Emotion Regulation and Relationship Satisfaction in Couple Therapy: What Happens Across Sessions?

ABSTRACT

While previous studies have demonstrated significant associations between partners' emotion regulation (ER) and overall well‐being, very few studies provide knowledge on ER in clinical couples, especially across couple therapy sessions. The sample consisted of 174 mixed‐sex couples, predominantly white and from the western region of the United States, who attended at least the initial eight sessions of couple therapy. The current study examined the intra‐ and inter‐personal effects of ER difficulties (i.e., lack of awareness, nonacceptance, limited strategies, and impulsivity) on the trajectory of relationship satisfaction over the initial eight sessions of couple therapy. The results provide a few important findings. First, gender differences exist in ER and relationship satisfaction. Second, across the initial eight couple therapy sessions, relationship satisfaction improves, and partners change at a similar rate (0.97 for women and 0.83 for men per session on a 100‐point scale; Wald test indicated no difference between these two rates of change). Third, ER difficulties have intra‐ and inter‐personal effects on the starting scores and rates of change in relationship satisfaction in couple therapy. It is worth noting that we found several counterintuitive positive associations between ER difficulties and relationship satisfaction, including lack of emotional awareness, nonacceptance of negative emotions, and limited regulation strategies. For clinical implications, we discussed that clinicians working with couples may find it beneficial to utilize modalities with a specific emphasis on the ER process, such as Greenberg's emotion‐focused couple therapy. The need for future studies with more diverse samples was recommended.

1. Introduction

In adulthood, romantic relationships may be the most intimate and challenging context that shapes many domains of people's lives. While challenging, the quality of couple relationships is irreplaceable in promoting personal and relational well‐being (Lebow and Snyder 2022). Meanwhile, recent findings show that emotion regulation (ER) might be an important target for promoting couple relationship functioning and satisfaction (Levenson et al. 2014; Snyder et al. 2005; Xu et al. 2023). Considering the significant role of couple relationships in people's overall well‐being, this study explores the links between partners' ER difficulties and the trajectory of relationship satisfaction across the initial eight sessions of couple therapy. Before presenting the design of our study, we first review key aspects of ER in couples, including its transdiagnostic role in treatment, core definitions and dyadic features, and implications for couple relationships.

1.1. Emotion Regulation: A Key Direction for Couple Therapy

Over the last 40 years, a substantial body of studies supports the efficacy and effectiveness of couple therapy in treating distressed couples, improving relationship satisfaction, promoting individual psychological and physical health, and advancing social functioning (Doss et al. 2022; Lebow and Snyder 2022; Snyder and Balderrama‐Durbin 2020). However, about 25% to 30% of couples do not benefit from therapy (Snyder et al. 2005), and this percentage is possibly higher in real‐world therapy settings. In recent years, researchers have proposed emotion regulation (ER) as an important target and direction across therapy modalities and presenting issues (Greenberg 2016; Johnson 2019; Levenson et al. 2014).

From a lifespan developmental perspective, Haase (2023) highlights that ER in couples becomes increasingly significant and consequential with age, underscoring its centrality for relationship quality and overall well‐being. Further, Levenson et al. (2014) pointed out, couples seeking therapy almost always experience ER issues in their relationship, regardless of the primary complaints or symptoms. Concurrently, Snyder et al. (2005) suggested that the presenting problems of distressed couples may vary greatly; yet, the intra‐ and inter‐personal difficulties in ER might be the crucial factor in developing and maintaining distress at both the relational and individual levels. Likewise, Snyder and Balderrama‐Durbin (2020), from a common factors perspective, suggested that different couple‐based treatments may produce comparable results by therapeutically impacting the ER processes. Moreover, scholars and clinicians in the broader mental health field have agreed that ER occupies a central role and is transdiagnostic across clinical issues and theoretical orientations (Greenberg 2016; Johnson 2019; Sloan et al. 2017). Therefore, ER is an important construct for further study with couples, especially couples seeking therapy.

1.2. Emotion Regulation: Definitions & Dimensions

Emotion regulation (ER) refers to the unfolding process during which an individual modulates what, when, and how emotions manifest (Gross 1998). Gratz and Roemer (2004) further suggested an integrative framework that specifies six dimensions of ER: (a) emotional awareness (awareness), (b) emotional clarity (clarity), (c) acceptance of emotions (acceptance), (d) regulation strategies (strategies), (e) goal‐directed behaviors (goal), and (f) controlling impulsive behaviors (impulsivity). According to prior studies, four ER dimensions (i.e., awareness, acceptance, strategies, and impulsivity) have significant associations with relationship satisfaction (Bloch et al. 2014; Rick et al. 2017; Xu et al. 2023). Thus, this study focuses on these four dimensions.

The awareness dimension examines the attention a person gives to their emotional states. Croyle and Waltz (2002) suggested that higher levels of emotional awareness are associated with lower relationship satisfaction for women in general social settings and romantic contexts. They further pointed out that discrepancies between partners' emotional awareness are negatively associated with relationship satisfaction for both women and men.

The acceptance dimension indicates that an individual is non‐judgmental towards their negative emotional states. Two cross‐sectional studies with clinical samples found that the higher partners' nonacceptance of negative emotions, the higher the reported relationship satisfaction for both partners (Rick et al. 2017; Xu et al. 2023). These findings are somewhat counterintuitive because prior studies with individuals have consistently suggested that higher nonacceptance is associated with adverse psychological outcomes (Wolgast et al. 2011). Regarding these inconsistent findings, Xu et al. (2023) suggested the possibility that when romantic partners engage in nonacceptance of their negative emotions at the intrapersonal level (i.e., internally within the partner), at the interpersonal level (i.e., externally in the relationship), they might choose to express fewer negative emotions and focus more on the positive aspects, which could promote relational engagement and meaningful conversations, and potentially higher relationship satisfaction for both partners. Nonetheless, these divergent findings underscore the need for further research. This also indicates that emotion regulation might operate quite differently in dyadic versus individual contexts.

The strategies dimension refers to an individual's beliefs and capacity in using effective regulation strategies. Researchers commonly focus on two strategies: reappraisal and suppression. A study with 111 couples in their first marriage and 108 couples in remarriages shows that while men's greater use of suppression worsens the negative association between conflict and relationship satisfaction, women's greater use of cognitive reappraisal reduces the strength of the negative link between conflict and relationship satisfaction (Frye et al. 2020).

