Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Oct 1.
Published in final edited form as: J Marital Fam Ther. 2018 Feb 11;44(4):561–574. doi: 10.1111/jmft.12311

Remembering the good times: The influence of relationship nostalgia on relationship satisfaction across time

AB Mallory 1, C Spencer 2, J G Kimmes 3, AM Pollitt 4
PMCID: PMC6245572  NIHMSID: NIHMS994230  PMID: 29430666

Abstract

We conducted two studies to understand if reminiscing about early parts of a romantic relationship can increase positive affect and relationship satisfaction. In Study 1, we examined the psychometrics of an adapted relationship nostalgia measure, if relationship nostalgia changes positive affect, and if relationship nostalgia and relationship satisfaction are associated. In Study 2, we tested the longitudinal link between relationship nostalgia and relationship satisfaction. Rather than increasing positive affect, relationship nostalgia is associated with a movement toward emotional homeostasis. Additionally, relationship nostalgia is positively associated with relationship satisfaction at the same time point but in the short term the association is negative—long term, this association disappears. Implications for using a relationship history in therapy are discussed.

Keywords: couples, relationship nostalgia, relationship satisfaction

For marital and couple therapists, taking a history of the relationship and asking couples to reminisce about the early parts of their relationship are standard practices addressed in most books about couple’s therapy. A number of specific methods for conducting a relationship history have been written about extensively (e.g., DeMaria, Weeks, & Hof, 1999; Gottman, 1999, Jacobson & Christensen, 1996), but the primary purpose of this practice is to gather information to develop an effective treatment approach. This process of reflecting about the early parts of the relationship can also be described as relationship nostalgia.

Some models of therapy (e.g. Gottman’s Method and Integrative Couples Behavioral Therapy), though they use different language, describe or use the relationship history as an intervention that can be used to foster positive emotions in a couple as they recollect about how they met and why they chose to stay in the relationship (Gottman, 1999; Jacobson & Christensen, 1996). However, despite strong anecdotal evidence, claims that relationship nostalgia elicited through the relationship history should be linked to better relational outcomes such as relationship satisfaction has, to our knowledge, been relatively unexplored. Previous research on general nostalgia has been found to be related to positive social and affective outcomes (Abeyta, Routledge, & Juhl, 2015; Wildschut, Sedikides, Arndt, & Routledge, 2006). However, only one previous study has specifically examined nostalgia in the context of romantic relationships and the authors found that nostalgia had a positive association with relationship satisfaction (Juhl, Sand, & Routledge, 2012).

A relationship history in couples therapy and research on general nostalgia share a common theme— they both represent how reflection on the past shapes current perspectives and emotions. One common factor across couple and family therapy models is that the presenting issue is conceptualized in the context of the family or couple’s relationship history (Sprenkle & Blow, 2004a; Davis & Piercy, 2007a). This relational conceptualization of the problem informs how the therapist will help clients shift their focus of the problem and direct the client’s attention through the therapist’s preferred therapy model to address current problems (Sprenkle & Blow, 2004a; 2004b; Davis & Piercy 2007a; 2007b). Therapists may help clients to think about the presenting issue through their negative interaction cycle (Emotion Focused Therapy; EFT), a problem saturated narrative (Narrative Therapy), or negative cognitions about the relationship (Integrative Behavioral Couples Therapy; IBCT). This focus on the problem as relational often gives couples and families a sense of uniting against the problem through a common conceptualization of it (Sprenkle & Blow, 2004a; Davis & Piercy, 2007a; 2007b). The relational focus of a problem also allows therapist to direct attention to positive instances (past or present) that challenge or change the relational issue. This may a brief moment where partners are able to express or attentively listen to unmet attachment needs (EFT); when a partners identify exceptions to a problem-saturated narrative (Narrative Therapy); or when they begin to challenge negative cognitions about a partner or the relationship (IBCT). Negative mood states trigger nostalgic memories, yet both the content and affect of nostalgic memories tend to be positively valenced (Routledge, et al., 2006). By nature, relationship histories involve discussing both the fondly and the un-fondly remembered parts of a relationship, but conceptualizing the problem as relational may help the couple feel more hopeful that therapy can help them (Sprenkle & Blow, 2004a; Davis & Piercy 2007a; 2007b) and may foster a sense of bonding (Gottman, 1999; Jacobson & Christensen, 1996).

Given the strong empirical evidence for nostalgia as a positive resource for relationships, and its conceptual similarity to the relationship history in the context of therapy, we use the empirical evidence from nostalgia research and theoretical conceptualizations of the relationship history as the guiding framework for this study. Using this integrated framework, the goal of our study is to examine if reminiscing about the early parts of a romantic relationship (measured through nostalgia) fosters positive emotions and a sense of bonding for people in romantic relationships. Our study will provide some empirical evidence for using a relationship history to bolster positive affect and a sense of bonding for romantic partners.

LITERATURE REVIEW

Relationship Histories

Inquiring about a couples’ relational history is generally considered standard practice for assessment in couples therapy and may take various forms besides a formal relationship history assessment. Relationship histories help therapists to understand the history of the conflict that brings the couple to therapy (Johnson, 2004), learn the couple’s attachment history (Johnson, 2004), assess for violence (Stith, McCollum, & Rosen, 2011), and take note of important relationship milestones (Titelman, 1998). Because it is common practice, conducting a relationship history in couples therapy is not widely cited but might be most credited to Murray Bowen (1966) and his use of the genogram to assess how couples met, fell in love, their important milestones, and how these processes influenced their current relationship.

