Abstract
ADHD is a neurodevelopmental disorder characterized by symptoms of inattention, hyperactivity, and impulsivity, causing clinically significant impairment in social, academic, and/or occupational functioning (American Psychiatric Association [APA], 2013). ADHD has traditionally been emphasized in children, but there is increasing recognition of its relevance in adults (Faraone et al., 2015). Parenting is an adult domain requiring cognitive, emotional, and behavioral endeavor and, thus, may be especially affected by elevated levels of ADHD symptoms (Johnston, Mash, Miller, & Ninowski, 2012; Mokrova, O’Brien, Calkins, & Keane, 2010). Studies find that parents with ADHD symptoms tend to use more ineffective discipline strategies, including overreactive and inconsistent discipline (Babinski et al., 2016; Banks, Ninowski, Mash, & Semple, 2008; Chen & Johnston, 2007; Chronis-Tuscano et al., 2008; Harvey, Danforth, McKee, Ulaszek, & Friedman, 2003; Mokrova et al., 2010; Murray & Johnston, 2006). In addition, parental ADHD symptoms have been associated with more parenting negativity and criticism of children (Babinski et al., 2016; Chronis-Tuscano et al., 2008; Harvey et al., 2003; Mazursky-Horowitz et al., 2015) and with less positive reinforcement (Chen & Johnston, 2007). Furthermore, even when considering the effect of variables known to affect both maternal ADHD symptoms and ineffective parenting, such as depression symptoms or child behavior problems, the relations between maternal ADHD symptoms and parenting remain significant (Chen & Johnston, 2007; Chronis-Tuscano et al., 2008). At the same time, however, studies have not found significant associations between parental ADHD symptoms and parental warmth, sensitivity, and emotional responsivity to children (Chen & Johnston, 2007; Chronis-Tuscano et al., 2008; Ellis & Nigg, 2009; Harvey et al., 2003; Mokrova et al., 2010).
A recent meta-analysis concluded that although parental ADHD symptoms are consistently associated with parenting behaviors, the effect size and the findings vary across studies (Park, Hudec, & Johnston, 2017). The varied findings suggest complex relations between parental ADHD symptoms and parenting, possibly involving individual parental factors as moderators. In Belsky’s determinant model of parenting (Belsky, 1984), parents’ psychological resources are the most proximal and potent predictors of parenting behavior. More recent research has demonstrated that parental cognitive control capacities, such as executive functioning, influence parenting behaviors (Crandall, Deater-Deckard, & Riley, 2015; Deater-Deckard, Wang, Chen, & Bell, 2012; Shaffer & Obradović, 2017). Individuals with ADHD typically experience problems in various components of executive functioning (Willcutt, Doyle, Nigg, Faraone, & Pennington, 2005) and self-regulation (Shiels & Hawk, 2010), and problems in either area could result in parenting impairments (Banks et al., 2008; Chen & Johnston, 2007; Johnston et al., 2012).
Inhibitory control is one of the primary domains of executive functioning (Miyake et al., 2000). It entails the ability to inhibit or restrain a prepotent dominant response to execute an appropriate behavior or make an effective response to achieve a goal (Barkley, 1997; Miyake et al., 2000). Inhibitory control has a central role in an individual’s ability to adaptively regulate cognition, emotion, and behavior (Posner & Rothbart, 2007), and has been proposed as a critical element of optimal parenting behaviors (Sanders & Mazzucchelli, 2013). Many of the parenting behaviors identified as less optimal among parents with elevated levels of ADHD, such as overreactive, intrusive, or inconsistent parenting, may be linked to parental deficits in self-regulation. Although scant, some empirical work has documented associations between parental inhibitory control and parenting among parents of typically developing children (Bridgett, Kanya, Rutherford, & Mayes, 2017; Shaffer & Obradović, 2017; Sturge-Apple, Davies, Cicchetti, Hentges, & Coe, 2017). In the case of mothers specifically, maternal inhibitory control has been positively associated with sensitive and supportive parenting during parent–child interactions among kindergarten-age children (Shaffer & Obradović, 2017). In addition, maternal inhibitory control has been positively associated with the ability to maintain sensitivity when eliciting compliance from 3- to 5-year-old children (Sturge-Apple, Jones, & Suor, 2017), and negatively associated with the negative parenting of infants (Bridgett et al., 2017). Overall, inhibitory control may be critical to parents’ ability to make reflective decisions about their actions in response to their child’s behaviors, to hold back from intervening, to suppress interference to maintain focus, and to regulate their own behavioral and emotional responses to appropriately respond to their child’s noncompliance or negative emotions (Shaffer & Obradović, 2017; Sturge-Apple et al., 2017). Given these findings, parental inhibitory control is a possible explanation of the link between parental ADHD and parenting impairments.
