Maternal attachment insecurity and poorer proficiency savoring memories with their children

Abstract

Relational savoring is the process of focusing one’s attention on the pleasurable parts of a relationship experience, and recent work has explored its relation to attachment styles. This study examined the relationship between attachment insecurity and the quality of responses to a relational savoring task among 63 mothers of children aged 9–12, testing the hypothesis that rumination mediated the relationship between greater attachment insecurity and lower quality savoring responses. Quality of savoring responses was measured two ways—on a macro level, via a global coding measure, as well as on a micro level, using word count analysis. Results indicated that greater attachment insecurity was related to lower quality relational savoring responses, and that in the case of attachment avoidance, maternal rumination mediated this relationship. These findings underscore the link between maternal attachment insecurity and poorer quality savoring, with preliminary cross-sectional findings pointing to rumination as an indirect effect explaining the link between attachment avoidance and poorer quality savoring. An inability to relish the positive aspects of parent–child interactions may be one factor underlying insecure parents’ insensitive caregiving and their diminished satisfaction in parenting. If replicated, these findings could have implications for clinical intervention using relational savoring.

Keywords

Attachment attachment anxiety attachment avoidance parenting relational savoring rumination

Attachment theory holds that individuals develop internal working models (IWMs), or cognitive affective schemas regarding attachment relationships, based on the quality of early interactions between themselves and their primary caregivers (Ainsworth & Bell, 1970; Bowlby, 1980). When caregivers are consistently responsive to infants’ needs, a secure attachment develops. However, when caregivers are either consistently unresponsive (avoidant attachment) or inconsistently responsive (anxious attachment), one of two different types of attachment insecurity results. Individuals high in attachment avoidance learn to systematically turn their attention away from attachment-related thoughts and feelings, whereas people high in attachment anxiety are preoccupied with attachment-related thoughts and feelings (Cassidy & Berlin, 1994; Crowell & Waters, 1996).

In turn, these IWMs are thought to guide behavior in attachment relationships and caregiving behavior throughout the life span; for instance, parents high in avoidance are less involved and supportive in their parenting (Karavasilis, Doyle, & Markiewicz, 2003; Rholes, Simpson, & Blakely, 1995) and parents high in attachment anxiety tend to misread children’s cues and discourage their exploration (Selcuk et al., 2010). To date, little work has examined how maternal insecurity might undermine people’s positive experiences in relationships, as has been called for by researchers (e.g., Mikulincer & Shaver, 2007b). Here, we focus on one positive relational process within the parent–child relationship, relational savoring, a practice involving sustained and direct awareness of the pleasure derived from a positive relational experience (Borelli, Rasmussen, Burkhart, & Sbarra, 2014) and explore its association with parents’ attachment insecurity.

Savoring

Savoring is the conscious, reflective, and active process of enjoying and attending to pleasure during any general experience (Bryant & Veroff, 2007). Bryant and Veroff proposed that in addition to other positive experiences, those occurring in relationships can be savored, which might result in a greater sense of personal connection (Bryant, Smart, & King, 2005). While any experience is open to being savored, previous research has delineated savoring specifically focused on something occurring within a relationship and with an emphasis on the positivity within the relationship itself as relational savoring (e.g., Borelli et al., 2014); this has been measured previously in the context of the parent–child relationship by asking a mother to recall and describe a positive moment in which she felt connected to her child. To date, only a single study examined parents’ attachment and response to relational savoring, finding that highly avoidant parents of toddlers show the greatest increases in relationship satisfaction and the greatest decreases in self-reported negative emotions (Burkhart, Borelli, Rasmussen, & Sbarra, 2015); no studies have explored links between attachment and quality of relational savoring.

Savoring ability has been measured using the savoring beliefs inventory ( Bryant, 2003), which relies on individuals’ reports of their savoring abilities and tendencies. However, there may be methodological limitations of using self-report as the index of savoring behavior, as the accuracy of this measure rests on people’s awareness and honesty regarding savoring; likewise, savoring may not occur as a reportable experience for everybody, as for some, savoring may be so habitual that it occurs without being noticed. For these reasons, we employ a behavioral measure of relational savoring by assessing the degree to which parents show evidence of prolonging, enhancing, and enjoying a positive relational experience with their children. To that end, we subject mothers’ relational savoring narratives to two main analyses. First, using a macro lens, we measured parental ability to savor an experience with their children using a global coding system and, second, adopting a micro lens, we measured the length of the mother’s responses on the task, arguing that lengthier responses may indicate greater task engagement.

Attachment insecurity and relational savoring

Both attachment avoidance and anxiety may undermine maternal attempts to relationally savor. Avoidance may do so via two avenues—reduced tolerance for relational closeness and difficulty experiencing positive emotion. Regarding the former, avoidant individuals demonstrate low tolerance for closeness, disengaging during social interactions (Sroufe, 2005; Tidwell, Reis, & Shaver, 1996), and engaging in limited post-processing of attachment-related cognitions and emotions (Dozier & Kobak, 1992; Mikulincer, Birnbaum, Woddis, & Nachmias, 2000; Mikulincer, Orbach, & Iavnielli, 1998). Second, while ample evidence suggests that people high in avoidance downplay or deny negative affect (e.g., Mikulincer & Florian, 1995; Mikulincer & Orbach, 1995), relatively little work examines how avoidance impacts one’s ability to reflect on positive relational experiences. Relative to secure mothers, highly avoidant mothers convey less joy and pleasure when discussing their relationships with their children (Slade, Belsky, Aber, & Phelps, 1999). Likewise, high avoidance individuals report experiencing fewer positive emotions in their everyday lives (Searle & Meara, 1999) and process positive emotion words differently than their secure counterparts (Cohen & Shaver, 2004). Given that relational savoring requires attention to the positive aspects of parent–child interactions, avoidant mothers may be ill-equipped to fully engage in this task.

Attachment anxiety may also undermine mothers’ ability to relationally savor in two ways related to negative emotion activation and access to fewer positive memories. Individuals high in attachment anxiety demonstrate an inability to suppress negative thoughts and feelings (Mikulincer, Dolev, & Shaver, 2004) even when asked to focus on the positive elements of an experience (Ben-Naim, Hirschberger, Ein-Dor, & Mikulincer, 2013). Thus, when asked to savor a positive moment with their child, anxiously attached mothers may be unable to focus their attention specifically on the positive aspects long enough to perform the task requested of them. Second, anxiously attached individuals are more stressed as parents (Nygren, Carstensen, Ludvigsson, & Frostell, 2012), leaving them with fewer positive memories from which to choose.

In the current study, we examine these two subtypes of attachment insecurity (attachment anxiety and attachment avoidance), individually and together, as predictors of relational savoring quality.

Mediating role of rumination

Rumination, a repetitive, passive, and intense narrowing of attention on depressed feelings, their antecedents, and consequences (Nolen-Hoeksema, 1991; Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008), is in many ways antithetical to savoring. While both involve focused attention, rumination invokes attention to the negative, whereas the process of savoring invokes attention to the positive aspects of a situation. Fredrickson’s (2001) broaden and build framework suggests that positive emotions can set off a cascade of other positive emotions, theoretically building resiliency and expanding one’s attentional focus. Rumination, on the other hand, narrows the individual’s focus to negative thoughts and feelings, drawing attention inward via a focus on the self and the self’s problems (Mor & Winquist, 2002). As such, high ruminators may struggle to savor positive experiences given their tendency to focus on the negative aspects of a given situation.

The positive association between attachment anxiety and rumination is well documented (Caldwell & Shaver, 2012; Reynolds, Searight, & Ratwick, 2014; Saffrey & Ehrenberg, 2007)—in general, individuals high in attachment anxiety tend to use emotion focused, rather than solution focused, coping behaviors, including rumination (see Mikulincer & Shaver, 2007a for a comprehensive review). For highly anxious individuals, rumination allows them to stay with the negative emotion, fitting with the general hyperactive regulatory profile.

Rumination may also be associated with attachment avoidance, although to date, no studies have directly examined this relation. Attachment avoidance is related to the systematic denial of negative affect and suppression of emotions (e.g., Mikulincer & Florian, 1995; Mikulincer & Orbach, 1995; Mikulincer & Shaver, 2003), and previous research has neglected to test the link between attachment avoidance and rumination for this reason (e.g., Caldwell & Shaver, 2012). However, avoidant individuals theoretically take pains to turn their attention away from attachment-related thoughts (Mikulincer, Orbach, & Iavnielli, 1998), and as an inherently self-focused (Mor & Winquist, 2002), inward-looking process, rumination may be one strategy by which highly avoidant adults avoid attachment-related thoughts. Despite its emotional focus, rumination might also uniquely serve to assist avoidant individuals in redirecting their attention away from the emotional aspects of their experiences—regardless of whether or not these emotional experiences are attachment related. Rumination involves cognitive processes (Nolen-Hoeksema, 1991), including the exploration of why one is depressed or down; in shifting mental energy into exploring the causes or consequences of the negative emotion, highly avoidant individuals may actually reduce their feeling of the negative. This proposed avoidance function of rumination is similar to the argument that has been levied to explain the function of worry in generalized anxiety disorder—that engaging in the largely cognitive, repetitive process of worry actually removes the individual from their strong and aversive emotions, thus constituting experiential avoidance (Borkovec, Alcaine, & Behar, 2004).

