufficient and not depend on this person or have this person depend on you. I n this relationship you felt this person wanted to be closer than you did, and his or her efforts to get closer made you feel nervous.
134 Table 3 10 Development of Anxious Attachment Prototype Priming Descriptor [ Blending of Earlier Study Descriptors f or Use in Current Study] Anxious attachment prime (Bartz & Lydon, 2004 [Study 1] Anxious attachment prime (Bartz & Lydon, 2004 [Study 2] Anxious attachment prime (Current Study) [Task #4; Group 3] Please think about a relationship you have had in which you have felt like the other person was reluctant to get as close as you would have liked. In this relationship you worried that the other person didn't really like you, or love you, and you worried that they wouldn't want to stay with you. In this relationship you wanted to get very close to the other person but you worried that this would scare the other person away. Please think about a relationship you have had in which you have felt like you wanted to be completely emotionally intimate with the other person but felt that the other person was reluctant to get as emotionally close as you would have liked. In this relationship you felt uncomfortable being alone and worried that the other person didn't value you as much as you valued them. In t his task, we want you to take a few moments to think about a relationship you have had in which you felt like the other person was reluctant to get as close as you would have liked. In this relationship you worried that the other person didn't really love you or value you as much as you did her or him, and you worried that he or she wouldn't want to stay with you. In this relationship, you wanted to get emotionally closer to this person, but you worried that if you tried to get closer you might scare this person away.
135 Items with Adequate Face Validity for Response to Partner Distress Feelings Thoughts Wishes 23 + 8 + 14 = 45 suitable items Empathy Items Distress Items Compassionate Caring Passive Identification False Superiority 15 items 16 items 14 items Average % of Inventory Variance Attributable to Each Factor Empathy Items Distress Items Compassionate Car ing Passive Identification False Superiority 26% 20% 14% % of Variance Explained by a Factor Determines # of Items Selected 30 Total Items Prioritizing High Factor Loadings and High Face Validity Empathy Items Distress Items Compassionate Caring Passive Identification False Superiority 13 items 10 items 7 items D ata analyses were performed utilizing full 17 item PEI distress index. Post h oc distress item analyses performed to examine avoidance/anxiety differentiating potential for the informing of future study. Passive Identification Factor Items False Superiority Factor Items Items thought to diff erentiate both avoidance and anxiety from attachment security equally well Items thought to sometimes sacrifice security/insecurity differentiation in favor of more effectively differentiating between avoidance and anxiety Figure 3 3. PEI SF R Instrumen t Development: Rationale and Process Pity Experience Inventories Beliefs Feelings Thoughts Wishes Behaviors 20 + 28 + 15 + 16 + 17 = 96 total items Inventories Relevant To Study Design Feelings Thoughts Wishes 28 + 15 + 16 = 59 relevant items
136 Distribution of Participant Task Sheet Packets, Clipboards, and Pens Seating and Welcoming of Participants / Opening Instructions Participant Task #1 (appendix E) 36 Item ECR R Assessing Intervening Variab le of Attachment Style Participant Task #2 (appendix F) Brief Memory Exercise Distracter Task Participant Task #3 (appendix G) Part icipants Select Memory of Partner in Distress for Later Reflection Participant Task #4 (appendices H thru K) Four Groups Assigned Differing Relationship Prototypes to Prime Accessibility of Differing Attachment Representations in Memory (Four Treatment Conditions along with Gender are Primary Independent Variables) Attachment Security Pr iming Attachment Avoidance Priming Attachment Anxiety Priming Control Group Neutral Condition Participant Task #5 (appendix L) Participants Recall, Reflect on, and Re experience Task #3 Memory Selected Participant Task #6 (appendices M and N) Participants Report on Feeli ngs, Thoughts and Wishes Experienced in Task #5 Using Two Dependent Variable Measures (44 Total Items) Participant Task #7 (appendix O) Survey of Present and Past Relationship Communication Quality (For Post Hoc Analyses of Secondary Independent Variables of Interest) Administrator Led Emotional Equilibrium Restoring Activity ( included in appendix P) Collection of Participant Packets / Expression of Thanks / Selection of Free Book Announcement about Free Semi nar / Reminder of Availability of Post Session Support Figure 3 4. Experimental Protocol Session Activities and Participant Tasks
137 CHAPTER 4 DATA ANALYSES AND RE SULTS The purpose of this study was to investigate the effects of attachment style and conte xtual activation of attachment security and attachment insecurity on partner empathy and distress responses to episodes of romantic partner distress. More specifically, this research was designed to examine whether the priming of secure or insecure attach ment memory representations influences emotional, cognitive, and motivational responses toward one s distressed spouse, and how one s more generalized attachment style influences these processes. The purpose of this chapter is to provide the following: ( 1) a description of the sample; (2) reliability findings for adapted instrumentation; (3) a description of the data; (4) testing of the research hypotheses; and (5) a summary of the findings. Tables referenced in this chapter can be located at the end of Chapter 4 unless they are included within the immediate text. Description of the Sample The participants for this study were married couples between the ages of 20 and 59 years, recruited from 10 churches of various denominations in and around Gainesville, Florida. The sample included 132 couples (264 individuals) who volunteered in response to recruitment announcements at their weekly church worship services. Participants were recruited from the following churches: Grace United Methodist (N= 44); Vineyard Christian Fellowship (N=24); Servants of Christ [Anglican] (N=10); Campus Church of Christ (N=22); The Family Church [Interdenominational] (N=78); Saint Andrews Episcopal (N=16); Creekside Community [Evangelical Free] (N=20); Westside Baptist (N=10); First Assembly [Assemblies of God] (N=28); and First Baptist of Alachua (N=12). Comparative ease of participant recruitment from among
138 church attendees provided a benefit inherently linked to limitations in generalizing study findings to broader populations. A discussion of these limitations can be found in Chapter 5 Sixty six participants (33 couples) were assigned to each of four groups: Attachment security priming condition (Group 1); attachment avoidance priming condition (Group 2); attachme nt anxiety priming condition (Group 3); or neutral condition (Group 4). Random assignment was stratified to achieve approximate group parity relative to the demographic variable of length of current marriage De facto stratification was achieved relative to the demographic variable of gender by ranged from zero years (less than six months) to 39 years, with a median of 12 years, a mean of 13.85 years, and a standard deviation of 10.66. No significant group differences were found relative to this demograp hic variable. A summary of length of current marriage group comparisons for the recruited sample can be found in Table 4 1. Table 4 1 Length of Current Marriage (Recruited Sample) Group N Mean Median Std Dev Min Max Range 1 66 14.03 10 11.44 0 36 36 2 66 14.12 13 10.44 0 39 39 3 66 13.70 13 10.04 0 31 31 4 66 13.55 10 10.90 0 37 37 Total 264 13.85 12 10.66 0 39 39 A total of 30 participant data sets from t he recruited sample were discard ed, and the resultant sample (N= 234; hereinafter labeled the utilized sample ) was used for all but one area of data analyses (see discussion of this exception later in this subsection). The decision rules governing data set elimination are described later in this chapter. The length of current marriage range remained unchanged for each of the four groups
139 after data set reduction, with only modest changes in the median, mode, and standard deviation for the four groups. As w ith the recruited sample, there were no significant group differences relative to the length of current marriage demographic variable. Table 4 2 displays length of current marriage for the utilized sample in five year increments, and Table 4 3 provides a comparison of the four groups in regard to this variable of group stratification. Table 4 2 Years of Marriage 5 Year Increment Breakdown (Utilized Sample) Years Married N 5 Years or Less 69 6 10 Years 42 11 15 Years 31 16 20 Years 21 21 25 Years 32 26 30 Years 22 31 35 Years 8 36 Years or More 9 Total 234 Table 4 3 Length of Current Marriage (Utilized Sample) Group N Mean Median Std Dev Min Max Range 1 61 14.23 10 11.49 0 36 36 2 60 14.27 13 10.58 0 39 39 3 55 13.27 12 10.04 0 31 31 4 58 13.29 10 10.66 0 37 37 Total 234 13.78 12 10.67 0 39 39 (RCHS) responses required further reduction of the utilized sample (N=234). The nature of analyses desired for this seg ment of the data required that only paired reduced sample required the a set exclusion decision rules ( see Description
140 of the Data subsection later in this chapter). This resultant sample of 103 matched pairs data sets (N=206; hereinafter labeled the matched pairs sample ) was used for ble 4 4 provides an overview of how the recruited sample was reduced to the utilized sample for all but one area of analyses, and then was further reduced to the matched pairs sample for analyses of the Relationship Communication History Survey responses. Table 4 4 Summary of Stages of Sample Reduction for Data Analyses N (Total) N (Male) N (Female) Sample Parameters 264 132 132 Recruited Sample (total study participants) 234 117 1 17 Utilized Sample (all but RCHS analyses) 206 103 103 Matched Pairs Sample (RCHS analyses only) Reliability Findings Three instruments created for use in earlier psychological research were revised or adapted f or use in this study. Chapter 3 inclu des sections describing the development and validation of the original instruments, as well as the rationale and processes for adapting or revising these instruments for use in the current study. The purpose of this section is to report the reliability st atistics calculated from the use of the revised and adapted instruments in this study (based on the N=234 utilized sample). These statistics are then compared to reliability data reported for use of the original instruments in earlier studies. Empathy and Distress Indices Adapted Version (BEDI AV) consistency of the two subscales of the BEDI AV, adapted from the original BEDI for coefficient of .96 was determined for the six item empathy scale and a coefficient of .90 was determined for the eight item distress scale.
141 coefficients of .94 (empathy) and .88 ( distress) reported for studies referenced in Chapter 3 Pity Experience Inventories Short Form Revised (PEI SF R) consistency for the three subscales of the PEI SF R, revised from the PEI SF by coefficient of .94 was determined for the 13 item compassionate caring scale (empathy), a coefficient of .79 was determined for the 10 item passive identifica tion scale (distress), and a coefficient of .78 was determined for the 7 item false superiority item combined distress scale. These measures of reliability compare to the o riginal PEI SF reported for studies referenced in Chapter 3 Separate distress scale reliability coefficients were not calculated by the original PEI SF authors because of meager use of items from the false superiority scale. Experiences in Close Relationships Questionnaire Revised Adapted Version (ECR R AV) consistency for the two subscales of t he ECR R AV, adapted from the original ECR R item attachment avoidance scale and a coefficient of .90 was determined for the 18 item attachment anxiety scale. These me asures of reliability compare to the original ECR R
142 anxiety) reported for studies referenced in Chapter 3 Table 4 5 provides comparisons of reliability estimates for the vari ous instruments and subscales revised or adapted for use in this study with those of the original instruments as reported in Chapter 3 Table 4 5 Reliability Comparisons of Revised /Adapted Instruments with Originals Revised or Adapted Instrument Sca le or Composite / Number of Items Original Instrument Scale or Composite / Number of Items ECR R AV Avoidance (18) 0.93 ECR R Avoidance (18) 0.90 Anxiety (18) 0.90 Anxiety (18) 0.90 BEDI AV Empathy (6) 0.96 BEDI Empathy (6) 0.94 Distress (8) 0.90 Distress (8) 0.88 PEI SF R Empathy (13) [A] 0.94 PEI SF Empathy (14) [A] 0.89 Distress (10) [B] 0.79 Not Reported N/A Distress (7) [C] 0.78 Not Reported N/A PEI Distress (17) 0.84 PEI Distress (10) 0.88 Combined Total Empathy (19) 0.96 No t Reported N/A Combined Total Distress (25) 0.91 Not Reported N/A R AV = Experiences in Close Relationships Questionnaire Revised Adapted Version; BEDI Distress Indices Adapted Version; PEI SF R = Pity Experience Inventories Short Form Revised; [A] = Compassionat e Cari ng (CC) subscale; [B] = Passive Identification (PI) subscale; [C] = False Superiority (FI) subscale; Reliability estimates calculated on sample size of N=234 Reliability estimates for the instruments revised and/or adapted for use in this study compa re favorably with reported reliability estimates for the original instruments. Therefore, these instruments may be regarded as reliable measures for subsequent data analyses. Description of the Data Data Collection Activities Participants completed an ele ven page assessment packet comprised of seven tasks. The intervening variables of attachment avoidance and anxiety were calculated from the 36 item ECR R AV [Task #1]. The dependent variables of empathy and
143 distress scores were calculated from the 14 ite m BEDI AV [Task #6, page one] and the 30 item PEI SF R [Task #6, pages two and three]. Subscale scores were calculated for the PEI distress score. Indices of predominant emotional response (BEDI AV PER) and (PEI SF R PER) were calculated by subtracting from its empathy score. A total index of predominant emotional response (Total PER) was calculated by merging the two PER indices after conversion to standard scores. Participants completed seven additional rating scales iden tified as follows: (a) a scale for reporting vividness of memory engagement with group specific (attachment or control) prototype (prototype memory vividness) [Task #4]; (b) a scale for reporting level of distress experienced by spouse in remembered episod e of spousal distress (memory distress level) [Task #5]; (c) a scale for reporting vividness of memory of episode of spousal distress [Task #5]; (d) a scale for reporting vividness of imagined recurrence of episode of spousal distress [Task #5]; and (e) th ree scales for reporting continued vividness of imagined recurrence of episode of spousal distress [participants completed this scale at the end of each of the three pages of Task # 6 eliciting the dependent variable empathy and distress responses]. The f ive scales described in (c), (d), and (e) were combined to form one partner in distress memory vividness mean. Task #2 data (distracter task) were not analyzed. Narrative data from Tasks #3 and #4 were used to inform exclusion of data sets of part icipants who noted failure to engage in dependent variable activating memory of spousal distress episode [Task #3] or in memory engagement of interaction with group specific prototype [Task #4]. Participant responses to the current marriage portion of the Relationship Communication History Survey (RCHS) [Task #7, page one] were analyzed as
144 matched pairs only data sets (N=206). RCHS items relative to earlier romantic relationships [Task #7, page two] were omitted from analyses because of high incidence of missing, incomplete, or suspect responses. An overview of participant data collection categories and their use in analyses can be located in Table 4 6 at the end of this chapter. Data Inclusion Decision Rules After study participation, data sets for 30 of the participants were excluded for one or more of the following reasons: (a) grossly repetitive and/or nonsensical responses provided (3 participants); (b) indication of failure to retrieve memory of spousal distress episode matching criteria specified (7 participants); (c) indication of failure to retrieve memory of relationship matching group specific attachment prototype described (10 participants); and (d) omission of responses to one or more of the 87 (non RCHS) analyses scaled response items (10 part icipants). A miniscule portion of the data in the (N=234) data sets retained for analyses presented as non whole number responses (.04%; N=9 items). These incidents included the circling of two adjoining numbers on one of the vividness or severity scales (a total of eight incidents), or providing a numerical response representing the midpoint between two available response optio ns on one of the dependent variable measurement questions (only one incident). Each of these responses was rounded down to the l ower of the two whole number options indicated. Data Analyses Analyses of participant data is provided in two sections: (a) correlational data analyses and (b) experimental data analyses. The correlational data analyses section contains an examinatio n of relationships between and among the 17 demographic,
145 intervening, memory measure, and dependent variable measures for the utilized sample (N=234). It also includes correlations and cross correlations of spousal responses to the Relationship Communicat ion History Survey (RCHS), as well as correlations and cross correlations between spousal responses and 14 of the afore mentioned variables (N=206). All correlations reported in this section are zero order correlations. The experimental data analyses sec tion includes an examination of differences observed between and among the four groups: Attachment security priming condition; attachment avoidance priming condition; attachment anxiety priming condition; and neutral condition (N=234). Correlational data analyses Pearson product moment correlation coefficients were calculated for 17 variables: (a) the two demographic variables of gender and length of marriage; (b) the two levels of the intervening variable of attachment style; (c) the three measures of memory engagement; and (d) the ten dependent variable measures of empathy and distress, including scales, subscales, and composite measures. For ease of presentation, the two demographic variables and the two intervening variables were combined in one matrix and correlated with each other (Table 4 7), with the three measures of memory engagement (Table 4 8), and with the ten dependent variable measures (Table 4 9). The three measures of memory engagement were correlated with each other (Table 4 10) and with the ten dependent variable measures (Table 4 11). And the ten dependent variable measures were correlated with each other in two parts (Tables 4 12 and 4 13). Demographic variables. A negative correlation was found between gender and attachmen t anxiety (r = .237 were found between gender
146 No ot her significant relationships were found between the demographic variables and the other variables of interest. Tables 4 7 thru 4 9 display the results of Pearson correlations for each of the two demographic variables with the other variables of interest. Intervening variable measures. A positive correlation was found between the two intervening variables of attachment avoidance and attachment anxiety (r = .53 ; between t hese two scales: (a) D evelopers of the ECR R (see discussion in Chapter 3 ) found that the dimensions underlying avoidance and anxiety are essenti ally orthogonal (r = .12). (b) G eneral population norms reported for ECR R usage reveal avoidance/anxiety scal es correlati on of r = .41 (see Table 3 7). S uch findings suggest that despite minimal overlap in what these two constructs measure, individual attachment styles represent a blending of co mpeting attachment strategies. T hus individual attachment dimension scores tend to overlap more highly than do the underlying dimensions they assess (see Chapter 3 discussion). A nd (c) The higher degree of o verlap between the two scales for the current study sample (r = .53) versus that reported in the general population n orms (r = .41) is likely attributable t o the lower mean and standard deviation Study: Av: M=2.69; SD=1.05; Ax: M=2.70; SD=0.97; General Population Married Norm s : Av: M=2.87; SD=1.27; Ax: M=3.64; SD=1.3 3). C hapter 5 includes a discussion of two literature supported distinctions between the current study sample population and the general population that may help to acc ount for these differences
147 A negative correlation was found between attachment av oidance and the memory engagement measure of partner in distress vividness (r = correlations were found between attachment avoidance and eight of the ten dependent variable measu res. Six of these correlations 1 level: BEDI Empathy (r = .373); BEDI PER (r = .334); PEI CC (r = .380); distancing (r = .223); PEI PER (r = .327); and Total PER (r = .349). The remaining two were found to be .132); and PEI FS (r = .1 67). A negative correlation was found between attachment anxiety and the demographic variable of length of current marriage (r = correlations were found between attachment anxiety and seven of the ten dependent variable measu r es. Six of these correlations Empathy (r = .186); BEDI Distress (r = .185); BEDI PER (r = .255); PEI FS (r = .231); PEI PER (r = .212); Total PER (r = .2 43). The remaining correlation was found to be signif 7 thru 4 9 display the results of Pearson correlations for each of the two intervening variables with the other variables of interest. Memory engagement measure s. Significant positive correlat ions were found between the first and third measures of memory engagement (group prototype vividness and partner in and third measures (partner distress level and partner in distress vividness; r = .457; ere was found
148 between partner in distress vividness and the intervening variable of a ttachment avoidance (r = A positive correlation was found between group prototype vividness and the .05). Significant correlations were found between partner distress level and eig ht of the ten dependent variable measures. Six of the correla tions were positive and were significant at the (r = .291); distancing (r = .230); and PEI Distress (r = 399). The remainin g two correlations were negative .151) and Total PER (r = .137). Significant positive correlations were found between partner in distress vividness and five of the ten dependent va level: BEDI Empathy (r = .322); BEDI Distress (r = .309); PEI CC (r = .347); PEI PI (r = .309); and PEI Distress (r = .268). Tables 4 8, 4 10, and 4 11 display the results of Pearson correlations for each of the memory engagement variables with the other variables of interest. Dependent variable measure s. Significant correlations between the ten dependent variable measures and the seven demographic, intervening, and memory engagement variables are described in the prev ious three subsections. Tables 4 9 and 4 11 display the results of all correlations between these variables. Concurrent validity of the two dependent variable empathy measures (BEDI Empathy and PEI CC) was high (r = .822), and each was highly correlated with the measures of predominant emotional response [PER] of which it was a component (correlations ranging from .695 to .781). Correlation of the two dependent variable distress measures (BEDI Distress
149 and PEI Distress) was less robust (r = .660) than wa s the correlation between their e mpathy counterparts, as were the negative correlations between these distress measures and the various PER measures to which they contributed (correlations ranging from .490 to .663). No significant relationship was fou nd between the BEDI Empathy and BEDI Distress measures. Significant correlations were found, however, between the PEI empathy measure (CC) and the PEI Distress subscales (PI and FS). The subscale designed to measure the more engaging aspects of response (PEI PI) was positively correlated with empathy (PEI CC) at r = .314. While the subscale designed to assess the more distancing aspects of such a response ( PEI FS) was negatively correlated with that same empathy measure (r = .334) Both correlations were significant at the p was found between the false superiority subscale (PEI FS) of PEI Distress and the BEDI Empathy measure (r = 12 and 4 13 display th e results of Pearson correlations for the ten dependent variable scales, subscales, and composites. These results are presented in two 5 variable by 10 variable matrices. Relationship Communication History S urvey (RCHS) response analyses. Participants res ponded to the Relationship Communication History Survey (RCHS) after completion of the dependent variable data gathering (Task #6). Although the differing group conditions may have influenced participant responses, areas of interest for the RCHS data were confined to gender differences in perceived frequency of distress (both self and partner ), perceived competency (both self and partner ) at being a source of comfort during such times of distress, and the relationships between those assessments
150 and (bo th self and partner ) dependent variable responses in the earlier imagined recurrence of an episode of partner distress. RCHS descriptive statistic s. Table 4 14 at the end of this chapter displays the descriptive statistics for gender comparisons in re gard to the four RCHS questions. In Question #1, husbands rated their own frequen cy of dis tress (M = 3.0874) lower than wives rated their own frequency of distress (M = 4.1262). This finding mirrored distress (M = 2.9320) were distress (M = 4.2816). Independent samples t tests were performed to test the equality of the means. Gender differences were found in regard frequently their spouses experience distress accompanied by a desire for their support. differences were found between the genders in their own effectiveness at providing comfort and support when their spouses are distressed. RCHS correlations. Pearson correlations were calculated for each of the four RCHS questions. Specifically, the response mean for each of the questions was
151 corre lated with the response mean for each of the other questions. Correlations were calculated for all participants (N=206), as well as for husbands (N=103) and wives (N=103) separately. The results of these correlations are presented in Tables 4 15 thru 4 1 7. Responses to each of the four RCHS questions were then correlated with 14 variables of interest from the primary study: the two demographic variables, the two intervening variables, and the ten dependent variable measures. The results of these correla tions can be found in Tables 4 18 and 4 19. Subsequently, two matched pair data sets (N=103) were constructed from the and vice versa. Correlations were then tendencies and their own perceptions of communication quality. Expanding the analyses to cross correlation of participant responses with spou sal data provided a results of these cross correlations can be located in Tables 4 20 thru 4 23. It should be noted that because of its nature as a fixed factor, no da ta was generated for gender in cross correlations. The following sections provide a description of the findings for both the participant RCHS correlations and the husbands/wives cross correlations. RCHS correlations: all responses with all responses. The strongest
152 as comfort giver) and Q#4 (own effectiveness as comfort Milder and negative stress frequen cy) and Q#4 (r = (r = 15 displays the full results of RCHS between response correlations. Simila he strongest positive correlation was between the Q#2 and Q#4 assessments of self and mate comforting competence (r = .46 correlations were found between Q#2 and Q#3 (r = .205; response correlations. correlation of Q#2 and Q #4 responses were in the same direction and at the same level sponses, a negative correlation was found between Q#3 and Q#4 (r = response correlations. RCHS correlations: gender. RCHS Q#1 was positively correlated with gender Because the values of male=1 and female=2 were assigned to the fixed factor of gender, this correlation indicates that wives reported higher incidence of distress and desire for spousal support than did husbands. Pred ictably, then, Q#3 was
153 negatively co rrelated with gender (r = incidence than did wives of their spouses experiencing distress and wanting support. No correlations between gender and RCHS responses were generated for the matched pair data sets because the nature of gender as a fixed factor precluded such analyses. Table 4 18 displays the full results of RCHS responses correlated with the demographic variable of gender. RCHS correlations: years of current marriage. Participants who had been ma rried to each other longer were more likely to rate their mates as more effective in providing support during their times of distress (Q#2). The correlation was significant rs as follows: The correlation f marriage was r = .212 (p between Frequency of distress with attempts to elicit c omfort was found to be negatively lengt h of marriage: The correlation length of current marriage was r = 01). Similarly the correlation between f their own distress and support eliciting frequency and s 4 18 4 20, and 4 22 display the full results of RCHS responses correlated with the demographic variable of years of current mar riage. RCHS correlations: intervening variables: all responses with own data. Attac hment avoidance was negatively
154 comforting; r = .505) and with Q#4 (effectiveness at comforting mate; r = .424). A positive c orrelation frequency; r = .305). Attachment anxiety was negatively correlated with Q#2 (r = .410) and with Q#4 (r = .269). A positive correlation was found between attachment anxiety and Q #1 (own frequency of distress; r = .314). All correlations were significant at the 18 displays the full results of RCHS responses correlated with the intervening variable of attachment style. RCHS correlations: intervening variabl data. tachment avoidance was negatively givers; r = .376) of their own effectiveness as comfort givers; r = as negatively correlated with ss as comfort givers; r = ir perceptions of their own effectiveness as comfort givers; r = .363). All of these correlations were significant at nt anxiety was positively correlated of significance (r = .210). Table 4 responses cross data. tachment avoidance was negatively givers;
155 r = fectivene ss as comfort givers; r = correlated attachment anxiety was negatively .347) and givers; r = .467). Table 4 22 display intervening variable data. RCHS correlations: dependent variables: all responses with own data. distress frequency (Q#1) was positively correlated with the following depen dent variable scales, subscales, and composites: BEDI Distress (r = .199); PEI PI (r = .210); PEI FS (r = .242); and PEI Distress (r = .257). All correlations were significant at giver (Q#2) was positively correlated with BEDI Empathy (r = .348); BEDI PER (r = .312); PEI CC (r = .359); PEI PER (r = .274); and Total PER (r = .305). Each of these c orrelations was responses were positively correlated with PEI PI stress) or Q#4 giver). Tables 4 18 and 4 19 display the full results of RCHS responses correlated with the dependent variable measures.