A common factor of couple and family therapy models is the conceptualization of presenting issues as relationally situated. However, not all models necessitate a formal relationship history, and few explicitly suggest using a relationship history to facilitate a sense of bonding and increase positive affect. In Integrative Behavioral Couples Therapy (IBCT; Jacobson & Christensen, 1996), as a part of the discussion of the therapeutic benefits of the initial interview, the authors state, “As we assess the couple’s developmental history and determine the basis for their attachment, the affect in the therapy room often becomes quite positive: hence the therapeutic benefit of gathering this assessment information” (p. 64). This suggests that assessing a couple’s history can be used to elicit positive affect. However, not every couple has a pleasant memory of the early parts of their relationship—Jacobson and Christensen (1996) go on to say, “…Of course, some couples have extremely depressing stories to tell regarding their initial meeting” (p. 65). However, even talking about the lows of a relationship can foster a sense of shared meaning as couples reflect about the changing dynamic of their relationship and similarly, nostalgic memories occasionally take the form of a redemption narrative (i.e. moving from negative to positive affect; Wildschut, et. al., 2006).

In Gottman’s (1999) model of marital therapy, assessment takes a prominent role prior to intervention with a couple. Gottman (1999) states, “Assessment is, by itself, a very powerful experience for couples and is no different from intervention to them” (p. 114). Here, Gottman refers to how discussing various parts of a couple’s history can bring up different emotions for clients. Although the purpose of the oral history interview in Gottman’s model is not necessarily to elicit positive affect, Gottman (1999) mentions that couples tend to enjoy the oral history interview—as he put it, “They often relish talking about how they met and telling the dramatic story of how they created a marriage and family” (p. 184). During the interview, therapists stay alert for instances of positive affect (e.g. body language, excitement, humor) as insight into the couples’ functioning (Gottman, 1999). It is this focus on the positive affect during the assessment process, which suggests that the relationship history can be used to elicit positive affect and foster a sense of boding. The fond remembrance of the story of how couples created a marriage and family has a similar affect to the content of nostalgia memories (Wildschut, et. al., 2006).

Although Gottman’s Model and IBCT are only two of many models of couples therapy, both models are well respected and some of the few that explicitly discuss the potential of the assessing a couple’s relationship history as a potential bonding experience for the couple. However, the question remains, does this assertion stand up to empirical testing?

Nostalgia

Perspectives on the nature of nostalgia have shifted over time from being seen as a disease or pathology to viewing nostalgia as a positive force (see Sedikides, Wildschut, Arndt, & Routledge, 2008 and Batcho, 2013 for short historical overviews). Although nostalgic memories are triggered by negative mood states (i.e., negative affect, loneliness, pessimism, and existential threats), the content of these memories typically involve the self along with those we share meaningful relationships with, such as family and friends, and are mostly positive experiences (Wildschut, et. al., 2006; Abeyta, Routledge, & Juhl, 2015; Routledge, et. al., 2011). The content of nostalgic memories closely resembles the focus of a relationship history. When therapists ask clients to reflect on their relationship, they may experience a host of positive feelings as the recall earlier parts of the relationship. Nostalgia has consistently been shown to generate positive affect, increase self-esteem, promote a sense of personal meaning, and can foster a sense of social connectedness with others (Abeyta, Routledge, & Juhl, 2015; Routledge et al., 2011; Wildschut, et al., 2006). The exception to the benefits of nostalgia seems to be when a person is high in avoidant related attachment (Wildschut, Sedikides, Routledge, Arndt, & Cordaro, 2010).

Adults who have a high level of avoidant attachment see others as unresponsive to their needs and do not seek out others for support— therefore, avoidant-attached people prefer distance in romantic relationships (Mikulincer & Shaver, 2012). The psychological benefits of nostalgia are socially orienting—that is, they help people feel more connected or that they can make social connections. Thus, people who do not normally seek out relationships as a source of comfort would not experience these benefits because it conflicts with their usual orientation towards people. The research on general nostalgic memories does suggest that having a nostalgic experience about one’s relationship could foster positive emotions and bonding between members of a couple in a way similar to that of a relationship history. Although there are many similarities between the concepts of nostalgia and relationship histories, there are some important differences particularly in terms of how nostalgia is measured.

Batcho (1995; 1998) developed the most commonly used measure of general nostalgia, the Nostalgia Inventory (NI). The NI assesses personal nostalgia by asking participants to rate how much they miss 20 items about things, places, people, and more abstract concepts (see Table 1). One critique of the NI is that the question prompt is asks about missing things from the past (Routledge, et. al., 2006). It is relatively untested if missing is a negative or positive (or both) quality but there is evidence that, in the context of nostalgia, it is more positive than negative. In one study, participants who scored high on the NI preferred happy song lyrics (Batcho, 2007), suggesting that the missing prompt of the NI may evoke a positive rather than negative emotional quality. In order to address the limitation of the NI conceptualization of nostalgia as missing, Routledge and colleagues (2006) assessed positive and negative affect in response to induced (through writing) nostalgia in their experiments and found it to increase positive rather than negative affect. Additionally, Routledge, Arndt, Sedikides, & Wildschut (2008) developed the South Hampton Nostalgia Scale in order assess participants proneness to nostalgia—an individual characteristic. Indeed, some people are simply more prone to nostalgia than others are and can benefit from the effects of nostalgia without a prompt to have nostalgic memories. Thus, it is important to consider nostalgia proneness when measuring nostalgia. Taken together, measurement of general nostalgia has received quite a bit of attention. Yet, these assessments of nostalgia have been focused on the individual and their general nostalgic memories—none have assessed nostalgia about a romantic relationship.

Table 1.

