Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Res Child Adolesc Psychopathol. 2021 Apr 15;50(2):117–131. doi: 10.1007/s10802-020-00754-0

A Multi-Method Investigation of Parental Responses to Youth Emotion: Prospective Effects on Emotion Dysregulation and Reactive Aggression in Daily Life

A L Byrd 1, V Vine 1, O A Frigoletto 1, S Vanwoerden 1, S D Stepp 1
PMCID: PMC8859862  NIHMSID: NIHMS1754486  PMID: 33856610

Abstract

Parental responses to negative emotion, one key component of emotion socialization, may function to increase (or decrease) reactive aggression over time via indirect effects on emotion dysregulation. However, despite its transdiagnostic relevance, very little research has examined this developmental risk pathway, and no studies have done so during the volatile and vulnerable transition to adolescence. The current study uses a sample of clinically referred youth (N = 162; mean age = 12.03 years; 47% female) and their parents to examine supportive and non-supportive parental responses to negative emotion using a multi-method (questionnaire, ecological momentary assessment [EMA], observation), multi-informant approach (child-, parent-, clinician-rated). Emotion dysregulation and reactive aggression were assessed via child report during a 4-day EMA protocol completed concurrently and 9 months later. Multivariate structural equation modeling was used to examine direct and indirect paths from parental responses to emotion to daily reports of emotion dysregulation and reactive aggression. Consistent with hypotheses, parental responses to emotion predicted reactive aggression via effects on emotion dysregulation. This indirect effect was present for supportive and non-supportive parental responses to emotion, such that supportive parental responses decreased risk, and non-supportive responses increased risk. Moreover, findings indicated differential prediction by informant, and this was specific to supportive parental responses to emotion, whereby child-reported support was protective, while parent-reported support, unexpectedly, had the opposite effect. The clinical significance of integrating supportive and non-supportive parental responses to negative emotion into etiological and intervention models of reactive aggression is discussed.

Keywords: Emotion socialization, Multi-informant, Adolescence, Reactive aggression, Emotion dysregulation, Ecological momentary assessment, Longitudinal

Reactive aggression is an impulsive, emotionally-driven form of aggression, and its persistence across adolescence is linked to severe and intractable trajectories of psychopathology (Loeber & Hay, 1997; Ostrov & Houston, 2008; Scott, Stepp, & Pilkonis, 2014). The high prevalence of reactive aggression in clinical settings underscores its transdiagnostic relevance and substantial public health burden, accentuating the importance of identifying developmental risk pathways and targets for intervention (Kazdin, 2003). Research suggests that parental responses to negative emotion, one key component of emotion socialization, may function to increase (or decrease) reactive aggression over time, via indirect effects on emotion dysregulation (Chang, Schwartz, Dodge, & McBride-Chang, 2003; Hubbard et al., 2002). The transition to adolescence represents a ripe period for assessing this developmental cascade, given notable increases in dysregulated emotion and behavior that coincide with the emergence of a psychopathology (Guyer, Silk, & Nelson, 2016; Spear, 2009; Steinberg & Morris, 2001). However, there is little research examining the influence of emotion socialization processes, like parental responses to emotion, on reactive aggression during this developmental period.

The current study examined this developmental risk pathway using a sample of clinically referred youth during the transition to adolescence. Supportive and non-supportive parental responses to emotion were assessed using a multi-method (questionnaire, observation, ecological momentary assessment [EMA]), multi-informant (child-, parent-, clinician-rated) approach. Direct and indirect associations from parental responses to emotion to daily reports of reactive aggression via emotion dysregulation were examined concurrently and 9 months later.

Reactive Aggression: The Role of Emotion Dysregulation and the Influence of Parenting

Reactive aggression is most commonly defined in terms of discrete verbal (screaming, yelling) and/or physical (hitting, kicking, punching) acts that occur in response to perceived threat or frustration (Berkowitz, 1989; Polman, Orobio de Castro, Koops, van Boxtel, & Merk, 2007). Notably, these behaviors manifest within the context of intense, dysregulated negative emotion – at the emotional peak of the freeze-flight-fight defense response – and are often characterized as seemingly unprovoked and/or disproportional to the situational context (Berkowitz, 1989; Gray, 1987). Unlike other forms of aggression, reactive aggression is hallmarked by emotion dysregulation and has been conceptualized as an extreme, maladaptive behavioral response to negative emotional states (Scarpa, Haden, & Tanaka, 2010; Vitaro, Brendgen, & Barker, 2006). Not surprisingly, it is evident across the internalizing and externalizing spectrum, highlighting its transdiagnostic relevance (Card & Little, 2006; Harty, Miller, Newcorn, & Halperin, 2009; Lahey, Miller, Gordon, & Riley, 1999; Marsee, Weems, & Taylor, 2008).

Etiological models of reactive aggression highlight the influential role of parenting factors on the development and persistence of reactive aggression (Dodge & Coie, 1987; Patterson, 1982). To date, such work has focused on various aspects of parenting, including global constructs like the family environment and more specific parenting styles or behaviors, like harsh and inconsistent discipline (see Pinquart, 2017 for review). Notably, these parenting factors are conceptualized as contributing to reactive aggression both directly and indirectly via their effects on emotion dysregulation. For example, empirical studies in early and late childhood have demonstrated that the direct effects of family emotional expressiveness and harsh parenting on aggression can be accounted for by indirect effects on emotion dysregulation (e.g., Chang et al., 2003; Ramsden & Hubbard, 2002). Further, a growing body of research suggests that parental responses to emotion, a key component of emotion socialization, may clarify one discrete way in which parents’ behavior functions to increase risk for emotion dysregulation, and in turn, contribute to increased risk for reactive aggression (Eisenberg, Cumberland, & Spinrad, 1998; Morris, Silk, Steinberg, Myers, & Robinson, 2007).

