Attachment as an Organizational Construct for Affect, Appraisals, and Coping of Late Adolescent Females

Journal of Youth and Adolescence, Vol. 28, No. 5, 1999

Attachment as an Organizational Constructfor Affect, Appraisals, and Coping of LateAdolescent Females

Julia C. Torquati1 and Alexander T. Vazsonyi2

Received October 14, 1998: accepted March 3, 1999

The purpose of this study was to 1) compare general affective dispositions (depres-sion and anxiety) and negative affect during interpersonal conflict as a functionof attachment security, 2) examine appraisals as a function of attachment styleand as predictors of coping, 3) compare strategies of coping with interpersonalconflict as a function of attachment style, and 4) investigate the roles of attachmentstyle, affect, and appraisals in predicting coping in the context of interpersonalconflict. Seventy-three late adolescent females participated. Insecure participantsreported higher levels of depression, anxiety, and negative affect during interper-sonal conflicts. Insecure participants were more likely to cope with interpersonalconflicts through support seeking or avoidance. Hierarchical regression analysisindicated that general and specific attachment style, affect, and appraisals signif-icantly predict coping strategies. Implications for general and specific models ofattachment as organizational constructs and attachment as a predictor of copingwith interpersonal and non-interpersonal stressors are discussed.

INTRODUCTION

This investigation examined the influence of attachment style, affect, and ap-praisals as predictors of coping strategies in the context of conflict with significantothers. First, attachment will be described as an organizational construct. Researchexamining the relationship between attachment style and affect will be reviewed

Portions of this paper were presented at the Biennial Meeting of the Society for Research on Adoles-cence, San Diego, CA, February 1994.

1Assistant professor, University of Nebraska-Lincoln. Research interests include developmental riskand competence. To whom correspondence should be addressed at 137 Child Development Lab,Lincoln, NE 68583-0830.

2 Assistant professor, Auburn University. Research interests include delinquency and adolescence.

0047-2891/99/1000-0545$16.00/0 C 1999 Plenum Publishing Corporation

545

to provide a context for the first objective of this study: to examine mental health(depression and anxiety) and negative affect during conflict as a function of at-tachment style. Next, studies are described to provide a context for the secondobjective: examination of stressor appraisals both as a function of attachment styleand as predictors of coping. The third objective is to examine strategies of copingwith interpersonal conflict as a function of attachment style. The fourth objectiveof this study is to test a multivariate model including attachment style, relationship-specific attachment models, negative affect (depression, anxiety, and contextualnegative affect), and appraisals as predictors of coping with interpersonal conflict.

Attachment as an Organizational Construct

Attachment theory has provided a framework for a great deal of research onthe coherence and organization of affect, cognition, and behavior in interpersonalrelationships. According to attachment theory, internal working models of rela-tionships, including representations of self, relationship, other, and the social worldin general, develop from an amalgam of early relationship experiences. Internalworking models serve as interpretive schemata in social contexts, organizing andguiding behavior, affect, and cognition. Attachment models are hypothesized to bedynamic, and individuals are capable of actively constructing and revising thembased on new experience (Bowlby, 1988). However, attachment models are alsohypothesized to be a mechanism of continuity in social development (Sroufe andWaters, 1977).

Working models of the social world in general and of specific attachment fig-ures provide an organizational construct for the development of strategies for regu-lating attachment behavior and emotions. General models of attachment compriseexpectations about the social world in general and provide a heuristic for the devel-opment and revision of relationship-specific models of attachment. Relationship-specific models of attachment include expectations about the behavior of the selfin the context of a specific relationship and of specific significant others. Internalworking models are activated in the context of a perceived threat, and thereforefunction as an organizational construct for affect, appraisals, coping behavior, andconflict resolution styles.

Attachment Style and Affect

Secure attachment style is characterized by expectations about efficacious be-havior on the part of the self and sensitive responding on the part of the attachmentfigure. Individuals with a secure attachment style anticipate sensitive respondingon the part of the attachment figure, and typically employ primary strategies involv-ing proximity seeking in response to activation of the attachment system. Insecure

546 Torquati and Vazsonyi

attachment style is characterized by anticipation of insensitive or inconsistent re-sponding on the part of the attachment figure, and is associated with secondarystrategies that involve relative "hyperactivation" (referred to as preoccupied) or"deactivation" (referred to as dismissing) of the attachment system (Kobak et al.,1993). These strategies are attempts to regulate negative emotion and/or monitoravailability of the attachment figure.

Investigations of the experience, expression, and regulation of emotion as afunction of attachment style have documented an association between insecureattachment styles and negative affect. For example, spouses with secure work-ing models demonstrated more constructive modulation of emotion than insecurespouses (Kobak and Hazan, 1991). Dismissing and preoccupied late adolescentsreport more negative affect in their romantic relationships while those with a se-cure attachment style report more positive affect, trust, and commitment (Simpson,1990). Feeney and Kirkpatrick (1996) reported higher levels of autonomic arousalin preoccupied and dismissing college women (compared to secure) during a lab-oratory separation, suggesting that the salience of the separation and the level ofdistress both differed as a function of attachment style.

Clinical studies of attachment style and psychopathology in adolescents andadults provide evidence of the association between insecure attachment style andemotion dysregulation. Hospitalized adolescents had higher rates of insecure at-tachment style than a community sample, and substance abuse and delinquencywere associated with dismissing attachment style (Allen et al., 1996). Dismissingattachment style was associated with conduct disorder and substance abuse, andpreoccupied attachment style was associated with affective disorders (Rosensteinand Horowitz, 1996). Preoccupied attachment style of adolescents was associ-ated with depressive symptoms (Kobak et al., 1991) and secure attachment withparents was associated with less depression and anxiety (Papini and Roggman,1992). Depressed college women reported greater preoccupation and fearful avoid-ance in romantic relationships in comparison to those who were non-depressed.Adult women recovering from depression also reported greater fearful avoidance(Carnelley et al., 1994).

