When Harry and Sally met Dick and Jane: Creating closeness … · 2010-08-06 · When Harry and Sally met Dick and Jane: Creating closeness between couples RICHARD B. SLATCHER Wayne

Personal Relationships, 17 (2010), 279–297. Printed in the United States of America.Copyright © 2010 IARR

When Harry and Sally met Dick and Jane: Creatingcloseness between couples

RICHARD B. SLATCHER

Wayne State University

AbstractThis study investigated how friendships between couples form and implications for within-couple process. Sixtycouples were randomly assigned to 1 of 2 conditions where they engaged in a 45-min interaction with anothercouple. In 1 condition, couples carried out self-disclosure tasks; in the other, couples engaged in nonemotional smalltalk. Compared to the small-talk condition, those in the high-disclosure condition felt closer to the couples theyinteracted with and were more likely to meet up with them again during the following month. Further, couples in thehigh-disclosure condition felt closer to their own partners. Actor–partner interdependence model analyses showedthese effects to be mediated by increases in positive affect. Implications for studying the interplay of social networksand romantic relationships are discussed.

Romantic relationships do not occur in a vac-uum. They begin, develop, and change withina larger environment. The individual charac-teristics of couple members (e.g., their per-sonalities, feelings of attachment) and theone-on-one interactions that couple membershave with each other are critical forces inshaping the future path of a romantic relation-ship, but the environmental context in whichcouples interact is vitally important as well.This environment includes both the physicalcontext (e.g., proximity and physical setting)and social context (e.g., family and friends) in

Richard B. Slatcher, Department of Psychology, WayneState University.

Portions of this research were funded by a grant fromthe National Institutes of Health (MH52391).

I would like to thank April Allen and SamanthaBeard for their assistance in data collection; David Kennyand Tessa West for their consultation regarding dataanalyses; James Pennebaker, Rebecca Bigler, SamuelGosling, Robert Josephs, and Timothy Loving for theirfeedback at various phases of this project; and TheodoreRobles and Anne Peplau for their helpful comments onan earlier draft of this manuscript.

Correspondence should be addressed to Richard B.Slatcher, Department of Psychology, Wayne State Uni-versity, 5057 Woodward Ave., 7th Floor, Detroit, MI,48202, e-mail: [email protected].

which couples are embedded. Couples’ friend-ships with those in their social networks maybe particularly relevant determinants of whatmakes for a strong and stable romantic rela-tionship (Agnew, Loving, & Drigotas, 2001;Milardo, 1982; Sprecher, Felmlee, Orbuch, &Willetts, 2002).

There is growing empirical evidence thatshared friendships are beneficial for couples.One of the most robust findings in this areaof research is that couples who have a largerpercentage of shared friends (vs. individualfriends) tend to have happier and longer last-ing relationships (Ackerman, 1963; Agnewet al., 2001; Milardo, 1988). From a soci-ological perspective, the more that couplesare integrated into their social networks, themore likely they are to have happy and satis-fying romantic relationships. However, it isunknown whether social networks have aninherent benefit for couples or just that peo-ple who are happier in their relationships aremore likely to make friendships with otherstogether as a couple. By studying how friend-ships between couples form in a controlledlaboratory setting, we may be able to bet-ter understand the processes through which

279

280 R. B. Slatcher

friendships between couples are generated,and gain insight into the directionality ofsocial network–romantic relationship qualitylinks. For example, when two couples firstmeet at a cocktail party or other social gath-ering, why do some hit it off and become fastfriends while others leave it at “nice to meetyou”? And when couples do hit it off, arethere residual benefits for the couples them-selves? The purpose of this article is to exam-ine how friendships between couples form andpotential implications for within-couple pro-cess (e.g., the effects of friendships betweencouples on feelings of closeness within acouple).

Examining the formation of friendshipsbetween couples using thecloseness-induction paradigm

One of the challenges facing researchers hasbeen finding ways to study how friendshipsbetween couples form in a controlled environ-ment. A paradigm that is potentially suitablefor addressing this challenge is the close-ness induction method by Aron and col-leagues (Aron, Melinat, Aron, Vallone, &Bator, 1997). In this procedure, closeness isgenerated between individuals through con-versational activities that primarily involveincreasing levels of self-disclosure. Sustained,escalating, and reciprocal self-disclosure isone of the keys to the establishment of closerelationships with others (Altman & Taylor,1973; Collins & Miller, 1994; Derlega, Metts,Petronio, & Margulis, 1993; Laurenceau,Feldman Barrett, & Pietromonaco, 1998; Lau-renceau, Rivera, Shaffer, & Pietromonaco,2004; Reis & Shaver, 1988; Slatcher & Pen-nebaker, 2006).

In a typical closeness-induction study, twostrangers are brought into the laboratory tohave a 45-min get-to-know-you conversation.They are randomly assigned to be in eithera high-disclosure condition or a small-talkcondition. Those in the high-disclosure con-dition are given questions to ask each otherthat are geared toward eliciting high levelsof self-disclosure from the respondent andother questions geared toward creating pos-itive affect and intimacy between the two

unacquainted individuals. Those in the small-talk condition are given questions to ask eachother that involve very minimal amounts ofself-disclosure and cover relatively unemo-tional topics. Compared to those in the small-talk condition, those in the high-disclosurecondition feel much closer to each other,regardless of how much they expect to likeeach other or whether they agree or disagreeon important topics (e.g., political attitudes)prior to the interaction.

By applying this paradigm to pairs of cou-ples and by following up with them afterthey leave the laboratory, one can examinehow friendships between couples form. Addi-tionally, this paradigm allows one to inves-tigate whether high levels of self-disclosurebetween couples result in stronger bondswithin couples and investigate possible mech-anisms underlying these effects.

Why might high levels of disclosure betweencouples increase closeness within couples?

One possibility for why experimentallyinduced disclosure between couples mightenhance closeness within couples is that itis a novel and arousing activity. A num-ber of recent studies by Aron and colleaguessuggest that couples’ participation in noveland arousing activities is associated withincreases in positive affect (Aron, Norman,Aron, McKenna, & Heyman, 2000; Strong &Aron, 2006). Such increases in positive affect,in turn, have been found to lead to increasesin self-reported relationship quality.

The theoretical foundation for this ideais based on Aron’s self-expansion model ofmotivation and cognition in close relation-ships (Aron, Aron, & Norman, 2001). Accord-ing to this model, when two people begin arelationship, each begins to “include the otherin the self.” Inclusion of other in the selfdescribes a process by which one begins toidentify his or her self-image as a new com-bination with the other’s self. The partner’sidentity, his or her beliefs, feelings, ideology,resources, and personality begin to becomeassociated with one’s own self. By associat-ing such unique aspects of one’s partner toone’s already defined self, the self expands

Creating closeness between couples 281

to include these new aspects of the partner.This process of self-expansion typically takesplace through mutual time spent with part-ner, shared activities, and common ideasand interests (Strong & Aron, 2006). Suchactivities are associated with rapid increasesin positive affect—indicated by feelings ofgreat pleasure, arousal, and excitement (Aronet al., 2001).