Parental Responses to Emotion as a Risk Factor

A parent’s response to their child’s emotion provides immediate feedback about the acceptability of emotions, shaping the way in which emotions are experienced and expressed in the future (Eisenberg et al., 1998). This transactional process may be particularly consequential when socializing negative emotions (Eisenberg, Fabes, & Murphy, 1996), given the central role that the dysregulation of negative emotion plays in the development and persistence of reactive aggression (Scarpa et al., 2010; Vitaro et al., 2006). Parental responses to negative emotion can be grouped broadly into behaviors that support emotional expression and promote adaptive emotion regulation, and those that are non-supportive of emotional expression and hinder adaptive emotion regulation (Eisenberg et al., 1998; Sanders et al., 2015).

When parents support emotional expression by acknowledging the negative emotional experience (e.g., listening, asking questions), providing comfort (e.g., hugging) or problem-solving ways to cope with negative emotions (e.g., giving advice, helping to fix the problem), they signal to their child that emotions are acceptable, and they give youth the opportunity to practice regulating their emotions (Eisenberg et al., 1998; Gottman et al., 1997). Such supportive responses have been linked to reduced risk for emotional and behavioral problems (Eisenberg, 2020). By contrast, when parents are not supportive of their child’s negative emotions, they may respond by neglecting the expression of emotion (e.g., ignoring, not noticing), by dismissing or minimizing emotional experiences (e.g., telling child to change their emotional experience, “don’t be angry”), by magnifying or intensifying the negative emotion (e.g., yelling back, getting angry), and/or or by punishing the emotion expression (e.g., mocking the emotional experience). To their child, these responses communicate that negative emotions are intolerable and unacceptable (Eisenberg et al., 1998), and they limit opportunities for youth to learn how to regulate their emotions, often leading to heightened distress and emotion dysregulation (Ramakrishnan et al., 2019). These non-supportive parental responses have also been linked to behavioral dysregulation (Buckholdt, Parra, & Jobe-Shields, 2014), indicating that they may be particularly relevant for understanding risk for reactive aggression.

Transition to Adolescence as Sensitive Developmental Window

While the majority of work examining parental responses to emotion has focused on early developmental periods, research suggests that parents continue to exert a critical influence as youth enter adolescence (Klimes-Dougan et al., 2007; Laursen & Collins, 2009; Morris, Criss, Silk, & Houltberg, 2017). For several reasons, the period between childhood and adolescence—the transition to adolescence—represents a sensitive window for assessing the impact of parental responses to emotion on reactive aggression. This developmental period is characterized by substantial hormonal and neurobiological changes that increase the intensity, variability, and frequency of negative emotions (Guyer et al., 2016; Steinberg & Morris, 2001), as well as the onset of various psychiatric disorders (Spear, 2009). Research suggests that the amplification of negative emotion, especially among clinically referred youth, may necessitate prolonged emotional scaffolding (Guyer et al., 2016; Morris et al., 2017; Steinberg & Silk, 2002). However, research has documented increases in non-supportive parental responses to emotion during this developmental window (Eisenberg et al., 1999) that are likely linked to marked structural changes in the parent–child relationship and increases in parent–child conflict (Branje, 2018; Granic, Hollenstein, Dishion, & Patterson, 2003; Steinberg & Silk, 2002). For all these reasons, we expect the transition to adolescence to have ample variability in key constructs, representing a rich period when clinically meaningful associations between parental responses to negative emotion, emotion dysregulation, and reactive aggression may be detected.

Optimizing Assessment of Parental Responses to Emotion

Parental responses to emotion have been assessed using a variety of methods and informants. Questionnaires are most commonly utilized to assess parent and/or youth perspectives on parents’ supportive and non-supportive responses to a range of negative emotions (Klimes-Dougan et al., 2007; Magai, 1996). Other studies prioritized more objective assessment of parental responses to emotion by coding observable behavior during parent–child interactions, often designed to elicit negative emotion (e.g., Thomas et al., 2017). Observational coding systems allow clinicians to identify the presence of supportive and non-supportive parental responses to emotion in-vivo, during dynamic interactions. While most studies utilize a single method or informant of parental responses to emotion, few studies have sought to combine across methods and informants (see Hastings, Klimes-Dougan, Kendziora, Brand, & Zahn-Waxler, 2014 for exceptions). Further, it appears no studies have attempted to assess parental responses to emotion in daily life using ecological momentary assessment (EMA). EMA has the benefit of incorporating repeated measures of parent and youth perspectives on parental responses to emotion as it occurs within the context of daily transactions, increasing ecological validity (Trull & Ebner-Priemer, 2013). Given that parent, child, and observer reports account for shared and unique variance in parenting practices (Gonzales, Cauce, & Mason, 1996), a comprehensive approach that taps into multiple perspectives and capitalizes on the strengths of various methodologies may have the greatest predictive utility.

Current Study

The present study examined the influence of supportive and non-supportive parental responses to emotion on reactive aggression via emotion dysregulation. This developmental risk pathway was examined using a sample of clinically referred youth (N = 162) and their parents during the sensitive transition to adolescence. Supportive and non-supportive parental responses to emotion were assessed comprehensively using a multi-method (questionnaire, observation, ecological momentary assessment [EMA]), multi-informant (child-, parent-, clinician-rated) approach. Emotion dysregulation and reactive aggression were assessed in daily life using EMA concurrently and at 9-month follow-up. We hypothesized that supportive parental responses to emotion would decrease risk for reactive aggression via reduced emotion dysregulation, and that non-supportive parental responses to emotion would increase risk for reactive aggression via increased emotion dysregulation.

Methods

Sample

Participants were 162 youth (Mage = 12.03 years, SD = 0.92 years, 47% female; 60% racial/ethnic minority1) and their caregivers (Mage = 39.84; SD = 7.25; 94% female; 48% racial/ethnic minority2). Youth and their caregivers (88% biological mothers, hereafter referred to as “parent”) were recruited from pediatric primary care clinics within a large, Mid-Atlantic, urban, academic hospital-based setting, and from ambulatory psychiatric treatment clinics in the same geographic region. All youth were receiving psychiatric treatment for any mood or behavior problem and were oversampled for emotional reactivity based on the Affective Instability subscale from the Personality Assessment Inventory-Adolescent version (M = 13.05, SD = 2.90; scores > 11 indicating clinical significance; Morey, 2007). This recruitment strategy was designed to capture a heterogenous sample of clinically referred youth, characterized by a range of psychopathology (Supplemental Table S1). Any youth with an IQ estimate < 70 (n = 5), not currently in treatment for a mood or behavior problem (n = 4), with an organic neurological medical condition (n = 1), diagnosed with an autism spectrum disorder, or in a current manic or psychotic episode were excluded. All parents had legal custody and primary physical custody (> 50% of the time). Parents reported having M = 3.24 children (SD = 1.68) in their home and 49% reported living with their romantic partners. While 64% of households had at least one employed parent, 19% reported an annual household income between $20,000—$39,000 and 31% reported annual income < $20,000. Additional demographic information is available upon request.

