Relational Aggression Targeting the Parent-Child Bond: A Pilot Study of Parent-Child Bond-Breaking Behavior

ABSTRACT

Relational aggression involves targeting and weaponizing relationships. Similarly, what is currently called parental alienation (PA) behavior, concerns targeting and weaponizing the parent–child relationship in the context of separation and divorce by one parent in order to disrupt the bond between the child and their other parent. In this study we report results from 46 separated and divorced parents evaluating a new parent-child bond-breaking measure and examining associations among parents’ attachment, romantic relational aggression, and bond-breaking. Recognition of parent-child bond-breaking as relational aggression links it to a robust scientific literature base, enhances measurement, and opens new lines of scientific inquiry.

KEYWORDS:

• Parental alienation
• parent-child bond-breaking
• relational aggression
• romantic relational aggression
• divorce
• coercive control

In this manuscript, we introduce a new measure of parent–child bond-breaking behavior and demonstrate, conceptually and empirically, that the behaviors that have thus far been considered under the umbrella of “parental alienation” are better understood within the wider and deeper literature of relational aggression and illustrate how that the conceptual linkage with relational aggression improves measurement and assessment of what was previously called the “parental alienation behaviour” construct. Our paper begins with brief reviews of both relational aggression behavior and parental alienation behavior. Then, we present logical links and highlight overlap between the two accepted definitions of each construct and go on to discuss problems with existing measures of parental alienation behavior and how they can be overcome. We conclude with a brief discussion of the predictors of engagement in parental alienation behavior – what we prefer to call parent-child bond-breaking for conceptual and measurement reasons – focusing on the construct of attachment.

Relational aggression

Relational aggression is a type of aggressive behavior where the perpetrator seeks to harm another person by targeting their relationships (Crick & Grotpeter, 1995); specifically, it is the “hostile manipulation of relationships and use of threats to control or dominate others” (Herrenkohl et al., 2007) (p. 387). Proactive relational aggression is intentional and coercive, and occurs in the absence of any other inciting factor, while reactive relational aggression occurs in response to an inciting factor such as real or perceived aggression on the part of another person, or another type of provocation (Murray-Close et al., 2010). Notably, these two subtypes of relational aggression are not mutually exclusive such that perpetrators can use both forms. Common relational aggression behaviors include giving the targeted person the silent treatment, excluding them from social activities and information, and spreading rumors about them; essentially any social action (or non-action) that denigrates and excludes the victim socially from valued relationships and undermines their social status. Relational aggression is distressing for victims who evidence poorer mental health outcomes (Marshall et al., 2015). In general, research suggests that relational aggression is more common among girls and women (Crick et al., 2007), but boys and men can be relationally aggressive as well (Bitsola & Kyranides, 2021; Burton et al., 2007).

Among children and young people, relational aggression is most commonly studied in the context of hierarchical and cliquish peer relationships in school settings; however, relationally aggressive behavior can persist into adulthood and into other relationship types. Specific to the context of romantic relationships, relational aggression behaviors can include threats to withdraw love and attention as well as flirting with other people to induce jealousy in one’s romantic partner (Dewi & Kyranides, 2022). Among adults, correlates of relational aggression include lower agreeableness and lower emotional intelligence (Burton et al., 2007) as well as elevated levels of narcissistic and psychopathic traits (Czar et al., 2011).

Relational aggressive behavior is expressed differently (i.e., the specific behaviors vary) depending on the type of social relationship (e.g., friends, colleagues, romantic partner), context where the relationship is embedded (e.g., school, workplace, home), and the phase of life of the individuals involved (e.g., children and young people, adults). Thus, it is consistent with the accepted definition of relational aggression to also consider parental alienation behavior as a context and life-stage specific pattern of relationally aggressive behavior between separating and divorcing parents that specifically targets the relationship between parents and children.

Parental alienation behavior

Parental alienation behavior (PAB) negatively affects the health and well-being of child and adult victims by negatively disrupting parent–child bonds (Harman et al., 2018; Miralles et al., 2023; Verrocchio et al., 2018). Parents of any gender can engage in alienation behaviors, which undermine the relationship between the co-parent (and ex-romantic partner) and a shared child(ren); it is most often studied in the context of separation and divorce (Harman et al., 2018). Parental alienation behaviors can include starting rumors and making accusations about the other parent that undermine their social status (e.g., they are a cheater, abandoner, abuser), impeding their physical contact and communication with their children actively or passively (e.g., blocking cell phone numbers, being unavailable at agreed phone contact times), blaming the other parent for things without justification (e.g., he/she is making us sell the car), gatekeeping (e.g., not passing along invitations to school events), and removing evidence that the other parent ever existed (e.g., getting rid of photos, gifts, toys, and keepsakes) (Baker & Darnall, 2006).