The emotional traits investigated in the studies reviewed above primarily de-scribe the relationship between mental health and attachment style. Fewer studieshave examined affect experienced in the context of a stressful event as a functionof attachment style. The first objective of this study is to replicate and extendprevious research comparing mental health (depression and anxiety) as a func-tion of attachment style, by comparing negative affect in the context of interper-sonal conflict as a function of attachment style. The following hypotheses willbe tested: 1) insecure participants will report greater depression and anxiety and2) participants with insecure attachment styles will report greater distress (negativeaffect) in the context of interpersonal conflicts with mothers, fathers, and datingpartners.

Attachment and Coping 547

Emotion and Coping

Coping has been defined as the process of addressing internal or externaldemands and is conceptualized as an active, purposeful process of responding tostimuli perceived as threatening or requiring resources (e.g., Billings and Moos,1981; Lazarus and Folkman, 1984). The functions of coping are to directly addressthe stressor(s) and/or to regulate associated emotions. Conceptual and empiricalefforts to define styles of coping parallel these distinct functions: problem-focusedcoping refers to efforts to specifically address the stressor, while emotion-focusedcoping refers to efforts to regulate emotions associated with the stressor (e.g., Bandand Weisz, 1988; Lazarus and Folkman, 1984).

There are several ways in which emotions may influence coping processes.For example, distinct types of emotions (e.g., sad, scared, angry) may predict dif-ferent coping strategies. Intensity of negative emotion may interfere with problemsolving or motivate decisive action. Alternatively, positive emotion may facilitateproblem solving. For example, Greene and Noice (1988) reported a linear relation-ship between positive emotion and problem solving in adolescents. Perhaps mostimportantly, emotion and cognition are integrated systems in the appraisal process,simultaneously regulating perceptions of threat and behavioral response choices.

Appraisals and Coping

The attachment system is activated in the context of perceived threat. Percep-tions of threat and degree of distress vary as a function of attachment style. Insecureattachment style has been associated with appraisal of greater threat and psycho-logical distress in the context of both interpersonal and non-interpersonal stressors.For example, anxious attachment style predicted emotional distress, which in turnpredicted increased probability of interpersonal conflict in romantic relationshipsof late adolescent females (Collins, 1996). Mikulincer and Florian (1995) reportedthat dismissing miliary recruits perceived the training as more threatening, andperceptions of threat were associated with emotion focused and distancing copingstrategies. No differences in problem focused coping as a function of attachmentstyle were observed. Dismissing recruits were less likely to use support seekingthan those who were secure or preoccupied.

An investigation of attachment styles, distress, and coping strategies of Israelicollege students in high- and low-danger areas during the Gulf War also revealedpatterns consistent with attachment theory (Mikulincer et al., 1993). Preoccupiedstudents reported more distress and used more emotion focused coping strategies,and dismissing students displayed more somatization, hostility, and trauma-relatedavoidance, and used more distancing coping strategies in high-danger situations.Secure students reported more support seeking in both low- and high-danger areas.

548 Torquati and Vazsonyi

There were no differences in problem focused coping as a function of attachmentstyle. Moreover, problem focused coping was inversely correlated with psycho-logical symptoms, and emotion focused coping was positively correlated withpsychological symptoms. Support seeking was not correlated with symptoms, butwas positively correlated with problem focused coping. Similarly, Armsden andGreenberg (1988) reported that stressful events were more strongly related tosymptomatology for insecurely attached adolescents than for securely attachedadolescents.

The influence of stable affective dispositions such as depression and anxietyon coping has also been examined. Endler and Parker (1990) reported a directrelationship between depression, anxiety, and emotion focused coping, a direct re-lationship between depression and avoidant coping for males only, and an inverserelationship between depression and problem focused coping for females only.Evidence for an inverse relationship between depression and problem focusedcoping and a direct relationship between depression and emotion focused copinghas also been found with adult samples (Billings and Moos, 1984; Rosenberg et al.,1987).

Attachment, Appraisals and Coping

Coping responses vary systematically with appraisals of stressors. Appraisalsof stressors include evaluations of event controllability, desirability, magnitude ofthreat, whether the event is expected, and the expected outcome of the event.Differential coping strategies as a function of appraisals have been examined withchild, adolescent, and adult samples (e.g., Compas et al., 1991; Compas et al.,1988b; Forsythe and Compas, 1987; Lazarus and Folkman, 1984). For example,attributions of internal control are consistently related to problem focused coping,especially in the context of interpersonal stressors in school aged children andadolescents (Compas et al., 1988b, 1991; Gamble, 1993).

Interpersonal stressors demand both coping and conflict resolution strategies.Styles of resolving interpersonal conflict vary as a function of attachment style.Adolescents with a secure attachment style demonstrate less anger, less avoid-ance of problem solving, and more constructive engagement during interactionswith their mothers (Kobak et al., 1993). Pistole (1989) compared conflict resolutionstyles as a function of attachment style and found that securely attached adults were(1) most likely to use integrating strategies in which a resolution is reached allowingeach partner to benefit (a "win/win" resolution) and (2) more likely than adults witha preoccupied attachment style to compromise, reaching a resolution in which eachindividual concedes something to preserve the relationship (a "win/lose win/lose"resolution). Moreover, preoccupied adults were more likely than dismissing adultsto oblige the partner by making a concession without the partner reciprocating.