The impulsivity dimension is about controlling impulsive behaviors amid emotional distress. Impulsive control difficulties are associated with lower relationship satisfaction (Bloch et al. 2014; Williams and Kuyken 2012), higher emotional arousal, risky and antisocial behaviors (Weiss et al. 2012), emotional and verbal aggression (Baker et al. 2018), as well as physical and psychological aggression in intimate relationships (Halmos et al. 2018).

1.3. Emotion Regulation in Couples: The Unique Features

Emotion regulation (ER) in couples is the dynamic process in which partners regulate emotions in the relationship, whether consciously or outside of awareness, with the aim of enhancing both relationship and individual well‐being (Bloch et al. 2014; Haase 2023; Levenson et al. 2014; Xu et al. 2023). In romantic relationships, ER is not simply summing both partners' regulatory activities but a dynamic process that requires partners to simultaneously perform multiple regulatory tasks: each partner needs to regulate their own emotions, attend to their partner's emotional state, and monitor the moment‐to‐moment quality of their relationship (Butler et al. 2014; Levenson et al. 2014). This multifaceted process creates an extremely complex, fluid, and dynamic emotional landscape that challenges partners' regulatory capacity and impacts their daily well‐being (Butler 2011).

1.3.1. Intra‐ & Inter‐Personal Levels

In couple emotion regulation, the resulting impact occurs at intra‐ and inter‐personal levels (Butler 2011; Horn and Maercker 2016; Levy‐Gigi and Shamay‐Tsoory 2017). A partner's regulatory activities impact both the self (i.e., intrapersonal) and the other partner (i.e., interpersonal). Thus, to bring positive outcomes into couple relationships, such as increasing relationship satisfaction, clinicians need to simultaneously target the intra‐ and inter‐personal levels of ER in couples (Ben‐Naim et al. 2013; Florean and Păsărelu 2019; Rick et al. 2017; Snyder et al. 2005). Meanwhile, during the process of ER in mixed‐sex couples, gender differences are worth noting.

1.3.2. Gender Differences

Prior research has pointed out the gender differences in ER. First, women typically serve as the “emotion regulator” in mixed‐sex romantic relationships, and they feel responsible for regulating negative emotions and maintaining a positive emotional environment. This imbalance of emotional work commonly exerts a significant toll on women's psychological health and impairs relational intimacy and closeness for both partners (Strazdins and Broom 2004). Thus, prior studies suggest that as the “emotion regulator” in relationships, women's emotion regulation might be particularly significant for relieving relational distress and enhancing relationship satisfaction for both partners (Bloch et al. 2014).

Second, across various socioeconomic and cultural backgrounds, women display higher emotional awareness than men (Ciarrochi et al. 2005). They also show more complex emotional knowledge and greater differentiation in articulating emotional experience in personal and social contexts (Barrett et al. 2000). However, these highly generalizable advantages in emotional awareness, clarity, and knowledge do not necessarily enable women to be more competent during emotional distress. Commonly, women use more internalized strategies (e.g., rumination; Brody et al. 2016) and verbal expression (e.g., negative talk to self or other; Horn and Maercker 2016) during negative emotional states. Meanwhile, men are socialized to adopt more externalized and behavioral coping strategies (e.g., avoidance, and distraction by work; Ciarrochi et al. 2005). Consequently, women tend to experience intense sadness and guilt during emotional distress, but men show “no emotion” by disengaging or tend to display anger (Barrett et al. 2000; Brody et al. 2016).

In short, emotion regulation in couples is a complex and dynamic process that includes both partners' regulatory activities, changes at intra‐ and inter‐personal levels, and gender differences. When working with couples, both researchers and clinicians need to intentionally incorporate these multiple aspects of emotion regulation.

1.4. Emotion Regulation and Couple Relationships

As Siegel said, “An emotion links various systems together to form a state of mind. It also serves to connect one mind to another” (Siegel and Bryson 2011, 392). Thus, it is impossible to develop an effective treatment without targeting the emotion regulation (ER) process. Prior studies show various intra‐ and inter‐personal consequences of ER difficulties in couples. These include depression and anxiety (Dubé et al. 2019; Levy‐Gigi and Shamay‐Tsoory 2017), adjustment disorder symptoms (Horn and Maercker 2016), lower interpersonal empathy (Florean and Păsărelu 2019), perception of hostile criticisms (Klein et al. 2016), physical and psychological aggression (Halmos et al. 2018), and nonproductive affect and conflict cycles (Ben‐Naim et al. 2013). Moreover, direct associations were found between partners' ER difficulties and relationship satisfaction.

Bloch et al. (2014) conducted a 13‐year 3‐wave study with 82 couples. Their results indicate intra‐ and inter‐personal effects of partners' ER concurrently and across time. Specifically, when women display less negative emotional experiences and behaviors, both partners report greater relationship satisfaction at Wave‐1. Longitudinally, both women's and men's reduced negative emotional behaviors predict positive changes in women's relationship satisfaction. However, these findings are from a laboratory study with middle‐aged (i.e., over 40 years old) married couples, so their generalizability to couple therapy might be limited.

Furthermore, the study by Rick et al. (2017) with 107 couples finds intra‐ and inter‐personal effects of partners' ER dimensions. Their study indicates positive intrapersonal effects of regulation strategies, negative intrapersonal effects of women's acceptance, negative intrapersonal effects of men's emotional awareness, and negative interpersonal effects of women's acceptance. Xu et al. (2023) replicated the study with a clinical sample of 275 couples. They also found the intrapersonal effects of regulation strategies and women's nonacceptance. Additionally, they reported negative intrapersonal effects of women's lack of awareness and positive intrapersonal effects of men's nonacceptance. This study also suggests the negative interpersonal effects of women's impulsivity and limited regulation strategies and the negative interpersonal effects of men's impulsivity.

Additionally, a study employed multilevel modeling to investigate whether baseline ER difficulties were associated with the change trajectory of relationship satisfaction in couple therapy (White VanBoxel et al. 2024). The study compared couples receiving emotionally focused therapy with those receiving treatment as usual. Their findings suggested that higher baseline ER difficulties were predictive of lower baseline relationship satisfaction, but not of changes in relationship satisfaction. They also found that the baseline and rate of change in relationship satisfaction did not differ between women and men. However, the study had only 28 couples in its final analysis, with 42.5% missing data at the end of therapy. Moreover, they relied on the total score of the ER scale rather than subscale scores, potentially due to limited sample size. Without a power analysis, the findings and their implications should be interpreted with caution, though they nevertheless highlighted the need for further research.