As with any form of aggression, these behaviors can be committed proactively – in the absence of any provocation on the part of the ex-partner AND/OR reactively – in response to a real or perceived threat from the ex-partner. Please consider the following examples of proactive parent–child bond-breaking. Parent A says to Parent B that the marriage is over and they want a divorce. Parent B immediately files a police report alleging that Parent A engages in domestic abuse and requests that Parent A is barred from occupying the family residence and having contact with the shared child(ren). Now, for this claim to be proactively aggressive, it would need to be known as false (i.e., a lie) to Parent B. Complicating matters, it could be objectively not true (i.e., not in keeping with reality) but perceived as true owing to what is known as a “reality distortion” in which Parent B believes it to be true, perhaps because they feel so threatened by the rejection of Parent A that they are genuinely in a state of panic and fear and cannot bear the thought of continuing to live with Parent A who has hurt them so profoundly. Reality distortions of this nature can be exacerbated by stress; with the stress of being rejected by a romantic partner being present among candidate stressors in situations of separation and divorce. An example of reactive parent–child bond-breaking would be the exemplar case proposed by women’s rights activists, namely that abusive fathers accused of abuse (e.g., neglect or overt abuse) knowingly and falsely allege “parental alienation” on the part of their female ex-partner to explain their shared children’s reticence for contact. Another example without physical or emotional abuse claims, could be both parents trading insults and involving the child(ren) in war of words – so badmouthing the other parent in front of the child, using the child as a go-between, making allegations about the other parent to the child, telling the child that the other parent does not love them, etc. In these cases, determining “who started it” becomes an irrelevant needle in the haystack. This more complex picture of co-relational aggression or otherwise complex and highly interactive model of parental alienation is consistent with the model put forth by Johnston and Sullivan (Johnston & Sullivan, 2020).

Prevalence estimates of experiencing bond-breaking and/or alienation depend on the sample such as the general population, families separating/divorcing, or families that have been referred for clinical support or family court processes, with prevalence increasing along this narrowing pathway as one would expect. Almost a quarter of divorced/separated families in the United States reported having at least one child who had experienced alienation from one of their parents (Harman et al., 2016). Studies have shown that these behaviors can eventually lead to rejection of the targeted parent by the child(ren) (Jaffe et al., 2017). In conjunction with the rejection, children generate specious and superficial justifications for their devaluation of the targeted parent despite their strong conviction to it, their narratives have a scripted adult-like quality to them, and their denigration and dismissiveness of the targeted parent is intense and inflexible (Baker & Damall, 2007). Recent methodologically rigorous reviews and analyses of study data using open science methodologies have added to the growing empirical and conceptual evidence for parental alienation behaviors and their negative psychological effects (Harman & Lorandos, 2020; Harman et al., 2018, 2019; Lee-Maturana et al., 2022).

Limitations of parental alienation behavioral measures

Most of the measures of “parental alienation behaviors” are victim self-report measures of either adults reporting retrospectively about their childhood experiences (Baker, 2009; Moné & Biringen, 2012) or self-identified alienated parents (Baker & Damall, 2007; Rowlands, 2019). Assessments of targeted children experiencing parent–child bond-breaking behaviors, or children who might be alienated, and are already involved in family court processes, are naturally much more complex and not the subject of the current manuscript (Bow et al., 2009; Rueda, 2004; Sîrbu et al., 2021; Warshak, 2020). Our focus is on the assessment of the perpetration of parent–child bond-breaking behaviors among parents. To our knowledge, there are no self-report measures that assess engagement in parent–child bond-breaking behaviors, leaving a significant gap in the literature. The lack of measurement tools of engagement in parent–child bond-breaking means that it is very difficult to adequately test and adjudicate among competing theories about why parents might engage in these behaviors.

It is important to note that there is the point of view that it is primarily the case that abusive fathers allege “parental alienation” as legal strategy to explain their children’s reasonable rejection of them (Barnett, 2020; Birchall & Choudhry, 2022; Lapierre & Côté, 2016). Rigorously testing this claim – that what these parents (primarily mothers) are doing is not alienation but a reasoned response to abuse and the claim that parental alienation is caused by parent–child bond-breaking behaviors in the absence of abuse, necessitates an objective, theoretically informed and empirically validated measure that can be used to evaluate both kinds of claims. This goal can be achieved by measuring parent–child bond-breaking behaviors without the presumption or embedding of questions related to alienation owing to instances where such “bond-breaking” could be perceived as protective along with measuring perceptions of danger and relationship quality prior to separation in the context of one instrument. Such a measure does not currently exist and it was one of the goals of this research project to create one. Returning to the issue of evaluating theories about why some parents are more likely to engage in parent–child bond-breaking than others, we next consider the unique role of attachment.