In addition, rumination may therefore explain why insecure individuals have difficulty engaging in relational savoring. For people high in attachment anxiety, who may be unable to move away from their negative emotions, rumination might undermine the ability to relationally savor by keeping them directly in the negative emotion. For people high in avoidance, rumination may allow them to avoid attachment-related thoughts and emotions that emerge through relational savoring. Thus, here we evaluate this proposed theoretical model regarding the interrelations between attachment insecurity, rumination, and demonstrated proficiency in relational savoring.

Current study

In this study, we assessed mothers’ relational savoring narratives, operationalizing quality of relational savoring using coder ratings of positive connection to others and verbosity on the narratives. We tested four hypotheses. First, we predicted that maternal attachment insecurity would be negatively associated with relational savoring quality. Second, we anticipated that attachment insecurity would be positively associated with rumination. Third, we hypothesized that rumination would be negatively associated with relational savoring quality. Finally, we predicted that rumination would mediate the association between attachment insecurity and relational savoring quality.

Due to our desire to isolate the aspects of relational savoring performance that were associated with parent factors (avoidance, anxiety, and rumination) over and above child factors (e.g., child attachment avoidance), in all analyses, we controlled for child attachment security. Because there were theoretical reasons to suspect that both attachment avoidance and anxiety might be associated with poor savoring quality and rumination, for each hypothesis, we evaluated the unique and combined roles of avoidance and anxiety. For the latter, we created a mean insecurity score (mean of avoidance and anxiety scale score).

Method

Participants

Participants were recruited via Internet posts, flyers, and word of mouth to participate in the first wave of data collection within a larger study. Mothers (N = 68) participated in an initial assessment (Time 1) in which they reported on demographic factors, attachment, and rumination. Mothers returned at Time 2, approximately 1.5 years later, and completed the relational savoring task. We were missing data from five participants, and a frequency check revealed no pattern to the missing data. The resulting sample of mothers (n = 63; M _age = 38.16 years; SD _age = 6.45) and their children (38% male; M _age = 9.9 years, SD _age = 1.41) was ethnically and socioeconomically diverse (41.3% Hispanic, 31.7% Caucasian, 15.9% African American, 4.8% Asian, 4.8% Native American, and 7.9% Other; 52.4% of the parents reported an income of less than USD$40,000 per year). Approximately 31.7% of the mothers were single, 58.7% were married or had a domestic partner, and 9.5% were widowed. Most mothers (50.8%) reported having one or two children in the home.

Materials and procedure

The study’s protocol had previously been approved by the Institutional Review Board. At each assessment, mothers and children provided informed consent and assent, respectively.

Self-reported attachment style

Participants completed the Revised Experiences in Close Relationships (ECR-R) Questionnaire, a 36-item survey designed for use with adults measuring attachment anxiety and attachment avoidance ( Fraley, Waller & Brennan, 2000). Adults rated 18 items assessing attachment avoidance (e.g., “I prefer not to show my partner how I feel deep down”) and 18 assessing attachment anxiety (e.g., “I often worry that my partner doesn’t really love me”). All items are rated on a 7-point scale, with scores ranging from 1 (strongly disagree) to 7 (strongly agree). Low scores on both the anxiety and avoidance scales signify secure attachment, whereas elevated scores on one or both of the scales signify insecure attachment (Brennan, Clark, & Shaver, 1998; Parker, Johnson, & Ketring, 2011). The ECR-R has demonstrated acceptable levels of reliability (Parker et al., 2011) and construct validity (Brennan et al., 1998). Cronbach’s α in this sample was good (avoidance = .91, anxiety =.88). In this study, we utilized both subscales individually and also calculated the mean of subscale scores (henceforth referred to as attachment insecurity, with higher scores indicating greater attachment insecurity).

Rumination

Mothers completed the rumination subscale of the Response Style Questionnaire (Nolen-Hoeksema & Morrow, 1991), a 22-item (e.g., “Think about all your shortcomings, failings, faults, mistakes”) questionnaire in which adults indicate how often they engage in particular actions when feeling down or depressed ranging from 1 (almost never) to 4 (almost always). Responses were totaled and averaged (range: 1–4); higher scores indicate greater rumination. The construct validity and reliability of this measure have been documented (Butler & Nolen-Hoeksema, 1994; Nolen-Hoeksema, 2000), and in this sample, Cronbach’s α was .94.

Child attachment

Children completed the Child Attachment Interview (CAI; Shmueli-Goetz, Target, Fonagy, & Datta, 2008; Target, Fonagy, Shmueli-Goetz, Datta, & Schneider, 1999) at Time 1. The CAI is a semi structured interview that utilizes 19 questions to elicit child reflection on attachment experiences (e.g., a time when the child was afraid and used the mother as a secure base) with their primary caregiver(s). Per CAI guidelines, interviews were conducted by trained researchers and videotaped interviews were transcribed and then coded according to 9-point rating scales. In this study, a certified coder rated all interviews; a second certified coder double-coded 16 cases, with adequate interrater reliability on the four attachment categories (four-way κ = .91, p < .001).

Consistent with recommendations (Roisman, Fraley, & Belsky, 2007), we conducted an exploratory factor analysis of the CAI data yielding two factors, an avoidance factor (called security/dismissal) and an anxious factor (called security/preoccupation; see Borelli, Somers, West, Coffey, Reyes, & Shmueli-Goetz, 2015, for more information), which is consistent with prior studies of the CAI (Borelli et al., 2015; Zachrisson, Røysamb, Oppedal, & Hauser, 2011). We utilized both CAI factors in analyses, which ranged from a score of 3 (highly secure) and −3 (highly dismissing or highly preoccupied, respectively).

Coder-rated relational savoring

Quality of relational savoring was measured using a writing task created for the current study but modeled after previously developed protocols (Borelli, Rasmussen et al., 2014; Borelli, McMakin, & Sbarra, 2010) aimed at measuring positivity and relational themes within savoring narratives. Mothers answered the following prompt at T2:
For this next task, we would like you to think of a positive emotional experience you have had with your child and write about this memory. This could be something as simple as enjoying a good meal together or laughing over a funny joke, or it could be something as major as taking a vacation or accomplishing a big task together. Try to focus on a single memory of a time when things seemed at their best between you two. Using as much detail as possible, describe the situation and how you felt at the time (think about specific details and how you were feeling—What were you doing? What kinds of feelings were you feeling?). Spend a few minutes thinking about your response before beginning to write, imagining what it was like to have been there. Write as much, or as little, as you feel comfortable with.

Prior to their completing the task, research assistants informed mothers that grammar and spelling would not count against them when answering the savoring prompt.

To globally evaluate relational savoring, we created a coding system based on prior work on relational savoring (Burkhart et al., 2015, Burkhart, Borelli, Rasmussen, Brody, & Sbarra, 2016). We operationalized relational savoring quality via three main factors: (1) the degree to which it involved a specific, explicitly positive memory between a mother and a particular child; (2) the degree to which the memory referenced the mother–child relationship or aspects of emotional connectedness between the mother and child; and (3) the degree to which memory involved utilization of adequate detail, including references to feelings, and self-other reflection (e.g., “I know he was feeling this too”). Eight trained coders rated each relational savoring response according to a 5-point scale, with greater numbers indicating greater ability to relationally savor. Prior to coding participant responses, coders were first trained and tested for agreement using 11 savoring texts not under analysis in the current study, achieving a high level of agreement (average measures intraclass coefficient [ICC] = .98). Afterward, coders who were blind to mother and child attachment classification scored the savoring transcripts for the whole sample; reliability for this sample was excellent (average measures ICC = .95). For the purposes of analyses, we computed the average savoring score across all eight coders for each participant (M_score = 2.84, SD = 0.99). See Figure 1 for examples of relational savoring responses.

Figure 1.
Coder-rated relational savoring excerpt.

Savoring word count

To measure the length of maternal savoring responses, data were subject to a textual analysis via the linguistic inquiry word count program (LIWC; Pennebaker, Francis, & Booth, 2001). First, transcribed responses were edited according to LIWC protocol (e.g., Pennebaker, Mehl, & Niederhoffer, 2003) that includes eliminating filler words, ensuring that all words are spelled correctly, and editing contractions to ensure proper tabulation (e.g., changing “Jeff’s going to the mall” to “Jeff is going to the mall”). Edited and checked transcripts were run through the LIWC computer program; we focused specifically on the length of savoring responses in the current study, using the word count variable (which tallies total words in participant responses after the editing process). See Figure 1 for examples.

Data analytic plan

To evaluate Hypotheses 1 and 2, we conducted hierarchical linear regressions. To conduct mediation analyses, we used PROCESS Model 4 (Hayes, 2012) that employs bootstrapping to calculate the indirect effect of an independent variable on a dependent variable through a mediating variable (see Hayes, 2013 for a comprehensive review). In all analyses, we controlled for child age, child sex, maternal age, and child attachment security/dismissal (analyses involving maternal attachment avoidance) or security/preoccupation (analyses involving maternal attachment anxiety).