156 data. stress frequency) were negatively tiveness) were positively correlated PEI CC (r Distress (r = own ef fectiveness as comf ort givers) were positively correlated 20 and 4 21 display dependent variable data. data. stress frequency) were negatively correl .268; .213; com
157 responses were negatively correlated responses (wives givers). Table s 4 22 and 4 correlated with Experimental data a nalyses Experimental data analyses performed on the utilized sample (N=234) included the following: (a) calculation of descriptive statistics for each group for each of 15 measures (two intervening variable measures; three measures (scales or composites) of quality of engagement in targeted memories; and ten dependent variable measures (including subscales and composites); (b) performance of one way analyses of variance (ANOVA) to determine level of significance of observed group differences; and (c) conducting of post hoc tests to explore group diffe rences revealed by ANOVA at Descriptive st atistics Tables 4 24 thru 4 26 at the end of this chapter display the descriptive statistics for the four groups for each of the 15 variables calculated. Measures of central tendency observed among these s tat istics fall within the range reported by the literature and anticipated in this study. Analyses of v ariance One way analyses of variance were performed to examine the significance of group differences observed for each of the descriptive statistics. Dif avoidance (df = 3 230
158 group specific prototype reflection vividness (df = 3 230 PEI false superiority distress scale measure (df = 3 230 No other between group differences were found. Tables 4 27 and 4 28 at the end of this chapter display the results of the one way analyses of variance for the 15 outcome me asures. Post hoc analyses of group differences tests for multiple comparisons were performed for the three measures displaying significant group way analyses of variance. Significant differences were confirmed for prototype reflection vividness and PEI false superiority. A discussion of possible explanations for these group differences is included in Chapter 5 The ANOVA finding of significant differences for attachment avoidance was not confirmed when account ing for multiple group comparisons. Table 4 tests for multiple comparisons for those measures showing significant group differences in the ANOVA. Research Hypotheses and Questions This study was designed to address the following four global research questions: (1) Does the priming of attachment security or insecurity differentially affect the experience of empathy and distress in response to one s distressed marriage partner? (2) Does gender differentially affect the experience of empathy and distress in response to one s distressed marriage partner? (3) How does attachment style interact with contextual attachment security or insecurity priming to affect the experience of empathy and personal distress in response to one s distressed marriage partner? And (4) Does attachment avoidance promote suppression of partner in distress memory vividness as an emotion regulation coping strategy?
159 The four global research questions were focused on the effects of the independent variable of group condition (Hypothesis 1, three Questions), the demographic variable of gender (Hypothesis 2, two Questions), the intervening variable of attachment style (Hypothesis 3, three Questions), and defensive processes that might influence level of engagement with stimuli intended to prompt dependent variable empathy and distress responses (Hypothesis 4, three Questions). One of the questions under Hypothesis 4 (Question 11) examined gender as a possible level of engagement correlate. The purpos e of this section is to examine the eleven research questions in the light of the data presented in the previous section. Pearson correlation coefficients were calculate d to investigate Questions 4 thru 8. And partial correlations were performed to investigate Questions 9 thru 11. An alpha level of .05 was selected to test the significance of the observed comparisons and correlations. Question 1 (H1): Attachment secur ity priming condition will result in higher empathy positive scores (empathy score minus distress score) than will attachment avoidance priming condition, attachment anxiety priming condition, or control condition The Null Hypothesis stated that there wou ld be no difference between the groups in regard to the dependent variable composite of predominant emotional response distress score from his or her empathy score. A higher e mpathy positive score is a PER score representing a larger positive number or a smaller negative number than the scores to which it is compared. One way analyses of variance (ANOVA) were esponses,
160 including all scales and subscales. The results of these analyses can be located in Tables 4 27 thru 4 29. No significant differences were observed for any of the empathy or distress scales or subscales, with the exception of the PEI Distress f alse s uperiority subscale PER dependent variable measures (BEDI PER, PEI PER, or Total PER). Thus, the Null Hypothesis was retained. Question 2 (H1): Attachment avoid ance priming condition will result in lower empathy scores than will attachment anxiety priming condition The Null Hypothesis stated that there would be no difference between the groups in regard to the dependent variable empathy. One way analyses of var iance (ANOVA) of these analyses can be located in Tables 4 27 and 4 28. No significant group differences were observed for either of the two empathy measures (BEDI Empathy or PEI Empathy). Thus, the Null Hypothesis was retained. Question 3 (H1): Attachment anxiety priming condition will result in higher distress scores than will attachment avoidance priming condition The Null Hypothesis stated that there would be no diffe rence between the groups in regard to the dependent variable distress. One way analyses of variance (ANOVA) of these analyses can be located in Tables 4 27 and 4 28. The only significant differenc e observed was for the PEI D istress false s uperiority subscale these differences were not observed between the attachment anxiety priming condition
161 (Group 3) and the attachment avoidance priming condition (Group 2). Thus, the Null Hypothesis was retained. Question 4 (H2 ): Females will exhibit higher empathy scores than will males in response to marriage partner distress The Null Hypothesis stated that there would be no difference between females and males in reg ard to the dependent variable empathy. Pearson correlations of the fixed factor gender with the two empathy measures revealed no significant relationship between gender and empathy responses to marriage partner distress (r = .00 BEDI Empathy; r = .04 PEI CC E mpathy). The results of these analyses can be located in Table 4 9. Thus, the Null Hypothesis was retained. Question 5 (H2): Males will exhibit greater use of distancing strategies than will females in response to marriage partner distress The Null Hypothesis stated that there would be no difference between males and females in regard to the dependent variable distancing score. Pearson correlation of the fixed factor gender with the distancing subscale of the PEI Distress false s uperiority subscale revealed no significant relationship between gender and the use of distancing strategies in response to marriage partner distress (r = .05). The results of these analyses can be located in Table 4 9. Thus, the Null Hypothesis was retained. Question 6 ( H3 ): Higher attachment avoidance scores will predict lower empathy positive scores (empathy score minus distress score) for all three experimental conditions and control condition in response to marriage partner distress The Null Hypothesis stated that t here would be no relationship between the intervening variable of attachment avoidance and the dependent variable composite of
162 predominant emotional response (PER). Pearson correlations of attachment avoidance scores (Av) with the three PER scores reveale d a significant negative relationship between attachment avoidance and predominant emotional response to the episode of marriage partner distress. The relationship between the variables was in the direction predicted. The Av with BEDI PER correlation was r = .334; the Av with PEI PER correlatio n was r = .327; and the Av with Total PER correlation was r = .349. All located in Table 4 9. Thus, the Null Hypothesis was rejected. Question 7 (H3 ): Higher attachment anxiety scores will predict lower empathy positive scores (empathy score minus distress score) for all three experimental conditions and control condition in response to marriage partner distress The Null Hypothesis stated that there would be n o relationship between the intervening variable of attachment anxiety and the dependent variable composite of predominant emotional response (PER). Pearson correlations of attachment anxiety scores (Ax) with the three PER scores revealed a signi ficant neg ative relationship between attachment anxiety and predominant emotional response to the episode of marriage partner distress. The relationship between the variables was in the direction predicted. The Ax with BEDI PER correlation was r = .255; the Av wi th PEI PER correlation was r = .212; and the Av with Total PER correlation was r = .243. All located in Table 4 9. Thus, the Null Hypothesis was rejected.
163 Questio n 8 (H3 ): Higher attachment avoidance scores will predict greater use of distancing strategies in response to marriage partner distress when controlling for experimental condition and gender The Null Hypothesis stated that there would be no relationship between the strategies in response to marriage partner distress when controlling for experimental condition and gender. Pearson correlation of attachment avoidance scores (Av ) with the distancing subscale of the PEI Distress false s uperiority subscale revealed a significant positive correlation tendency to employ distancing strategies in response to the episode of marria ge partner distress (see Table 4 9). The relationship between the two variables when controlling for group and gender (r = .238) was in the direction predicted and was significant at the Question 9 ( H4 ): Higher attachment avoidance scores will predict lower partner in distress memory vividness ratings when controlling for experimental condition and gender The Null Hypothesis stated that there would be no relationship between the intervening variable of attachment avoidance and the (5 score composite mean) vividness rating of the partner in distress episode. The zero order correlation between these two variables is displayed in Table 4 8. Partial correlations were calculated tachment avoidance scores (Av) and partner in distress vividness means, controlling for group and gender. A significant negative correlation
164 the hypothesis (r = .137) Thus, the Null Hypothesis was rejected. Question 10 (H4 ): Attachment avoidance priming condition will result in lower partner in distress memory vividness ratings when controlling for attachment style and gender The Null Hypothesis stated that there w ould be no difference between the groups in regard to the (5 score composite mean) vividness rating of the partner in distress episode. Table 4 in distress vividness ratings. The means ranged from a l ow of 5.14 (control condition) to a high of 5.42 (attachment avoidance priming condition). There were no significant differences found between the groups regarding the partner in distress vividness ratings when conducting partial correlations controlling for attachment avoidance, attachment anxiety, and gender. Thus, the Null Hypothesis was retained. Question 11 (H4 ): Males will exhibit lower partner in distress memory vividness ratings when controlling for experimental condition and attachment style The Null Hypothesis stated that there would be no difference between males and females in regard to the (5 score composite mean) vividness rating of the partner in distress episode. The zero order correlation between these two variables is displayed in T able 4 partner in distress vividness means, controlling for group, attachment avoidance, and attachment anxiety. Although in the direction predicted, the relationship between the var retained.
165 Summary In Chapter 4 I have provided a description of the sample, an examination of reliability findings for adapted instrumentation, a reporting of data analyses pro cedures and results, and the conclusions of outcomes testing of the research hypotheses. The null hypotheses were retained for seven of the research questions posed in the study design. The null was rejected in regard to the other four research questions Additional findings from the primary and secondary analyses of the collected data were reported. The relevance of these findings to the existing body of marital research, along with the imp lications for future research is discussed in Chapter 5
166 Tabl e 4 6 Overview of Participant Data Collection Categories and their Use s Task # Data Label Subscale # of Items Description 1 ECR R AV Full (36) Avoidance (Av) 18 Intervening Variable Anxiety (Ax) 18 Intervening Variab le 2 Distracter Discarded (per design) 3 Narrative PID Memory Selection 4 Narrative Prototype Memory Selection 4 PT Mem Viv 1 Group Specific Prompt 5 Mem Viv PID Mem Viv 1 5 Item Mean (1 of 5) 5 Mem Dis Lev 1 Partner Distress Level 5 Imag Viv PID Mem Viv 1 5 Item Mean (1 of 5) 6 Imag Viv PID Mem Viv 3 5 Item Mean (3 of 5) 6 BEDI AV Full (14) Empathy 6 Dependent Variable Distress 8 Dependent Variable 6 PEI SF R Full (30) Empathy (CC) 13 Dependent Variable Distress (PI) 1 0 Dependent Variable Distress (FS) 7 Dependent Variable Distancing (3) PEI FS Subscale Total Items 87 All But RCHS Analyses (N=234) 7 RCH Survey (page 1) 4 Communication Quality 7 RCH Survey (page 2) Discarded (flawed data) Total Items 4 RCHS Analyses (N=206) Note: ECR R AV = Experiences in Close Relationships Questionnaire Revised Adapted Version; PID = Partner In Distress; PT Mem Viv = Prototype Memory Vividness rating; PID Mem Viv = Vividness of Memory of partner in distr ess episode; Mem Dis Lev = Level of partner distress in episode remembered; Imag Viv = Vividness of Imagined recurrence of partner in distress episode; BEDI and Distress Indices Adapted Version; PEI SF R = Pity Experience Inventories Short Form Revised; CC = Compassionate Caring subscale; PI = Passive Identification subscale; FS = False Superiority subscale; RCH Survey (RCHS) = Relationship Communication History Survey
167 Table 4 7 Pearson Correlations (1 of 7) [Demographic and Interv ening Variables with Each Other] Gender Years Married Av Ax Gender (M/F) Pearson Correlation 1 0.029 0.075 0.07 Sig. (2 tailed) 0.656 0.256 0.287 N 234 234 234 234 Years Married Pearson Correlation 0.029 1 0.102 .237 ** Sig. (2 tailed) 0.656 0.121 0 N 234 234 234 234 Attachment Avoidance Score (Av) Pearson Correlation 0.075 0.102 1 .528 ** Sig. (2 tailed) 0.256 0.121 0 N 234 234 234 234 Attachment Anxiety Score (Ax) Pearson Correlation 0.07 .237 ** .528 ** 1 Sig. (2 tailed) 0.287 0 0 N 234 234 234 234 Note: ** Cor relation is significant at the .01 level (2 tailed)
168 Table 4 8 Pearson Correlations (2 of 7) [Demographic and Intervening with Memory Engagement Variables] Group Prototype Vividness Partner Distress Level PID Mean Vividness Level Gender (M/F) Pearson Correlation .240 ** 0.122 0.121 Sig. (2 tailed) 0 0.062 0.064 N 234 234 234 Years Married Pearson Correlation 0.028 0.067 0.089 Sig. (2 tailed) 0.668 0.304 0.173 N 234 234 234 Attachment Avoidanc e Score (Av) Pearson Correlation 0.071 0.01 .129 Sig. (2 tailed) 0.277 0.876 0.049 N 234 234 234 Attachment Anxiety Score (Ax) Pearson Correlation 0.069 0.027 0.039 Sig. (2 tailed) 0.294 0.686 0.554 N 234 234 234 Note: ** Correlation is signi ficant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 tailed); PID = Partner In Distress
169 Table 4 9 Pearson Correlations (3 of 7) [Dependent Variables with Demographic and Intervening Variables] Gender Years Marrie d Av Ax BEDI Empathy Pearson Correlation 0 0.045 .373 ** .186 ** Sig. (2 tailed) 0.995 0.495 0 0.004 N 234 234 234 234 BEDI Distress Pearson Correlation .144 0.064 0.087 .185 ** Sig. (2 tailed) 0.027 0.332 0.186 0.004 N 234 234 234 234 BEDI PER Pearson Correlation 0.09 0.073 .334 ** .255 ** Sig. (2 tailed) 0.17 0.263 0 0 N 234 234 234 234 PEI CC Pearson Correlation 0.039 0.02 .380 ** 0.118 Sig. (2 tailed) 0.557 0.76 0 0.071 N 234 234 234 234 PEI PI Pearson Correlation 0.09 0 .132 0.057 Sig. (2 tailed) 0.168 0.998 0.043 0.383 N 234 234 234 234 PEI FS Pearson Correlation 0.101 0.055 .167 .231 ** Sig. (2 tailed) 0.124 0.404 0.01 0 N 234 234 234 234 Distancing Score Pearson Correlation 0.046 0.031 .223 ** 0.116 Sig. (2 ta iled) 0.485 0.64 0.001 0.077 N 234 234 234 234 PEI Distress Pearson Correlation 0.11 0.027 0.008 .150 Sig. (2 tailed) 0.095 0.686 0.902 0.022 N 234 234 234 234 PEI PER Pearson Correlation 0.113 0.002 .327 ** .212 ** Sig. (2 tailed) 0.085 0.9 8 0 0.001 N 234 234 234 234 Total PER Pearson Correlation 0.11 0.033 .349 ** .243 ** Sig. (2 tailed) 0.094 0.618 0 0 N 234 234 234 234 Note: ** Correlation is significant at the .01 level (2 tailed); Correlation is significant at the .05 level (2 Predominant Emotional Response; PEI = Pity Experience Inventories; CC = Compassionate Caring subscale; PI = Passive Identification subscale; FS = False Superiority subscale
170 Table 4 10 Pearson Correlations (4 of 7) [Memory Engagement Variables with Each Other] Group Prototype Vividness Partner Distress Level PID Mean Vividness Level Group Prototype Vividness Pearson Correlation 1 0.08 .538 ** Sig. (2 tailed) 0.225 0 N 234 234 234 Partner Distress Level Pearson Correlation 0.08 1 .457 ** Sig. (2 tailed) 0.225 0 N 234 234 234 PID Mean Vividness Level Pearson Correlation .538 ** .457 ** 1 Sig. (2 tailed) 0 0 N 234 234 234 Note: Correl ation is significant a t the .01 level (2 tailed) ; PID = Partner In Distress
171 Table 4 11 Pearson Correlations (5 of 7) [Dependent Variables with Memory Engagement Variables] Group Prototype Vividness Partner Distress Level PID Mean Vividness Level BEDI Empathy Pearso n Correlation .154 0.121 .322 ** Sig. (2 tailed) 0.018 0.065 0 N 234 234 234 BEDI Distress Pearson Correlation 0.049 .387 ** .309 ** Sig. (2 tailed) 0.452 0 0 N 234 234 234 BEDI PER Pearson Correlation 0.085 .151 0.048 Sig. (2 tailed) 0.197 0. 021 0.465 N 234 234 234 PEI CC Pearson Correlation 0.114 .199 ** .347 ** Sig. (2 tailed) 0.083 0.002 0 N 234 234 234 PEI PI Pearson Correlation 0.027 .384 ** .309 ** Sig. (2 tailed) 0.683 0 0 N 234 234 234 PEI FS Pearson Correlation 0.03 .291 ** 0 .124 Sig. (2 tailed) 0.647 0 0.059 N 234 234 234 Distancing Score Pearson Correlation 0.031 .230 ** 0.042 Sig. (2 tailed) 0.639 0 0.518 N 234 234 234 PEI Distress Pearson Correlation 0.003 .399 ** .268 ** Sig. (2 tailed) 0.958 0 0 N 234 234 234 PEI PER Pearson Correlation 0.097 0.114 0.11 Sig. (2 tailed) 0.139 0.082 0.092 N 234 234 234 Total PER Pearson Correlation 0.097 .137 0.09 Sig. (2 tailed) 0.137 0.037 0.17 N 234 234 234 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 Predominant Emotional Response; PEI = Pity Experience Inventories; CC = Compassionate Caring subscale; PI = Passive Identification su bscale; FS = False Superiority subscale
172 Table 4 12 Pearson Correlations (6 of 7) [Dependent Variables with Each Other (First Half)] BEDI Empathy BEDI Distress BEDI PER PEI CC PEI PI BEDI Empathy Pearson Correlation 1 0.05 .781** .822** .314** Sig. (2 tailed) 0.45 0 0 0 N 234 234 234 234 234 BEDI Distress Pearson Correlation 0.05 1 .663** 0.062 .613** Sig. (2 tailed) 0.45 0 0.345 0 N 234 234 234 234 234 BEDI PER Pearson Correlation .781** .663** 1 .578** .148* Sig. (2 tailed) 0 0 0 0.023 N 234 234 234 234 234 PEI CC Pearson Correlation .822** 0.062 .578** 1 .445** Sig. (2 tailed) 0 0.345 0 0 N 234 234 234 234 234 PEI PI Pearson Correlation .314** .613** .148* .445** 1 Sig. (2 tailed) 0 0 0.023 0 N 234 234 234 234 234 PEI FS Pearson Correlation .334** .512** .571** .146* .485** Sig. (2 tailed) 0 0 0 0.026 0 N 234 234 234 234 234 Distance Score Pearson Correlation .489** .396** .614** .362** .283** Sig. (2 tailed) 0 0 0 0 0 N 234 234 234 234 234 PEI Distress Pearson Correlation 0.05 .660** .375** .228** .906** Sig. (2 tailed) 0.45 0 0 0 0 N 234 234 234 234 234 PEI PER Pearson Correlation .684** .422** .777** .711** .265** Sig. (2 tailed) 0 0 0 0 0 N 234 234 234 234 234 Total PER Pearso n Correlation .766** .551** .919** .695** .230** Sig. (2 tailed) 0 0 0 0 0 N 234 234 234 234 234 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 Distress Indices; PER = Predominant Emotional Response; PEI = Pity Experience Inventories; CC = Compassionate Caring subscale; PI = Passive Identification subscale; FS = False Superiority subscale
173 Table 4 13 Pearson Correlations (7 of 7) [Dependent Variables with Each Other (Second Half)] PEI FS Distance Score PEI Distress PEI PER Total PER BEDI Empathy Pearson Correlation .334** .489** 0.05 .684** .766** Sig. (2 tailed) 0 0 0.45 0 0 N 234 234 234 234 234 BEDI Distress Pearson Correlation .512** .396** .660** .422** .551** Sig. (2 tailed) 0 0 0 0 0 N 234 234 234 234 234 BEDI PER Pearson Correlation .571** .614** .375** .777** .919** Sig. (2 tailed) 0 0 0 0 0 N 234 234 234 234 234 PEI CC Pearson Correlation .146* .362** .2 28** .711** .695** Sig. (2 tailed) 0.026 0 0 0 0 N 234 234 234 234 234 PEI PI Pearson Correlation .485** .283** .906** .265** .230** Sig. (2 tailed) 0 0 0 0 0 N 234 234 234 234 234 PEI FS Pearson Correlation 1 .825** .809** .712** .693** Si g. (2 tailed) 0 0 0 0 N 234 234 234 234 234 Distance Score Pearson Correlation .825** 1 .589** .743** .731** Sig. (2 tailed) 0 0 0 0 N 234 234 234 234 234 PEI Distress Pearson Correlation .809** .589** 1 .522** .490** Sig. (2 tailed) 0 0 0 0 N 234 234 234 234 234 PEI PER Pearson Correlation .712** .743** .522** 1 .962** Sig. (2 tailed) 0 0 0 0 N 234 234 234 234 234 Total PER Pearson Correlation .693** .731** .490** .962** 1 Sig. (2 tailed) 0 0 0 0 N 234 234 234 234 234 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 Predominant Emotional Response; PEI = Pity Experience Inventories; CC = Compas sionate Caring subscale; PI = Passive Identification subscale; FS = False Superiority subscale
174 Table 4 14 Gender Comparisons for RCH Survey Questions Gender N Mean Standard Deviation Q#1 Your Distress Frequency Male 103 3.09 1.18 Fema le 103 4.13 1.31 Q#2 Mate's Comforting Effectiveness Male 103 4.51 1.53 Female 103 4.32 1.47 Q#3 Mate's Distress Frequency Male 103 4.28 1.35 Female 103 2.93 1.29 Q#4 Your Comforting Effectiveness Male 103 4.00 1.31 Female 103 4.15 1.30 Tab le 4 15 RCH Survey Between Response Correlations: All Participants Q #1 Q #2 Q #3 Q #4 Q #1 Your Own Distress Frequency Pearson Correlation 1 0.051 .154 0.078 Sig. (2 tailed) 0.466 0.027 0.267 N 206 206 206 206 Q #2 Mate's Comforting Effec tiveness Pearson Correlation 0.051 1 0.093 .457 ** Sig. (2 tailed) 0.466 0.185 0 N 206 206 206 206 Q #3 Mate's Distress Frequency Pearson Correlation .154 0.093 1 .213 ** Sig. (2 tailed) 0.027 0.185 0.002 N 206 206 206 206 Q #4 Your Own Com forting Effectiveness Pearson Correlation 0.078 .457 ** .213 ** 1 Sig. (2 tailed) 0.267 0 0.002 N 206 206 206 206 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 tailed)
175 Table 4 16 RCH Survey Between Response Correlations: Husbands Only Q #1 Q #2 Q #3 Q #4 Q #1 Your Own Distress Frequency Pearson Correlation 1 0.067 0.102 0.139 Sig. (2 tailed) 0.501 0.307 0.161 N 103 103 103 103 Q #2 Mate's Comforting Effectiveness Pear son Correlation 0.067 1 .270 ** .462 ** Sig. (2 tailed) 0.501 0.006 0 N 103 103 103 103 Q #3 Mate's Distress Frequency Pearson Correlation 0.102 .270 ** 1 .205 Sig. (2 tailed) 0.307 0.006 0.038 N 103 103 103 103 Q #4 Your Own Comforting Effect iveness Pearson Correlation 0.139 .462 ** .205 1 Sig. (2 tailed) 0.161 0 0.038 N 103 103 103 103 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 tailed) Table 4 17 RCH Survey Bet ween Response Correlations: Wives Only Q #1 Q #2 Q #3 Q #4 Q #1 Your Own Distress Frequency Pearson Correlation 1 0.118 0.041 0.08 Sig. (2 tailed) 0.236 0.678 0.423 N 103 103 103 103 Q #2 Mate's Comforting Effectiveness Pearson Correlation 0 .118 1 0.006 .462 ** Sig. (2 tailed) 0.236 0.948 0 N 103 103 103 103 Q #3 Mate's Distress Frequency Pearson Correlation 0.041 0.006 1 .217 Sig. (2 tailed) 0.678 0.948 0.028 N 103 103 103 103 Q #4 Your Own Comforting Effectiveness Pearson Corr elation 0.08 .462 ** .217 1 Sig. (2 tailed) 0.423 0 0.028 N 103 103 103 103 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 tailed)
176 Table 4 18 Correlation of RCHS Responses with Variables of Interest (1 of 6) Part 1] Q #1 Q #2 Q #3 Q #4 Gender Pearson Correlation .386 ** 0.065 .458 ** 0.056 Sig. (2 tailed) 0 0.355 0 0.425 N 206 206 206 206 Years of Current Marriage Pear son Correlation 0.083 .207 ** 0.096 0.127 Sig. (2 tailed) 0.233 0.003 0.171 0.069 N 206 206 206 206 Attachment Avoidance Score (Av) Pearson Correlation 0.034 .505 ** .305 ** .424 ** Sig. (2 tailed) 0.629 0 0 0 N 206 206 206 206 Attachment Anxiet y Score (Ax) Pearson Correlation .314 ** .410 ** 0.076 .269 ** Sig. (2 tailed) 0 0 0.276 0 N 206 206 206 206 BEDI Empathy Pearson Correlation 0.046 .348 ** .224 ** .402 ** Sig. (2 tailed) 0.514 0 0.001 0 N 206 206 206 206 BEDI Distress Pearson Corre lation .199 ** 0.074 0.076 0.063 Sig. (2 tailed) 0.004 0.288 0.277 0.367 N 206 206 206 206 BEDI PER Pearson Correlation 0.094 .312 ** .219 ** .346 ** Sig. (2 tailed) 0.181 0 0.002 0 N 206 206 206 206 Note: ** Cor relation is significant at the 01 level (2 tailed) Empathy and Distress Indices; PER = Predominant Emotional Response