Comparison of Nostalgia Inventory with the measure adapted for our study

Nostalgia Inventory (Batcho, 1995) Adapted Measure for this Study
Rate on a 5-point scale (1 = Not at all; 5 =Very much) how you miss the following from when you were younger. Please rate how much you miss the following aspects of the early parts of your relationship with your current partner on the following scale from 1 = Do not miss at all to 5 = Miss very much/very frequently
1. Family 1. The places that you went together a
2. Heros/Heroines 2. The music that you listened to together b
3. Not having to worry 3. Quality time that you spent together
4. Places a 4. The feelings you had when you were together c
5. Music b 5. Feeling safe and secure when you were together
6. Someone you loved 6. The excitement of the start of the relationship
7. Friends 7. The intimacy you shared with this person
8. Things you did f 8. TV shows or movies you watched together d
9. Toys 9. Time spent at home together e
10. The way people were 10. Times when this person made you laugh
11. Feelings you had c 11. First times kissing this person
12. TV shows, movies d 12. The activities you did together f
13. School 13. Sexual experiences you had with this person
14. Having someone to depend on 14. Times this person comforted you
15. Holidays 15. Learning new things about each other
16. The way society was 16. Times you spent talking with this person
17. Pets ------
18. Not knowing sad or evil things ------
19. Church/religion ------
20. Your house e ------
Open in a new tab

Notes: The items with the same letters were adapted for the measure in our study.

Nostalgia and Romantic Relationships

Little research has investigated the association between nostalgia and relationship satisfaction. One of the first studies to connect these two constructs examined if nostalgia orients people towards their romantic relationship (Juhl, Sand, & Routledge, 2012). When participants were prompted to write about a nostalgic experience, those who were low in avoidant attachment had higher relationship satisfaction than those who were high in avoidant attachment. The authors concluded that nostalgia orients people towards romantic relationships when they have low avoidant attachment, but away from relationships when they have high avoidant attachment. One limitation of this study was the use of a measure of general nostalgia. In our study, we will examine is if relationship nostalgia is associated with relationship satisfaction regardless of avoidant attachment.

STUDY 1 CURRENT STUDY

The purpose of Study 1 is to establish a domain-specific construct—nostalgia about a romantic relationship—that can be used to assess whether there is empirical support for having couples discuss their relationship histories to elicit positive affect or bonding between the partners. The main research question we address in Study 1 is, what are the associations between relationship nostalgia, positive affect and relationship satisfaction? To answer this question, we take four steps to elucidate the associations between relationship nostalgia and positive affect as well as relationship nostalgia and satisfaction.

First, we examine the internal consistency and factor structure of a measure of nostalgia that we adapted to assess relationship nostalgia. Second, we use the relationship nostalgia measure to asses if positive and negative affect change from before to after completing the relationship nostalgia measure to understand the affect elicited when participants think about the early parts of their relationship. Based on previous research we would expect that positive affect would increase after completing the relationship nostalgia measure. Third, we examine the association between relationship nostalgia and relationship satisfaction when accounting for meaningful covariates. Because it has yet to be tested if people who are less relationally satisfied (a negative mood state) experience higher levels of nostalgia, we test both relationship nostalgia and relationship satisfaction as dependent variables given that they may influence one another. Finally we test avoidant attachment as a moderator of the association between relationship nostalgia and relationship satisfaction. If romantic relationship nostalgia really is a generally positive experience, even in distressed couples presenting to therapy (Gottman, 1999; Jacobson & Christensen, 1996), then avoidant attachment should not moderate the association between relationship nostalgia and relationship satisfaction.

STUDY 1 METHOD

Sample and Procedure

Data were collected as part of an online survey promoted to students enrolled in four human development and three sociology undergraduate courses at a large, public Midwestern university. The researchers were not instructors for any of these courses. Students were given extra credit for participation in the study or for completing an alternative extra credit opportunity. Data were collected in the spring 2015 semester (N = 605). The data were limited to students who described their relationship as dating exclusively which brought the operational sample to n = 206. The sample was mostly White (88.6 %), followed by Latino (3.3%), Black (3.3%), Asian (2.8%), Native American (1.4%), and different racial identity (0.5%). Participants were mostly women (69%) with a mean age of 19.59 years (SD = 1.47) and mean relationship length of 18.64 months (SD = 17.21).

Measures

Relationship nostalgia.

Participants were asked to rate the degree to which they missed various aspects of the early parts of their relationship with their current partner (e.g., “Times when this person made you laugh”, “Times you spent talking with this person”) with 16 items. This scale was adapted from the nostalgia inventory (Batcho, 1995), which measures nostalgia by what is missed from earlier in a person’s life (e.g., school, the way society was, music). Items were adapted whenever possible (e.g., “Places” was changed to “places that you went together”) and items were added for experiences that typically happen early in a relationship (e.g. “learning new things about each other”; See Table 1). Responses ranged from 1 = Do not miss at all to 5 = Miss very much/very frequently. The responses were coded such that higher mean scores represent more nostalgia for the early aspects of their relationship.

Psychometric tests support initial use of this scale for measuring nostalgia about a romantic relationship and that this measure of relationship nostalgia is a unidimensional construct. The scale had an acceptable internal consistency (α = .98) and inter-item correlation (r = .70). We used the Kaiser-Meyer-Olkin measure of sampling adequacy (Kaiser, 1974) to examine if the items in our measure might be explained by an underlying factor. The Kaiser-Meyer, Olkin measure of sampling adequacy (Kaiser, 1974; MOj = .96) was above the .5 cut off which suggests that our items are explained by an underlying factor. We used Bartlett’s Test of Sphericity (1973) to examine if our items were related to one another and appropriate for a factor analysis. Bartlett’s Test of Sphericity was significant (X2(120) = 3932.38, p < .001), suggesting that the minimum conditions were met to conduct a factor analysis. Taken together both The Kaiser-Meyer, Olkin measure of sampling adequacy (Kaiser, 1974) and Bartlett’s Test of Sphericity (1973) suggested it was appropriate to conduct a factor analysis. We used maximum likelihood estimation for extraction in order to be consistent with the estimation method we used for the confirmatory factor analysis. Only one factor was extracted, which explained 72.60% of the variance.

Relationship satisfaction.