Results

Table 1 reports means and standard deviations for all continuous variables. A one-way analysis of variance revealed that length of savoring response did not significantly differ by participant educational attainment, F(5, 57) = .484, p = .787; likewise, number of children present in the home did not significantly correlate with any outcome measure. Table 2 reports the results of all bivariate correlations between key variables. As predicted, coder-rated savoring was negatively associated with attachment avoidance (r = −.235, p < .05, one-tailed), attachment anxiety (r = −.284, p < .05), attachment insecurity (r = −.293, p < .05), and rumination (r = −.297, p < .05). Further, savoring word count was negatively associated with attachment avoidance (r = −.265, p < .05), attachment anxiety (r = −.333, p < .01), attachment insecurity (r = −.338, p < .01), and rumination (r = −.303, p < .05). Finally, rumination was positively associated with attachment avoidance (r = .375, p < .01), attachment anxiety (r = .624, p < .01), and attachment insecurity (r = .563, p < .01).

Table 1.
Key variable descriptive statistics by child gender.

Total Males Females

Measures N = 63 n = 24 n = 39

Maternal attachment avoidance^a 2.65 (1.16) 2.82 (1.54) 2.54 (0.85)

Maternal attachment anxiety^a 2.71 (1.11) 2.60 (0.97) 2.78 (1.20)

Maternal attachment insecurity^a 2.68 (1.00) 2.71 (1.17) 2.67 (0.90)

Maternal rumination scores 1.71 (0.53) 1.78 (0.53) 1.68 (0.52)

Coder-rated maternal savoring 2.84 (0.99) 2.89 (1.17) 2.81 (0.88)

Maternal savoring word count 109.67 (67.47) 119.92 (81.24) 103.36 (57.66)

Child attachment anxiety^b 0.18 (0.97) 0.09 (1.04) 0.13 (.94)

Child attachment avoidance^b 0.06 (1.07) −0.24 (1.01) 0.24 (1.07)

Note. CAI = Child Attachment Interview; ECR-R = Revised Experiences in Close Relationships.

^aMeasured by the ECR-R.

^bChild attachment anxiety/avoidance as measured by the CAI—high scores equal low avoidance/high security or low anxiety/high security.

Table 2.
Correlation matrix for key variables.

Variable 1 2 3 4 5 6 7 8 9 10

1 Child age –

2 Maternal age −.033 –

3 Maternal coder-rated Savoring −.018 .002 –

4 Maternal savor word count −.059 −.079 .626** –

5 Maternal attachment avoidance .030 −.011 −.235^a −.265* –

6 Maternal attachment anxiety .000 −.078 −.284* −.333 .559 –

7 Maternal attachment insecurity .018 −.050 −.293* −.338 .887 .878** –

8 Child avoidance^b .056 −.219 .199 −.012 .045 .081 .071 –

9 Child anxiety^b −.041 .114 .083 .144 −.184 −.314* −.281* −.067 –

10 Maternal rumination .192 .170 −.297* −.303* .375 .624 .563** .080 −.161 –

Note. CAI = Child Attachment Interview.

^aSignificant p < .05 one-tailed.

^bHigh scores equal high attachment security/low anxiety/avoidance (in the CAI scoring system, these are actually referred to as preoccupation and dismissal, but for consistency’s sake, we refer to them in terms parallel to the adult attachment style literature)

^* p < .05 (two-tailed); ** p < .01 (two-tailed).

Hypothesis 1: Does maternal attachment predict savoring quality?

After controlling for covariates (child sex, child age, parent age, and child attachment), attachment avoidance was negatively associated with coder-rated savoring, ΔR ² = .062, β = −.252, p = .05 (see Table 3); given the directional nature of our hypothesis, we interpret this result as being statistically significant. Similarly, after controlling for covariates, attachment avoidance was negatively associated with savoring word counts, ΔR ² = .081, β = −.288, p = .027.

Table 3.
Attachment avoidance predicting savoring quality and rumination.

Mean savor score Savor word count Rumination

b SE 95% CI b SE 95% CI b SE 95% CI

Step 1 R ² 0.06 0.03 0.10

Constant 4.12* 1.25 [1.62, 6.63] 243.52* 85.64* [72.04, 415.00] 0.11 0.62 [−1.14, 1.35]

Child age −0.06 0.09 [−0.25, 0.12] 0.02 6.31 [−12.62, 12.65] 0.08 0.05 [−0.01, 0.17]

Child sex −0.17 0.28 [−0.73, 0.38] −25.43 19.07 [−63.61, 12.76] −0.11 0.14 [−0.39, 0.17]

Parent age 0.004 0.02 [−0.04, 0.04] −1.27 1.37 [−4.01, 1.47] 0.02 0.01 [−0.01, 0.04]

CAI avoidance^a 0.22 0.12 [−0.02, 0.47] 0.99 8.26 [−15.56, 17.53] 0.06 0.06 [−0.06, 0.18]

Step 2 ΔR ² 0.06* 0.08* 0.12*

Attachment avoidance −0.22* 0.11 [−0.43, 0.00] −16.81* 7.40 [−31.62, −1.99] 0.16* 0.05 [0.05, 0.27]

Note. CAI = Child Attachment Interview.

^aCAI avoidance: Highly secure to highly avoidant.

p < .05.

After controlling for covariates, attachment anxiety was negatively associated with coder-rated savoring, ΔR ² = .075, β = −.290, p = .034 (see Table 4); a second regression revealed that after controlling for covariates, attachment anxiety was negatively related with savoring word count, ΔR ² = .088, β = −.314, p = .019.

Table 4.
Attachment anxiety predicting savoring quality and rumination.

Mean savor score Savor word count Rumination

b SE 95% CI b SE 95% CI b SE 95% CI

Step 1 R ² 0.02 0.05 0.11

Constant 4.40 1.28 [1.85, 6.96] 255.02* 84.49 [85.84, 424.20] −0.34 0.50 [−1.34, 0.661]

Child age −0.08 0.09 [−0.27, 0.11] −1.18 6.19 [−13.58, 11.22] 0.10* 0.04 [0.02, 0.17]

Child sex 0.04 0.28 [−0.52, 0.59] −16.64 18.36 [−53.41, 20.13] −0.20 0.11 [−0.41, 0.02]

Parent age −0.003 0.02 [−0.04, 0.04] −1.46 1.33 [−4.12, 1.21] 0.02 0.01 [−0.001, 0.03]

CAI anxiety^a −0.01 0.14 [−0.29, 0.26] 4.40 9.10 [−13.82, 22.61] 0.02 0.05 [−0.09, 0.13]

Step 2 ΔR ² 0.08* 0.08* 0.40*

Attachment anxiety −0.26* 0.12 [−0.50, −0.02] −19.01* 7.87 [−34.78, −3.25] 0.31* 0.05 [0.22, 0.41]

Note. CAI = Child Attachment Interview.

^aCAI anxiety: Highly secure to highly anxious.

p < .05.

After controlling for covariates, insecurity was negatively associated with coder-rated savoring, ΔR ² = .086, β = −.307, p = .020 (see Table 5); similarly, after controlling for covariates, attachment insecurity was negatively associated with savoring word count, ΔR ² = .099, β = −.329, p = .013.

Table 5.
Attachment insecurity predicting savoring quality and rumination.

Mean savor score Savor word count Rumination

b SE 95% CI b SE 95% CI b SE 95% CI

Step 1 R ² 0.07 0.05 0.13

Constant 4.40 1.25 [1.89, 6.91] 262.52* 85.24 [91.75, 433.28] −0.29 0.55 [−1.40, 0.82]

Child age −0.07 0.09 [−.25, .11] −0.34 6.20 [−12.76, 12.08] 0.08* 0.04 [0.002, 0.16]

Child sex −0.13 0.27 [−.68, .42] −22.46 18.62 [−59.76, 14.83] −0.13 0.12 [−0.38, 0.11]

Parent age −0.003 0.02 [−.04, .04] −1.42 1.35 [−4.14, 1.29] 0.02 0.01 [−0.001, 0.03]

CAI avoidance^a 0.23 0.12 [−.01, .47 1.47 8.13 [−14.82, 17.76] 0.05 0.05 [−0.06, 0.15]

CAI anxiety^b 0.01 0.13 [−.26, .27] 5.03 9.01 [−13.02, 23.07] −0.01 0.06 [−0.12, 0.11]

Step 2 ΔR ² 0.09* 0.10* 0.29*

Insecure attachment^c −0.30* 0.13 [−.56, −.05] −22.16* 8.65 [−39.50, −4.83] 0.29* 0.06 [0.18, 0.41]

Note. CAI = Child Attachment Interview.

^aCAI avoidance: Highly secure to highly avoidant.

^bCAI anxiety: Highly secure to highly anxious.

^cMean score of attachment avoidance and attachment anxiety subscales.

p < .05.

In addition, results show that all three metrics of attachment (avoidance, anxiety, and insecurity) were associated with both the coder-rated and the linguistic indices of poorer relational savoring quality.

Hypothesis 2: Does attachment predict rumination?

