The effect of emotionally focused couple therapy (EFCT) on shame and intimacy in couples: a randomized controlled trial (RCT)

Abstract

Objectives

Intimate relationships can enhance quality of life and reduce the risk of psychological disorders. Self-disclosure, a fundamental component of intimacy, occurs only when vulnerability is met with acceptance rather than rejection. However, shame disrupts this process by inhibiting vulnerability and promoting defensive strategies, thereby undermining intimacy and relational security. This study aimed to address shame within an Emotionally Focused Couple Therapy framework to enhance intimacy.

Method

A randomized controlled design was employed. Twenty-six married couples were randomly assigned to either an experimental group or a control group. The Experience of Shame Scale and the Personal Assessment of Intimacy in Relationships were used as assessment tools.

Results

Compared with the control, EFCT significantly increased intimacy and reduced shame in the experimental group (p < 0.05).

Conclusion

Shame may disrupt both intimacy and the therapeutic relationship and should be addressed during treatment. The discussion outlines possible mechanisms of change.

Trial registration

This paper has been registered with the Clinical Trial ID (71773). It was registered on 2024/10/04. (Clinical Trial Number IRCT20230801058994N1)

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-025-03415-3.

Introduction

Since humans are inherently social beings who rely on connections, experiencing an intimate relationship plays a crucial role in quality of life [1, 2]. The need to communicate with others throughout one’s lifespan develops the nervous system [3, 4], coping strategies for stress [5, 6], daily emotions [7], and interpersonal issues [8]. In addition, an intimate relationship reduces the risk of depression, and high levels of intimacy are negatively related to loneliness and depression [9–11].

Moreover, a lack of sexual and emotional intimacy in couples is associated with marital dissatisfaction [12–14], which may lead to divorce [15]. The divorce-to-marriage ratio has increased more than twofold in the last decade [16]. Considering the given information and the growing demand for couples therapy among clients, it is necessary to pay attention to intimacy enhancement and relationship improvement [17].

Intimacy, fears, and desire

Intimacy can be characterized as a multifaceted interpersonal phenomenon that encompasses various dimensions of a relationship, such as social, emotional, and sexual intimacy [18]. Moreover, intimacy involves the establishment of a deep emotional connection between two individuals who engages multiple aspects of self-worth [19]. For this purpose, intimacy can be defined through two aspects: self-disclosure and validation [20]. Therefore, intimacy involves sharing something personal and receiving validating, accepting, and caring responses [19].

The need for intimacy evolves throughout childhood and adolescence [21]. Childhood experiences shape a person’s perception of the world, which affects the behavioural patterns in relationships throughout the lifespan [22]. Intimacy increases in relationships when two individuals reciprocally engage in vulnerable self-disclosure and find their partner accepting and understanding [23]. Simultaneously, as a person is about to express needs and desires, fear of rejection is encountered. It is closely linked to shame and can serve as a significant barrier to intimacy. Consequently, needs may be either suppressed or expressed with anger [24]. Therefore, to overcome obstacles to intimacy, shame should be addressed and acknowledged.

Self-disclosure can occur only when vulnerability is met with acceptance rather than rejection. In other words, intimacy cannot flourish if one partner’s vulnerable self-expression is dismissed or perceived as threatening [19]. Intimacy is progressively shaped and deepened through this iterative process of sharing and accepting vulnerable disclosures. However, shame functions as a critical disruptor in this process. Shame, whether rooted in current relationship insecurities or past relational wounds, inhibits individuals from disclosing vulnerable aspects of themselves. This interference with self-disclosure impairs the development of intimacy and contributes to relational insecurity [23]. Thus, the presence of shame acts as a formidable barrier to emotional safety and connection, impeding the mutual acceptance necessary for intimacy to emerge and thrive. We ensure that this conceptual nuance is more explicitly articulated in the revised abstract and further elaborated in the introduction for greater clarity [24].

Shame, hidden emotion

Shame is a complex self-conscious emotion with an emotional desire to disappear as well as a deep perception of being incomplete, inadequate, and unacceptable [25]. Shame is an emotion that arises as a result of self-evaluation from the other’s perspective, resulting in a negative perception of oneself [26]. Although ‘state shame’ is transitory and can be experienced by anyone throughout life, trait shame or shame proneness refers to the pervasive experience of shame, also known as core shame [27, 28]. In this context, adaptive shame is useful for socialization, whereas maladaptive shame emerges with any deficit [29]. This maladaptive shame has symptoms such as humiliation and devaluation, self-loathing and self-mockery, idealization, perfectionism, and a strong desire to control, which can be some of the symptoms of maladaptive and chronic shame [28].

Shame explanations, such as the experience of personal failure or wrong behaviour, are common but incomplete. In contrast, shame is described as an interpersonal emotion or an immediate internal reaction to threats to connection or emotional bonds [30]. In other words, while enthusiasm is expected in a relationship, the underlying fears of being lovable and valuable are a concern [31].

It seems that shame proneness has a developmental root. Repeated disruptions in the caregiver–infant relationship may constitute relational trauma, resulting in shame internalization and leading to impaired autonomic responses to emotional arousal and interpersonal difficulties [29]. Infants are likely more prone to experience shame in the absence of sufficient emotional or physical attunement from their caregivers [32].

When the child seeks to share an experience with the caregiver but the caregiver does not respond appropriately, shame is provoked. Therefore, every child has experienced shame; however, the parent’s action reattunes and repairs the relational rupture may determine whether state/trait shame is shaped. In this context, attachment is an essential issue that should be noted [27].

Shame-prone individuals might be more capable of engaging in maladaptive relationships in adulthood. Shame can predict lower satisfaction, poor relationship quality, and increased fear of intimacy [33]. As self-disclosure and corresponding validation are essential for intimacy and occur in a secure relationship, the experience of shame can be an obstacle to an individual’s self-disclosure in relationships and therapeutic alliances [26].