Attachment and Coping 549

Simpson et al. (1996) examined distress and changes in perception of partner af-ter discussing a major or minor problem and found differences as a function ofattachment style. Preoccupied men and women perceived their partner and rela-tionship less positively, and reported greater distress, while less ambivalent menand women had more positive perceptions after discussing a major problem. To-gether these results suggest that preoccupied adults appraise the conflict as a greaterthreat to the relationship and are therefore reluctant to request that the partner com-promise, instead putting the relationship before their personal interests and goals.

Coping strategies have been examined as moderators of the relationship be-tween attachment and marital adjustment (Lussier et al., 1997). Secure attachmentpredicted problem focused coping, and insecure attachment predicted emotionfocused coping. However, problem focused coping attenuated the relationship be-tween insecure attachment and marital satisfaction, while emotion focused copingamplified the relationship. Support seeking as a coping strategy can be both emo-tion and problem focused because individuals can receive both emotional andinformational support. Simpson et al. (1992) found that secure women were morelikely than avoidant women to seek support under stress.

Secure attachment models are characterized by a sense of social self-efficacyand a generally positive perception of the social world in general and of attach-ment figures in particular. It is plausible that the self-efficacy characteristic ofsecure attachment style will be associated with a greater sense of control overthe outcome of a conflict situation. Similarly, a secure, positive perception of at-tachment figures is expected to be associated with minimal perceived threat to therelationship in the context of conflict. The second objective of this research is toexamine appraisals as a function of attachment style and as predictors of coping.The following hypotheses will be tested: (1) Insecure participants will perceivethe conflict as a greater threat; (2) secure participants will appraise greater controlover outcome; and (3) appraisals of control (over outcome) will be associated withproblem focused coping.

The third objective of this research is to examine coping strategies as a functionof attachment style. It is hypothesized that insecure participants will rely more onemotion focused coping in comparison to secure participants, and that secureparticipants will be more likely to engage in problem focused coping.

The research summarized above describes attachment style as an organiza-tional construct for emotion, appraisals, and coping. The fourth objective of thisstudy is to integrate the previously described findings by testing a multivariatemodel including attachment style, relationship-specific attachment models, neg-ative affect (depression, anxiety, and contextual negative affect), and appraisalsas predictors of coping with interpersonal conflict. The following hypotheses willbe tested: (1) Insecure attachment to a specific other will predict emotion focusedcoping and (2) secure attachment to a specific other will predict problem focusedcoping.

550 Torquati and Vazsonyi

METHODS

Sample and Procedures

Seventy-three females ranging in age from 18-22 years (mean = 20.6 years)participated. The sample was predominantly Caucasian (83.6%; Hispanic, 13.7%;African American, 2.7%). Participants completed questionnaires during a 75-minute social science class. Each participant completed her own questionnaireand marked responses on a computer scan sheet. Graduate research assistants gaveinstructions and were available for questions during the entire class period. Thissample is restricted to females because the class enrollment was predominantlyfemale, and there were too few males who completed the survey (n = 7) to justifyanalyzing the data separately or together.

Measures

Attachment Style

General attachment style was assessed using the 18-item self-report measuredeveloped by Collins and Read (1990). This measure assesses three dimensions ofattachment: (1) comfort with closeness (Close), (2) comfort depending on othersand having others depend on the respondent (Depend), and (3) Anxiety. Participantsrated each statement on a scale from 1 (not at all true for me) to 4 (very true for me).

Three additional items were developed by the authors for the present researchto assess "activation" of the attachment system (Kobak et al., 1993): (1) "Relation-ships with others are important to me" (indicates valuing attachment characteristicof security); (2) "Sometimes I spend so much time thinking about my relationshipswith others that I have a hard time getting things done" (activation); and (3) "Idon't have time to be concerned about close relationships" (deactivation).

The hypotheses for the present research were based on the conceptualiza-tion of attachment categories. Collins (1996) developed a system for assigningindividuals to categories based on the continuous scores on the dimensions Close(a = .71), Depend (a = .69), and Anxiety (a = .70). Individuals scoring ator above scale midpoint on each dimension are considered to be "high" on thatdimension, and those scoring below the midpoint are considered "low." Close andDepend were combined into a single mean score, and in this sample they weresignificantly correlated in the expected direction (r = .29; p < .05). This score issimilar to the Secure-Dismissing dimension constructed by Simpson et al. (1992),except that higher scores indicate greater security rather than greater avoidance.The items described above for valuing attachment (#1) and dismissing (#3, reversecoded) were included in the Close/Depend score. The preoccupation item (#2)

Attachment and Coping 551

was included in the Anxiety score. Close and Anxiety were significantly corre-lated in the expected direction (r = —.32, p < .01) as were Depend and Anxiety(r = —.44, p < .01). Categories were assigned as follows: 1) secure = highClose/Depend, low Anxiety; 2) preoccupied = high Close/Depend, high Anxiety;3) dismissing = low Close/Depend, low Anxiety; 4) fearful = low Close/Depend,high Anxiety. Proportions of the sample were not large enough in each of theseparate insecure categories to examine variation as a function of type of inse-cure attachment. Therefore, all insecure types were combined into a single cate-gory for comparison of means (insecure: n = 29; secure: n = 44). Dimensionalscores (Close/Depend, Anxiety) were used in parametric analyses. Items from thesame measure were adapted to represent specific attachment style with mother,father, and dating partner; all subscales demonstrated good reliability (a range =.69-.88).

Participants only completed measures relevant to current relationships. Forexample, participants only completed the section of the survey describing rela-tionships with their dating partner if they were "currently dating." There was nominimum length of time required to define "currently dating." All 73 participantsreported on dating relationships and relationships with mothers. Seventy partici-pants reported on relationships with fathers.

