Despite IWMs being relatively sensitive to the changing quality of relations with caregivers in the first years of life, as the years go by, these become increasingly stable and eventually remain available throughout the entire life span. According to this framework, as a core component of personality, IWMs would impact a range of psychological functioning domains such as emotion regulation capacities or interpersonal functioning (Bowlby, 1969, 1973). A key life domain that is greatly impacted is romantic interpersonal functioning (Knies et al., 2021), with several authors stating that the quality of romantic relationships is rooted in the vicissitudes of early attachment experiences (Owens et al., 1995; Roisman et al., 2005; Velotti et al., 2014).

Although the question of the continuity of IWMs from childhood to adulthood is still under debate, a consensus has been reached towards the utility of investigating the topic of romantic relationships through the lens of the attachment theory (Hazan and Shaver, 1987; Crowell et al., 1999; Simpson and Rholes, 2012; Gray and Dunlop, 2019). Indeed, it has been stated that the attachment system would drive the individual to establish attachment bonds also with extra-familial significant others being typically friends (in adolescence) and romantic partners (in late adolescence and adulthood) (Berlin et al., 2008). The psychological functions of the old attachment figures (e.g., the parents) would be carried out by a new significant other (e.g., the romantic partner) who would be expected to satisfy the individual’s attachment needs. Again, the implicit interpersonal knowledge regarding the ways the individual’s attachment needs are framed and satisfied within this specific relationship would shape expectations towards the self, the other, and the individual’s emotional experience in the relationship. Because this specific IWM – referred to the specific romantic partner – will structure the individual’s behaviors in this current romantic bond it is expected to greatly impact on a wide range of outcomes related to this relationship.

The strong theoretical framework advancing the idea that attachment might be a key construct in the understanding of romantic relationships led several authors to develop research tools to investigate the topic. In line with the general trend in the field of research on attachment, most authors preferred the use of self-report questionnaires to assess romantic adult attachment styles. However, the Current Relationship Interview (CRI), a well-known tool to evaluate romantic IWM has been developed by Crowell and Owens (1996) to grasp the current attachment representations and stimulated the whole field of research (San Martini and Zavattini, 2011). Despite the soundness of the theoretical framework underlying the instrument and the spread of its use in several countries, it is noteworthy that no data have been published regarding the properties of the Italian version of the interview to date. An additional gap in the existing literature is related to the scarcity of data brought by the scientific community regarding the predictive value of romantic attachment, measured with the CRI, and some key outcome variables related to romantic relationship functioning. In particular, as better illustrated in the paragraph below, evidence is lacking regarding the capacity of romantic attachment – as measured by the Italian version of the CRI – to predict the perception of dyadic adjustment and styles of conflict resolution with the romantic partner.

Romantic relationship quality has a relevant impact on individuals’ wellbeing. For individuals involved in an intimate relationship, such dimension assumes a central role in their life, being either a resource and/or a source of significant stress (Velotti et al., 2013; Farero et al., 2019). A plurality of terms have been used to refer to romantic relationship wellbeing including dyadic/marital adjustment (Locke and Wallace, 1959; Spanier, 1976), an umbrella term that describes the wellbeing of the relationship as an entity (Farero et al., 2019). This refers to both overall wellbeing in a relationship along with specific components (cohesion, consensus, affective expression, and satisfaction) related to the resolution of relationship difficulties (Spanier, 1976, 1979; Busby et al., 1995). The Dyadic Adjustment Scale (DAS; Spanier, 1976) is one of the most widely utilized self-report measures in clinical and research settings for measuring relationship wellbeing (Carey et al., 1993; Sabourin et al., 2005; Herrington et al., 2008; South et al., 2009). Theoretically, dyadic adjustment levels are expected to be associated with a positive, secure attachment to the partner. More precisely, these levels are expected to vary according to the impact of the individuals and their partner IWMs as well as according to the impact of the interaction between these two components (Velotti and Zavattini, 2011). Regarding this last factor, contributions highlight that beyond the quality of the attachment bond, the matching status between attachment types across the partners (i.e., being both secure, both insecure, or being mismatched), may have relevant implications for the relationship functioning (Simpson, 1990; Strauss and Morry, 2012; González-Ortega et al., 2017; Velotti et al., 2022; Cataudella et al., 2023). Empirically, few but promising pieces of research supported this perspective showing that security in attachment to partner significantly and positively predicted dyadic adjustment during the transition to parenthood (Velotti et al., 2011; Castellano et al., 2014). Of note, despite the predictive role in longitudinal studies, the cross-sectional association of security of romantic attachment as measured by the Italian version of the CRI with dyadic adjustment levels has not been tested yet.

