Adolescence and youth are periods involving changes that affect on a socioemotional, physical, and psychological level (Branje, 2018; Calvete, Orue et al., 2020). These changes can generate conflicts within the family environment (Branje, 2018; Calvete et al., 2011). These conflicts escalate when family members resort to aggressive behaviors as a problem-solving method (O’Hara et al., 2017), which can lead to more severe problems such as child-to-parent violence (CPV) (Calvete et al., 2017). CPV are recurrent psychological, physical, and economic violent behaviors by children against parental figures (Pereira et al., 2017). This kind of intrafamilial violence is exercised to a greater extent toward mothers (Calvete et al., 2017; Gámez-Guadix et al., 2012). Meanwhile, there is no agreement regarding the child’s gender. Some studies have found no differences in CPV among gender groups (Calvete, Orue, Gámez-Guadix & Bushman, 2015; Loinaz & De Sousa, 2020; Gámez-Guadix et al., 2012), yet a recent study suggests that gender differences vary depending on the aggression form (Simmons et al., 2018). In terms of age, CPV shows its highest prevalence between the ages of 12 and 17 (Calvete, Orue et al., 2020). However, it can also emerge in childhood or persist into early adulthood (18 to 25 years) (for review, see Simmons et al., 2018). This continuity may be linked to factors such as financial difficulties hindering independence, as well as the persistence of violent dynamics within the parent–child relationship (Ibabe, 2020; Rogers & Ashworth, 2024). To date, few studies have specifically focused on CPV during this stage of development, highlighting the need to generate evidence that can help address this gap in literature.

Regardless of the presence of CPV in society, there is not a clear consensus about its current prevalence (Calvete & Orue, 2016; Contreras & Cano, 2014), mainly due to the variety of definitions (Gámez-Guadix et al., 2012) and disagreements regarding measurement (Calvete et al., 2013). According to research in community and international samples, psychological CPV occurs in a range from 33% to 95.7% in national studies (Calvete et al., 2017; Ibabe & Bentler, 2016) and from 45% to 87.2% of the cases in international studies (Beckmann et al., 2021; Calvete et al., 2018). In addition, findings suggest that physical CPV occurs in 2.6% to 22% of cases in national studies (Contreras et al., 2020; Calvete et al., 2017) and in 5.5% to 47.5% in international studies (Beckmann et al., 2021; Lyons et al., 2015; Margolin & Baucom, 2014).

Even though these studies show high rates of CPV there are still gaps in the research of this growing phenomenon. In addition, most previous studies have centered on family risk factor detection, emphasizing exposure to family violence as a powerful element (Bautista-Aranda et al., 2023; Contreras et al., 2020; Warren et al., 2023). Moreover, harsh discipline strategies, such as physical (Harries et al., 2023) and psychological punishment (Jiménez-Granado et al., 2023; Gámez-Guadix et al., 2012), are also associated with increased CPV. Some research indicates that parents who are physically and psychologically abusive to their children are more likely to experience CPV than those in whom violence does not occur (Beckmann et al., 2021). On the other hand, although little research has attempted to evaluate the individual risk factors of adolescents regarding the development of CPV, there are studies that give us an idea about them. Thus, adolescents’ impulsivity (Loinaz et al., 2020; Rico et al., 2017) and substance use (Calvete, Orue, Gámez-Guadix, Del Hoyo-Bilbao et al., 2015; Del Hoyo-Bilbao et al., 2020) have received the most empirical attention. However, there are few studies aiming to identify its protective factors (Beckmann et al., 2021; Loinaz et al., 2023). Even fewer studies have tried to identify factors that can buffer the predictive relation between exposure to family violence and CPV (Contreras et al., 2020; Del Hoyo-Bilbao et al., 2020).

Knowing and detecting CPV protective factors while simultaneously working with those that protect youths that are exposed to family violence at home is imperative. The literature shows how mindfulness-based interventions can be beneficial for youths in preventing emotional and behavioral problems (Shorey et al., 2015), improving mental health outcomes (Ortiz & Sibinga, 2017), and being helpful in psychological interventions with perpetrators that have been exposed to violence during their childhood (Ferreira et al., 2019). In addition, mindfulness interventions seem to have a positive effect on the decrease of impulsive (Shorey et al., 2015) and aggressive behaviors in adolescents, thus improving anger management (Vandana & Singh, 2017). Along these lines, the results of a meta-analysis found that mindfulness training interventions generated benefits in terms of the development of dispositional mindfulness (DM) dimensions (Quaglia et al., 2016). Thus, DM dimensions seemed to have a positive effect on the decrease of psychological symptoms in adolescence and youths (Zoogman et al., 2015).