Taken together, these findings point out that ER difficulties play significant roles in couple functioning at both intra‐ and inter‐personal levels. For researchers and clinicians, this underscores the need to examine not only individual ER activities but also how regulatory processes unfold and are reshaped within relationships. Specifically, rather than translating results from individual‐focused studies, advancing our understanding of ER in couples requires dyadic analyzes. For example, in the current study, by using dyadic latent growth curve modeling, we could simultaneously examine both partners' regulatory processes while accounting for the interdependent nature of constructs in couple relationships.

In conclusion, ER in couples has gained increased attention, and previous research demonstrates the significant intra‐ and inter‐personal effects of partners' ER (Bloch et al. 2014; Rick et al. 2017; Xu et al. 2023). Yet, considering the inconsistency in findings and the cross‐sectional nature of the studies with clinical samples, we need more studies on understanding ER in the process of couple therapy.

2. Current Study

In this study, we aimed to examine the intra‐ and inter‐personal effects of emotion regulation difficulties (i.e., lack of awareness, nonacceptance of negative emotions, limited regulation strategies, and impulsivity) on the trajectory of relationship satisfaction across the initial eight couple therapy sessions. Our hypotheses were:

Hypothesis 1

Gender differences exist in emotion regulation and relationship satisfaction.

Hypothesis 2

Relationship satisfaction improves over the initial eight couple therapy sessions.

Hypothesis 3

A partner's emotion regulation difficulties are associated with their own (i.e., intrapersonal/actor effects) and their partner's (i.e., interpersonal/partner effects) starting scores and rates of change in relationship satisfaction across the initial eight couple therapy sessions.

Like the study by Xu et al. (2023), the present study draws on data from the same pilot project, the MFT‐PRN (Johnson et al. 2017; www.mft‐prn.net), and uses the DERS and CRS measures. Xu et al. (2023) examined cross‐sectional associations between emotion regulation (ER) difficulties and couple relationship satisfaction (CRS) at baseline, including gender differences. The current study is intentionally designed as both a conceptual replication and a longitudinal extension of that work. Specifically, Hypothesis 1 replicates prior findings on gender differences within a longitudinal sample of couples in therapy; meanwhile, Hypotheses 2 and 3, using dyadic growth curve modeling, extend the literature by examining how baseline ER difficulties predict intra‐ and inter‐personal change trajectories in CRS over the initial eight sessions of couple therapy. This distinction allows the present study to move beyond cross‐sectional associations and contribute novel insight into the dynamic processes through which ER difficulties shape couple therapy outcomes over time.

3. Method

3.1. Participants

The sample consisted of 174 mixed‐sex couples, predominantly from the western region of the United States, who attended at least the initial eight sessions of couple therapy. Data were pooled from the participating clinics in the Marriage and Family Therapy Practice Research Network (MFT‐PRN; Johnson et al. 2017; www.mft‐prn.net). Data were not available on which therapeutic approach was used (i.e., TAU‐treatment as usual vs. a manualized approach). The study was approved by the institutional review board at Brigham Young University.

The average age was 29.53 years (SD = 10.29) for women and 30.40 years (SD = 10) for men. Most women (78.74%, n = 137) and men (83.33%, n = 145) were White. For education, 72.99% (n = 127) of women and 63.22% (n = 110) of men had an associate's degree or higher. 58.05% (n = 101) of women and 52.87% (n = 92) of men reported less than $39,999 in household income. The average relationship length was 6.24 years (SD = 9.01), and 40.23% (n = 70) of the couples reported having one or more children, with a median of 2.

3.2. Measures and Analyzes

3.2.1. Couple Relationship Satisfaction

The Couple Relationship Scale evaluates emotional intimacy, commitment, trust, safety, cohesion, acceptance, conflict, physical intimacy, overall happiness, and personal well‐being in romantic relationships (Anderson et al. 2021). Item responses range on a 0 to 100 analog scale, and higher scores represent greater relationship satisfaction. The scale has a clinical cutoff of 70.9. The item responses are anchored on each end. For example, “I feel all alone (0) – like a team (100) in my relationship with my partner.” In this study, average scores were used, and Cronbach's alpha ranged from 0.93 to 0.97.

3.2.2. Emotion Regulation Difficulties

The Difficulties in Emotion Regulation Scale Short form is an 18‐item scale (Kaufman et al. 2016). It examines clinical problems in six emotion regulation dimensions, including lack of emotional awareness, nonacceptance of negative emotions, limited regulation strategies, and difficulties in controlling impulsivity. The responses range on a 5‐point Likert‐type scale from 1 (almost never = 0%–10% of the time) to 5 (almost always = 91%–100% of the time). Higher scores represent greater emotion regulation difficulties. Items include “I pay attention to how I feel [reverse code]” for the lack of awareness subscale; “When I'm upset, I become embarrassed for feeling that way” for the nonacceptance subscale; “When I'm upset, it takes me a long time to feel better” for the limited strategies subscale; and “When I'm upset, I become out of control” for the impulsivity subscale. In this study, average scores of subscale items were used, and Cronbach's alpha ranged from 0.72 to 0.89.

3.2.3. Missingness

For relationship satisfaction, missingness ranged from 4.6% (n = 8) to 18.39% (n = 32). Potential bias was addressed using the maximum likelihood estimator in Mplus Version 8.7 (Muthén and Muthén 2017).

3.2.4. Dyadic Growth Curve Models

To account for interdependence, the dyadic version of growth curve modeling was utilized (Whittaker et al. 2014). All assessments were finished before the corresponding session. Thus, in our growth curve models, we set the factor loadings at sessions #1–#8 as 0–7. We also applied 5000 bootstrap iterations to all model estimations (see Figure S1 for the conceptual conditional model).

We conducted the model fitting in two steps. In step one, we compared models with and without quadratic effects of the slopes. The quadratic effects were not significant, and the delta χ ² test result did not support the inclusion of non‐linear slopes (χ ² (2) = 0.22, p = 0.639). In step two, considering prior studies (Rick et al. 2017; Xu et al. 2023) included clarity and goal dimensions of emotion regulation, we compared models without and with these two dimensions. The model did not fit better when adding clarity (χ ² (8) = 10.40, p = 0.238) or goal (χ ² (8) = 12.95, p = 0.114), and no significant path was associated with these two dimensions. Thus, our conditional model included four other emotion regulation dimensions as predictors, and relationship length as a control variable (Jose and Alfons 2007).

3.2.5. Power Analysis

For the final dyadic latent growth model, we conducted a Monte Carlo simulation power analysis with 1000 replications, using Mplus Version 8.7 (Muthén and Muthén 2017). Results indicated that the study was well powered to detect actor effects on the latent intercepts (i.e., baseline relationship satisfaction), with the exception of the lack of awareness dimension. Power for path effects on the slopes (i.e., average rates of change) was moderate for a few predictors but generally insufficient, and partner effects were largely underpowered. Taken together, these results suggest that significant effects on the intercepts can be interpreted with confidence, whereas nonsignificant slope and partner effects should be interpreted cautiously. Detailed, parameter‐specific results from the power analysis are available from the corresponding author upon request.

