Attachment, Jealousy, and Conjugal Violence

Abstract Psychological aggression is the most common manifestation of intimate partner violence. Literature demonstrates that attachment and jealousy can predict intimate partner violence. The aim of this study was to test a theoretical model in which jealousy is a mediator between attachment and psychological aggression. This is a quantitative study, with cross-sectional and explanatory design. A sample of 600 people, aged between 18 and 65 years old, residing in 13 Brazilian states, was investigated. A Sociodemographic data questionnaire, Psychological Aggression Sub-scales (CTS2), Questionnaire on the Affective Relationships (QAR), Interpersonal Jealousy Scale (IJC) and Adult Attachment Scale (AAS 2) were used. Structural equation modeling indicated that jealousy functioned as a mediator between attachment and psychological aggressions, alerting to the possibility of prevention in conjugality through early interventions.

Psychological aggression is considered as any act or omission that causes or aims to cause damage to self-esteem, identity or to development (Secretaria de Políticas de Saúde, 2001), which includes threats, humiliations, blackmail, behavior demands, discrimination, exploitation, criticism of sexual performance, provoking isolation or preventing the use of one's own money.Thus, it causes fear, dread, and possibly rejection, depreciation, discrimination, humiliation and disrespect (Dokkedahl et al., 2019;Jarnecke& South, 2021).
Several researches have been carried out about intimate partner violence in its different forms of manifestation and jealousy is commonly found as a factor associated with its occurrence (Brem et al., 2018;Centeville& Almeida, 2014), including jealousy in virtual social networks (Daspe et al., 2018).Jealousy is understood as a threat related to fear of loss (real or imaginary) of the partner, which is influenced by feelings (such as possession, love and competitiveness).It is understood that jealousy is circumscribed to a certain cultural, social and historical context (through experiences in the family of origin, the romantic love model, for example), occurring in an individual or relational perspective.Jealousy may lead to the occurrence of violence, at a psychological, physical or sexual level.
A study conducted by Costa et al. (2016), with 264 Brazilian college students who heard two audios, one with and the other without a discussion sparked by jealousy, indicated that people rated it as more morally inappropriate when the husband had been aggressive in the discussion that did not involve jealousy.Thus, the authors identified that university students tend to minimize the meaning of aggression committed in situations involving jealousy as compared to aggression in situations that do not involve jealousy.In contrast (STILL), a study conducted in Nicaragua, with 199 men and 201 women, to evaluate predictors of violence between intimate partners, identified that possessive jealousy was not associated with violence (Buunk&Massar, 2019), evidencing inconsistencies in the literature.Jealousy is socially interpreted as a form of caring (Zampiroli, 2017) and this kind of understanding minimizes the severity of its manifestations, such as controlling the partner's behavior.
Early experiences of violence in the family of origin are also understood as factors that lead to minimizing the severity of aggressions in intimate relationships.Several studies point to this relationship between experiences in the family of origin and intimate partner violence (Colossi, Marasca & Falcke, 2015;Neppl et al., 2019).Some consider that experiencing the context of violence in the family of origin can lead to the tendency of repetition of patterns (Razera et al., 2014) and even that people with a history of violence experienced in the family of origin are more predisposed to perceive violence as justifiable in the marital relationship, naturalizing the phenomenon (Temple et al., 2013).
The attachment theory describes the quality of the child's interactions with their caregivers in the family of origin, helping to shape expectations or work models of future relationships, being perpetuated across generations.In the same way as in childhood, the importance of attachment in love relationships spawns from a protective function, when in times of difficulties there is, for example, regulation of affection, mutual support and solidarity (Semensato& Bosa, 2013).Studies indicate that attachment styles persist in adulthood (Scheeren et al., 2015;Sommer et al., 2016) and that attachment influences affections, cognitions and behaviors in intimate relationships.Bowlby (1998aBowlby ( , 1998b) ) highlights three types of attachment: Secure Attachment, Anxious-Resistant Attachment, and Anxious-Ambivalent Attachment.Secure Attachment develops when caregivers are perceived as available and sensitive to needs.In this way, people with Secure Attachment believe that others can be trusted and feel loved, as their attempts at closeness are met and so they do not worry about the possibility of abandonment.Anxious-Resistant Attachment occurs when there is little sharing of affection.After separations, even when brief, the caregiver is avoided and, when the caregiver is present, there is lack of confidence and doubt about whether they will meet the needs.In love relationships, individuals with this type of attachment tend to experience lower levels of trust, satisfaction, intimacy, and stability (Kirkpatrick & Davis, 1994).In Anxious-Ambivalent Attachment, the caregiver is available to help in some situations and not in others, generating threats of separation and abandonment.Thus, the individuals believe that they are not worthy of love, worry that they cannot count on the caregiver if it were necessary and are afraid of rejection (Bowlby, 1998a(Bowlby, , 1998b)).In these cases, once they become adults, they tend to suspect that their partner may have another person, increasing vigilance and threats, monitoring the behavior of the partner, intensifying the fear, anger and sadness, with tendencies to develop pathological jealousy (Guerrero, 1998) and situations of intimate partner violence (Becker &Crepaldi, 2019;Sommer et al., 2016).
A study conducted in Brazil, with 428 men and women, with the goal to evaluate the predictive role of attachment styles in the resolution of marital conflict, found that insecure attachment styles were predictors of a destructive style in conflict resolution.On the other hand, secure attachment was a predictor of constructive conflict resolution (Scheeren et al., 2015).Within this direction, an explanatory study, with 2,500 university students in the United States (Gover et al., 2008), pointed out that exposure to violence during childhood was a predictor for involvement in violent relationships.In this sense, it is worth mentioning that the relationship established with the maternal and paternal figures is important for the quality of the conjugal bond.Follingstad et al. (2002) developed a structural equation model to test the relationship between ambivalent attachment, anger, and attempts to control the partner as predictors of the severity and frequency of violence in dating.In a sample of 412 university students (roughly balanced by gender), 80 with a history of violence, ambivalent attachment was related to the occurrence of violence through the variables anger and control, which are commonly associated with jealousy.Mauricio et al. (2007) examined 192 men with warrants to participate in an aggressor intervention program and who completed measures for insecure attachment.The authors used structural equation modeling to test the hypotheses that the relationships between ambivalent attachment and physical and psychological violence were mediated by personality disorders.The results indicated that personality disorders entirely mediated the relationship between ambivalent attachment and physical violence and partially mediated the relationship between ambivalent attachment and psychological violence.In the same direction, the study by Sommer et al. (2016), with 163 couples, examined the relationship between attachment and violence.The results indicated that ambivalent attachment was associated with the perpetration of physical violence and sexual coercion.
In this sense, considering that attachment experiences impact on development, being reflected in marital relations, the aim of this study was to test an explanatory model in which attachment can be associated with the occurrence of psychological aggression, being mediated by the presence of jealousy in the marital relationship.