Controlling for covariates (child sex, child age, parent age, and child attachment) on Step 1, attachment avoidance was positively related to maternal rumination, ΔR ² = .123, β = .354, p = .004, 95% CI [0.054, 0.269] (see Table 3); similarly, attachment anxiety was positively related to maternal rumination, ΔR ² = .395, β = .666, p = .000, 95% CI [0.221, 0.407] (see Table 4). Finally, after controlling for covariates, R ² = .126, p =* .163, attachment insecurity was positively related to maternal rumination, ΔR ² = .285, β = .557, p = .000, 95% CI [0.180, 0.405] (see Table 5). Thus, all three metrics of attachment insecurity were associated with higher rumination.

Hypothesis 3: Does rumination predict relational savoring quality?

After controlling for covariates (child sex, child age, and parent age), R ² = .058, p = .478, greater maternal rumination was negatively associated with coder-rated savoring scores, ΔR ² = .104, β = −.339, p = .010, 95% CI [−1.123, −.157]. Similarly, controlling for covariates, R ² = .027, p = .808, greater maternal rumination was negatively related to savoring response word counts, ΔR ² = .095, β = −.325, p = .016, 95% CI [−75.329, −8.046]. Thus, rumination was associated with poorer quality relational savoring as assessed by both coder-rated and linguistic measures.

Hypothesis 4: Does rumination mediate between attachment and savoring quality?

The results of Hypotheses 2 and 3 provided the foundation for testing three mediation models. In each model, we tested the hypothesis that rumination mediated the link between attachment and savoring quality using PROCESS (Hayes, 2012); in PROCESS, the point estimate and confidence interval for the indirect effect of the mediator are calculated, and if the confidence interval does not eclipse 0, one can conclude that the indirect effect is significant (p < .05); see Tables 6 and 7 for complete mediation results.

Table 6.
Mediation analyses predicting coder-rated savoring scores.

DV: Coder-rated savoring

Avoidance Anxiety Insecurity

b SE 95% C.I. b SE 95% C.I. b SE 95% C.I.

Constant 4.18* 1.22 [1.74, 6.62] 4.28* 1.28 [1.72, 6.85] 4.27* 1.24 [1.78, 6.76]

Child age −0.02 0.09 [−0.20, 0.16] −0.05 0.10 [−0.25, 0.15] −0.03 0.09 [−0.22, 0.15]

Child sex −0.23 0.27 [−0.77, 0.32] −0.03 0.28 [−0.60, 0.54] −0.18 0.27 [−0.73, 0.36]

Parent age 0.01 0.02 [−0.03, 0.05] 0.002 0.02 [−0.04, 0.04] 0.01 0.02 [−0.03, 0.05]

CAI avoidance^a 0.25 0.12 [0.02, 0.49] – – – 0.25* 0.12 [0.01, 0.49]

CAI anxiety^b – – – −0.003 0.14 [−0.28, 0.27] 0.006 0.13 [−0.26, 0.26]

Rumination −0.53^d 0.26 [−1.05, −0.01] −0.35 0.34 [−1.03, 0.33] −0.43 0.30 [−1.03, 0.17]

Avoidance −0.13 0.11 [−0.36, .010] – – – – – –

Anxiety – – – −0.15 0.16 [−0.47, 0.17] – – –

Insecurity^c – – – – – – −0.18 0.15 [−0.49, 0.13]

Note. CAI = Child Attachment Interview.

^aCAI avoidance: Highly secure to highly avoidant.

^bCAI anxiety: Highly secure to highly anxious.

^cMean score of attachment avoidance and attachment anxiety subscales.

^dIndirect effect in this pathway was significant, point estimate = −0.09, 95% CI [−.23, −.006].

p < .05

Table 7.
Mediation analyses predicting savoring word count.

DV: Savor word count

Avoidance Anxiety Insecurity

b SE 95% C.I. b SE 95% C.I. b SE 95% C.I.

Constant 246.86 84.08 [78.44, 415.28] 248.03* 84.94 [77.87, 418.20] 256.25* 85.39 [85.12, 427.38]

Child age 2.53 6.35 [−10.20, 15.25] .78 6.56 [−12.35, 13.92] 1.43 6.42 [−11.45, 14.30]

Child sex −28.83 18.82 [−66.52, 8.86] −20.70 18.91 [−58.58, 17.19] −25.31 18.80 [−62.99, 12.37]

Parent age −0.79 1.37 [−3.54, 1.95] −1.15 1.38 [−3.90, 1.60] −1.08 1.39 [−3.87, 1.71]

CAI avoidance^a 2.77 8.17 [−13.60, 19.14] – – – 2.50 8.19 [−13.91, 18.91]

CAI anxiety^b – – – 4.86 9.12 [−13.41, 23.13] 4.92 9.00 [−13.12, 22.96]

Rumination −31.82^d 17.88 [−67.65, 4.00] −20.64 22.45 [−65.62, 24.34] −21.40 20.56 [−62.61, 19.81]

Avoidance −11.68 7.81 [−27.33, 3.97] – – – – – –

Anxiety – – – −12.53 10.58 [−33.72, 8.65] – – –

Insecurity^c – – – – – – −15.91 10.53 [−37.01, 5.19]

Note. CAI = Child Attachment Interview.

^aCAI avoidance: Highly secure to highly avoidant.

^bCAI anxiety: Highly secure to highly anxious.

^cMean score of attachment avoidance and attachment anxiety subscales.

^dIndirect effect in this pathway was significant, point estimate = −5.13, 95% CI [−14.70, −0.28].

*p < .05

The indirect effect of rumination on the relationship between attachment avoidance and both coder-rated savoring (point estimate = −0.09, 95% CI [−0.23, −0.004]) and savoring word count (point estimate = −5.13, 95% CI [−14.70, −0.28]) was significant. In contrast, the indirect effects of rumination on the relationship between attachment anxiety and both coder-rated savoring (point estimate = −.11, 95% CI [−0.35, 0.09]) and savoring word count (point estimate = −6.48, 95% CI [−23.47, 6.01]) were not significant. Similarly, the indirect effect of rumination on the relationship between attachment insecurity and both coder-rated savoring (point estimate = −.13, 95% CI [−0.38, 0.03]) and savoring word count (point estimate = −6.25, 95% CI [−20.40, 3.58]) was not significant.

As an additional step, we reran all analyses switching the independent and mediating variable—none of the indirect effects were significant. We also reran analyses switching the mediator and dependent variable—here only one indirect effect (word count as a mediator of the association between avoidance and rumination, point estimate = 0.03, 95% CI [0.001, 0.079]) was statistically significant.

Thus, our findings suggest that rumination only acted as a statistically significant indirect effect in explaining the association between attachment avoidance, but not attachment anxiety or mean insecurity, and relational savoring quality.

Discussion

Savoring has emerged as a promising technique within the research and clinical literatures (e.g., Burkhart et al., 2015; McMakin, Siegle, & Shirk, 2011). Here, we evaluated the link between maternal attachment insecurity and relational savoring quality as well as whether rumination explains this link.

Attachment insecurity and relational savoring quality

In line with our predictions, we found that mothers with greater attachment insecurity (avoidance, anxiety, and mean insecurity) demonstrated lower quality savoring of relational memories, as measured behaviorally using both macro and micro-level measures. Relational savoring involves focusing on and consciously attending to positive moments of emotional connection with another person; for a number of reasons, this may not be an emotionally straightforward task for parents high in attachment insecurity.

With regard to attachment avoidance, we believe there are at least three possible reasons for the observed effect. First, attachment avoidance is characterized by an inability or unwillingness to access attachment-related memories or experiences (Crowell & Waters, 1996; Mikulincer et al., 1998). Therefore, highly avoidant mothers may have fared worse on the task due to an inability to access an emotionally charged, or relationally relevant, episode occurring within the mother–child relationship. Second, it could be that mothers high in avoidance never formed the close memories the task elicits. Using diaries of participants’ social interactions, Tidwell, Reis, and Shaver (1996) found that the entries of avoidant adults were more socially disengaged, and highly avoidant adults tend to habitually distance themselves from others (Mikulincer et al., 1998); thus, it remains possible that mothers high in attachment avoidance lacked the initial engagement in previous interactions with their children, making it difficult for them to recall a moment of closeness at all.

A third interpretation of the findings pertains to the processing and interpretation of positive emotion. Avoidant adults report experiencing significantly less positive emotion in their everyday lives than their secure counterparts (Searle & Meara, 1999) and process positive emotions differently than secure individuals (Cohen & Shaver, 2004). It is possible that mothers high in attachment avoidance generated poorer quality relational savoring narratives due to difficulty in accessing a memory they would count as a “positive” one—that is, having experienced less positive emotion in their everyday interactions with their child, the mothers in our study may have been unable to have attended to the positivity in the first place. It is also possible that for avoidant mothers, closeness to one’s child does not feel subjectively positive, and the task was therefore difficult given that the two qualities of the experiences being probed (close memories and positive memories) were incompatible with one another. Future work ought to test whether a similar pattern emerges when adults high in attachment avoidance savor memories of other close relationships—for example, with their romantic partners—to further clarify the contextual factors that might influence the results observed in the current study.