Self-expansion theory suggests that novelactivities—such as having a highly disclos-ing conversation with another couple—createstrong positive affect, which becomes asso-ciated with one’s partner through positivereinforcement, in turn leading to greater feel-ings of closeness toward one’s partner. Thefirst link in this proposed meeting anothercouple→positive affect→relationship close-ness chain of associations is suggested inseries of two daily experience sampling stud-ies conducted by Larson and colleagues (Lar-son & Bradney, 1988; Larson, Mannell, &Zuzanek, 1986). In these studies, participantsreported their highest levels of positive affectwhen they were engaging in activities withtheir spouses and friends together, comparedto activities alone, with spouse only or withfriends only. These findings provide prelimi-nary evidence that increases in positive affectmay be one mechanism through which friend-ships with other couples may lead to enhancedcloseness within couples.

When two couples interact in an intense,self-disclosing fashion, rapid expansion of theself may be especially strong after the initialexhilaration of a new relationship starts tofade. Take, for example, relationship partnerswho have been dating for a year and havebecome more and more accustomed to eachother as their relationship has progressed. Thetwo people, once very fresh and exciting toeach other, seem less and less novel andself-expansion slows or comes to a halt.At this point, people may become boredand dissatisfied in their relationship (Aron &Aron, 1986). For long-term couples, intenseself-disclosure with another couple may besufficiently novel and arousing enough tolead to increased positive affect and, in turn,heightened feelings of closeness toward one’spartner.

There are, of course, alternative expla-nations for why self-disclosing interactionsbetween couples might lead to greater feel-ings of closeness within couples. The mostlikely alternative is that people may learnnew things about their romantic partners aftergoing through a closeness-induction interac-tion with another couple as a result of the highlevels of self-disclosure that are elicited inthis type of interaction. Indeed, learning newthings about one’s partner via partner disclo-sures is linked to greater feelings of closenessin the early phases of one’s relationship (Aronet al., 2001); it is possible that learning newthings about one’s partner at later phases ofa relationship may lead to increased close-ness as well. Second, it is possible that thesheer novelty of this type of laboratory-basedinteraction might make couple members feelcloser to each other. Although it has beenargued that positive affect is the mediatingmechanism through which novel and arous-ing activities enhance relationship functioning(Strong & Aron, 2006), the effects of noveltyand positive affect have not yet been directlycompared.

Overview of the current studyand research questions

In the current study, 60 heterosexual couplesin committed dating relationships completedbaseline measures of relationship quality andpositive affect and then were randomly as-signed to one of two conditions where theyengaged in a 45-min interaction with anothercouple. In the high-disclosure condition, pairsof couples carried out self-disclosure tasksthat gradually escalated in intensity. In thesmall-talk condition, pairs of couples engagedin nonemotional small-talk discussions. Mea-sures were administered to assess feelings ofcloseness toward the other couple, measuresof closeness within couples, and measuresassessing possible mediators of the effectsof the manipulation on feelings of closenesswithin couples, including positive affect, nov-elty, and learning new things about one’s part-ner. One month later, participants were askedto complete a brief online follow-up measureto assess long-term effects of the manipulation

282 R. B. Slatcher

on perceptions of the other couple, feelingsof closeness toward romantic partners andwhether or not they had contacted the othercouple.

Using this design, the following hypothe-ses were tested: (a) couples in the high-disclosure condition would feel closer to theother couples compared to those in the small-talk condition and would be more likely tomeet up with them again after the experi-ment, (b) couples in the high-disclosure con-dition would feel closer to their own partnerscompared to those in the small-talk condi-tion, and (c) the effects of the manipulation oncloseness within couples would be mediatedby increases in positive affect. This designalso made it possible to test whether effectof the manipulation on closeness within cou-ples alternatively might be due to the noveltyof the interaction or to simply learning newthings about one’s partner. Further, with thefollow-up data, it was possible to investigatewhether couples who attempt to form friend-ships with other couples are happier in theirrelationships to begin with than those whodo not. Finally, potential moderators of theeffects of the manipulation—including gen-der, ethnicity, and length of time dating—alsowere examined.

Method

Participants

Sixty couples in dating relationships fromthe Austin, Texas metropolitan area, wererecruited via flyers and advertisements postedon the websites Craig’s List and Facebook.Couples were recruited on the basis that theywere unmarried and had been dating theirpartner for at least 1 year. Participants rangedin age from 19 to 26 (M = 20.83, SD = 1.73).The ethnic makeup of the sample was 62.5%White/Caucasian, 16.7% Asian, 14.2% His-panic/Latino, 1.7% Black/African American,and 5% indicating other ethnicity. Coupleshad been dating from 1 to 5 years (M = 2.04,SD = 1.05) and indicated that they eitherwere in a serious dating relationship (85.8%)or life partnership (14.2%); none were mar-ried or had children. Each couple was paid$20 for participating in the study.

Procedure

Unacquainted couples came to the laboratoryin pairs. On couples’ arrival at the labora-tory, the experimenter checked to make surethat the pairs of couples did not know eachother; none did. After getting informed con-sent from participants, the experimenter ledeach person to one of four separate roomsto complete baseline questionnaires. The pairsof couples then were brought into a room toengage in one of two types of discussion tasksdescribed below. Subsequently, couples filledout postinteraction questionnaires; all mea-sures at baseline and after the interaction werecounterbalanced to prevent order effects.

One month later, both members of eachcouple were contacted to fill out a final onlinemeasure, assessing their perceptions of theother couple, feelings of closeness towardtheir own partner, and the extent to which theytalked to each other about the interaction withthe other couple over the past month. Addi-tionally, they were asked whether they or theirromantic partner had been in contact with andmet up with other couple. Of the original120 participants in the study, 103 (86%) com-pleted the follow-up measure. Participation atTime II was completely voluntary and unpaid.

Experimental manipulation: Discussion topic(closeness or small talk)

After completing baseline measures, the pairsof couples were randomly assigned to engagein a discussion primarily involving increasingamounts of self-disclosure (high-disclosurecondition) or a discussion involving unemo-tional everyday topics (small-talk condition)for 45 min (a complete description of this pro-cedure may be found in Aron et al., 1997).After being brought into a room with twosets of comfortable chairs, one participantwas given a packet with three sets of slips.The experimenter then read the task instruc-tions aloud to the group and then left theroom. These basic initial instructions were thesame for both the closeness and small-talkconditions.

Groups then began at once with the firstSet I slip. The participant with the slips would


read the question printed on each slip tothe group, and each member of the groupanswered the question in turn. After 15 min,the experimenter came back into the room totell the couples to stop, put away the Set Islips, and begin Set II; after another 15 min,to begin Set III; and after a final 15 min,to stop completely. In the high-disclosurecondition, questions on the slips were gearedto elicit high levels of self-disclosure, with theintensity of the questions gradually increasing(three sets are used so that pairs of coupleswho go slowly through the tasks will doat least some of the fairly intense Set IIItasks). Examples of some of these questionsinclude “For what in your life do you feel themost grateful?” and “How close and warm isyour family? Do you feel your childhood washappier than most other people’s?”

Those in the small-talk condition simi-larly went through three sets of slips, but inthis condition, questions focused on everyday,unemotional activities involving very littlepersonal disclosure, such as, “When was thelast time you walked for more than an hour?”and “Describe where you went and what yousaw.” After the 45-min interaction sessionscame to an end, participants were separatedto complete postinteraction questionnaires inindividual laboratory rooms. They then werepaid and dismissed from the laboratory. Cou-ples were neither encouraged nor discouragedfrom contacting each other after completion ofthe study. Any exchanges of contact informa-tion occurred outside of the experiment, forexample, couples providing each other withphone numbers as they left the building.