Children are equally important determinants of parenting (Belsky, 1984; Collins, Maccoby, Steinberg, Hetherington, & Bornstein, 2000). The conceptualization of parenting as a reaction to children, not an action (Kerr & Stattin, 2003), emphasizes the relevance of parents’ control capacities, especially in challenging parent–child interactions and in child noncompliance and/or misbehavior (Deater-Deckard et al., 2012). In the context of the high hereditary nature of ADHD (Nikolas & Burt, 2010), where approximately 40% to 55% of children with ADHD have at least one parent with ADHD (Takeda et al., 2010), it is probable that parenting challenges among parents with elevated symptoms of ADHD reflect, to some extent, the presence of ADHD in their children (Johnston & Lee-Flynn, 2011). Children with ADHD experience behavioral, social, and emotional problems, as well as functional impairments in key domains, such as academic performance and peer relations (Becker, Langberg, Vaughn, & Epstein, 2012; Connor et al., 2003; Frazier, Youngstrom, Glutting, & Watkins, 2007). In addition, parents of children and adolescents with ADHD report more frequent parent–child conflicts (Barkley, 2008; Deault, 2010; Edwards, Barkley, Laneri, Fletcher, & Metevia, 2001; Markel & Wiener, 2014; Mulligan et al., 2013) and a greater tendency to mutually express negative behaviors, such as exerting control or noncompliance, than do parents of typically developing children (Edwards et al., 2001). Not surprisingly, children with ADHD place greater demands on parenting and evoke more overreactive, less patient, and more inconsistent responses from their parents (Johnston & Chronis-Tuscano, 2015; Waschbusch, 2002). The result is likely to be elevated stress in parents of children with ADHD (Deault, 2010; Theule, Wiener, Tannock, & Jenkins, 2013), and this, in turn, may cause parents to become more rejecting, controlling, and reactive (Bögels, Lehtonen, & Restifo, 2010).
It has been suggested that the stressful nature of parenting a child with ADHD will be compounded if parents struggle with their own ADHD, in what researchers call the “similarity–fit” hypothesis (Psychogiou, Daley, Thompson, & Sonuga-Barke, 2007; Psychogiou, Daley, Thompson, & Sonuga-Barke, 2008). Some studies find parents with ADHD who have children with ADHD express greater warmth, sensitivity, positivity, and emotional responsivity, and are less irritable and corrective of their children than parents with ADHD whose children do not have ADHD (Psychogiou et al., 2007; Psychogiou et al., 2008). Other studies demonstrate the opposite effect, that is, the “similarity–misfit” hypothesis, whereby parental ADHD exacerbates the negative impact of child ADHD on parenting (Babinski et al., 2016; Chronis-Tuscano et al., 2008; Harvey et al., 2003; Zisser & Eyberg, 2012). For example, in one study, mothers with ADHD who also had an adolescent with ADHD experienced more conflict than mothers without ADHD who had an adolescent with or without ADHD (i.e., control; Babinski et al., 2016). Accordingly, when considering the impact of parents’ ADHD symptoms on their parenting, we cannot ignore the presence of ADHD in their children.
The Current Study
Despite recent advances in explaining the links between parental ADHD symptoms and parenting, many more issues may be involved in these complex relations. The goal of our study was to extend current understanding and address several gaps in the research. In line with previous findings of the different operation of symptom dimensions in parenting (Park et al., 2017), we examined the effect of both maternal inattention and hyperactivity–impulsivity symptoms. Across studies, there is consistent evidence of the deleterious effect of parents’ inattention symptoms on parenting, including inconsistent discipline, deficits in monitoring child behaviors, and reduced positive and involved parenting (Chen & Johnston, 2007; Harvey et al., 2003; Murray & Johnston, 2006; Williamson, Johnston, Noyes, Stewart, & Weiss, 2017). At the same time, findings on the associations between parents’ hyperactivity–impulsivity symptoms and parenting are inconsistent (Johnston et al., 2012), at times associated with ineffective parenting, such as laxness or harsh parenting (Chen & Johnston, 2007; Harvey et al., 2003), and, at other times, with increased positive parenting (Lui, Johnston, Lee, & Lee-Flynn, 2013).
One possible explanation of the dichotomous findings is that studies have used either self-reported measures of parenting or observations of parent–child interactions. Sole reliance on self-reporting of parental ADHD symptoms and parenting behaviors increases the potential for bias from shared variance. In addition, mothers with ADHD may overestimate their parenting behaviors, inflating the correlation between self-reported harsh parenting and parental ADHD (Lui et al., 2013). Accordingly, we used both observation and self-report measures to assess parenting behaviors. Importantly, by examining inhibitory control via direct neurocognitive assessment, we decreased the impact of mothers’ assessments of their own regulation capacity.