177 Table 4 19 Correlation of RCHS Responses with Variables of Interest (2 of 6) Part 2] Q #1 Q # 2 Q #3 Q #4 PEI CC Pearson Correlation 0.097 .359 ** 0.094 .415 ** Sig. (2 tailed) 0.165 0 0.179 0 N 206 206 206 206 PEI PI Pearson Correlation .210 ** .150 0.005 0.101 Sig. (2 tailed) 0.002 0.032 0.941 0.148 N 206 206 206 206 PEI FS Pearson Corr elation .242 ** 0.09 0.134 .194 ** Sig. (2 tailed) 0 0.197 0.054 0.005 N 206 206 206 206 Distancing Pearson Correlation 0.08 0.128 .153 .209 ** Sig. (2 tailed) 0.253 0.066 0.028 0.003 N 206 206 206 206 PEI Distress Pearson Correlation .257 ** 0. 057 0.068 0.025 Sig. (2 tailed) 0 0.412 0.335 0.725 N 206 206 206 206 PEI PER Pearson Correlation 0.1 .274 ** 0.131 .382 ** Sig. (2 tailed) 0.154 0 0.06 0 N 206 206 206 206 Total PER Pearson Correlation 0.103 .305 ** .175 .388 ** Sig. (2 tail ed) 0.142 0 0.012 0 N 206 206 206 206 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 tailed); PEI = Pity Experience Inventories; CC = Compassionate Caring subscale; PI =Passive Identifi cation subscale; FS = False Superiority subscale; PER = Predominant Emotional Response
178 Table 4 20 Correlation of RCHS Responses with Variables of Interest (3 of 6) Part 1] Q #1 Q #2 Q #3 Q #4 Gend er Pearson Correlation Sig. (2 tailed) N 103 103 103 103 Years of Current Marriage Pearson Correlation 0.157 .212 .371 ** 0.159 Sig. (2 tailed) 0.114 0.032 0 0.108 N 103 103 103 103 Attachment Avoidance Score (Av) Pearson Correla tion 0.039 .376 ** 0.077 .320 ** Sig. (2 tailed) 0.698 0 0.44 0.001 N 103 103 103 103 Attachment Anxiety Score (Ax) Pearson Correlation 0.173 .374 ** .210 .363 ** Sig. (2 tailed) 0.08 0 0.033 0 N 103 103 103 103 BEDI Empathy Pearson Correlation 0.088 .236 0.035 .208 Sig. (2 tailed) 0.376 0.016 0.728 0.035 N 103 103 103 103 BEDI Distress Pearson Correlation 0.137 .236 .213 0.165 Sig. (2 tailed) 0.168 0.017 0.031 0.095 N 103 103 103 103 BEDI PER Pearson Correlation 0.158 .341 ** 0.104 .274 ** Sig. (2 tailed) 0.112 0 0.295 0.005 N 103 103 103 103 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 tailed) Indices; PER = Predomi nant Emotional Response
179 Table 4 21 Correlation of RCHS Responses with Variables of Interest (4 of 6) Part 2] Q #1 Q #2 Q #3 Q #4 PEI CC Pearson Correlation .202 .195 0.109 .210 Sig. (2 tail ed) 0.041 0.048 0.274 0.033 N 103 103 103 103 PEI PI Pearson Correlation 0.132 0.08 .226 0.014 Sig. (2 tailed) 0.185 0.423 0.022 0.891 N 103 103 103 103 PEI FS Pearson Correlation 0.117 .239 .273 ** 0.075 Sig. (2 tailed) 0.238 0.015 0.005 0 .453 N 103 103 103 103 Distancing Pearson Correlation 0.106 0.174 0.098 0.119 Sig. (2 tailed) 0.286 0.079 0.323 0.231 N 103 103 103 103 PEI Distress Pearson Correlation 0.03 0.168 .282 ** 0.045 Sig. (2 tailed) 0.761 0.089 0.004 0.65 N 103 1 03 103 103 PEI PER Pearson Correlation 0.163 .297 ** 0.1 .223 Sig. (2 tailed) 0.101 0.002 0.316 0.023 N 103 103 103 103 Total PER Pearson Correlation 0.169 .330 ** 0.107 .256 ** Sig. (2 tailed) 0.088 0.001 0.283 0.009 N 103 103 103 103 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 tailed); PEI = Pity Experience Inventories; CC = Compassionate Caring subscale; PI =Passive Identification subscale; FS = False Superiority subscale ; PER = Predominant Emotional Response
180 Table 4 22 Correlation of RCHS Responses with Variables of Interest (5 of 6) Part 1] Q #1 Q #2 Q #3 Q #4 Gender Pearson Correlation Sig. (2 tailed) N 103 103 103 103 Years of Current Marriage Pearson Correlation .313 ** .202 0.168 0.095 Sig. (2 tailed) 0.001 0.04 0.089 0.342 N 103 103 103 103 Attachment Avoidance Score (Av) Pearson Correlation .288 ** .464 ** 0.07 .446 ** Sig. (2 taile d) 0.003 0 0.485 0 N 103 103 103 103 Attachment Anxiety Score (Ax) Pearson Correlation 0.112 .347 ** 0.011 .467 ** Sig. (2 tailed) 0.262 0 0.912 0 N 103 103 103 103 BEDI Empathy Pearson Correlation .268 ** .286 ** 0.055 0.103 Sig. (2 tailed) 0.006 0.003 0.579 0.299 N 103 103 103 103 BEDI Distress Pearson Correlation 0.037 0.063 0.04 0.073 Sig. (2 tailed) 0.712 0.525 0.686 0.465 N 103 103 103 103 BEDI PER Pearson Correlation .207 .237 0.011 0.119 Sig. (2 tailed) 0.036 0.016 0.912 0.2 32 N 103 103 103 103 Note: ** Cor relation is significant at the .01 level (2 tailed) ; Correlation is significant at the .05 level (2 tailed) Indices; PER = Predominant Emotional Response
181 Table 4 23 Correlat ion of RCHS Responses with Variables of Interest (6 of 6) Part 2] Q #1 Q #2 Q #3 Q #4 PEI CC Pearson Correlation .224 .236 0.032 0.115 Sig. (2 tailed) 0.023 0.017 0.748 0.246 N 103 103 103 103 PEI PI Pearson Correlation 0.048 0.097 0.11 0.172 Sig. (2 tailed) 0.633 0.329 0.268 0.083 N 103 103 103 103 PEI FS Pearson Correlation 0.144 0.177 0.057 0.071 Sig. (2 tailed) 0.146 0.074 0.567 0.474 N 103 103 103 103 Distancing Pearson Correla tion 0.183 .266 ** 0.088 0.162 Sig. (2 tailed) 0.065 0.007 0.375 0.102 N 103 103 103 103 PEI Distress Pearson Correlation 0.035 0.016 0.102 0.085 Sig. (2 tailed) 0.726 0.871 0.304 0.396 N 103 103 103 103 PEI PER Pearson Correlation .213 .209 0.049 0.034 Sig. (2 tailed) 0.031 0.034 0.626 0.732 N 103 103 103 103 Total PER Pearson Correlation .223 .233 0.025 0.073 Sig. (2 tailed) 0.024 0.018 0.799 0.464 N 103 103 103 103 Note: ** Cor relation is significant at the .01 level (2 tai led) ; Correlation is significant at the .05 level (2 tailed); PEI = Pity Experience Inventories; CC = Compassionate Caring subscale; PI =Passive Identification subscale; FS = False Superiority subscale; PER = Predominant Emotional Response
182 Table 4 24 Descriptive Statistics for Attachment Style and Memory Measures Data Label Group N Mean Std Dev Min Max Attachment Avoidance Score (Av) 1 61 2.89 1.11 1.33 6.06 2 60 2.83 1.03 1.17 5.67 3 55 2.39 0.97 1.06 5.61 4 58 2.63 1.04 1.00 5.28 Total 234 2.69 1.05 1.00 6.06 Attachment Anxiety score (Ax) 1 61 2.73 0.98 1.17 6.22 2 60 2.85 1.06 1.56 5.22 3 55 2.65 0.99 1.06 5.28 4 58 2.55 0.83 1.22 5.06 Total 234 2.70 0.97 1.06 6.22 Group Specific Prototype Memory Vividness 1 61 5.90 1.12 2.00 7. 00 2 60 5.75 1.07 3.00 7.00 3 55 5.91 0.97 2.00 7.00 4 58 5.33 1.10 3.00 7.00 Total 234 5.72 1.09 2.00 7.00 Partner in Distress Episode Severity Level 1 61 5.69 1.10 3.00 7.00 2 60 5.87 1.16 2.00 7.00 3 55 5.80 1.18 3.00 7.00 4 58 5.74 1.21 2.00 7.00 Total 234 5.77 1.16 2.00 7.00 Partner in Distress Vividness Scales (5 Score Composite) 1 61 5.29 0.84 3.20 7.00 2 60 5.42 0.90 2.60 7.00 3 55 5.32 1.01 3.00 7.00 4 58 5.14 0.90 3.20 7.00 Total 234 5.29 0.91 2.60 7.00
183 Table 4 2 5 Descriptive Statistics for Dependent Variable Measures (Part 1) Data Label Group N Mean Std Dev Min Max BEDI Empathy 1 61 4.66 1.65 1.00 7.00 2 60 4.84 1.81 1.00 7.00 3 55 4.83 1.54 1.00 7.00 4 58 4.64 1.72 1.00 7.00 Total 234 4.74 1.68 1.00 7 .00 BEDI Distress 1 61 3.59 1.29 1.00 6.13 2 60 3.59 1.44 1.00 6.13 3 55 3.33 1.40 1.00 6.38 4 58 3.91 1.45 1.13 6.38 Total 234 3.61 1.40 1.00 6.38 BEDI PER (BEDI Empathy minus BEDI Distress) 1 61 1.07 1.98 4.50 6.00 2 60 1.25 2.21 4.33 4.96 3 55 1.51 2.28 5.00 6.00 4 58 0.73 2.45 4.83 4.63 Total 234 1.14 2.24 5.00 6.00 PEI Compassionate Caring (CC) 1 61 4.95 1.28 1.46 6.85 2 60 4.96 1.42 1.62 7.00 3 55 5.06 1.18 1.23 6.69 4 58 5.00 1.20 1.54 7.00 Total 234 4.99 1.27 1. 23 7.00 PEI Passive Identification (PI) 1 61 4.19 1.22 1.30 6.60 2 60 4.21 1.15 1.20 6.40 3 55 3.97 1.23 1.50 6.60 4 58 4.27 1.21 1.70 6.80 Total 234 4.16 1.20 1.20 6.80 Note Emoti onal Response; PEI = Pity Experience Inventories
184 Table 4 26 Descriptive Statistics for Dependent Variable Measures (Part 2 ) Data Label Group N Mean Std Dev Min Max PEI False Superiority (FS) 1 61 2.87 1.07 1.00 5.29 2 60 2.71 1.13 1.00 5.71 3 55 2 .55 1.23 1.00 5.86 4 58 3.17 1.42 1.29 6.86 Total 234 2.83 1.23 1.00 6.86 Distancing Score (3 item FS subset) 1 61 2.55 1.50 1.00 6.67 2 60 2.41 1.53 1.00 6.67 3 55 2.33 1.50 1.00 7.00 4 58 2.88 1.78 1.00 7.00 Total 234 2.54 1.59 1.00 7. 00 PEI Distress (PI + FS) 1 61 3.64 1.01 1.41 5.88 2 60 3.59 0.93 1.29 5.65 3 55 3.38 1.08 1.41 5.82 4 58 3.82 1.15 1.59 6.41 Total 234 3.61 1.05 1.29 6.41 PEI PER (PEI CC minus PEI Distress) 1 61 1.31 1.21 1.24 3.63 2 60 1.37 1.6 0 2.29 4.68 3 55 1.68 1.49 2.59 4.66 4 58 1.18 1.48 3.26 3.88 Total 234 1.38 1.45 3.26 4.68 Total PER (scaled score merging BEDI PER and PEI PER) 1 61 51.65 9.64 27.35 67.81 2 60 52.33 12.38 25.77 75.97 3 55 54.47 11.83 24.34 76.97 4 58 50 .26 12.23 15.61 71.71 Total 234 52.14 11.58 15.61 76.97 Note: PEI = Pity Experience Inventories; FS = False Superiority; PI = Passive Identification; PER = Predominant Emotional Response; CC = Compassionate Indices
185 Table 4 27 One Way Analyses of Variance (Part 1) Sum of Squares D f Mean Square F Av Between Groups 8.82 3 2.94 2.711 0.046* Within Groups 249.44 230 1.085 Total 258.26 233 Ax Between Groups 2.947 3 0.982 1.045 0.374 Within Groups 216.311 230 0.94 Total 219.258 233 PT Mem Viv Between Groups 12.963 3 4.321 3.794 0. 011* Within Groups 261.981 230 1.139 Total 274.944 233 Mem Dis Lev Between Groups 1.06 3 0.353 0.262 0.853 Within Groups 309.936 230 1.348 Total 310.996 233 PID Mem Viv Between Groups 2.418 3 0.806 0.969 0.408 Within Groups 191.349 23 0 0.832 Total 193.766 233 BEDI Empathy Between Groups 2.046 3 0.682 0.24 0.868 Within Groups 653.22 230 2.84 Total 655.265 233 BEDI Distress Between Groups 9.712 3 3.237 1.667 0.175 Within Groups 446.637 230 1.942 Total 456.349 233 BEDI PER Between Groups 17.965 3 5.988 1.2 0.311 Within Groups 1,147.91 230 4.991 Total 1,165.87 233 Note: Correlation is significant at the .05 level (2 tailed); Av = Attachment Avoidance score; Ax = Attachment Anxiety score; PT Mem Vi v = Prototype Memory Vividness rating; Mem Dis Lev = Level of partner distress in episode remembered; PID Mem Viv = Vividness of Memory of partner in Empathy and Distress Indices; PER = Predominant Emotional Response
186 Ta ble 4 28 One Way Analyses of Variance (Part 2) Sum of Squares D f Mean Square F PEI CC Between Groups 0.45 3 0.15 0.092 0.965 Within Groups 375.971 230 1.635 Total 376.421 233 PEI PI Between Groups 2.916 3 0.972 0.675 0.568 Within Groups 331.396 230 1.441 Total 334.311 233 PEI FS Between Groups 12.125 3 4.042 2.729 0.045* Within Groups 340.615 230 1.481 Total 352.74 233 Distancing (FS subset) Between Groups 10.258 3 3.419 1.364 0.255 Within Groups 576.704 230 2.507 Total 586.962 233 PEI Distress (PI + FS) Between Groups 5.439 3 1.813 1.663 0.176 Within Groups 250.709 230 1.09 Total 256.147 233 PEI PER Between Groups 7.499 3 2.5 1.19 0.314 Within Groups 483.208 230 2.101 Total 490.707 233 Total PER Between Groups 520.243 3 173.414 1.299 0.276 Within Groups 30,711.7 230 133.529 Total 31,231.9 233 Note: Cor relation is significant at the .05 level (2 tailed) ; PEI = Pity Experience Inventories; CC = Compassionate Caring subscal e; PI = Passive Identification subscale; FS = False Superiority subscale; PER = Predominant Emotional Response
187 Table 4 29 tests for Multiple Comparisons [Group Differences] Group (I) Group (J) Mean Difference (I J) Attachment Avoidance Score (Av) 1 2 0.059 0.989 3 0.501 0.050 4 0.260 0.523 2 1 0.059 0.989 3 0.442 0.108 4 0.201 0.721 3 1 0.501 0.050 2 0.442 0.108 4 0.240 0.611 4 1 0.260 0.523 2 0.201 0.721 3 0.240 0.611 G roup Specific Prototype Memory Vividness rating 1 2 0.152 0.863 3 0.007 1.000 4 .574(*) 0.019* 2 1 0.152 0.863 3 0.159 0.855 4 0.422 0.141 3 1 0.007 1.000 2 0.159 0.855 4 .582(*) 0.021* 4 1 .574(*) 0.019* 2 0.422 0.141 3 .582(*) 0.021* Pity Experience Inventories False Superiority subscale (PEI FS) 1 2 0.157 0.893 3 0.318 0.496 4 0.306 0.519 2 1 0.157 0.893 3 0.161 0.893 4 0.463 0.167 3 1 0.318 0.496 2 0.161 0.893 4 .624(*) 0.035* 4 1 0.306 0.519 2 0.463 0.167 3 .624(*) 0.035* Note: Cor relation is significant at the .05 level (2 tail ed)
188 CHAPTER 5 DISCUSSION The purpose of this study was to investigate the effects of attachment style and contextual activation of attachment se curity and attachment insecurity on partner empathy and distress responses to episodes of romantic partner distress. The purpose of this chapter is to discuss the findings of that investigation. This chapter begins with brief overviews of the literature review conducted and the methodology utilized in this study. It then moves to an in depth discussion of the results of the study protocol. Subsequent to this discussion, implications of the study are stated, study limitations are identified, and recommend ations for future research are offered. This chapter then concludes with a brief summary. Literature Review The theoretical foundation for this study was built upon a substantial body of literature relating to adult attachment, cognitive structuring, e motion regulation, and memory systems. Well established findings in these overlapping domains informed the findings to an area of clinical interest helped also to shape the design of the current study. The review of the literature foundational to this study was organized into six areas of inquiry. The first area was an examination of whether attachment is a relevant lens for viewing romantic relationships. That is, whe ther romantic relationships include functional similarities to earlier developmental stage attachment relationships. The second area involved an exploration of the malleability of earlier shaped attachment dynamics. Or more specifically, the degree to wh ich earlier established ways of relating
189 in attachment relevant contexts remains open to influence. The third area centered on internal working models of attachment, as likely mediators of attachment style continuity and change. And the fourth area gave focus to the role that memory systems play in resisting accommodation of new attachment relevant information to existing cognitive and affective schema. The fifth area of inquiry related to the relationship betw een attachment security and pro social motiva tion. Specifically, it involved an examination of whether and how state s of dispositional and/or contextually f elt attachment security enhance emotional inclination, to help others. And t he sixth area pertained to stu dies investigating secure base and safe haven behaviors in romantic relations hips including discussion of a gap in this area of inquiry that has helped inform the design of the current study. Methodology This study was designed to facilitate examination o f the independent and interactional effects of attachment style and attachment context as influences in how group design (three experimental conditions and one control co ndition) was used to facilitate manipulation of attachment context. Existing instruments were adapted to operationalize attachment style and responses to partner distress. A brief overview of participants and procedures is provided in this section. Study Participants The participants in this study were 132 married couples (264 individuals) between the ages of 20 and 59 years, recruited from 10 churches in Alachua County, Florida. Length of current marriage ranged from a few months to 39 years, w ith a median of 12 years and a mean of just under 14 years. Random stratification, with length of current
190 marriage as the stratifying variable, was used to assign 33 couples (66 individuals) to each of four groups. A total of 30 data sets were excluded f or missing or unusable data, leaving 234 data sets for analyses. Examination of couple communication quality questions required the use of only matched pairs data sets. For this area of analyses only, 28 sets of otherwise useable data were omitted, becau se their corresponding spousal data sets had been among those earlier excluded. Analyses of this component, then, were limited to 103 matched pair data sets (206 individuals). Study Procedures Participation required pen and paper responses to seven tasks, encompassing eleven pages, and requiring approximately one hour of time. Six of the seven tasks were identical for all participants, with the remaining task representing the four group differing conditions of the study. Separate administrations of the s tudy protocol often included multiple couples, representing more than one group. This was possible because the four group differing tasks were parallel in structure and length, permitting the protocol session administrator to simultaneously guide particip ants through divergent content with uniform timing prompts. In the first task assessment of the intervening variable of attachment style was item Experiences in Close Relationships Questionnaire Revise d baseline levels of attachment avoidance and attachment anxiety were calculated from the two 18 item subscales of this instrument. The second task was a distracter task, in which participants were asked to recall several rather innocuous events from their past, and to rate the certainty of their recollections. This task was designed both to divert participant attention from the task
191 just completed, and to frame subsequent tasks as memory exercises, with the hope of eliciting more eager engagement in subsequent tasks involving the revisiting of less pleasant memories. In the third task participants were asked to select a memory of a time when their spouse was highly distressed and desiring their comfort and support. Memories selected were later used as stimuli for eliciting participant dependent variable empathy and distress responses. Memory selection criteria were provided, with the goal that both the level and the nature of participant emotional arousal induced would serve as an ample stimulus for later dependent variable responses. The fourth task was designed to stimulate within participants the differing group specific variations of attachment context. Participants in the first group were guided in reflecting upon a relationship fitting a prototype provided, designed to heighten accessibility of secure attachment memory representations. Similarly, the group two prototype was intended to heighten attachment avoidance, and the group three prototype to heighten attachment anxiety. The control condition (group four) involved a neutral reflection task a prototype related memory activity designed not to activate any aspect of the attachment system. At the end of this task, participants in all four groups rated the vividness of their immersion in memory with the person fitting the prototype provided. In the fifth task participants were asked to revisit the memory of partner distress selected in task three. Prompts provided were designed to encour age participants to immerse themselves in the memory, after which they were asked to rate the degree of
192 Subsequent to this activity, participants were asked to imagine a recurrence of this episode of partner distress. Again, prompts encouraged participant immersion in the imagined episode recurrence and vividness level ratings were again recorded. The sixth task hy and distress responses to the spousal distress episode stimulus. Empathy and distress use in this study: (a) the 14 d (b) a 30 item version of the Pity Experience Inventories. These 44 items were presented in three pages, and at the end of each page participants again rated their success at sustaining immersion in the imagined recurrence of the spousal distress episode In the seventh task participants were asked to provide rating responses to four questions assessing empathy eliciting and empathy providing behaviors in their current marriage: (a) frequency of their own distress accompanied by desire for sp ousal own responses at such times. Additional questions elicited participant ra tings of self and partner as empathy givers in significant relationships predating the current marriage. This data was not processed, however, for reasons discussed in Chapter 4 Data collected from the various procedures were in the form of 7 poin t Likert responses. There were 91 such responses: (a) 36 items assessing the intervening variable of attachment style (Task #1); (b) 44 items assessing the dependent variable of empathy and distress responses (Task #6); (c) 7 items assessing various aspec ts of
193 memory engagement (Tasks #4, #5, and #6); and (d) 4 items assessing spousal communication quality (Task #7). Results This section includes discussion of results relative to instrument reliability, attachment style effects, attachment context effec ts, gender effects, and other effects. Relationship Communication History Survey (RCHS) findings are discussed, as are certain instrument development findings of note. Instrument Reliability Internal consistency estimates for the three instruments adapted for use in this study were comparable to those reported in the literature for use of the original instruments. A summary of these compariso ns is displayed in Table 4 5 The reliability coefficients for the avoidance scale ( = .93) and for the anxiety sc ale ( = .90) of the adapted version of the ECR R used in this study were similar to those commonly reported for the two scales ( ECR R (Fraley, Waller, & Brennan, 2000; Sibley & Liu, 2004). Relia bility estimates for 6) and for the distress scale the BEDI used in this study were comparable to those reported for the original BEDI (2001) studies that the current study was designed to extend. version of the PEI to which it was compared. T he internal consistency estimate for the
194 used in the current stud y differed substantively from the version used by Mikulincer et al. (2001), current study versus earlier version reliability measures represent comparisons of response consistency to an array of both shared and differing items. The Instrument Development s ection later in this chapter includes an extensive selection criteria may have contributed to their respective internal consistency estimates. Notwithstanding these caveats subsequently discussed, reliability coefficients for five of the six current study scales were equal to or higher than those reported in earlier uses of the instruments adapted. And the one exception was only moderately lower. Therefore, with the current study reliability estimates ranging from for use in this study may be regarded as reliable. Effects of Attachment Style esearch questions (Q6, Q7, Q8, and Q9) related to the effects of the intervening variable of attachment style on the dependent variable empathy and distress responses to partner distress. These research questions are enumerated below along with a discussi on of the findings. Question 6 (H3 ): Higher attachment avoidance scores will predict lower empathy positive scores (empathy score minus distress score) for all three experimental conditions and control condition in response to marriage partner distress T he hypothesis represented by this research question was confirmed. correlations we re negative, and they ranged from r = .327 to r = .349. Because
195 predominant emotional response (PER) is a composite score derived by subtracting the distress score from the empathy score, an examination of the component scores may provide insight into t he emotion regulation processes specific to attachment avoidance. Correlations between attachment avoidance and the two empathy measures (BEDI and PEI CC) were negative and even stronger than the avoidance/PER correlations (r = .38 that attachment avoidance prompts suppression of empathic feelings, thoughts and motivations toward partner distress as a means of regulating personal distress. This interpretation seems to be supported by the component avoidance/distress scale correlations, where the PEI passive identification sub scale score was negatively correlated with attachment avoidance (r = false superiority sub scale score was positively correlated with A discussion of the distinction between these two sub scales can be found in the Instrument Development section later in this chapter. For purposes of the present discussion, it is interesting to note that attach ment avoidance was negatively correlated (PEI PI), but was positively correlated with the more highly defensive and egoistic feelings, thoughts, and wishes of the PEI FS measure. The combined distress measures revealed no significant correlation with attachment avoidance (BEDI distress: r = .087; PEI total distress: r = .008). Taken together, these findings suggest that the inter and intrapersonal strategies underlying attachment avoidance include the following: (a) an increased tendency to suppress empathy related emotions, cognitions, and motivations toward a distressed partner; (b)
196 a decreased tendency to seek to calm a distressed partner as a strategy for self regulation of negative affect; (c) an increased tendency to self regulate by avoidance of interpersonal and intrapersonal distance to achieve the overall goal of not ex ceeding a description of attachment avoidance dynamics (1969/1982; 1973; 1980), as well as with summaries of adult attachment literature exploring attachment avoidance in romantic Question 7 (H3 ): Higher attachment anxiety scores will predict lower empathy positive scores (empathy score minus distress score) for all three experimental conditions and cont rol condition in response to marriage partner distress The hypothesis represented by this research question was confirmed. measures of predominant emotional response were somewhat lower than the Av/PER were negative also, ranging from r = .212 to r = .255. The results of an examination of the component scores comprising these measures seem to suggest which emot ion regulation processes underlying attachment anxiety may be parallel to processes undergirding attachment avoidance, and which may be divergent. Correlations between attachment anxiety and the two empathy measures (BEDI and PEI CC) were much milder than those between attachment avoidance and empathy. The BEDI empathy with Ax correlation was r = BEDI with Av: r =