Procedure

The present data were drawn from this longitudinal study, where youth and their parent completed three assessments, every 9 months. The current study focuses on the first and second assessments. Of the initial dyads, 146 (90%) dyads participated in the second assessment.3

Laboratory assessments included semi-structured diagnostic interviews, questionnaires, and a parent–child interaction task. Following the first and second laboratory assessments, participants completed a 4-day EMA protocol, on separate lab-provided smartphones. Parents’ supportive and non-supportive responses were assessed during the first assessment using a multi-method approach, combining child report, parent report and observational data across questionnaire, EMA, and a parent–child interaction task. Emotion dysregulation and reactive aggression were assessed in daily life during the first and second assessment using EMA. All study procedures were approved by the Human Research Protection Office (HRPO) and the Clinical and Translational Science Institute (CTSI) pediatric practice-based research network. Youth and their parents provided written informed consent and were compensated for their time.

Parent–child conflict discussion.

Parents and youth identified areas of disagreement using a 25-item questionnaire that included common areas of parent-child conflict (e.g., behavior toward siblings, internet/cell phone usage, behavior in school). If an area of conflict was endorsed, respondents each rated its frequency (1 = once in past month to 6 = more than once per day) and intensity (1 = not at all bad to 5 = extremely bad). Research assistants identified two conflict topics that were rated highly in terms of frequency (M = 5.23; SD = 1.13) and intensity (M = 3.99; SD = 0.92) by both members of the dyad. Dyads were then asked to discuss topics with a goal of resolving disagreements in the future.

EMA.

All youth and their parents completed a 4-day EMA protocol consisting of 10 time-based prompts over the course of a 4-day period, with two of the days always including Saturday and Sunday (e.g., Friday: midday, nighttime; Saturday-Sunday: morning, midday, nighttime; Monday: midday, nighttime). Youth and their parents were prompted to complete assessments via a “beep” on a study-provided phone. Compliance was high, with the average youth completing 91.12% of assessments at time 1 and time 2, and the average parent completing 90.00% of assessments at time 1, and 90.70% at time 2.

Parental Responses to Emotion

Supportive and non-supportive parental responses were assessed at the first assessment (T1) using multiple informants and methods: 1) child report, via questionnaire and EMA; 2) parent report via EMA; and 3) clinician-rated observation during the parent–child conflict discussion. Six separate measurement models were tested using the indicators described below: two latent factors (supportive and non-supportive) for each of the three informants (child report, parent report, clinician rated observation).

Child Report.

Youth completed the Emotion Socialization Measure (ESM; Klimes-Dougan et al., 2007), adapted from the Emotions as a Child (EAC) Scale (Magai, 1996). Using a 5-point Likert scale (1 = not at all to 5 = very much), youth rated how likely their parents were to respond to their sadness, anger, fear, and shame with supportive (i.e., validate/reward) and unsupportive (i.e., neglect, dismiss, magnify, or punish) responses. Youth also reported on their parent’s supportive and non-supportive responses during daily interactions. At each EMA prompt, youth were asked “Have you had an argument or been upset with your parent?” or “Have you talked with your parent about a problem or worry?” since the last prompt. If they responded “yes” to either (n = 161), they were asked to indicate how supported or unsupported they felt using a 4-point Likert scale (0 = not at all to 4 = very).

For the latent factor of child-reported supportive parental responses, three indicators were included: validation/reward subscale of ESM (e.g., “talk it out with you”, “comfort you”), and two EMA items indexing support (“how encouraging or helpful was your parent?”, “how loving was your parent?”). The small number of indicators (n = 3) necessitated a just-identified model, with factor loadings ranging from 0.53–0.92 (validation/reward subscale, “how loving was your parent”, respectively). The latent factor of child-reported non-supportive parental responses included five initial indicators: four subscales from the ESM: neglect (“ignore you”, “not notice”), dismissive/override (“joke with you”, “tell you not to worry”), magnify (“yell back at you”, “get upset themselves”), punish (“say you should be ashamed”, “tell you to grow up”), and one EMA item (“how critical was your parent of you?”). Initial model fit was poor (χ2(5) = 12.93, p < 0.05; CFI = 0.92; RMSEA = 0.10) and evaluation of indicators revealed one non-significant factor loading (p > 0.05; i.e., dismissive/override). After removing this indicator, model fit with the remaining four indicators was good (χ2(2) = 1.62, p > 0.05; CFI = 1.00; RMSEA < 0.001), with factor loadings ranging from 0.34–0.93, (“how critical was your parent”, punishment subscale, respectively).

Parent Report.

Parents reported on supportive and non-supportive responses during conflict. At each EMA prompt, parents were asked “Have you had an argument or been upset with your child?”. If they responded “yes” (n = 102), they were asked to indicate whether or not they responded with any of the supportive or non-supportive responses listed below (multiple responses allowed), derived from the ESM (Klimes-Dougan et al., 2007).