Moreover, emotional inhibition is a concern in attaining effective treatment. Addressing shame and its impact on interpersonal functioning and therapeutic relationships should be considered a primary focus of intervention [34].

Attachment theory: connecting and integrating

It can be assumed that attachment theory is the first and only theory concentrating on childhood relationships and their impact on individuals throughout their lifespan [35]. Attachment theory explains that humans will barely survive without an attachment feature. Attachment theory holds that emotional dysregulation could be the result of not satisfying childhood needs such as physical and emotional availability and responsiveness [36].

In addition, emotional interactions in an attachment-based relationship might lead to the emergence of the brain structure and function. The sustainable nature of attachment styles might lead to the development of fixed emotion regulation and rigid strategies in social bonding. Infants and toddlers develop common interpersonal patterns with their caregivers. These patterns, which are called ‘internal working models’, are maintained during the lifespan [37]. Nevertheless, attachment needs are not limited to childhood; rather, in adulthood, emotional closeness and validation are considered two fundamental needs of attachment in close relationships, especially couple’s relationships [37]. Attachment injuries in marriage and romantic relationships are likely to heighten the risk of physical disease and mental disorders and increase marital distress [38].

Attachment styles are often discussed as relatively stable personality-based typologies (e.g., anxious, avoidant, secure). They are typically treated as trait-like tendencies formed in early caregiving environments. On the other hand, attachment strategies are understood as context-dependent regulatory behaviours that individuals use to manage closeness, vulnerability, and emotional threat within relationships (e.g., protest behaviours, withdrawal, hyperactivation). These are often more fluid and responsive to change, especially in the therapeutic context [1].

From a neurological perspective, the left hemisphere is verbal, and relatedness is not established until it is integrated with the right hemisphere, which is emotional and nonverbal. The right hemisphere regulates emotions, which should be reflected primarily by parents. When caregivers fail to coregulate the child’s emotional arousal—due to misattunement or emotional unavailability—the child may experience chronic anxiety in emotionally charged contexts and accordingly might be prone to drive rational strategies by the left hemisphere to consider and act. In other words, when parents’ right hemisphere is barely in harmony with the child’s right hemisphere, the connection between the two hemispheres becomes weak or even cut off. In this context, a lack of attunement may lead to shame-proneness [39]. Considering that attunement is a principal part of emotionally focused therapy, it is assumed that shame treatment might be effective in this framework.

Emotionally focused couple therapy

EFCT is based on attachment theory and applies humanistic and systemic principles to improve functioning in couples’ relationships, which is achieved by creating a more secure attachment bond [35, 40]. Emotionally Focused Couple Therapy (EFCT) is one of the most effective treatments in repeated measurement studies from pretreatment to two-year follow-up. EFCT has a significant effect on approximately 70% of cases on average, and in approximately 82% of these cases, the stability of the change has been confirmed during follow-up [41, 42].

The empiricism and humanistic nature of EFT have caused principles such as empathy and unconditional positive attention to be included in the context of therapy. In emotionally focused therapy, it is assumed that attachment is the heart of the emotion regulation process. Therefore, it aims to achieve self-regulation through coregulation, which in turn enhances empathy and validation [40].

Furthermore, EFCT focuses on a couple’s emotional engagement and de-escalates the negative interaction cycle between them to smooth the emotional experience. Enactment is an important technique in EFCT, leading to more engagement in emotional exchange and easing the expression of vulnerability [40, 42].

Therefore, the anticipated result of EFCT is balanced interdependence, on the basis of Attachment theory, human beings shape their identity and emotions in relationship with others [43]. Given that the attachment process focuses meaningfully on the client’s interpersonal patterns, establishing a secure base in which the client re-engages in past relational dynamics/patterns and experiences primary emotions to understand the need behind it and improve attachment wounds is essential. A healthy relationship in psychotherapy can help the patient discover the self and regulate emotions [44].

Although Greenberg and Goldman [45] devoted a great attention to shame within the framework of Emotion-Focused Therapy for Couple (EFT-C), several key differences exist between their approach and the current study. While their work provides valuable insight, the present research is grounded in Johnson’s protocol [35], wherein shame is considered one of the five core emotions and is addressed implicitly throughout the therapeutic process. In contrast, Greenberg’s model emphasizes identity formation and mastery, interpreting shame primarily through the lens of dominance and submission dynamics. However, this conceptualization does not fully incorporate shame into the broader framework of modern attachment theory, which explains relational trauma and the need for right hemisphere attunement [29, 32].

Furthermore, while Greenberg and Goldman explore the behavioural outcomes of shame, other important relational and emotional consequences—particularly those relevant to therapeutic alliances—are underdeveloped in their model. In the present study, shame is examined not only as a core emotion but also as a relational construct with significant implications for the therapeutic relationship and emotional regulation within couples.

This study

Shame can cause behaviours in relationships that undermine transparency and closeness [27, 46]. Moreover, experiencing shame and defence mechanisms against it might lead to destructive behavioural patterns such as control or withdrawal behaviours that negatively affect intimate relationships and psychological well-being [47]. Shame predicts lower satisfaction and poorer relationship quality [26]. In addition, shame has been identified as a mediating factor linking relationship problems and depression [48].

Above all, shame is important in the therapeutic relationship. Psychotherapy will be effective if safe therapeutic communication is established [49]. This obstacle must be resolved since the therapeutic alliance is one of the most consistent predictors of therapeutic success [50, 51]. Internalized shame-coping strategies may have negative impacts on both treatment and intimate relationships. Four primary shame-coping strategies have been identified, namely, self-attack, attack on others, withdrawal, and avoidance. Individuals who use a withdrawal coping strategy are more likely to have an ineffective therapeutic relationship that negatively affects therapy progress [34]. Experiencing maladaptive shame may shape a vicious cycle of shame isolation that causes self-fragmentation [52]. Assessing shame and coping strategies during the evaluation stage of therapy and interventions focused on shame in the early stages of treatment can improve patient prognosis by providing insight into possible problems in the establishment of therapeutic relationships [31, 34].