With regard to attachment anxiety, we believe there are two possible interpretations of our results, related to emotion and exploration. First, previous research indicates that anxious mothers adopt a strategy to keep their children close that involves intermittent responsiveness more heavily weighted toward their child’s expressions of negative emotion, such as fear (Cassidy & Berlin, 1994; Haft & Slade, 1989). This runs counter to the savoring prompt utilized in the current study—to recall and expand upon a memory that is both positive and relational. For anxious mothers, it is possible that feeling positively about interactions with their child manifests primarily through their child’s expressions of negative emotion—not positive emotion—and that this leads to some difficulty in both recalling a memory that fits the task and expanding upon that memory in a way that preserves the intent of the task. Indeed, one study found that anxious mothers had difficulty staying positive when directed to do so (Ben-Naim et al., 2013), and future work could explore whether anxious mothers perform better when asked to savor a personal memory rather than a relational one.

A second explanation for the pattern of observed results might be that anxious mothers’ ability to relationally savor was undermined by the focus on secure base or exploration-focused memories (experiences when the mother was supporting the child’s growth or exploration; e.g., “accomplishing a big task together”) in the relational savoring task. Anxious mothers have difficulty supporting their children’s exploratory behaviors and often interfere with exploration even when their child is not demonstrating a need for their attention (see Cassidy & Berlin, 1994). Reflecting on the child’s exploration may have activated anxious mothers’ ambivalence toward their children’s exploration, as the focus on the two of them doing something growth oriented may be less comfortable than reflecting on a time when the mother comforted the child. Instead of improving high anxiety mothers’ emotional states, focusing on these memories may dysregulate them (e.g., Haft & Slade, 1989).

Attachment insecurity and rumination

We also found that greater attachment avoidance, greater attachment anxiety, and greater attachment insecurity were all individually associated with higher rumination. To our knowledge, no other studies have demonstrated the link between attachment avoidance and rumination; in fact, when it comes to rumination, attachment avoidance has been largely overlooked. This may be due to the fact that the hyperactive profile of high attachment anxiety phenotypically resonates more with clinical conceptualizations of rumination (Reynolds et al., 2014; Saffrey & Ehrenberg, 2007)—a relationship that is well established (e.g., Caldwell & Shaver, 2012) and again demonstrated here. Despite this, we believe it makes conceptual sense to think that avoidance and rumination might be associated. For instance, we theorized that as with the argument posed with respect to the avoidance function of worry (Borkovec et al., 2004), rumination may offer some utility in an avoidant individual’s quest to detach from their emotions—that it might be one strategy, among a possible many, which prevents avoidant individuals from accessing relational memories and allows them to avoid deep emotional processing (Dozier & Kobak, 1992). Engaging in a repetitive search for causes and consequences surrounding the negative mood may assist high avoidance adults in suppressing their emotions (Mikulincer & Florian, 1995; Mikulincer & Orbach, 1995; Mikulincer & Shaver, 2003).

By definition, rumination is also self-focused; indeed, scholars argue that negative mood can increase self-focus and that avoidant individuals may use this self-focus as a way to avoid attachment-related thoughts and feelings (as in highlighting self-other differences; Mikulincer, 1994; Mikulincer et al., 1998). Future work should seek to further explore this relationship, in the interest of identifying whether attachment avoidance and rumination are consistently linked, and under what conditions this association is strengthened or weakened.

Rumination and savoring quality

In addition, rumination was negatively associated with quality of maternal relational savoring; to our knowledge, this is the first study to examine these two constructs in tandem. As previously mentioned, savoring involves the invocation of, and prolonged attention to, positive memories (Bryant & Veroff, 2007); conversely, rumination involves a perseverative focus on negative affect, as evidenced by its role in both depression and anxiety (Nolen-Hoeksema, 2012). There are two possible explanations for this relationship. First, ruminative mothers’ attention may be focused on negative experiences, and therefore they may be less able to access or focus their attention on a positive memory with their child. Second, cognitive overload might occur when mothers high in rumination are asked to savor; that is, the incongruence between predominant emotional state (negative) and the task (e.g., savoring a positive memory) may place greater pressure on self-control resources necessary to refocus one’s attention. Certain forms of behavioral performance suffer under conditions of emotional and cognitive misalignment, such as when individuals are induced to feel a negative emotion and then subject to working memory and Stroop tasks (Gray, 2001; Storbeck, 2012). Further, people who are induced to feel positive emotion display more cognitive flexibility on behavior tasks requiring less fixed responses (Dreisbach & Goschke, 2004). In the current study, high ruminating mothers described memories with children with less engagement and positivity, potentially as a result of low cognitive flexibility exacerbated by the misalignment of their emotional focus (negative) with the demands of the positively valenced task.

Rumination as mediator of the link between attachment insecurity and savoring quality

Finally, we found that rumination mediated the association between attachment avoidance and savoring quality, suggesting that rumination may be one mechanism undermining avoidant individuals’ ability to relationally savor. In contrast, rumination did not mediate the association between attachment anxiety or attachment insecurity and maternal ability to relationally savor. While avoidance and its relation to the experience of positive emotion (Searle & Meara, 1999) and savoring (Burkhart et al., 2015) have been explored previously, this is the first study to explore rumination as a mechanism through which avoidant individuals less fully explore and describe their positive emotions. For avoidant adults, rumination may allow them to shift away from attachment-related, and possibly emotionally provocative, feelings, and move them into a realm of perseverative and self-focused cognition (e.g., thinking about the causes and correlates of their feelings, rather than the feelings themselves). The tendency to ruminate may explain why people high in avoidance demonstrate lower quality relational savoring, a task that requires a focus on experiences of closeness in the parent–child relationship.

That rumination did not mediate the relationship between attachment anxiety and savoring quality was unexpected, but this may mean that other factors explain the link between attachment anxiety and savoring quality, including negative emotion itself, which may be manifested in dampening or spoiling of positive memories. However, attachment anxiety entails hyperactive attention to the emotional elements present in attachment relationships (Cassidy & Berlin, 1994)—and although anxiety and rumination share a hyperactive quality, rumination’s underpinning in cognition, rather than emotion, may preclude it as a dominant manner in which anxious mothers’ ability to relationally savor is undermined.

Although speculative at this point, this pattern of findings could suggest that it is the relational and emotional aspects of relational savoring, rather than the positivity aspect, that renders it incompatible with a ruminative response style—this could explain why rumination uniquely mediates the link between attachment avoidance and not anxiety. Individuals high in avoidance may be more likely to struggle with the relational and emotional demands of the relational savoring task, whereas people high in anxiety may be adept at maintaining focus on relational and emotional experiences but may struggle with preserving attentional focus on positive emotion. This proposition contains many untested elements, and as such is a question that should be tested in future investigations. One strategy for testing this idea would be to individually code the relational, emotional, and positive aspects of relational savoring narratives and examine the association between these three dimensions of relational savoring quality for their association with avoidance and anxiety.

Implications

If replicated with a larger sample, these findings may have important implications. Previous research has examined savoring as a treatment outcome measure (e.g., Meyer, Johnson, Parks, Iwanski, & Penn, 2012) as well as an intervention method (McMakin et al., 2011). Savoring reduces depression, increases happiness, increases positive emotion, reduces negative emotion, and increases relationship satisfaction ( Borelli, Rasmussen, et al., 2014; Borelli, Sbarra, et al., 2014; Bryant et al., 2005; Burkhart et al., 2015; McMakin et al., 2011; Schueller, 2010). Given this trend, understanding the mechanisms underlying the efficacy of savoring is important. Our results suggest that attachment avoidance, attachment anxiety, and attachment insecurity are associated with lower proficiency in savoring an experience within the parent–child relationship, and for avoidant individuals, the mutual association with rumination may explain this relationship. Thus, clinical interventions centered on helping avoidant individuals to become comfortable savoring relational experiences may be most successful if constructed within a framework of gradated exposure. Further, future research should test whether an intervention aimed at reducing rumination is efficacious in improving the relational savoring ability of avoidant adults or whether a focus on improving relational savoring ability of avoidant adults reduces rumination.

These findings warrant further exploration of pathways explaining the relationship between attachment anxiety/insecurity and poor savoring quality; future research may focus on emotion or interpersonal constructs rather than cognitive ones as potential mediators. In general, anxious adults might experience difficulty savoring secure base moments with their children but do better savoring safe haven moments—contexts in which closeness vis-à-vis child negative emotion are more readily established. Future research might explore whether experimentally manipulating the relational focus of the savoring (e.g., secure base versus safe haven) produces different results.

Limitations and future directions

We note here three main limitations with regard to the current study. First, in measuring the savoring construct, participants typed their responses to the prompt, and the length of participant responses varied widely. This could have been influenced by participants’ ability to express themselves in writing, despite assurances that grammar and spelling would not count. Although response length did not differ as a function of participant income or educational attainment, it is possible that shorter responses (which we believe indicated less engagement with the relational savoring task) may have been due to writing fatigue rather than task disengagement. To complement these findings, researchers may wish to encourage participants to verbalize their savoring responses in future work.

