Epidemiological studies allowed us to conclude that the worldwide prevalence of attention-deficit/hyperactivity disorder (ADHD) ranges from 2% to 10% (Catalá-López et al., 2012; Danielson et al., 2018; Polanczyk et al., 2007; Rowland et al., 2002; Song et al., 2021). ADHD is a neurodevelopmental disorder characterized by pervasive inattentive symptoms (impaired attention abilities), disorganization, and impulsive/hyperactivity (American Psychological Association [APA, 2013]). A specific ADHD subtype (inattentive or hyperactive subtype) can be diagnosed depending on the predominance of inattentive or hyperactivity symptoms, while a combined subtype of ADHD can be diagnosed when an individual presents pervasive symptoms of both types (Epstein & Loren, 2013).

Several meta-analyses and systematic reviews have linked ADHD with violence proneness and/or antisocial behaviors, establishing a positive association between both (Mohr-Jensen & Steinhausen, 2016; Pratt et al., 2002; Saylor & Amann, 2016). From a neurodevelopmental perspective, the absence of a correct early ADHD diagnosis and its concurrent treatment during childhood might entail a considerable risk for the onset of antisocial behaviors during adolescence and adulthood, which is reflected by the number of arrests and/or incarcerations, as well as recidivism rates after incarceration or interventions (Mohr-Jensen & Steinhausen, 2016; Pratt et al., 2002; Saylor and Amann, 2016). Moreover, the type of violence perpetrated by ADHD individuals tend to be characterized by reactivity and/or impulsivity, more often perpetrated by individuals with a combined subtype ADHD (Saylor & Amann, 2016), that is, individuals with concurrent inattention and hyperactivity symptoms.

Due to the fact that emotional and behavioral dysregulation (e.g., violence proneness, antisocial behaviors…) tend to be affected by the severity of ADHD symptoms (Mansour et al., 2017), a correct diagnosis of ADHD becomes of great clinical interest as it could reduce reoffending or recidivism after an individual completes tailored interventions or their prison sentence (Loyer Carbonneau et al., 2021). In this regard, the rate of criminal recidivism decreased among those patients who were diagnosed with ADHD and were receiving their corresponding treatment (e.g., receiving stimulants, psychotherapy, etc.) (Buitelaar et al., 2014; Lichtenstein & Larsson, 2013; Pappadopulos et al., 2006). The deeper our knowledge regarding the main causes of violence (e.g., correct ADHD diagnosis and moderator factors), the better our ability for designing more specific interventions and therefore reducing recidivism rates.

In recent years, the importance of ADHD for perpetrating or being a victim of intimate partner violence (IPV) has been highlighted (Buitelaar et al., 2020; Buitelaar et al., 2014; Fang et al., 2010; Theriault and Holmberg, 2001; Wymbs et al., 2012; Wymbs et al., 2017). Most of this violence among heterosexual couples is perpetrated by male against their female partners. Estimates of worldwide prevalence of this kind of violence revealed that approximately 27% of women aged 15 to 49 years have suffered some type of physical or sexual violence from their male partner throughout their life (Martín-Fernández et al., 2019, 2020; World Health Organization [WHO, 2021]).

It cannot be concluded that ADHD necessarily or directly facilitates IPV perpetration. Alcohol and drug misuse and type B personality traits (e.g., antisocial, borderline, etc.), among others, may act as confounding factors, that is, ADHD in combination with the above-mentioned factors might drive individuals to carry out risky behaviors such as violence, and avoiding those confounding variables considerably diminishes the ability of ADHD to predict IPV (Buitelaar et al., 2020). However, these variables (e.g., alcohol or drug misuse, antisocial personality traits, etc.) do not offer too much information regarding the cognitive processing of surrounding information of individuals affected by this disorder. Therefore, it becomes important to explore cognitive variables that clarify how individuals with ADHD might be prone to violence, given that these processes are closely and/or directly related with behavioral and emotional regulation, which in turn, affect violence proneness.