We measured relationship satisfaction with a five item version of the Quality of Marriage Index (QMI; Norton, 1983) The QMI was adapted so that all items assess a romantic relationship rather than marriage (e.g. “we have a good relationship” rather than “we have a good marriage”). Participants were asked to indicate the degree to which they agree or disagree with each item using a 7-point scale (1 = strongly disagree to 7 = strongly agree). Higher scores reflected more relationship satisfaction (α = .98).

Positive and negative affect.

We measured positive and negative affect with the Positive and Negative Affect Schedule (PANAS; Watson and Clark, 1994), a 20 item measure of state mood that asks participants to what extent they feel ten positive (e.g., cheerful) and ten negative (e.g., disgusted) feelings and emotions using a five point scale 1 = not at all to 5 = extremely. The PANAS was given before and after the relationship nostalgia measure to capture changes in participants’ affect after completing the relationship nostalgia measure. Positive affect and negative affect scores from before and after the relationship nostalgia measure were calculated separately for paired sample t-tests The internal consistency was α = .94 for the positive affect items and α = .95 for the negative affect items.

Covariates of Romantic Relationship Nostalgia and Relationship Satisfaction

Nostalgia frequency.

We used the 5-Item South Hampton Nostalgia Scale (Routledge, Arndt, Sedikides, & Wildschut, 2008). The scale asks about the frequency of nostalgic experiences, proneness to nostalgic experiences, the importance of nostalgia experiences, and how often participants bring to mind nostalgic experiences on a seven point scale. Higher scores reflect more frequent nostalgic experiences. The alpha coefficient was α = .89.

Adult attachment.

We used a nine-item version of The Experiences in Close Relationships-Revised (Fraley, Waller, & Brennan, 2000) to asses avoidant attachment (e.g., “I prefer not to show this person how I feel deep down”; α = .89) and anxious attachment (e.g., “I often worry that this person doesn’t really care for me”). Participants indicated agreement with each statement on a seven-point scale (1 = strongly disagree to 7 = strongly agree; α = .92).

Benign relationship attributions.

We used the Relationship Attribution Measure (RAM; Fincham & Bradbury, 1992) to assess attributions about the romantic partner’s behaviors. Participants were presented with four hypothetical negative partner behaviors (e.g., “Imagine that your partner criticizes something you say”) which is then followed by six items about reactions to that negative behavior (e.g. “The reason my partner criticized me is not likely to change”). Participants were asked to indicate the degree to which they agree or disagree with each item using a 6-point scale (1 = strongly disagree to 6 = strongly agree). The items were coded such that a higher score reflected more benign attributions about the partner (α = .86).

Neuroticism.

We used the Eysenck Personality Questionnaire-Brief Version (Sato, 2005) to assess neuroticism with twelve items (e.g. “Are you an irritable person”). Participants indicated the degree to which each statement matched their experience on a scale from 1 = not at all to 5 = extremely. Higher scores reflect more neuroticism (α = .94).

Self-Esteem.

We measured self-esteem with the Rosenberg Self–Esteem Scale (Rosenberg, 1979) which consists of 10 items (e.g., I am able to do things as well as most other people) ranging from 1 = strongly disagree to 4 = strongly agree with higher scores reflecting higher levels of self-esteem (α = .92).

Relationship length.

We control for relationship length as measured in months.

Analysis Plan

Analyses for Study 1 were run in SPSS (IBM SPSS 21) with the exception of the confirmatory factor analysis (CFA) which was run in Mplus 7.4 (Muthén & Muthén, 1998–2015). First, to determine reliability and validity of the relationship nostalgia measure, we ran a CFA to assess the factor structure we found with the EFA prior to running other analyses. Next we ran a paired sample t-test between the positive and negative affect scores before and after the relationship nostalgia measure. We interpreted both the t-value for significant mean differences in affect, and calculated Cohen’s d to assess the magnitude of the effect of relationship nostalgia on affect. Next, two ordinary least squares regressions were calculated—one with relationship satisfaction as the dependent variable and the other with relationship nostalgia as the dependent variable. Variables were entered in four blocks. For the relationship satisfaction model (Model 1a and 1b), we first entered control variables (relationship length, neuroticism, and self-esteem). Next, we entered controls specific to relationship nostalgia (i.e. nostalgia frequency), followed by relationship nostalgia, anxious attachment, and avoidant attachment, with an interaction between relationship nostalgia and avoidant attachment entered last. We tested the interaction in order to assess if relationship nostalgia is associated with relationship satisfaction regardless of a person’s level of avoidant attachment. For the relationship nostalgia model (Model 2), we first entered control variables (i.e. relationship length, neuroticism, and self-esteem). We then entered nostalgia frequency followed by anxious and avoidant attachment. Lastly, we entered relationship satisfaction. Only, the final models are discussed and presented in Table 2.

Table 2.

Study 1 Regressions for Relationship Satisfaction and Relationship Nostalgia (n = 206)

Model 1a (relationship satisfaction) Model 1b (relationship satisfaction) Model 2 (relationship nostalgia)
Variable B SE(B) β B SE(B) β B SE(B) β
Neuroticism −0.16 0.11 −0.11 −0.17 0.11 −0.12 0.19 0.10 0.15
Rel Attribution 0.31* 0.13 0.16* 0.28* 0.13 0.14* −0.35*** 0.12 −0.20***
Self Esteem −0.02 0.17 −0.01 −0.01 0.17 0.00 0.18 0.16 0.09
Rel Length 0.01* 0.00 0.12* 0.01 0.00 0.12 0.00 0.00 0.05
Nostalgia Freq −0.01 0.05 −0.02 −0.01 0.05 −0.01 0.14*** 0.05 0.21***
Anx Attach −0.03 0.09 −0.03 −0.04 0.09 −0.04 0.00 0.09 0.00
Avoid Attach −0.76*** 0.13 −0.43* −0.78*** 0.13 −0.44*** 0.40*** 0.13 0.26***
Rel Satisfaction -- -- -- -- -- -- 0.16* 0.07 0.17*
Nostalgia 0.16* 0.07 0.15* 0.17* 0.07 0.15* -- -- --
Nostalgia X Avoid attach -- -- -- 0.08 0.08 0.06 -- -- --
R2 0.291 0.294 0.161
F for change in R2 n/a 0.003 n/a
Open in a new tab
*

Note: p < .05.