Our capacity to evaluate a causal model in the current study was also limited given its correlational nature. Although we incorporated a short-term longitudinal design, suggesting that attachment insecurity and rumination temporally preceded poorer quality savoring, our data (e.g., only having each measure at one time point) do not permit us to ascertain whether insecurity causes poorer quality savoring (in the case of avoidance, through rumination), whether poorer quality savoring causes insecurity, or whether these two factors work transactionally. Future work should use an experimental design to more adequately assess the question of causality, such as one utilizing attachment security primes (e.g., Mikulincer & Shaver, 2007b) to alter relational savoring ability, or use of a longitudinal randomized control intervention to examine the possibility that relational savoring reduces insecurity or rumination. Further, although our findings suggest that rumination is an indirect effect explaining the link between avoidance and savoring quality, undoubtedly there are numerous mediators of this association that were not tested in the current investigation.

There are several promising future directions for continued research. For example, recent studies involving savoring have manipulated elements of time—directing participants to savor events that happen in the future, for instance (e.g., Quoidbach, Wood, & Hansenne, 2009). While it is reasonable to expect that rumination may only undermine the savoring of previously occurring events given its retrospective nature, future research might explore whether the pattern of results found here apply when the temporal focus is placed ahead, rather than behind, the participant.

Conclusions

This study informs our understanding of maternal attachment insecurity and its association with relational savoring. The results suggest that attachment avoidance, attachment anxiety, and attachment insecurity, in general, are all associated with lower quality of relational savoring narratives and that in the case of attachment avoidance, this may be driven by greater rumination. Perhaps highly avoidant adults have a difficult time engaging with the emotional demands of a relational savoring task, leading to avoidance via rumination, which in turn is associated with lower quality relational savoring. The present work marks one small step in the process of elucidating the processes through which attachment insecurity in general, and avoidance in particular, are associated with difficulties embracing the positive, relational aspects of parenthood.

	Total	Males	Females
Maternal attachment avoidance^a	2.65 (1.16)	2.82 (1.54)	2.54 (0.85)
Maternal attachment anxiety^a	2.71 (1.11)	2.60 (0.97)	2.78 (1.20)
Maternal attachment insecurity^a	2.68 (1.00)	2.71 (1.17)	2.67 (0.90)
Maternal rumination scores	1.71 (0.53)	1.78 (0.53)	1.68 (0.52)
Coder-rated maternal savoring	2.84 (0.99)	2.89 (1.17)	2.81 (0.88)
Maternal savoring word count	109.67 (67.47)	119.92 (81.24)	103.36 (57.66)
Child attachment anxiety^b	0.18 (0.97)	0.09 (1.04)	0.13 (.94)
Child attachment avoidance^b	0.06 (1.07)	−0.24 (1.01)	0.24 (1.07)

	Variable	1	2	3	4	5	6	7	8	9	10
1	Child age	–
2	Maternal age	−.033	–
3	Maternal coder-rated Savoring	−.018	.002	–
4	Maternal savor word count	−.059	−.079	.626**	–
5	Maternal attachment avoidance	.030	−.011	−.235^a	−.265*	–
6	Maternal attachment anxiety	.000	−.078	−.284*	−.333**	.559**	–
7	Maternal attachment insecurity	.018	−.050	−.293*	−.338**	.887**	.878**	–
8	Child avoidance^b	.056	−.219	.199	−.012	.045	.081	.071	–
9	Child anxiety^b	−.041	.114	.083	.144	−.184	−.314*	−.281*	−.067	–
10	Maternal rumination	.192	.170	−.297*	−.303*	.375**	.624**	.563**	.080	−.161	–

	Mean savor score	Savor word count	Rumination
Step 1 R ²	0.06			0.03			0.10
Constant	4.12*	1.25	[1.62, 6.63]	243.52*	85.64*	[72.04, 415.00]	0.11	0.62	[−1.14, 1.35]
Child age	−0.06	0.09	[−0.25, 0.12]	0.02	6.31	[−12.62, 12.65]	0.08	0.05	[−0.01, 0.17]
Child sex	−0.17	0.28	[−0.73, 0.38]	−25.43	19.07	[−63.61, 12.76]	−0.11	0.14	[−0.39, 0.17]
Parent age	0.004	0.02	[−0.04, 0.04]	−1.27	1.37	[−4.01, 1.47]	0.02	0.01	[−0.01, 0.04]
CAI avoidance^a	0.22	0.12	[−0.02, 0.47]	0.99	8.26	[−15.56, 17.53]	0.06	0.06	[−0.06, 0.18]
Step 2 ΔR ²	0.06*			0.08*			0.12*
Attachment avoidance	−0.22*	0.11	[−0.43, 0.00]	−16.81*	7.40	[−31.62, −1.99]	0.16*	0.05	[0.05, 0.27]

	Mean savor score	Savor word count	Rumination
Step 1 R ²	0.02			0.05			0.11
Constant	4.40*	1.28	[1.85, 6.96]	255.02*	84.49	[85.84, 424.20]	−0.34	0.50	[−1.34, 0.661]
Child age	−0.08	0.09	[−0.27, 0.11]	−1.18	6.19	[−13.58, 11.22]	0.10*	0.04	[0.02, 0.17]
Child sex	0.04	0.28	[−0.52, 0.59]	−16.64	18.36	[−53.41, 20.13]	−0.20	0.11	[−0.41, 0.02]
Parent age	−0.003	0.02	[−0.04, 0.04]	−1.46	1.33	[−4.12, 1.21]	0.02	0.01	[−0.001, 0.03]
CAI anxiety^a	−0.01	0.14	[−0.29, 0.26]	4.40	9.10	[−13.82, 22.61]	0.02	0.05	[−0.09, 0.13]
Step 2 ΔR ²	0.08*			0.08*			0.40*
Attachment anxiety	−0.26*	0.12	[−0.50, −0.02]	−19.01*	7.87	[−34.78, −3.25]	0.31*	0.05	[0.22, 0.41]

	Mean savor score	Savor word count	Rumination
Step 1 R ²	0.07			0.05			0.13
Constant	4.40*	1.25	[1.89, 6.91]	262.52*	85.24	[91.75, 433.28]	−0.29	0.55	[−1.40, 0.82]
Child age	−0.07	0.09	[−.25, .11]	−0.34	6.20	[−12.76, 12.08]	0.08*	0.04	[0.002, 0.16]
Child sex	−0.13	0.27	[−.68, .42]	−22.46	18.62	[−59.76, 14.83]	−0.13	0.12	[−0.38, 0.11]
Parent age	−0.003	0.02	[−.04, .04]	−1.42	1.35	[−4.14, 1.29]	0.02	0.01	[−0.001, 0.03]
CAI avoidance^a	0.23	0.12	[−.01, .47	1.47	8.13	[−14.82, 17.76]	0.05	0.05	[−0.06, 0.15]
CAI anxiety^b	0.01	0.13	[−.26, .27]	5.03	9.01	[−13.02, 23.07]	−0.01	0.06	[−0.12, 0.11]
Step 2 ΔR ²	0.09*			0.10*			0.29*
Insecure attachment^c	−0.30*	0.13	[−.56, −.05]	−22.16*	8.65	[−39.50, −4.83]	0.29*	0.06	[0.18, 0.41]

	DV: Coder-rated savoring
Constant	4.18*	1.22	[1.74, 6.62]	4.28*	1.28	[1.72, 6.85]	4.27*	1.24	[1.78, 6.76]
Child age	−0.02	0.09	[−0.20, 0.16]	−0.05	0.10	[−0.25, 0.15]	−0.03	0.09	[−0.22, 0.15]
Child sex	−0.23	0.27	[−0.77, 0.32]	−0.03	0.28	[−0.60, 0.54]	−0.18	0.27	[−0.73, 0.36]
Parent age	0.01	0.02	[−0.03, 0.05]	0.002	0.02	[−0.04, 0.04]	0.01	0.02	[−0.03, 0.05]
CAI avoidance^a	0.25	0.12	[0.02, 0.49]	–	–	–	0.25*	0.12	[0.01, 0.49]
CAI anxiety^b	–	–	–	−0.003	0.14	[−0.28, 0.27]	0.006	0.13	[−0.26, 0.26]
Rumination	−0.53^d	0.26	[−1.05, −0.01]	−0.35	0.34	[−1.03, 0.33]	−0.43	0.30	[−1.03, 0.17]
Avoidance	−0.13	0.11	[−0.36, .010]	–	–	–	–	–	–
Anxiety	–	–	–	−0.15	0.16	[−0.47, 0.17]	–	–	–
Insecurity^c	–	–	–	–	–	–	−0.18	0.15	[−0.49, 0.13]

	DV: Savor word count
Constant	246.86*	84.08	[78.44, 415.28]	248.03*	84.94	[77.87, 418.20]	256.25*	85.39	[85.12, 427.38]
Child age	2.53	6.35	[−10.20, 15.25]	.78	6.56	[−12.35, 13.92]	1.43	6.42	[−11.45, 14.30]
Child sex	−28.83	18.82	[−66.52, 8.86]	−20.70	18.91	[−58.58, 17.19]	−25.31	18.80	[−62.99, 12.37]
Parent age	−0.79	1.37	[−3.54, 1.95]	−1.15	1.38	[−3.90, 1.60]	−1.08	1.39	[−3.87, 1.71]
CAI avoidance^a	2.77	8.17	[−13.60, 19.14]	–	–	–	2.50	8.19	[−13.91, 18.91]
CAI anxiety^b	–	–	–	4.86	9.12	[−13.41, 23.13]	4.92	9.00	[−13.12, 22.96]
Rumination	−31.82^d	17.88	[−67.65, 4.00]	−20.64	22.45	[−65.62, 24.34]	−21.40	20.56	[−62.61, 19.81]
Avoidance	−11.68	7.81	[−27.33, 3.97]	–	–	–	–	–	–
Anxiety	–	–	–	−12.53	10.58	[−33.72, 8.65]	–	–	–
Insecurity^c	–	–	–	–	–	–	−15.91	10.53	[−37.01, 5.19]

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Ainsworth

M. D.

Bell

S. M.

(1970). Attachment, exploration, and separation: Illustrated by the behavior of one-year-olds in a strange situation. Child Development, 41, 49–67. doi:10.2307/1127388

Ben-Naim

Hirschberger

Ein-Dor

Mikulincer

(2013). An experimental study of emotion regulation during relationship conflict interactions: The moderating role of attachment orientations. Emotion, 13, 506–519. doi:10.1037/a0031473

Borelli

J. L.

McMakin

Sbarra

(2010). Mental reflection task manual: Pomona deployment study. Pomona College. Unpublished Manuscript.