Most research has focused on parent–child interactions in the context of play or teaching activities (Chronis-Tuscano et al., 2008; Lowry, Schatz, & Fabiano, 2015; Lui et al., 2013; Mazursky-Horowitz et al., 2015; Psychogiou et al., 2008; Williamson et al., 2017), but we examined parent–child conflict, specifically conflict discussion, as this may be more challenging for parents to manage (Babinski et al., 2016; Grimbos & Wiener, 2016). In the context of conflict, parents require the ability to be patient, to provide emotional support; explanations, and positive reinforcement; and to respond in a restrained and well-thought-out manner even when the child is not cooperative. Conflicts are frequent between children with ADHD and their parents and are characterized by high emotional intensity (Edwards et al., 2001; Markel & Wiener, 2014; Walther et al., 2012). At the onset of transitioning into adolescence, parent–child relationships become even more conflictual (Allison & Schultz, 2004; Holmbeck, 1996), especially if the child has ADHD (Barkley, Fischer, Edelbrock, & Smallish, 1991; Edwards et al., 2001).
Finally, the specificity of the relations between parental ADHD and parenting needs to be examined by considering several covariates known to predict individual differences in parenting behavior. One is the nature of the child himself or herself. Simply stated, parents’ behaviors are affected by their children (Collins et al., 2000). Managing conflicts, exerting control, and setting limits are more challenging with an overly exuberant or disagreeable child than with a more compliant and inhibited one (Kochanska, Aksan, & Joy, 2007). Children with ADHD demonstrate a higher rate of behavior problems, and these are consistently associated with ineffective and negative parenting (Chen & Johnston, 2007; Chronis-Tuscano et al., 2008; Ellis & Nigg, 2009). Thus, it is important to control for children’s behavior problems when examining parenting behaviors.
Co-occurring behavior problems among children with ADHD present challenges to parents (Deault, 2010). It is, therefore, not surprising that parents of children with ADHD often experience higher levels of parenting stress than parents of typically developing children (Theule et al., 2013; Wiener, Biondic, Grimbos, & Herbert, 2016). Much like parental ADHD symptoms, parenting distress is associated with greater use of authoritarian or permissive parenting styles (Hutchison, Feder, Abar, & Winsler, 2016) and also found to correlate with parental ADHD symptoms (Theule et al., 2013). Clearly, parenting distress needs to be considered in any examination of the specificity of parental ADHD symptoms on parenting.
Our study was a multimethod investigation of the interplay between maternal ADHD symptoms, child ADHD, and parental inhibitory control in contributing to self-reported and observed parenting behaviors among mothers of children with and without ADHD. Specifically, we examined how the association between maternal ADHD symptoms and parenting varied across mothers’ inhibitory control and their children’s ADHD. We hypothesized that maternal ADHD symptoms would be associated with higher levels of negative parenting and lower levels of supportive parenting behaviors. In addition, we hypothesized that maternal inhibitory control would be associated with higher levels of negative dimensions of parenting. Because no prior literature has examined the moderation effect of maternal inhibitory control, we did not formulate hypotheses. Last, in terms of child ADHD, according to the similarity–fit hypothesis, child ADHD should moderate the effect of maternal ADHD symptoms on parenting. However, according to the similarity–misfit model, moderation should occur, but parenting should be more impaired when both parent and child have high levels of ADHD symptoms. Based on recent findings, we hypothesized that child ADHD would moderate the effect of maternal ADHD symptoms on negative dimensions of parenting behavior, thus supporting the similarity–misfit hypothesis (Babinski et al., 2016; Grimbos & Wiener, 2016). However, if supportive parental behavior were included in the analysis, the similarity–fit hypothesis would be supported (Johnston, Williamson, Noyes, Stewart, & Weiss, 2016).
Method
Participants
The sample included 141 mothers (age = 29-52 years, M = 40.18 years, SD = 4.37 years) and their 8- to 12-year-old biological children (M = 9.44 years, SD = 0.98 years), 72 boys and 69 girls, 61 with ADHD and 80 without ADHD. The sample was relatively diverse in terms of socioeconomic status (SES); family income and maternal education levels were diverse across the sample. Participant demographic information is presented in Table 1.
Sample Demographic and Children Characteristics by Child ADHD Status.
p < .01. ***p < .001.
To be included in the ADHD group, children had to be diagnosed with ADHD by a psychiatrist, neurologist, pediatrician, or a family doctor qualified to diagnose ADHD (as per Israeli Health Department official instructions) according to the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; APA, 2000). Children’s diagnosis was confirmed using the following criterion: Mothers completed the ADHD-IV Rating Scale (DuPaul et al., 1998), a well-validated measure reflecting the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; APA, 1994) criteria for ADHD (APA, 2000). The scale includes 18 items, nine of which describe symptoms of inattention (e.g., “Has difficulty sustaining attention in tasks or play activities”) and nine of which describe symptoms of hyperactivity–impulsivity (e.g., “Fidgets with hands or feet or squirms in seat”). Mothers rated each item on a 4-point Likert-type scale ranging from 0 (never or rarely) to 3 (very often). Children were included in the ADHD group only if mothers indicated “often” or “very often” for at least six of the nine symptoms on either the inattentiveness or the hyperactive–impulsive scales. Children included in the control group did not meet this rating criterion. To be included in the study, children could not be diagnosed with a developmental disability (e.g., autism spectrum disorder, intellectual disability) or have a chronic health issue.