Coping Style

Fifteen items of the Child and Adolescent Problem Solving Inventory(Gamble, 1993) were used to assess three styles of coping: (1) problem solving(a: Mom = .79; Dad = .81; Partner = .69); (2) support seeking (a: Mom = .69;Dad = .69; Partner = .80); and (3) avoidance (A: Mom = .71; Dad = .71; Part-ner = .71). Participants described the topic of a recent conflict with each rela-tionship partner (mother, father, dating partner), then rated the frequency of usingthree different coping strategies during each conflict on a 4-point response scaleranging from never (1) to always (4). Five items comprise each of three subscales.Sample items for each type of coping include: (1) problem solving: "I try to figureout what is wrong and do something about it"; "I think about the problem andtry to fix it"; (2) support seeking: "I talked to a friend about what happened"; "Iasked someone I respected for advice"; and (3) avoidance: "I did something elseand tried to forget about it"; "I went on as if nothing had happened."

Appraisals

Participants rated appraisals of control ("When you have a conflict with ,how much control do you have over the way things will turn out?") and threat("When you have a disagreement with , how much of a problem is it foryou?") on a 3-point scale. Higher scores on the threat question indicated that the

552 Torquati and Vazsonyi

conflict was a "big problem," and higher scores on the control question indicatedthat the participant had "little or no control" over the way things turn out.

Negative Affectivity

Participants separately rated each of eight specific emotions (angry/mad, sad,jealous, afraid/scared, happy (reverse coded), guilty, lonely, disappointed, rejected,worried) according to how they felt during the conflict they described with mother,father, and dating partner on a scale ranging from 1 (did not feel at all) to 3 (reallyfelt). A negative affect intensity score was comprised of the mean of the eightemotion ratings for each conflict (Mom A = .71; Dad A = .68; Dating Partnera = .63).

The trait subscale of the State-Trait Anxiety Inventory (STAI) (Spielberger,1983) and the 13-item short form of the Beck Depression Inventory (BDI) (Becket al., 1961) were indicators of general negative affectivity. The STAI includes 20items describing symptoms of anxiety, which participants rated on a scale from 1(least anxious) to 4 (most anxious) (a = .79). For each item of the BDI, participantschose 1 of 4 statements describing affective, cognitive, physical, and behavioralsymptoms of depression. Scores for each item ranged from 0 (not depressive) to 3(most depressive symptom) for a total possible score of 39 (A = .90).

RESULTS

The first objective of this study was to examine variation in both mental health(depression and anxiety) and negative affect in the context of interpersonal conflictas a function of attachment style. Mean scores on depression and anxiety werecompared using one-tailed t-tests. Consistent with hypotheses, participants with aninsecure attachment style rated themselves higher on both depression and anxiety.Mean scores on negative affect in the context of conflict were also compared usingone-tailed t-tests. Insecure participants scored higher than secure participants onnegative affect during conflict with mother, father, and dating partner, confirmingthe second hypothesis (Table I).

The second objective of this study was to compare appraisals of control andthreat as a function of attachment style, and appraisals as predictors of coping. One-tailed t -tests were used to compare means of appraisals for secure and insecureparticipants (Table I). Hypotheses regarding appraisals of control were partiallysupported: Insecure participants rated themselves lower on control over how theconflict with father will turn out (F(2,68) = 4.8; p < .05) (higher scores reflect lesscontrol). Comparisons of control appraisals for mother and dating partner were notsignificant. Hypotheses regarding appraisals of threat as a function of attachmentsecurity were partially supported: Insecure participants regarded conflict with dat-ing partner as a greater threat (F ( 2 , 7 1 ) = 17.7; p < .001). Comparisons of appraisal

Attachment and Coping 553

Torquati and Vazsonyi

of threat during conflict with mother and father as a function of attachment securitywere not significant.

Hypotheses regarding appraisals as predictors of coping strategies were par-tially supported. Appraisals of control were significantly correlated with problemsolving in conflicts with father and dating partner, but not mother. Control wasalso significantly correlated with support seeking during father conflict. Highercontrol scores reflect lower appraisals of control, so the relationship is actuallyinverse: appraisals of lower control were associated with problem solving withfathers and dating partners, and with support seeking in response to conflict withmothers. Appraisals of threat were significantly correlated with support seekingand problem focused coping during conflicts with mother and dating partner, andwith problem focused coping during conflict with father (Table II).

The third objective of this study was to compare coping strategies as a functionof attachment style. One-tailed t-tests partially supported hypotheses. Insecureparticipants were more likely to use avoidance in all three relationships and to usesupport seeking in response to conflict with mother and dating partner. However,secure and insecure participants did not significantly differ on their endorsementof problem focused strategies (Table III).

The fourth purpose of this research was to test a multivariate model in-cluding attachment style, relationship-specific attachment models, negative affect(depression, anxiety, and contextual negative emotion), and appraisals to predictthree types of coping strategies: problem solving, support seeking, and avoidance.

554

Table I. Negative Affect and Appraisals as a Functionof Attachment Style

Secure (n = 44)M (SD)

Insecure (n = 29)M (SD) F

General negative affectDepressionAnxiety

12.5 (2.8)12.6(3.4)

17.0 (6.7)14.6 (3.4)

16.1c

5.2a

Negative affect during conflictMom conflictDad conflictPartner conflict

1.4 (0.3)1.4 (0.3)1.7 (0.3)

1.7 (0.4)1.7 (0.5)1.9 (0.4)

10.8b

7.3a

7.5b

AppraisalsControl-MomControl-DadControl-PartnerThreat-MomThreat-DadThreat-Partner

2.0 (0.7)2.0 (0.7)1.8 (0.5)1.5 (0.6)1.4 (0.6)1.8 (0.6)

2.2 (0.8)2.3 (0.6)1.9 (0.7)1.8 (0.8)1.8 (0.9)2.3 (0.5)

2.54.8a

0.22.73.3

17.7c

ap < .05.bp < .01.cp < .001.