Baseline measures

Relationship satisfaction

Baseline relationship satisfaction was mea-sured using Hendrick’s (1988) RelationshipAssessment Scale (RAS). The RAS is a vali-dated measure of relationship satisfaction thatconsists of seven items on a 5-point Likert-type scale such as, “In general, how satisfiedare you with your relationship?” It has goodinternal reliability (α = .83 in the current sam-ple) and correlates strongly with measures

of love, commitment, investment, and dyadicadjustment.

Relationship closeness

Baseline closeness was measured using theSternberg Intimacy Scale (SIS; Sternberg,1997). The SIS contains twelve 7-point Likert-type scale items (e.g., “I have a warm relation-ship with my partner”) and correlates moder-ately with other measures of closeness andintimacy. Internal consistency (α) for the SISin this sample was .86. The SIS was chosenfor two main reasons: (a) so that participants’responses on the IOS scale administered afterthe experiment would be less biased by base-line responses and (b) in the initial validationof the IOS, the measure that correlated moststrongly with it (r = .45) was the SIS (Aronet al., 1992).

Positive affect

Baseline positive affect was measured usingthe Vigor subscale of the Profile of MoodStates (POMS; McNair, Lorr, & Droppleman,1971). This subscale of the POMS containseight feeling descriptors such as lively, active,and energetic on a scale from 0 (not at all ) to4 (extremely). The POMS is a broad measureof trait-level positive affect and has exhib-ited very good internal consistency (α = .84in the current sample). The scale was cho-sen because of its wide use and because ofits strong correlation (r = .43 in this sam-ple) with the positive affect subscale of thePositive Affect and Negative Affect Scale(PANAS; Watson & Clark, 1994), which wasthe postinteraction measure of positive affectused in this study.

Postinteraction measures

Closeness with the other couple

Closeness with the other couple was measuredusing a tailored version of Aron colleagues’Inclusion of Other in the Self Scale (IOS;Aron, Aron, & Smollan, 1992). The IOSconsists of seven pairs of circles labeled“Self” and “Other” that overlap to variousdegrees, creating a 7-point interval scale.

284 R. B. Slatcher

Respondents choose the pair that they feelbest describes their relationship. The IOShas exhibited high test–retest and alternate-form reliability (.85 and .92, respectively)and convergent and discriminant validity withother measures of closeness and intimacy. Inthis case, the measure was altered so that eachof the seven pairs of circles was labeled “Mypartner and I” and “The other couple.”

Relationship closeness

Postinteraction closeness with romantic part-ners was measured using the standard ver-sion of Aron colleagues’ (1992) previouslydescribed IOS scale. The standard IOS hasconvergent and discriminant validity withother measures of closeness and intimacy andhas been shown to be a good predictor of rela-tionship stability—as good as or better thanseveral longer and more elaborate measuresof closeness (Aron et al., 1992).

Positive affect

Postinteraction positive affect was measuredusing the Positive Affect subscale of thePANAS (Watson, Clark, & Tellegen, 1988),one of the most widely used measures ofaffect. The Positive Affect scale on thePANAS is made up of 10 items (e.g., “inter-ested,” “strong”) on a Likert-type scale rang-ing from 1 (not at all ) to 7 (extremely) thatparticipants use to rate how they feel at a par-ticular moment. The internal reliability of forthis scale is very good (α = .93 in the currentsample).

Newly learned knowledge about one’spartner

In order to test the alternative hypothesisthat increased closeness within couples is afunction of simply learning new informationabout one’s partner in the high-disclosurecondition, a single-item question on a 9-pointLikert-type scale asked, “To what extent doyou feel as though you gained new knowledgeabout your partner today?”

Novelty of interaction

In order to test the other alternative hypothesisthat increased closeness within couples is

explained by the novelty of the interaction,a multi-item scale of interaction novelty wasconstructed for this study. A copy of thescale may be found in the Appendix. Internalconsistency for the measure was α = .72 inthis sample.

Time II measures

Time II measures were designed to be veryshort to reduce attrition; items were chosento be as broad and encompassing as possible.The first item assessed the extent to whichthe participant had talked about the interac-tion with his or her romantic partner afterthe study. The second item assessed inter-est in “hanging out” with the other coupleagain in the future. The third item assessed thehope that the other couple would stay togetherin the future. The fourth item assessed howmuch participants thought the other couplehoped that their relationship (that of the par-ticipant and his or her partner) would staytogether; these first four items were all on a9-point scale (1 = not at all, 9 = very much).The fifth item was the IOS scale, used toassess closeness with one’s romantic part-ner. In addition to these items, the ques-tionnaire contained an item asking whetherthe respondent or their romantic partner hadbeen in contact with the other couple sincethe study ended; this item was at the couplelevel to maximize statistical power in case ofhigh attrition and/or low frequency of contactbetween couples at Time II. Finally, an itemwas included asking whether participants andtheir partners had together met in person withthe other couple since the study ended.

Results

Baseline relationship characteristics

As shown in Table 1, couples did not differ atbaseline between conditions in length of timedating or positive affect (POMS). However,there was a trend of couples in the small-talk condition having slightly higher levelsof relationship satisfaction (RAS) comparedto those in the high-disclosure condition, andthose in the small-talk condition felt signifi-cantly closer (SIS) to their romantic partners


Table 1. Baseline relationship characteristics as a function of group

Experimental condition

Small-talk High-disclosureMeasure M (SD) M (SD) t p Cohen’s d

Length of time dating partner 2.01 (1.00) 2.07 (1.11) 0.32 .75 0.06Closeness with romantic partner (SIS) 6.72 (0.35) 6.55 (0.51) 2.19 .03 0.40Relationship satisfaction (RAS) 4.58 (0.41) 4.45 (0.54) 1.52 .13 0.30Positive Affect (POMS) 18.87 (5.14) 18.23 (5.87) 0.63 .53 0.12

Note. N = 15 groups (four participants in each group) in each condition. SIS = Sternberg Intimacy Scale; RAS =Relationship Assessment Scale; POMS = Profile of Mood States.

than did those in the high-disclosure con-dition. Accordingly, the posttest measure ofrelationship closeness controlled for baselinecloseness. Additionally, the posttest measureof positive affect controlled for baseline posi-tive affect so that it reflected changes in pos-itive affect over the course of the interactionrather than simply mean levels.

Overview of data analytic strategy: Theactor–partner interdependence model

A unique characteristic of dyadic data is thatthe data from two couple members are notindependent. For example, people who aresatisfied in their romantic relationship tendto have romantic partners who also are sat-isfied, people who are optimistic tend tohave optimistic romantic partners, and soon. In order to account for this noninde-pendence in statistical analyses, relationshipresearchers in recent years often have framedtheir analyses in the actor–partner interdepen-dence model (APIM; Kashy & Kenny, 2000;Kenny, 1996).

APIM is a technique designed to addressnonindependence in dyadic analysis. Thistechnique allows researchers to estimate, forexample, the influence of one person’s behav-ior (e.g., expressions of positive affect) on herown feelings of closeness toward her romanticpartner—actor effects—as well as the effectsof her behavior on her partner’s feelings ofcloseness toward her—partner effects. Fur-thermore, APIM can be used to test mediationand moderation (Campbell & Kashy, 2002).For example, APIM can be used to estimatethe effects of the experimental manipulation

used here on within-couple relationship close-ness and whether the effects of the manipula-tion on closeness are mediated by a person’sown level of positive affect (actor effect), aswell as by his or her partner’s level of positiveaffect (partner effect).