Attachment

Attachment was popularized in the context of parent-infant research conducted by John Bowlby and Mary Ainsworth (Bretherton, 1992). Briefly, it is theorized that when infants feel secure in their relationship with their caregivers that they can explore their environment, grown, learn and thrive with the confidence that they can depend on their caregiver for safety if needed. These early attachment bonds and psychosocial behavioral patterns pave the way for scaffolding positive peer relationships, future romantic relationships and finally a next generation of parent–infant relationships (Fairchild & Finney, 2006). “Secure” attachments were theorized to be associated with a variety of positive psychological, educational and mental health outcomes, with insecure attachment types (e.g., anxious, avoidant, disorganized) associated with increased risk for negative outcomes (Cassidy & Berlin, 1994; Cassidy et al., 2013; Mikulincer & Shaver, 2012). Recent research has focused on illuminating a deeper dimensionality of the construct, creating novel assessment paradigms that are culturally sensitive, and using interdisciplinary approaches (Fearon & Roisman, 2017; Rothbaum et al., 2002; Stern et al., 2022).

When important relationships breakdown, such as in the case of marital or romantic relationships involving children, it can activate deeply embedded attachment-related emotions, cognitions, and thought-to-behavioral routines that influence within-person and between-person functioning, or put another way, both intra- and inter-personal patterns of thinking and acting (Sbarra & Borelli, 2019). Sbarra and Borelli (2019) argue that attachment anxiety and attachment avoidance are particularly powerful in predicting adjustment following separation and divorce. Attachment anxiety can manifest in different ways but generally it involves excessive worry about a breakdown in a relationship and is correlated with feelings of low self-worth, “clingy” and impulsive behaviors. Paradoxically, these actions can then drive partners away creating the scenario that the individual was so fearful of in the first place. People prone to avoidant attachments are more distant and may at times appear cold, they often prefer to be independent so they do not have to rely on other people, even if this is something that they might actually desire (Li & Chan, 2012). Attachment theory has successfully informed approaches to psychotherapy and treatment for couples and individuals for decades (Daly & Mallinckrodt, 2009; Meyer & Pilkonis, 2001).

In the context of parental alienation research, to our knowledge, despite popularization of the notion that attachment plays a key a role in engagement in parent–child bond-breaking (Childress, 2015), there is no empirical research examining how parental attachment may translate into intra- and inter-personal behaviors toward the ex-partner and their child, and specifically into parent–child bond-breaking behavior. Following separation and divorce attachment reorganization must take place, not only between parents and children but also among the parents themselves. That is each adult must reevaluate who they are in the context of the relationship breakdown. This process can be difficult, but it is fundamentally necessary in order to move forward in a healthy way (Sbarra & Borelli, 2019). Related to the process of post-separation adjustment is that high conflict divorce has been clearly linked to negative outcomes for children (Johnston, 1994) and the important role of having positive (i.e., high quality) and stable relationships with both mothers and fathers is strongly supported empirically (Brand & Klimes-Dougan, 2010). Consequently, limiting parental engagement in parent–child bond-breaking behavior, which could be viewed as an ineffective and harmful coping strategy, is crucially important. An adult who is anxiously attached for example, may react very poorly to the breakdown of the relationship with the co-parent; they might cling to their child(ren) instead, thus leading to enmeshment behaviors (Barber & Buehler, 1996; Garber, 2011). Fearful of yet another abandonment and dissolution of the family relationships, they may turn to engaging in parent–child bond-breaking behavior to inhibit the children from spending time with their other parent potentially owing to anger toward that other parent, but also out of a deep fear of “losing the kids” and their identity as a parent (after having lost their identity as a husband or a wife), which might be related to their self-esteem.

Thus, in the current pilot study, we explored attachment avoidance and anxiety as key correlates of engagement in parent–child bond-breaking behavior in a sample of separated and divorced adult parents and evaluated our central claim that this behavior is a sub-form of relational aggression by evaluating links between parent–child bond-breaking and romantic relational aggression (i.e., to test its construct validity by evaluating indicators of convergent validity). To summarize, we tested two hypotheses (1) there will be a positive association between attachment anxiety and parent–child bond-breaking and (2) there will be a positive association between romantic relational aggression and parent–child bond-breaking, owing to there being an underlying proclivity for engaging in socially and relationally aggressive behaviors towards one’s partner.

Methods

Participants

A convenience sample of adults (N = 53) was recruited through online study advertisements, word of mouth, and publicly posted flyers from May 2023 to July 2023. Study materials expressly did not mention or otherwise reference parental alienation to avoid collecting a biased sample of adults who either endorse or oppose the concept, and then further biasing participants through the use of this term once enrolled in the study. To be eligible for the study, participants had to be over the age of 18, fluent in written and spoken English, have at least one child and be separated or divorced from the child’s other parent. After reading a study information sheet, participants indicated their informed consent to participate by completing an online form. They were then directed to a secure survey platform to complete a set of questionnaires and open-ended questions (these were not the focus of the current report). The survey was designed and deployed using best-practice guidance to deter and to detect “bots” (Hallberg, 2022). No identifying information was collected. At the conclusion of the study participants were provided with a debrief sheet pointing them toward resources for separated and divorced parents and thanked for their time. The study protocol was reviewed and approved by the University of Edinburgh’s Department of Clinical and Health Psychology’s Ethical Review Board. A description of the surveys and measurements follows below.