Borelli

J. L.

Rasmussen

H. F.

Burkhart

M. L.

Sbarra

D. A.

(2014). Relational savoring in long-distance romantic relationships. Journal of Social and Personal Relationships, 32, 1083–1108. doi:10.1177/0265407514558960

Borelli

J. L.

Sbarra

D. A.

Snavely

J. E.

McMakin

D. L.

Coffey

J. K.

Ruiz

S. K.

… Chung

S. Y.

(2014). With or without you: Preliminary evidence that attachment avoidance predicts nondeployed spouses’ reactions to relationship challenges during deployment. Professional Psychology: Research and Practice, 45, 478–487. doi:10.1037/a0037780

Borelli

J. L.

Somers

West

J. L.

Coffey

J. K.

Reyes

Shmueli-Goetz

(2015). Associations between attachment narratives and self-report measures of attachment in middle childhood: Extending evidence for the validity of the child attachment interview. Journal of Child and Family Studies. doi:10.1007/s10826-015-0310-8

Borkovec

T. D.

Alcaine

Behar

(2004). Avoidance theory of worry and generalized anxiety disorder. In Heimberg

R. G.

Turk

C. L.

Mennin

D. S.

(Eds.), Generalized anxiety disorder: Advances in research and practice (pp. 77–108). New York, NY: Guilford Press.

Bowlby

(1980). Attachment and loss: Vol. 3. Loss. New York, NY: Basic Books.

Brennan

K. A.

Clark

C. L.

Shaver

P. R.

(1998). Self-report measurement of adult attachment: An integrative overview. In Simpson

J. A.

Rholes

&W. S.

(Eds.), Attachment theory and close relationships (pp. 46–76). New York, NY: Guilford.

10.

Bryant

(2003). Savoring Beliefs Inventory (SBI): A scale for measuring beliefs about savouring. Journal of Mental Health, 12, 175–196. doi:10.1080/0963823031000103489

11.

Bryant

F. B.

Smart

C. M.

King

S. P.

(2005). Using the past to enhance the present: Boosting happiness through positive reminiscence. Journal of Happiness Studies, 6, 227–260. doi:10.1007/s10902-005-3889-4

12.

Bryant

F. B.

Veroff

(2007). Savoring: A new model of positive experience. New Jersey, NJ: Lawrence Erlbaum Associates, Inc.

13.

Burkhart

Borelli

J. L.

Rasmussen

H. F.

Brody

Sbarra

D. A.

(2016, under review). If feelings could talk: Parental mentalizing explains the association between attachment anxiety and parenting satisfaction.

14.

Burkhart

Borelli

J. L.

Rasmussen

H. F.

Sbarra

D. A.

(2015). Cherish the good times: Relational savoring in parents of infants and toddlers. Personal Relationships, 22, 692–711. doi:10.1111/pere.12104

15.

Buter

L. D.

Nolen-Hoeksema

(1994). Gender differences in responses to depressed mood in a college sample. Sex Roles, 30, 331–346. doi:10.1007/BF01420597

16.

Caldwell

J. G.

Shaver

P. R.

(2012). Exploring the cognitive-emotional pathways between adult attachment and ego-resiliency. Individual Differences Research, 10, 141–152.

17.

Cassidy

Berlin

L. J.

(1994). The insecure/ambivalent pattern of attachment: Theory and research. Child Development, 65, 971–981. doi:10.2307/1131298

18.

Cohen

M. X.

Shaver

P. R.

(2004). Avoidant attachment and hemispheric lateralisation of the processing of attachment-and emotion-related words. Cognition and Emotion, 18, 799–813. doi:10.1080/02699930341000266

19.

Crowell

J. A.

Waters

(1996). Discriminant validity of the adult attachment interview. Child Development, 67, 2584–2599. doi:10.1111/1467-8624.ep9706060186

20.

Dozier

Kobak

R. R.

(1992). Psychophysiology in attachment interviews: Converging evidence for deactivating strategies. Child Development, 63, 1473–1480. doi:10.1111/1467-8624.ep9308195014

21.

Dreisbach

Goschke

(2004). How positive affect modulates cognitive control: Reduced preservation at the cost of increased distractibility. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30, 343–353. doi:10.1037/0278-7393.30.2.343

22.

Fraley

R. C.

Waller

N. G.

Brennan

K. A.

(2000). An item response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365. doi:10.1037//0022-3514.78.2.350

23.

Fredrickson

B. L.

(2001). The role of positive emotions in positive psychology: The broaden- and-build theory of positive emotions. American Psychologist, 56, 218–226. doi:10.1037/0003-066X.56.3.218

24.

Gray

J. R.

(2001). Emotional modulation of cognitive control: Approach withdrawal states double-dissociate spatial from verbal two-back task performance. Journal of Experimental Psychology: General, 130, 436–452. doi:10.1037/0096-3445.130.3.436

25.

Haft

W. L.

Slade

(1989). Affect attunement and maternal attachment: A pilot study. Infant Mental Health Journal, 10, 157–172. doi:10.1002/1097-0355(198923)10:3<157::AID-IMHJ2280100304>3.0.CO;2-3

26.

Hayes

A. F.

(2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. Retrieved from http://www.afhayes.com/public/process2012.pdf

27.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

28.

Karavasilis

Doyle

A. B.

Markiewicz

(2003). Associations between parenting style and attachment to mother in middle childhood and adolescence. International Journal of Behavioral Development, 27, 153–164. doi:10.1080/0165025024400015

29.

McMakin

D. L.

Siegle

G. J.

Shirk

S. R.

(2011). Positive Affect Stimulation and Sustainment (PASS) module for depressed mood: A preliminary investigation of treatment-related effects. Cognitive Therapy and Research, 35, 217–226. doi:10.1007/s10608-010-9311-5

30.

Meyer

P. S.

Johnson

D. P.

Parks

Iwanski

Penn

D. L.

(2012). Positive living: A pilot study of group positive psychotherapy for people with schizophrenia. The Journal of Positive Psychology, 7, 239–248. doi:10.1080/17439760.2012.677467

31.

Mikulincer

(1994). Human learned helplessness: A coping perspective. New York, NY: Plenum.

32.

Mikulincer

Birnbaum

Woddis

Nachmias

(2000). Stress and accessibility of proximity related thoughts: Exploring the normative and intraindividual components of attachment theory. Journal of Personality and Social Psychology, 78, 509–523. doi:10.1037/0022-3514.78.3.509

33.

Mikulincer

Dolev

Shaver

P. R.

(2004). Attachment-related strategies during thought suppression: Ironic rebounds and vulnerable self-representations. Journal of Personality and Social Psychology, 87, 940–956. doi:10.1037/0022-3514.87.6.940

34.

Mikulincer

Florian

(1995). Appraisal of and coping with a real-life stressful situation: The contribution of attachment styles. Personality and Social Psychology Bulletin, 21, 406–414. doi:10.1177/0146167295214011

35.

Mikulincer

Orbach

(1995). Attachment styles and repressive defensiveness: The accessibility and architecture of affective memories. Journal of Personality and Social Psychology, 68, 917–925. doi:10.1037/0022-3514.68.5.917

36.

Mikulincer

Orbach

Iavnieli

(1998). Adult attachment style and affect regulation: Strategic variations in subjective self–other similarity. Journal of Personality and Social Psychology, 75, 436–448. doi:10.1037/0022-3514.75.2.436

37.

Mikulincer

Shaver

P. R.

(2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 35, pp. 53–152). New York, NY: Academic Press.

38.

Mikulincer

Shaver

P. R.

(2007a). Attachment in adulthood: Structure, dynamics, and change . New York, NY: Guilford Press.

39.

Mikulincer

Shaver

P. R.

(2007b). Boosting attachment security to promote mental health, prosocial values, and inter-group tolerance. Psychological Inquiry, 18, 139–156. doi:10.1080/10478400701512646

40.

Mor

Winquist

(2002). Self-focused attention and negative affect: A meta-analysis. Psychological Bulletin, 128, 638–662. doi:10.1037/0033-2909.128.4.638

41.