Applying a cognitive model for understanding which mechanisms explain violence proneness in individuals affected by ADHD, it can be stated the existence of considerable differences between individuals affected by ADHD and those unaffected by this or other disorders in several cognitive domains such as IQ, processing speed, working memory, attention, cognitive flexibility, planning, and inhibition abilities (Loyer-Carbonneau et al., 2021; Pievsky & McGrath, 2018), as well as in facial emotion recognition (Bora & Pantelis, 2016). The broader the cognitive deficits in ADHD individuals, the higher their proneness to anger expression (McDonagh et al., 2019) and emotional dysregulations (Banaschewski et al., 2012). Additionally, cognitive deficits have also been associated with IPV proneness (Humenik et al., 2020; Romero-Martínez, Lila, et al., 2022; Romero-Martínez, Santirso, et al., 2022). This can be explained by a reduced ability to process surrounding information and act or make consequent decisions with this analysis. In other words, when IPV perpetrators present a lowered ability for processing emotional stimulus, verbalizing inner states, coping with stressful stimulus, or making decisions, there are a greater risk of IPV perpetration (Expósito-Álvarez et al., 2021; Howieson, 2019; Humenik et al., 2020; Lezak et al., 2012). Their low ability for coping with ambiguous stimulus (e.g., couple arguments, stressful workplace situations, etc.) might facilitate violence intake when those deficits are combined with other hostile cognitive schemas (e.g., low perceived severity of IPV, sexism, etc.; Juarros-Basterretxea et al., 2022; Martín-Fernández et al., 2022), and a lack of abilities for searching alternative ways to respond. Assessing neuropsychological variables might clarify the association between ADHD and IPV recidivism, which in turn, would help us to prevent or reduce it. Unfortunately, to date there is a gap in the scientific literature regarding how these variables explain the potential association between ADHD and IPV.

One of the main problems with IPV perpetrators is the considerable percentage of them that abandon standard batterer interventions prematurely (i.e., dropout; Arce et al., 2020; Lila et al., 2020; Santirso, Gilchrist, et al., 2020; Soleymani et al., 2018). This is a key problem for IPV perpetrators, as dropout constitutes a valuable factor when predicting IPV perpetrators’ recidivism (Lila et al., 2019). Nonetheless, as far as we know, the usefulness of ADHD symptoms to explain dropout in IPV perpetrators has not yet been analyzed. However, there is evidence that highlights ADHD as a risk factor for recidivism (Buitelaar et al., 2014; Lichtenstein & Larsson, 2013; Pappadopulos et al., 2006) and points out how cognitive deficits might allow to partly explain IPV perpetrators’ dropout and recidivism (Romero-Martínez et al., 2021; Romero-Martínez, Lila, et al., 2022; Romero-Martínez, Santirso, et al., 2022). Hence, it would be necessary to conduct empirical research to assess whether ADHD in combination with cognitive deficits in IPV perpetrators considerably increases our ability to explain dropout and recidivism in those men.

In sum, the present study was aimed primarily at answering three questions: 1) whether IPV perpetrators affected by ADHD would present a differentiated neuropsychological profile from IPV perpetrators without ADHD and, especially, in comparison with non-violent men (or controls). In line with what has been previously concluded in this field of research (Bora & Pantelis, 2016; Loyer-Carbonneau et al., 2021; Pievsky & McGrath, 2018), we expected that ADHD IPV perpetrators (combined type) would present higher cognitive impairments (i.e., working memory, attention, cognitive flexibility, planning abilities, and emotion decoding) than IPV perpetrators without ADHD and controls, but not in comparison with IPV perpetrators affected by ADHD inattentive type; 2) we also aimed to assess whether to abandon the intervention program before it ends (dropout) and recidivism was directly related to an ADHD diagnosis in IPV perpetrators; accordingly, based on previous empirical research in this field (Mohr-Jensen & Steinhausen, 2016; Pratt et al., 2002; Saylor & Amann, 2016), we hypothesized that the higher the severity of the diagnosis (e.g., ADHD combined subtype vs unaffected or inattentive subtype), the greater the dropout, and official recidivism rates; last, 3) the main novelty of this study was to measure whether neuropsychological impairments would mediate the association between ADHD symptoms with dropout and recidivism among IPV perpetrators. To date, a scarce number of studies have explored the mediator role of neuropsychological deficits between the above-mentioned variables. However, neuropsychological deficits have been demonstrated to have a considerable predictor value for dropout and recidivism in violent individuals, including IPV perpetrators (Hundozi et al., 2016; Meijers et al., 2017; Miura & Fuchigami, 2017; Norman et al., 2022; Romero-Martínez et al., 2021; Romero-Martínez, 2022; Sánchez de Ribera et al., 2021). Therefore, we consider that neuropsychological deficits, specifically executive functioning, would mediate the association between ADHD, specifically the combined subtype, with dropout and recidivism.

As can be seen in Table 1, we checked for differences in age, demographic variables, alcohol, and drug misuse between groups. Even though there were no differences in any of the variables considered, that is, age, number of children, educational level, nationality, and/or working status, groups differed in terms of alcohol, drug misuse, and physical and psychological IPV perpetration. Thus, alcohol and drug misuse were considered in a later analysis to control their potential role.