In this study, the APIM analyses are some-what more complex because they involve twocouples interacting with each other rather thantwo individuals interacting. In this case, thepeople within each couple are distinguishableby gender, but the two couples within thegroup are indistinguishable from each other(Olsen & Kenny, 2006). The model con-structed for this special case of the APIM maybe found in Figure 1.

In this mediation model, experimentalcondition (small-talk or closeness, dummycoded as 0 and 1, respectively) directlyaffects the hypothesized mediator, X (positiveaffect), as well as directly affecting each per-son’s outcome variable, Y (e.g., within-couplecloseness). There are also indirect effectsfrom experimental condition that are mediatedthrough X to affect Y. In this model, each per-son’s behavior affects his own outcome (actoreffect) as well as the outcome of his conver-sational partners (partner effects)—includingthe outcome of his romantic partner as wellas the outcomes of each member of the othercouple. This model can allow one to investi-gate, for example, the extent to which exper-imental condition drives each person’s levelof positive affect, and how each person’s levelof positive affect in turn drives his ownlevel of closeness as well as driving everyoneelse’s levels of closeness.

286 R. B. Slatcher

g r

g r

0, t

e

0, t

e

h s

s

0, t1

e

e

k

o

l

l

p

k

p

j

n

m

0, i e

0, i e

f1

1

0, i1 e

0, i1 e

f2

f3

b

b

b1

n

m

c1

c

b1

c1

c

o

j

h

q

q

Experimental Condition

XF1

XF2

XM1

XM2

YF1

YF2

YM1

YM2

f2

f4 f4 w4w4

w1 w2

w2w3

0, t1

Figure 1. Actor–partner interdependence model (APIM) used to estimate the mediated effectsof experimental condition on outcome variables with pairs of couples.Note. F1 = female in couple 1; F2 = female in couple 2; M1 = male in couple 1; M2 = malein couple 2. The parameters of this model include direct paths (b and b1) from the predictor(experimental condition) to the mediators (XF1, XF2, XM1, and XM2), direct paths (c and c1)from the predictor to the outcome variables (YF1, YF2, YM1, and YM2), direct actor effects (jand k), direct partner effects (l, m, n, o, p, q), mediator intercepts (g and h), mediator residualvariances (i and i1), outcome intercepts (r and s), outcome residual variances (t and t1), andmediator residual (f1, f2, f3, and f4), and outcome residual (w1, w2, w3, and w4) covariances.

The results of APIM analyses are shownfirst for the main effects of condition on theoutcome variables of interest (Hypotheses 1and 2) and then for the mediation analysis(Hypothesis 3). Except when otherwise noted,structural equation modeling (SEM) was usedfor all analyses.

Postinteraction group differences

As shown in Table 2, participants in the high-disclosure condition felt closer to the othercouples they interacted with than did those inthe small-talk condition. Further, those in thehigh-disclosure condition felt closer to theirromantic partners following the interaction

than did those in the small-talk condition. Thisfinding is particularly striking considering thatthe mean romantic partner closeness for par-ticipants in both conditions was greater than6 on a 7-point scale—a large ceiling effect.Those in the high-disclosure condition alsoreported greater increases in positive affectafter the interaction, reported that the inter-action was more novel, and reported that theylearned more new things about their roman-tic partner during the interaction compared tothose in the small-talk condition. However,both groups scored below the midpoint of 5on this scale, in essence signaling that peo-ple learned very few new things about theirpartners though this exercise, regardless of


Table 2. Group differences in postinteraction outcome measures


Small talk High disclosureMeasure M (SD) M (SD) β p Cohen’s d

Closeness with other couple (IOS) 3.35 (1.39) 4.75 (1.31) .50 .000 1.05Closeness with romantic partner (IOS) 6.37 (0.80) 6.63 (0.58) .20 .039 0.39Positive affect (PANAS) 4.31 (1.28) 5.26 (0.91) .47 .000 0.86Novelty of interaction 6.07 (1.57) 7.23 (1.46) .38 .000 0.78Learned new things about romantic partner 2.63 (1.68) 4.12 (2.59) .31 .000 0.68

Note. N = 15 groups (four participants in each group) in each condition. βs represent standardized beta weights fromactor–partner interdependence model analyses. Means for closeness with romantic partner and positive affect areadjusted to control for respective baseline measures. IOS = Inclusion of Other in the Self Scale; PANAS = PositiveAffect and Negative Affect Scale.

condition. There were no gender differencesin any of these effects.

Group differences at Time II (1-monthfollow-up)

Of the 60 participants in the small-talk con-dition, 50 completed the follow-up mea-sure 1 month after interacting with the othercouple, while 53 out of 60 participantsin the high-disclosure condition completedit; at least 1 member from every cou-ple completed the follow-up measure. Attri-tion analyses detected no differences in anyTime I measures (both at baseline and post-interaction) between those who completed theTime II measures and those who did not.

As shown in Table 3, couples in the high-disclosure condition talked significantly moreto each other about the couple they had metin the month after initially taking part in thestudy than did those in the small-talk con-dition. They were also more interested in“hanging out” with the other couple again inthe future, hoped more that the other cou-ple would stay together in the future, andthought that the other couple would hopemore that they (the respondent and their part-ner) would stay together in the future aswell. This suggests that compared to cou-ples in the small-talk condition, couples inthe high-disclosure condition were more com-mitted to the success of the other coupleand thought that the other couple was more

committed to theirs. Additionally, there wasa marginally significant effect of those inthe high-disclosure condition to feel closer totheir romantic partners in the month after thestudy compared to those in the small-talk con-dition. There were no gender differences forany of the Time II measures.

The follow-up questionnaire also assessedwhether couples had been in contact with theother couples they met in the study and, ifso, whether they had actually gotten togetherwith them. In the high-disclosure condition,10 out of the 30 couples had contacted thecouple they had met in the study—either byphone, by e-mail, or in person. In contrast,none of the couples in the small-talk condi-tion had initiated contact with the couples theyhad met. A chi-square analysis conducted atthe group level showed that couples in thehigh-disclosure condition were significantlymore likely than those in the small-talk con-dition to contact each other, χ2(1) = 12.00,p < .001, � = .45. Further, among those whohad made contact with the other couples, fourhad reported getting together on two sepa-rate occasions and two had reported gettingtogether on three separate occasions. A chi-square analysis showed that couples in thehigh-disclosure condition were significantlymore likely than those in the small-talk con-dition to get together with the other couple,χ2(1) = 6.67, p < .01, � = .33. Thus, cou-ples in the high-disclosure condition not onlywere more likely than those in the small-talk

288 R. B. Slatcher

Table 3. Group differences at time II (1-month follow-up)


Small talk High disclosureMeasure M (SD) M (SD) β p Cohen’s d

Amount of talk about the othercouple after the study wasfinished

1.58 (0.78) 3.00 (2.47) .31 .000 0.77

Interest in hanging out with thiscouple again in the future

3.24 (1.87) 5.21 (2.20) .45 .000 0.97

Hope that the other couple staystogether in the future

5.64 (2.17) 7.30 (1.60) .42 .000 0.88

How much participant thinks theother couple hopes participantand partner stay together

4.98 (2.04) 6.91 (1.82) .48 .000 1.01

Closeness with romantic partner(IOS)

5.65 (1.36) 6.08 (1.11) .19 .076 0.34

Note. n = 50 in small-talk condition and 53 in high-disclosure condition. Mean Inclusion of Other in the Self Scale(IOS) scores adjusted to control for baseline closeness. βs represent standardized beta weights from actor–partnerinterdependence model analyses.

condition to have lasting positive feelingstoward the couples they had met, but theywere also more likely to contact them after-ward and to meet them in person—precursorsto real friendship.