METHOD Participants
In this study, which was quantitative, cross-sectional, and characterized by explanatory design, the participants were 384 women and 216 men (N=600), in a heterosexual relationship, with ages ranging from 18 to 65 years (M= 31.69,SD= 9.74).The sample consisted of a non-clinical population, the relationship time varied between 1 and 40 years (M= 8.60, SD= 8.30) and the participants were residents of 16 Brazilian states (Bahia, Ceará, Goiás, Mato Grosso do Sul, Minas Gerais, Paraíba, Paraná, Pernambuco, Piauí, Rio de Janeiro, Rio Grande do Sul, Rio Grande do Norte, Rondônia, São Paulo, Santa Catarina and Distrito Federal).Table 1 shows sociodemographic characteristics of the sample.

Instruments
Sociodemographic data questionnaire.Composed of 13 items to map the characteristics of the participants.
Psychological Aggression Sub-scales CTS2-Revised Conflict Tactics Scale (Straus et al., 1996).It is a scale to evaluate tactics for conflict resolution; the subscale used verifies psychological violence in love relationships.It is presented as a checklist and the measure is divided into minor (insults, curses, screams, offenses, threats and turning your back in the middle of a fight) or severe (offending from ugly, fat, "bad in bed" or something like that, destroying a companion's personal object).The Brazilian version, adapted and translated by Moraes et al. (2002), was used.The following alphas were obtained: 0.61 severe psychological aggression and 0.82 minor psychological aggression.
Questionnaire on the Affective Relationships (QAR) (Marazziti et al., 2003) aims to measure the occurrence of behaviors resulting from thoughts related to jealousy and to distinguish normal from obsessive jealousy.It consists of 30 items on a likert scale of four points.The translated version used was adapted by Costa et al. (2013).In this study, the alpha of 0.82 was obtained.
Interpersonal Jealousy Scale (IJC) (Mathes & Severa, 1981).This instrument aims to measure predictive factors of jealousy, intensity and beliefs related to jealousy, self-esteem, romantic love and insecurity.It has 28 questions, and the answers vary on a nine-point Likert scale.The translated version used was adapted by Costa et al. (2013).In this study, the alpha of 0.93 was obtained.
Adult Attachment Scale (AAS 2) (Collins & Read, 1990).Created with the aim of identifying the three Bowlby bonding patterns (Secure Attachment, Anxious-Resistant Attachment, and Anxious-Ambivalent Attachment) developed by adult individuals in the relationship with their partners, contemplating the bonding patterns experienced in childhood, it consists of 18 items, on a five-point Likert scale.The following alphas were obtained: 0.84 anxious-ambivalent, 0.69 anxious-resistant, 0.61 secure.