The main three objectives of our study were: 1) to find out whether IPV perpetrators with and without ADHD differed among themselves and compared to non-violent men in several cognitive domains evaluated by neuropsychological tests, 2) whether there were differences between IPV perpetrators (with and without ADHD) relative to dropout (treatment compliance) and official recidivism and, most importantly, 3) if the ADHD diagnosis and neuropsychological performance interact to predict dropout and official recidivism among IPV perpetrators.

Our results indicate that all the groups of IPV perpetrators (with and without ADHD) presented worse performance in all cognitive domains assessed (i.e., IQ, working memory, verbal fluency, cognitive flexibility, planning abilities, and emotion decoding abilities) than controls. Additionally, ADHD IPV perpetrators presented worse performance in all cognitive domains than unimpaired IPV perpetrators, but not in comparison with inattention IPV perpetrators. Additionally, IPV perpetrators affected by ADHD presented the highest rate of dropout and official recidivism. Lastly, even though neuropsychological impairments and IPV perpetrators’ group variables separately significantly explained dropout and official recidivism, their interaction did not increase the amount of explained variance and, therefore, were not significant to predict each one. All these results remained significant after controlling the effect of drug misuse.

We initially hypothesized that ADHD IPV perpetrators (combined subtype) would present worse performance compared to unimpaired IPV perpetrators and controls, but not among subtypes of ADHD groups, which was in line with previous research in this field (Loyer-Carbonneau et al., 2021; Pievsky & McGrath, 2018). Sure enough, our data pointed out that IPV perpetrators with ADHD combined subtype presented neuropsychological differences in all cognitive domains in comparison with unimpaired IPV perpetrators and controls, planning abilities and emotion decoding abilities compared to unimpaired IPV perpetrators. Moreover, it was particularly interesting that only a group of IPV perpetrators, specifically ADHD IPV perpetrators, presented worse working memory performance compared to unimpaired IPV perpetrators and controls. Our data also reinforced that all groups of IPV perpetrators presented worse neuropsychological performance than controls, as it has been previously stated (Babcock et al., 2008; Cohen et al., 2003; Easton et al., 2008; Romero-Martínez, Lila, et al., 2022; Walling et al., 2012), with the effect size of these differences being moderate or large, specifically, for verbal fluency and cognitive flexibility, and low for the rest of the variables. Therefore, IPV perpetrators’ performance could not be considered as a deficit (1.5-2 SD below performance of control group), situating their performance slightly below the control group. A priori, those neuropsychological differences could be attributed to different patterns of alcohol and/or drug misuse acting as confounding factors. In fact, the presence of a substance use disorder might be masking the performance of individuals affected by ADHD (Slobodin, 2020). Curiously all these differences remained significant after controlling their effect. Thus, we could consider both factors as complementary risk factors instead of drugs developing a moderator role between ADHD and IPV, as their inclusion in statistical analysis increased the amount of explained variance without affecting the significant effect of group above neuropsychological performance.

Results of the rate of dropout and recidivism among IPV perpetrators with different severity of ADHD symptoms were in line with conclusions proposed by Wymbs et al. (2017) but contrasted with those obtained by the same authors (Wymbs et al., 2019; Wymbs et al., 2015). That is, our study highlighted that ADHD (combined subtype) presented the highest rate of dropout and recidivism, even after controlling the effect of alcohol and drug misuse. Hence, as stated above, alcohol and drug misuse should be considered a relatively independent factor in the association between ADHD and IPV perpetration. In any case, we need to keep in mind that we assessed two final intervention outcomes (e.g., dropout and official recidivism) instead of IPV levels. Furthermore, we considered a sample of adults convicted of IPV, whose drug misuse levels tend to be higher than normative population (Hines & McDouglas, 2012), instead of young normative adults without IPV conviction as Wymbs et al. (2015), Wymbs et al. (2017), and Wymbs et al. (2019) did before. Given that we cannot attribute violence proneness in ADHD individuals to drug misuse, our data highlighted the need to incorporate or develop coadjutant intervention strategies focused on ADHD symptoms differentiated from others, paying attention to substance use disorders, although this does not mean or contradict that certain individuals need both intervention strategies together to reduce dropout and recidivism by attending to individual needs.

The main novelty of this research was the results related to the third objective, to measure whether neuropsychological performance joined with the ADHD diagnosis to explain dropout and official recidivism. We hypothesized that ADHD, particularly combined subtype, with the highest neuropsychological deficits would explain IPV perpetrators’ dropout, as well as official recidivism. Contrary to our expectations, the inclusion of both factors increased the amount of explained variance, but the interaction between both did not reach significance for explaining the above-mentioned dependent variables. Thus, as stated before, we cannot consider that the neuropsychological deficits mediate and/or moderate the association between ADHD symptoms and dropout and official recidivism for all the IPV perpetrators with combined or inattention subtypes, being, hence, both risk factors relatively independent. Thus, these factors should be treated independently (e.g., pharmacological + cognitive training).