Measures

Sociodemographic information

For the purpose of characterizing the sample we asked parents to self-report their age, race/ethnicity, sexual orientation, sex assigned at birth and current gender (if different), current relationship status, parental status (e.g., mother, father), educational attainment, current employment status, their number of children, if they have parental rights and responsibilities for the children they share with their ex-partner, and their primary country of residence.

Parent-child bond-breaking

A new instrument was created for the purpose of the study. We desired a simple 10-item instrument in lay language that could be used in both clinical and community samples. It was critical to word items as concretely and objectively as possible to encourage honest responding. It was important for this reason to avoid alienation terminology. Indeed, many of the behaviors captured on the measure may genuinely be perceived as being protective of the adult and of the child(ren), and because alienation is a potential product of the parent–child bonds being broken we did not want to conflate the antecedent behavior with its potential consequent on this instrument as it would weaken its construct validity.

Parent–child bond-breaking behavior was assessed through a new measure that included 10 items with Likert-type response choices anchored as Strongly Disagree (1) to Strongly Agree (5), based on the authors’ subject matter expertise in psychometrics, aggression, parenting and child safety and on prior research related to alienation strategies corresponding to: negative messaging, limiting contact and communication, erasing, encouraging betrayal of trust, and undermining authority (Baker & Darnall, 2006; Verrocchio et al., 2018). Items were written as follows – (1) Negative messaging: “I let my child know that their mum/dad isn’t a great parent,” “I let my child know that their mum/dad can’t be trusted;” (2) Limiting Contact and Communication: “I restrict my ex-partner’s access to information about my child,” “I restrict my ex-partner’s access to our child”; (3) Erasing: “I remove reminders of my ex-partner from our child’s life, like gifts and photographs,” “I avoid talking with my child about the good times we had as a family before the split”; (4) Encouraging Child to Betray Trust: “I don’t care if my child is mean to their mum/dad,” “I encourage my child to say negative things about their mum/dad;” and (5) Undermining authority: “I’ve let my ex-partner know that their opinion about how our child is raised is not that important,” “I’ve let other people in my child’s life know that my ex-partner can’t be trusted.” Of all of these items, encouraging the child to betray trust are theoretically the most closely linked with overt aggression, but even with these, it is possible to argue that after years of a toxic relationship with an ex-partner, a parent could legitimately not care if their child is mean to the co-parent. This does not mean that it is wise parenting decision, but rather the truth that the ex-partner just does not care if this happens. It is more difficult to rationalize that encouraging a child to say a negative thing about their other parent is a reasonable parenting behavior, even in the context of a high conflict separation and abusive or reckless behavior on the part of that parent; as such, we believe strong endorsement of these two items, and in particular this latter one, would indeed be able to capture more serious relationally aggressive behavior. In measuring deviant behavior significant care must be taken to encourage honest responding. We argue that using the term parental alienation is likely to be very biasing to potential participants, dissuading perpetrators from participating in research that they may feel is blaming or casts them in a pejorative light.

Two True or False items were used to establish if the ex-partner has been convicted in a court of law of a crime related to abuse or neglect (within the bounds of the accuracy of the person doing the reporting) and the reporting parents’ perception that the relationship between the ex-partner and children was better prior to the separation/divorce: (1) My ex-partner has been convicted in a court of law of committing a crime related to abuse or neglect and (2) My ex-partner’s relationship with our child was better before the separation or divorce. We included these two items in order to adjust for engagement in parent–child bond-breaking due to the perception of a justified need, particularly because the ex-partner was unsafe in some way. This approach was crucial to isolate the construct of interest, which are the parent–child bond-breaking acts, not hypothesized antecedent factors (e.g., parental motivations or intentions precipitating the behavior) as well as to keep measurement of the behaviors clear of conflation with potential consequences of the behaviors (e.g., alienation). Thus, the parent–child bond-breaking measure is not a measure of parental alienation, but rather it is a measure of the kind of relationally aggressive behavior that might lead to it.