Attachment and Coping 555

Table II. Correlations Between Negative Affect, Appraisals, and Coping Styles

DepressionAnxietyNegative affectControlMagnitude (threat)

Mother

Sup

.37b

.21

.61b

.20

.37b

Avoid

.46b

.18

.53b

.15

.23

Prob

.19

.15

.33b

.19

.44b

Father

Sup

.02

.13

.15

.26a

.16

Avoid

.30a

.06

.36b

-.10-.11

Prob

-.18.07

-.18.33b

.26a

Dating Partner

Sup

.44b

.24

.40b

.04

.45b

Avoid

.29a

.20

.21

.01

.06

Prob

-.07.15.21.28a

.41b

ap < .05.bp < .01.

Table III. Coping Strategies as a Function of Attachment Style

Coping Strategies

Conflict with motherSupport seekingProblem solvingAvoidance

Conflict with fatherSupport seekingProblem solvingAvoidance

Conflict with dating partnerSupport seekingProblem solvingAvoidance

Secure (n = 44)M (SD)

1.8 (0.4)2.2 (0.6)2.1 (0.4)

2.1 (0.6)2.6 (0.9)1.9 (0.6)

1.9 (0.3)2.5 (0.6)2.2 (0.3)

Insecure (n = 29)M (SD)

2.1 (0.6)2.5 (0.7)2.4 (0.5)

2.4 (0.6)2.7 (0.7)2.1 (0.5)

2.2 (0.4)2.5 (0.6)2.5 (0.4)

F

8.9b

2.913.0c

3.10.14.1"

14.9c

0.412.5b

ap < .05.bp < .01.cp < .001.

Stepwise set hierarchical regression analysis was used, entering variables in thefollowing order: Step (1), general attachment style; Step (2), specific attachmentstyle; Step (3), affect (depression and negative affect in the context of conflict);Step (4), appraisals of control and threat; Step (5), control x threat interaction term.Variables were entered as a set, so the R 2 T, F, and p values presented in the ta-bles are estimates for the step, rather than for individual variables. Anxiety wasomitted from regression analysis because it was not significantly correlated withcoping styles. The dimensional scores for general and specific attachment style(Close/Depend, Anxiety) were entered in the regressions. Attachment variableswere entered first because they were conceptualized for this research as an orga-nizational construct for emotions and appraisals, which are proximal predictors ofcoping.

Torquati and Vazsonyi

Table IV. Set Hierarchical Regressions of Attachment, Affect, and Appraisals as Predictorsof Coping With Mother Conflicta

Step 1: General attachmentClose/DependAnxiety

Step 2: Specific attachmentClose/DependAnxiety

Step 3: AffectConflict emotionsDepression

Step 4: AppraisalsThreatControl

Step 5: Interaction effectThreat x Control

Total R2

Problem Solving

Beta

.08

.13

.03

.25

.35

.10

.16

.09

.14

R2T

.02

.05

.06

.02

.00

.15

F

0.7

1.2

1.6

1.4

1.2

Support Seeking

Beta

.23

.33

-.06.35

.59

.23

.25-.00

-.42

R2T

.16

.09

.20

.03

.02

.50

F

6.4c

5.6c

8.5"

7.3"

7.0d

Avoidance

Beta

.05

.33

-.07.16

.56

.32

.34

.02

-.96

R2T

.11

.02

.21

.12

.04

.44

F

4.3b

2.5b

5.7d

5.8d

5.4d

aVariables are entered as a set; K2 T's and F-'s are for the step.bp < .05.cp < .01.dp < .001.

Conflict With Mother

All three regression equations significantly predicted coping strategies dur-ing conflict with mother (Table IV). All five steps significantly predicted supportseeking and avoidance. Women who were more anxious in their general and spe-cific attachment, who were more depressed and experienced greater intensity ofnegative emotions in the conflict, and who perceived the conflict as a greater threatwere more likely to seek support and to avoid their mothers. Only the affect xappraisal interaction predicted problem solving.

Conflict With Father

Regression equations predicting problem solving and avoidance during con-flict with fathers were significant (Table V). All five steps were significant in eachequation. Standardized regression coefficients for attachment and emotion vari-ables indicated relationships in the expected direction: General attachment secu-rity was positively associated with problem solving and inversely associated withavoidance. Depression was positively associated with avoidance, and negativelyassociated with problem solving.

556

557

Conflict With Dating Partner

None of the predictors were significantly correlated with problem solvingduring conflict with a dating partner, so the multivariate model was not testedfor problem solving. General and specific attachment, affect, appraisals, and theaffect x appraisal interaction predicted support seeking and avoidance. Resultsare presented in Table VI.