Procedure
The study obtained ethical approval from the University Ethics Committee and the chief scientist of the Ministry of Education. Informed consent was obtained from participants at the outset of the study. Families were recruited through advertisements at schools, community centers and other public places, parent support groups, mailings to local ADHD advocacy groups, public bulletin boards, and health professionals working with children diagnosed with ADHD and their families. Two home visits were scheduled, approximately 1 week apart; each lasted about 1.5 hr. The first meeting consisted of an interview discussing the child’s ADHD diagnosis, the administration of two inhibitory control neurocognitive computerized tasks to mothers, and the completion of maternal self-report measures. During the second meeting, mothers and their children engaged in a conflict discussion task. Mothers and children who were being treated by psychostimulants were administrated the computerized task assessment and participated in parent–child interactions after a 24-hr washout before the home visit.
Measures
Maternal ADHD symptoms
Mothers rated their own levels of ADHD symptoms using the Adult ADHD Self-Report Scale (ASRS; Kessler et al., 2005; Zohar & Konfortes, 2010). The ASRS was developed by the World Health Organization to assess ADHD symptoms in adults (Adler et al., 2006; Kessler et al., 2005). It includes 18 items describing ADHD symptoms given in the DSM-IV-TR (APA, 2000). Nine describe inattention symptoms (e.g., “How often are you distracted by activity or noise around you?”), and nine describe hyperactive/impulsive symptoms (e.g., “How often do you find yourself talking too much when you are in a social situation?”). Mothers rated the frequency of the described behaviors in the last 6 months on a 5-point Likert-type scale ranging from 0 (never) to 4 (very often). The Hebrew ASRS version has demonstrated good psychometric properties, 0.6 to 0.9 test–retest reliability, and differentiates individuals with and without prior or current diagnoses of ADHD (Zohar & Konfortes, 2010). In our sample, the internal consistency of inattention symptoms was α = .85; for hyperactivity–impulsivity symptoms, it was α = .83.
Maternal inhibitory control
To assess inhibitory control, mothers were administered a version of the go/no-go computerized task (Tsal, Shalev, & Mevorach, 2005). A sequence of 16 possible stimuli that included a square, a circle, a triangle, or a star in red, blue, green, or yellow appeared in the center of a computer screen. Participants were instructed to respond by pressing the space bar with their preferred index finger as soon as they saw a red square and to withhold responses to all other stimuli. The target appeared in 70% of the trials. Put otherwise, participants had to respond to the majority of trials and inhibit response to the others. The task consisted of a single block of 320 trials preceded by 15 practice trials and lasted approximately 12 min. Scores were based on the percent of commission errors (i.e., false alarms/incorrect responses) out of all trials. Large values of this measure reflect an impaired ability to inhibit prepotent responses to a stimulus. The reported split-half reliability of commission errors in this task is .83 (Geva-Regev, 2015).
Negative and supportive maternal parenting
To assess parenting, we used both mothers’ self-reports and our observations of mother–child interactions. We used the Parenting Scale (Arnold, O’leary, Wolff, & Acker, 1993) to assess dysfunctional parenting practices. Each item on the scale consists of an ineffective parenting behavior paired with an effective counterpart to form high and low anchors on a 7-point scale. Parents indicate their tendency to employ specific discipline strategies, with 7 indicating a high probability of using a dysfunctional discipline strategy and 1 indicating a high probability of using an effective, alternative discipline strategy. The scale includes two subscales, a lax parenting discipline subscale (e.g., “When my child does something I don’t like . . . ,” where 1 = I do something about it every time it happens and 7 = I often let it go) and an overreactive discipline subscale (e.g., “When my child misbehaves . . . ,” where 1 = I speak to him calmly and 7 = I raise my voice and yell; Rhoades & O’Leary, 2007; Salari, Terreros, & Sarkadi, 2012). The coefficient alphas for the overreactive parenting and lax parenting scales were .70 and .80, respectively.