Procedures for Data Collection and Analysis
This research followed the guidelines of resolutions 466/2012 of the Brazilian National Health Council (Conselho Nacional de Saúde, 2012) on researches with human beings.It was approved by the Ethics Committee under protocol 13/182.The study was conducted through an online questionnaire, which was made available to participants on an internet address.The data relating to the identification of all participants of the survey, such as I.P. address and email were kept confidential.
The sampling process involved the sending of invitations by e-mail and social networks, informing users of the objectives and procedures of the research, as well as the confidentiality of the data and the anonymity of their collaboration.The sampling criterion was also considered "snowball effect", since participants were asked to forward the invitation to their acquaintances via email or messages.
Participants had access to the free and informed consent form (ICF) and answered the online questionnaire according to gender to make reading easier.After the collection, a database was created in the program SPSS 22.0 (Statistical Package for Social Science).The data were analyzed using descriptive statistics to calculate means and standard deviation, and Pearson correlation analysis.Following that, the AMOS 20.0 program was used, in which confirmatory factor analyses were performed and, from the correlation analyses obtained, structural equation modeling (SEM) was performed.
The proposed model was composed of three latent variables: attachment (independent and exogenous variable), jealousy (dependent, mediator, endogenous and exogenous variable) and psychological aggression (dependent, endogenous variable).The model used is recursive (Hair et al., 2009), having as peculiarity the absence of covariance between disorders of endogenous variables and the unidirectionality of determination relationships between the variables.
The estimation method used was of Maximum Likelihood (MLE), because it estimates the parameters that maximize the likelihood of observing the sample covariance matrix (Hair et al., 2009).The adequacy analysis of the structural equation model was based on the following general adjustment measures (Hair et al., 2009): a) absolute adjustment measures -Chi-Square (χ 2 ), statistical significance of χ 2 (p), root of mean square error of approximation (RMSEA); b) incremental adjustment measure -Tucker-Lewis Index (TLI); c) parsimonious adjustment measure -Normed Fit Index (NFI), Confirmatory Fit Index (CFI) and weighted Chi-square (χ 2 /df).

RESULTS
The results indicate that 81.5% of the sample reported perpetrating psychological aggressions of minor intensity, through insults, cursing, screams, threats and others.Behaviors such as offenses, destruction of objects, among others, which are characterized as severe psychological aggression, are described by 29.8% of the sample.Correlations between all variables alternate between moderate and weak and can be observed in Table 2.
With regard to SEM, the structural model refers to the existence of relationships between constructs (Hair et al., 2009).Thus, using significance and magnitude analyses of correlations, it was possible to identify which observable variables (predictors, mediators and outcomes) would make the composition of the latent variables of the model.The proposed structural model was based on the hypothesis that jealousy mediates attachment and the occurrence of psychological aggressions.The model adjustment results are shown in Table 3.
Model 1 presented inadequate results for a structural equation model, because it revealed insufficient TLI and NFI values.For Model 2, the prediction arrow of the relationship between attachment and psychological aggression was removed, considering that the data were not significant (β=-0.022,p=0.76).For Model 3, the covariance between secure and anxious avoidant attachment was adjusted.With these adjustments, the model presented satisfactory results.The final model can be observed in Figure 1.
When examining the standardized regression coefficients of the model, it is identified that the magnitude of the impact of attachment on jealousy is moderately negative (β=-0.63,p<0.001), which, in turn, moderately and positively impacts psychological aggression (β=0.52,p<0.001).Thus, according to Hair et al. (2009), the analyses point out that jealousy is a total mediator between attachment and psychological aggression.