Relational romantic aggression

Relational Romantic Aggression was assessed using a subscale from the Self-report Measure of Aggression and Victimisation Questionnaire (Linder et al., 2002; Murray-Close et al., 2010). Participants respond to 5 items on a 7-point scale, Not at all True (1) to Very True (7). In keeping with prior research, participants were asked to report on their current romantic partner and if not in a relationship to report on their prior partner. The five items are: (1) “I have threatened to break up with my romantic partner in order to get him/her to do what I wanted”; (2) “I try to make my romantic partner jealous when I am mad at him/her;” “I have cheated on my romantic partner because I was angry at him/her;” “I give my romantic partner the silent treatment when s/he hurt my feelings in some way;” and” If my romantic partner makes me mad, I will flirt with another person in front of him/her.” Prior research has indicated strong reliability for this measure α = .72 (Linder et al., 2002).

Attachment

Attachment was assessed using the Experiences in Close Relationships – Revised Questionnaire. It is a 36-item questionnaire with two sub-scales, Avoidance and Anxiety, each consisting of 18 items scored on a 7-point scale, Disagree Strongly (1) to Agree Strongly (7). An example item on the Avoidance sub-scale is that “I find it difficult to allow myself to depend on romantic partners” and an example item on the Anxiety sub-scale is “I often worry my romantic partner doesn’t really love me;” the scales have strong reliability, α = .93 and α = .92 respectively (Fairchild & Finney, 2006).

Hypotheses and data analysis plan

We hypothesized that romantic relational aggression will be positively linked with propensity to engage in parent–child bond-breaking owing to it being a form of relational aggression and further that attachment anxiety would be positively linked with propensity to bond-break. Observing these relationships, particularly a strong pattern of stability in relational aggression (or consistency in engaging in relational aggression toward romantic partners), would be a marker of the construct validity of parent–child bond-breaking behavior. These hypotheses were evaluated through correlational and regression analyses predicting differences in propensity to engage in parent–child bond-breaking by individual differences in engagement in relational romantic aggression and attachment anxiety adjusting for parent relationship type (mother, father), parent report of the ex-partner having a prior conviction related to abuse or neglect (yes, no), and the parent reporting that the ex-parent had a better relationship with the child(ren) prior to the separation (yes, no). We included Attachment Avoidance as it is commonly measured along with Attachment Anxiety but had no a priori hypotheses about its relationship with bond-breaking and were primarily concerned with Attachment Anxiety.

Results

Following data cleaning procedures used to remove duplicate responses (e.g., bots), there were 46 evaluable participants. There was no indication that these remaining respondents were in anyway invalid. Twenty-eight (61%) of participants were from the United Kingdom (England or Scotland), 13 were from an Asian country (e.g., China, India), 2 from Europe, and 1 from North America. Two people chose not to report this information but completed the rest of the survey.

Twenty-nine (63%) of participants were mothers, identified as female, and were an average age of M(SD) = 43(9.00) years old. Most mothers in our sample were White (n = 22), followed by Asian (n = 5), and Other (n = 2). Fourteen had a postgraduate or doctorate degree, 7 had an undergraduate degree, and 7 had finished secondary school or achieved a diploma. Regarding employment, 1 was unemployed, 3 were actively seeking employment, 2 were students, 16 worked full-time, and 3 indicated an “Other” employment status. Regarding current relationship status, 4 identified as single and never married, 7 were divorced but currently single, 7 were in relationship but were not currently married, 5 were separated, 3 were divorced but did not indicate another status outside this, and 3 were re-married. Twenty-five identified as heterosexual or “straight,” 3 identified as bi-sexual and 1 identified as lesbian. Fourteen mothers reported having 1 child, 7 reported having 2 children, 6 reported having 3 children and 1 reported having 4 children; parents were asked to report on all biological or adopted children. Twenty-six mothers reported having parental rights and responsibilities, 2 reported that they did not, and 1 was not sure. On average, mothers were separated/divorced from the father/other parent for M(SD) = 7.00(6.84) years.

The remaining 17 participants were fathers, identified as male, and were an average age of M(SD) = 39.71(10.73) years old. Five fathers had post-graduate or doctorate degrees, 7 had an undergraduate degree, and 5 had completed secondary school or achieved a college diploma. One father reported being unemployed and 16 reported being in fulltime employment. Two fathers reported being currently single and never married, 6 reported being divorced but currently single, 4 reported being in a relationship but were not currently married, 3 reported being divorced but did not indicate another status outside this, and 2 reported being separated. No fathers reported being re-married and all fathers reported being heterosexual. Nine fathers reported having 1 child, 5 reported having 2 children, 2 reported having 3 children, and 1 reported having 5 children; parents were asked to report on all biological or adopted children. Fifteen fathers had full parental rights and responsibilities, one did not, and one was not sure. On average, fathers were separated/divorced from the mother/other parent for M(SD) = 7.94(6.77) years.