Mother–child conflict discussion
To assess observed parenting negativity and supportive parenting behaviors, we employed a conflict discussion task. In this task, mother–child dyads were asked to have a discussion on a topic on which they had recently argued or disagreed and try to reach a solution. To choose a discussion topic, children and their mothers separately completed the Issues Checklist (Robin & Foster, 1989). The measure lists common parent–child conflict topics, such as bedtime, fights with siblings, homework, and the use of electronics. For each topic, mothers and children were asked to indicate whether they had disagreed about the issue during the last month. Mother and child reports were compared by a research assistant, and the topic rated by both the mother and child as most conflictual was chosen for the first topic of discussion (DeLambo, Ievers-Landis, Drotar, & Quittner, 2004). Children and mothers were instructed to discuss the topic for a full 5 min, with the goal of trying to resolve the conflict. At the fourth minute, the researcher indicated to participant dyads that they had an additional minute to reach a solution.
Mothers’ behaviors were coded with the following coding schemes: (a) codes for supportive parenting included offering thoughtful comments, asking questions, providing explanations, and considering the child’s opinions and suggestions (Recchia, Ross, & Vickar, 2010), and giving positive reinforcement and emotional support (Lunkenheimer, 2014); (b) codes for parenting negativity included criticism, dismissive behavior, rejection of the child (Dishion, Rivera, Verberkmoes, Jones, & Patras, 2002; Neitzel & Stright, 2003), and the expression of negative affect (Lunkenheimer,2014). Behavioral observations were recorded using the INTERACT 16.05 observational software (Mangold, 2017) by two trained research assistants, who were not aware of study hypotheses or given details about child ADHD diagnosis status. Coding was done on a second-by-second basis, and codes were mutually exclusive. The total number of supportive and negative behaviors was calculated. To overcome time differences in conflict discussions, we computed proportion scores. The total number of negative behaviors and the total number of supportive behaviors was each divided by the total number of behaviors observed during the conflict and then multiplied by 100. Interobserver agreement kappa coefficients based on 15% of the observations were .76 and .72 for negative and supportive behaviors, respectively.
Children’s conduct problems
To control for children’s disruptive behaviors, mothers completed the five-item Conduct Problem subscale of the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997). The psychometric properties of the SDQ are well established (Goodman, 2001; Palmieri & Smith, 2007). The Conduct Problem subscale assesses delinquent behaviors (e.g., “Often fights with other children or bullies them”) on a 3-point Likert-type scale, ranging from 1 (not true) to 2 (somewhat true), and 3 (certainly true). The internal consistency was α = .64, much like previous studies (Cummings, Koss, & Davies, 2015).
Maternal parenting distress
To control for mothers’ parenting distress, mothers completed the Parenting Distress subscale from the Parenting Stress Index–Short Form (PSI-SF; Abidin, 1995). The Parenting Distress subscale consists of 12 items (e.g., “I feel that I cannot handle things”) measuring the distress a parent feels in her or his parental role. Mothers rated each item rated on a 5-point Likert-type scale, ranging from 0 (strongly disagree) to 5 (strongly agree). The PSI-SF has been shown to be a reliable and valid measure (McKelvey et al., 2009; Wietecha et al., 2012); in this study, the internal consistency of the mothers’ Parenting Distress scale was .81.
Data Analysis
Before conducting our analysis, we examined the demographic characteristics of the sample and ran descriptive statistics for all variables. Next, we performed bivariate analysis to examine the relations between demographics and study variables, and among study variables themselves. To address the primary research questions, that is, the independent and interactive effects of maternal ADHD symptoms, maternal inhibitory control, and child ADHD diagnostic status (0 = control, 1 = ADHD) on observed and self-reported maternal parenting, we conducted a series of hierarchical linear regressions. We conducted regression analyses for each parenting measure (i.e., parenting negativity, supportive parenting, laxness, overreactivity). Mothers’ inattention symptoms and hyperactivity–impulsivity symptoms were included in the same regression models. As mothers’ inattention symptoms and hyperactivity–impulsivity symptoms were moderately correlated (r = .52, p < .01), the level of multicollinearity was appropriate in terms of the risk for Type II error (Grewal, Cote, & Baumgartner, 2004). Accordingly, both symptoms could be included in the same regression model.
In the first step of the regression models, we entered child sex, child conduct problems, and parenting distress as control variables. In the second step, we entered child ADHD diagnosis status, maternal ADHD symptoms (both mothers’ inattention symptoms and hyperactivity–impulsivity symptoms), and maternal inhibitory control, so that we could examine their independent main effects. In the third step, to determine the moderating role of child ADHD and maternal inhibitory control on the association between maternal ADHD symptoms and self-reported and observed parenting, we entered the interaction of maternal ADHD symptoms and child ADHD and the interaction of maternal ADHD symptoms and maternal inhibition. Following model-testing recommendations, we mean-centered the continuous variables before creating interaction terms (Cohen, Cohen, West, & Aiken, 2003). Given the presence of a significant interaction, we used the PROCESS macro for SPSS (Hayes, 2013). The PROCESS macro calculates conditional effects (i.e., simple slopes) for the effects of a predictor on the outcome at specified values of the moderator (one SD above and below the M). Simple slopes and the Johnson–Neyman technique were used to identify the regions of significant tests. Last, visual plots of the significant interactions were generated.