Are couples who contact each other happierin their own relationships?

Although voluntary contact with another cou-ple does not constitute a true shared friend-ship per se, it does suggest proactive stepstoward friendship. With the couples in thehigh-disclosure condition in this study, we cansee whether couples who contacted the othercouples were happier in their own relation-ships at baseline compared to couples whomade no attempt at contacting the other cou-ples. This would provide evidence that cou-ples who are happier in their relationshipsare more likely to seek out friendships withother couples than couples who are less happyin their own relationships. However, analysesindicated that compared to couples who didnot make contact with the other couples, cou-ples who contacted the other couples felt nocloser to their partners (contact M = 6.68 andSD = 0.37; no contact M = 6.48 and SD =

0.56), t (58) = 1.34, p = .17, d = 0.35, andno more satisfied with their partners (contactM = 4.54 and SD = 0.34; no contact M =4.40 and SD = 0.62), t (58) = 0.99, p = .33,d = 0.26, at baseline. Although this providespreliminary evidence that the formation offriendships between couples is not driven bypreexisting relationship quality, interpretationis qualified by the small sample size of thegroup and the definition of friendship (contactwith other couples) used here.

Mediation of the effects of the manipulationon within-couple closeness by positive affect

Mediation of the effect of the manipula-tion on within-couple closeness by positiveaffect next was tested. The APIM media-tion analyses presented below follow the foursteps of mediation recommended by Baronand Kenny (1986). Following this approach,mediation is inferred when: (a) the indepen-dent variable significantly affects the media-tor, (b) the independent variable significantlyaffects the dependent variable in the absenceof the mediator, (c) the mediator has a sig-nificant unique effect on the dependent vari-able, and (d) the effect of the independent


variable on the dependent variable shrinksor is reduced to zero upon the addition ofthe mediator to the model. A direct pathfrom the independent variable to the medi-ator that is not statistically different fromzero would support mediation. Formal testsof mediated paths followed the recommenda-tions of Shrout and Bolger (2002). Specifi-cally, a bootstrap method in SEM was used toobtain 95% confidence intervals (CIs) and sig-nificance tests of mediated paths. This methodis slightly more accurate than the conventionalSobel test and is recommended when conduct-ing mediation analyses with small to moderatesample sizes.

As shown in Figure 2, positive affect sig-nificantly mediated the association betweenexperimental condition and closeness withromantic partners. There were significant

actor effects for women, significant partnereffects from the woman in one couple to theman in the other couple, and from the womanin one couple to the man in the same couple.

The bootstrap tests indicated that the medi-ated paths were significant: mediated paths forwomen (standardized indirect effect) = .09,p = .03, 95% CI = (.01, .55); mediated pathsfor men (standardized indirect effect) = .19,p = .02, 95% CI = (.02, .43). The analysesalso suggested that the mediation was full forboth women and men. After controlling forthe mediated paths, the standardized directpath from experimental condition to within-couple closeness dropped from .20 to .08 andwas no longer significant for women (p =.54), and dropped from .20 to .01 and wasno longer significant for men (p = .95). Con-straining the direct path to be equal to zero for

.37*

.22*

.31*

.37*

.22*

.20* (.01)

.20* (.08)

.20* (.01)

.20* (.08)

.29*

.29*

Positive Affect F1

Positive Affect F2

Experimental Condition

.31*

Within-CoupleCloseness

M1


M2


F2


F1

Figure 2. Actor–partner interdependence mediation model of the effects of experimentalcondition on changes in feelings of closeness with one’s romantic partner by changes in positiveaffect.Note. F1 = female in couple 1; F2 = female in couple 2; M1 = male in couple 1; M2 =male in couple 2. Significant paths shown; intercepts, residual variances, and residual covari-ances not shown. Overall model fit was good (comparative fit index [CFI] = .91; root meansquare error of approximation [RMSEA] = .07).*p < .05.

290 R. B. Slatcher

women did not significantly worsen the fit ofthe model, �χ2(1, N = 60) = .36, p = .55;constraining the direct path to be equal tozero for men also did not significantly worsenthe fit of the model, �χ2(1, N = 60) = .09,p = .92. Thus, the extent to which peoplefelt close to their romantic partners was, forwomen, driven by their own level of positiveaffect and, for men, by the levels of positiveaffect of both of their own partner and thewoman in the other couple.

Might the effects of the experimentalmanipulation on positive affect be mediatedby increased feelings of closeness towardone’s partner rather than vice versa? Amodel testing this reverse-mediation modelwas next constructed. The analysis showedthat women’s positive affect was significantlypredicted by their own partner’s feelings ofcloseness (β = .28, p < .05), but the pathfrom the closeness to positive affect fromthe man in the one couple to the woman inthe other couple was no longer significant(β = .18, p > .10), nor was the path from onewoman’s feelings of closeness to her ownpositive affect significant (β = .16, p > .10).Further, a bootstrap test indicated that medi-ation of the effect of condition on women’spositive affect via their partner’s closenesswas only marginally significant (standardizedindirect effect) = .12, p = .08, 95% CI =(−.01, .31). However, this reverse-mediationmodel was not significantly worse-fitting thenthe original mediation model tested above,�χ2(1, N = 60) = 1.14, p = .29. Thus, theresults of this reverse-mediation model anal-ysis were somewhat inconclusive.

Novelty

There were no significant actor or partnereffects for how novel the interaction was.Although those in the closeness conditionreported that their experience was more novelthan those in the small-talk condition, varyinglevels of perceived novelty did not mediatethe effects of the manipulation on feelings ofcloseness toward one’s own romantic partner.

Learning new things about one’s romanticpartner

There were also no significant actor or part-ner effects for learning new things aboutone’s romantic partner. This indicates thatthe effects of the experimental manipulationon feelings of closeness toward one’s partnercannot be explained by the fact that they mayhave simply been learning new things abouttheir partner during the interaction.

Moderators

Did the experimental manipulation work forsome people but not others? Additional anal-yses were conducted to see whether gen-der, ethnicity, or length of time dating mod-erated the effects of the manipulation oncloseness with the other couple and close-ness with one’s romantic partner. In theseanalyses, SEM again was used, with exper-imental condition (dummy coded 0 and 1for small talk and closeness, respectively)entered with the moderator (e.g., length oftime dating) and the product of experimen-tal condition and the moderator. As shownin Figure 3, the effects of the manipulationon feelings of closeness with the other cou-ple were moderated by length of time dating(interaction term β = .25, p = .04); length oftime dating was positively associated withfeelings of closeness with the other couple forthose in the closeness condition but not forthose in the small-talk condition. This find-ing suggests that the longer that two peopleare dating each other, the more comfortablethey may be about being open and disclos-ing around other couples. Neither gender norethnicity—including whether couples werematched on ethnicity—moderated the effectsof the manipulation.