The latent factor of parent-reported supportive responses included four indicators. Counts for each of the 4 responses (“listened”, “asked questions about it”, “provided advice”, and “helped fix the problem”) were used. Model fit was acceptable (χ2(2) = 5.50, p > 0.05; CFI = 0.99; RMSEA = 0.13), with factor loadings ranging from 0.66–0.93. For the parent-reported non-supportive responses, seven initial indicators were included: punishing (“punished child”, “said do not approve of situation or problem”), magnifying (“raised voice or yelled back”), dismissive (“change your attitude”, “told a joke”, “distracted child”), and neglectful (“ignored the problem”) responses. Counts for each of the 7 responses were used and initial model fit was poor (χ2(14) = 80.872, p < 0.05; CFI = 0.67; RMSEA = 0.22). Evaluation of indicators revealed two non-significant factor loadings (p > 0.05; i.e., “told a joke”, “distracted child”). After removing these indicators, final model fit with the remaining 5 indicators was good (χ2(5) = 7.83, p > 0.05; CFI = 0.98; RMSEA = 0.07), with factor loadings ranging from 0.33–0.80 (“punished child”, “raised voice or yelled back”, respectively).

Clinician-Rated Observation.

Supportive and non-supportive parental responses to emotion were behaviorally coded from the recorded parent–child conflict discussion. A trained coding team coded three dimensions of supportive responses (i.e., emotional support, respect for autonomy, positive affect) and three dimensions of non-supportive responses (i.e., rejection, coercion, anger/frustration) using an adapted version of the System for Coding Interactions and Family Functioning (SCIFF) coding system (Lindahl & Malik, 2000). Each dimension was scored using a 5-point Likert scale (1 = very low to 5 = high) and scores on each of these dimensions were used as manifest indicators of the latent supportive and non-supportive construct (described below). To reach acceptable reliability, coders had to rate two consecutive training assignments and achieve agreement with a ‘master coder.’ Fifteen percent of recordings were also coded for reliability, with acceptable agreement for each dimension (average ICCs: emotional support = 0.84; respect for autonomy = 0.81; positive affect = 0.79; rejection = 0.82; coercion = 0.72; anger/frustration = 0.84).

The latent factor of observed supportive parental responses included 3 indicators: 1) emotional support was defined as recognizing emotional needs and providing comfort or reassurance; 2) respect for autonomy was defined as the facilitation of youth opinions and contribution to decision making in the conversation, including actively acknowledging their opinion and validating them; and 3) positive affect was defined as the overall level of genuine affection expressed by the parent toward the child, as evidenced by tone of voice, facial expression and body language (e.g., smiling, laughing). The small number of indicators (n = 3) necessitated a just-identified model, with factor loadings ranging from 0.55–0.92 (positive affect, emotional support, respectively). For the latent factor of observed non-supportive parental responses, 3 indicators were used: 1) rejection/invalidation defined as the frequency and intensity with which a parent made cruel, critical, insulting, blaming, unkind, rude, or insensitive statements to their child; 2) coercion defined as the frequency with which a parent made threatening, bullying, shaming, or embarrassing, statements to their child; and 3) anger/frustration defined as the overall level of anger, frustration, tension, and irritation expressed by the parent toward their child. The small number of indicators (n = 3) necessitated a just-identified model, with factor loadings ranging from 0.63–0.85 (coercion, anger/frustration, respectively).

Supportive and non-supportive latent factors were then examined within a three-factor model and a higher-order factor model. The supportive three-factor model demonstrated good model fit (χ2(32) = 57.55, p < 0.05; CFI = 0.95; RMSEA = 0.07); however, the three supportive factors were not significantly correlated (range = −0.08–0.16, p > 0.05), suggesting there was not sufficient shared variance between latent indicators to justify a higher order latent factor. The supportive higher order factor model failed to demonstrate significant factor loadings (ps > 0.05); thus, the supportive three-factor model was retained. By contrast, all non-supportive factors were positively correlated (range: 0.23–0.53, p < 0.05) within the three-factor model, suggesting significant shared variance between latent indicators. The non-supportive higher-order factor model demonstrated good model fit (χ2(51) = 75.48, p < 0.05; CFI = 0.92; RMSEA = 0.05), with significant factor loadings for all three indicators (child report: β = 0.87, p < 0.001; parent report: β = 0.61, p < 0.001; observation: β = 0.38, p = 0.001); thus, the non-supportive higher order factor model was retained.

Emotion Dysregulation and Reactive Aggression in Daily Life

Emotion Dysregulation.

At each EMA prompt (n = 10), youth rated a series of negative (i.e., angry, sad, stressed, nervous) and positive emotions (i.e., happy, relaxed, excited, energetic, proud) on a 4-point Likert scale (0 = not at all to 3 = a lot). Specifically, youth responded to a series of prompts asking, “In the past 15 min, how much have you felt…”. Responses to these questions were used to create 3 core components of emotion dysregulation. To capture the overall level of negative emotion, the mean for each negative emotion across all EMA assessments was calculated. To capture variability in negative emotion, the mean squared successive difference (MSSD) was calculated for each negative emotion and then averaged. MSSD assesses the difference between consecutive emotion assessments (e.g., anger at T2–anger at T1; anger at T3–anger at T2) to capture the magnitude of changes (i.e., stronger emotional ups-and-downs) while taking temporal dependency into account (Trull et al., 2008). Higher MSSD indicates a more variable or unstable emotional profile, representative of larger moment-to-moment fluctuations and heightened reactivity to internal or external events. Finally, to assess the relative frequency of negative emotion, the ratio of negative: positive emotion was calculated by dividing the overall level of negative emotion by the overall level of positive emotion. These indices were used as manifest indicators of a latent dysregulated emotion construct at each timepoint. Overall model fit was good (χ2(9) = 13.49, p > 0.05; CFI = 0.99; RMSEA = 0.06), and factor loadings ranged from 0.600.80 at the first assessment (variability in negative emotion, frequency of negative emotion, respectively), and 0.70–0.72 at 9-month follow-up (frequency of negative emotion, variability in negative emotion, respectively).

Reactive Aggression.