Another central feature of romantic relationship functioning is related to the way conflicts are experienced and managed in the couple. Conflict is a natural outcome of interpersonal interactions when the parties perceive themselves as being in opposition to each other. This opposition may involve preferred outcomes, attitudes, values, and behaviors (Elsayed-Ekhouly and Buda, 1996; Chakrabarty et al., 2002). The growing interest in research on couple conflict stimulated the development of a plurality of assessment tools. For instance, the Rahim Organizational Conflict Inventory questionnaire, initially developed for and validated in organizational contexts, has been used in research on close relationships (Hammock et al., 1990; Castellano et al., 2009). Of note, IWMs may be especially influential for how partners perceive and respond to conflict (Lin, 2003; Mikulincer and Shaver, 2005; Schudlich et al., 2013). Indeed, in couples, the different points of view of each partner can lead to the onset of friction as well as explicit conflicts that affect individual and interpersonal wellbeing (Castellano et al., 2009). There may be important differences between securely and insecurely attached individuals in their approach to managing conflict within romantic relationships. Insecure attachment has been associated with relatively less adaptive or supportive conflict tactics and behaviors (Kobak and Hazan, 1991; Cohn et al., 1992; Cowan et al., 1996; Simpson et al., 1996; Guerrero et al., 2009; Sierau and Herzberg, 2012; Ricco and Sierra, 2017). However, empirical evidence regarding the link between attachment security to partner measured with the CRI and conflict resolution strategies is still lacking. Roisman et al. (2005) observed that security levels assessed with the CRI significantly discriminated between couples showing a good relationship quality in a conflict task and couples being rated with a poor relationship quality. Also, levels of coherence of the CRI transcripts resulted to be positively and significantly associated with the dyadic capacity to solve conflict assessed through an observational measure (Haydon et al., 2012). Moreover, the security of attachment to a partner measured with the CRI has been negatively correlated with the frequency of marital disagreements over the last week (Waldinger et al., 2015). An Italian study found that insecurity predicts lower levels of adaptive conflict resolution strategies during the transition to parenthood such as the cooperative integration strategy (Castellano et al., 2014).

As briefly illustrated, despite the utility of the attachment theory in explaining romantic relationship-related processes and the uniqueness of the tool developed by Crowell and Owens (1996) – the CRI – information regarding the properties of its Italian version is lacking. In addition, the promising empirical evidence regarding the predictive role of attachment security as measured by the CRI and both dyadic adjustment and conflict resolution strategies profiles still need to be replicated and extended by additional studies. The present study aims to partially fulfil these gaps by administrating the Italian version of the CRI to a sample of adults and measuring its predictive role towards these two outcomes.

The study involved a total of 102 adult Italian heterosexual couples (50% males) with a mean age of 36.42 years (S.D. = 4.69). Regarding education levels, 39.4% of the sample report to have a middle and/or high school degree with the others having a university degree. Most couples were married (71.4%) and with children (84.1%). The group was drawn from a normal population and all participants were recruited through a convenience sampling technique. Specifically, the study was promoted thought university announcements, and students of psychology courses as well as colleagues of researchers were asked to promote the study in their networks of friends and family. Also, after the procedure was completed with a couple, participants were asked to promote the study in their personal networks.