4. Results

4.1. Preliminary Findings

Relationship satisfaction at session #8 was significantly higher than session #1 for both women, t(160) = 6.04, p < 0.001, and men, t(153) = 5.14, p < 0.001 (see Table S1 for additional preliminary statistics).

4.2. H1: Gender Differences

Compared to men, women at session #1 reported less difficulties in awareness, t(174) = −5.87, p < 0.001, but had more nonacceptance, t(174) = 3.08, p = 0.001, limited strategies, t(174) = 3.63, p < 0.001, and higher impulsivity, t(174) = 1.95, p = 0.026. Further, women had lower relationship satisfaction at session #1, t(174) = −3.66, p < 0.001, but not at session #8 than men.

4.3. H2: Trajectory of Couple Relationship Satisfaction Across Eight Sessions

The unconditional growth curve model provided a good fit to the data (χ ² (114), N = 174) = 169.81, p < 0.001; RMSEA = 0.053; CFI = 0.977; TLI = 0.976; SRMR = 0.072.

The intercept of relationship satisfaction was 70.73 (p < 0.001) for women and 73.83 (p < 0.001) for men. We first compared the intercepts with the clinical cutoff value (i.e., 70.9) of relationship satisfaction. The results show that women's baseline relationship satisfaction was not significantly lower than the cutoff value (p = 0.90), whereas men's baseline relationship satisfaction was significantly higher than the cutoff (p = 0.012). In other words, on average, couples in our sample were not clinically distressed at the outset of therapy. We further conducted the Wald test of parameter constraints to compare them. The results indicate that women and men start couple therapy with different levels of relationship satisfaction, Wald χ ² (1) = 8.01, p = 0.005. Also, the variances of both intercepts were significant (women's variance = 267.29, p < 0.001; men's variance = 201.36, p < 0.001), indicating a significant amount of individual variability in the starting scores of relationship satisfaction within each group.

The slope of relationship satisfaction was 0.97 (p < 0.001) for women and 0.83 (p < 0.001) for men, meaning women increased by 0.97 and men by 0.83 in relationship satisfaction per session. The Wald test for comparing slopes was not significant, Wald χ ² (1) = 0.64, p = 0.426, revealing that women and men might change at a similar rate in the initial eight couple therapy sessions. However, the variance of slopes was significant for both groups (women's variance = 2.49, p < 0.001; men's variance = 0.96, p = 0.010).

Regarding associations between women's and men's growth factors (i.e., intercepts and slopes), the only significant correlation (r = 0.59, p < 0.001) was between the two intercepts (i.e., women's and men's starting scores of relationship satisfaction). This suggests that women's and men's relationship satisfaction are positively related to each other at the first session. Partners' rates of change in relationship satisfaction were not associated with each other (p = 0.921). This indicates that partners might not necessarily experience parallel trajectories of improvement in relationship satisfaction during the initial eight sessions of couple therapy.

Considering the significant variations in intercepts and slopes of relationship satisfaction, and prior findings on the role of emotion regulation, a conditional model was needed to further explore the trajectory of relationship satisfaction across the initial eight sessions of couple therapy. Based on the nonsignificant Wald test results for the slopes, we constrained women's and men's slopes of relationship satisfaction to be equal in the conditional model.

4.4. H3: Intra‐ & Inter‐Personal Effects of Emotion Regulation Difficulties

The conditional model provided a good fit to the data (χ ² (223), N = 174) = 320.09, p < 0.001; RMSEA = 0.05; CFI = 0.962; TLI = 0.954; SRMR = 0.060. The variance that this model accounted for was 35.5% in women's intercept, 19.3% in women's slope, 36.4% in men's intercept, and 24.2% in men's slope. The control variable, relationship length, was associated with women's intercept (β = −0.25; p < 0.001) and slope (β = 0.23; p = 0.021) (see Table 1 for detailed results).

TABLE 1.

Conditional GCM with predictors and a control variable.

Variables	Women's intercept			Women's slope
	β	SE	p	β	SE	p
Actor Awareness	−0.05	0.07	0.460	−0.06	0.10	0.567
Partner Awareness	−0.07	0.07	0.317	−0.004	0.11	0.973
Actor Nonacceptance	0.21*	0.10	0.026	−0.03	0.13	0.810
Partner Nonacceptance	0.03	0.08	0.701	0.14	0.14	0.306
Actor Limited Strategies	−0.30**	0.10	0.003	0.30**	0.14	0.032
Partner Limited Strategies	0.11	0.10	0.277	−0.23	0.14	0.093
Actor Impulsivity	−0.28**	0.09	0.002	0.11	0.16	0.464
Partner Impulsivity	−0.21*	0.09	0.022	0.03	0.15	0.858
Actor Awareness	−0.11	0.08	0.164	0.32*	0.14	0.018
Partner Awareness	−0.08	0.08	0.265	0.04	0.13	0.742
Actor Nonacceptance	0.19*	0.08	0.015	−0.08	0.16	0.609
Partner Nonacceptance	0.33***	0.09	0.000	−0.26	0.19	0.166
Actor Limited Strategies	−0.29**	0.10	0.002	−0.01	0.18	0.944
Partner Limited Strategies	−0.21*	0.10	0.024	0.16	0.18	0.397
Actor Impulsivity	−0.10	0.09	0.248	−0.09	0.18	0.629
Partner Impulsivity	−0.30**	0.10	0.002	0.29	0.18	0.109
Actor Awareness	−0.11	0.08	0.164	0.32*	0.14	0.018

Note: 5000 bootstrap iterations were added to the analysis, and slopes of partners' relationship satisfaction were constrained to be equal. N = 174 couples. Results are standardized (β). GCM = dyadic latent growth curve modeling for relationship satisfaction and Control = Length of relationship.

^* p < 0.05.

^** p < 0.01.

^*** p < 0.001.

4.4.1. Lack of Awareness

Men's lack of emotional awareness at session #1 was linked to men's slope of relationship satisfaction (β = 0.32, p = 0.018). This indicates that when men report more difficulties in emotional awareness at session #1, there might be more improvement per session in their relationship satisfaction.