DM is the innate ability to consciously attend to the present moment, focusing on sensations, emotions, and thoughts under a non-judging paradigm (Williams et al., 2007), an ability that everyone has to a greater or lesser extent (Kabat-Zinn, 2003). In recent years, there has been growing interest in DM as a protective factor due to its positive impact on mental health in adult (Bergin & Pakenham, 2016), adolescent (Cortazar & Calvete, 2023), and young adult samples (Calvete et al., 2017; Cortazar & Calvete, 2019; Salem & Karlin, 2023). Nevertheless, there are challenges to drawing conclusions about their beneficial role in mental health, mainly given there is no clear consensus on the construct’s definition and evaluation. Some authors define it unidimensionally (Brown & Ryan, 2003), while others consider it a multidimensional construct (Baer et al., 2006; Bishop et al., 2004). However, there is growing consensus over its multifaceted definitions. For instance, Ruth Baer et al. (2006) conducted an exploratory factorial analysis of the five most used tools for DM measurement, suggesting the existence of five distinguishable variables: describing (translating emotional aspects into words); non-judging (accepting sensations, emotions, and thoughts without judgement); acting with awareness (paying attention to perceptions, thoughts, and emotions); observing (disposition to focus on perceptions, thoughts, and emotions); and non-reactivity (preventing getting carried away by thoughts and feelings).

Following the five-factor model, recent studies suggest that the role of each dimension will depend on the psychological problem. Describing was associated with lower internal symptomatology (i.e., anxiety and depression) in young university student samples (Bergin & Pakenham, 2016). Non-judging is associated with lower levels of internalizing symptomatology both at the cross-sectional (Van Son et al., 2015) and the longitudinal level in adult samples, while other studies identify it as contributing to externalizing symptoms in adolescents (Cortazar & Calvete, 2019). Acting with awareness has the most benefits toward internalizing symptoms in both adults (Bullis et al., 2014) and adolescents (Ciesla et al., 2012; Cortazar & Calvete, 2019). Furthermore, this dimension presented benefits against externalizing symptomatology in adolescents (Calvete et al., 2017; Cortazar & Calvete, 2019) and in young adult samples (Shorey et al., 2014). Observing showed an ambivalent role depending on the sample type (Aldao et al., 2010). In a general sample, results identified it as prejudicial against internalizing symptomatology in adults (Bullis et al., 2014; Christopher et al., 2012), adolescents, and emerging adults (Royuela-Colomer & Calvete, 2016). Nevertheless, a recent study found that the ability to observe internal emotions benefited the reduction of externalizing symptomatology, as well as aggressive behaviors over time (Cortazar & Calvete, 2019). Finally, non-reactivity is negatively associated with internalizing symptoms (Van Son et al., 2015) and externalizing symptoms, reducing physical and psychological aggressions perpetrated in dating violence in young adults’ samples (Shorey et al., 2014), yet is not associated with either internalizing or externalizing symptomatology in studies of adolescents’ samples (Cortazar & Calvete, 2019).

Some studies have assessed the role of DM dimensions in stressful situations. Nerea Cortazar and Esther Calvete (2019) pointed out that describing and non-judging predicted lower internalizing symptomatology when adolescents were exposed to higher levels of stress. Describing was also protective against externalizing symptomatology, following other findings that suggest this ability holds protective potential against stress (Bergin & Pakenham, 2016; Bullis et al., 2014). Alethea Desrosiers et al. (2013) consider it especially important in highly stressful situations. Meanwhile, non-judging, non-reactivity, and acting with awareness did not associate with lower levels of externalizing symptomatology in stressful situations (Cortazar & Calvete, 2019). Nevertheless, in the study by Calvete et al. (2017), acting with awareness protected adolescents against externalizing symptomatology in stressful situations. Finally, observing seems to increase internalizing symptomatology in stressful situations (Cortazar & Calvete, 2019).

In spite of the growing evidence centered on the benefits of DM dimensions on mental health, few studies have focused on DM’s potential to prevent externalizing symptoms compared to internalizing ones (Calvete et al., 2017; Cortazar & Calvete, 2019). Moreover, the different facets of DM play a beneficial role in contexts characterized by high levels of stress (Cortazar & Calvete, 2019; Desrosiers et al., 2013). These types of contexts may share features with situations of family violence, suggesting that facets of DM could also play a protective role against CPV in contexts of high exposure to family violence. Few studies have assessed the role of the five DM dimensions, and none of them evaluate the possible relations with CPV aggressive behavior. The main aim of this study was to explore the role of DM dimensions in the relation between exposure to family violence and CPV (physical and psychological) toward mother and father.