Intimacy is more extensive, complicated, and misunderstood than other concepts in explaining interpersonal relationships and needs to be further clarified [21]. In this context, many studies have demonstrated the effects of EFCT on issues such as intimacy [53], marital satisfaction [54, 55], sexual assertiveness [56, 57], emotional dysregulation [58, 59], and health in married life [60], whereas few studies have investigated factors and obstacles related to intimacy enhancement [61].

On the basis of the above considerations, shame, as an interpersonal emotion, can be vital in the emergence and maintenance of intimate relationships [62]. Despite the potential impact of clients’ experiences of shame and various theories indicating this [30, 63–65], researchers have barely evaluated and addressed shame as a core emotion in therapeutic communication [66–68]. Although some studies have suggested assumptions about the association between shame and intimacy, the literature rarely concentrates on different aspects of intimacy [61, 69].

The current study examines the effectiveness of methods for communicating with clients’ experience of shame and aims to investigate the effectiveness of EFCT on shame and intimacy. Although research has confirmed the effects of EFT on social anxiety [70] and PTSD symptoms [71], with a focus on shame to our knowledge, there are no studies on the effectiveness of EFCT on shame in the nonclinical public [72]. Although some studies emphasize shame in the treatment of couples [49, 68] and intimacy [67], it is still more of a theory than a clinical intervention. While earlier work suggested that individuals with avoidant attachment may not benefit significantly from EFCT [55], more recent studies have provided evidence for changes in avoidant attachment through mechanisms such as withdrawer re-engagement [57]. However, these studies have not directly addressed the role of shame in the process of change. Given that shame is particularly salient in avoidant individuals and may inhibit emotional engagement [34, 73, 74], the current study aims to explore how addressing shame can facilitate therapeutic progress in this population.

The hypotheses of the present study are as follows:

Hypothesis 1

EFCT can reduce shame, or there is a significant difference between the experimental and control groups in the post-test scores for shame (experience of shame scale).

Hypothesis 2

EFCT can increase intimacy, or there is a significant difference between the experimental and control groups in the post-test scores for intimacy (personal assessment of intimate relationships).

Hypothesis 3

Working with shame in the frame of EFCT can reduce avoidance and withdrawal in the middle of therapy and enrich the process (on the basis of a nonverbal and behavioural assessment of shame and coping strategies against shame).

Method

Participants and procedure

This research employed a randomized controlled trial that used a pre-test/post-test model and included a control group. The sampling flowchart is presented in Fig. 1. A research announcement was distributed to all couples attending the three counselling and psychotherapy centres that had agreed to collaborate in the study. A total of 100 married couples were approached during the sampling period. Of these, 40 couples expressed interest in participating. Following the initial screening interviews, 4 couples were excluded because they did not meet the inclusion criteria, leaving 36 eligible couples. These couples were randomly (using a table of random numbers) assigned to two groups of 18: one group participated in couples’ therapy sessions, while the other group attended workshop sessions. The power analysis was conducted via G*Power 3, which is based on an expected medium effect size for EFTs (f² = 0.35), an alpha level of 0.05, and a desired statistical power of 0.80. The required sample size for detecting significant effects in a covariate analysis was estimated to be 67 individuals (or 34 couples that were divided into two groups of 17 experimental and control couples). The first sample size (18 couples in each group) exceeds this threshold.

Fig. 1.

Sampling flowchart

During the study, 3 couples from the therapy group and 2 couples from the workshop group withdrew for personal reasons. Additionally, in the final phase, 2 couples from the workshop group and one couple from the therapy group did not complete the post-test questionnaires. Furthermore, one couple in the therapy group decided to separate, resulting in the exclusion of their data from the analysis. Ultimately, data from 13 couples in each group were included in the final analysis.

The entry criteria included at least one year of cohabitation, the absence of addiction and violence and extramarital relationships, reading and writing literacy, the motivation to participate in the therapy sessions, not suffering from a serious illness, and not using psychiatric drugs. The exit criteria included unwillingness to continue the cooperation, critical events, and absence in more than one meeting per month.

After the interview and pre-test stages, the experimental group was planned to engage in couple therapy sessions, whereas the control group participated in a workshop titled “Relationship Improvement”. The experimental group received 20 sessions of emotionally focused couple therapy, and the control group participated in 20 sessions of the workshop.

First, each couple participated in a 90-minute joint interview session, followed by individual 60-minute sessions where each partner responded to questions about their attachment history and relational experiences. These sessions also focused on establishing a therapeutic contract and building a strong therapeutic alliance.

In addition to self-reported assessments of shame at the pre- and post-intervention stages, qualitative and behavioural data were gathered via therapist observations, clinical notes, and supervision discussions. Behavioural signs of shame (e.g., gaze aversion, posture withdrawal, tone shifts, sudden silence) and shame-based coping strategies (e.g., avoidance, self-attack, criticism) were documented session by session.

The therapeutic process then followed the standard EFT phases: de-escalation of the negative interaction cycle, identifying and validating secondary emotions, accessing primary emotions, and reframing the couple’s interactions. In the subsequent phase, both partners were encouraged to engage emotionally, empathize with each other’s attachment injuries, and express their feelings and needs more openly (accepting vulnerability). Finally, the last phase involved consolidating the new positive cycle and discussing possible future concerns and challenges.