DISCUSSION

The purpose of this research was to examine attachment style as an organi-zational construct for affect, appraisal, and coping. Results largely supported hy-potheses. Insecure general attachment was significantly correlated with all indicesof negative affect, and insecure specific attachment was significantly correlatedwith depression and with negative affect in the context of interpersonal conflict.Similarly, all five comparisons of affect as a function of attachment security weresignificant, indicating that participants with an insecure attachment style reportedmore negative affect in the context of interpersonal conflict, and higher levels of

Attachment and Coping

Table V. Set Hierarchical Regressions of Attachment, Affect, and Appraisals as Predictorsof Coping With Father Conflicta

Step 1: General attachmentClose/DependAnxiety

Step 2: Specific attachmentClose/DependAnxiety

Step 3: AffectConflict emotionsDepression

Step 4: AppraisalsThreatControl

Step 5: Interaction effectThreat x Control

Total R2

Problem Solving

Beta

.35

.06

.20-.21

-.06-.20

-.04.33

-.10

R2T

.11

.05

.03

.09

.00

.28

F

4.0b

3.2b

2.5b

2.9c

2.6b

Support Seeking

Beta

.21

.13

.14-.15

.34-.02

.14

.13

.02

R2T

.05

.03

07

.02

.00

.18

F

1.8

1.4

1.9

1.6

1.5

Avoidance

Beta

-.27.28

-.53.16

.06

.10

.06-.13

-.37

R2T

.14

.25

.01

.02

.00

.42

F

5.4b

10.3d

6.9d

5.3d

4.7d

aVariables are entered as a set; TR2's and F's are for the step.bp < .05.cp < .01.dp < .001.

Torquati and Vazsonyi

Table VI. Set Hierarchical Regressions of Attachment, Affect, andAppraisals as Predictors of Coping With Dating Partner Conflict

Step 1: General attachmentClose/DependAnxiety

Step 2: Specific attachmentClose/DependAnxiety

Step 3: AffectConflict emotionsDepression

Step 4: AppraisalsThreatControl

Step 5: Interaction effectThreat x Control

Total R2

Support Seeking

Beta

.12

.38

-.09.24

.29

.29

.27-.08

.95

R2T

.16

.03

.13

.06

.01

.39

F

6.4b

3.9b

5.1c

4.8c

4.5c

Avoidance

Beta

-.00.30

.24

.10

.14

.22

.24

.24

.13

R2T

.09

.04

.05

.09

.03

.30

F

3.4a

2.5a

2.4a

2.9b

3.0b

ap < .05.bp < .01.cp < .001.

depression and anxiety. This replicates previous work linking attachment stylewith affective disorder (e.g., Carnelley et al., 1994). Affective disorder and sit-uational emotions are qualitatively different constructs. Investigation of reportedemotions during conflict with mother, father, and dating partner represents a uniquecontribution to understanding attachment as an organizational construct for affect.

Attachment was also examined as an organizational construct for appraisals.Results partially supported hypotheses. Secure attachment style is associated withefficacious perceptions of the self in social situations and is analogous to greaterperceived control. Secure participants perceived greater control in conflicts withfathers, but comparisons of perceived control in conflicts with mothers and dat-ing partners were nonsignificant. Participants with an insecure attachment styleperceived conflict with their dating partners as a greater threat than did secureparticipants.

Appraisals of threat and control were examined as predictors of coping. Ap-praisals of threat were positively associated with problem solving in all threerelationships, and positively associated with support seeking with mother and dat-ing partner. It appears that perception of threat in the context of conflict can be amotivator for problem focused action. Appraisals of control significantly predictedproblem solving with father and dating partner, and support seeking with father.

Results also largely supported hypotheses regarding attachment as an organi-zational construct for coping: Insecure participants were more likely to use avoidant

558

strategies to cope with conflict in all three relationships and were more likely toseek support in response to conflict with mothers and dating partners. These re-sults are consistent with previous research indicating that insecure attachment isassociated with more emotion focused coping (Lussier et al., 1997). However, itshould be noted that seeking emotional and informational support from others canbe both emotion and problem focused.

Comparison of problem solving as a function of attachment security yieldedno significant differences; however, both general and specific attachment predictedproblem solving with fathers in the multivariate model. This is also consistentwith findings of Lussier et al. (1997), who reported an association between se-cure attachment style and problem focused coping in the context of interpersonalconflict. Previous research has not documented an association between attach-ment and problem focused coping in response to non-interpersonal stressors (e.g.Mikulincer and Florian, 1995), underscoring the importance of attachment as anorganizational construct for appraisals specific to relational stressors.

General affective dispositions and situational emotions were examined aspredictors of coping. It appears from these data that state anxiety is not significantlyrelated to coping. Conversely, depression was associated with avoidance in all threerelationships and with support seeking in response to conflict with mothers anddating partners.

Consistent with hypotheses, negative affect in the context of conflict waspositively related to avoidance in conflicts with mothers and fathers, and positivelyassociated with support seeking during conflicts with mothers and dating partners.Contrary to hypotheses, negative emotion was positively associated with problemsolving in conflicts with mother and with dating partner. This is inconsistent withprevious research (e.g., Greene and Noice, 1988). However, this finding raises aninteresting question: Why invest effort in solving a problem if it is not distressing foryou or your relationship partner? Negative emotion can serve an adaptive functionin relationships. Just as love and protectiveness function to motivate adults to carefor children and security needs promote proximity seeking, anger and fear cansignal disequilibrium in the relationship and motivate action to regain equilibrium.Anger preceding aggression in abusive relationships should be avoided, and canbe a signal to flee. Avoidance is an appropriate response in high danger situations,or situations in which there is little or no control over outcomes. Avoidance isproblematic if it is inappropriately generalized across stressor situations. Similarly,negative affect can be adaptive if it motivates action that moves a relationship toequilibrium, and can be maladaptive if it is disruptive to the relationship and/or tothe health and development of individuals in the relationship.

Cognitive appraisals have been extensively examined in coping literature. Em-phasis on cognitive appraisals reflects a rational bias in models of behavior. Resultsof this research indicate that the role of emotions in predicting coping should beexamined more closely. Contextual emotions were as important as cognitive ap-praisals in these analyses. Moreover, negative emotions predicted problem solving

Attachment and Coping 559

with mothers—indicating that negative emotion can be a motivator for action.Negative emotion is typically viewed as maladaptive. However, negative emotioncan serve an adaptive function in relationships.