During the EMA protocol, participants completed an end-of-day assessment each day (n = 4): “When you were feeling your worst, did you let your feelings out by….”: 1) “crying/screaming,” 2) “yelling at someone,” or 3) “hitting, slamming or punching someone or something.” Youth could endorse any of these items, and a count score (range = 0–4) was generated for each behavior reflecting the number of days each behavior was endorsed (i.e., 0 = did not engage in behavior on any day, 4 = engaged in behavior every day). At the first assessment, nearly half of the sample reported verbal aggression at least once (crying/screaming [41.2%], yelling at someone [43.8%]), and 20.3% reported physical aggression. At the 9-month follow-up, over one-third of the sample reported verbal aggression at least once (crying/screaming [33.9%], yelling at someone [37.7%]), and 18.8% reported physical aggression. Counts for each behavior were used as manifest indicators of a latent reactive aggression construct at each timepoint. Overall model fit was good (χ2(8) = 9.54, p > 0.05; CFI = 0.99; RMSEA = 0.04), and factor loadings ranged from 0.44–0.88 at the first assessment (crying/screaming, hitting, slamming, punching someone or something, respectively), and from 0.400.89 at 9-month follow-up (crying/screaming, hitting, slamming, punching someone or something, respectively).

Covariates

Demographic Variables.

Parent and youth age and minority status (0 = Caucasian; 1 = minority status), and youth sex (0 = male; 1 = female) were included as covariates.

Receipt of Public Assistance.

Parents reported on their annual income and whether or not they received any financial assistance (e.g., WIC, food stamps, welfare, etc.). A dichotomous variable (0 = no public assistance; 1 = received public assistance) was included as a covariate.

Data Analytic Plan

Primary study hypotheses were tested using two structural equation models in MPlus version 8 (Muthén & Muthén, 2012). Models were estimated using weighted least squares mean and variance adjusted (WLSMV), a robust estimator provides the best option for modelling ordinal or categorical data (Li, 2016). Overall model fit was evaluated using the chi-square statistic (χ2), comparative fit index (CFI; Bentler, 1990), Tucker-Lewis index (TLI; Tucker & Lewis, 1973), and root mean square error of approximation (RMSEA; Hu & Bentler, 1999). For the CFI and TLI, conventional cutoff values of 0.90 or greater indicate acceptable fit and 0.95 or greater indicate good fit (McDonald & Ho, 2002). RMSEA values between 0.05 and 0.08 represent an acceptable fit, with values below 0.05 indicating good fit (Kline, 2005; McDonald & Ho, 2002). In both models (supportive and non-supportive), all possible direct and indirect paths were estimated while controlling for the effects of demographic covariates (i.e., parent and youth age and minority status, youth sex, and receipt of public assistance).4 Each model estimated direct paths between parental responses to emotion and emotion dysregulation, and parental responses to emotion and reactive aggression, concurrently (T1) and 9 months later (T2). Additionally, the stability of emotion dysregulation and reactive aggression was estimated as well as prospective effects of emotion dysregulation on reactive aggression, and vice versa.

Results

Descriptive statistics for all study variables are shown in Table 1 and bivariate correlations between extracted latent variables are reported in Table 2. Child-reported supportive parental responses were negatively associated with emotion dysregulation assessed concurrently (T1) and 9 months later (T2). Unexpected positive associations were shown between parent-reported supportive parental responses and reactive aggression at T1, as well as between observed supportive parental responses and emotion dysregulation at T1. Child-reported non-supportive parental responses were positively associated with reactive aggression at T1 and T2, and parent-reported non-supportive parental responses were positively associated with emotion dysregulation at T2.

Table 1.

Descriptive statistics for all study variables

Mean/% SD Min Max N
Demographics Parent Age 40.35 7.23 26.73 62.38 162
Parent Sex (Female) 94% --- --- --- 162
Parent Minority Race 48% --- --- --- 162
Youth Age 12.59 0.95 10.60 14.10 162
Youth Sex (Female) 47% --- --- --- 162
Youth Minority Race 60% --- --- --- 162
Receipt of Public Assistance 52% --- --- --- 162
Supportive Responses to Emotion Child Report
 ESM: Validate/Reward 47.70 17.13 16.00 78.00 146
 EMA: “How encouraging or helpful?” 2.11 0.86 0.00 3.00 161
 EMA: “How loving?” 2.58 0.67 0.40 3.00 161
Parent Report
 EMA: “Listened” 1.12 1.28 0.00 7.00 102
 EMA: “Asked questions” 1.10 1.27 0.00 7.00 102
 EMA: “Provided advice” 0.85 1.12 0.00 7.00 102
 EMA: “Helped fix problem” 0.60 0.85 0.00 4.00 102
Observation during Parent-Child Conflict
 Emotional Support 2.94 1.13 1.00 5.00 157
 Respect for Autonomy 3.41 1.06 1.00 5.00 157
 Positive Affect 1.95 1.17 1.00 5.00 156
Non-Supportive Responses to Emotion Child Report
 ESM: Neglect 19.08 7.86 9.00 46.00 146
 ESM: Magnify 13.38 4.22 5.00 25.00 146
 ESM: Punish 20.30 7.05 2.00 46.00 146
 EMA: “How critical?” 0.29 0.48 0.00 2.88 137
Parent Report
 EMA “Ignore” 0.09 0.35 0.00 2.00 102
 EMA “Change attitude” 0.68 1.04 0.00 6.00 102
 EMA: “Said did not approve” 1.20 1.22 0.00 8.00 102
 EMA: “Yelled back” 0.99 1.21 0.00 7.00 102
 EMA: “Punish” 0.36 0.66 0.00 3.00 102
Observation during Parent-Child Conflict
 Rejection/Invalidation 1.34 0.68 1.00 5.00 157
 Coercion 1.81 1.09 1.00 5.00 156
 Anger/Frustration 2.20 1.19 1.00 5.00 155
Emotion Dysregulation Time 1
 Intensity of Negative Emotion 0.32 0.39 0.00 2.14 162
 Variability in Negative Emotion 0.50 0.58 0.00 2.80 162
 Frequency of Negative Emotion 0.36 0.78 0.00 5.36 162
Time 2
 Intensity of Negative Emotion 0.28 0.34 0.00 1.58 142
 Variability in Negative Emotion 0.45 0.54 0.00 3.11 142
 Frequency of Negative Emotion 0.33 0.59 0.00 4.25 142
Reactive Aggression Time 1
 Screaming/ Crying 0.72 1.07 0.00 4.00 153
 Yelling at Someone 0.76 1.06 0.00 4.00 153
 Hitting Someone or Something 0.30 0.72 0.00 4.00 153
Time 2
 Screaming/Crying 0.50 0.78 0.00 3.00 133
 Yelling at Someone 0.59 0.92 0.00 4.00 133
 Hitting Someone or Something 0.25 0.60 0.00 3.00 133
Open in a new tab

EMA = ecological momentary assessment; ESM = Emotion Socialization Measure

Table 2.