197 r = .118; not significant (compared to PEI CC wit h Av: r = in contrast to the negative correlation between attachment avoidance and the somewhat other oriented PEI PI distress measure (r = with Ax correlation (though meager), was in the opposing direction (r = .057; not significant). Taken together, these findings suggest that attachment anxious individuals are less likely than are those who are attachment avoidant to suppress empathy related emotions, cognitions, and motivations, or to resist ot her identifying orientations, as emotion regulation strategies in the face of partner distress. A comparison of Ax scores with the various component distress scores, however, may indicate that this difference comes at a cost. In contrast to the insignific ant correlations between attachment avoidance and both BEDI and PEI total distress (r = .087 and r = .008 respectively), the Ax with BEDI distress correlation (r = .185; suggest lesser efficacy for attachment anxiety spawned emotion regulation strategies when encountering partner distress. These findings seem to support the characterization of anxiously along with the approach avoidance conflict these individuals subsequently experience when such merging leads to emotional flooding (Bowlby, 1969/1982; 1973; 1980; Bretherton & Munholland, 1999). This difference between anxious and avoidant attachment emo tion regulation strategies may help to explain the even stronger positive Ax with PEI FS correlation explanation is that compared to those whose attachment insecurity manifests as
198 avoidance, the anxious individual is less able to avert emotional flooding by suppression of empathy related cognitions, affect, and motivations, and thus must more strongly revert to more rigidly defensive processes (such as those measured by the PEI FS t anxiety correlated more strongly with the PEI FS than did attachment anxiety (2000). Question 8 (H3 ): Higher attachment avoidance scores will predict greater use of distancing strategies in response to marriage partner distress when controlling fo r experimental condition and gender The hypothesis represented by this research question was confirmed. The three item distancing subscale of the seven item PEI false superiority sub scale was composed of those items that indicate the motivation to use ph ysical or emotional distance as a means of self regulation in response to partner distress [PEI Q6: I find myself wanting to run away from the scene ; PEI Q8: I find myself wanting to forget about all this that is happening ; and PEI Q25: I find myself wanti ng to emotionally distance from the scene ]. The correlation between attachment avoidance and the PEI contrast, the positive correlation between attachment anxiety a nd the PEI FS distancing score did not rise to the level of significance (r = .116). The fact that the Av with PEI FS distancing correlation (r stronger than the Ax with PEI FS distancing correlation (r = .116; not significant), yet the PEI FS correlatio some Av/Ax differentiating dynamic in the remaining four PEI FS scale items strong enough to account for su ch a major swing.
199 Indeed such findings were observed A moderate difference was found in attachment style correlations with PEI Q18 [ I find myself afraid that I am going to get hurt as this situation unfolds lation with Ax .01]. Much larger differences, however, were found in regard to PEI Q11 [ I find myself ; correlation with Av: r = .053; 22 [ I find myself feeling concerned about the future [ I find myself feeling guilty about not being able to help my husband or wife ; correlation with Av: r = .005; correla item correlations with attachment avoidance and anxiety scores can be found in Tables 5 1 and 5 2 at the end of this chapter. These examples, however, seem to provide compelling evidence th at attachment avoidance prompts emotional distancing strategies as a means of negative affect regulation, while attachment anxiety results in more self focused introspective distress ruminations in response to partner distress. These findings are consiste affects of attachment style on intra psychic processes (2007). Question 9 (H4): Higher attachment avoidance scores will predict lower partner in distress memory vi vidness ratings when controlling for experimental condition and gender. As with the earlier three questions regarding the effects of attachment style, the hypothesis representing this research question was also confirmed. The correlation between attachmen t avoidance and partner in distress memory vividness ratings (when controlling for experimental condition and gender) was negative ( r = .129), and was
200 vividness correlation (r = .039), and is consistent with earlier discussions in this subsection of the emotion regulation strategies characteristic of attachment avoidance. Specific to this research question, participants higher in attachment avoidance exhibited a significantly greater tendency to suppress engagement in the memory and the imagined recurrence of the episode of partner distress, as a means of regulating negative affect associated with such visualizations. By contrast, participants higher in attachment anxiety did not attempt (or were not successful in their attempts) to regulate their negative affect by such strategies. The fact that all four hypotheses relative to attachment style were confirmed seems all the more significan t in light of the lower mean and standard deviation of compared to the norms repor ted for the married population in general (Av: M=2.87; SD=1.27; Ax: M=3.64; SD=1.33) (Note: scores reflect 7 point Likert ratings) The lower mean scores in the current study may have served to compress the variability of the scores, and consequently infl ated the level of correspondence between the two sets of scores (r = .53) compared to the level of correspondence between attachment avoidance and attachment anxiety scores reported for the general populati on (r = .41; see Table 3 7). An alternative hypo thesis is that the comparative range restriction of of range of scores, and the resul tant higher Av/Ax scores correspondence, it seems possible that findings for the general population might yield even stronger correl ations
201 than those calculated in testing the above four hypotheses relative to the effects of attachment style. Effects of Attachment Context the effects of the differing group conditions of attachment security priming, attachment avoidance priming, attachment anxiety priming, or neutral priming, on dependent variable empathy and distress responses to partner distress. These research questions are enumerated below, and are then discussed as a group. Question 1 (H1): Attachment security priming condition will result in higher empathy positive scores (empathy score minus distress score) than will attachment avoidance priming condition, attachment anxiety priming condition, or control condition Question 2 (H1): Attachment avoidance priming condition will result in lower empathy scores than will attac hment anxiety priming condition Question 3 (H1): Attachment anxiety priming condition will result in higher distress scores than will attachment avoidance priming condition Question 10 (H4 ): Attachment avoidance priming condition will result in lower p artner in distress memory vividness ratings when controlling for attachment style and gender None of these four research hypotheses relating to the effects of attachment context were confirmed. The following discussion addresses four (potentially overlap ping) reasons that might explain these findings: (a) the likelihood that the effects of attachment style dwarfed the effects of attachment context; (b) the likelihood that attachment context priming resistance dynamics at work in responses to partner distr ess are greater than those at work in responses to stranger distress; (c) the
202 possibility that the methodology chosen to prime specific attachment mental representations in memory was less than optimum for the combination of target participants (couples) a nd the target empathy and distress invoking stimulus (episode of partner distress); and (d) that there are nuanced factors at work when priming the full spectra of attachment memory representations (security avoidance and anxiety compared with neutra l priming) in the current study, that were not equally operational in and positive affect priming. The first possibility in considering the absence of significant group differences, attributable to the priming of group differing attachment relevant memory mode of responding in attachment relevant contexts, dwarfed any effects contributed by the attachment priming. That attachment style itself had significant effects on each of the variables central to these research questions (predominant emotional response; empathy responses; distress responses; partner in distress vividness ratings) was ev ident from discussions earlier in this section. A complicating factor may have been that despite random assignment, the (latent) level of attachment avoidance for the attachment security priming group (group one) fell just short of being significantly hig her attachment avoidance and anxiety (attachment style) was large enough to overrun any actual affects from the differing attachment context priming conditions o f the various groups.
203 The second possibility in considering the absence of significant group differences is that there are more malleable dynamics at play in priming an altruistic mindset toward a stranger in distress, than there are in contextual attempt response tendencies toward spousal distress. The suggestion of this possibility is not to be confused with the more cynical observation that it is easier to be kind to a stranger than it is to be kind to your spouse. Instead, it is intended to give weight to the likelihood that a heightened sense of felt security may increase an altruistic sense of resource expendability on behalf of a distressed stranger (earlier studies: e.g., Mikulincer et al., 2001; Bartz & Lydon, 2004; Mikuli ncer et al., 2005), in ways that are less easily influenced within the multi faceted milieu of habitual and reciprocal partner response histories (current study). The third possibility for explaining the absence of group differences relates to e xperimental design deficiencies. It is possible that a more powerful attachment priming effect might have been achieved by utilizing a partner based attachment prototype. Specifically, the effects of the manipulation of attachment context priming might h ave been more powerful had participants been asked to reflect upon an interaction with their spouse that fit the group specific prototype. Although attachment theory tenets relative to internal working model function and memory system interplay suggest th at earlier attachment prototypes comingle with and influence current attachment relationships (see Chapter 2 discussion), direct targeting of secure avoidant or anxious attachment schema embedded within memories of current partner interactions may ser ve as more effective stimuli for priming attachment context.
204 Imagining the recurrence of an earlier episode of partner distress may also have represented a less than optimum design distinctive. Theoretically, the heightened or diminished sense of attach ment security resulting from the group specific priming condition should serve to frame an already lived episode in a new light and thus to increase or decrease accessibility of certain altruistic or egoistic cognitions, affect, and motivations (see discus sion in Chapter 2 ). But practically speaking, memory of actual thoughts, feelings, and wishes in the original episode may have served to limit a fresh experience of the episode, thus diluting the effects of contextual attachment priming as a means of brin ging a new set of eyes to the earlier experienced event. The potential limitations of this study design feature were anticipated when exploring methodological alternatives, but were overcome by a paucity of design alternatives lending themselves to implem entation in a group setting. The fourth possibility for no significant differences found between the groups is a more nuanced one, and relates to the complexity of altruistic and egoistic processes that seem to be at play when the opportunity or need to support a distressed other is considered in an effort to explain an unexpected group difference, where group four (neutral priming) participants exhibited significantly highe r PEI FS distancing scores occurrence seemed particularly unlikely in that PEI FS scores were positively (and significantly) correlated with attachment avoidance and anxiety, yet group four participants ex hibited the second lowest latent Av scores and the lowest latent Ax scores of any of the four groups. Initial attempts to make sense of this intriguing
205 outcome centered on the nature of the neutral priming activity imagining and reflecting upon an acquain tance, who might be effective serving as mayor of a town or small city. It was considered that perhaps what had been intended as a neutral exercise had instead tapped into a sense of cynicism associated with the political climate at the time of protocol a dministration (the months immediately preceding the 2008 national elections). Yet analyses of individual item response group differences seemed to suggest another cause. Most of what distinguished group four PEI FS responses from those of the other group s was related to the more highly self protective by distancing wishes sampled by the PEI FS scale (see discussion earlier in this section). In other words, it seemed that there was something about the group four priming stimulus that served to render part icipants comparatively defenseless in the face of an imagined recurrence of partner distress. An intriguing possibility for explaining this phenomenon may be extrapolated from the discussion of findings under the earlier Effects of Attachment Style sectio n. Although involving more complexity than a brief explanation can encompass, the reasoning underlying this possibility might be summarized in the following three propositions: That the process of effectively responding empathically to partner distress does were antithetical to the giving of empathy, but rather it calls for the ability to manage that distress as a means of other oriented perspective taking, resulting in i ntuitive formulations of empathic response. That attachment security serves to enhance distress management by increasing distress management by contributing to defensive processe s that abet inter and intra personal distancing from its source. And attachment anxiety serves to manage distress by inducing ambivalent merging with, and sometimes subsequent abrupt withdrawal from, its source.
206 That however imperfectly the priming of attachment security, avoidance, or anxiety relevant strategies for managing their own distress relative to the remembered partner in d istress episode, the attachment systems of those in the first three groups had been activated prior to the event revisiting task. By contrast, group four participants were comparatively ambushed, prompting them to resort to more overt distancing strategies as a means of distress management. Clinical and research implic ations associated with this finding of group difference, and aris ing from the possibility that this difference may be accounted for by the hypothesis proposed, will be discussed later in this chapter. Effects of Gender en research questions (Q4, Q5, and Q11) related to the effects of gender on dependent variable empathy and distress responses to partner distress. These research questions are enumerated below, and are then discussed as a group. Question 4 (H2 ): Females will exhibit higher empathy scores than will males in response to marriage partner distress Question 5 (H2): Males will exhibit greater use of distancing strategies than will females in response to marriage partner distress Question 11 (H4 ): Males wi ll exhibit lower partner in distress memory vividness ratings when controlling for experimental condition and attachment style None of these three research hypotheses relating to the effects of gender were confirmed. The original PEI authors (Florian et al., 2000) found gender differences in both the PEI empathy measure ( compassionate caring scale) and in its somewhat other oriented distress measure ( passive identification sub scale). In both cases, they cited significantly higher scores for women. Becau se these scores were virtually
207 analyses of data were conducted in an effort to explain this unexpected outcome. The only significant gender difference found among the var iables analyzed related to the vividness level reported for participant visualization of the group specific prototype, where women provided vividness ratings that were significantly higher than dif ferences may be artifacts of a gender vividness rating bias, as opposed to evidence that men actually had greater difficulty with (or expended lesser effort toward) engaging vividly in memory with the person/relationship fitting the group specific prototyp e. Yet lower male ratings of partner in distress mean vividness approached significant earlier suggested lens of competing altruistic and egoistic processes precipita ted by the empathy shortfall through emotion regulation strategies, including suppression of visual engagement in unpleasant stimuli related to the attachment prototype a nd the remembered episode of partner distress. [Note: See also the finding of Fehr, 2008 (cited by Fehr, 2010) that gender differences in regard to empath y expression was not maintained when the target was ] Other Effects Two additional effects are worthy of note in discussing findings of significance in the current study. The first is the finding that the demographic variable length of current marriage was negatively correlated with attachment anxiety (r = That is, attachment anxiety scores tended to go down the longer one was married. Three possibilities are suggested for this phenomenon: One possibility is that a greater
208 sense of attachment security is achieved as a couple negotiates a mutually supportiv e relationship over time. Another possibility is that higher anxiety relationships tend to be winnowed out through divorce over time, with the effect of skewing individual anxiety levels downward among relationships that survive. And yet another possibil ity is that longer married higher anxiety couples may have been less likely to volunteer for the study. The correlation between attachment avoidance and marriage longevity was in the same direction, but fell short of statistical significance (r = = .12). These R normative data (Table 3 7), indicating that attachment anxiety declines with age, while attachment avoidance er, and the median age of 24 years somewhat limits the applicability of this normative data as a The second finding of note not earlier discussed represents an artifact of study design that is statistically significant but of seeming little practical import. This finding obvious reason for this difference is that reflection on the individual fitting the group four (neutr al priming) condition prototype did not lend itself to a level of vividness commensurate with the secure avoidant or anxious attachment priming prototype reflection tasks of the three non control groups. Specifically, participants in the first three g roups were asked to reflect upon a relationship that matched a provided attachment prototype, while the group four participants were asked to imagine a casual acquaintance functioning in the role of mayor of a town or small city. Because reflection on a h istorically experienced relationship is arguably more likely to be vivid than
209 imagination of an event never experienced, the finding of lower control group prototype vividness levels seems both expected and non problematic. Relationship Communication His tory Survey ( RCHS ) Participants provided ratings to four items in t he RCHS intended to assess the of partner empathy that was provided on such occasions. In re sponse to the first question (Q #1), participants reported their own frequency of distress accompanied by support seeking, and in the s econd question (Q effectiveness in responding supporti vely. In the third question (Q #3), participan ts reported how often their spouses looked to them for support when distressed and in the fourth question (Q #4), they provided an assessment of their own competence at giving support at such times. Analyses of between question correlations spoke to the q uestion of whether and how self and other distress frequency was related to other and self comforting effectiveness Responses to the four questions were subsequently correlated with the demographic, intervening, and dependent variable responses in th e main study, and gender differences were examined. These analyses were an attempt to answer the question of tendencies translate into marital distress and empathy exchange dynamics Finally, matched p study data, and vice versa. These analyses examined the question of distress and empathy exchange dynamics.
210 Correlates of distress frequency with desire for spousal comfort (Q#1) their own distress accompanied by bids for spousal support, than did husbands. Argui ng against the possibility that this finding represents a gender bias in self reports of distress frequency, was the fact that Q#3 responses were also positively correlated with gender (r = higher distress frequency incidence for wives An alternative hypothesis to the conclusion that wives are more frequently distressed than are husbands, may be extrapolated from the stressing thoughts, feelings, and wishes as an emotion regulation strategy. Such may suggest that rather than being less frequently distressed, husbands may merely be less likely than wives to seek partner support during such times. This hypothesis is se emingly correlated with reports of mate distress frequency (Q#3) (r = this may suggest that spouses differ in the degree to which they choose other regulated versus self regulated processes to ameliorate distress. Other findings subsequently discussed seem to lend weight to this hypothesis. (r = .314; p .01), and consequently with each of the four distress measures correla ting significantly with attachment anxiety (ranging from r = .199 to r = .257; p .01). one significant correlation was observed (at the p .05 level). Yet when linkin
211 the p .01 level; four at the p with revealed a more liberal use of distancing strategies, perhaps as a means of regulating their own distress precipitated by more frequent spousal bids for emotion regulation as p 01), the data went opposing directions when linking Q#1 responses with spousal data: Husbands whose wives were higher in anxiety were less likely to seek spousal support (r = .173; = .08), while wives whose husbands were higher in anxiety were more l ikely to seek spousal support (r = .112; = .26). Although the individual breakdowns fall short of statistical significance, such data continues to be consistent with the earlier discussed male bias of distancing from the source of distress as an emotion regulation strategy. In e distress frequency (r = .039). A possible explanation for this difference emerges when considering the Mikulincer and Shaver, 2007), coupled with earlier discussed f indings of gender bias greater tendencies toward the use of egoistic strategies in the face of partner distress
212 may permit them to default to self regulation of distres s when married to attachment avoidant wives, more readily than can wives who are married to attachment avoidant husbands. orrelate of the length of current marriage (r = ors may be contributing to this pronounced gender difference. One possibility may relate to the finding that both husbands and wives reported significantly greater incidence among wives of eliciting partner support during times of distress (see earlier di scussion). Consequently, there was more room for that this factor accounts for some of the decrease in distressed support seeking as a function of marriage length, this does not explain the change in the opposite direction accompanying marital longevity for husbands. An additional contributing factor, then, may be related to gender differences in regard to empathy (see Chapter 2 discussion). Consistent fin dings of wo provide and garner empathy (e.g., Fehr, 2010), suggest that wives may be more successful than their husbands at support network formation. And that having more reliable alternative others as sources of empathy, w ives may be better able than their husbands to adapt their partner directed empathy seeking bids to the level of empathy giving competence their spouses have demonstrated.