Results reported here are consistent with those of Simpson (1990) and Collins(1996), in which both preoccupied and dismissing participants reported greaterdistress in their romantic relationships, and with those of Feeney and Kirkpatrick(1996) in which both preoccupied and dismissing college women experiencedgreater autonomic arousal during a laboratory separation than secure participants. Itis noteworthy that similar results have been obtained in each of the studies using dis-similar methods. However, findings differ from those of Simpson et al. (1992), pos-sibly because in this research the origin of the threat is within the relationship, andin the Simpson research the locus of the threat was in the research context—makingthe dating partner a potential source of support in the context of perceived threat.

This raises an interesting question regarding attachment as an organizationalconstruct for coping. Attachment theory suggests that internal working models ofrelationships provide a heuristic for processing social information. To what extentdo attachment style and patterns of appraisal, emotion, and coping within relation-ships generalize to non-interpersonal stressors? The influence of attachment styleon coping with interpersonal (Lussier et al., 1997) and non-interpersonal stressorshas been examined (e.g., Mikulincer and Florian, 1995). Investigating attachmentas an organizational construct for coping with both types of stressors in the samesample is an important question for future research.

This research tested the direct contributions of attachment, affect, and ap-praisals in the prediction of coping with conflict. However, results suggest that amediating model in which attachment predicts affect and appraisals, which in turnpredict coping is also plausible, and would be consistent with the conceptualizationof attachment as an organizational construct.

Limitations of the present research should be noted. Proportions of the sam-ple in each of the insecure attachment style categories were not sufficient to testdifferential hypotheses as a function of type of insecure attachment. It will beimportant for future research to delineate variability in coping strategies as a func-tion of type of insecure attachment because, for example, relative activation anddeactivation can be considered strategies to regulate attachment related emotionsand cognitions.

The current sample was limited to females. Including males in future re-search is important for several reasons, including delineation of the differentialinfluence of attachment style with mothers and fathers on attachment and copingwith dating partners. Finally, there is potential for significant method variance asall data are self-report, although items describing attachment style are primarilycognitive and affective, and items describing coping are behavioral. Appraisals ofthreat and control were each measured by a single item, so reliability estimatedcould not be computed. There was potential for multicollinearity because of thesignificant correlations between negative affect and attachment styles. Replication

560 Torquati and Vazsonyi

and refinement should include a larger, more diverse sample, and employ multiple,dissimilar measures of constructs.

REFERENCES

Allen, J. P., Hauser, S. T., and Borman-Spurrell, E. (1996). Attachment theory as a framework forunderstanding sequelae of severe adolescent psychopathology: An 11-year follow-up study.J. Consult. Clin. Psychol. 64: 254-263.

Armsden, G.C., and Greenberg, M. T. (1987). The inventory of parent and peer attachment: Individualdifferences and their relationship to psychological well-being in adolescence. J. Youth Adoles.16(5): 427-454.

Baldwin, M. (1992). Relational schemas and the processing of social information. Psychol. Bull. 112:461-484.

Band, E. B., and Weisz, J. R. (1988). How to feel better when it feels bad: Children's perspectives oncoping with everyday stress. Develop. Psychol. 24(2): 247-253.

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., and Erbaugh, J. (1961). An inventory measuringdepression. Arch. Gen. Psychiatry 4: 53-63.

Billings, A. G., and Moos, R. H. (1981). The role of coping responses in attenuating the impact ofstressful life events. J. Behav. Med. 4: 139-157.

Billings, A. G., and Moos, R. H. (1984). Coping, stress, and social resources among adults with unipolardepression. J. Personal. Soc. Psychol. 46: 877-891.

Bowlby, J. (1980). Attachment and Loss: Loss, Sadness, and Depression. BasicBooks, New York.Bowlby, J, (1988). A Secure Base: Parent-Child Attachment and Healthy Human Development. Basic

Books, New York.Carnelley, K. B., Pietromonaco, P. R., and Jaffe, K. (1994). Depression, working models of others, and

relationship functioning. J. Personal. Soc. Psychol. 66: 127-140.Cavell, T. A., and Constantin, L. (1991). Attachment and social competence in early adolescence. Paper

presented at the Biennial Meeting of the Society for Research in Child Development, Seattle, WA.Collins, N. L. (1996). Working models of attachment: Implications for explanation, emotion, and

behavior. J. Personal. Soc. Psychol. 71: 810-832.Collins, N. L., and Read, S. J. (1990). Adult attachment, working models, and relationship quality in

dating couples. J. Personal. Soc. Psychol. 58: 644—663.Compas, B. E. (1987). Coping with stress during childhood and adolescence. Psychol. Bull. 101:

393-403.Compas, B. E., Forsythe, C. J., and Wagner, B. M. (I988a). Consistency and variability in causal

attributions and coping with stress. Cog. Therapy Res. 12: 305-320.Compas, B. E., Malcarne, V. L., and Fondacaro, K. M. (1988b). Coping with stressful events in older

children and young adolescents. J. Consult. Clin. Psychol. 56: 405-411.Compas, B. E., Banez, G. A., Malcarne, V., and Worsham, N. (1991). Perceived control and coping

with stress: A developmental perspective. J. Soc. Issues 47: 23-34.Dodge, K. A. (1991). Emotion and social information processing. In Dodge, K. A., and Garber, J.

(eds.), The Development of Emotion Regulation and Dysregulation. Cambridge University Press,Boston, pp. 159-181.