Nolen-Hoeksema

(1991). Responses to depression and their effects on the duration of depressive episodes. Journal of Abnormal Psychology, 100, 569–582. doi:10.1037/0021-843X.100.4.569

42.

Nolen-Hoeksema

(2000). The role of rumination in depressive disorders and mixed anxiety/depressive symptoms. Journal of Abnormal Psychology, 109, 504–511. doi:10.1037/0021-843X.109.3.504

43.

Nolen-Hoeksema

(2012). Emotion regulation and psychopathology: The role of gender. Annual Review of Clinical Psychology, 8, 161–187. doi:10.1146/annurev-clinpsy-032511-143109

44.

Nolen-Hoeksema

Morrow

(1991). A prospective study of depression and posttraumatic stress symptoms after a natural disaster: The 1989 Loma Prieta earthquake. Journal of Personality and Social Psychology, 61, 115–121. doi:10.1037/0022-3514.61.1.115

45.

Nolen-Hoeksema

Wisco

B. E.

Lyubomirsky

(2008). Rethinking rumination. Perspectives on Psychological Science, 3, 400–424. doi:10.1111/j.1745-6924.2008.00088.x

46.

Nygren

Carstensen

Ludvigsson

Frostell

A. S.

(2012). Adult attachment and parenting stress among parents of toddlers. Journal of Reproductive and Infant Psychology, 30, 289–302. doi:10.1080/02646838.2012.717264

47.

Parker

M. L.

Johnson

L. N.

Ketring

S. A.

(2011). Assessing attachment of couples in therapy: A factor analysis of the experiences in close relationships scale. Contemporary Family Therapy, 33, 37–48. doi:10.1007/s10591-011-9142-x

48.

Pennebaker

J. W.

Francis

M. E.

Booth

R. J.

(2001). Linguistic Inquiry and Word Count (LIWC): LIWC 2001 . Mahwah, NJ: Erlbaum.

49.

Pennebaker

J. W.

Mehl

M. R.

Niederhoffer

K. G.

(2003). Psychological aspects of natural language. use: our words, our selves. Annual Review of Psychology, 54, 547–77. doi:10.1146/annurev.psych.54.101601.145041

50.

Quoidbach

Wood

A. M.

Hansenne

(2009). Back to the future: The effect of daily practice of mental time travel into the future on happiness and anxiety. The Journal of Positive Psychology, 4, 349–355. doi:10.1080/17439760902992365

51.

Reynolds

Searight

H. R.

Ratwik

(2014). Adult attachment style and rumination in the context of intimate relationships. North American Journal of Psychology, 16, 495–506.

52.

Rholes

W. S.

Simpson

J. A.

Blakely

B. S.

(1995). Adult attachment styles and mothers’ relationships with their young children. Personal Relationships, 2, 35–54. doi:10.1111/j.1475-6811.1995.tb00076.x

53.

Roisman

G. I.

Fraley

R. C.

Belsky

(2007). A taxometric study of the Adult Attachment Interview. Developmental Psychology, 43, 675–686. doi:10.1037/0012-1649.43.3.675

54.

Saffrey

Ehrenberg

(2007). When thinking hurts: Attachment, rumination, and postrelationship adjustment. Personal Relationships, 14, 351–368. doi:10.1111/j.1475-6811.2007.00160.x

55.

Schueller

S. M.

(2010). Preferences for positive psychology exercises. The Journal of Positive Psychology, 5, 192–203. doi:10.1080/17439761003790948

56.

Searle

Meara

N. M.

(1999). Affective dimensions of attachment styles: Exploring self-reported attachment style, gender, and emotional experience among college students. Journal of Counseling Psychology, 46, 147–158. doi:10.1037/0022-0167.46.2.147

57.

Selcuk

Guaydin

Sumer

Harma

Salman

Hazan

… Ozturk

(2010). Self-reported romantic attachment style predicts everyday maternal caregiving behavior at home. Journal of Research in Personality, 44, 544–549. doi:10.1016/j.jrp.2010.05.007

58.

Shmueli-Goetz

Target

Fonagy

Datta

(2008). The Child Attachment Interview: A psychometric study of reliability and discriminant validity. Developmental Psychology, 44, 939–995. doi:10.1037/0012-1649.44.4.939

59.

Slade

Belsky

Aber

Phelps

J. L.

(1999). Mothers’ representations of their relationships with their toddlers: Links to adult attachment and observed mothering. Developmental Psychology, 35, 611–619. doi:10.1037/0012-1649.35.3.611

60.

Sroufe

L. A.

(2005). Attachment and development: A prospective, longitudinal study from birth to adulthood. Attachment & Human Development, 7, 349–367. doi:10.1080/14616730500365928

61.

Storbeck

(2012). Performance costs when emotion tunes inappropriate cognitive abilities: Implications for mental resources and behavior. Journal of Experimental Psychology: General, 141, 411–416. doi:10.1037/a0026322

62.

Target

Fonagy

Shmueli-Goetz

Datta

Schneider

(1999). The child attachment interview (CAI) protocol. Unpublished manuscript, University College London.

63.

Tidwell

M. O.

Reis

H. T.

Shaver

P. R.

(1996). Attachment, attractiveness and social-interaction: A diary study. Journal of Personality and Social Psychology, 71, 729–745. doi:10.1037/0022-3514.71.4.729

64.

Zachrisson

H. D.

Røysamb

Oppedal

Hauser

(2011). Factor structure of the child Child Attachment Interview. European Journal of Developmental Psychology, 8, 744–759. doi:10.1080/17405629.2011.631293

	Total	Males	Females
Measures	N = 63	n = 24	n = 39
Maternal attachment avoidance^a	2.65 (1.16)	2.82 (1.54)	2.54 (0.85)
Maternal attachment anxiety^a	2.71 (1.11)	2.60 (0.97)	2.78 (1.20)
Maternal attachment insecurity^a	2.68 (1.00)	2.71 (1.17)	2.67 (0.90)
Maternal rumination scores	1.71 (0.53)	1.78 (0.53)	1.68 (0.52)
Coder-rated maternal savoring	2.84 (0.99)	2.89 (1.17)	2.81 (0.88)
Maternal savoring word count	109.67 (67.47)	119.92 (81.24)	103.36 (57.66)
Child attachment anxiety^b	0.18 (0.97)	0.09 (1.04)	0.13 (.94)
Child attachment avoidance^b	0.06 (1.07)	−0.24 (1.01)	0.24 (1.07)

	Mean savor score			Savor word count			Rumination
	b	SE	95% CI	b	SE	95% CI	b	SE	95% CI
Step 1 R ²	0.06			0.03			0.10
Constant	4.12*	1.25	[1.62, 6.63]	243.52*	85.64*	[72.04, 415.00]	0.11	0.62	[−1.14, 1.35]
Child age	−0.06	0.09	[−0.25, 0.12]	0.02	6.31	[−12.62, 12.65]	0.08	0.05	[−0.01, 0.17]
Child sex	−0.17	0.28	[−0.73, 0.38]	−25.43	19.07	[−63.61, 12.76]	−0.11	0.14	[−0.39, 0.17]
Parent age	0.004	0.02	[−0.04, 0.04]	−1.27	1.37	[−4.01, 1.47]	0.02	0.01	[−0.01, 0.04]
CAI avoidance^a	0.22	0.12	[−0.02, 0.47]	0.99	8.26	[−15.56, 17.53]	0.06	0.06	[−0.06, 0.18]
Step 2 ΔR ²	0.06*			0.08*			0.12*
Attachment avoidance	−0.22*	0.11	[−0.43, 0.00]	−16.81*	7.40	[−31.62, −1.99]	0.16*	0.05	[0.05, 0.27]

	DV: Coder-rated savoring
	Avoidance			Anxiety			Insecurity
	b	SE	95% C.I.	b	SE	95% C.I.	b	SE	95% C.I.
Constant	4.18*	1.22	[1.74, 6.62]	4.28*	1.28	[1.72, 6.85]	4.27*	1.24	[1.78, 6.76]
Child age	−0.02	0.09	[−0.20, 0.16]	−0.05	0.10	[−0.25, 0.15]	−0.03	0.09	[−0.22, 0.15]
Child sex	−0.23	0.27	[−0.77, 0.32]	−0.03	0.28	[−0.60, 0.54]	−0.18	0.27	[−0.73, 0.36]
Parent age	0.01	0.02	[−0.03, 0.05]	0.002	0.02	[−0.04, 0.04]	0.01	0.02	[−0.03, 0.05]
CAI avoidance^a	0.25	0.12	[0.02, 0.49]	–	–	–	0.25*	0.12	[0.01, 0.49]
CAI anxiety^b	–	–	–	−0.003	0.14	[−0.28, 0.27]	0.006	0.13	[−0.26, 0.26]
Rumination	−0.53^d	0.26	[−1.05, −0.01]	−0.35	0.34	[−1.03, 0.33]	−0.43	0.30	[−1.03, 0.17]
Avoidance	−0.13	0.11	[−0.36, .010]	–	–	–	–	–	–
Anxiety	–	–	–	−0.15	0.16	[−0.47, 0.17]	–	–	–
Insecurity^c	–	–	–	–	–	–	−0.18	0.15	[−0.49, 0.13]