Discussion

In a romantic relationship, one person’sthoughts, feelings, and behaviors are not inde-pendent from the other person’s. So, too, arethe thoughts, feelings, and behaviors of bothpeople not independent from the social worldin which a relationship is embedded. How do


1

2

3

4

5

6

7

Low Length of Time Dating

High Length of TimeDating

Clo

sene

ss w

ith

Oth

er C

oupl

e High-Disclosure Condition

Small-Talk Condition

Figure 3. Moderation of effects of experimental manipulation on feelings of closeness withthe other couple by length of time dating (split at the median for graphical representation).

couples form friendships with others outsideof their relationship, and what are the impli-cations of those friendships?

The goal of this article was to investi-gate how friendships between couples areforged in a controlled laboratory settingand the impact of those developing friend-ships on within-couple processes. The methodused was relatively simple and straightfor-ward, lasting only 45 min. Couples tookpart in discussion activities primarily involv-ing increasing amounts of self-disclosure orin discussion activities involving minimaldisclosure. This design provided a frame-work for investigating whether it is possibleto generate interpersonal closeness betweencouples and for examining the effects ofinteractions between couples on closenesswithin couples. Following up with couples1 month later provided an opportunity toexamine the long-term effects of the manip-ulation on whether couples advanced fur-ther in the friendship-formation process (e.g.,contacted the other couple or met up withthem), and other long-term effects on per-ceptions of the other couple and feelingsof closeness toward one’s own romanticpartner.

Between-couple effects

Compared to couples in the small-talk condi-tion, couples in the high-disclosure conditionfelt much closer to the other couples. Of the

moderators examined in this study—gender,ethnicity, and length of time dating—onlylength of time dating moderated the effects ofthe manipulation, with length of time datingpositively associated with feelings of close-ness with the other couple for those in thecloseness condition but not for those in thesmall-talk condition.

Anecdotally, based on experimenter obser-vations made during and after the studysessions, a large number of couples in thehigh-disclosure condition enjoyed their inter-action with the other couple immensely andseemed to be on a path toward real friendship.On several occasions, when the experimenterrequested that couples in the high-disclosurecondition stop what they were doing and goon to the next set of slips, she was chas-tised for interrupting and given looks of dis-appointment. Similar moments occurred at theend of the 45-min interaction, when cou-ples were asked to separate. Couples in thehigh-disclosure condition often balked at thisrequest, asking for more time to discuss what-ever topic they had at that moment; manydid not want to leave or end their conver-sation with the other couple. In contrast,almost every couple in the small-talk condi-tion seemed ready to depart and emotionallyneutral at the end of the experiment.

At the 1-month follow-up, compared tothose in the small-talk condition, those in thehigh-disclosure condition were significantly

292 R. B. Slatcher

more committed to the success of the othercouple’s relationship and thought the othercouple was more committed to the successof theirs. Further, 10 of the 30 couples inthe high-disclosure condition contacted theother couple they had met—6 actually metup in person; none in the small-talk conditioncontacted the couples they had met.

Are couples who seek out friendships withother couples happier in their romantic rela-tionships to begin with compared to those whodo not? The findings from the follow-up phaseof this study preliminarily suggest otherwise:Those in the high-disclosure condition whocontacted the other couples felt no closer ormore satisfied in their relationships comparedto those who did not. However, this finding isqualified by the fact that only a small numberof couples contacted each other—limiting thestatistical power of analyses—and becausethe couples in this study were by and largehappy to begin with. It is unknown whethercouples who are truly unhappy in their rela-tionships would be willing to put the effortinto couple-based friendships.

Within-couple effects

Couples in the high-disclosure condition alsofelt closer to their own partners immedi-ately following the interaction and marginallycloser to them a month later. This suggests thepossibility of relationship-enhancing effects ofevolving friendships between couples, partic-ularly in the short term.

Positive affect mediated the effects of themanipulation on feelings of closeness to one’sromantic partner for both men and women.The greater the positive affect reported bythe women, the closer they felt to their part-ners. For the men, the greater the positiveaffect reported by both of the women in thegroup—their own partner and the womanin the other couple—the closer they felt totheir partner; neither novelty nor learning newthings about one’s partner mediated the effectsof the manipulation.

The findings relating to positive affect arein line with current thinking on the mecha-nisms underlying the benefits of novel andarousing activities for couples (Strong &

Aron, 2006). As couples naturally becomemore accustomed to each other over timeand relational boredom becomes more likely,going out on a double-date with another cou-ple may be an easy way for couples to recre-ate the initial feelings of excitement in theirrelationship. It has been theorized that self-expansion must be rapid in order for there tobe rises in positive affect (Aron et al., 2001).The self-expansion that couple members wentthrough in this paradigm was rapid enough tolead to rises in positive affect and, in turn,increased feelings of closeness within couples.However, in the real world, it is likely that thisrapid self-expansion would lessen on futuredouble-dates with other couples as the nov-elty of meeting a new couple wears off. Theprocess through which interacting with a newcouple enhances relationship closeness maybe different from the process through whichgetting together with existing couple friendsmight enhance closeness. Thus, while previ-ous research has hinted that interactions withexisting couple friends are beneficial for cou-ples (Agnew et al., 2001; Milardo, 1988),the extent to which previous findings mapon to those reported here requires furtherinvestigation.

Although it was predicted that increases inpositive affect would mediate the effects ofthe manipulation on within-couple closeness,the partner effects of positive affect on close-ness were somewhat unexpected. The linksbetween positive affect and feelings aboutone’s relationship typically have been thoughtof from an intrapersonal perspective ratherthan an interpersonal one (Laurenceau, Troy,& Carver, 2005; Strong & Aron, 2006). How-ever, the actor–partner analyses reported heresuggest that, for men at least, there maybe interpersonal processes at work. In otherwords, while the feeling of positive affectis an intrapersonal experience, the outwardexpression of positive affect influences otherpeople and is thus highly interpersonal. It isimportant that future work further investigatethese processes—incorporating questionnairemeasures as well as observational measuresof affect—to see whether these interpersonaleffects are direct or involve more complexprocesses and attributions. It is conceivable,


for example, that the effect of increased pos-itive affect from the woman in one coupleleading to the man in the other couple feelingcloser to his own partner is due to the maninterpreting that the woman in the other cou-ple enjoyed spending time with him and hispartner, which in turn led the man to feel bet-ter about his relationship. The lack of similareffects from the men to the women is notable.While men and women reported similar levelsof positive affect during the interaction, dif-ferences in men and women’s visible levels ofpositive affect may have been a contributingfactor. In other words, the women may nothave picked up the men’s positive affect haseasily as the men picked up on theirs becauseof gender differences in the ways in whichpositive affect is outwardly displayed.