4.4.2. Nonacceptance of Negative Emotions

Women's nonacceptance at session #1 was predictive of women's intercept of relationship satisfaction (β = 0.21, p = 0.026). Also, women's (β = 0.33, p < 0.001) and men's (β = 0.19, p = 0.015) nonacceptance were associated with men's intercept of relationship satisfaction. This implies that the greater women's and men's nonacceptance of negative emotions at session #1, the higher men's starting score of relationship satisfaction.

4.4.3. Limited Strategies

Women's limited regulation strategies were associated with women's intercept (β = −0.30, p = 0.003), men's intercept (β = −0.21, p = 0.024), and women's slope (β = 0.30, p = 0.032) of relationship satisfaction. These imply, when women have more difficulties in regulation strategies, both partners might report lower relationship satisfaction at session #1, while women might experience faster change in relationship satisfaction across the initial eight sessions of couple therapy. Besides, men's difficulties in regulation strategies were linked to men's intercept of relationship satisfaction (β = −0.29, p = 0.002). A series of Wald tests was utilized to test these path effects. The strongest was the actor effect of women's limited strategies on women's slope of relationship satisfaction (β = 0.30).

4.4.4. Impulsivity

Women's impulsivity was associated with the intercepts of women's (β = −0.28, p = 0.002) and men's (β = −0.30, p = 0.002) relationship satisfaction. Men's impulsivity was linked with women's intercept of relationship satisfaction (β = −0.21, p = 0.022), suggesting the possibility that the more impulsive men at session #1, the lower women's beginning relationship satisfaction.

5. Discussion

Overall, the dyadic growth curve models provided some novel findings. First, we found gender differences in emotion regulation difficulties and couple relationship satisfaction. Second, relationship satisfaction improves in the initial eight sessions of couple therapy, and partners change at a similar rate. Third, emotion regulation difficulties have intra‐ and inter‐personal effects on the starting scores and rates of change in relationship satisfaction across the initial eight sessions of couple therapy. Meanwhile, some counterintuitive findings are worth particular attention, such as the positive association between nonacceptance of negative emotions and relationship satisfaction.

5.1. Preliminary Findings

From session #1 to #8, both partners reported an increase in relationship satisfaction, which again gave evidence for the value of couple therapy in reducing relational distress (Doss et al. 2022). In couple therapy, clinicians help partners engage in healthy interactional patterns, decrease emotional avoidance, increase positive emotions, and constructively process the dysregulated or heightened emotions, which could lead to higher relationship satisfaction for both partners (Johnson 2019).

5.2. H1: Gender Differences

Our study concurs with previous findings on gender differences in mixed‐sex couples. Compared to men, women show lower relationship satisfaction (Jackson et al. 2014) and greater emotional awareness (Ciarrochi et al. 2005). But women also report more difficulties in accepting negative emotions (Barrett et al. 2000), employing regulation strategies, and controlling impulsive behaviors (Xu et al. 2023). These gender differences in emotion regulation might be largely due to the socialization process. In most societies, women are expected to achieve an optimum emotional state for both the self and others, while men need to be less expressive and demonstrate fewer or no emotions to fit in the stereotypes (Brody et al. 2016). Thus, women commonly take on the “emotion regulator” role in their relationships as well as social contexts in general (Bloch et al. 2014). This imbalance of emotional work might not only contribute to more emotion regulation difficulties for women (Xu et al. 2023) but also significantly increase women's psychosocial impairments and health risks. This might partially explain the typical gender gap in couple relationship satisfaction across the literature (Strazdins and Broom 2004).

At the same time, our results suggest that couple therapy may attenuate some of these gender disparities. While women reported lower satisfaction at baseline, no gender differences were observed at session #8, indicating that therapy may help rebalance relational dynamics and promote more equitable relationship satisfaction across partners. This result further provides some support for the efficacy of couple therapy in promoting a mutually functioning and satisfying relationship (Doss et al. 2022).

Clinically, these results underscore the importance of therapists being attentive to gendered patterns of emotional labor in therapy. Women may carry disproportionate responsibility for regulating emotions, which can obscure men's role in relational distress. Therapy provides an opportunity to challenge these imbalances by encouraging both partners to take shared responsibility for emotion regulation processes. Such an approach may not only reduce the psychosocial burden placed on women but also foster healthier and more reciprocal relational patterns over the course of couple therapy.

5.3. H2: Trajectory of Couple Relationship Satisfaction Across Therapy Sessions

In this study, both partners' relationship satisfaction improves across the initial eight sessions of couple therapy, which again affirms that “couple therapy works” (Lebow and Snyder 2022, 4). Over the past 40 years, multiple meta‐analyzes and reviews attest to the effectiveness of couple therapy in treating distressed couples. While most prior studies focus on comparing individuals receiving couple therapy with those who have not (Lebow and Snyder 2022), this study with dyadic latent growth curve modeling provides another perspective. Our findings show, on average, relationship satisfaction increases by 0.97 for women and 0.83 for men per session on a 100‐point scale (please note, Wald test indicated no difference between these two rates of change). The couple relationship scale has a reliable change index of 16 (Anderson et al. 2021). Thus, it might take roughly 16–20 sessions for both partners to obtain clinically reliable improvements in their relationship satisfaction. Please note that this should not be interpreted as the actual trajectory of change in couple therapy. Rather, this is intended to be an illustration of the magnitude of change in terms of relationship satisfaction during the initial eight sessions of couple therapy.

However, the only association between partners' growth factors (i.e., the baseline and rates of change) was between the starting scores (r = 0.59, p < 0.001). This might suggest that women and men are similar or interdependent in terms of their perceived relationship satisfaction. The other non‐significant links between partners' growth factors potentially indicate that the starting scores of partners' relationship satisfaction have no associations with their average rates of change in relationship satisfaction. This finding is consistent with the study by Bartle‐Haring (2024), which suggests that couples receiving therapy change in different trajectories across therapy sessions, and there are no consistent associations between the starting scores and rates of change in relationship satisfaction.

Besides, it is particularly worth noting that while the Wald test results show women's and men's rates of change of relationship satisfaction are not statistically different from each other, their rates of change in therapy were not related. A potential explanation for this is that partners working with the same clinician might have shared resources both in therapy and outside of therapy in their day‐to‐day life, including the therapeutic alliance, relationship distress and dissolution, and financial resources, etc. These shared resources are likely to inform a similar rate of change in therapy for both partners. Yet, each partner's unique individual traits, such as gender, childhood trauma experiences, overall perceived distress, and commitment to therapy and change, might have a significant role in the change process. These individual traits could potentially explain the non‐significant association between partners' rates of change in relationship satisfaction. Still, we need more studies to further clarify the links between partners' starting scores and rates of change in relationship satisfaction over the course of couple therapy.