**

p < .01.

***

p < .001.

STUDY 1 RESULTS

We first ran a CFA to assess the factor structure of our relationship nostalgia measure. The CFA for the relationship nostalgia scale as one factor had acceptable model fit (X2 (104) = 260.146, p <.001; RMSEA = 0.09, 90% CI [.072, .098]; CFI = .93 TLI = .92; SRMR = .03; scaling correct factor for MLR= 1.64). Next, in order to assess if relationship nostalgia shifts affect, we ran a paired sample t-test between the positive and negative affect scores before and after the relationship nostalgia measure. The paired sample t-test analyses supported that there was a change in affect before and after completing the relationship nostalgia measure. There was a significant decrease in the scores for positive affect before (M = 3.20, SD = 0.82) and after (M = 3.01, SD = 0.88) the relationship nostalgia measure (t (205) = 5.42, p < .001). This was a small effect (d = .40). The same was true for negative affect scores before (M = 1.96, SD = 0.77) and after (M = 1.88, SD = 0.80) the relationship nostalgia measure (t (205) = 2.67, p <.01). This effect was negligible (d = .19). Upon closer inspection of item-level paired sample t-tests (available upon request), nine of the ten positive affect items were significantly lower while only three of ten negative affect items significantly differed after the relationship nostalgia measure. This suggests a stronger attenuation of positive affect than negative affect.

Following the t-test analyses, we next ran two ordinary least squares linear regression models, one with relationship satisfaction as the dependent variable (Models 1a and 1b) and one with relationship nostalgia as the dependent variable (Model 2). For regression Model 1a, which examined relationship satisfaction as the dependent variable (see Table 2), we found that more benign attributions (β = 0.16, p <.05), a longer relationship (β = 0.12, p <.05), and more relationship nostalgia (β = 0.15, p <.05) were positively associated with higher relationship satisfaction. Avoidant attachment (β = −0.43, p < .001) was negatively associated with relationship satisfaction. For Model 1b, the interaction between relationship nostalgia and avoidant attachment was not significant (β = 0 .08, p = .34). This suggests that nostalgia specific to a romantic relationships and its association with relationship satisfaction is not influenced by a person’s level of avoidant attachment. Model 1a accounted for 29% of the variance in relationship satisfaction.

For regression Model 2, which examined relationship nostalgia as the dependent variable (Table 2), we found that having higher levels of avoidant attachment (β = 0.26, p <.001), frequency of nostalgic experiences (β = 0.21, p <.001), and relationship satisfaction (β = 0.17, p <.05) were associated with higher levels of relationship nostalgia. Consistent with research on general nostalgia (Abeyta, Routledge, & Juhl, 2015), more benign attributions towards a partner (β = −0.20, p < .001) were linked to lower levels of relationship nostalgia. This model accounted for 16% of the variance in relationship nostalgia.

STUDY 1 DISCUSSION

In Study 1, we sought to answer if there was an association between relationship nostalgia, positive affect, and relationship satisfaction by adapting the nostalgia inventory (Batcho, 1995) such that the questions assessed missing the early parts of a romantic relationship. Study 1 established that our adapted measure of relationship nostalgia adequately assesses reminiscing about earlier parts of a romantic relationship which appears to have the benefit of decreasing some negative affect in the short term, but also decreases positive affect in the short term. This is contrary to previous research on general nostalgia (Wildschut, et. al., 2006) and the notion that thinking of the earlier parts of a relationship increases positive affect (Gottman, 1999, Jacobson & Christensen, 1996). Nostalgia about a romantic relationship appears to be distinct from general nostalgia in that it does not promote an increase of positive affect but seems to induce a movement to emotional homeostasis. However, consistent with the notion that a relationship history can foster a sense of bonding (Gottman, 1999, Jacobson & Christensen, 1996), relationship nostalgia is positively associated with relationship satisfaction. This association was not qualified by participant’s level of avoidant attachment suggesting that relationship nostalgia has positive benefits for people regardless of their level of avoidant attachment. In addition, we found that relationship nostalgia and relationship satisfaction are positively associated.

These results paint a complex picture about the nature of relationship nostalgia and its link with relationship satisfaction. On one hand, relationship satisfaction and relationship nostalgia have a positive association, yet at the same time, relationship attributions, and avoidant attachment have an opposite relationship with relationship nostalgia than they do with relationship satisfaction in our study. However, this may be best explained by previous research which suggests that negative mood states trigger nostalgia (Routeldge et. al., 2006). Less benign (more pessimistic) relationship attributes may increase a person’s relationship nostalgia because these attributions provide more opportunities to have negative mood states. Because avoidant-attached people prefer distance in romantic relationships (Mikulincer & Shaver, 2012), relationship nostalgia provides a safe cognitive space to feel psychologically close to a partner through remembering good times in the past. Relationship histories may have the benefit of providing a medium for avoidant attached people to talk about their relationship safely, promote a movement to emotional homeostasis, and promote a sense of bonding with their partner.

Taken together, these findings suggest that a relationship history may be well suited to couples presenting for therapy where one or both partners have high levels of avoidant attachment or pessimistic relationship attributions, especially since relationship nostalgia appears to move people towards emotional homeostasis and is positively associated with relationship satisfaction. However, at least one issue remains which is to better understand the ordering of the associations between relationship nostalgia and relationship satisfaction, which we focus on in Study 2. The purpose of Study 2 is to test the order of associations between relationship nostalgia and relationship satisfaction with longitudinal data.