During the early and middle phases of therapy, individual sessions were also held with a focus on repairing shame, tailored to each client’s attachment strategies and experiences. This was a slight deviation from the standard EFCT protocol. However, this deviation was not fundamental. Johnson noted that one reason for holding individual sessions is the emergence of therapeutic obstacles—shame being one of them [58]. While the original protocol does not address this directly, our study sought to examine the hypothesis that, owing to the close relationship between shame, attachment, and therapeutic alliance, shame should be assessed more explicitly—although in a way that would not further shame the client.

To address shame without shaming the client, the therapist relied on strategies rooted in EFCT principles of emotional safety, validation, and attunement. The most important thing that was mentioned in the article was avoiding direct confrontation or labelling. The term “shame” was rarely used explicitly in early sessions. Instead, the therapist focused on exploring clients’ emotional experiences in their own language, allowing shame-related material to emerge gradually and safely. Before approaching vulnerable material, the therapist prioritized establishing a strong alliance and modelling nonjudgmental curiosity. This included a warm tone, a soft pace, and affirming the client’s courage in sharing difficult emotions.

Shame-related defences (e.g., withdrawal, silence, and sarcasm) were framed as understandable responses to emotional pain, with no signs of dysfunction. Clients were helped to see their reactions as protective strategies developed in response to unmet attachment needs. Validation and normalization were used for this purpose.

The therapist monitored subtle signs of activation (e.g., gaze aversion, frozen expression, shifts in posture) and responded with empathy and pacing rather than pushing for immediate disclosure. The goal was to invite clients into contact with their shame experience while maintaining a compassionate and accepting therapeutic stance.

The second group participated in sessions under the title of a relationship enhancement workshop. However, these were not truly workshops focused on EFT or shame. Rather, the participants discussed their relational problems and challenges while the therapist and others listened in a group discussion setting. The logic behind these sessions was that shame is a relational emotion, and if the second group had responded to the questionnaires without any relational engagement with a therapist, the observed differences between the two groups might have been due solely to the presence of an empathic listener rather than the effectiveness of EFT on shame and intimacy. In addition, couples in the control group were authorized to choose to participate in therapy at the end of the research.

One therapist performed therapy sessions and the workshop. She was trained in clinical psychology; completed a ‘therapy in practice’ workshop, an ‘Emotionally Focused Couple Therapy’ workshop, and a ‘principles of attachment’ workshop; and underwent a 6-month supervision period with a couple therapist supervisor. Given that the therapist was a Ph.D. candidate (first author), it was necessary to ensure rigorous training and supervision to guarantee the fidelity of the treatment and the research process. To this end, the overall treatment plan was first aligned with the original Emotionally Focused Couple Therapy (EFT) manuals by Dr Sue Johnson [40, 58]. It was then reviewed, revised, and finalized by the research team. Throughout the intervention period, the therapist was under the supervision of an independent supervisor who had no conflict of interest, had over 10 years of experience in couples therapy, and had received formal training and supervision in EFT.

Adherence to the intervention was monitored through regular check-ins by the research team and by referring to Johnson’s sources [35] to ensure adherence to the correct stages and techniques. In addition, the therapy sessions were held under the supervision of an emotionally focused couple therapist with no conflicts of interest. The CONSORT reporting guidelines were used [75].

Progression through the EFCT stages was monitored via a combination of therapist session notes, clinical checklists based on Johnson’s (2004, 2019) model, and supervision discussions. Specific indicators—such as the de-escalation of the negative cycle, increased emotional engagement, and client readiness for enactments—guided therapists in determining movement between stages.

Therapists looked for a reduction in the intensity and frequency of negative patterns, increased emotional accessibility, and initial vulnerability between partners. When clients could identify their role in the cycle and take emotional risks without high levels of blame or withdrawal, therapists moved into Stage 2 work.

Stage 2 work—focused on consolidation and new interaction patterns—was attempted only when the spouses showed consistent responsiveness, emotional regulation, and secure bonding moments. This included clear, corrective emotional experiences (e.g., withdrawer re-engagement, blamer softening) that were observed and discussed in supervision sessions.

In this study, Stage 3 was introduced selectively and modestly. It typically involves reinforcing new emotional responses, reflecting growth, and planning future relational resilience. We have revised the manuscript to clarify that not all couples fully completed Stage 3 and that fidelity to the model was maintained through supervised adherence to clinical indicators for each phase.

If the therapist deviated from EFCT fidelity in any session (e.g., moved too quickly into enactments, overused cognitive interpretations, or failed to fully assess emotion), the supervisor provided feedback to redirect the therapeutic process back in alignment with EFCT principles. Specific moments were reviewed, and alternative interventions were explored. Deviations were noted and reflected upon by the therapist. These reflections were used to enhance learning and maintain intentionality in subsequent sessions. Rather than being excluded, minor deviations were integrated into supervision as natural challenges in learning EFCT. When needed, course corrections were implemented in the following session.

Measures

The experience of shame scale (ESS)

This scale consists of 25 items. This scale measures three dimensions of shame, including experiential (Have you felt ashamed of any of your personal habits? ), cognitive (Have you worried about what other people think of any of your personal habits? ), and behavioural (Have you tried to conceal any of your personal habits? ). Answering and scoring are based on a 4-point Likert scale from 1 (not at all) to 4 (very much), and the total score is between 25 and 100. The Cronbach’s alpha of all the subscales is greater than 0.80 [76]. In the present study, the Cronbach’s alpha coefficient of the scale was 0.80.