Results
Descriptive Statistics and Bivariate Correlations
Descriptive statistics and bivariate correlations for all variables are presented in Table 2. As expected, we found several significant correlations between the covariates (i.e., parenting distress, children’s conduct problems, children’s sex) and variables of interest. Mothers of children with ADHD tended to demonstrate more negative and fewer supportive behaviors during conflict discussions than mothers of children without ADHD. However, no associations were found between child ADHD and mothers’ self-reported parenting. Maternal inattention and hyperactive–impulsive symptoms were positively associated with self-reported overreactive parenting, but only maternal inattention symptoms were associated with self-reported lax parenting. Maternal inhibitory control was significantly associated with reduced parenting negativity and overreactivity and with more supportive parenting behaviors. Overall, the effect size of the significant bivariate correlations between mothers’ ADHD symptoms and parenting and maternal inhibition control was low (Cohen, 1992).
Means, Standard Deviations, and Correlations Among the Study Variables (N = 141).
Note. Ch. = child; M = mother; hyper–impul = hyperactivity–impulsivity.
p < .05. **p < .01. ***p < .001.
Regression Analysis
Results of multiple regression models predicting observed parenting and self-reported parenting are presented in Table 3 and Table 4 respectively.
Hierarchical Multiple Regression Results Predicting Observed Parenting Behaviors.
Note. Ch. = child; M = mother; hyper–impul = hyperactivity–impulsivity.
p < .05. **p < .01.
Hierarchical Multiple Regression Results Predicting Self-Reported Parenting Disciplinary Behaviors.
Note. Ch. = child; M = mother; hyper–impul = hyperactivity–impulsivity.
p ⩽ .05. **p < .01.
Observed parenting negativity
At Step 1, F(3, 137) = 2.57, with all covariates included in the model, only child conduct problems had a significant effect on parenting negativity. Simply stated, mothers of children with more conduct problems demonstrated more parenting negativity. At Step 2, F(7, 133) = 3.01, the addition of child ADHD diagnosis, mothers’ hyperactive–impulsive symptoms, inattention symptoms, and inhibitory control accounted for an additional 8% variance in parenting negativity. Inspection of the predictors revealed that child ADHD diagnosis and maternal inhibitory control had a significant main effect on parenting negativity, such that mothers of children without ADHD and with higher inhibitory control demonstrated lower parenting negativity. At Step 3, F(11, 129) = 3.04, the four interaction terms, that is, maternal ADHD symptoms (inattention or hyperactive–impulsive) with child ADHD and maternal ADHD symptoms (inattention or hyperactive–impulsive) with maternal inhibitory control, resulted in a significant 6% increase in explained variance. Inspection of the specific predictors revealed a significant interaction between mothers’ hyperactive–impulsive symptoms and inhibitory control. Analyses of simple slopes revealed that maternal hyperactive–impulsive symptoms were positively related to parenting negativity in mothers with low inhibitory control (b = 19.38, SE = 7.56, p = .011) but not significantly related to parenting negativity in mothers with high inhibitory control (b = −8.17, SE = 6.69, p = .22), indicating that inhibitory control serves as a protective factor in the context of high hyperactive–impulsive symptoms (see Figure 1).

The interaction between maternal hyperactive–impulsive symptoms and inhibitory control on parenting negativity.
Observed supportive parenting
At Step 1, F(3, 137) = 3.13, the model’s covariates accounted for 6% of the variance in supportive parenting. Among the covariates, only child conduct problems were a significant predictor, with higher conduct problems predicting less supportive parenting behaviors. At Step 2, F(7, 133) = 2.85, adding child ADHD diagnosis, maternal hyperactive–impulsive symptoms, inattention symptoms, and inhibitory control accounted for a significant 7% increment in the explained variance. Among these predictors, maternal hyperactive–impulsive symptoms and inhibitory control were both significant predictors of supportive parenting. Specifically, higher levels of maternal inhibitory control and maternal hyperactive–impulsive symptoms predicted increased supportive parenting. There were no significant interactions, Step 3, F(11, 129) = 2.830.