Inclusion criteria were the following: (a) to be more than 18 years old; (b) to have a good understanding of Italian language; (c) to be involved in a romantic relationship for at least 1 year. Exclusion criteria were the following: (a) to suffer from an acute psychotic episode; (b) to have been severely intoxicated by substance or alcohol intake within the past 3 months; (c) to suffer from a neurological disease; (d) to have received a diagnosis of cognitive deficit.

Before the involvement of each participant in the research procedure, research’s aims and scopes were briefly exposed and information on privacy and anonymity was delivered. Upon reading and approval of the informed consent, participants were asked to fill several self-report questionnaires under the supervision of a psychologist. Then, a semi-structured interview was administrated and audiotaped. No compensation was given. All procedures complied with the official directions established by the American Psychological Association.

Participants completed an initial survey (information sheet), created specially for the purpose of the study, collecting demographic information such as age, gender, educational levels and profession, marital status, and family situation.

Afterward, the nature of romantic attachment representations was investigated with the Italian version of the Current Relationship Interview (CRI; Crowell and Owens, 1996; Santona and Zavattini, 2007). The Italian version of the CRI is available upon request from the corresponding author. The purpose of the interview is to reveal how participants mentally represent attachments in romantic relationships, as reflected in their manner of speaking about their relationship. The questionnaire was built using the Adult Attachment Interview (AAI) as a model; it consists of 15 questions, the formulation of which considers the reciprocal nature of adult romantic relationships. The interview contains questions about the participant’s dating history; the nature of the present relationship and characteristics of the partner; and routine behaviors within the relationship, especially those related to providing and seeking support from the partner. To elicit an overview of the relationship, questions include topics such as what they have learned from each other and their hopes and concerns about the future of the relationship. The evaluation is based on the transcript and allows the individual to be placed in two groups: Secure (S) and Insecure depending on the profile that emerges from the scores (from 1 to 9) obtained on 18 evaluation scales. The classification system of the CRI distinguishes, within the group of individuals with insecure attachment, those who avoid talking about attachment in terms of a secure basis or who devalue it (distancing/devaluing profile - D), those who, instead, place a particular emphasis on these aspects and compulsively try to control them (preoccupied profile - P) and finally those who fall into the Unresolved category (U). Importantly, despite the administration of the interview does not require a specific training, the coding of the transcripts does. In the current study, both the administration and the coding of the transcript were performed by trained researchers.

Participants were then asked to fill two self-report questionnaires.

The way in which people tend to respond to interpersonal conflicts have been investigated using the Rahim Organizational Conflict Inventory questionnaire, Section II (ROCI II; Rahim and Majer, 1995). The questionnaire consists of 28 items on a 5-point Likert scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree) investigating the modalities of discussion with the other according to five independent responding styles: Collaborating Style, Accommodation Style, Competing Style, Avoidant Style, Compromising Style measured by 7, 6, 5, 6 and 4 statements, respectively. The internal consistencies of the scale was 0.71.

Romantic relationship quality was assessed using the Dyadic Adjustment Scale (DAS; Spanier, 1976; Gentili et al., 2002). This is a 32-item rating instrument that may be completed by either one or both partners in a relationship. Respondents are asked to rate each of the items on different Likert-type scales choosing the most suitable response options by indicating the extent of agreement or disagreement between the individual and his/her partner for each item. The most useful way of interpreting DAS is through the subscale scores: Dyadic Consensus (13 items; the degree to which the couple agrees on matters of importance to the relationship), Dyadic Satisfaction (10 items; the degree to which the couple is satisfied with their relationship), Dyadic Cohesion (5 items; the degree of closeness and shared activities experienced by the couple), and Affective Expression (4 items, the degree of demonstrations of affection and sexual relationships). A person taking this test can obtain a score from 0 to 151. The lower scores on DAS are indicative of having a problem, while the higher the score the better the person’s adjustment to the relationship. The instrument provides an estimation of the dyadic adjustment level perceived by each partner separately and by the couple (i.e., the average score obtained by each partner). In our study, the reliability of the instrument was confirmed by Cronbach’s α, with values ranging from 0.63 to 0.93.