Factor-analysis

Because we used a new measure, the 10 items of the parent–child bond-breaking measure were subjected to principal factor analysis using a Varimax rotation. The Kaiser–Meyer–Olkin Measure of Sampling Adequacy = .754 and Bartlett’s Test of Sphericity was significant (p = .000) thus supporting the factorability of the data matrix. Following rotation, 3 factors were retained (Table 1). This solution was the most parsimonious numerical and theoretical accounting of the data. Factor 1 explained 51.42%, Factor 2 explained 14.45%, and Factor 3 explained 12.32% of the variance respectively; so cumulatively, 78.19% of the total variance was explained. For the three subscales, for factor one (Encouraging child’s aggression toward the co-parent) we observed M(SD) = 1.55(.84) and a range of 1–5, for factor two (Editing, Excluding and Re-Writing the Narrative) we observed M(SD) = 1.93(.95) and a range of 1–4, and for factor three (Devaluing the Co-parent) we observed M(SD) = 2.00(.89) and a range of 1–5. A composite measure computed by a simple average of all 10 items yielded, α =.89, M(SD) = 1.89(.76) and ranged from 1.52 to 2.20.

Table 1. Parent–child bond-breaking, three-factor solution.

	Factor
	Encouraging Child’s Aggression Towards the Co-Parent α = .88	Editing, Excluding and Re-Writing the Narrative α = .86	Devaluing the Co-Parent α = .80
9. I encourage my child to say negative things about their mum/dad.	.894	.250	.083
4. I don’t care if my child is mean to their mum/dad.	.761	.170	.123
2. I restrict my ex-partner’s access to information about my child.	.297	.763	.301
8. I avoid talking with my child about the good times we had as a family before the split.	.103	.663	.158
3. I remove reminders of my ex-partner from our child’s life, like gifts and photographs.	.613	.645	.075
7. I restrict my ex-partner’s access to our child.	.491	.593	.118
6. I let my child know that their mum/dad can’t be trusted.	.570	.085	.794
1. I let my child know that their mum/dad isn’t a great parent.	.583	.143	.628
5. I’ve let my ex-partner know that their opinion about how our child is raised is not that important.	.008	.577	.614
10. I’ve let other people in my child’s life know that my ex-partner can’t be trusted.	−.020	.180	.564

Extraction Method: Principal Axis Factoring. Rotation Method: Varimax with Kaiser Normalization.

Descriptive statistics

No differences were observed by parental status (mother vs father) for the overall average parent–child bond-breaking score or any of its subscales: parent–child bond-breaking, Average: M(SD)_fathers = 1.91(.85) and M(SD)_mothers = 1.88(.90); F(1, 45) = .16, p = .89; De-Valuing: M(SD)_fathers = 1.86(.83) and M(SD)_mothers = 2.10(.93); F(1, 45) = .78, p = .38; Encouraging Child’s Aggression: M(SD)_fathers = 1.65(1.11) and M(SD)_mothers = 1.50(.64); F(1, 45) = .33, p = .57; and Excluding and Re-writing the Narrative: M(SD)_fathers = 2.09(1.03) and M(SD)_mothers = 1.84(.90); F(1, 45) = .70, p = .41.

To create a count of parent–child bond-breaking behaviors, we re-scored each item into a dichotomous variable with a 1 being assigned if the parent reported agreeing or strongly agreeing with the behavior and then we summed the 10 items, thus possible scores could range from 0 to 10. Using this count index we learned that 25 parents reported no to minimal engagement in these behaviors with a behavioral count score of 0. This was the most common count score we observed (54%) as it comprised just over half of the sample of parents. Nine parents (20%) reported engaging in one bond-breaking behavior, 4 parents reported engaging in 2 behaviors, 3 parents reported engaging in 3 behaviors, 4 parents reported engaging 4 behaviors and 1 parent reported engaging in 6 bond-breaking behaviors. The most frequently endorsed behaviors were “letting other people in my life know my ex-partner can’t be trusted”(17%); “letting my ex-partner know that their opinion about how our child is raised is not that important” (17%); “I restrict my ex-partner’s access to information about my child”(17%); “I avoid talking with my child about the good times we had as a family before the split” (11%), and “I remove reminders of my ex-partner from my child’s life, like gifts and photographs” (11%).

Inferential statistics

Two participants did not complete the attachment and relational aggression measure, thus there were 44 evaluable participants for the regression analysis. As predicted, the overall average of the parent–child bond-breaking scale, and all three subscales, demonstrated significant positive associations with relational romantic aggression and attachment anxiety (Table 2). Based on this pattern of associations we used the average scale score for all subsequent analyses.

Table 2. Bivariate correlation matrix (spearman’s rho).

	Aggression Towards the Co-Parent	Re-Writing the Narrative	Devaluing	Parent-Child Bond-Breaking Average	Attachment Avoidance	Attachment Anxiety	Romantic Relational Aggression
Encouraging Child’s Aggression	1
Re-Writing the Narrative	.625**	1
De-valuing	.571**	.722**	1
Parent-child bond-breaking Average	–	–	–	1
Attachment Avoidance	−.026	.096	.193	.109	1
Attachment Anxiety	.498**	.505**	.629**	.630**	.140	1
Romantic Relational Aggression	.445**	.479**	.431**	.516**	−.110	.398*	1

*p < .05; **p < .001.