213 Correlates of perceived spousal comforting competence (Q#2) interwoven. Specifically, it may suggest that emotional equilibrium restoring outcomes arising from more predictably predispose a spouse to reciprocate when the shoe is on the other foot. Husbands, who reported more frequent distress among their wives (Q#3), also rated their wives less successful as comforters (r = relationship between those two variables. Viewing this distinction through the lens of earlier discussed gender differences in employing distance as a means of emotion regulation, it may be that husbands are more readily deterred than are wives from seeking support from a partner whose own emotion regulation strategies rely on amplification rather than suppression o f distress. In other words, it is possible that the wives) toward self regulation strategies, averting the perceived potential of spousal distress induction resulting from t heir bids for support. were positively attachment were negatively correlated with Q#2 responses (Avoidance (Av): r = .505; these correlations, dependent variable responses that were significantly correlated with Av and Ax scores in the main study showed predictable direction and strength of correlation when linked with Q#2 responses. Interestingly, the strength of negative c
214 strategies (r = ficacy (r = .174; earlier discussed findings suggesting that overcoming a default tendency to use distance as an emotion regulation strategy may be more central comforting efficacy than it is to that of their wives. Correlates of spousal distress frequency with desire for comfort (Q#3) The frequency of partner distress (Q#3) was negatively correlated with self perceptions of efficacy as a comforter (r = virtually equal levels for both husbands and wives. It is not difficult to imagine why this is so. Similar to the findings discussed relative to Q#1, frequency of spousal distress accompan ied by desire for partner support varied with the length of current marriage frequency of distress accompanied by bids for empathy was negatively correlated with length of marriage (r = of distress accompanied by bids for their support increased with marriage longevity (r = .168). Thus, both the gap, and the length of marriage related opposing direction s, were perceived by both husbands and wives (of themselves and of each other), lending weight to the unlikelihood that such differences were chance occurrences, and perhaps lending credence to the underlying causes hypothesized in earlier discussion. distress frequency (r = .305; p .01) This finding was in tune with frequent clinical observations that the more dismissive strat egies with which attachment avoidant
215 individual s seek to regulate their own e motions tend to be both unwelcome and ineffective as strategies for amelioratin g partner distress, and inde ed may serve to exacerbate it. Assessments of partner distress frequency were negatively correlated .224; p .01) and were positively correlated with their distancing scores (r = .1 53; p .05). These findings suggest that distress managing distancing strategies as substitutes for empathy provision serves to exacerbate frequency of spousal distress. Reciprocally, they may also suggest dynamics of empathy fatig ue leading to the use of distancing strategies in the face of unrequited partner distress. In examining gender differences associated with Q#3 responses, it was noted that scores (r = .210; p .05) and all four of their distress scale scores (ranging from r = .213; p .05 to r = .282; p frequently distr essed and support seeking were more likely to be high in latent attachment anxiety, and consistently reported higher levels of distress in response to the partner in distress episode task of the main study. In stark contrast, husbands whose wives reported them as more frequently distressed and seeking their support showed insignificant main study distress elevations (correlations with the four distress scales ranged from r = .04 to r = .11; alpha levels ranged from .27 to .69). And correlation with the di stancing scale was also mild (r = .09; = .38). Viewed in the light of earlier discussions, this finding may indicate that husbands who more frequently seek
216 support from their wives in times of distress may not be more frequently distressed than those wh o do not, but instead may merely be choosing other oriented strategies over more egoistic strategies for mediating episodes of personal distress. The sense of oneself as an effective comfort g iver rose with length of marriage, attachment avoidance (r = attachment anxiety (r = Comforting efficacy correlated positively with the two empathy measures (BEDI: emotional response (ranging from r = .346 to r correlations are only moderately lower than empathy score correlations reflects the finding that overall distress scores were not significantly elevated among those rating themselves more highly in spousal comfort ing effectiveness. The data indeed indicates moderately positive distress scale correlations with comforting competence ratings, but interesting these are largely offset by the negative correlation with the PEI FS distancing subscale (r = These findings indicate that higher levels of both avoidance and anxiety hinder spousal comforting efficacy, and that the ability to to distancing strategies is a signif icant factor in such efficacy. As with several findings earlier discussed, an examination of the cross correlated data suggested significant gender differences regarding spousal characteristics cy. Although a more minor difference, it is worthy of note that both attachment avoidance and attachment anxiety
217 were more highly contraindicated for husbands whose wives rated themselves as effective comforters (Av: r = .01), than for wives whose husbands gave themselves higher comfort giving ratings (Av: r = r = in comforting efficacy scored higher on both empathy mea main study scores were significantly correlated comforting competence. Earlier findings (see Table 4 9) showed both Av and Ax negatively correlated with both empathy measures and all three PER measures. Taken alone, this would argue for higher empathy ratings among those lower in avoidance and anxiety (husbands of women rating themselves higher in comforting efficacy). Yet the reverse direction appeared. Once again, this suggests a gender difference in empathy eliciting and empathy providing tendencies among the subject couples, and one that supports a view of women as more effective empathy givers (Fehr, 2010). It may be that higher empathy in wives contributes to a climate that permits men to overcome distancing tendencies in favor of giving them comfort w hen distressed thus making emotion regulation a collaborative venture. By contrast, it may be that wives do not need the comparative safety of higher empathy husbands to achieve comfort giving competency Instrument Development In contrast to the ECR R AV and the BEDI AV of the current study, representing relatively minor adaptations of the original instruments, the version of the PEI used in
218 this study represents a more ambitious instrument develop ment effort. The version of the PEI used by Mikulincer et al. (2001) utilized 24 items (14 empathy items and 10 distress items) from an earlier developed 59 item pool of (PEI) feelings thoughts and wishes sampling items. The version of the PEI devel oped for use in this study utilized 30 items (13 empathy items and 17 distress items) from that same 59 item pool. Figure 3 3 (at the end of Chapter 3 ) provides a schematic summary of the rationale and item selection process utilized in development of the Mikulincer and his colleagues (2001) did not list the 14 empathy items they selected for their short form version of the PEI (PEI SF), from among the 18 available empathy items in the original PEI item pool. Therefore, apart from the three item how many of the 13 empathy items selected for use in the form (PEI SF R) overlap with those used in the earlier version. Given the fairly modest used version ( within the range of chance. Two factors converge to argue for at least considering another possibility, however. The first is that although the two scales of necessity shared a significant number of items in common, they were not identical. And the panel of experts, assessing which items exhibited optimum face validity as viable is possible that this test
219 PEI empathy scale version, and consequently to higher internal consistency estimates. As with the empathy scale, Mikulincer et al. (2001) did not provide a listing of the 10 distress items they used in their short form version of the PEI. Complicating this item passive identification subscale, item false superiority subscale Yet two of the three distress items the authors cite as PEI SF examples were indeed taken false superiority subsc ale. Whereas a significant overlap can be the available 18 items), the earlier and curr to be far more divergent. There are three reasons for this assumption: First, the probability of any given item correspondence between the two versions is diminished because of the higher number of distress ite ms (N=41) than empathy items (N=18) in the 59 earlier version; 17 items in the current stu scale ensures divergent item selection. And third, the earlier authors stated their intent to achieve relative distress item h passive identification distress subscale (despite their apparent inadvertent straying into the false
220 superiority item pool). But the design of the current study argued for intentional and generous sampling from b oth PEI distress subscales. A more thorough discussion of the rationale behind the use of a less homogenous distress item scale in the current study can be found in Chapter 3 For the purposes of this discussion, that rationale can be summarized as follow s: In contrast to the multiple studies by Mikulincer et al. (2001), that were designed to compare the effects of secure attachment priming with positive affect priming and neutral priming, the current study was designed to compare the effects of secure avoidant anxious and neutral attachment priming on empathy and distress responses to a distressed other. In the earlier studies, therefore, it was not essential to account for the phenomenon that attachment avoidance and attachment anxiety descr ibe divergent emotion regulation strategies for compensating for a void in felt security, because study design dictated the need only to distinguish between attachment secure and attachment insecure r, did argue for such an attempt to differentiate attachment avoidance promoted responses from attachment anxiety promoted responses. Little was found in the literature to inform the attempt to distinguish between (what I have labeled) attachment insecurit y homogenous items (those useful merely for differentiating attachment security from insecurity in general) and attachment insecurity divergent items (those useful for distinguishing between the two types of attachment insecurity). The one exception to th is silence was the observation of Florian, Mikulincer, and Hirschberger (2000) that the 23 item false superiority distress subscale correlated
221 ugh these correlations revealed greater shared variance (insecurity versus security) than unshared variance (anxiety versus avoidance), the level of differentiation revealed for the false superiority scale as a whole argued for the possibility that specifi c scale items might display greater levels of divergence. An additional factor argued for more generous sampling of the false superiority scale items insecurity divergent items. A lthough somewhat of an oversimplification, it would not be inaccurate to view the distinction between the empathy scale and the two separate distress scales as follows: The empathy ( compassionate caring ) scale samples other oriented (altruistic) feeling s, thoughts, and wishes directed The passive identification distress subscale samples self and other oriented responses motional equilibrium. And the false superiority distress subscale samples self oriented responses that indicate a higher level of intra psychic enmeshment in, or regulation. Given these distinctions coupled with the need to differentiate attachment avoidance motivated responses from attachment anxiety motivated responses, in overall scale internal consistency (see Chapter 3 discussion). Simply put, it seemed unlikely that attachment avoidant responses could be distinguished from attachment anxiety responses without drawing generously from the pool of distress items sampling the more extreme defensive responses to from that item pool would of necessity provide a downward skew to the homogeneity of
222 full distress scale responses, because participants biased toward attachment avoidance would likely give lower ratings to attachment anxiety biased items and those with a bent toward attachment anxiety would provide lower marks to avoidance biased items. The fruits of this tradeoff were evident in the (17 items) than in the earlier studies (10 items). Th at this lower internal consistency estimate is attributable to this tradeoff seems likely from the fact that current study reliability coefficients were of equal magnitude or higher than those reported in the literature for each of the other five scales ad apted for use in this study (see Figure 4 5 and Instrument Reliability discussion earlier in this chapter). Because no significant group differences were found that could be attributable to t he differential attachment priming conditions (see discussion ear lier in this chapter under Effects of Attachment Context ), the methodological consideration of examining such differences through the lens of attachment insecurity differentiating items (see C hapter 3 discussion) became moot. However, the presentation of distress item analyses findings may be useful for informing future research, where refining of methodologies for the priming of differing types of attachment representations in memory might lead to future observations of significant dif ference. These findings have been displayed in three tables at the end of this chapter, incorporating both the 17 1 provides distress item correlations
223 with (ECR R AV) attachment avoidance scores, ranked from strongest to weakest. Six correlations were Table 5 2 provides distress item correlations with (ECR R AV) attachment anxiety scores, ranked from strongest to weakest. Seven correlations were significant at the p 3 displays the absolute value attachment constructs (avoidance and anxiety). The items in Table 5 3 are presented in a manner that rank orders (highest to lowest) the level of item property effectiveness in discriminating attachment avoidance from attachment anxiety. Absolute value differences ranged from a high of r = .260 to a low of r = .007). As suggested by the original P EI authors (Florian, Mikulincer, and Hirschberger, 2000), the false superiority distress items correlated more highly with attachment anxiety than with attachment avoidance. In fact in the rank order of strength of correlation, four PEI false superiority distress items and five BEDI distress items preceded the highest correlated passive identification item (Table 5 2). By contrast, passive identification items were better represented among those items most highly correlated with attachment avoidance than were those from the false superiority scale (Table 5 1). Both PEI distress subscales contributed amply to the list of items most effectively discriminating avoidant from anxious responses (Table 5 3), validating the methodological decision to sample gener ously from the false superiority sub scale in constructing the overall PEI distress scale. Implication s of the Study Multiple implications of the findings of this study may be suggested to inform attachment related
224 dynamics. Several of these research and clinical implications will be discussed in this section. Additional implications may be extrapolated from subsequent discussions in the section s addressing study limitations and suggestions for future research. One implication of study findings that may inform future research relates to the complexity of inter and intra personal empathy and distress exchange dynamics between couples, in comparison to the dynamics involved in responses to strange r distress, which were the subject of earlier studies informing the design of the current study. Heightening a sense of felt security to stimulate more altruistic momentary responses to a stranger, a transaction neither embedded in historical context with that individual nor engendering future expectations by that individual, can more easily be manipulated than can the dynamics associated with the long embedded interactional n future research efforts designed to examine couples interactions, more individual couple tailored attac hment priming methodologies may be necessary to create a significant attachment context effect. This conclusion suggests an implication of the study f indings for clinical practice. Because the effects of attachment insecurity adversely, and at a significant level, interfere with optimum couple empathy exchange in response to partner distress, and because clinical interventions aimed at influencing (glo bal) attachment style tend to result in only incremental gains, clinical strategies designed to help couples create more relationship specific attachment contexts appear to be the most promising t related relationship
225 dynamics. Although by no means a new thought, the critical nature of such strategies was affirmed by the findings of this study. Another implication for clinical practice emerges from the gender differences observed in the main stud y, that were even more pronounced in correlations of post study participant ratings of distressed partner empathy bid frequency and other partner response quality. These gender differences argue for a component in the earlier recommended clinical strategi es for helping couples create relationship specific attachment contexts. This component relates to the current study findings of greater tendency on the part of husbands to use distancing strategies in response to partner distress, the likelihood that the use of such strategies serves to exacerbate partner distress, and the partner perceived comforting effectiveness of husbands who resist the tendency to distance. Clinical interventions aimed at helping couples bring to conscious awareness the processes u nderlying these emotion regulation strategies, and helping them construct alternative means of bidding for support and framing interpretations of such bids, may be instrument al in helping couples create an attachment context specific to their relationship. Such a context, reinforced by more beneficial outcomes over time, may provide a more fruitful route to bypassing resistant to change problematic attachment schema. As with the earlier clinical implication cited, this suggestion is by no means a new one. But the findings of this study shed additional light on the value of such clinical strategies. An unanticipated study finding suggests implications both for future research and for clinical practice. This finding was the unexpected observation of differ ences in the PEI FS distress scale for the control (neutral priming) group. Although this gro up was
226 baseline attachment avoidance and anxiety scores, an artifact of random assignment that would suggest lower PEI FS scores the group instead scored significantly higher on the PEI FS measure of more extreme distress response. And this difference was accounted for largely by the distancing subscale. In earlier discussion a hypothesis was proposed, suggesting that despite th e differing priming conditions of the three treatment groups, what the participants shared in common was that all were induced into an attachment state of mind prior to the task of reflecting on the partner in distress episode, whereas participants in the neutral priming condition were not. And that however imperfectly these emotion regulation systems were functioning for these participants, the group four participants were by comparison ambushed by the partner in distress episode and thus forced to resort to more extreme distancing strategies as their means of emotion regulation. empathy and distress dynamics are encouraged to incorporate study design elements suitable for eva luating this proposed framework for understanding this unanticipated finding. Whether subsequent research reinforces the hypothesized causes, or merely suggests that the finding of difference was a statistical anomaly, the occasion for attempting to under stand this phenomenon draws further attention to the earlier stated value of utilizing clinical strategies that help couples develop more attachment secure relationship contexts. Specifically, reflection on potential causes underlying this finding calls a ttention to the phenomenon that avoidant and anxious attachment strategies are motivated by attempts to be adaptive to current relationship experiences in such a way that the vital priority of emotion regulation is maintained. Such attempts, shaped by
227 a ttachment history biased interpretations of earlier and current attachment contexts, are often maladaptive. Consequently, the occasion of this unexpected finding of difference highlights the value of clinical strategies aimed at bringing this phenomenon t o light with client couples, accompanied by a coaching of their communication exchanges in ways for maintaining emotional equilibrium. In his own clinical practice, th is author has observed that approaches to this theme that combine insight and skills practice tend to defuse couple recriminations aimed at partner default emotion regulation strategies, and consequently to enhance couple collaboration in constructing inte ractional environments more conducive to empathy exchange. An additional implication of study findings that may serve to inform future research relates to instrument development. Included in Chapter 3 is a discussion of the rationale and the processes inv olved in developing for use in this study a short form version of the Pity Experience Inventories (Florian, Mikulincer, & Hirschberger, 2000) from the original PEI item pool. Such development was necessary because an earlier short form version of the PEI developed by Hirschberger ( 2000 ) was inadequate to distinguish between distress responses motivated by attachment avoidance and those motivated by attachment anxiety. In the PEI short form version developed for the current study, item selection from the o riginal PEI item pool was guided by the goal of including items with properties suitable to distinguish between the two types of insecure instrument are discussed earlier in t his chapter and are displayed in Tables 5 1 thru 5 3 at the end of this chapter.
228 It seems likely that Table 5 properties ma y prove to be underestimates when used with participants with the higher Av and Ax scores representing general population norms. This likelihood is supported by a comparison of Av/Ax score correlations for current study participants (r = .53) with the smaller degree of overlap reported by Fraley (2010) ( r = .41), a difference doubt less attributable to the current than general population Av/Ax scores. These findings of difference suggest that PEI distress items nearer the top of Table 5 3 may exhibit even stronger Av/Ax discriminating properties for the gen eral population than were demonstrated by their use in the current study. This author is unaware of any instrument in current use that attempts to make the attachment style motivated distinctions validated by the use of this instrument in the current stud y. Future researchers of adult attachment dynamics may find these instrument development findings helpful in studies w here distinguishing the underlying forces motivating nuanced insecure attachment dynamics is an area of focus. In addition to the clinical and research implications earlier discussed, t he current s certain implications both for counsel ing theory and for the Theor etical approaches that focus on coa ching couples to alter problematic communic ation dynamics as a means of enhancing relationship stability (e.g., Gottman, 1994) may be informed by the current insecurity related emotions and cognitions that may be motiv ating the behavioral patterns the couple is seeking to change. Similarly, theoretical approaches that focus on helping couples attend to emotional dynamics facilitating or impeding safe haven communication behaviors (e.g., Johnson, 1996) may
229 be informed b s findings concerning the cognitive and motivati onal processes heightened by differin g attachment states of mind and inextricably linked with those affective processes. Certain e xperiential approaches to counselor training (e.g., Renni e, 1998) advocate role playing activities in which tra inees alternately occupy the positions of client and counselor, as a means of better learning the skills of process identification in the counseling dyad setting A learning module that permits counsel or followed by g roup discussions of the emotions, cognitions, and motivations they experienced in response to the partner in dis tress episode, could serve as an effective ass ist in helping marriage and family counseling trainees more intuitively grasp the attachment related emotion regulation processes they will be called upon to help couples identify as a prerequisite to their building of more supportive dyadic exchanges. Limitations of the Study Limitations of the study may be categorized as both sample related and protocol design related. Sample related limitations to generalizability included the use of volunteer participant couples. It may be that those se lf selecting for participation in a study examining re sponses to partner distress diff er as a group from those choosing not to respond to the invitation to participate. Another limitation was the somewhat underrepresentation of racial minorities in the sa mple, compared to such representations in the population at large. And yet another limitation to generalizability was the recruitment of participants from evangelical churches. In their development of the Pity Experience Inventories, Florian, Mikulincer, and Hirschberger (2000) found that those identifying themselves as religious scored higher on the empathy scale (PEI CC) and the other oriented distress scale (PEI PI) than did those who identified themselves
230 as non religious. [Not e: But see also the fin ding of Sprecher and Fehr, 2005 (cited by Fehr, 2010) that spirituality/religiosity was a correlate of empathic response to stranger distress but not to partner distress]. In light of the inverse relationship between empathy and both manifestations o f at tachment insecurity, religiosity and/or other factors associated with the context of participant recruitment may help to explain why current study subjects score d lower than the overall ECR R norms (see Table 3 7) published by Fraley (2010) on both attachm ent avoidance (Av) and attachment anxiety (Ax) [Av : current study = 2.69; married norm = 2.87 ; A x: current study = 2.70; married norm = 3.64]. In considering what portion of the observed differences might be attributable to subject religiosity, it should be noted in regard to the larger Ax gap th had a median age of 24 years, that his findings predicted declining Ax scores with ascending age, and that current study findings reflected a negative correlation between anxiety scores and l norm data also predicted Av score increases with ascending age. Therefore, it would be a mistake to attribute the full (above noted) Ax gap to religiosity and/or other factors related to self selection of participants, and a wider Av gap (than that noted above) should be assumed because of the current study subject religiosity as well as other sampling and self selection criteria l ikely contributed to the obtaining of a sample with higher levels of attachment security, and consequently with greater tendencies toward empathic responding than those observed in the broader population. Although this sampling limitation cannot be disreg arded, it can conversely be argued that given the relative absence of extremes in participant attachment
231 insecurity, findings of difference correlated with avoidance and anxiety scores may be all the more noteworthy. Several limitations rela ted to the design of the study itself are worthy of note. Some of these limitations were foreseeable, yet were necessary tradeoffs involved in using a larger number of participants. Included among these limitations were potential response biases associat ed with pen and paper data collection. Partial compensation for such response biases could have been made by the inclusion of a social desirability scale collection to parti cipant pen and paper responses was that observation of participant behaviors in response to actual episodes of partner distress might have provided a fuller picture of couple interactional dynamics associated with bids for empathy and partner responses to those bids. Florian, Mikulincer, and Hirschberger (2000) developed a PEI behaviors inventory in addition to the ir thoughts, feelings, and wish es inventory used in this stu dy. But observation s of couple behaviors i n response to partner distress would have required a commitment of time and reso urces not possible within the scope of the current study The finding of n o sig nificant differences between the groups in regard to differin g attachment priming conditions may have been related to both sample and prot ocol design limitations. In regard to the sample itself, r mean and consequently restricted range avoidance and anxiety scores (see discussion earlier in this chapter) may have served to suppress the effect size attributable to the differing attachment priming conditions Two separate yet interrelated issues associated with protocol design may have provided limitations to the study as well. First, it seems likely
232 that t he comparative effect size of the partne r in distress episode stimulus, coupled with the pull of historically establishe d couple interactional dynamics, may have dwarfed the treatme nt effect itself. And second, it seems possible that att achment priming stimuli that we re more partner specific might have created a gre ater effect size than the ones provided. For example, instead of assigning group one to refle ct upon a relationship that fit the secure attachment prototype, p articipants might have been asked to reflect upon a time with their spouse th at was characterize d by the [secure attachment] description provided. Such a design change might have help ed to mediate the problems associated with the dependent variable inducing stimulus of imagining the recurrence of a pas t episode of partner distress. Given the const raints (including perhaps the ethical ones) of precipitating real time partner in distress episodes as response stimuli, it was necessary recurrence was informed by attac hment theory research suggesting the effects of attachment state s of relevant event. Imagining a recurrence of the partner in distress episode, then, was intended to prompt participants to re experience the event in their currently induced attachment state of mind, rather than merely reporting what they thought, felt, and cited limitations of reporting on memory reenactme nts, rather than observing real time behaviors, the priming of attachment state of mind with actua l spouse specific memories may prompt participants to re experience partner in distress episodes wit h different eyes, and thus may serve as more ef fective pri ming prompts.