5.4. H3: Intra‐ & Inter‐Personal Effects of Emotion Regulation Difficulties

5.4.1. Lack of Awareness

Our study shows that the lower the men's initial emotional awareness, the faster they may change across the initial eight sessions of couple therapy (β = 0.32, p = 0.018). This finding is counterintuitive, as prior research generally links higher emotional awareness with positive outcomes such as greater motivation for emotional work, richer emotional knowledge, increased social participation, and more positive moods (Barrett et al. 2000; Ciarrochi et al. 2005; Cooper et al. 2020). In contrast, low awareness has been associated with relationship distress (Croyle and Waltz 2002), dissolution, and other clinical issues such as depression and eating disorders (Ciarrochi et al. 2005; Eckland et al. 2021; Racine and Wildes 2013).

For this counterintuitive finding, one explanation may be that men with initially lower awareness have more room to benefit from couple therapy. This aligns with a systematic review suggesting that higher baseline severity predicts greater symptom reduction in therapy (Scholten et al. 2023). Across therapy modalities, emotion is one of the key components for facilitating therapeutic change for both partners (Snyder et al. 2005). In couple therapy, clinicians guide both partners to express and experience their emotions, through which they might gain more awareness of their own and their partner's emotions (Greenberg 2016). Emotional awareness is the foundational step in emotion regulation. Men obtaining more emotional awareness in couple therapy are likely to feel motivated to engage with their partner, express their vulnerabilities, have higher clarity over emotions, and experience more positive emotions (Ciarrochi et al. 2005). As therapy progresses, men with lower starting awareness may experience the greatest overall improvements in both emotional engagement and relationship satisfaction. What's more, couple therapy might bridge the discrepancy between women's and men's emotional awareness (Barrett et al. 2000) through the shared emotional experience in therapy (Johnson 2019). The reduced discrepancy between partners' emotional awareness might further lead to more positive changes in relationship satisfaction (Croyle and Waltz 2002).

5.4.2. Nonacceptance of Negative Emotions

We found positive associations between nonacceptance and relationship satisfaction. Specifically, higher women's nonacceptance at baseline predicted higher women's initial relationship satisfaction (β = 0.21, p = 0.026). In addition, both partners' nonacceptance was associated with higher men's initial relationship satisfaction (women β = 0.33, p < 0.001; men β = 0.19, p = 0.015). These findings run contrary to the common assumption that nonacceptance is maladaptive. Meanwhile, they might be explained through protective buffering, a relationship‐focused coping skill for emotion regulation in interpersonal contexts. Protective buffering involves an individual's effort in avoiding or hiding negative emotions and accommodating others' emotions, to reduce distress or dissolution. When couples have different experiences and reactions, protective buffering is likely to decrease negative emotions within the regulator (i.e., the individual who uses the skill) and protect the other partner's feelings, which might potentially lead to higher perceived relationship satisfaction for both partners (Butler et al. 2014). This interpretation aligns with prior evidence that nonacceptance is associated with greater interpersonal avoidance in intimate relationships (Wolgast et al. 2011). Partners who struggle to accept their negative emotions may attempt to minimize or avoid conflict. For couples in therapy, who may already perceive their relationship as fragile, such tendencies may dampen conflict signals, reduce the acknowledgment of distress, and create a temporary sense of stability or peace that is experienced as satisfaction.

Similarly, a prior study suggests, when partners are more judgmental (e.g., feel ashamed) with their own negative emotions (i.e., at the intrapersonal level), between the partners (i.e., at the interpersonal level), there might be less negative emotional exchanges and more positive emotional engagement and constructive conversations (Xu et al. 2023). Particularly, women in general feel more responsible and are more competent than men in detecting and articulating emotions (Barrett et al. 2000). Therefore, in romantic relationships, when women feel bad (e.g., feel ashamed or guilty) about having negative emotions, they might show fewer negative emotions, and men is likely to experience less spillover of negative emotions (Butler 2011; Cooper et al. 2020), and become more comfortable in avoiding their own negative emotions (Ciarrochi et al. 2005; Strazdins and Broom 2004). Potentially, men might thus encounter fewer difficulties in emotion regulation and are likely to experience higher relationship satisfaction. But this does not imply that women have more responsibilities in regulating emotions, as Bloch et al. (2014) suggested. Instead, it should gather more attention on the imbalance of emotional work in mixed‐sex couple relationships.

Besides, the scale we used for measuring emotion regulation difficulties only assessed a partner's difficulties in regulating negative emotions, without information on the regulation of positive emotions, especially the ratio of negative to positive emotions. It is possible that when partners show greater nonacceptance, they might express fewer negative emotions and engage in more positive interactions to repair perceived ruptures or conflicts (Gottman and Levenson 1992). This seems especially plausible given that couples in our sample were not clinically distressed according to the cutoff score of the relationship satisfaction scale (Anderson et al. 2021).

However, nonacceptance of negative emotions has many adverse effects from studies on individual emotion regulation, including heightened anxiety (Kashdan et al. 2006), depressive moods (Williams and Kuyken 2012), lower self‐awareness (Hayes and Wilson 2003), and individual distress and interpersonal avoidance of emotions (Wolgast et al. 2011).

Specifically in mixed‐sex relationships, the positive effects of nonacceptance might come at the sacrifice of women's emotional well‐being. To “win” men's happiness and relationship satisfaction, women might have to “lose” their authenticity in negative emotional states. They do so by avoiding negativity and focusing on positive elements of the relationship (Weber et al. 2019). Butler et al. (2014) also pointed out that, as a relationship‐focused coping skill for emotional distress, protective buffering (i.e., hiding negative emotions) could help regulate negative emotions in relationships, but it may have come at a cost. Because it is associated with lower levels of positive emotions and less engagement for both partners in a romantic relationship, which could potentially lead to decreased relationship satisfaction over time (Levenson et al. 2014).

What's more, when men experience nonacceptance of negative emotions, they tend to display more relational avoidance (Wolgast et al. 2011). Simultaneously, women might perceive higher pressure on hiding negative emotions for preserving a positive environment (Strazdins and Broom 2004; Weber et al. 2019). Eventually, men's nonacceptance of negative emotions might lead to lower relational closeness and quality for women (Bloch et al. 2014; Weber et al. 2019; Xu et al. 2023). As such, Xu et al. (2023) suggested that clinicians should pay particular attention to this “lose‐win” pattern. When working with couples, clinicians need to help them experience their own and their partner's negative emotions authentically and constructively. Gradually, couples might be able to experience a balanced regulation process in their relationship, with both partners feeling empowered and competent to give and receive vulnerable feelings and emotional experiences.