We evaluated the contributions of attachment avoidance, attachment anxiety, and relational romantic regression toward parent–child bond-breaking behavior. We adjusted for parental role (mothers, fathers), report that the ex-partner was convicted of a crime related to abuse or neglect in a court of law (yes, no), and if the respondent’s perception of whether or not the relationship between the ex-partner and the shared children was better prior to the separation or divorce (yes, no). The inclusion of these two True/False items adjusted for any variation in parent–child bond-breaking behavior owing to the potential for the parent respondent to want to distance or otherwise disrupt the parent–child bond for what they perceived to be legitimate safety reasons. Note, according to our conceptual definition this would still be relationally aggressive behavior; it would just be reactive to the perception of a threat.

Sex at birth, ex-partner’s reported conviction status, and perception of relationship quality prior to separation were entered at Step 1 and explained 2% of the variance F(1,43) = .674, p = .573. Romantic relational aggression was entered in Step 2 and with its addition the model explained 23.5% of the variance; F(5,43) = 4.23, p = .006, and finally attachment avoidance and anxiety were entered in Step 3 and with their addition the model explained 39.1% of the variance in parent–child bond-breaking behavior; F(7,43) = 5.61, p < .001.

To further explore the joint contributions of relational romantic aggression and attachment anxiety, because they were the only significant variables in the initial regression model and in the bivariate correlation analysis, these variables were regressed onto the average parent–child bond-breaking scale score. None of the previously evaluated hypothesized confounders (e.g., being a mother or father, ex-partner with a conviction) contributed significantly to the prior model so they were also dropped. This reduced model accounted for 45% of the variation in parent–child bond-breaking behavior: relational romantic aggression: β = .37, 95% CI[.018, .090], p = .004 and attachment anxiety: β = .47, 95% CI[.008, .025], p < .001.

Discussion

The results from our pilot study provide initial evidence for the validity of a new measure of parent–child bond-breaking behaviors, a type of family-based relational aggression most commonly occurring in the context of a separation and divorce. Evidence for convergent validity was found through the observation of significant associations between individual differences in romantic relational aggression being positively associated with parent–child bond-breaking. These findings also suggest that targeted parents, that is those parents who are victims of bond-breaking attacks, may have a history of romantic relational aggression victimization by the same partner. We also saw support for strong linkages between attachment anxiety and propensity for parent–child bond-breaking behavior even after adjusting for parental role, ex-partner’s abuse history, and prior relationship quality with the shared children. Collectively, there was no evidence for parent–child bond-breaking behavior being related to parental roles; mothers and fathers reported similar levels of engagement in the behavior. We adjusted for a history of the ex-partner’s child abuse and neglect offending behavior and pre-separation parent–child relationship quality and it had no effect on the pattern of associations we observed.

One aim of this paper was to propose that by conceptually demonstrating that parental alienation behaviors are a context-specific subtype of relational aggression, we can overcome Jangle. Jangle fallacies occur when different terms are used to explain very similar if not identical concepts, actions, behaviors, or constructs and then people adopt the position that if we are using different terms then it must be that underlying constructs are also different. Additionally, we can leverage violence prevention strategies and learnings from the relational aggression field while still incorporating the growing empirical literature that substantiates parental alienation behaviors (although we strongly prefer the less loaded term of parent–child bond-breaking) as a concerning and pervasive type of family violence that is connected to other aggressive behaviors such as romantic relational aggression, emotional and psychological abuse, and coercive control. The term parental alienation behaviors, in our view, is too focussed on the experience of the parent. It is a term created by adults for adults and does not give sufficient attention to the bond between parents and their children.

Parental alienation, a potential product of parent–child bond-breaking behavior that we did not assess in this current study, is known to occur within a milieu of other factors such as mental health conditions, parenting skills and practices, other forms of abuse, and relational issues making it practically difficult for outside professionals (e.g., psychologists, social workers, teachers, lawyers, judges) to identify and excise the parental alienation behaviors and to delineate the unique contribution of parental alienation’s role in any given situation and custody dispute (Johnston & Sullivan, 2020). As such, researchers and practitioners need to be sensitive to these behaviors occurring in contexts where there are, and are not, substantiated allegations of other forms of abuse and other contributing factors. Targeted fathers in particular are faced with limited resources and are reluctant to self-identify as victims due to prevailing social norms and stereotypes about masculinity (e.g., shame, disbelief and confusion about their victimization) (Hine et al., 2021). Increased awareness among professionals in public health, medicine, law, social work and education about what parent–child bond-breaking behavior is, how it is measured, and how it affects families is crucial.