DISCUSSION
The data on how much the participants reported perpetrating psychological aggressions found in this research are greater than those found in studies from different countries (Montesino et al., 2014;Wolford-Clevenger et al., 2015) and quite similar with other Brazilian studies (Colossi & Falcke, 2013;Colossi, Marasca & Falcke, 2015;Madalena et al., 2015;Wagner et al., 2015).The investigation carried out by Colossi, Razera et al. (2015) presented higher indices than those observed in this study with respect to psychological aggression of minor intensity (86.5% for women and 85.6% for men) and lower indices than the findings in this study in severe aggression (25.9% for women and 16.1% for men).Considering that constant offenses generate severe emotional wounds, disturbing self-esteem, safety and confidence (Martins et al., 2014), it is necessary to invest in more space for discussion, prevention and creation of public policies to confront this type of violence.
Among the types of violence, psychological aggression is the most difficult to identify and situations of aggression of this type can occur for a long time, causing emotional impact (Colossi, Marasca & Falcke, 2015), such as anxiety, depression, and impulsivity (Costa, 2010;Priolo Filho et al., 2019), which may even lead to the occurrence of suicide (Secretaria de Políticas de Saúde, 2001).In addition to direct combating situations of violence in love relationships, attention should also be paid to the origins of violent behavior.
A secure attachment contributes to the development of more lasting interpersonal relationships throughout life (Fearon et al., 2010).Dutton and White (2012) report that insecure attachment is the greatest predictor of intimate partner violence, as it contributed to fear and distrust in relationships, preventing the development of self-efficacy, intimacy, positive social expectations and ability to emotional regulation (Fearon et al., 2010).
The data of this study corroborate the data of the studies by Follingstad et al. (2002), Mauricio et al. (2007) and Sommer et al. (2016), who found relationships between attachment and violence.It may be thought that attachment can generate constant insecurity between the desire to maintain the love relationship and the constant anticipation of the fear of rejection, of being exchanged for a supposed rival.This fear can be externalized through jealousy, which leads to the occurrence of episodes of psychological aggression.
It is also worth noting that it is difficult to assess when jealousy can be considered pathological, but it becomes relevant to demystify the commonsense idea that jealousy is a manifestation of care, affection or love (Mendes & Claudio, 2010).It is understood that jealousy cannot mask dysfunctional behaviors, through its naturalization and, even less, minimize the impact of the occurrence of episodes of aggression, as in the study by Costa et al. (2016) which identified that in the presence of jealousy college students minimize the significance of aggression.

FINAL CONSIDERATIONS
Previous studies have already highlighted how much the experiences in the family of origin and insecuritypromoting contexts constitute precipitators of violence in love relationships.In this research, the role of jealousy as a mediator between attachment experiences based on early bonds and the occurrence of psychological aggression in love relationships is highlighted, warning of the possibility of preventive interventions of intimate partner violence, acting in cases where there is the presence of jealousy.
In this study, jealousy appeared as a result of attachment experiences and act as a bridge to the occurrence of violence.Therefore, interventions should be planned in order to strengthen attachment bonds, which, more securely, tend to decrease the manifestations of jealousy and, consequently, psychological aggression.
The results obtained bring important contributions, also relevant for future research.Its cross-sectional design does not allow access to the understanding of underlying psychological processes, which constitutes a limitation.Longitudinal studies can contribute to a better understanding of the interaction between evolving variables.In addition, a replica of this study with data related to marital dyad and covering gender issues can provide more subsidies for intervention with couples in situations of violence, considering that, with this, it may be possible to identify complexities of the dyad and promote more functional strategies for conflict resolution.Furthermore, it is necessary to deconstruct the socially widespread idea that jealousy is linked to proof of love, care or affection, as well as that marriage can be the salvation for family stories of emotional deprivation.Such conceptions generate suffering and possible deleterious reverberations for the relationship, such as affective dependence and violence.In this sense, it is understood that paying attention to the early manifestations of jealousy in relationships can contribute to minimizing the occurrence of violent behavior.

Table 1
Socio-demographic characteristics of the sample.

Table 2
Correlation between all variables.

Table 3
Model adjustment test.