STUDY 2 CURRENT STUDY

Study 2 builds upon the first study by testing the temporal ordering of relationship satisfaction and relationship nostalgia. In Study 1, we found that, at the same time point, relationship nostalgia and relationship satisfaction are positively associated. However, we know from previous research that relationship satisfaction changes over time (Karney & Bradbury, 1995) but a lack of longitudinal research means it is unclear how or if nostalgia changes over time. It is also not clear whether thinking about the early parts of a relationship would have a positive or negative association with relationship satisfaction as the relationship length increases. It is also possible that the reverse is true: A person’s level of relationship satisfaction may influence how nostalgic they are about the early parts of their relationship. We will test one main research question with Study 2: how are relationship nostalgia and relationship satisfaction related over time? We use longitudinal data and utilize an autoregressive cross-lagged model to test our research question.

STUDY 2 METHODS

Data were collected from two large human development courses during the fall 2015 semester at three time points with six weeks in between each wave (Wave 1 (N = 462), Wave 2 (N = 422) and Wave 3 (N = 414)). The data were screened to ensure participants were in the same relationship across all three waves. First, participants who reported being single were dropped from the sample (N = 147). Next, participants who were not in a committed relationship at all three time points or responded “yes” to the question “During the past six weeks have you broken up with a romantic partner?” were dropped from the sample (n = 130). Finally, participants who were not in an exclusive dating relationship were dropped (n =117). This resulted in a final sample of 103 participants. The final sample was mostly White (86.4%), followed by Latino (6.8%), Asian (2.9%), Native American (1.9%), and other or mixed race (1.9%) participants. The majority of the participants were women (82.5%). The mean relationship length in months at Wave 1 was M = 23.69 (SD = 19.56). The mean age at base line = 19.20 years (SD = 1.18).

Measures

There were minor measurement differences between Study 1 and Study 2. We used the full Experiences in Close Relationships—Revised (Fraley, Waller, & Brennan, 2000) to measure avoidant attachment (18-items) in Study 2—a decision made as part of a larger study. However, since the internal consistency was already high with the shorter version, this does not substantially alter Study 2. The relationship nostalgia and the QMI had alphas above .97 at each wave. Wave 1 neuroticism, benign relationship attributions, avoidant attachment, and nostalgia frequency had alphas above .87.

Analysis Plan

We ran an autoregressive cross-lagged model to test whether relationship nostalgia and relationship satisfaction predict themselves and each other over time. We ran the autoregressive cross-lagged path analysis in Mplus 7.4 (Muthén & Muthén, 1998–2015) with full maximum likelihood estimation to handle missing data and to estimate robust standard errors due to the negatively skewed distribution of relationship satisfaction.

STUDY 2 RESULTS

The autoregressive cross-lagged path analysis had acceptable fit (X2 (31) = 48.624, p = .02; RMSEA = .07, 90%CI [.03, .11]; CFI = .92; TLI = .89; SRMR = .07; scaling correct factor for MLR= .92). We found that Wave 1 relationship satisfaction was negatively associated with Wave 2 relationship nostalgia (β = −0.13, p <.05). However, Wave 2 relationship satisfaction was not related to Wave 3 relationship nostalgia (β = −0.10, p = .17). Additionally, higher levels of relationship nostalgia at Wave 1 were related to lower relationship satisfaction at Wave 2 (β = −0.20, p < .01), but Wave 2 relationship nostalgia was not related to Wave 3 relationship satisfaction (β = 0.05, p = .67). The model accounted for 12.2% of variance in relationship nostalgia at Wave 1, 59% at Wave 2, and 57.6% at Wave 3. This model accounted for 14% of the variance of relationship satisfaction at Wave 1, 8.4% at Wave 2, and 15.4% at Wave 3 though the R2 for relationship satisfaction Wave 2 and Wave 3 was not statistically significant (p =0.08; p = 0.16).

STUDY 2 DISCUSSION

The purpose of Study 2 was to examine the association between relationship nostalgia and relationship satisfaction over time. Our results suggest that relationship nostalgia and satisfaction are negatively associated with each other over time. However, this negative association only remained significant from Wave 1 to Wave 2, suggesting a short-term association between relationship nostalgia and relationship satisfaction. This may also be due to the brief “dosage” of nostalgia in the study, or that link between relationship nostalgia and relationship satisfaction is less important over time.

GENERAL DISCUSSION

In Study 1, we found that relationship nostalgia was linked with slight declines in both positive affect and negative affect with a small but noticeable effect size for positive affect. The small effect is important, given the short time period between measurements of the participants’ affect, because it suggests that relationship nostalgia could have a substantial effect on couples’ affect when experienced for an extended period of time, for example, during a relationship history. Gottman (1999) and Jacobson & Christensen (1996) have written about how taking a relationship history can elicit positive emotions and can be a beneficial experience in therapy. However, our findings suggest that it may have the effect of moving people towards emotional homeostasis, which may be beneficial for models of therapy that target cognitions or behaviors as mechanisms of change. Our findings from Study 1 do support that reflecting about the early parts of a relationship can foster a sense of boding via relationship satisfaction.

Given that we found that nostalgia about a romantic relationship and relationship satisfaction are positively correlated at one time point, reminiscing about the early parts of a relationship may foster a person’s sense of satisfaction about their romantic relationship. Further, the level of nostalgia that a couple expresses when describing how they first met, or major relationship milestones, may provide information about the couples’ level of current relationship satisfaction. If a couple does not bring up many nostalgic memories that they shared during the relationship history, it could indicate that the couple is currently experiencing low levels of relationship satisfaction. This would be consistent with Gottman’s (1999) finding that in successful relationships, partners have a depth of knowledge about their partner’s psychological worlds and maintain “cognitive space” for their relationship. If the couple does bring up many fond memories, this could be an indication that the couple is still experiencing a higher level of relationship satisfaction, and a therapist could use this information to highlight strengths in the couples’ relationship. Although taking a relationship history can focus on the progression of the conflict that drove the couple to seek couples therapy, it may be equally beneficial to draw attention to the positive memories that the couple shared together and integrate both into the relational conceptualization of how the couple can reach their therapy goals.