Cronbach’s alphas were computed to explore the internal consistency of continuous measures. Then, descriptive analyses were carried out namely frequencies, means, and standard deviations. Comparisons between groups were performed through Kruskal-Wallis tests and chi-square difference tests. All statistical analyses were performed using SPSS v.25 software for Mac.

Differences in dyadic adjustment levels according to romantic attachment were investigated following complementary approaches. First, we tested the hypothesis that individual perceptions of dyadic adjustment would differ according to individual romantic attachment type in the whole sample and in men and women separately. Then, we explored whether individual dyadic adjustment differed according to the partners’ romantic attachment style. Afterward, we explored whether the dyadic adjustment differed according to individuals’ romantic attachment type. Lastly, we tested the hypothesis that both individuals and dyadic estimation of adjustment would differ according to the matching between romantic attachment styles between the partners.

As displayed in Table 2, in the whole sample, individuals classified with a preoccupied attachment compared to individuals with a secure attachment obtained lower scores in the general level of dyadic adjustment. Also, men with a preoccupied attachment, compared to those with a secure attachment, obtained lower scores on the cohesion subscale of the DAS. No other significant differences were identified.

Then, we investigated if women and men DAS scores significantly differed according to their attachment types. As displayed in Table 3, no significant differences were observed.

Also, the hypothesis that dyadic evaluation of adjustment would differ according to attachment type was tested in the sample of women and men separately. Results, illustrated in Table 4, did not identify any significant difference.

Lastly, we explored the possibility that matching status between attachment types (either matching or no matching) would have discriminant between dyadic adjustment levels. This hypothesis has been tested regarding couple levels of dyadic adjustment as well as dyadic adjustment as perceived by men and women separately. As displayed in Table 5, we found that women involved in a relationship characterized by matching attachment types between partners reported significantly lower levels of consensus compared to women involved in relationships characterized by an attachment type mismatch. To further deepen this result, we test the presence of significant differences on cohesion levels perceived by women, between couples matched on secure (n = 17) versus insecure (n = 9) attachment types. However, no significant difference emerged (p = 0.839).

To test whether the distribution of constructive versus destructive conflict resolution strategies differed according to attachment type, a series of chi-square tests were performed. First, we found that conflict resolution style was not significantly more frequent among individuals with a secure, dismissing, or preoccupied attachment towards their partner. This result was replicated in the whole sample (p = 0.261), among men only (p = 0.715), and among women only (p = 0.115).

Despite the value of this study, the reader should appreciate its results in light of some important limitations. First, we used a convenience sampling procedure, therefore self-selected participants may not be fully representative of the whole population of Italian couples, limiting the generalizability of our findings. This is especially true for couples seeking psychotherapy, which may imply specificities related to their clinical conditions. In addition, the small sample size limits the heterogeneity of variables investigated, reducing in turn the type of analyses that we were able to perform. For instance, only one participant was classified as having an “other” type of attachment model and this category was therefore excluded from further analyses. Also, the poor heterogeneity regarding conflict resolution strategy types did not allow for testing more complex hypotheses regarding this variable. Another issue is related to the absence of measurement of some variables that may moderate the relationships observed. For instance, emotion regulation capacities are considered to be tightly related to the attachment model and may greatly impact the outcomes measured and especially the type of conflict resolution strategies (Halperin, 2014; Garofalo et al., 2016). Lastly, the use of self-report measures to assess both dyadic adjustment and conflict resolution strategies may be considered a limitation of the study. Indeed, most of the contrasting results brought by past studies employed observational measures. This methodological issue may be further investigated in future studies to test whether the type of instrument measuring these variables may significantly impact the estimation of their link with current attachment types.

As a whole, this study highlights the complexity of the issue regarding the impact of current attachment in romantic relationships and calls for future studies investigating the topic through the CRI.