Easterbrook, J. A. (1959). The effects of emotion on cue utilization and the organization of behavior.Psychol. Rev. 66: 183-201.

Elicker, J. G. (1991). Attachment history, interpersonal sensitivity, and peer competence in preadoles-cence. Poster presented at the Biennial Meeting of the Society for Research in Child Development,Seattle, WA.

Endler, N. S., and Parker, J. D. A. (1990). Multidimensional assessment of coping: A critical evaluation.J. Personal. Soc. Psychol. 58: 844-854.

Eysenck, M. W. (1982). Attention and Arousal. Springer-Verlag, New York.Feeney, B. C., and Kirkpatrick, L. A. (1996). The effects of adult attachment and presence of romantic

partners on physiological responses to stress. J. Personal. Soc. Psychol. 70: 255-270.Forsythe, C. J., and Compas, B. E. (1987). Interaction of cognitive appraisals of stressful events and

coping: Testing the goodness of fit hypothesis. Cog. Therapy Res. 11: 473-485.

Attachment and Coping 561

Gamble, W. C. (1993). Perceptions of controllability and other stressor event characteristics as deter-minants of coping among young adolescents and young adults. J. Youth Adolesc, 23: 65-84.

George, C., Kaplan, N., and Main, M. (1985). The Attachment Interview for Adults. Unpublishedmanuscript, University of California, Berkeley, Department of Psychology.

Greene, T. R., and Noice, N. (1988). Influence of positive affect upon creative thinking and problemsolving in children. Psychol. Rep. 63: 895-899.

Griffin, D., and Bartholomew, K. (1994). Models of the self and other: Fundamental dimensionsunderlying measures of adult attachment. J. Personal. Soc. Psychol. 67: 430-445,

Hazan, C., and Shaver, P. (1987). Romantic love conceptualized as an attachment process. J. Personal.Soc. Psychol. 52: 511-524.

Kobak, R. R., and Hazan, C. (1991). Attachment in marriage: Effects of security and accuracy ofworking models. J. Personal. Soc. Psychol. 60: 861-869.

Kobak, R. R., Sudler, N., and Gamble, W. (1991). Attachment and depressive symptoms during ado-lescence: A developmental pathways analysis. Develop. Psychopathol. 3: 461-474.

Kobak, R., Cole, H., Ferenz-Gillies, R., Fleming, W., and Gamble, W. (1993). Attachment and emotionregulation during mother-teen problem solving: A control theory analysis. Child Develop. 64:231-245.

Kwakman, A. M., Zuiker, J. M., Schippers, G. M., and Wuffel, F. J. (1988). Drinking behavior, drinkingattitudes, and attachment relationship of adolescents. J. Youth Adolesc. 17: 247-253.

Lazarus, R. S., and Folkman, S. (1984). Stress, Appraisal, and Coping. Springer, New York.Lussier, Y., Sabourin, S., and Turgeon, C. (1997). Coping strategies as moderators of the relationship

between attachment and marital adjustment. J. Soc. Pers. Relat. 14: 777-791.Main, M., Kaplan, N., and Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move

to the level of representation. In Bretherton, I., and Waters, E. (eds.), Growing Points of Attach-ment Theory and Research. Serial No. 209. Monographs of the Society for Research in ChildDevelopment.

Mandler, G. (1975). Mind and Emotion. Wiley, London.Mikulincer, M., and Florian, V. (1995). Appraisal of and coping with a real-life stressful situation: The

contribution of attachment styles. Personal. Soc. Psychol. Bull. 21:406—414.Mikulincer, M., Florian, V., and Weller, A. (1993). Attachment styles, coping strategies, and posttrau-

matic psychological distress: The impact of the Gulf War in Israel. J. Personal. Soc. Psychol. 64:817-826.

Moss, R. H., and Billings, A. G. (1982). Conceptualizing and measuring coping resources and processes.In Goldberger, L., and Breznitz, S. (eds.), Handbook of Stress: Theoretical and Clinical Aspects.Macmillan, New York.

Papini, D. R., and Roggman, L. A. (1992). Adolescent perceived attachment to parents in relation tocompetence, depression, and anxiety: A longitudinal study. J, Early Adolesc. 12: 420-440.

Patterson, J. M., and McCubbin, H. I. (1987). Adolescent coping style and behaviors: Conceptualizationand measurement. J. Adolesc. 10: 163-186.

Pistole, M. C. (1989). Attachment in adult romantic relationships: Style of conflict resolution andrelationship satisfaction. J. Soc. Pers. Relat. 6: 505-510.

Rosenberg, S. J., Peterson, R. Z., and Hayes, J. R. (1987). Coping behaviors among depressed andnondepressed medical inpatients. J. Psychosom. Res. 31: 653-658.

Rosenstein, D. S., and Horowitz, H. A. (1996), Adolescent attachment and psychopathology. J. Consult.Clin. Psychol. 64: 244-253.

Simpson, J. A. (1990). Influence of attachment styles on romantic relationships. J. Personal. Soc.Psychol. 59: 971-980.

Simpson, J. A., Rholes, W. S., and Nelligan, J. S. (1992). Support seeking and support giving withincouples in an anxiety-provoking situation: The role of attachment styles. J. Personal. Soc. Psychol.62: 434-446.

Simpson, J. A., Rholes, W. S., and Phillips, D. (1996). Conflict in close relationships: An attachmentperspective. J. Personal. Soc. Psychol. 71: 899-914.

Speilberger, C. D. (1983). Manual for the State-Trait Anxiety Inventory. Consulting PsychologistsPress, Palo Alto, CA.

Sroufe, L., and Waters, E. (1977). Attachment as an organizational construct. Child Develop. 48:1184-1199.

562 Torquati and Vazsonyi