Self-reported overreactive parenting
At Step 1, F(3, 137) = 7.63, all covariates included in the model (i.e., child’s sex, child’s conduct problems, parenting distress) were significant predictors, accounting for 14% of the variance in overreactive parenting. Specifically, mothers of girls tended to be more overreactive than mothers of boys; mothers whose children had more conduct problems and mothers with higher levels of parenting distress also self-reported higher levels of overreactive parenting. At Step 2, F(7, 133) = 5.75, child ADHD status, maternal ADHD symptoms (both inattention and hyperactivity–impulsivity), and maternal inhibitory control explained an additional 9% of the variance in overreactive parenting. Inspection of the main effect revealed that a higher level of maternal hyperactive–impulsive symptoms predicted a higher level of overreactive parenting. In addition, mothers of children with ADHD reported lower overreactive parenting behavior than mothers of children without ADHD. Adding the four interaction terms to the model at Step 3 did not reveal an additional significant increment in explained variance, Step 3, F(11, 129) = 3.98.
Self-reported lax parenting
At Step 1, F(3, 137) = 8.76, child sex, conduct problems, and parenting distress accounted for 16% of the explained variance in lax parenting. Among the covariates, child conduct problems and parenting distress were both significantly and positively associated with lax parenting. At Step 2, F(7, 133) = 4.03, the addition of child ADHD, maternal ADHD symptoms, and maternal inhibitory control did not significantly account for additional variance in lax parenting. At Step 3, F(11, 129) = 4.00, the addition of the four interaction terms accounted for additional 8% of the variance in maternal lax parenting. We found a significant interaction between maternal inattention symptoms and child ADHD diagnosis. Analyses of simple slopes revealed that maternal inattention symptoms were related to lax parenting only among mothers of children without ADHD (b = 0.45, SE = 0.19, p = .018; see Figure 2). In addition, the interaction between maternal inattention symptoms and inhibitory control was significant. Analyses of simple slopes revealed that maternal inattention symptoms were related to lax parenting among mothers with high inhibitory control (b = 0.469, SE = 0.23, p = .003) but not among mothers with low inhibitory control (b = 0.22, SE = 0.22, p > .05; see Figure 3).

The interaction between maternal hyperactive–impulsive symptoms and inhibitory control on lax parenting.

The interaction between maternal inattention symptoms and inhibitory control on lax parenting.
Discussion
A growing body of research has found associations between elevated levels of parental ADHD symptoms and parenting impairments (Park et al., 2017). Our study extends this line of research by examining whether maternal inhibitory control and child ADHD moderate the relations between maternal ADHD symptoms and self-reported or observed parenting behaviors among mothers of children with and without ADHD. Although inhibitory control is known to be related to difficulties among individuals with ADHD, until now, it has not specifically been explored in the context of ADHD symptoms and parenting.
A novel contribution of the study is our finding that maternal inhibitory control moderated the associations between maternal hyperactive–impulsive symptoms and parenting negativity, such that high inhibitory control reduced the risk of increased parenting negativity. Specifically, mothers with a higher level of hyperactive–impulsive symptoms displayed more parenting negativity if they had a lower level of inhibitory control. These findings suggest that when mothers have elevated hyperactive–impulsive symptoms but, at the same time, can withhold responses and regulate their behavior in challenging situations, such as mother–child conflicts, they are better able to regulate their behaviors toward their children.
We also found that higher inhibitory control was associated with increased supportive parenting behavior during mother–child conflicts. The finding is consistent with and extents previous studies demonstrating the contribution of inhibitory control to positive parenting behaviors when parents are teaching and/or playing with young children (Shaffer & Obradović, 2017; Sturge-Apple et al., 2017). Maternal inhibitory control seems to be an important resource for parenting, particularly in affectively laden mother–child interactions, as it enables mothers to be attentive to their children while regulating their own reactions. This finding is in line with emerging studies on the contribution of specific executive function skills to parenting (Deater-Deckard, Sewell, Petrill, & Thompson, 2010).
Mothers’ ability to control their reactions to their children’s misbehavior was reflected in the interaction between maternal inattention symptoms, and maternal inhibitory control on lax parenting and the interaction between maternal inattention symptoms and child ADHD on lax parenting. Specifically, maternal inattention symptoms were positively related to lax parenting but only in the context of high maternal inhibitory control and when their children did not have ADHD. As previously reported, mothers with ADHD symptoms have difficulty organizing and consistently monitoring their children’s actions and their parenting reactions (Chen & Johnston, 2007; Harvey et al., 2003; Murray & Johnston, 2006; Park et al., 2017). Yet, if they have high inhibitory control, they can regulate their reactions and overlook their children’s misbehavior. It seems that child ADHD plays a similar role in maternal reactions; when the child does not have ADHD, mothers with inattention symptoms will not be challenged by the difficult behaviors of children with ADHD. As a result, they can allow themselves to exhibit lax parenting and inconsistency. Overall, although further research is needed to replicate these findings, our study points to the importance of considering both maternal regulatory abilities, such as inhibitory control, and child ADHD status when probing the contribution of maternal ADHD symptoms to parenting.