Working memory functioning explained dropout, whereas IQ, working memory, and cognitive flexibility significantly explained official recidivism. To understand the reduced rate of treatment compliance, we can speculate that ADHD IPV perpetrators might feel overwhelmed by the content of intervention programs, as they did not have the appropriate cognitive skills to deal with a complex content such as interrogating their own feelings, changing hostile cognitive schemas regarding women, improving decision-making processes, etc. In this sense, working memory processes sustain other more complex processes for making and planning decisions. This, in turn, might considerably increase the risk of recidivism, given that they do not change certain hostile cognitive schemas or behavioral regulation, among others, to deal appropriately with certain conflictive and demanding context.

Our study highlights the need for complementing current standard batterer programs with cognitive training attending to IPV perpetrators’ cognitive needs in order to decrease the risk of dropout and official recidivism. In this sense, there is an example of a randomized controlled pilot study, which compared a standard intervention with another group of IPV perpetrators receiving this program in combination with cognitive training (e.g., including pen and pencil tasks and videos, among others). This last group experienced considerable improvements in several cognitive domains, such as speed processing and cognitive flexibility, as well as a considerable reduction in the risk of reoffending (Romero-Martínez, Santirso, et al., 2022). Cognitive training has also been concluded to present a certain, although limited, positive impact by reducing working memory and executive functioning deficits in people affected by ADHD, which also has an impact on reducing the severity of ADHD behavioral dysregulation (Cortese et al., 2015). Hence, an initial neuropsychological assessment to identify cognitive needs and other neurological diagnosis and, consequently, implementing coadjutant intervention programs should be incorporated.

The present study had several strengths and offer an interesting perspective of IPV, but there were also several limitations which should be kept in mind for conducting future research. First, the cross-sectional nature of the study design and the relatively reduced sample limited the causality of the associations between the variables. Thus, it would be necessary to replicate the results in future studies with a longitudinal study and a larger sample size. Second, we would like to point out that it would be suitable for future studies to measure temporal stability of ADHD diagnosis. Furthermore, it would be important to take into account previous ADHD diagnosis from other mental health professionals or whether participants were diagnosed during their childhood or adolescence. Particularly, whether IPV perpetrators experience notorious changes in several cognitive abilities after intervention. Furthermore, it would be important to include other ADHD screening tools during the initial assessment of IPV perpetrators to further complement the diagnosis by facilitators. In fact, the role of CPT-3 performance to clearly differentiate ADHD from other disorder diagnosis, such as reading disorders or dyslexia, has been criticized (McGee et al., 2000; Miranda et al., 2012). Third, it would be important to consider a larger number of neuropsychological tests to determine whether IPV perpetrator typologies clearly differ in their cognitive profiles (e.g., decision-making, inhibitory control, planning abilities with more complex tests, etc.). In this sense, we would like to highlight the need for developing new instruments for assessing emotion decoding abilities, as the Eyes Test showed poor internal consistency, but in line with what is indicated in a systematic review (Oakley et al., 2016). It is important to bear in mind that this low internal consistency reduces the value for conclusions. Four, participants were codified as reoffenders and non-reoffenders, but this did not offer too much information regarding the type of re-offense (e.g., physical, psychological violence, violation of a restraining order, etc.). This would help in future studies for a wider understanding of reoffending/recidivism. Five, it would be important to include other groups of IPV perpetrators (e.g., ADHD impulsive subtype) and/or controls affected by ADHD but without IPV conviction for contrasting and complementing current results. Furthermore, we considered it particularly important to assess how neuropsychological performance of IPV perpetrators would be associated with different strategies when coping with couple arguments in future research.

In conclusion, our data underline the importance of ADHD, especially the combined subtype, and the neuropsychological deficits for explaining dropout and official recidivism in IPV perpetrators, independently of alcohol and drug misuse. Moreover, we support the need for developing coadjutant cognitive training programs for neuropsychological dysfunctions and treatment for ADHD symptoms, which clearly interfere with current adherence to intervention programs for IPV perpetrators and recidivism. In addition, we do not reject incorporating other alternatives such as pharmacological strategies and non-invasive brain techniques, when necessary, which could increase the success of current interventions because they make it possible to increase the success of psychological interventions. Thus, future research in this field should consider all these variables, which will allow us to build a biopsychosocial model of treatment compliance and recidivism.