Findings from Study 2 help elucidate the association between relationship satisfaction and relationship nostalgia over time. Both relationship satisfaction and nostalgia at wave 1 were negatively related to the other only at Wave 2. These findings suggest that highly satisfied couples are less nostalgic over time and more nostalgic couples may be less satisfied over a short period of time. This is consistent with research on general nostalgia, as nostalgia is often triggered by negative mood states (Wildschut, et. al., 2006; Routledge, et. al., 2011). These findings also suggest that early on, relationship nostalgia can provide some information about a couples’ satisfaction with the relationship, but that this effect disappears over time. Additionally, a high amount of nostalgia for the early parts of the relationship may indicate relationship problems that will lead to a decline in relationship satisfaction over time.

Taken together, the findings from Study 1 and Study 2 seem to support using the relationship history as a means to promote a remembrance of the good times in a relationship to foster a sense of bonding (Gottman, 1999, Jacobson & Christensen, 1996). However, the effect of using the relationship history in this way may have a short-term positive effect, which may be desired for an assessment conducted early in the therapy process but should not be a sole intervention to foster bonding over time.

Future Directions and Limitations

There are several limitations to our study that are important to consider. One limitation is that the adapted Batcho’s Nostalgia Inventory (1995) was used as a proxy to measure nostalgia about the early parts of participant’s current relationship. Nostalgia researchers (Wildschut, et. al., 2006) have noted that the focus on missing may be semantically problematic. However, our adapted scale might not suffer from this limitation as both positive and negative affect declined. Regardless, the measure will need further validation and development. Our ability to understand the emotional affect elicited by the relationship nostalgia measure was limited by the short period of time between measurement periods. However, our goal was not to necessarily evoke the deep nostalgic experiences that a person would have in the therapy room but rather to briefly capture missing the early parts of their current relationship. Yet, even in this brief context of this study, we did find significant small effect sizes. This is promising for future studies. A limitation of both studies were the sample size and the homogeneity of the sample—participants were mostly white young adult women in college who were in relatively short and highly satisfied dating relationships. Although we controlled for relationship length, there may be important differences in longer relationships especially for those that span decades and have weathered numerous challenges. It is also important to note that our exploration of nostalgia and relationship satisfaction measured these variables across a relatively short period of time (i.e., 3.5 months). The previous research on nostalgia used experimental designs and the findings from this research suggest that there is a connection between general nostalgia and relationship satisfaction. Future research could examine a relationship history with couples in long-term committed or married relationships and use an experimental design to test the individual and dyadic effects of eliciting nostalgic memories about their relationship on numerous relationship outcomes.

Clinical Implications

We tentatively offer a few clinical implications based on the findings of our studies. First, our findings seem to suggest that intentionally using a relationship history to evoke nostalgia about the relationship can at a minimum move affect towards neutrality. A sense of bonding can also be fostered through nostalgia about the romantic relationship, which may help with motivation about potential effectiveness of therapy early in the process. However, this approach may have a short-term effect making it appropriate for an early assessment but not a repeat intervention to change affect. Even when couples do report fond memories about the early parts of their relationship, this may actually indicate dissatisfaction with the current state of the relationship over time. Conversely, it seems that highly dissatisfied couples may be reflecting on early parts of the relationship that they miss—conducting a relationship history early in therapy seems quite appropriate for couples who may be in this state of mind and through a relational conceptualization help foster a sense of bonding or hope that therapy can be helpful. Overall, asking about relationship histories in first session can simply serve the purpose of assessment to better understand a couples’ relationship but also to potentially shift the affect and sense of bonding for a couple early in the therapeutic process.

CONCLUSION

We conducted two studies to elucidate the association between relationship nostalgia and relationship satisfaction. We used this approach to provide an initial test of if relationship nostalgia potentially elicited by relationship histories can be used to increase positive affect and a sense of bonding for a couple. Our findings suggest that nostalgia about a romantic relationship is distinct from general nostalgia in a number of ways and that rather than increase positive affect seems to trigger a movement to emotional homeostasis. However, in the short-term, relationship nostalgia and relationship satisfaction are positively associated suggesting a sense of satisfaction with the relationship can occur from this sort of reminiscing. In the long-term, a high amount of nostalgia for the early parts of the relationship may indicate dissatisfaction with some aspects of the current relationship

Figure 1.

Figure 1

Open in a new tab

Cross-lagged auto regressive path model with relationship nostalgia and relationship satisfaction. Model Fit: X2 (31) = 48.624, p = .02; RMSEA = .07, 90%CI [.03, .11]; CFI = .92 TLI = .89; SRMR = .07; scaling correct factor for MLR= .92. Significant paths are bolded and dashed paths are not significant. Covariates include Wave 1 relationship nostalgia are relationship attributions, nostalgia frequency, positive affect, negative affect, and relationship length. Covariates for relationship satisfaction include Wave 1 relationship attributions, avoidant attachment, neuroticism and relationship length. Standardized coefficients are reported. Note: W = wave; *p < .05, **p < .01, ***p < .001.