To sum up, considering the inconsistent findings on nonacceptance of negative emotions, future studies are needed to delineate the intra‐ and inter‐personal aspects of partners' nonacceptance of negative emotions in couple relationships.

5.4.3. Limited Strategies

Our results show, when women experience more difficulties in regulation strategies, both women (β = −0.30, p = 0.003) and men (β = −0.21, p = 0.024) are likely to report lower relationship satisfaction at session #1, but women might gain more changes in relationship satisfaction per session (β = 0.30, p = 0.032). Commonly, women show higher levels of emotional awareness and knowledge, and they are expected to maintain positive emotions for themselves and others (Barrett et al. 2000; Ciarrochi et al. 2005). Yet, these expected emotional works might make women more susceptible to the difficulties in using functional regulation strategies (Brody et al. 2016; Strazdins and Broom 2004). When women experience challenges in regulation strategies, they tend to show more self‐blame and less acceptance of their upset feelings. Potentially, they might engage in more emotional avoidance and dysfunctional regulation strategies, and are less conducive to emotional involvement, effective communication, and empathy towards their partner (Florean and Păsărelu 2019; Halmos et al. 2018). These negative consequences are likely to not only put women at risk for physical and psychological impairments (Strazdins and Broom 2004) but also might negatively impact the relationship satisfaction for their partner (Brandão et al. 2020; Frye et al. 2020; Mazzuca et al. 2019). In couple therapy, when clinicians help women gain more emotion regulation tools and skills, especially at times of emotional distress, women might feel more capable and confident in regulating their emotions, and their partner might experience less spillover of dysregulated emotions (Butler 2011). Therefore, as the partner, men might feel more secure in participating in the emotional process in couple therapy (Greenberg 2016; Johnson 2019), which is pivotal for enhancing relationship satisfaction and promoting positive changes in the course of treatment (Levenson et al. 2014).

Meanwhile, men's difficulties in regulation strategies have adverse links with men's starting score of relationship satisfaction. This finding supported that when individuals perceive themselves as less capable of regulating unpleasant emotions, they might display more emotional avoidance, have less positive views regarding themselves and their partner, and are likely to experience lower relationship satisfaction (Mazzuca et al. 2019).

5.4.4. Impulsivity

Our findings suggest that greater women's impulsivity is connected with lower relationship satisfaction for both partners (women β = −0.28, p = 0.002; men β = −0.30, p = 0.002). These are consistent with prior findings (Bloch et al. 2014; Xu et al. 2023). Impulsive behaviors, particularly amid negative emotions, could significantly damage both partners' emotional and physical safety through reckless expressions and rash actions (Halmos et al. 2018; Watkins et al. 2014). In couple therapy, clinicians help couples stop the negative affect reciprocity cycles, so they might be able to interact with each other with fewer or no reckless behaviors. When women can express feelings and thoughts constructively with the tools from therapy, they are likely to engage less in dysfunctional coping skills (e.g., destruction of property and risky behaviors). By using adaptive regulation skills instead of impulsive behaviors, they might experience less shame and guilt, but a sense of being in control and satisfaction. At the same time, their partner might experience higher levels of emotional and physical safety (Halmos et al. 2018), more intimate connections (Dubé et al. 2019), and greater relationship satisfaction (Omidi and Talighi 2017; Xu et al. 2023).

Also, we find that more men's impulsivity is associated with lower women's relationship satisfaction at session #1. This finding provides extra support to the study by Watkins et al. (2014). They showed that impulsivity compromises partners' relationship satisfaction and their physical and emotional safety. Particularly, men have a higher risk than women of reinforcing impulsivity, especially physical aggression, in the relationship.

Moreover, results from the Wald tests of path quality show no statistical difference between intra−/actor and inter‐personal/partner effects of emotion regulation (ER) difficulties. These findings might suggest that it is equally important to target ER at intra‐ and inter‐personal levels for both partners during couple therapy (Horn and Maercker 2016; Xu et al. 2023). Further, we found more intra‐ than inter‐personal effects of ER difficulties on the trajectory of relationship satisfaction in the initial eight sessions of couple therapy. This might be due to the power limitations of our analysis in detecting partner/interpersonal effects, which are often smaller than actor/intrapersonal effects in dyadic studies. Additionally, one potential rationale is that the scale we used for measuring ER difficulties was specifically developed to capture an individual's difficulties in regulating their own emotions, and their perceptions of the resulting impact on themselves (i.e., intrapersonal effects). It is possible that, while dyadic latent growth curve modeling could estimate part of the interpersonal effects, the limitation of our data on ER difficulties might create a gap between the actual interpersonal effects and those estimated in our study.

Lastly, there are more interpersonal/partner effects of women's ER difficulties than men's. This is in line with previous studies that men do not seem to impact women as much as women influence men (Gottman and Levenson 1992), especially during negative emotional states (Bloch et al. 2014; Xu et al. 2023). Regarding this finding, Bloch et al. (2014) concluded that effective women ER should be viewed as both a characteristic and an indicator of satisfied romantic relationships. Thus, they proposed that women should responsively and competently fulfill their role in regulating the emotional balance in mixed‐sex relationships. On the contrary, we believe that in a systemic approach to couple therapy, clinicians should emphasize the importance of empowering men to take on more emotional work, encouraging women to step down from the “emotion regulator” position, and helping both partners achieve authentic emotional experiences (Xu et al. 2023).

5.5. Clinical Implications

This study provides vital implications for clinicians working with couples, especially those with prominent emotional distress in their relationships. First, targeting emotion regulation (ER) in couples might be a key change mechanism for reducing the percentage (i.e., 25%–30%) of couples who gain no progress or get worse in therapy (Snyder et al. 2005). As Levenson et al. (2014) pointed out, almost all couples seeking therapy have troubled ER. Therefore, to maximize the efficiency and efficacy of couple therapy, we encourage clinicians working with couples to deliberately incorporate ER into the treatment process. For example, using an adequate measure to assess partners' difficulties and differences in ER, to customize treatment plans that would target certain ER issues in the relationship (Xu et al. 2023). To facilitate an increase in relationship satisfaction, clinicians are recommended to focus on enhancing women's regulation strategies and men's emotional awareness, which show an impact on the rates of change (i.e., slopes) in relationship satisfaction across couple therapy sessions.