Personal assessment of intimate relationships (PAIR)

This scale includes 36 items divided into 6 subscales: emotional intimacy (My partner listens to me when I need someone to talk to), social intimacy (We enjoy spending time with other couples), sexual intimacy (I am satisfied with our sex life), intellectual intimacy (My partner helps me clarify my thoughts), recreational intimacy (We enjoy the same recreational activities), and the conventionality scale (My partner has all the qualities I have ever wanted in a mate) [18]. The scoring method of this scale is based on a 5-point Likert scale from 1 (does not apply to me at all) to 5 (extremely applies to me). Therefore, the minimum and maximum scores on this scale range from 36 to 180. Cronbach’s alpha of the subscales has been reported to be above 0.70 [18]. In the present study, the total Cronbach’s alpha coefficient was 0.87.

Data analysis

ANCOVA was used as the primary statistical method, with shame and intimacy as the main variables of interest. This study did not include missing data. After all assumptions for the analysis were verified, the covariate status of the pre-test scores were confirmed through correlation analysis, and the effectiveness of the covariates was assessed via a built model. The analysis was then conducted for the two main variables, followed by separate analyses for each variable subscale. The analysis was accomplished via SPSS v26. Repeated measures analysis was used to study the effect of therapy on the dependent variables.

Results

Preliminary analysis

Thirteen couples participated in the experimental group. In this group, 4 people (15%) had diplomas, 12 people (48%) had bachelor’s degrees, 6 people (22%) had master’s degrees, and 4 people (15%) had Ph. D. The age of the experimental group was between 25 and 46 years, and the average age of the subjects was 37.16 years. Additionally, between 1 and 18 years of age had passed since people’s marriage, and on average, 5.64 years had passed since their marriage.

Thirteen couples participated in the control group. In this group, 1 person (3.8%) had a diploma, 9 (34.6%) had a bachelor’s degree, 10 (38.5%) had a master’s degree, and 6 (23.1%) had a Ph. D. The age range of the participants was between 26 and 57 years, and the average age was 32.58 years. The duration of marriage among people is between 1 and 31 years, and on average, 7.77 years have passed since marriage.

A chi-square test of independence revealed no significant difference between the experimental and control groups in terms of education level (χ² _{(3, N=52)} = 6.25, p = 0.09 > 0.05) and job (χ² _{(2, N=52)} = 1.81, p = 0.40 > 0.05). In addition, there was no significant difference in the duration of marriage between the two groups (t _(1,50) =-1.29, p = 0.20 > 0.05). Independent sample t tests revealed a significant difference in age between the two groups (t _(1,50) =-2.33, p = 0.24 < 0.05). However, in the ANCOVA model including age and pre-test scores as covariates, the interaction effect between group and age was not significant (F _{(1, 50)} = 1.62, p = 0.10 > 0.05), indicating that age did not moderate the effect of the intervention. Therefore, age was retained as a covariate in the final model, but it did not influence the treatment outcome.

In the pre-test stage, there was no significant difference between the scores of shame (t _(1,50) =-0.49, p = 0.62 > 0.05) and intimacy in the two groups (t _(1,50) =-1.93, p = 0.06 > 0.05). The scores were normal according to the Shapiro‒Wilk test (P = 0.20 > 0.05). The means and standard deviations of the shame and intimacy subscales for each group and stage are presented in Table 1.

Table 1.

Descriptive statistics of the experimental and control groups

	Experimental Group				Control Group
	Pre-test		Post-test		Pre-test		Post-test
	M	SD	M	SD	M	SD	M	SD
Experience	17.81	5.71	16.62	4.89	18.69	5.15	17.81	4.36
Cognition	18.92	6.50	16.88	5.63	19.50	6.72	20.65	6.59
Behaviour	17.42	5.95	14.92	5.57	18.12	4.27	18.19	4.69
Shame	54.15	16.56	48.42	15.14	56.31	14.79	56.65	13.84
Emotional Intimacy	17.58	7.14	20.85	7.68	21.35	6.29	20.19	7.16
Social Intimacy	16.77	3.88	17.88	4.38	19.65	5.48	19.42	6.67
Sexual Intimacy	20.96	6.38	22.70	6.32	22.08	5.97	19.23	4.90
Intellectual Intimacy	18.62	6.01	20.23	5.93	20.88	5.49	21.77	5.86
Recreational Intimacy	19.35	5.62	19.96	5.93	22.31	5.86	21.23	5.82
Conventionality Scale	19.27	6.00	20.12	6.41	20.50	4.92	19.85	5.72
Intimacy	112.54	26.94	121.73	29.72	126.80	26.10	121.69	27.35

Multivariate covariance analysis

Multivariate covariance analysis was used for the analysis of the results. Since the independent variable (Group or type of treatment) is nominal and the dependent variables (post-test scores of shame and intimacy) are scale-type and the assumptions of the analysis are met (including independence of groups, random sampling, normal distribution of dependent variables, and homogeneity of variance), multivariate covariance analysis is allowed. Therefore, a multivariate covariance analysis design was used to investigate the effects of Emotionally Focused Couple Therapy on shame and intimacy.

The results of the box test of the equality of covariance matrices revealed that the covariance matrices of shame (M = 11.92, p = 0.08 > 0.05) and intimacy (M = 37.37, p = 0.052 > 0.05) are equal in both groups. In addition, the results of Levin’s test to check the homogeneity of the variance of the dependent variables in the groups are presented in Tables 2 and 3.

Table 2.

Levene’s test of the equality of error variances for shame subscales

Variable	F	df1	df2	Sig.
Experience	0.238	1	50	0.628
Cognition	1.838	1	50	0.181
Behaviour	0.007	1	50	0.934

Table 3.

Levene’s test of equality of error variances for intimacy subscales

Variable	F	df1	df2	Sig.
Emotional Intimacy	0.145	1	50	0.70
Social Intimacy	1.30	1	50	0.26
Sexual Intimacy	0.04	1	50	0.85
Intellectual Intimacy	0.90	1	50	0.35
Recreational Intimacy	0.01	1	50	0.92
Conventionality Scale	1.20	1	50	0.28

The homogeneity test of regression coefficients was investigated through the interaction of the dependent variables and the independent variable (intervention method) in the pre-test and post-test stages. The interaction of these pretests and post-tests with the independent variable was insignificant and indicated the homogeneity of the regression slope. Therefore, this assumption is also valid. In accordance with the assumptions of multivariate covariance analysis, the use of this test was allowed.