233 One outcome arising from the finding of no significant group difference was that participant responses in the Relationship Communication History Survey could be correlated with their dependent variable responses to the partner in distre ss stimulus, with no muddying of the data by the (earlier task) differing group conditions. This outcome was a somewhat redemptive one, with some of the most intriguing findings emerging from this portion of the data. The design of the current study rest ed upon a foundation that had been substantively laid within its major areas of inquiry, yet it also reached into a domain hitherto not thoroughly explored. Like any such reach, it was rudimentary and inherently in need of refinement. Future researchers are invited to build upon its incremental reach toward informing clinical strategies designed to enhance secure base and safe Suggestions for Future Research Future researchers may wish to extend the findings of t his study to more racially diverse populations by providing invitations to participate at less racially homogenous gatherings than did the current study. They may also wish to recruit participants from social networks unidentified with any religious persu asion. Overcoming the limitation of self selection is a more difficult undertaking, however. Involuntary participation seems likely to diminish the level of engagement and response authenticity in regard to study tasks that are both personally intrusive and unpleasant. And the de facto captive audience for much psychological research (university students) poses its own self selection problems (aptitude and perhaps economic status), not to mention generalizability problems to the target population (marrie d couples, from newlywed to seasoned). Future researchers are encouraged to weigh the tradeoffs of self selection with other means of securing participation as well as to sample from a more diverse
234 array of venues with the goal of minimizing current study sample limitations earlier discussed. Future researchers who seek to build upon this study using pen and paper data gathering my wish to consider embedding a social desirability scale within participant response options. They may also wish to utilize m ore partner specific attachment priming stimuli for the independent variable group differing conditions. Specifically, it is suggested that future researchers redesign the attachment context priming prototypes presented in Tables 3 8 thru 3 10 in such a way that participants are instructed to think about a time in their relationship with their spouse that fit the attach ment criteria cited, rather than think ing about a specific (non spousal) relationship that was characterized by the prototype description. Future researchers with fewer time (and other resource) constraints are encouraged to consider use of real time laboratory observations of spousal behaviors in response to partner in distress episodes accompanied by bids for support, as an alternative to self reports of current responses to remembered events. Finally, given the intriguing gender and attachment style differences observed in participant responses to the Relationship Communication History Survey (RCHS), future researchers are encouraged to attempt to refi ne this rudimentary instrument and/or to cross validate its findings with the use of other measurements of relationship satisfaction. Summary avoidance an d anxiety, coupled with the effects of the priming of specific attachment representations in memory, on their emotions, cognitions, and motivations in response to a remembered episode of spousal distress. Significant effects were found for the
235 two part in tervening variable of attachment avoidance and anxiety, in correlations with dependent variable empathy and distress responses to partner distress. No significant effects were found for the group differing conditions of attachment priming. Significant ef post study communication study empathy and distress responses to spousal distress. Implications of these findings were di scussed, along with limitations of the study and recommendations for future research.
236 Table 5 1. Distress Item Correlations with Attachment Avoidance [Av] (in rank order from greatest to smallest) [N=234] Index Q # Distress Item Core Phrase Pearson Co rrelation BEDI 6 find self becoming perturbed by spouse's behavior .323** PEI PI 27 sharing spouse's grief; feeling spouse's pain .322 ** PEI PI 16 feeling sadness at what spouse is going through .305 ** PEI FS 6 find self wanting to run away from scen e .232** PEI PI 2 thinking spouse did not deserve for this to happen .213 ** PEI FS 25 wanting to emotionally distance from scene .209** PEI FS 8 find self wanting to forget about what is happening .145* PEI FS 18 afraid I'm going to get hurt as situat ion unfolds .137* PEI PI 4 feeling helpless to relieve spouse's suffering .134* PEI PI 17 wishing a miracle would happen 0 .121 BEDI 10 find self grieved by whole situation 0 .097 BEDI 5 find self becoming alarmed about situation 0 .095 BEDI 8 find sel f feeling troubled 0 .083 BEDI 3 find self becoming disturbed 0 .077 BEDI 13 find self becoming worried 0 .062 PEI FS 11 feeling ashamed don't know how to fix situation 0 .053 PEI PI 14 find self thinking -why did this have to happen? 0 .051 PEI PI 12 wis hing situation would turn out to be a mistake 0.037 PEI PI 20 find self wishing I could turn back clock 0.020 PEI FS 22 feeling concerned about future 0.014 BEDI 1 feeling upset about what is happening 0.011 BEDI 12 find self becoming upset about whol e thing 0.010 PEI PI 24 find self having feelings of despair about situation 0.008 PEI FS 29 feeling guilty about inability to help spouse 0.005 PEI PI 7 feeling anger at cause of spouse's suffering 0.002 Note: ** Correlation is significant at the 01 level (2 tailed); Correlation is significant at the .05 level (2 Experience Inventories; PI = Passive Identification subscale; FS = False Superiority subscale; Av = Experiences in Close Relationships Questionnaire Attachment Avoidance subscale
237 Table 5 2. Distress Item Correlations with Attachment Anxiety [Ax] (in rank order from greatest to smallest) [N=234] Index Q # Distress Item Core Phrase Pearson Correlation BEDI 6 find self becoming perturbed by spouse's behavior .256** PEI FS 11 feeling ashamed don't know how to fix situation .208** PEI FS 22 feeling concerned about future .197** PEI FS 18 afraid I'm going to get hurt as situation unfolds .196** BEDI 13 find self becomi ng worried .183** BEDI 5 find self becoming alarmed about situation .181** PEI FS 6 find self wanting to run away from scene .174** BEDI 3 find self becoming disturbed .158* BEDI 8 find self feeling troubled .156* PEI PI 4 feeling helpless to relieve spouse's suffering .152* BEDI 1 feeling upset about what is happening .147* PEI FS 29 feeling guilty about inability to help spouse .146* PEI PI 24 find self having feelings of despair about situation 0.114 PEI PI 16 feeling sadness at what spouse is g oing through 0.106 BEDI 12 find self becoming upset about whole thing 0.104 PEI PI 17 wishing a miracle would happen 0.103 PEI PI 12 wishing situation would turn out to be a mistake 0.100 PEI PI 2 thinking spouse did not deserve for this to happen 0. 092 PEI FS 25 wanting to emotionally distance from scene 0.073 PEI PI 27 sharing spouse's grief; feeling spouse's pain 0.062 PEI FS 8 find self wanting to forget about what is happening 0.060 PEI PI 14 find self thinking -why did this have to happen? 0.057 BEDI 10 find self grieved by whole situation 0.039 PEI PI 20 find self wishing I could turn back clock 0.025 PEI PI 7 feeling anger at cause of spouse's suffering 0.009 Note: ** Correlation is significant at the .01 level (2 tailed); Correlat ion is significant at the .05 level (2 Experience Inventories; FS = False Superiority subscale; PI = Passive Identification subscale; Ax = Experiences in Close Relationships Questionnaire At tachment Anxiety subscale
238 Table 5 3. Distress Items Discriminating Attachment Avoidance from Anxiety (in rank order from highest discriminating to lowest discriminating ) [N=234] Index Q # Distress Item Root Av Ax Difference (Absolute Value) PEI PI 27 feeling spouse's grief and pain 0.322 0.062 0.260 PEI PI 17 wishing a miracle would happen 0.121 0.103 0.224 PEI PI 16 feeling sadness at spouse's plight 0.305 0.106 0.199 PEI FS 22 feeling concerned about future 0.014 .197** 0.183 PEI FS 11 sham e about inability to fix 0.053 .208** 0.155 PEI FS 29 guilt about inability to help 0.005 0.146 0.141 BEDI 1 upset about what is happening 0.011 .147* 0.136 PEI FS 25 wanting to emotionally distance .209** 0.073 0.136 PEI PI 24 feeling despair over sit uation 0.008 0.114 0.122 BEDI 13 find self becoming worried 0.062 .183** 0.121 PEI PI 2 thinking spouse didn't deserve this 0.213 0.092 0.121 BEDI 12 becoming upset over whole thing 0.010 0.104 0.114 PEI PI 14 wondering why this had to happen 0.05 1 0.057 0.108 BEDI 5 becoming alarmed about situation 0.095 .181** 0.086 PEI FS 8 wanting to forget what is happening .145* 0.060 0.085 BEDI 3 finding self becoming disturbed 0.077 .158* 0.081 BEDI 8 finding self feeling troubled 0.083 .156* 0.073 BED I 6 perturbed by spouse's behavior .323** .256** 0.067 PEI PI 12 wishing it would be a mistake 0.037 0.100 0.063 PEI FS 18 fearing getting hurt .137* .196** 0.059 BEDI 10 finding self grieved 0.097 0.039 0.058 PEI FS 6 wanting to run away from scene .232** .174** 0.058 PEI PI 20 wanting to turn back clock 0.020 0.025 0.045 PEI PI 4 helpless to relieve suffering .134* .152* 0.018 PEI PI 7 anger at cause of suffering 0.002 0.009 0.007 Note: ** Correlation is significant at the .01 level (2 tailed ); Correlation is significant at the .05 level (2 tailed); PEI = Pity Experience Inventories; PI = Passive Identification Indices; Av = Experiences in Close Relationships Questionnaire Attachment Avoidance subscale; Ax = Experiences in Close Relationships Questionnaire Attachment Anxiety subscale
239 APPENDIX A INSTITUTIONAL REVIEW BOARD APPROVAL FORM
240 APPENDIX B LETTER TO PASTORS: PERMISSION TO RECRUI T Dear Pastor ________________________, I am a private practice counselor who often receives marriage counseling referrals from local pastors. It has long been an interest of mine to better understand the emotional dynamics that serve to facilitate or to hinder responding empathically to ones distressed spouse. Indeed, exploring that phenomenon has become such a passion to me that gaining greater insight into that mystery has become the focus of my doctoral research. The purpose of my writing to you is to requ est your permission and support in my recruitment of couples to participate in that study. I am intending to recruit 120 couples from a total of 6 12 churches, to participate in a session that is estimated to require about 50 minutes of their time. Your help and theirs will assist me in learning more about this vital area of interest, and it will also facilitate my completion of my doctoral studies. If you feel that you are able to support me in this endeavor, I would seek to meet with you in the near fu ture to work out the details of presenting the opportunity to your congregant couples. At a minimum, the invitation to participate would be provided through a bulletin insert (prepared by me and subject to your approval), designating a room for interested couples to meet briefly after the church service. Ideally, the written invitation would be reinforced by a brief announcement from me at the service where the bulletin insert is provided. At the brief meeting after the church service, any questions can be answered, couples agreeing to participate can sign the informed consent form, and a brief information sheet can be completed by participating couples providing (among other information) contact information so that I can notify them about date and time o ptions for their participation. If possible, it would be best if we could offer two or more date and time options where your couples could participate in the 50 minute session somewhere on your church campus. If this can not be scheduled around times whe n the campus is already open, I will be happy to bear any additional utility and/or monitoring staff expenses incurred. I will be contacting you a few days after your receipt of this letter to inquire whether we might meet to answer any questions you may have in hopes of securing your support. Thank you for your willingness to consider helping with this need. Sincerely, Stephen C. Figley, L.M.F.T., L.M.H.C.
241 APPENDIX C SAMPLE CHURCH RECRUITMENT ANNOUNCMENT [FRONT] Announcing a Couples' Research Study and a Free Seminar In my practice as a marriage counselor, some client couples imagine that my own marital communication is never short of blissful. Sadly, there are times when my wife is distressed and wanting my empathy and support and I miss my cue. Sometimes, I miss it by a lot! At times, my failure is due to selfishness or being distracted. At other times, however, I genuinely want to support my wife in her distress, but my energies get used up regulating my own emotions. In my efforts t o restore my own Why this phenomenon occurs and what couples can do about this obstacle to supportive communication is the focus of my Ph.D. research study. A majority of the 1 20+ couples that I am recruiting for this study have already enrolled, and many have already completed the one hour testing. It is my hope that about twenty couples from your church, ages 20 59, will join me in this research. I will be available after th e service in the church foyer to answer any questions for couples who are interested in participating As part of my THANK YOU for your participation, I am inviting you to attend at no cost a one day seminar designed to teach and to model how to provide em pathetic and validating responses to your spouse during times of distress. Seminar attendance is not required for participation in the study, but you are invited to attend all or part of whichever seminar you choose: Nov 8 @ [Seminar at Name of Chu rch Location] [ 9 AM 4 PM ] [ UF Football Away @ Vanderbilt ] (8:00 PM Kickoff ) Nov 22 @ [Seminar at Name of Church Location] [ 9 AM 4 PM ] [ UF Football Home vs The Citadel ] (1:30 PM Kickoff)
242 APPENDIX C (CONTINUED) SAMPLE CHURC H RECRUITMENT ANNOUNCMENT [BACK] Available Times for Couples Testing October 6 Monday 7:00 PM October 10 Friday 6:00 PM October 10 Frida y 7:30 PM October 11 Saturday 10:30 AM October 11 Saturday 1:30 PM [Note: Oct 11 UF/LSU Home Kick off is at 8:00 PM] October 12 Sunday 2:00 PM @ [Name of Church] Fellowship Hall October 16 Thursday 7:00 PM October 17 Friday 6:00 PM October 17 Friday 7:30 PM October 18 Saturday 10:30 AM October 18 Saturday 1:30 PM October 18 Saturday 3:00 PM O ctober 18 Saturday 4:30 PM [Note: There is NO UF football game on October 18] Testing at My Office Except Oct 12 th You may choose the on site opportunity provided on October 12 th or you may schedule any one of the other opportuniti es provided at my office [4001 Newberry Road, Suite C 4] for your one hour participation. Multiple couples will be tested at one time. We reserve the right to reschedule a session with 72 hour advance notice to you if fewer than three couples have signe d up for it.
243 APPENDIX D INFORMED CONSENT FOR M [FRONT] Dear Prospective Study Participants: I am a graduate student in the Department of Counselor Education at the University of Florida, conducting research for my doctoral dissertation under the s upervision of Dr. Harry Daniels. The purpose of this study is to examine how the availability of memories of relationships with parents and close others affects the way couples respond to one another during times of distress. If you agree to participat e in this study, we will use the contact information you provide to notify you of the times that the study is offered at your church. We estimate that your participation in the study session itself will require a little less than 60 minutes. As part of t he study, you will be asked to perform a series of memory exercises, including thinking about relationships from the past and remembering a time when your spouse needed your support. You will also be asked to respond to three surveys measuring how you exp erience relationships in general, and how you feel when your spouse needs your support. There are no physical risks involved in your participation in the study. The memory exercises may result in some participants experiencing uncomfortable emotions fo r a time. But it is expected that such feelings will be reduced by the end of the study session. The person administering the study session has worked for more than twenty years in vocational church ministry settings. She will be available after the ses sion to help you talk through any uncomfortable feelings that may remain. You will not be asked to interact directly with your spouse as part of this study, but voluntary participation by both the husband and wife is essential to achieve the goals of the study. Neither you nor your spouse will be required to participate in any activity that you wish to omit, nor to answer any question that you wish to skip. Either or both of you are free to withdraw from the study at any time without consequence. You wi ll not receive any direct benefits by your participation in this study. But it is hoped that your participation will benefit those who seek ways to teach couples how to communicate more effectively. As a way of my thanking each couple who volunteers to p articipate, you will receive a free copy of a book that was written to help couples improve their communication, and you will be given the opportunity to attend at no cost to you a one day marriage communication seminar that I will present after study comp letion. Your identity will be protected by the assignment of a confidential identifier, and only I will have access to the master list linking names with participant identifiers. Once the data is analyzed, the master list will be destroyed. Your identit y will be kept confidential to the extent provided by law, and your identity will not be revealed in the dissertation provided to the graduate school, in any later publication of results provided to research journals, or in any other context or setting. N either your spouse nor any other participant will see your responses.
244 APPENDIX D (CONTINUED) INFORMED CONSENT FORM [BACK] If you have any questions about this research study, feel free to contact me by e mail at figle firstname.lastname@example.org or at my counseling office at (352) 380 0209, or you may contact my faculty supervisor, Dr. Harry Daniels, at (352) 392 0731 (ext 226). Questions or concerns about your rights as a research participant may be directed to the IRB02 Office, Un iversity of Florida, Box 112250, Gainesville, FL 32611 2250; (352) 392 0433. Thank You, Stephen C. Figley, M.Ed., Ed.S. Agreement: We have read and understood the information provided above. By signing our names below, we voluntarily agree to partic ipate in the study and we acknowledge that we have received a copy of this consent document. _________________________________ _________________________________ Date Date _________________________________ _______ __________________________ _________________________________ ______ _____ Month Day Year How may we contact you to schedule your participation and to notify you of the free seminar? E mail: ______________________________ E mail: ______________________________ Phone: (_____) _____ ________ H W C Phone: (_____) _____ _______ H W C
245 APPENDIX E PARTICIPANT TASK #1 (ECR R AV) Participant Task #1 Experiences in Close Relationships Revised Adapted Version The following statements concern your experiences in close relationships. The majority of the questions will inquire about your experiences with your current marriage partner. Some of the others will ask about your various experiences in romantic relationsh ips throughout adulthood. And the rest relate to your experiences in close relationships in general. Please respond to each statement by indicating how much you agree or disagree with it. Write the number in the space provided, using the following ratin g scale: 1 2 3 4 5 6 7 Disagree Strongly Disagree Disagree Slightly Neutral/ Mixed Agree Slightly Agree Agree Strongly ____ 1. My experience has been to feel uncomfortable opening up to romantic partners. ____ 2. I worry a lot about my relationships. ____ 3. I usually discuss my problems and concerns with my husband/wife. ____ 4. My experience has been to feel comfortable depending on romantic partners. ____ 5. I get uncomfortable whe n my wife/husband wants to be very close. ____ 6. I rarely worry about my husband/wife leaving me. ____ 7. My wife/husband makes me doubt myself. ____ 9. My experience has been to be ve ry comfortable being close to romantic partners. ____ 10. I often worry that my husband/wife will not want to stay with me. ____ 11. My desire to be very close sometimes scares people away. ____ 12. I talk things over with my wife/husband. ____ 14. My experience has been to prefer not to be too close to romantic partners. ____ 15. It helps to turn to my wife/husband in times of need. ____ 16. I prefer not to show my husband/wife how I fe el deep down. ____ 17. My experience has been to find it difficult to let myself depend on romantic partners. Go On To Next Page >>>
246 APPENDIX E (CONTINUED) PARTICIPANT TASK #1 (ECR R AV) 1 2 3 4 5 6 7 Disagree Strongly Disagree Disagree Slightly Neutral/ Mixed Agree Slightly Agree Agree Strongly ____ 19. I tell my husband/wife just about everything. ____ 20. My experience h as been to become nervous when romantic partners have gotten too close to me. ____ 22. I do not often worry about being abandoned. ____ 23. My husband/wife only seems to notice me ____ 24. My experience has been that when I show my feelings for romantic partners, I become afraid that they will not feel the same about me. ____ 26. It makes me husband/wife. ____ 27. When my wife/husband is out of sight, I worry that she/he might become interested in someone else. ____ 28. My experience has been to be afraid that once a romanti c partner gets to know me, he ____ 29. My husband/wife really understands me and my needs. ____ 30. I find it relatively easy to get close to my wife/husband. gs for me were as strong as my feelings for him/her. ____ 32. My experience has been to find it easy to depend on romantic partners. ____ 33. My experience has been that sometimes romantic partners have changed their feelings about me for no appare nt reason. as I care about them. ____ 35. I feel comfortable sharing my private thoughts and feelings with my wife/husband. ____ 36. My experience has bee n that my romantic partners have tended not to want to get as close as I would like.
247 APPENDIX F PARTICIPANT TASK #2 (DISTRACTER TASK) Participant Task #2 Memory Warm Up Exercises The following questions are designed to help you begin to exercise your long term memory, so that you can begin to practice this skill before you are called on to use it for later tasks. For each question below, please write your answer in the space to the right. Then, next to your answer, write the number (fro m the scale below) that reflects how sure you are of the answer you gave. 1 2 3 4 5 Just a Guess I Think Maybe Somewhat Sure Pretty Sure Absolutely Sure Question Answer How Sure Are You? What was the name of your first grade teacher? Who was the first person you remember What was your hourly wage in your first job working for a regular bu siness? Where were you when the clock struck midnight and Y2K arrived (01 01 00)? At what restaurant did you last order an entre costing more than $25?