Bivariate correlations between primary study variables

1 2 3 4 5 6 7 8 9
Supportive Responses to Emotion
1. Child Report
2. Parent Report 0.04
3. Observation 0.10 −0.06
Non-Supportive Responses to Emotion
4. Child Report 0.01 0.28 ** −0.15
5. Parent Report −0.21 * 0.42 ** −0.22 * 0.34 **
6. Observation −0.14 −0.06 −0.68 ** 0.27 ** 0.16†
Emotion Dysregulation
7. Emotion Dysregulation: T1 −0.26 ** 0.10 0.17 * 0.07 0.17† −0.15
8. Emotion Dysregulation: T2 −0.29 ** 0.07 0.16*** 0.02 0.22 * −0.14 0.61 **
Reactive Aggression
9. Reactive Aggression: T1 0.03 0.32 ** −0.11 0.24 ** 0.19† 0.08 0.15 0.11
10. Reactive Aggression: T2 0.00 0.18 −0.16 0.20 ** 0.13 0.13 0.21 ** 0.28 ** 0.75 **
Open in a new tab

Significant correlations are bolded and represent associations between extracted latent variables. T1 = first assessment; T2 = 9-month follow-up

**

p < 0.01;

*

p < 0.05;

†

p < 0.10

Supportive Parental Responses to Emotion

Figure 1 depicts the model examining direct and indirect pathways from supportive parental responses to emotion dysregulation and reactive aggression, assessed concurrently (T1) and 9 months later (T2), while accounting for the stability of emotion dysregulation and reactive aggression over time, and covariates. The model demonstrated good fit and explained 25% of the variance in emotion dysregulation at T1 and 40% at T2, as well as 34% of the variance in reactive aggression at T1 and 48% at T2.

Fig. 1.

Fig. 1

Open in a new tab

Multivariate structural equation model testing direct and indirect paths from supportive parental responses to emotion dysregulation and reactive aggression

As expected, emotion dysregulation at T1 predicted emotion dysregulation at T2, and reactive aggression at T1 predicted reactive aggression at T2 (Table 3). Emotion dysregulation at T1 predicted reactive aggression at T2, while reactive aggression at T1 was unrelated to emotion dysregulation at T2. Child-reported supportive parental responses were negatively associated with emotion dysregulation at T1 and T2, and observed supportive parental responses negatively predicted reactive aggression at T2. Contrary to hypotheses, parent-reported supportive parental responses were positively associated with reactive aggression at T1. As predicted, there was an indirect effect of supportive parental responses on reactive aggression at T2 via emotion dysregulation at T1, and this was specific to child report (Table 3). Child-reported supportive parental responses were associated with less emotion dysregulation at T1, which predicted less reactive aggression at T2. Additionally, child-reported supportive parental responses were indirectly associated with emotion dysregulation at T2 via emotion dysregulation at T1, such that supportive parental responses were associated with less emotion dysregulation at T1, which predicted less emotion dysregulation at T2. Finally, there was an indirect effect of parent-reported supportive responses on aggression at T2 via aggression at T1, with higher levels of parent-reported supportive responses predicting reactive aggression over time.

Table 3.

Multivariate structural equation model testing direct and indirect paths from supportive parental responses to emotion to emotion dysregulation and reactive aggression

Supportive Responses (T1) Emotion Dysregulation in Daily Life (T1) Reactive Aggression in Daily Life (T2) β SE P
Emotion Dysregulation → Emotion Dysregulation 0.476 0.047 0.000
Reactive Aggression → Emotion Dysregulation 0.105 0.123 0.396
Emotion Dysregulation → Reactive Aggression 0.313 0.091 0.001
Reactive Aggression → Reactive Aggression 0.566 0.125 0.000
Direct Effects Child Report → Emotion Dysregulation −0.153 0.058 0.008
Child Report → Reactive Aggression 0.009 0.097 0.928
Child Report → Emotion Dysregulation −0.227 0.064 0.000
Child Report → Reactive Aggression −0.031 0.106 0.768
Indirect Effects Child Report → Emotion Dysregulation → Emotion Dysregulation −0.073 0.027 0.007
Child Report → Reactive Aggression → Emotion Dysregulation 0.001 0.010 0.929
Child Report → Emotion Dysregulation → Reactive Aggression −0.048 0.022 0.026
Child Report → Reactive Aggression → Reactive Aggression 0.005 0.055 0.928
Direct Effects Parent Reports → Emotion Dysregulation 0.109 0.078 0.160
Parent Reports → Reactive Aggression 0.453 0.081 0.000
Parent Reports → Emotion Dysregulation 0.067 0.091 0.457
Parent Reports → Reactive Aggression −0.149 0.108 0.166
Indirect Effects Parent Reports → Emotion Dysregulation → Emotion Dysregulation 0.052 0.037 0.158
Parent Reports → Reactive Aggression → Emotion Dysregulation 0.047 0.057 0.202
Parent Reports → Emotion Dysregulation → Reactive Aggression 0.034 0.030 0.249
Parent Reports → Reactive Aggression. → Reactive Aggression 0.257 0.080 0.001
Direct Effects Observation → Emotion Dysregulation 0.090 0.060 0.134
Observation → Reactive Aggression −0.060 0.090 0.506
Observation → Emotion Dysregulation 0.115 0.067 0.088
Observation → Reactive Aggression −0.221 0.098 0.024
Indirect Effects Observation → Emotion Dysregulation → Emotion Dysregulation 0.055 0.033 0.095
Observation → Reactive Aggression → Emotion Dysregulation −0.006 0.012 0.599
Observation → Emotion Dysregulation → Reactive Aggression 0.036 0.024 0.139
Observation → Reactive Aggression → Reactive Aggression −0.034 0.051 0.509
Open in a new tab

Overall model fit was good (χ2(294) = 349.29, p < 0.05; CFI = 0.94; TLI = 0.93; RMSEA = 0.03). Significant associations are bolded and represent effects after accounting for parent and youth age and minority status, youth sex, and receipt of public assistance. T1 = first assessment; T2 = 9-month follow-up

Non-Supportive Parental Responses to Emotion

Figure 2 depicts the model examining direct and indirect pathways from non-supportive parental responses to emotion dysregulation and reactive aggression, assessed concurrently (T1) and 9 months later (T2), while accounting for the stability of emotion dysregulation and reactive aggression over time, and controlling for covariates. The model demonstrated good fit to the data and explained 36% of the variance in emotion dysregulation at T1 and 36% at T2, as well as 31% of the variance in aggression at T1 and 40% at T2.