The correlations between the scores of shame and intimacy in the two stages of the test were calculated. On this basis, there was a significant correlation between ‘shame pre-test’ and ‘shame post-test’ (r ₍₅₀₎ = 0.54, p < 0.01) and between ‘intimacy pre-test’ and ‘intimacy post-test’ (r ₍₅₀₎ = 0.80, p < 0.01). Therefore, pre-test stage scores can be used as covariates. Despite our assumption based on previous research, there was no significant correlation between shame and intimacy in either of the stages. Therefore, multivariate covariance was performed for each one independently. Additionally, in the built-term model, the pretest scores had a significant effect on the variance, whereas the demographic data (age, education, marriage time) had no significant effect.

Next, multivariate covariance analysis was performed to determine the differences between the groups. The results indicated that the experimental group and control group were significantly different in terms of shame (F _(1,50) = 3.38, p = 0.03 < 0.05) and intimacy (F _(1,50) = 3.30, p = 0.02 < 0.01). The results of the multivariate analysis are presented in Tables 4 and 5. As a result, the EFCT can significantly reduce shame and increase intimacy in couples.

Table 4.

Multivariate analysis for shame

Effect	Value	F	Hypothesis df	Error df	Sig.	Partial Eta Squared	Observed Power
Pillai’s Trace	0.18	3.39	3.00	47.00	0.03	0.18	0.73
Wilks’ Lambda	0.82	3.39	3.00	47.00	0.03	0.18	0.73
Hotelling’s Trace	0.22	3.39	3.00	47.00	0.03	0.18	0.73
Roy’s Largest Root	0.22	3.39	3.00	47.00	0.03	0.18	0.73

Table 5.

Multivariate analysis for intimacy

Effect	Value	F	Hypothesis df	Error df	Sig.	Partial Eta Squared	Observed Power
Pillai’s Trace	0.31	3.30	6.00	44.00	0.01	0.31	0.90
Wilks’ Lambda	0.70	3.30	6.00	44.00	0.01	0.31	0.90
Hotelling’s Trace	0.45	3.30	6.00	44.00	0.01	0.31	0.90
Roy’s Largest Root	0.45	3.30	6.00	44.00	0.01	0.31	0.90

Next, the univariate analysis results of the subscales were analysed. As presented in Table 6, for the shame subscale, there was a significant effect for cognition (F (1,50) = 4.93, p = 0.03 < 0.05) and behaviour (F (1,50) = 5.30, p = 0.03 < 0.05), but there was no significant effect for experience (F (1,50) = 0.60, p = 0.44 > 0.05).

Table 6.

Univariate analysis for shame subscales

Dependent Variable	Sum of Squares	df	Mean Square	F	Sig.	Effect size	Observed Power
Experience	8.89	1	8.89	0.60	0.44	0.01	0.12
Cognition	147.37	1	147.37	4.93	0.03	0.09	0.58
Behaviour	111.49	1	111.49	5.30	0.03	0.10	0.62

The univariate analysis results of the subscales were analysed. As presented in Table 7 for the intimacy subscales, there was a significant effect for emotional intimacy (F (1,50) = 4.36, p = 0.04 < 0.05), sexual intimacy (F (1,50) = 13.33, p = 0.00 < 0.05) and the conventionality scale (F (1,50) = 6.21, p = 0.02 < 0.05), but there was no significant effect for social intimacy (F (1,50) = 0.01, p = 0.94 > 0.05), intellectual intimacy (F (1,50) = 0.42, p = 0.52 > 0.05) and recreational intimacy (F (1,50) = 0.60, p = 0.45 > 0.05).

Table 7.

Univariate analysis for intimacy subscales

Dependent Variable	Sum of Squares		df	Mean Square	F	Sig.	Effect size	Observed Power
Emotional Intimacy		134.40	1	134.40	4.36	0.04	0.08	0.53
Social Intimacy		0.15	1	0.15	0.01	0.94	0.00	0.05
Sexual Intimacy		311.60	1	311.60	13.33	0.00	0.21	0.95
Intellectual Intimacy		7.01	1	7.01	0.42	0.52	0.01	0.10
Recreational Intimacy		10.34	1	10.34	0.60	0.45	0.01	0.11
Conventionality Scale		94.14	1	94.14	6.21	0.02	0.11	0.68

Clinical significance

Clinical significance was examined via the RCI [77]. The “cut-off” was used for calculations, which is the midpoint between the normative mean and the pretherapy mean. Our categories of clinical significance are deterioration (reliable change in a negative direction; separation, or drop out of treatment because of doing poorly), no change (no reliable improvement in either direction), improvement (reliable change, change in a positive direction but not reaching the normal range), and recovery (reliable change in a positive direction and reaching the normal direction). The categories of clinical significance are shown in Table 8.

Table 8.

Clinical significance of the experimental groupe

Measure	Deteriorated	Unchanged	Improved	Recovered
Shame (ESS)	5 (18%)	7 (26%)	9 (33%)	6 (23%)
Intimacy (PAIR)	3 (12%)	5 (18%)	8 (30%)	11(40%)

Discussion

This study aimed to investigate the effectiveness of Emotionally Focused Couple Therapy on intimacy. In addition, it was assumed from the literature that shame can be an obstacle to attaining intimacy [46, 68]. Therefore, this research emphasized and addressed shame in therapy and examined its changes during the therapy process.