Limitations and future directions

These findings have some notable limitations.Most importantly, a very reasonable argu-ment can be made that the direction of theeffects between positive affect and within-couple closeness could be reversed becausethey were measured at the same point intime. In other words, it could be that feelingclose toward another couple lead to increasesin positive affect, and not vice versa. Whilereverse-mediation analysis did not lend sup-port to the idea that increases in within-couple closeness mediated the effects of themanipulation on positive affect, future studiesexamining these questions with larger samplesfollowed over time are necessary to rule outissues of directionality. Second, with the num-ber of couples in this study, only medium tolarge effects could be detected with substan-tial power; while no evidence was found forthe hypothesis that more satisfied couples aremore likely to seek out shared friendships, thenull effects may have been in part a functionof low statistical power. Finally, the couplesin this study were young couples in datingrelationships from the United States. Althoughthe couples in this study had been together forat least a year, it may be that the effects gen-erated here may differ somewhat for marriedand older couples and for those from othercultures.

There are a number of potential futuredirections for this research. Perhaps mostobvious is investigating when and under whatspecific conditions relationship closeness isenhanced through interactions with other cou-ples. Self-disclosure is undoubtedly part of theprocess. However, the questions that couplesasked each other in this procedure covered abroad array of topics—some very disclosing,some not. Manipulating these questions wouldallow for the isolation of different effects.For example, whether disclosures about neg-ative experiences versus positive experiencesare differentially closeness inducing could beinvestigated; although much of the researchexamining disclosure in close relationshipshas focused on negative disclosures and thesupport from partners that they elicit (Pasch& Bradbury, 1998), recent findings suggestthat how people respond to positive event dis-closures is important as well (Gable, Reis,Impett, & Asher, 2004). Further, some ofthe questions used in this procedure—albeita small percentage—potentially involved aperson’s romantic relationship (e.g., “Whatwould constitute a “perfect” day for you?”).To what extent are these types of ques-tions responsible for the greater feelings ofcloseness toward romantic partners in thehigh-disclosure condition? Relatedly, to whatextent were increases in within-couple close-ness attributable to self-disclosures to andfrom one’s own partner? Although media-tion analyses showed that the effects of themanipulation were not due to simply learn-ing new things about one’s partner, it is pos-sible that partner disclosure led to greaterfeelings of closeness through other pathways.Adding a comparison condition in which cou-ples go through the closeness-induction pro-cedure without other couples present wouldallow one to test this idea directly. Finally,investigating of the effects of other types ofnovel and arousing activities—such as play-ing games or physical activities—would helpclarify the conditions under which interac-tions between couples enhance relationshipfunctioning.

Although positive affect may partiallyexplain why interactions between couplescan enhance within-couple closeness, other

294 R. B. Slatcher

mechanisms should be considered as well. Forexample, previous research has demonstratedthe tendency to perceive one’s own relation-ship to be superior to others’ relationships(Rusbult, Van Lange, Wildschut, Yovetich, &Verette, 2000; Van Lange & Rusbult, 1995).Thus, when couples interact with other cou-ples they may generally be inclined to thinkthat their relationship is better than the othercouples’ relationships. This downward com-parison, in turn, may make couples feel bet-ter about their own relationship and leadto greater feelings of closeness toward theirpartner.

Implications

Social and clinical psychologists, sociologists,and others interested in the study of rela-tionships would benefit from understandinghow and why couples are affected by peo-ple outside their relationships. Studying howfriendships between couples form can leadto clearer conceptions not only of how out-siders affect the quality of couples’ relation-ships but also of even more basic questionsabout how friendships develop. For example,are my partner and I at our best when we arealone together or at a crowded party? If myromantic partner and I want to get to knowanother couple, what sorts of questions shouldwe ask them? And how do we respond whenthey tell us personal things about themselves?Despite how often we come into contact withother people in the context of our relation-ships, little is known about how these peopleaffect our relationships and how our relation-ships affect them.

From a clinical standpoint, the findingsfrom this study offer preliminary evidence thathaving close, intimate, and shared interactionswith others outside of one’s relationship offerrelationship-enhancing benefits. Interventionsdeveloped for distressed couples potentiallycould integrate aspects of social networksinto their approaches, either through struc-tured interactions such as the one used hereor by providing couples with strategies tostrengthen current friendships or form newones.

These findings have exciting practicalimplications as well. By seeking out

friendships with other couples, couples maybe able to build closeness within their ownrelationship, particularly when interactionswith other couples involve increasing amountsof disclosure. Of course, in the real world,these processes would not be expected tooccur as quickly as they do in the labora-tory. Indeed, over 30 years of self-disclosureresearch indicates that when people disclosepersonal information too quickly in newfriendships they are actually more apt to beless well liked (Collins & Miller, 1994). In thelaboratory—when disclosures are facilitatedby an experimenter—early and intense self-disclosure may not violate social norms in thesame way that it does outside the lab. Thus, inthe real world, couples would probably be bestadvised to engage in personal self-disclosureswith other couples over several occasions andmore extended periods of time. Naturalisticlongitudinal research with pairs of couples isneeded to test the ideal time course of disclo-sures and the long-term efficacy of interac-tions between couples in boosting closenesswithin couples. Such studies would allow oneto test, for example, whether additional inter-actions with another couple would continue toboost within-couple closeness after the initialexcitement of meeting a new couple starts tofade.

Conclusion

The closeness-induction method provides auseful framework for examining how devel-oping friendships between couples affect thewithin-couple process in a controlled labora-tory setting. The results reported here indicatethat closeness between couples can be gener-ated relatively quickly and that this processcan enhance people’s relationships. Further,with recent statistical advances, the effects ofoutside friendships on relationship processesnow can be estimated fairly easily.

It is not known whether “real” friend-ships between couples are formed after goingthrough this type of closeness-inducingexercise. The closeness that is generated isprobably most akin to the closeness pro-duced after a very good first double-date.The fact that many couples progressed onthe path toward friendship—contacting each


other and, in some cases, meeting up for asecond date—is indication that the closenessproduced with this method is authentic.

The results reported here and elsewheresuggest that shared interactions with otherscan be beneficial for couples. Of course,in some instances, interactions with othercouples may have downsides—for example,when there are extradyadic attractions or whenone perceives the other couple as being hap-pier or more stable. We are only beginningto understand how and under what conditionsthese types of interactions are associated withpositive relationship outcomes. However, thepreliminary evidence suggests that investigat-ing interactions between couples and outsiderswill help elucidate the paths to intimacy inclose relationships and, more broadly, thepaths to friendship.

References

Ackerman, K. (1963). Affiliations: Structural determi-nants of differential divorce rates. American Journalof Sociology, 69, 13–20.

Agnew, C. R., Loving, T. J., & Drigotas, S. M. (2001).Substituting the forest for the trees: Social networksand the prediction of romantic relationship state andfate. Journal of Personality and Social Psychology,81, 1042–1057.

Altman, I., & Taylor, D. (1973). Social penetration:The development of interpersonal relationships. NewYork, NY: Holt, Rinehart, & Winston.

Aron, A., & Aron, E. (1986). Love and the expansion ofself: Understanding attraction and satisfaction. NewYork, NY: Hemisphere.

Aron, A., Aron, E., & Norman, C. C. (2001). Self-expansion model of motivation and cognition inclose relationships and beyond. In G. J. O. Fletcher& M. Clark (Eds.), Blackwell handbook of socialpsychology: Interpersonal processes (pp. 478–501).Malden, MA: Blackwell.

Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusionof other in the self scale and the structure of inter-personal closeness. Journal of Personality and SocialPsychology, 63, 596–612.

Aron, A., Melinat, E., Aron, E. N., Vallone, R., &Bator, R. (1997). The experimental generation ofinterpersonal closeness: A procedure and some pre-liminary findings. Personality and Social PsychologyBulletin, 23, 363–377.