Many studies have tried to identify CPV risk factors (Calvete & Veytia, 2018; Del Hoyo-Bilbao et al., 2020). Research identifies exposure to family violence as a potential predictor of CPV (Contreras et al., 2020; Gallego et al., 2019), but there are no studies analyzing the interacting factors. Some DM dimensions have demonstrated a protective role against aggressive behaviors (Calvete et al., 2017; Cortazar & Calvete, 2019). Therefore, the general aim of this study was to assess the role of the five DM dimensions in the relation between exposure to family violence and CPV (physical and psychological) toward mother and father (Figure 1). Taking the previous review into consideration, we first expected to find positive associations between exposure to family violence and physical and psychological CPV toward mother and father. Second, we expected negative associations between all five DM dimensions and all CPV forms toward mothers and fathers. Finally, the five DM dimensions were expected to buffer the relation between exposure to family violence and CPV (physical and psychological) toward mother and father.

The sample was intentional, and it was composed of 415 adolescents and young adults, 311 girls (74.9%), 103 boys (24.8%), and 1 non-binary person (0.2%). The ages of the participants ranked from 18 to 25 years (M_age = 22.54; SD_age = 1.72). 97.8% of the participants were from Spain, 0.9% were from other European countries (0.2% Germans, 0.5% Rumanians, and 0.2% Swiss), and the remaining 1.3% were from South America (0.2% from Brazil, 0.7% from Colombia, 0.2% from Uruguay, and 0.2% from Venezuela). Regarding the family situation, 81% of the participants lived with their parental figures, 6.8% with friends, 5.8% with a partner, 1.9% lived alone, and 4.3% did not specify their situation. Finally, regarding employment status, 72.1% were students, 20.7% were employed, 3.8% were unemployed, and 3% did not specify their employment situation.

Child-to-parent violence. For CPV assessment, the Child-to-Parent Aggression Questionnaire was used (CPAQ; Calvete et al., 2013). This self-report questionnaire is formed by 20 parallel items, 10 items directed to the maternal figure and another 10 to the paternal figure. Questionnaire items measure different physical (seven items) (e.g., “You have kicked or punched”) and psychological CPV (three items) (e.g., “You have insulted or said bad words”) behaviors. The response scale was Likert-type with four response options ranging from 0 (never) to 3 (often, it happened six or more times). CPAQ has shown excellent psychometric properties in Spanish adolescent community samples (Calvete, Morea et al., 2019; Calvete & Orue, 2016) and emerging adults. In this study, Cronbach alpha coefficients were: psychological CPV mother = .72, physical CPV mother = .72, psychological CPV father = .70, physical CPV father = .82.

Dispositional mindfulness. For the assessment of DM, the Spanish version of the Five Facet Mindfulness Questionnaire (FFMQ; Cebolla et al., 2012) was administered. This questionnaire was compounded by 39 self-report items that assesses the five DM dimensions: describing (eight items) (e.g., “I am good at finding words to describe my feelings”), non-judging (eight items) (e.g., “I criticize myself for having absurd and irrational ideas”), acting with awareness (eight items) (inverse item; e.g., “I get easily distracted”), observing (eight items) (e.g., “I perceive smells and the aroma of things”), and non-reactivity (seven items) (inverse item; e.g., “I perceive my feelings and emotions without opposing them”). The response scale was Likert type with five response options ranging from 1 (never) to 5 (usually). FFMQ-A has shown acceptable psychometric properties in adolescent and emerging adult community samples (Cortazar & Calvete, 2019). In this study, Cronbach alpha coefficients were: .91 for describing; .89 for non-judging; .73 for observing; and .78 for non-reactivity.

Exposure to family violence. Exposure to family violence was assessed using the Exposure to Violence Scale (CEV; Orue & Calvete, 2010). This self-reported questionnaire is compounded by 21 items with a Likert-point scale of five options ranging from 0 (never) to 4 (everyday). This study used the sub-scale of exposure to family violence compound by six items that assesses direct violence (three items) (e.g., “How often did they hit or physically harm you at home”), and indirect violence (three items) (e.g., “How often have you seen a person hitting or physically harming someone at home”). This tool had appropriate psychometric properties in general population samples at the national level (Del Hoyo-Bilbao et al., 2020; Orue & Calvete, 2010). The main study used the global index of exposure to family violence, which has a .89 Cronbach alpha coefficient.