EFCT can reduce shame, or there is a significant difference between the experimental and control groups in the post-test scores for shame (experience of shame scale)

The results indicated a significant reduction in shame scores within the experimental group from pre- to post-treatment, as well as a significant post-treatment difference between the experimental and control groups. These findings fit with those of previous studies [59, 65].

A relationship grows through self-disclosure, whereas shame is a barrier to self-disclosure in close relationships [33, 66] and in therapy [50, 51], which may prevent the formation of strong relationships. Individuals who experience extreme levels of shame may find it difficult to relate to others because of underlying fears of being unlovable [66, 67]. These beliefs and confrontational responses prevent clients from establishing a safe therapeutic relationship. Dealing with clients’ experiences of shame in therapy facilitates the treatment process and reduces their avoidance [35, 49]. On the basis of attachment theory, mechanisms against shame are likely to correspond to internal working models. Therefore, integrating shame work as a central component in an attachment-based intervention strengthens the therapeutic relationship and the effectiveness of treatment [34].

Notably, the Experience of Shame Scale primarily measures maladaptive shame; however, individuals who habitually avoid or deny the experience of shame may become aware of and accept the shame they experience as a first step in therapy, even if it is unpleasant. Such increases should not be interpreted solely as negative outcomes but rather as a reflection of emotional awareness. Consequently, cognitive and behavioural shame significantly differed between the two groups. Nevertheless, experiential shame was not significantly different. Therefore, it can be concluded that when individuals become aware of shame, they might experience it more, while it influences their behaviours and relationships less [69].

EFCT can increase intimacy, or there is a significant difference between the experimental and control groups in the post-test scores for intimacy (personal assessment of intimate relationships)

The results demonstrated a significant increase in intimacy in the experimental group after therapy compared with the control group. These findings are consistent with previous research [53, 56–58].

Based on EFT, the cycle of negative interaction begins when partners are trying to express their emotional needs, but the way they do that triggers each other’s emotional vulnerabilities. When the communication process is explained from an attachment perspective, experiencing secondary and primary emotions is accessible respectively [53, 59]. Additionally, processing the shame that has caused an inability to see their vulnerability and show it to their spouse leads to a conversation about vulnerabilities and increased understanding and empathy. Therefore, new communication patterns can replace anxious or avoidance behaviours [42].

Given that the EFCT focuses meaningfully on the client’s emotional experience, establishing a secure base in which the client performs previous patterns while being aware of them without drowning in those patterns is principal. Within the context of a safe relationship, clients are able to experience their emotions, and intimacy can be improved [35, 66].

In addition, creating a safe relationship provides accommodations in which the strategies used by the client to prevent the experience of shame are discussed. This leads to an increase in a client’s emotional acceptance [60]. As a result, if individuals experience negative emotions that neither they nor their partner could regulate before, they are capable of self-regulation and coregulation [49]. They can talk to each other about their problems and accept differences in their opinions and life experiences. In other words, when shame is resolved, emotional experience is enhanced [43, 51].

Compared with other intimacy dimensions, the present study revealed greater gains in sexual intimacy. EFT focuses primarily on emotional connections and is particularly concerned with expressing and uncovering unspoken needs. Consequently, when spouses with marital conflicts show signs of sexual problems, including sexual assertiveness [56, 57], EFT interventions reduce their conflicts and improve their sexual problems [35, 56].

Working with shame in the frame of EFCT can reduce avoidance and withdrawal in the middle of therapy and enrich the process (on the basis of a nonverbal and behavioural assessment of shame and coping strategies against shame)

Notably, in this study, the coping strategy used to address shame was observed during treatment. Individuals try four main coping strategies, namely, self-attack, attack others, withdrawal, and avoidance [64]. The literature indicates that individuals who use withdrawal as a coping strategy are more likely to have an ineffective therapeutic relationship that negatively affects therapy progress [34, 53]. In addition, clients who use self-attacks are more likely to have dysfunctional intimate relationships, which can negatively impact their mental health [44]. During the therapy sessions in this study, whenever signs of withdrawal were detected, shame was tracked to determine whether the decision to stop the sessions was defensive. In addition, any other shame-related behaviours (self-attacks, others’ attacks, and avoidance) was shown to enhance the experience of shame and, consequently, the primary emotion. This increased awareness could facilitate intimacy [46, 49, 67].

In this study, some individuals experienced increases in both shame and intimacy while simultaneously exhibiting a reduction in defensive behaviours related to shame and destructive behaviours within intimate relationships [53, 67]. In contrast, participants who were already aware of their shame but attempted to manage it through strategies such as self-attacks or others’ attacks (for example, by blame or criticism) experienced a shame reduction and a corresponding increase in intimacy [66].

Furthermore, on the basis of interview data, increases in intimacy and decreases in shame appeared to be less associated with participants’ attachment styles and more closely related to their attachment-based strategies [30]. Specifically, individuals who tended to withdraw emotionally during periods of emotional arousal—whether their attachment style was avoidant or anxious—often experienced increases in both shame and intimacy following the intervention. In contrast, those who tended to use self-attacking strategies more commonly reported decreases in shame and increases in intimacy [64, 69].

These findings may help explain the contradictory and often puzzling outcomes observed in treatments involving individuals with avoidant attachment styles. Overall, a primary therapeutic task is to help withdrawn individuals remain engaged in treatment and become aware of their shame [53]. Emotionally Focused Couple Therapy then facilitates the resolution of shame as a gateway to accessing and processing other core emotions [33, 40]. Even in cases where the stabilization phase had been completed, some couples reported improved emotional tolerance and effective communication in high-tension situations due to emotional awareness and processing. It is natural for clients to initially feel a temporary decrease in intimacy as they move through emotionally painful material; however, this process is often a necessary passage toward deeper connections.