Research on relational aggression has, over a period of several decades, demonstrated it is a valid construct that spans cultures, that is observed in men and women, and that it has a negative impact on victims. Its tenets are uncontroversial: aggressors target and weaponize social relationships to exclude and to malign victims. These same tenets are indisputably also the core tenets of parent–child bond-breaking behavior, which are goal-directed actions undertaken by one parent with the aim to rupture the other parent’s relationship with their child. Conceptually, parent–child bond-breaking is clearly a specific form of relational aggression. Empirically, as evidenced in this study, different types of releational aggression (romantic and parent-child bond-breaking) are positively correlated. Consequently, there is both theoretical, logical, and empirical support for the validity of the parent–child bond-breaking behavior construct. We have further shown parent-child bond-breaking can be validly measured using the new 10-item assessment created for the purpose of this study, although additional research to further test the tool with larger and diverse samples is essential.

We hope that clearly re-positioning parental alienation behaviors under the sturdy umbrella of relational aggression, giving it a clearer name – parent–child bond-breaking behavior, and improving its measurement through the development of a more rigorous assessment instrument that acknowledges criticisms of parental alienation research, namely that it could use better measurement methods, will help to shift the dialogue about parental alienation in a productive and scientific manner. A report from the BBC had the following to say, “Women’s rights organisations argue parental alienation is used to gaslight abused women. Fathers’ rights organisations claim that some mothers make up allegations of abuse to prevent them from seeing their children. And children are caught in the middle.” (BBC Radio 4—Analysis, Parental Alienation, 2021). We would agree that children are indeed caught in the middle. We are optimistic, however, that a principled scientific approach can be used to build bridges and inform science and practice in this space.

Limitations, strengths and future research

We recruited a community sample of adults and not a forensic sample. It is likely that parent–child bond-breaking behaviors exists on a continuum and we did not observe more extreme and pervasive behaviors that would lead to severe alienation of a child from their parent. The average parent–child bond-breaking behaviors score in our study ranged from 1.5 to 2.0 which is on the lower end of what the measure can capture. However, we did observe use of the upper ends of the scale with parents indicating that they agreed or strongly agreed with the behaviors. So, for example, 17% of the sample agreed or strongly agreed that they let their ex-partner know that their opinion about how their child is raised is not that important, but only 2% agreed or strongly agreed that they did not care if their child was mean to their mum/dad. This indicates that parents are willing to admit to these behaviors and that more extreme behaviors may be less frequently observed. Future directions for research with this measure can include testing it in different populations and determining best-practices for scoring. For example, an average full-scale score like we used in the current analysis can provide a general assessment but other approaches such summing the number of behaviors that a parent agreed or strongly agreed with to create a behavioral count score can also be explored as we demonstrated. This way of scoring the measure is agnostic to which behaviors are being engaged in and is concerned more with the total spread of engagement.

Future research can determine if certain behaviors are more indicative of moderate to severe alienation and further explore use of the sub-scale scores. We did not observe changes in the relationships among the variables based on a history of conviction status or pre-separation relationship quality but that could be related to our small community sample of participants. We strongly recommend that these items are retained on the parent–child bond-breaking behavior assessment we created to allow for continued study of the contribution of those behaviors toward both the probability of alienation and the commission of the parent–child bond-breaking behaviors. Finally, in this study, we examined attachment anxiety as a key correlate of parent–child bond-breaking behaviors and found evidence to support this. Other constructs of interest are Dark Triad Traits (e.g., narcissism, psychopathy, and Machiavellianism). For example, research has linked psychopathic traits with increased engagement in relationally aggressive behavior (Czar et al., 2011), in particular with proactive relational aggression (White et al., 2015), and also with intergenerational attachment problems (Kyranides et al., 2023; van der Zouwen et al., 2018). Additional research on this topic could help to illuminate processes and associations between proactive and reactive parent–child bond-breaking behavior and mental health outcomes for parents and children. Longitudinal and quasi-experimental designs can be leveraged in future research to compare groups (e.g., cultural, offending severity), to study change over time, and to isolate other risk and protective factors. Strengths of the study included a robust and theoretically grounded measurement tool and validated measures of theoretically linked constructs (e.g., attachment, relational aggression) – thus, the internal validity of the study is very strong. We advise caution in generalizing the effect sizes we observed to other populations owing to the convenience sampling methods we used to collect the sample.

In summary, we hope that by drawing logical links between the hallmark bond-breaking behavior associated with parental alienation and the core behaviors associated with relational aggression we can overcome jangle, and build bridges to a more productive and unbiased scientific and practical dialogue that can ultimately lead to tangible public health and psychological benefits for children and their families. We hope that this new measure can be further researched by other teams interested in evaluating a diversity of claims related to parent–child bond-breaking.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.