Data collection and contact with the participants were carried out through social media and QR codes that were published around different universities of Bizkaia. First, participants were informed of the aim and nature of the study. Second, individuals who wanted to participate received an active informed consent form where detailed information about their willingness to participate, anonymity, and data confidentiality was provided, along with a notification that results could be requested at any moment if they so desired it. Thirdly, contact with the researchers was indicated in case they needed to solve any doubts regarding the study. The study was conducted with young adults, with age as the sole exclusion criterion. Accordingly, only participants aged 18 years or older were eligible to participate. Two of the participants rejected the consent for data use (0.48%). The study meets the ethical standards required in psychological research.

Correlation analysis among variables of the study and descriptive statistics were analyzed through IBM SPSS Statistics 26 (2012). structural equation model was carried out using version 8.80 of LISREL attending to the robust maximum likelihood method (rml). The paths were analyzed according to the Satorra-Bentler χ²-(S-B χ²) scale and requires the estimation of the variances and covariances. The hypothetical model presented direct paths from exposure to family violence and the five DM dimensions (describing, non-judging, acting with awareness, observing, and non-reactivity) to physical CPV toward mothers, physical CPV toward fathers, psychological CPV toward mothers, and physical CPV toward fathers. The model also included paths from interactions between exposure to family violence and the five dimensions of DM (exposure to family violence X describing, exposure to family violence X non-judging, exposure to family violence X acting with awareness, exposure to family violence X observing, and exposure to family violence X non-reactivity) to physical and psychological CPV toward mothers and fathers. Independent variables were transformed to z punctuations to generate moderation terms (Frazier et al., 2004). The adequation of the model was analyzed according to the fit indexes of the model: NNFI indexes were .90 or higher, RMSEA indexes were .06 or lower, and CFI indexes were .90 or higher; the model fit perfectly (Hu & Bentler, 1999).

Table 1 shows the correlation coefficients and descriptive analysis among all studied variables. On the one hand, exposure to family violence was positively associated with all CPV forms and with describing and observing dimensions, while negatively with non-judging. On the other hand, non-judging was negatively associated with psychological CPV toward the mother and positively with physical CPV toward the father. Non-reactivity was negatively associated with all CPV forms except for physical CPV toward fathers. Table 2 shows the difference between boys and girls in all the study variables. The results show no gender differences between variables, excluding the acting with awareness dimension, where girls obtained higher scores with a small effect size.

The final model showed excellent Satorra-Bentler adjustment χ²(32, N = 415) = 56.97, RMSEA = .043, (90% CI: .070; .094), NNFI = .96, CFI = .99 y NFI = .97. The model explained 14% of the variance regarding psychological CPV toward mothers, 28% of physical CPV toward mothers, 16% of psychological CPV toward fathers, and 29% of physical CPV toward fathers. Figure 2 shows the significant paths obtained in the parsimonious model. The results showed that exposure to family violence was associated with all CPV forms. Non-judging was negatively associated with psychological CPV toward mothers and positively with psychological CPV toward fathers. Observing was negatively associated with physical CPV toward mothers.

Three moderation terms were associated with changes in CPV: exposure to family violence X describing was associated with physical CPV toward mothers and fathers. Exposure to family violence X observing was negatively associated with psychological CPV among fathers. Exposure to family violence X non-reactivity was negatively associated with psychological CPV among mothers and fathers and with physical CPV toward mothers.

The significant interactions are shown in various figures for their interpretation. Figure 3 shows the structural interaction between exposure to family violence X describing in relation to physical CPV among father. Thus, when the exposure to family violence is high, physical CPV toward fathers increases only when describing abilities are low. The interaction between exposure to family violence X describing in relation with physical CPV toward father was not reflected because the interaction form was similar. Figure 4 shows the association between exposure to family violence X observing and psychological CPV toward fathers. Thus, when exposure to family violence is high and the observing dimension is low, psychological CPV toward fathers increases. Figure 5 indicates the association obtained between exposure to family violence X non-reactivity in relation to physical CPV toward mothers. Thus, when exposure to family violence is high and the levels of non-reactivity are low, physical CPV toward mothers increases. Figure 6 shows the association between exposure to family violence x non-reactivity and psychological CPV toward fathers. Thus, when exposure to family violence is high, and non-reactivity is low, psychological CPV toward fathers increases. The interaction between exposure to family violence X non-reactivity in the relation with psychological CPV toward mothers was not reflected because the interaction form was similar.