Fig. 2.

Fig. 2

Open in a new tab

Multivariate structural equation model testing direct and indirect paths from non-supportive parental responses to emotion dysregulation and reactive aggression

Associations between emotion dysregulation and aggression at T1 and T2 were similar to the supportive model (Table 4). As expected, non-supportive parental responses were positively associated with emotion dysregulation and reactive aggression at T1. Also consistent with hypotheses, there was a significant indirect effect of non-supportive parental responses on reactive aggression at T2 via emotion dysregulation at T1 (Table 4). Specifically, non-supportive parental responses were associated with greater emotion dysregulation at T1, which predicted reactive aggression at T2. Non-supportive responses were also indirectly associated with emotion dysregulation at T2 via emotion dysregulation at T1, and with reactive aggression at T2 via reactive aggression at T1, demonstrating that non-supportive parental responses predicted emotion dysregulation and reactive aggression over time.5

Table 4.

Multivariate structural equation model testing direct and indirect paths from non-supportive parental responses to emotion to emotion dysregulation and reactive aggression

Non-Supportive Responses (T1) Emotion Dysregulation in Daily Life (T1) Reactive Aggression in Daily Life (T2) β SE P
Emotion Dysregulation → Emotion Dysregulation 0.538 0.048 0.000
Reactive Aggression → Emotion Dysregulation 0.032 0.126 0.770
Emotion Dysregulation → Reactive Aggression 0.267 0.104 0.010
Reactive Aggression → Reactive Aggression 0.534 0.126 0.000
Direct Effects Higher-order Factor → Emotion Dysregulation 0.351 0.071 0.000
Higher-order Factor → Reactive Aggression 0.406 0.086 0.000
Higher-order Factor → Emotion Dysregulation 0.146 0.104 0.161
Higher-order Factor → Reactive Aggression 0.015 0.145 0.916
Indirect Effects Higher-order Factor → Emotion Dysregulation → Emotion Dysregulation 0.188 0.041 0.000
Higher-order Factor → Reactive Aggression → Emotion Dysregulation 0.013 0.044 0.771
Higher-order Factor → Emotion Dysregulation → Reactive Aggression 0.094 0.040 0.019
Higher-order Factor → Reactive Aggression → Reactive Aggression 0.217 0.077 0.005
Open in a new tab

Overall model fit was good (χ2(366) = 456.430, p < 0.05; CFI = 0.90; TLI = 0.88; RMSEA = 0.04). Significant associations are bolded and represent effects after accounting for parent and youth age and minority status, youth sex, and receipt of public assistance. T1 = first assessment; T2 = 9-month follow-up

Discussion

Previous research suggests that parental responses to negative emotion, one key component of emotion socialization, may function to increase (or decrease) reactive aggression over time via indirect effects on emotion dysregulation (Eisenberg et al., 1999; Morris et al., 2007). However, despite its transdiagnostic relevance, very little research has examined this developmental risk pathway, and no studies have done so during the volatile and vulnerable transition to adolescence. The current study examined associations between parental responses to negative emotion, emotion dysregulation, and reactive aggression in a sample of clinically referred youth, both concurrently and prospectively, using a multi-method, multi-informant assessment of parental responses to emotion and youth reports of emotion dysregulation and reactive aggression in daily life. Consistent with hypotheses, parental responses to emotion predicted reactive aggression via emotion dysregulation. This indirect effect was present for supportive and non-supportive parental responses to emotion, such that supportive parental responses decreased risk, and non-supportive responses increased risk. Moreover, findings indicated differential prediction by informant, whereby child report of supportive parental responses to emotion were protective while parent-report, unexpectedly, had the opposite effect.

These findings provide evidence supporting a developmental cascade in which neglectful, minimizing, magnifying and punishing reactions to the expression of negative emotion function to increase risk for reactive aggression over time via emotion dysregulation. Specifically, youth who experienced non-supportive parental responses when expressing negative emotion reported higher levels of emotion dysregulation in daily life, which in turn predicted their daily reports of reactive aggression 9 months later. These results extend etiological models of reactive aggression, which have traditionally focused more broadly on the family environment and parenting styles (Pinquart, 2017), by highlighting the impact of discrete parental responses to negative emotion as a risk factor for reactive aggression via indirect effects on emotion dysregulation. Moreover, findings expand on emotion socialization research that documents the influence of parental responses to emotion on dysregulation into adolescence (Buckholdt et al., 2014; Morris et al., 2017) and demonstrates its impact on reactive aggression specifically.

Results also highlight the protective role that supportive parental responses to negative emotion can have on reactive aggression via indirect effects on emotion dysregulation. Specifically, youth who experienced supportive or validating responses to their expression of negative emotion, reported lower levels of emotion dysregulation, and this predicted less reactive aggression 9 months later. These findings echo research that consistently demonstrates the protective effect of supportive parental responses to emotion with regard to broader social and emotional competence (Eisenberg, 2020). It also extends this work by demonstrating this protective effect within a sample of emotionally dysregulated, clinically referred youth, further underscoring the significance of prolonged emotional scaffolding among vulnerable populations. Taken together, our findings fit into a broader literature and suggest that during this developmentally sensitive window, at-risk youth may be differentially susceptible to the influence of parental responses to emotion (Belsky & Pluess, 2009), which has implications for altering developmental trajectories of reactive aggression.