Second, we recommend treatment modalities that emphasize the emotion regulation (ER) process. For example, Gottman couple therapy, emotionally focused couple therapy by Johnson (2019), and emotion‐focused couple therapy by Greenberg (2016); Lebow and Snyder (2022). Specifically, in Greenberg's emotion‐focused couple therapy, clinicians guide couples to distinguish between biological, social, and cultural elements of emotions and the resulting impacts on each partner and the couple dyad. Considering the gender and the intra−/inter‐personal differences in ER, this emphasis on the multifaceted nature of emotions might be of particular help in breaking the gender imbalance in ER and promoting authenticity and emotional balance in couples (Xu et al. 2023). Furthermore, while most couple therapy models mainly target the decrease in negative or maladaptive emotions (Levenson et al. 2014), Greenberg's emotion‐focused couple therapy presents detailed steps and tasks for navigating both the negative and positive emotional states of each partner and the couple dyad. Hence, Greenberg's emotion‐focused couple therapy might potentially help both partners achieve a maximum emotional state in the relationship by targeting ER at both the intra‐ and inter‐personal levels, also both negative and positive emotions (Levenson et al. 2014; Xu et al. 2023).

Third, there are significant gender differences in partners' emotion regulation difficulties, especially emotional awareness. We encourage clinicians working with couples to pay particular attention to bridging the gap between women's and men's emotional awareness. Because the imbalance might threaten women's psychosocial well‐being, and both partners' relational intimacy and cohesion (Strazdins and Broom 2004). While most couple therapy modalities have elements for promoting emotional awareness, some interventions (e.g., reflecting poignant emotions) from emotionally focused couple therapy (Johnson 2019) might be particularly helpful. Besides, Ciarrochi et al. (2005) found that both women's and men's emotional awareness increased significantly after reading a passage about the significance of emotional awareness. Thus, psychoeducation on emotional awareness, especially for men, might be another approach to breaking the gender imbalance in emotional awareness.

Fourth, across literature and as evidenced in the current study, emotion regulation strategies have both intra‐ and inter‐personal effects on the trajectory of couple relationship satisfaction. In advancing both partners' psychological well‐being and relationship satisfaction (Strazdins and Broom 2004), clinicians should be aware of equipping partners with intra‐ and inter‐personal regulation strategies, especially those for coping with negative emotions. The Process Model of Emotion Regulation by Gross (1998) proposes two main categories of emotion regulation strategies: (1) antecedent‐focused strategies for coping with anticipated emotionally challenging events before the emotions are generated. For instance, mental preparation for an anxiety‐triggering conversation (i.e., financial struggles). (2) response‐focused strategies for regulating emotions that are generated after a stressful event. For example, a reappraisal of the stressful event as an opportunity for personal growth and relational intimacy.

5.6. Limitations and Future Directions

Like all studies, this one is not without limitations, and we will focus on the following. First, the scale used for emotion regulation (ER) difficulties in our study only collected data on partners' intrinsic regulation (i.e., regulation of one's own emotions), without information on extrinsic regulation (i.e., regulation of other's emotions) and the reciprocal effects due to the back‐and‐forth process of ER in couples. Additionally, the scale was developed to examine clinical ER difficulties in individuals (Kaufman et al. 2016). Thus, the data could not provide information on the regulation process of positive emotions in couples. Specifically, we had no information on the ratio of negative to positive emotions, and the related process and consequences in the participating couples' relationship (Gottman and Levenson 1992). These shortcomings call for attention to the interpretation and application of the findings. On the other hand, to future scholars, these limitations further point out the need for an adequate scale for assessing ER in couples (Levenson et al. 2014; Xu et al. 2023).

Second, the couples in our sample are mixed‐sex and primarily white, with an average relationship length of 6.24 years. Because emotions frequently arise in social contexts (Butler 2011), ER is highly sensitive to contextual factors (i.e., age and race/ethnicity). Therefore, the generalizability of the findings in this study might be limited due to the demographic features. Future researchers are encouraged to use more diverse samples, such as couples with potentially no gender or sex differences, couples who are at the initial developmental stage of their relationship, or couples with different cultural backgrounds.

Another limitation of this study is that information on the specific therapeutic approaches used with the participating couples was not available. As a result, it is unclear whether and how differences in treatment modalities may have influenced the observed associations between ER difficulties and relationship satisfaction. For instance, manualized treatments that explicitly emphasize emotion regulation, such as emotionally focused couple therapy (Johnson 2019), may differentially shape emotion regulation compared to more behaviorally oriented treatments. Without this information, our findings should be interpreted as reflecting common therapeutic processes across modalities rather than the effects of a particular model. Furthermore, the dataset did not include detailed information on therapist type or specific inclusion criteria for couple therapy settings (e.g., licensed clinicians vs. student therapists), which may limit the generalizability of our findings to a particular treatment context. Future research that systematically tracks treatment approaches and clinician characteristics will be valuable for clarifying the role of ER‐focused interventions in shaping the trajectories of relationship satisfaction in couple therapy.

Lastly, our sample consisted of couples who attended at least the initial eight sessions of couple therapy; however, most participants did not report clinically significant distress in their relationship (i.e., average scores of relationship satisfaction approximated or exceeded the clinical cutoff). This might have influenced the intensity of observed emotion regulation (ER) difficulties, which could fluctuate the intra‐ and inter‐personal effects of ER on the trajectories of couple therapy outcomes. Additionally, with data from the first eight sessions of couples therapy, the findings should be interpreted as reflecting early treatment processes rather than change trajectories across the full course of therapy. Accordingly, future studies should replicate these findings with couples displaying different levels of relationship distress and ER difficulties across the full course of therapy (e.g., 12 or 24 sessions).

6. Conclusion

This study, to the best of our knowledge, is one of the first studies using a clinical sample to investigate emotion regulation difficulties and the trajectory of relationship satisfaction during the initial eight sessions of couple therapy. With 174 mixed‐sex couples in couple therapy, this study provides a few important findings. First, gender differences exist in emotion regulation difficulties and couple relationship satisfaction. Second, across the initial eight couple therapy sessions, relationship satisfaction improves at a similar rate for women and men. Third, emotion regulation difficulties have intra‐ and inter‐personal effects on the starting scores and change of relationship satisfaction in the initial eight sessions of couple therapy. Surprisingly, we found several positive associations between emotion regulation difficulties and greater relationship satisfaction, including lack of emotional awareness, nonacceptance of negative emotions, and limited regulation strategies. We discussed relevant clinical implications and highlighted the need for future studies with diverse samples.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Appendix S1: famp70125‐sup‐0001‐AppendixS1.docx.

Xu, M. , Johnson L. N., Anderson S. R., Bradford A., and Witting A. B.. 2026. “Emotion Regulation and Relationship Satisfaction in Couple Therapy: What Happens Across Sessions?.” Family Process 65, no. 1: e70125. 10.1111/famp.70125.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.