248 APPENDIX G PARTICIPANT TASK #3 (MEMORY SELECTION TA SK) Participant Task #3 Memory Selec tion Task Box #1 Task Overview Stored within our memories are events where we have observed other people upset or in distress and needing someone to reach out to them with support. When we are not in close relationship with the one in distress, we may feel more freedom to choose whether or how to respond. If the one in distress is our spouse, however, the expectation that we respond in supportive ways is usually much higher. One of the aims of this study is to better understand the feelings, th oughts, and wishes people experience when their marriage partner is upset or in distress and wanting their support. To help us accomplish this aim, we will be asking you to go back into your memory and bring up a time when you experienced this. You will not have to tell us what happened. Box #2 Zeroing in on a Memory Times of distress are a normal part of living. So you can probably think of quite a few times when your spouse was in distress or really upset and wanted your support. Because we are interested in a certain type of event where this happened though, we ask you to use the following guidelines when you choose an event to focus on. ( 1 ) where your spouse was visibl y upset or shaken or agitated by something that had happened. It will be much easier to measure your responses to something that was A BIG DEAL! ( 2 ) Please choose an event where you were not primarily to blame for what had happened that caused your spou upset about, we usually feel defensive and it is harder to respond in supportive ways. ( 3 ) Please choose an event that did not impact you directly as much as it impacted your distr essed spouse If your spouse lost his or her job, and your thoughts became too focused on ( 4 ) Please choose an event that did not result in a grand once an d for all resolution We what you were feeling, thinking, and wishing as your spouse was in distress. ( 5 ) Please choose an event that you could imagine happeni ng again in the future Not that it could happen again the exact same way it happened before. But that it would not be completely surprising for your spouse to become distressed again over similar kinds of happenings. Please take a few moments to sort t hrough your memory to find an event that fits these five guidelines as closely as possible. When you feel that you have settled on a memory that fits the guidelines well, please write a brief phrase in the blank below that will serve to remind you of the memory you have chosen. We will ask you to come back to this memory in later tasks. ________________________________________ ____________
249 APPENDIX H PARTICIPANT TASK #4 (GROUP 1) Participant Task #4 Relationship Memory Task (Group 1) Box #1 Beginni ng Instructions In this task, we want you to take a few moments to think about a relationship you have had in which you found it relatively easy to be emotionally close to the other person. In this relationship, you felt comfortable depending on thi s person and by this person or about this person wanting to be closer than you wanted her or him to be. In this g alone or about this person not accepting you. It is important that the relationship you choose to think about be with someone who had (or has) a fairly significant role in your life. This person can be someone from your past or someone with whom you ar e currently in relationship. It can be a parent or parent substitute, a grandparent or other relative, a teacher or coach, a close friend or confidante, or someone else who fits this description. If your spouse fits this description but you are able also to think of someone else who does, it is preferred for this assignment that you reflect on the other relationship. Box #2 Zeroing in on a Person/Relationship to Think About As you narrow your choice to one relationship that best fits the descrip tion above, please use the blank line below to ..??? ___________________________ __________________ Box #3 Fully Entering the Memory Now take a few moments and try to get a visual image in your mind of this person. How are you doing at returning to that place in your memory? Think about a time when you were interacting with this person and experiencing the is it like being with this person? What would he or she do or say to you? What would you do or say in return? How do you feel when you are with this person? How would you feel if this person were here with you now? Stay with the memory of this person for a few moments longer. Box #4 Relationship Memory Reflection Feedback Please write a sentence or two about what you were thinking and feeling about yourself in relation to the person you thought about: _____________________________________________ _________________________ _______ ___________________________________________________________ __________________ _____________________________________________________________________________ _____________________________________________________________________________ How clear and vivid was your memory of being with this person? Please circle a number 1 2 3 4 5 6 7 Unable to Recall Rather Faint Mildly Clear Fairly Clear Rather Clear Very Clear It Was Like Reliving It
250 APPENDIX I PARTICIPANT TASK #4 (GROUP 2) Participant Task #4 R elationship Memory Task (Group 2 ) Box #1 Beginning I nstructions In this task, we want you to take a few moments to think about a relationship you have had in which you felt somewhat uncomfortable being emotionally close to the other person. In this relationship you found it difficult to trust this pe rson completely, or you felt it was very important to be independent and self sufficient and not depend on this person or have this person depend on you. In this relationship you felt this person wanted to be closer than you did, and his or her eff orts to get closer made you feel nervous. It is important that the relationship you choose to think about be with someone who had (or has) a fairly significant role in your life. This person can be someone from your past or someone with whom you are currently i n relationship. It can be a parent or parent substitute, a grandparent or other relat ive, a teacher or coach, a former friend or romantic partner or someone else who fits this description. If your spouse fits this description but you are able also to th ink of someone else who does, it is preferred for this assignment that you reflect on the other relationship. Box #2 Zeroing in on a Person/Relationship to Think About As you narrow your choice to one relationship that best fits the description a bove, please use the blank line below to ...??? ___________________________ __________________ Box #3 Fully Entering the Memory Now take a few moments and try to get a visual image in your mind of this person. How are you doing at returning to that place in your memory? Think about a time w hen you were interacting with this person and experiencing the is it like being with this person? What would he or she do or say to you? What would you do or say in return? How do you feel when you are with this person? How would you feel if this person were here with you now? Stay with the memory of this person for a few moments longer. Box #4 Relationship Memory Reflection Feedback Please write a senten ce or two about what you were thinking and feeling about yourself in relation to the person you thought about: _____________________________________________ _________________________ _______ _________________________________________________________________ ____________ _____________________________________________________________________________ _____________________________________________________________________________ How clear and vivid was your memory of being with this person? Please circle a numbe r 1 2 3 4 5 6 7 Unable to Recall Rather Faint Mildly Clear Fairly Clear Rather Clear Very Clear It Was Like Reliving It
251 APPENDIX J PARTICIPANT TASK #4 (GROUP 3) Participant Task #4 R elationship Memory Task (Group 3 ) Box #1 Beginning Instruc tions In this task, we want you to take a few moments to think about a relationship you have had in which you felt like the other person was reluctant to get as close as you would have liked. In this relationship you worried that the other person di person, but you worried that if you tried to get closer you might scare this person away. It is important that the relationship you choose to think about be with someone who had (or has) a fairly significant role in your life. This person can be someone from your past or someone with whom you are curr ently in relationship. It can be a parent or parent substitute, a grandparent or ot her relative, a teacher or other mentor a close friend or confidante, or someone else who fits this description. If your spouse fits this description but you are able als o to think of someone else who does, it is preferred for this assignment that you reflect on the other relationship. Box #2 Zeroing in on a Person/Relationship to Think About As you narrow your choice to one relationship that best fits the descri ption above, please use the blank line below to ...??? ___________________________ __________________ Box #3 Fully Entering the Memory Now take a few moments and try to get a visual image in your mind of this person. How are you doing at returning to that place in your memory? Think about a time when you were interacting with this person and experiencing the is it like being with this person? What would he or she do or say to you? What would you do or say in return? How do you feel when you are with this person? How would you feel if this person were here with you now? Stay with the memory of this person for a few moments longer. Box #4 Relationship Memory Reflection Feedback Please write a sentence or two about what you were thinking and feeling about yourself in relation to the person you thought about: _____________________________________________ _________________________ _______ __________________________________________________________ ___________________ _____________________________________________________________________________ _____________________________________________________________________________ How clear and vivid was your memory of being with this person? Please circle a number 1 2 3 4 5 6 7 Unable to Recall Rather Faint Mildly Clear Fairly Clear Rather Clear Very Clear It Was Like Reliving It
252 APPENDIX K PARTICIPANT TASK #4 (GROUP 4) Participant Task #4 Acquaintance Memory Task (Group 4 ) Box #1 Beginning Instructions In this task, we want you to take a few moments to think about a casual acquaintance whom you have known or currently know, who is of above average intelligence, shows good people seem to mind being in the spotlight. The casual acquaintance that we want you to think about is one who you could imagine enjoying and being successful at being the mayor of a town or small city (5,000 to 25,000 people ). It is important that the acquaintance you choose to think about be someone you have observed enough to have an idea what she or he is like, but also someone you have not been particularly close to. This person should not be a relative or someone you have lived with. This person should not be someone you were ever romantically drawn to or someone you have been in serious conflict with. This person can be someone from your past or someone you currently know. He or she can be someone you knew or know from school, work, church, or community. Box #2 Zeroing in on the Acquaintance to Think About As you think about people you know casually who you could imagine serving as mayor of a town or small city, try to narrow your choice to one acquaintance who you think best demonstrates the skills an d qualities listed in Box #1 above. Then use the blank line below to tell us who you chose (for example: : ______________________________________________ Box #3 Fully Entering the Imagined Scenario Now take a few moments and try to get a visual image in your mind of this person serving in the role of mayor of a town or small her or him meeting with city workers to resolve grievances. Picture him or her fielding complaints from citizens who are upset about a broken water line or a pothole in their road. Imagine this person shaking hands and kissing babies while campaigning for re election. How are you doing at re turning to that place in your memory and imagining this scenario? Box #4 Imagination Exercise Feedback Please write a sentence or two about what it was like to imagine this person in this role? How mentally taxing was this exercise for you ? __ ___________________________________________ _______________________________ _____________________________________________________________________________ _____________________________________________________________________________ _________________________ ____________________________________________________ How clear and vivid was your imagination of this scenario ? Please circle a number 1 2 3 4 5 6 7 Unable to Imagine Rather Faint Mildly Clear Fairly Clear Rather Clear Very Clear It Was Like L iving It
253 APPENDIX L PARTICIPANT TASK #5 (MEMORY REVISITING T ASK) Participant Task #5 Memory Revisiting Task Box #1 Returning to Event Memory Chosen in Task # 3 In this exercise, we would like to test an aspect of memory that we refer to as relations hip emotional memory. To do this, we need you to return to the memory you chose earlier about a time when your wife or husband was in distress and wanted your support. It is important that you be able to imagine yourself back in that situation so that yo u can emotionally tune in to what you were feeling, thinking, and wanting to do as it was happening. Try to remember what things were like when your spouse was so upset. Think about the things moments. How vivid was your memory of the distressing ev ent you re called? (circle appropriate #) 1 2 3 4 5 6 7 Unable to Recall Rather Faint Somewhat Clear Fairly Clear Quite Clear Very Vivid It Was Like Rel iving It ate #) 1 2 3 4 5 6 7 Unable to Recall Fairly Mild Starting to Rise Moderately High Quite High Very High Extremely High Box #2 Imagining a Similar Event Happening Again The final memory function we would like to test is known as visualization or imagination. Visualization and imagination are actually memory based activities that permit us to explore future possibilities based on our happening and thinki ng about how we might respond or react based on similar experiences we have had in the past. To t est this memory function, we would like you to imagine that the event you have been remembering (or something quite similar) is happening again. Tak e a few moments to see if you can imagine something similar occurring all over again. Try to observe every detail. Notice are feeling and thinking. Notice h ow you are wanting to respond to your mate. As challenging as this task is, see if you can stay with this picture and these feelings for a few moments. How vivid was your imagined recurrence of the distressing event? (circle appropriate #) 1 2 3 4 5 6 7 Unable to Imagine Rather Faint Somewhat Clear Fairly Clear Quite Clear Very Vivid It Was Like Rel iving It
254 APPENDIX M PARTICIPANT TASK #6 (BEDI AV ) Participant Task #6 Participant F eedback Ratings Page One We would like for you to continue to th ink about the distressing event with your husband or wife that you are imagining is happening again. As you continue focusing on this, we are interested in what you are feeling. For each of the sentences below, please place the number that best indicates how much you are feeling the emotion expressed. A rating of that emotion at an extremely high level. 1 2 3 4 5 6 7 Not at all Slightly Somewhat A fair amount Quite a bit Very much Extremely As I continue to focus on the event I imagine is happening again ____ 1. I find myself feeling upset about what is happening. ____ 2. I find myself feeling compassionate toward my husband or wife. ____ 3. I find myself becoming disturbed. ____ 4. I find myself feeling warm toward my wife or husband. ____ 5. I find myself feeling alarmed by the situation. ____ 6. I find myself becoming perturbed by the way my husband or wife is behaving. ____ 7. I find myself feeling soft hearted toward my wife or husband. ____ 8. I find myself feeling troubled. ____ 9. I find myself having tender feelings toward my husband or wife. ____ 10. I find myself grieved by the whole situation. ____ 11. I find myself feeling sympathetic toward my wife or husband. ____ 12. I find myself becoming distressed by the whole thing. ____ 13. I find myself becoming worried. How are you doing at stayin distress? Please circle the number below that shows how clear the scenario is now: 1 2 3 4 5 6 7 Unable to Experience Rather Faint Somewhat Clear Fairly Clear Quite Clear Very Vivid It is Like Rel iv ing It
255 APPENDIX N PARTICIPANT TASK #6 (PEI SF R ) Participant Task #6 Participant F eedback Ratings Page Two We know that this is no fun, but it is important that you continue to focus on the event you imagined, similar to the one experienced earlier. expressions and body language? Can you put yourself fully in that imagined event? As you try to stay in that place, please continue rating each item using the following scale. 1 2 3 4 5 6 7 Not at all Slig htly Somewhat A fair amount Quite a bit Very much Extremely As I continue to focus on the event I imagine is happening again ____ 1. I find myself feeling closeness and warmth for my wife or husband. ____ 2. I find myself thinking that my husban d or wife did not deserve for this to happen. ____ 3. I find myself wanting to better understand what my wife or husband is feeling. ____ 4. I find myself feeling helplessness about how to relieve his or her suffering. ____ 5. I find myself wantin g to hug and caress my wife or husband. ____ 6. I find myself wanting to run away from the scene. ____ 7. I find myself feeling anger at the people or circumstances that caused this situation. ____ 8. I find myself wanting to forget about all thi s that is happening. ____ 9. I find myself thinking about how to help my husband or wife to feel better. ____ 10. I find myself wanting to give comfort to my wife or husband. tuation. ____ 12. I find myself wishing that the situation would turn out to be a mistake. ____ 13. I find myself having feelings of love for my husband or wife. ____ 14. I find myself thinking to myself why did this situation have to happen? ____ 15. I find myself thinking about how to encourage my wife or husband. Please circle the number below that shows how clear the scenario remains in your mind: 1 2 3 4 5 6 7 Unable to Experience Rather Faint Somewhat Clear Fairly Clear Quite Clear Very Vivid It is Like Rel iving It
256 APPENDIX N (CONTINUED) PARTICIPANT TASK #6 (PEI SF R) Participant Task #6 Participant F eedback Ratings Page Three Thanks for staying with us so far this is your last page of responses where it is necessary to stay focused on the event you imagined. Can you still picture this distressing scene? What is your spouse feeling, saying, and doing? As you try to stay in that place, please continu e rating each item using the following scale. 1 2 3 4 5 6 7 Not at all Slightly Somewhat A fair amount Quite a bit Very much Extremely As I continue to focus on the event I imagine is happening again ____ 16. I find myself feeling sadness as I see what my husband or wife is having to go through. ____ 17. I find myself wishing a miracle would happen. ____ 18. I find myself afraid that I am going to get hurt as this situation unfolds. ____ 19. I find myself wanting to get close to my wife or husband. ____ 20. I find myself wishing I could turn back the clock. ____ 21. I find myself thinking to myself that I need to try to comfort my husband or wife. ____ 22. I find myself feeling concerned about the future. ____ 23. I find myself wan ting to talk to my wife or husband about her or his distress. ____ 24. I find myself having feelings of despair about this situation. ____ 25. I find myself wanting to emotionally distance from the scene. ____ 26. I find myself thinking about ways to ch ange the situation. feeling her or his pain. ____ 28. I find myself wanting to calm my wife or husband. ____ 29. I find myself feeling guilty about not being able to help my husband or wife. __ __ 30. I find myself wanting to encourage my wife or husband and to cheer up him or her. Please circle the number below that shows how clear the scenario remained in yo ur mind: 1 2 3 4 5 6 7 Unable to Experience Rather Faint Somewhat Clear Fairly Clear Quite Clear Very Vivid It is Like Rel iving It
257 APPENDIX O PARTICIPANT TASK #7 (RCHS) Participant Task #7 Relationship Communication History Survey Instructions: Tha nk you for staying with us until this the final task. Telling us something about your experiences communicating with your spouse as well as with earlier romantic relationship partners will add to the benefit of our findings. The questions on the front of this page relate to how well you and your marriage partner respond The questions on the back of this page are exploring the same topic but are inquiring about sign ificant relationships you may have experienced before your current marriage. In responding to the questions on the back of this page, if you had more than three earlier romantic relationships (marriages or exclusive relationships of a year or longer), ch oose the three that had the biggest impact on you. If you had fewer than three, leave the excess boxes blank. Remember that your responses will be kept confidential For each question below, please circle the number (1 7) to the right that best rep resents your experiences with your current husband or wife How often do you find yourself upset or distressed and wanting/needing your spouse to calm, sooth, comfort, or in some other way support you? 1 2 3 4 5 6 7 Hardly On Quite Almost Ever Occasion a Bit Daily How effective is your spouse at responding in the ways that you want or need during such times? 1 2 3 4 5 6 7 Not Making Often Right on at All an Effort Helpful Target Ho w often does it seem that your spouse is upset or distressed and wanting/needing you to calm, sooth, comfort or in some other way support him or her? 1 2 3 4 5 6 7 Hardly On Quite Almost Ever Occasion a Bit Daily How effective do you feel like you are at responding in the ways that your spouse wants and needs during such times? 1 2 3 4 5 6 7 Not Making Often Right on at All an Effort Helpful Target Go On To Next Page >>>
258 APPENDIX O (CONTINUED) PARTICIPANT TASK #7 (RCHS) Instructions: For each box below please tell us the type of relationship and its length in years. Please circle the number at the right of each question that best reflects your experience. Most significant earlier partner relationship / we were together for ______ years. ____ married ____ living together but not married ____ exclusive dating relationship In this relationship, how effective was your partner at responding in the wa ys that you wanted or needed when you were upset or distressed? 1 2 3 4 5 6 7 Not Making Often Right on at All an Effort Helpful Target In this relationship, how effective do you feel like you were at responding in the ways that your partner wanted or needed when he or she was upset or distressed? 1 2 3 4 5 6 7 Not Making Often Right on at All an Effort Helpful Target Next most significant earlier partner relati onship / we were together for ______ years. ____ married ____ living together but not married ____ exclusive dating relationship In this relationship, how effective was your partner at responding in the ways that you wanted or needed when you w ere upset or distressed? 1 2 3 4 5 6 7 Not Making Often Right on at All an Effort Helpful Target In this relationship, how effective do you feel like you were at responding in the ways that your partner wanted or needed when he or she was upset or distressed? 1 2 3 4 5 6 7 Not Making Often Right on at All an Effort Helpful Target Next most significant earlier partner relationship / we were together for ______ years. ____ married ____ li ving together but not married ____ exclusive dating relationship In this relationship, how effective was your partner at responding in the ways that you wanted or needed when you were upset or distressed? 1 2 3 4 5 6 7 Not Making Often Right on at All an Effort Helpful Target In this relationship, how effective do you feel like you were at res ponding in the ways that your partner wanted or needed when he or she was upset or distressed? 1 2 3 4 5 6 7 Not Making Often Right on at All an Effort Helpful Target
259 APPENDIX P EXPERIMENTAL PROTOCOL ADMINISTRAT OR SCRIPT [7 PAGES] statements in directing th e participant group session. Instructions in bold italics are intended to be read to the participant group verbatim Before the Session Configure chairs in a manner that ensures adequate personal space [Note: Participants will be seated in chairs and us ing clipboards as the writing surface] Greet participants as they arrive providing each with the test packet bearing her or his name, a clipboard, and a pen Advise participants not to open packets until instructed; share that they may sit together until it is time to begin after which you will be required to separate them Accommodate unscheduled participants who have earlier enrolled, by retrieving their packets grouped by church [Note: Among the make up session options for those who have missed their own c site testing time(s) is for them to join participants of other churches at their on site testing; to facilitate this, the packets of all enrollees absent site testing are taken to each subsequent testing opportunity w hether such enrollees confirmed attendance at that testing session or not] Complete any Informed Consent matters not already resolved [For Example: One spouse not available for sign up during recruitment; couple learned of the study through someone else in their church, were sent an e mail copy of the informed consent, and need to execute the document itself prior to testing] Shortly before beginning the testing procedure itself, guide participants in selecting seating that precludes tactile contact with sp ouse and limits visual contact Beginning the Session Begin the session when all scheduled participants have arrived and are seated, and the announced starting time has arrived; if awaiting later arrivers, closely monitor participant restlessness before p ermitting a start time delay; in any event, do not permit a start time delay of more than ten minutes Begin the session by reading the following, verbatim: Pardon the formality, but I am required to begin the session by reading to you the following script: Welcome! Thank you for the sacrifice of time you have made to participate in this study. We apologize that we found it necessary to seat you separate from your husband or wife. One of the assignments during this hour requires that you retrieve a me mory of a time of interacting with your spouse, and we thought it might be easier to focus on that memory if you were sitting alone.
260 APPENDIX P (CONTINUED) [PAGE 2 OF 7] In a moment, I am going to ask you to open the envelope bearing your name, and to remove a packet of worksheets. From your completion of these worksheets, we hope to learn something about your experiences in close relationships including your relationship with your wife or husband. The more open and honest you can be in your response s, the more valuable your participation will be to our study. After you have finished your participation today, feel free to share with your spouse whatever you want of your responses. But please know that your spouse will not see the responses on your worksheets, I will not see your responses, and by the time anyone reviews your responses, your worksheets will be identified only by a code and not by your name. So you are safe in being completely honest. Make certain the envelope in front of you bears your name on the outside. You may now open your envelope and remove the packet of worksheets placing it face down on your clipboard. Task #1 Introduce this task by reading the following script: Items on the first two pages inquire about w hat you tend to feel or experience in close relationships. These relationships include but are not limited to -your relationship with your spouse. After reading the instructions, you will be asked to rate your level of agreement with each of 36 statement s. Do not labor very long over any one item. Most people have been able to complete these items in about 7 8 minutes. After you have completed the first page, move on to the second page. Feel free to raise your hand if you are confused about anything y ou are asked to do. You may now turn your packet over and begin Task #1. At the 4 minute mark, provide the following prompt: If you are not on the second page by now, you might consider not lingering so long over each item. At the 6 minute mark, provide the following prompt: A couple more minutes. At the 7 minute mark, provide the following prompt: One more minute if you need it. If everyone is done before the 8 minute mark, you may end this task. If people are still working, continue until the 8 mi nute mark and then announce: If you were unable to complete every item, the responses you provided will still be tabulated in our study. At attention before going on to the ne xt task.
261 APPENDIX P (CONTINUED) [PAGE 3 OF 7] Task #2 Introduce this task by reading the following script: This next task is a brief one, and is designed to help you begin to exercise your long term memory. Please read the instructions and very quick ly provide your best recollection to the five questions below rating how confident you are of your memory. You will be given 3 4 minutes for this task. You may now begin Task #2. At the 3 minute mark, if anyone appears still to be writing, provide the f ollowing prompt: If you are not finished, take another half minute or so. At the 4 minute mark, provide the following prompt: to turn your packet face down, and then turn your attention to me as I introduce the next ta sk. Task #3 Introduce this task by reading the following script: In the next task, you will be asked to retrieve a more personal memory a time when your wife or husband was upset and looking to you for support. After reading the overview of the assi gnment, pay particular attention to the guidelines that help you to choose one specific memory that will be most relevant to our research. We will take 6 7 minutes for this task. You may now begin work on Task #3. At the 4 minute mark, provide the follo wing prompt: By this time, you should be mulling over episodes in your memory that might fit the descriptions provided. It is OK if you cannot retrieve a memory that completely fits every guideline provided It is very important, however, that the memor y you select be one where the distress level of your husband or wife was quite high and your support was wanted or needed. At the 6 minute mark, provide the following prompt: need to ask you to take the next half minute or so to finalize your choice of one specific memory, and to write a phrase on the line at the bottom that will remind you of the memory you have chosen. You will be asked to come back to that memory in a later task. At the 7 minute mark, provide the fo llowing prompt: to turn your packet face down, and then turn your attention to me as I introduce the next task.
262 APPENDIX P (CONTINUED) [PAGE 4 OF 7] Task #4 Introduce this task by reading the following script: In a moment, I am going to ask you to turn to the next page, where you will be asked to exercise your memory in yet another way. Because different ones of you have been assigned different versions of Task #4, I will be limited in how specific I can be in my verbal instructions as you work through this task. So carefully follow the instructions in each of the boxes and feel free to raise your hand if any part of the assignment confuses you. You will have about 6 7 minutes to complete this task. You may now turn your packet over and begin Task #4. At the 4 minute mark, provide the following prompt: A couple more minutes. At the 6 minute mark, provide the following prompt: is important that you move on to the last box and provide the two areas of feedback about your experience in this task. At the 7 minute mark, ask the following question: Is there anyone who has not yet circled a number on the scale at the bottom of the page? When you are confident that everyone h as provided this item of feedback, make the following statement: going to ask you to turn your packet face down, and then turn your attention to me as I introduce the next task. Task #5 Introduce Task #5 with the following script: The ne xt task requires that you return to the earlier memory when your spouse was upset and needed your support. We ask that you read the instructions and then follow the prompts designed to help you vividly recall that memory. When you are done with the first box, move on to the second one. Although this may be an unpleasant task, your putting yourself fully into this memory will be of real benefit to the study results. We have allotted about 7 8 minutes for this portion of the study. You may now turn your packet over and begin Task #5. At the 4 minute mark, provide the following prompt: way point of our time. At the 7 minute mark, provide the following prompt: you have not yet completed all of the three scales, please take a moment to finish up and give us this feedback. When you are completed, please put your pen down and look up at me.
263 APPENDIX P (CONTINUED) [PAGE 5 OF 7] Task #6 When everyone has looked up indicating that all are d one Introduce Task #6 with the following script: In the next part of our study, we want you to report on your experience in the memory task you just completed. For your feedback to be effective, it is important that you maintain your focus on the episode you have been thinking about. To remind you of this, each of the feedback pages ends by asking you to report on how well you are maintaining that focus. Spend only a few seconds on each item or about three minutes per page. You may now turn to Task #6, read the instructions, and begin rating the items on the three pages. At the 3 minute mark, provide the following prompt: You should be finishing up with the first page by now make sure that you complete the focus scale at the bottom and move on to the next page. At the 6 minute mark, provide the following prompt: You should be finishing up with the second page by now make sure that you complete the focus scale at the bottom and move on to the next page. At the 9 minute mark, provide the following pro mpt: You should be finishing up with the third page by now make sure that you complete the focus scale at the bottom when you are done. At the 10 minute mark, provide the following prompt: Even if you were unable to complete all of the items, please tak e a moment to respond to the focus scale on the bottom of the third page then set your pen down and look up at me. Task #7 When everyone has looked up indicating that all are done Introduce Task #7 with the following script: thanks f or staying with us to the end! The last task is a two page survey. On the first page, we ask you for brief feedback about an aspect of your communication with your wife or husband. On the second page, we ask for similar feedback on communication pattern s in any previous significant romantic relationships you may have had before your current marriage. This brief survey should take only 5 6 minutes. If anything is confusing to you, please raise your hand. You may now turn to Task #7 and complete the two pages. At the 3 minute mark, provide the following prompt: About three more minutes. At the 5 minute mark, provide the following prompt: your pen down and look up. We will give just a moment for any who may not be fini shed.
264 APPENDIX P (CONTINUED) [PAGE 6 OF 7] At the 6 minute mark, provide the following prompt: At this time, please return your packet to the envelope bearing your name. When you have done so, please give me your attention for some closing instructions Emotional Equilibrium Restoring Activity The following script should be read: For some of you, this may have been a rather unpleasant experience of having to focus on difficult memories. To help ensure that these memories are less likely to intrude upon your thoughts as you leave here, we want to take a couple moments to shift our focus to something more comforting. when I am feeling distressed and need to shif t my focus to God and the comfort he gives: those who are crushed in spirit. A righteous [person] may have many troubles, but the LORD delivers [them] from them all. [Psalms 34:18,19] tly for the LORD; he turned to me and heard my cry. He lifted me out of the slimy pit, out of the mud and mire; he set my feet on a rock; and gave me a firm place to [Psalms 40:1, 2] He alone is my rock and my salvation; he is my fortress, I will never be shaken. Trust in him at all times, O people; pour [Psalms 62:1, 2, 8] will restore my life again; from the depths of the earth you will again bring me up. You will increase my honor and comfort [Psalms 71:20, 21] He stilled the storm to a whisper; the waves of the sea were hushed. They were glad when it [Psalms 107:28 30] If you continue to feel distressed about what you have thought about in this session, and it would help you to stay behind and debrief with me or to have me pray with you -I would welcome you to do that. If this experience has surfaced issues that are troubling to you and you would like my feedback in helping you to find a pastoral or pro fessional counselor or a class where you might explore these issues further I will be happy to assist you in this way.