Acknowledgments

This research was supported by grant, R24HD042849, Population Research Center, and the grant, T32HD007081, Training Program in Population Studies awarded to the Population Research Center at The University of Texas at Austin by the Eunice Kennedy Shriver National Institute of Child Health and Human Development. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

REFERENCES

  1. Abeyta A, Routledge C, & Juhl J (2015). Looking back to move forward: Nostalgia as a psychological resource for promoting relationship goals and overcoming relationship challenges. Journal of Personality and Social Psychology, 109(6). [DOI] [PubMed] [Google Scholar]
  2. Bartlett MS (1937). “Properties of sufficiency and statistical tests”. Proceedings of the Royal Statistical Society, Series A 160, 268–282 [Google Scholar]
  3. Batcho KI (1995). Nostalgia: A psychological perspective. Perceptual and Motor Skills, 80, 131–143. [DOI] [PubMed] [Google Scholar]
  4. Batcho KI (1998). Personal nostalgia, world view, memory, and emotionality. Perceptual and Motor Skills, 87, 411–432. [DOI] [PubMed] [Google Scholar]
  5. Batcho K (2013). Nostalgia: The bittersweet history of a psychological concept. History of Psychology, 16(3), 165–176. [DOI] [PubMed] [Google Scholar]
  6. Batcho KI (2007). Nostalgia and the emotional tone and content of song lyrics. The American Journal of Psychology, 120(3), 3. [PubMed] [Google Scholar]
  7. Bowen M (1966). The use of family theory in clinical practice. Comprehensive psychiatry, 7(5), 345–374. [DOI] [PubMed] [Google Scholar]
  8. Davis SD, & Piercy FP (2007). What clients of couple therapy model developers and their former students say about change, part I: Model-Dependent common factors across three models. Journal of Marital and Family Therapy, 33(3), 318–343. [DOI] [PubMed] [Google Scholar]
  9. Davis SD, & Piercy FP (2007). What clients of couple therapy model developers and their former students say about change, part II: Model-Independent common factors and an integrative framework. Journal of Marital and Family Therapy, 33(3), 344–363. [DOI] [PubMed] [Google Scholar]
  10. DeMaria R, Weeks G, & Hof L (1999). Intergenerational assessment of individuals, couples, and families: Focused genograms. Philadelphia: Brunner/Mazel. [Google Scholar]
  11. Fraley RC, Waller NG, & Brennan KA (2000). An item response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365. [DOI] [PubMed] [Google Scholar]
  12. Gottman JM (1999). The marriage clinic: A scientifically-based marital therapy. WW Norton & Company. [Google Scholar]
  13. Jacobson NS, & Christensen A (1996). Integrative couple therapy: Promoting acceptance and change. WW Norton & Co. [Google Scholar]
  14. Johnson SM (2004). The practice of emotionally focused couple therapy: Creating connection (2nd ed). New York: Brunner-Routledge. [Google Scholar]
  15. Juhl J, Routledge C, Arndt J, Sedikides C, & Wildschut T (2010). Fighting the future with the past: Nostalgia buffers existential threat. Journal of Research in Personality, 44(3), 309–314. [Google Scholar]
  16. Juhl J, Sand EC, & Routledge C (2012). The effects of nostalgia and avoidant attachment on relationship satisfaction and romantic motives. Journal of Social and Personal Relationships, 29(5), 661–670. [Google Scholar]
  17. Karney BR, & Bradbury TN (1995). The longitudinal course of marital quality and stability: A review of theory, methods, and research. Psychological bulletin, 118(1), 3. [DOI] [PubMed] [Google Scholar]
  18. Kaiser H 1974. An index of factor simplicity. Psychometrika 39: 31–36. [Google Scholar]
  19. Kenny DA, Kashy DA, & Cook WL (2006). The analysis of dyadic data. New York: Guilford. [Google Scholar]
  20. Mikulincer M, & Shaver PR (2012). Adult attachment orientations and relationship processes. Journal of Family Theory & Review, 4(4), 259–274. [Google Scholar]
  21. Rosenberg M (1979). Conceiving the self. New York: Basic Books. [Google Scholar]
  22. Routledge C, Arndt J, Sedikides C, & Wildschut T (2008). A blast from the past: The terror management function of nostalgia. Journal of Experimental Social Psychology, 44(1), 132–140. [Google Scholar]
  23. Routledge C, Wildschut T, Sedikides C, & Juhl J (2013). Nostalgia as a resource for psychological health and well-being. Social and Personality Psychology Compass, 7(11), 808–818. [Google Scholar]
  24. Sato T (2005). The Eysenck personality questionnaire brief version: Factor structure and reliability. The Journal of Psychology, 139(6), 545–552. [DOI] [PubMed] [Google Scholar]
  25. Sedikides C, Wildschut T, Arndt J, & Routledge C (2008). Nostalgia: Past, present, and future. Current Directions in Psychological Science, 17, 304–307. [Google Scholar]
  26. Sprenkle DH, & Blow AJ (2004a). Common factors and our sacred models. Journal of Marital and Family Therapy, 30, 113–130. [DOI] [PubMed] [Google Scholar]
  27. Sprenkle DH, & Blow AJ (2004b). Common factors are not islands—they work through models: A response to Sexton, Ridley, and Kleiner. Journal of Marital and Family Therapy, 30, 151–158. [DOI] [PubMed] [Google Scholar]
  28. Stith SM, McCollum EE, & Rosen KH (2011). Couples therapy for domestic violence: Finding safe solutions. Washington; DC: American Psychological Association. [Google Scholar]
  29. Titelman P, (2014). Family Systems Assessment Based on Bowen Theory In Titelman P (Ed.) Clinical applications of Bowen family systems theory. Hoboken: Taylor and Francis. [Google Scholar]
  30. Watson D Clark LA, & Tellegen A (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 55, 1063–1070. [DOI] [PubMed] [Google Scholar]
  31. Wildschut T, Sedikides C, Arndt J, & Routledge C (2006). Nostalgia: Content, triggers, functions. Journal of Personality and Social Psychology, 91, 975–993. [DOI] [PubMed] [Google Scholar]
  32. Wildschut T, Sedikides C, Routledge C, Arndt J, & Cordaro F (2010). Nostalgia as a repository of social connectedness: The role of attachment-related avoidance. Journal of Personality and Social Psychology, 98, 573–586. [DOI] [PubMed] [Google Scholar]

RESOURCES