Cultural issues

Notably, shame and intimacy are both culture-dependent variables and are sensitive to environmental differences. Therefore, the data should be interpreted conservatively. Dissociation from groups that you belong to and not meeting standards would mean inconsistency or inadequacy and might provoke shame [78]. A lack of shame awareness might lead to defence mechanisms that can compromise mental health [79].

In Iran, as a country with a collectivist culture, having intimate relationships is assumed to be an important priority in quality of life. In contrast, owing to the essence of technology and social growth, close relationships are rare but vital. As a result, maintaining a relationship might require more concealment of the self and the suppression of needs [80]. On this basis, a long-term decrease in intimacy is a concern; however, research has indicated that gender-role ideology affects intimacy more than culture type does [81]. Nevertheless, it can be implied that gender roles and cultural types should be noted mutually. Given that romantic love is not still remarkable in collectivist societies, the scale used in this study could not offer notable power for intimacy differences.

Limitations and future research

The first limitation of this research is the small sample size, which constrains the generalizability of the findings. In addition, the use of self-reporting tools was another limitation of the research. Although an interview about attachment, relationships, and experience of shame was performed, the results might not be capable of strong assumptions due to the number of samples and could be used as a qualitative study in another paper. Therefore, employing implicit tools for assessing shame and intimacy in a larger number of samples in future research is suggested. Furthermore, it should be assumed that a control group with no interaction with the therapist could clarify the results. Future research could use three groups to examine this more accurately.

Additionally, the data collection and examination stages were two serious limitations in this study. All therapy sessions and completing scales were online. Therefore, the data were collected online because of the COVID-19 epidemic, and some data were missed or might have been collected with less accuracy. Conducting research with a situation of completing the forms in the presence of the therapist or researcher and more follow-ups will lead to results with greater accuracy and explanatory power.

In addition, randomization was performed to divide couples into groups, and the first stage of sampling was voluntary. On the basis of our observations, most of the couples were nonclinical but severely distressed or insecurely attached, were not motivated and were only trying their last chance in free therapy. Hence, performing research with complete randomization would deepen these findings.

Despite the random assignment of participants to the intervention and control groups, a significant difference in age was found between the two groups. The interaction between age and group was analysed in the ANCOVA model and found to be nonsignificant, indicating that the age difference did not affect the treatment outcome. However, given the relatively small sample size, this imbalance should be considered a potential limitation of the study.

Another important limitation is the lack of a fidelity check with standard ratings via voice or video of therapy sessions. While a formal fidelity rating tool was not used in this study, therapist fidelity was monitored through multiple strategies, including session documentation, structured supervision, and treatment planning grounded in EFCT principles [82]. Although the therapist had completed Core Skills training, they were not certified in the EFCT. This limits the assurance that treatment fidelity was consistently maintained throughout the study. In addition, while inter-rater reliability was not formally calculated for the shame assessment, consensus was reached through joint session reviews. Moreover, the same therapist conducted both arms of the intervention, which may have introduced a confound, as therapist-related factors (e.g., style, rapport, skill) could have influenced outcomes independent of treatment modality.

Finally, this research was a clinical trial and could not examine other factors that might mediate or moderate the relationship between shame and intimacy. While few studies have addressed this issue, it should be investigated in the future. In addition, dyadic analysis was planned to be performed. The correlations between the actor shame score and the partner shame score (r ₍₅₀₎ = 0.05, p > 0.05) and intimacy score (r ₍₅₀₎ = 0.16, p > 0.05) were not significant for the dyadic template.

Clinical implications

The results of this study have several important clinical and theoretical implications. For the first time, specific shame coping styles have been implicated in the development of an effective therapeutic alliance and intimate relationship functioning. Early interventions to identify unhelpful shame coping styles may ultimately help resolve some of these relationship problems.

One important clinical implication of this study concerns how shame can be addressed without reinforcing or reactivating shame in the client. Therapists working within the EFCT framework must remain attuned to subtle signs of shame—such as withdrawal, silence, or emotional numbing—and respond with validation, pacing, and empathic curiosity. Clinicians may support clients by reflecting secondary emotions, tracking nonverbal cues, and gradually accessing primary emotions in a safe and non-pathologizing manner. This study underscores the importance of creating an emotionally secure therapeutic alliance where shame can be processed without blame, judgment, or exposure, aligning with EFCT’s emphasis on emotional safety and attachment repair.

Conclusion

The findings of this study can be useful for couples and family therapists. Emotionally focused Couple Therapy can be used as an effective treatment method for couples who are experiencing shame and intimacy issues.

Ultimately, the results of this research revealed that paying attention to the therapeutic relationship, with a focus on shame, leads to a reduction in shame in different dimensions and an increase in the client’s potential for intimacy. Adding this part to emotion-based therapies, such as Emotionally Focused Couple Therapy, which was used in this research, is suggested because of the strong connection between attachment, shame, and emotion regulation in individual and couple therapies.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

EFCT

Emotionally focused couple therapy

EFT-C

Emotionally focused therapy for couples

ESS

Experience of shame scale

PAIR

Personal assessment of intimate relationships

RCI

Reliable change index

Author contributions

Z.M: Data collection, Data analysis, Writing report J.M: Research conceptualization, Advisor, and Observer H.H: Research conceptualization, Advisor, and Editor M.R.T: Research conceptualization, Advisor, and Editor.

Funding

This research has received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data cannot be provided now. It would be available after acceptance and in confidential circumstances.

Declarations

Ethics approval and consent to participate

The research started after the code was coordinated and obtained in accordance with the Declaration of the Ethics Committee for Research on Human Subjects at Shiraz University. This study was reviewed and approved by the Ethics Committee for Research on Human Subjects at Shiraz University (IRB number: IR.US.REC.1401.034). All participants provided written informed consent before their inclusion in the study. Informed consent to participate was obtained from all of the participants in the study. Participants entered the research after completing the consent form.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.