265 APPENDIX P (CONTINUED) [PAGE 7 OF 7] Please remember to select your free book one per couple before you leave. Amazon reviews are ava ilable for your perusal, and I am happy to provide you feedback about my opinion of each book. We encourage you to let us know by e mail whether you plan to attend the free seminar, and whether you anticipate attending the one November 8 th at The Family C hurch, or the one November 22 nd at Grace at Fort Clarke. We ask that you avoid talking about this session with couples who have not yet participated. After the Session Participant packet envelopes will be collected by the s ession administrator as participants are leaving The session administrator will make herself available for approach by any participants who may want additional help as offered at the close of the emotional equilibrium restoring activity The session adminis trator will ensure that all materials are retrieved and organized for departure The session administrator will deliver the packets and other materials to the principal investigator, along with verbal and written feedback concerning any unique happenings in the testing session itself that may affect participant data
266 LIST OF REFERENCES Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale, NJ: Erlbaum. Alexandro v, E. O., Cowan, P. A., & Cowan, C. P. (2005). Couple attachment and the quality of marital relationships: Method and concept in the validation of the new couple attachment interview and coding system. Attachment and Human Development, 7 123 152. Allen, J. P., & Land, D. (1999). Attachment in adolescence. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 319 335). New York: Guilford Press. Allen, P. A., Kaut, K. P., & Lord, R. R. (2008). Emoti on and episodic memory. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 115 132). Amsterdam: Elsevier. Anderson, J. R., & Schooler, L. J. (2000). The adaptive nature of memory In E. Tulving & F. I. M. Craik (Eds. ), The Oxford handbook of memory (pp. 557 570 ). New York: Oxford University Press. Atance, C. M. (2008). From the past into the future: The developmental origins and trajectory of episodic future thinking. In E. Dere, A. Easton, L. Nadel, & J. P. Huston ( Eds.), Handbook of Episodic Memory (pp. 99 114). Amsterdam: Elsevier. Babcock, J. C., Jacobson, N. S., Gottman, J. M., & Yerrington, T. P. (2000). Attachment, emotional regulation, and the function of marital violence: Differences between secure, preoccu pied, and dismissing violent and nonviolent husbands. Journal of Family Violence, 15 391 409. Bakermans K ranenburg, M. J., & van IJzendoorn, M. H. (2009). The first 10,000 Adult Attachment Interviews: Distributions of adult attachment representations in clinical and non clinical groups. Attachment & Human Development, 11 223 263. Bartholomew, K, & Horowitz, L. M. (1991). Attachm ent styles among young adults: A test of a four category model. Journal of Personality and Social Psychology, 61 226 244. Bartz, J. A., & Lydon, J. E. (2004). Close relationships and the working self concept: Implicit and explicit effects of priming att achment on agency and communion. Personality and Social Psychology Bulletin, 30 1389 1401. Batson, C. D. (2010). Empathy induced altruistic motivation. In M. Mikulincer & P. R. Shaver (Eds.) Prosocial motives, emotions, and behavior: The better angels of our nature (pp. 15 34). Washington, D. C.: American Psychological Association.
267 Batson, C. D., Batson, J. G., Griffit, C. A., Barrientos, S., Brandt, J. R., Sprengelmeyer, P., & Bayly, M. J. (1989). Negative state relief and the empathy altruism hypothesis. Journal of Personality and Social Psychology, 56 922 933. Batson, C. D., Bolen, M. H., Cross, J. A., & Neuringer Bene fiel, H. E. (1986). Where is the altruism in the altruistic personality? Jo urnal of Personality and Social Psychology, 50 212 220. Batson, C. D., Fultz, J., & Schoenrade, P. A. (1987). Distress and empathy: Two qualitatively distinct vicarious emotions with different motivational consequences. Journal of Personality, 55 19 39. self reported distress and empathy on egoistic versu s altruistic motivation to help. Jo urnal of Personality and Social Psychology, 45 706 718. Belsky, J. (1999). Modern evolutionary theory and patterns of attachment. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 141 161). N ew York: Guilford Press. Ben Ari, A., & Lavee, Y. (2005). Dyadic characteristics of individual attributes: Attachment, neuroticism, and their relation to marital quality and closeness. American Journal of Orthopsychiatry, 75 621 631. Birnbaum, G. E. (2007). Attachment orientations, sexual functioning, and relationship satisfaction in a community sample of women. Journal of Social and Personal Relationships, 24 21 35. Bowlby, J. (1969 /1982 ). Attachment and loss: Vol. 1. Attachment New York : Basic Books. Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. New York: Basic Books. Bowlby, J. (1980). Attachment and loss: Vol. 3. Sadness and depression. New York: Basic Books. Bowlby, J. (1988). A secure base: Parent chi ld attachment and healthy human development. New York: Basic Books. Brand, M., & Markowitsch, H. J. (2008). The role of the prefrontal cortex in episodic memory. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 317 341). Amsterdam: Elsevier.
268 Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self report measurement of adult romantic attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp. 46 76). New York: Guilford Press. Bretherton, I., & Munholland, K. A. (1999). Internal working models in attachment relationships: A construct revisited. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 89 111 ). New York: Guilford Press. Brown, S. C., & Craik, F. I. M. (2000). Encoding and retrieval of information In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 93 107 ). New York: Oxford University Press. Campbell, L., S impson, J. A., Boldry, J., & Kashy, D. A. (2005). Perceptions of conflict and support in romantic relationships: The role of attachment anxiety. Journal of Personality and Social Psychology, 88 510 531. Cassidy, J. (1994). Emotion regulation: influences of attachment relationships. In Fox, N. A. (Ed.), The development of emotion regulation: biological and behavioral considerations. Monographs of the Society for Research in Child Development 59 228 49. Chappell, K. D., & Davis, K. E. (1998). Attachm ent, partner choice, and perception of romantic partners: An experimental test of the attachment security hypothesis. Personal Relationships, 5 327 342. Christman, S. D., & Propper, R. E. (2010). Episodic memory and interhemispheric interaction: Handedne ss and eye movements. In G. M. Davies & D. B. Wright (Eds.), Current issues in applied memory research (pp. 185 205). New York: Psychology Press. Cialdini, R. B., Schaller, M., Houlihan, D., Arps, K., Fultz, J., & Beaman, A. (1987). Empathy based helping : Is it selflessly or selfishly motivated? Journal of Personality and Social Psychology, 52 749 758. Clinton, T., & Sibcy, G. (2006). Why you do the things you do: The secret to healthy relationships. Nashville: Integrity Publishers. Collins, N. L., Fo rd, M. B., Guichard, A. C., Kane, H. S., & Feeney, B. C. (2010). Responding to need in intimate relationships: Social support and caregiving processes in couples. In M. Mikulincer & P. R. Shaver (Eds.), Prosocial motives, emotions, and behavior: The better angels of our nature (pp. 367 389). Washington, D. C.: American Psychological Association.
269 Collins, N. L., Guichard, A. C., Ford, M. B., & Feeney, B. C. (2004). Working models of attachment: New developments and emerging themes. In W. S. Rholes & J. A. Simpson (Eds.), Adult attachment: Theory, research, and clinical implications (pp. 196 239). New York: Guilford Press. Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Persona lity and Social Psychology, 58 644 663. Conway, M. A. (2008). Exploring episodic memory. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 19 29). Amsterdam: Elsevier. Christman, S. D., & Propper, R. E. (2010). Episodic memory and interhemispheric interaction: Handedness and eye movements. In Davies, G. M., & Wright, D. B. (Eds.), Current Issues in Applied Memory Research (pp. 185 205). New York: Psychology Press. Davila, J., Bradbury, T. N., & Fincham, F. (1998 ). Negative affectivity as a mediator of the association between adult attachment and marital satisfaction. Personal Relationships, 5 467 484. Davila, J., Karney, B. R., & Bradbury, T. N. (1999). Attachment change processes in the early years of marr iage. Journal of Personality and Social Psychology, 76 783 802. Feeney, J. A. (1994). Attachment style, communication patterns, and satisfaction across the life cycle of marriage. Personal Relationships, 1 333 348. Feeney, J. A. (1999). Adult romantic attachment and couple relationships. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical application (pp. 355 377). New York: Guilford Press. Feeney, J. A. (2004). Transfer of attachment from parents to romantic partners: Effects of individual and relationship variables. Journal of Family Studies, 10 220 238. Feeney, J. A., & Hohaus, L. (2001). Attachment and spousal caregiving, Personal Relationships, 8 21 39. Feeney, J. A., & Noller, P. (1990). Attachment st yle as a predictor of adult romantic relationships. Journal of Personality and Social Psychology, 58 281 291. Fehr, B. (2010). Compassionate love as a prosocial emotional. In M. Mikulincer & P. R. Shaver (Eds.) Prosocial motives, emotions, and behavior: The better angels of our nature (pp. 245 265). Washington, D. C.: American Psychological Association.
270 Florian, V., Mikulincer, M., & Hirschberger, G. (2000). The anatomy of a problematic emotion: The conceptualization and measurement of the experience of pity. Imagination, Cognition, and Personality, 19 3 25. Fonagy, P. (1999). Psychoanalytic theory from the viewpoint of attachment theory and research. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory research, and clinical applic ations (pp. 595 624). New York: Guilford Press. Fraley, R. C. (2002). Attachment stability from infancy to adulthood: Meta analysis and dynamic modeling of developmental mechanisms. Personality and Social Psychology Review, 6 123 151. Fraley, R. C. (201 0 ). Information on the Experiences in Close Relationships Revised (ECR R) adult attachment questionnaire. Retrieved August 15, 2010, from R. C. University of Illinois at Urbana Champaign Psychology Department Web site: http://www.psych.illinois.edu/~rcfraley/measures/ecrr.htm Fraley, R. C., & Brumbaugh, C. C. (2004). A dynamical systems approach to conceptualizing and studying stability and change in attachment security. In W S. Rholes & J. A. Simpson (Eds.), Adult attachment: theory, research, and clinical implications (pp. 86 132). New York: Guilford Press. close friendships and romant ic relationships. Personal Relationships, 4 131 144. Fraley, R. C., & Shaver, P. R. (1999). Loss and bereavement: Attachment theory and In J. Cassidy & P. R. Shaver (Eds.), Handbo ok of attachment: Theory research, and clinical applications (pp. 735 759 ). New York: Guilford Press. Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item response theory analysis of self report measures of adult attachment. Journal of Personal ity and Social Psychology, 78 350 365. Fujii, T. (2008). The basal forebrain and episodic memory. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 343 362). Amsterdam: Elsevier. Gainesville Council for Economic Outrea ch. Retrieved August 15, 2010 from http://www.g ceo.com/media/docs/CEO/PDFs/GainesvilleOverview2009.pdf Gillath, O., Shaver, P. R., Mikulincer, M., Nitzberg, R. E., Erez, A., & van IJzendoorn, M. H. (2005). Attachment, caregiving, and volunteer ing: Placing volunteerism in an attachment theoretical framework. Personal Relationships, 12 425 446.
271 Gottman, J. M. (1994). What predicts divorce?: The relationships between marital processes and marital outcomes. Hillsdale, NJ: Earlbaum. Griffin, D. W., & Bartholomew, K. (1994). The metaphysics of measurement: The case of adult attachment. In K. Bartholomew & D. Perlman (Eds.) Advances in personal relationships: Vol. 5. Attachment processes in adulthood (pp. 17 52). London: Jessica Kingsley. Hazan C., & Shaver, P. R. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology 52, 511 524. Hazan, C., & Shaver, P. R. (1990). Love and work: An attachment theoretical perspective. Journal of Personalit y and Social Psychology, 59 270 280. Hazan C., & Ziefman, D. (1999). Pair bonds as attachments: Evaluating the evidence. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, a nd clinical application (pp. 336 354). New York: Guilford Press. Hesse, E. (1999). The Adult Attachment Interview: Historical and current perspectives. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 395 433). New York: Guilford Press. J oe, V. C. (1974). Perceived personal control and attribution of causality. Perceptual and Motor Skills, 38 323 329. Johnson, S. M. (1996). The practice of emotionally focused marital therapy: Creating connection. Florence, KY: Bruner/Mazel. Johnson, S. M., & Whiffen, V. E. (Eds.), (2003). Attachment processes in couple and family therapy. New York: Guilford Press. Kagan, J. (1996). Three pleasing ideas. American Psychologist, 51 901 908. Klohnen, E. C., & Luo, S. (2003). Interpersonal attraction and personality: What is attractive self similarity, ideal similarity, complementarity or attachment security? Journal of Personality and Social Psychology, 85 709 722. Kobak, R. (1999). The emotional dynamics of disruptions in attachment relationships: Implications for theory, research, and clinical intervention. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 21 43). New York: Guilford Press. Kobak, R. R., Cole, H. E., Ferenz Gillies, R. and Fleming, W. S. (1993). Attachment and emotion regulation during mother teen problem solving: a control theory analysis. Child Development 64 231 45.
272 Kobak, R., & Hazan, C. (1991). Attachment in marriage: Effects of security and security accuracy of working models. Journal of Personality and Social Psychology, 60 861 869. Kunce, L. J., & Shaver, P. R. (1994). An attachment theoretical approach to caregiving in romantic relationships. In K. Bartholomew & D. Perlman (Eds.) Advances in personal relat ionships: Vol. 5. Attachment processes in adulthood (pp.205 237). London: Jessica Kingsley. Latty Mann, H., & Davis, K. E. (1996). Attach ment theory and partner choice: Preference and actuality. Journal of Social and Personal Relationships, 13 5 23. Ma in, M. (1999). Attachment Theory: Eighteen points with suggestions for future studies. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 845 887). New York: Guilford Press. Main, M. B. (2010). Adult Attachment Interview protocol. Retrieved October 25, 2010, from Stony Brook University, Department of Psychology Web site: http://www.psychology.sunysb.edu/attachment/measures/content/aai_interview.pdf Main, M., Hesse, E., & Kaplan, N. (2005). Pred ictability of attachment behavior and representational processes at 1, 6, and 19 years of age: The Berkeley longitudinal study. In K. E. Grossmann, K. Grossmann, & E. Waters (Eds.), Attachment from infancy to adulthood: The major longitudinal studies (pp. 245 304). New York: Guilford Press. Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move to the level of representation. Monographs of the Society for Research in Child Development, 50 66 104. Marchand, J. E. attachment orientations, depressive symptoms, and conflict resolution behaviors. Attachment and Human Development, 6 99 112. Mayseless, O. (2004). Home leaving to military service: Attac hment concerns, transfer of attachment functions from parents to peers, and adjustment. Journal of Adolescent Research, 19 533 558. Mikulincer, M., Florian, V., & Tolmacz, R. (1990). Attachment styles and fear of personal death: A case study of affect re gulation. Journal of Personality and Social Psychology, 58 273 280. Mikulincer, M., Gillath, O., Halevy, V., Avihou, N., Avidan, S., & Eshkoli, N. (2001). sense of attachme nt security promotes empathic responses. Journal of Personality and Social Psychology, 81 1205 1224.
273 Mikulincer, M. & Shaver, P. R. (2007). Attachment in adulthood: Structure, dynamics and change. New York: The Guilford Press. Mikulincer, M., Shaver, P. R., Gillath, O., & Nitzberg, R.A. (2005). Attachment, caregiving, and altruism: Boosting attachment s ecurity increases compassion and helping. Journal of Personality and Social Psychology, 89 817 839. Moscovitch, M., Winocur, G., Ryan, L., and Nade l (2008). Functional neuroanatomy of remote, episodic memory. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 239 269). Amsterdam: Elsevier. Moulin, C. J. A., & Chauvel, P. (2010). Dj vu: Insights from the dream y state and the neuropsychology of memory. In Davies, G. M., & Wright, D. B. (Eds.), Current Issues in Applied Memory Research (pp. 206 235). New York: Psychology Press. Nadel, L., Hupbach, A., Hardt, O., & Gomez, R. (2008). Episodic memory: Reconsolidati on. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 43 56). Amsterdam: Elsevier. Pearlin, H. J. & Schooler, C. (1978). The structure of coping. Journal of Health and Social Behavior, 19 2 22. Rennie, D. L. (19 98). Person centred counselling: An experiential approach. London: Sage Publications. Ryan, L., Hoscheidt, S., & Nadel, L. (2008). Perspectives on episodic and semantic memory retrieval. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 5 18). Amsterdam: Elsevier. Sax, G. (1989). Principles of educational measurement and evaluation (3rd ed.). Belmont, CA: Wad s worth. Schwartz, S. H. (1992). Universals in the content and structure of values: Theoretical advances a nd empirical tests in 20 countries. Advances in Experimental Social Psychology, 25 1 65. Shaver, P. R., Belsky, J., & Brennan, K. A. (2000). The Adult Attachment Interview and self reports of romantic attachment: Associations across domains and methods. Personal Relationships, 7 25 43. Shaver, P. R., & Mikulincer, M. (2006). A behavioral systems approach to romantic love relationships: Attachment, caregiving, and sex. In R. J. Sternberg & K. Weis (Eds.), The new psychology of love (pp. 35 64). New Ha ven, CT: Yale University Press.
274 Shaver, P. R. & Mikulincer, M. (2007). Adult attachment theory and the regulation of emotion. In J. J. Gross (Ed.), Handbook of emotion regulation (pp. 446 465). New York: Guilford Press. Shrager, Y., & Squire, L. R. ( 2008). The medial temporal lobe: Visual perception and recognition memory. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 271 281). Amsterdam: Elsevier. Sibley, C. G., & Liu, J. H. (2004). Short term temporal sta bility and factor structure of the revised experiences in close relationships (ECR R) measure of adult attachment. Personality and Individual Differences, 36 369 375. Simpson, J. A. (1990). Influence of attachment styles on romantic relationships. J ournal of Personality and Social Psychology, 59 971 980. Smalley, G., & Paul, R. S. (2006). The DNA of relationships for couples. Branson, MO: Smalley Publishing Group. Smith, D. M. (2008). The hippocampus, context processing and episodic memory. In E Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 465 481). Amsterdam: Elsevier. Sperling, M. B., Berman, W. H., & Fagen, G. (1992). Classification of adult attachment: An integrative taxonomy from attachment and p sychoanalytic theories. Journal of Personality Assessment, 59 239 247. Sprecher, S., & Fehr, B. (2005). Compassionate love for close others and humanity. Journal of Social and Personal Relationships, 22 629 651. Sprecher, S., Fehr, B., & Zimmerman, C. (2007). Expectation for mood enhancement as a result of helping: The effects of gender and compassionate love. Sex Roles, 56 543 549. Stevenson Hinde, J. (2005). The interplay between attachment, temperament, and maternal style: A Madingley perspective. In K. E. Grossmann, K. Grossmann, & E. Waters (Eds.), Attachment from infancy to adulthood: The major longitudinal studies (pp. 198 222). New York: Guilford Press. Suddendorf, T., & Corballis, M. C. (2008). Episodic memory and mental time travel. In E. D ere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp.31 42). Amsterdam: Elsevier. Thompson, R. A. (1999). Early attachment and later development. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical Applications (pp. 265 286). New York: Guilford Press.
275 Configurations of adult attachment representations and their implications for marital functioning. Developmental Psychology, 40 295 314. Trimble, M. R., & Cavanna, A. E. (2008). The role of the precuneus inepisodic memory. In E. Dere, A. Easton, L. Nadel, & J. P. Huston (Eds.), Handbook of Episodic Memory (pp. 363 377). Amsterdam: Elsevier. Trinke, S. J., & Bart holomew, K. (1997). Hierarchies of attachment relationships in young adulthood. Journal of Social and Personal Relationships, 14 603 625. Tulving, E. (2001). Origin of autonoesis in episodic memory. In H. L. Roediger, III, J. S. Nairne, I. Heath, & A. M Surprenant (Eds.), The nature of remembering: Essays In honor of Robert G. Crowder (pp. 17 34). Washington, D. C.: American Psychological Press Van IJzendoorn, M. H. (1995). Adult attachment representations, parental responsiveness, and infant attach ment: A meta analysis on the predictive validity of the Adult Attachment Interview. Psychological Bulletin, 117 387 403. Vaughn, B. E., & Bost, K. K. (1999). Attachment and temperament: Redundant, independent, or interacting influences on interpersonal a daptation and personality development? In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical a pplications (pp. 265 286). New York: Guilford Press. Wampler, K. S., Shi, L., Nelson, B. S., & Kimball, T. G. (2003). The A dult Attachment Interview and observed couple interaction: Implications for an intergenerational perspective on couple therapy. Family Process, 42 497 515. Waters, E., Merrick, S., Treboux, D., Crowell, J., & Albersheim, L. (2000). Attachment security i n infancy and early adulthood: A twenty year longitudinal study. Child Development, 71 684 689. Wheeler, M. A. (2000). Episodic memory and autonoetic awareness. In E. Tulving & F. I. M. Craik (Eds.), The Oxford handbook of memory (pp. 597 608). New York : Oxford University Press.
276 BIOGRAPHICAL SKETCH Stephen Carroll Figley was b orn in Marion, Ohio, in 1953. Aft er graduating from West Liberty Salem High School in rural west central Ohio, he attended G race College, later transferring to Florida Bible College, where he r Pastoral S tudies in 1975. Steve served in pastoral roles in the years subsequent to graduation often concurrent with other employment. After se ven years working as a municipal government administrator, he ser ved seven years as a Christian s chool headmas t er in Okeechobee, Florida During that tenure he and his family spent four summers in Winona Lake, Indiana, where he was enrolled in a graduate program hosted by Grace Theological Seminary utilizing vi siting education professors. Steve earned a Master of A rts degree in Christian School Administration from that program in 1992, receiving the outstanding graduate student award and strong encouragement to pursue his doctorate Steve subsequently enrolled in the University of Florida College which he recognized as much more in tune with his pass ion and calling. En route to earning this change of direction doctorate, Steve was awarded the Ma ster of Education and Specialist in Education degrees in m arriage and family c ounseling from the University of Florida in 1996. After his supervised experien ces, he established his own privat e practice in 1998. Steve is both a Licensed Marriage and Family Therapist and a Licensed Mental Health Counselor. counseling, and the integration of counsel ing with the Christian faith. M any of his referrals come from pastors and previous clients whose receptivity to counseling is enhanced by such integration.
277 change of educ ational and career direction resulted in a more ci rcuitous route to achieving his original goal of earning a doct orate Emerging priorities included a battle with cancer to overcome, teenage children to launch, an d a private practice to build so that life could be sustained in the process. Several benefits arose, however, from this sometimes burdensome ad venture of a later in life doctoral process: study design was very much informed and refined by his clinical pract ice focus of couples Reciprocally his clinical appro aches wer e significantly influenced by the adult attachment themed literatur e informing his research study. And his established presence in the community facilitated his ready recruitment of a large number of couples willing to serve as his study subjects. Steve is married to Joanna, who is the f ounder and executive director of G race Encourageme nt Ministries. Under the umbrella of this not for profit corporation Steve and his wife conduct marriage seminars and retreats. They are in the process of developin g a marriage curr iculum designed to translate well across diverse culture s, and recently piloted that curriculum with a group of 35 African couples during a one month visit to Botswana Steve and his wife have two adult children one grandchild and another soon to be born. Steve looks forward to training for a marathon and p erhaps someday fulfilling the dream of qualifying for and r unning at Boston. He is buoyed by the anticipation of family ad ventures unencumbered by dissertation deadlines, and after a season of unwinding, looks forward to vision casting with his wife about shared ventures of ministry to couples both in the U.S. and overseas.