Consistent with previous findings, maternal hyperactive–impulsive symptoms were positively related to supportive parenting behaviors, independent of child ADHD status, suggesting these mothers are more positive toward their children (Johnston et al., 2016; Lui et al., 2013). The benefit of hyperactive–impulsive symptoms in mothers has been attributed to related personality characteristics (Johnston et al., 2016). Findings from recent studies indicate that ADHD symptoms have positive associations with certain personality traits, such as greater interpersonal openness (i.e., seen in esthetic sensitivity, intellectual curiosity, need for variety, and nondogmatic attitudes; Gomez & Corr, 2014; Van Dijk et al., 2017). These characteristics, in turn, might be reflected in the supportive parenting behaviors we observed in the mother–child conflict discussions, such as offering thoughtful comments, asking questions, providing explanations, and considering the child’s opinions and suggestions.
At the same time, maternal hyperactive–impulsive symptoms and inattention symptoms had a negative effect on parenting. Specifically, mothers with a higher level of ADHD symptoms reported more overreactive parenting, even when child conduct problems and parenting distress were controlled, again echoing previous findings (for a review, see Johnston et al., 2012). Yet, the lack of associations between maternal ADHD symptoms and observed parenting negativity during mother–child conflict discussions suggests the possibility that mothers with elevated ADHD symptoms have a tendency to overestimate their ADHD symptoms and difficulties with parenting in self-report measures, inflating the correlation between maternal ADHD symptoms and self-reported harsh parenting (Lui et al., 2013; Molina, Donovan, & Belendiuk, 2010). In general, the correlations between mothers’ self-reported overreactive parenting and observed parenting negativity are found to be relatively low (Hendriks, Van der Giessen, Stams, & Overbeek, 2018). The type of measure (self-reports vs. observation) is arguably one reason for inconsistency across studies (Park et al., 2017), highlighting the importance of including both observations and self-reports to measure parenting and pointing to one of the strengths of our study.
In terms of children’s effect on parenting, we found no support for either the “similarity–fit” hypothesis or for the “similarity–misfit” hypothesis. In previous studies finding support for either hypothesis, the interaction effect was small (approximately 2%-5% of the explained variance in parenting) or marginally significant. In addition, the moderation effect of child ADHD was apparent in a limited number of parenting dimensions, including positive behavior, maternal empathy, and conflict quality, and all findings were based on parental self-reports (Grimbos & Wiener, 2016; Johnston et al., 2016).
Consistent with previous studies, we found significant positive correlations between children’s conduct and both observed and self-reported parenting, including higher parenting negativity, overreactivity, laxness, and lower levels of supportive parenting (Arney, Rogers, Baghurst, Sawyer, & Prior, 2008; Karazsia, van Dulmen, & Wildman, 2008; Rhoades & O’Leary, 2007). Clearly, children play a significant role in shaping parenting; a child who is angry and oppositional can be especially challenging to manage (Scaramella & Leve, 2004).
Limitations, Implications, and Future Study
The findings of this study must be considered in the context of some limitations. First, we used a dimensional approach to examine maternal ADHD. Our study did not include clinical assessment of maternal ADHD diagnosis and comorbid disorders. Thus, we cannot rule out the presence of potential comorbid disorders with symptoms similar to those of ADHD (e.g., anxiety, mood disorders). Using mothers’ self-reported symptoms is an additional limitation; future studies of maternal ADHD may benefit from multi-informant assessment procedures, such as self and collateral reports (Belendiuk, Clarke, Chronis, & Raggi, 2007). Future studies should also use diverse methodologies, including observations or other ratings of parenting behaviors (e.g., child reports, partner reports), to clarify the apparent differences across methods and parenting behaviors. Finally, given the number of observed p values, there is a risk of Type I error. Hence, further studies are needed to replicate these findings.
Despite these limitations, the study sheds light on the link between maternal ADHD symptoms and maladaptive maternal behaviors (Johnston et al., 2012; Park et al., 2017). Even so, we stress the need to continue to explore other cognitive and effective control capacities that might be involved in the complex associations between parental ADHD (e.g., working memory, cognitive flexibly) and parenting.
The study also has clinical implications; our findings suggest that maternal ADHD symptoms combined with poor maternal inhibitory control may be an important factor to address during parental interventions. According to our findings, it may be useful to include components that address parental ADHD-related deficits such as inhibitory control. Poor inhibitory control could interfere with the implementation of effective responses to child misbehavior. Effective parental interventions should include practicing inhibitory control, so that parents learn how to inhibit their tendency to respond to children’s misbehavior impulsively and in a noneffective manner. Mindful parenting might be a useful intervention for this purpose; parents would learn to pay attention to their children nonjudgmentally, increase their awareness of the present moment with their child, and reduce their automatic (negative) reactions to the child (Bögels, Hellemans, van Deursen, Römer, & van der Meulen, 2014).
Footnotes
Acknowledgements
We are grateful to the parents and children who participate in the study.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