Aron, A., Norman, C. C., Aron, E. N., McKenna, C., &Heyman, R. E. (2000). Couples shared participationin novel and arousing activities and experiencedrelationship quality. Journal of Personality and SocialPsychology, 78, 273–284.

Baron, R. M., & Kenny, D. A. (1986). The modera-tor–mediator variable distinction in social psychological

research: Conceptual, strategic and statistical consid-erations. Journal of Personality and Social Psychol-ogy, 51, 1173–1182.

Campbell, L., & Kashy, D. A. (2002). Estimating actor,partner, and interaction effects for dyadic data usingPROC MIXED and HLM: A user-friendly guide.Personal Relationships, 9, 327–342.

Collins, N. L., & Miller, L. C. (1994). Self-disclosureand liking: A meta-analytic review. PsychologicalBulletin, 116, 457–475.

Derlega, V. J., Metts, S., Petronio, S., & Margulis, S. T.(1993). Self-disclosure. Newbury Park, CA: Sage.

Gable, S. L., Reis, H. T., Impett, E. A., & Asher, E. R.(2004). What do you do when things go right? Theintrapersonal and interpersonal benefits of sharingpositive events. Journal of Personality and SocialPsychology, 87, 228–245.

Hendrick, S. S. (1988). A generic measure of relationshipsatisfaction. Journal of Marriage and the Family, 50,93–98.

Kashy, D. A., & Kenny, D. A. (2000). The analysisof data from dyads and groups. In H. T. Reis &C. M. Judd (Eds.), Handbook of research methodsin social psychology (pp. 451–477). New York, NY:Cambridge University Press.

Kenny, D. A. (1996). Models of interdependence indyadic research. Journal of Social and Personal Rela-tionships, 13, 279–294.

Larson, R. W., & Bradney, N. (1988). Precious momentswith family members and friends. In R. M. Milardo(Ed.), Families and social networks (pp. 107–126).Thousand Oaks, CA: Sage.

Larson, R., Mannell, R., & Zuzanek, J. (1986). Dailywell-being of older adults with friends and family.Psychology and Aging, 1, 117–126.

Laurenceau, J.-P., Feldman Barrett, L., & Pietromonaco,P. R. (1998). Intimacy as an interpersonal process:The importance of self-disclosure and perceivedpartner responsiveness in interpersonal exchanges.Journal of Personality and Social Psychology, 74,1238–1251.

Laurenceau, J.-P., Rivera, L. M., Shaffer, A. R., &Pietromonaco, P. R. (2004). Intimacy as an interper-sonal process: Current status and future directions. InD. Mashek & A. Aron (Eds.), Handbook of closenessand intimacy (pp. 61–78). Mahwah, NJ: Erlbaum.

Laurenceau, J.-P., Troy, A. B., & Carver, C. S. (2005).Two distinct emotional experiences in romantic rela-tionships: Effects of perceptions regarding approachof intimacy and avoidance of conflict. Personality andSocial Psychology Bulletin, 31, 1123–1133.

McNair, D. M., Lorr, M., & Droppleman, L. F. (1971).Manual: Profile of Mood States. San Diego, CA:Educational and Industrial Testing Service.

Milardo, R. M. (1982). Friendship networks in develop-ing relationships: Converging and diverging socialenvironments. Social Psychology Quarterly, 45,162–172.

Milardo, R. M. (1988). Families and social networks. Anoverview of theory and methodology. In R. M. Milardo

296 R. B. Slatcher

(Ed.), Families and social networks (pp. 13–47). New-bury Park, CA: Sage.

Olsen, J. A., & Kenny, D. A. (2006). Structural equationmodeling with interchangeable dyads. PsychologicalMethods, 11, 127–141.

Pasch, L. A., & Bradbury, T. N. (1998). Social support,conflict, and the development of marital dysfunction.Journal of Consulting and Clinical Psychology, 66,219–230.

Reis, H. T., & Shaver, P. (1988). Intimacy as an inter-personal process. In S. W. Duck (Ed.), Handbook ofpersonal relationships (pp. 367–389). Oxford, Eng-land: Wiley.

Rusbult, C. E., Van Lange, P. A. M., Wildschut, T.,Yovetich, N. A., & Verette, J. (2000). Perceived supe-riority in close relationship: Why it exists and per-sists. Journal of Personality and Social Psychology,79, 521–545.

Shrout, P. E., & Bolger, N. (2002). Mediation in exper-imental and nonexperimental studies: New proce-dures and recommendations. Psychological Methods,7, 422–445.

Slatcher, R. B., & Pennebaker, J. W. (2006). How doI love thee? Let me count the words: The socialeffects of expressive writing. Psychological Science,17, 660–664.

Sprecher, S., Felmlee, D., Orbuch, T. L., & Willetts,M. C. (2002). Social networks and change in personal

relationships. In A. L. Vangelisti, H. T. Reis, & M. A.Fitzpatrick (Eds.), Stability and change in relation-ships (pp. 257–284). New York, NY: Cambridge Uni-versity Press.

Sternberg, R. J. (1997). Construct validation of a triangu-lar love scale. European Journal of Social Psychology,27, 313–335.

Strong, G., & Aron, A. (2006). The effect of shared par-ticipation in novel and challenging activities on expe-rienced relationship quality: Is it mediated by highpositive affect? In K. D. Vohs & E. J. Finkel (Eds.),Self and relationships: Connecting intrapersonal andinterpersonal processes (pp. 342–359). New York,NY: Guilford.

Van Lange, P. A. M., & Rusbult, C. E. (1995). My rela-tionship is better than—and not as bad as—yoursis: The perception of superiority in close relation-ships. Personality and Social Psychology Bulletin, 21,32–44.

Watson, D. W., & Clark, L. A. (1994). The PANAS-X:Manual for the positive and negative affect schedule-expanded form. Unpublished manuscript, Universityof Iowa.

Watson, D. W., Clark, L. A., & Tellegen, A. (1988).Development and validation of brief measures ofpositive and negative affect: The PANAS scales.Journal of Personality and Social Psychology, 54,1063–1070.

Appendix

Novelty of Interaction Scale

Below are a number of questions that apply to what you thought about the interaction youjust had with the other couple. Please answer these questions as accurately and honestly aspossible. Your answers will be completely confidential.

1. This interaction was a very novel experience for my romantic partner and me.

1 2 3 4 5 6 7 8 9

Not true Definitely true

2. This interaction was quite different than anything I’ve done with my romantic partner before.

1 2 3 4 5 6 7 8 9


3. This interaction was quite similar to other times that my partner and I have met other couplesfor the first time (reverse-scored).

1 2 3 4 5 6 7 8 9



4. This interaction was quite different from how my partner and I usually spend time together.

1 2 3 4 5 6 7 8 9


5. My romantic partner does this kind of thing all the time (reverse-scored).

1 2 3 4 5 6 7 8 9


6. Having this type of interaction with other people outside our relationship was a very newexperience for my romantic partner and me.

1 2 3 4 5 6 7 8 9


7. I’ve never done anything like this with my partner before.

1 2 3 4 5 6 7 8 9


Documents

When Harry and Sally met Dick and Jane: Creating closeness … · 2010-08-06 · When Harry and Sally met Dick and Jane: Creating closeness between couples RICHARD B. SLATCHER Wayne