Consistent with previous work (Maurizi et al., 2012), findings highlight discrepancies among informant reports, and this was specific to supportive parental responses to emotion. The protective effect of supportive parental responses to emotion was circumscribed to youths’ perception of parental support, suggesting that youths’ subjective experience of parental support has the greatest impact on their behavior. At the same time, parent-reported supportive responses to emotion significantly predicted reactive aggression, but in the opposite direction. Specifically, youth with parents who reported higher levels of supportive responses to conflict in daily life (e.g., asking questions, providing advice, helping to solve the problem) were more likely to engage in reactive aggression concurrently, and this indirectly increased risk for reactive aggression 9 months later. While this finding is contrary to our hypotheses and stands in stark contrast to prior work demonstrating the protective effects of parent-reported supportive responses to emotion (Eisenberg, 2020), we note several potential explanations here. First, it is possible that parents’ report more accurately reflects their intent to support, rather than the function of their behavior, consistent with our finding that youths’ subjective experience of support is most protective. Second, this could reflect parents’ over-reliance on problem-solving in the absence of validation, which has been shown to predict emotion and behavioral dysregulation over time (Dixon-Gordon, Whalen, Scott, Cummins, & Stepp, 2016). While parents may perceive their attempts to assist with problem-solving as supportive, youth may experience this as unsupportive or invalidating, especially in the midst of conflict, a speculation that would be interesting for future studies to verify. Lastly, this counterintuitive effect is consistent with the notion that at-risk youth have difficulty benefiting from the positive effects of their social environment, which may reflect selective attention to non-supportive parental responses to emotion or the dismissal of parental responses that contradict overall representations of the parent–child relationship (Fonagy, Luyten, Allison, 2015).

Notably, observed supportive responses during conflict were directly related to reduced reactive aggression at 9-month follow-up. This is in line with research suggesting that supportive parenting behaviors reduce risk for aggression (Houltberg et al., 2016), and suggests that supportive parental responses ‘in the heat of the moment’ may divert youth away from maladaptive escalations to aggression. While we failed to demonstrate an indirect effect on reactive aggression via emotion dysregulation, it is possible that our global observational coding system was unable to capture the nuanced nature of parental responses to emotion within the context of specific dyadic transactions. Along these lines, coding systems that account for more dynamic patterns of behavior suggest that predictive value lies in moment-to-moment exchanges (Lougheed et al., 2019; Moed et al., 2015) and future work may incorporate momentary coding to more accurately capture specific transactions between parents and children. Ultimately, better characterizing these dyadic interpersonal processes could point to malleable treatment targets.

Limitations

Findings from the current study should be considered within the context of several limitations. First, while results from our clinically referred sample mirror those demonstrated in community samples (Eisenberg, 2020), continued replication is needed to determine whether the developmental risk pathway identified generalizes to larger samples of youth. Replication in larger, community samples would enhance reproducibility and afford the inclusion of resampling methods when testing indirect effects (Koopman, Howe, Hollenbeck, & Sin, 2015). Second, though our analyses included minority status as a covariate, this indicator was dichotomized and utilized in statistical models as a person-centered variable. We acknowl- edge that race is a non-discrete, socially created construct and reflects the effects of numerous risk factors for which minority status is a proxy (Kaufman & Cooper, 2001; Richeson & Sommers, 2016). Future work is needed to examine the direct impact of these risk factors on emotion socialization and dysregulated emotion and behavior (Jones & Neblett, 2017). Third, though we controlled for the effects of youth sex, our sample size did not allow us to examine the moderating effects of sex. Given that sex may influence the emotion socialization process and the expression of emotion and behavior dysregulation (Chaplin, Cole, & Zahn-Waxler, 2005; Klimes-Dougan et al., 2007; Vanwoerden, Hofmans, & De Clercq, 2020), future work should aim to explore this potential moderator. Additionally, we were unable to assess differential effects of sex within the dyad (i.e., mother versus father to son versus daughter), and future studies could investigate these relationships with regard to reactive aggression (Katz et al., 2014; Walton & Flouri, 2010). Fourth, the current study focused on one key aspect of emotion socialization, parental responses to emotion. Future work may examine other relevant characteristics, such as parents’ own emotion regulation, which could influence youth both by modeling means of coping with negative emotion, and/or by impacting parents’ ability to respond effectively (Chang et al., 2003). Finally, while our study demonstrated prospective associations between parents’ response to emotion and emotion dysregulation and reactive aggression, it focused on only two time points. Continued longitudinal work with additional follow-ups would allow for the examination of reciprocal associations between parental emotion socialization and youth outcomes in line with transactional models (Patterson, 1982; Sameroff, 1975) and previous empirical work in this area (Eisenberg et al., 1999).

Clinical Implications

The current study illustrates the importance of supportive and non-supportive parental responses to emotion during the developmentally sensitive transition to adolescence. These findings also highlight the significance of integrating parental responses to negative emotion into etiological models of reactive aggression and underscore potential treatment implications. While traditional parenting interventions often focus on implementing consistent behavioral consequences following aggression (Kazdin, 2003), our findings suggest that targeting how parents respond to displays of negative emotion could function to decrease both emotion dysregulation and reactive aggression over time. Moreover, given the extent to which emotion dysregulation and reactive aggression are intricately intertwined, strategies focused on helping parents to separate the valid emotional experience from the maladaptive aggressive behavior may be useful. Finally, results suggest that utilizing youth’s subjective experience of parental support is critical when attempting to effectively target emotion dysregulation and reactive aggression.

Supplementary Material

Supplemental Materials

Acknowledgements

We are grateful to all the families who took part in this study, and to the MoodY study team, which includes interviewers and their supervisors, data managers, student workers, and volunteers.

Funding This study was supported by grants awarded to Dr. Stephanie Stepp and Dr. Amy Byrd from the National Institute on Mental Health (R01 MH101088, F32 MH110077). Additional funding from the National Institute of Health also supported this work (T32 MH018269, K01 MH119216).

Footnotes

1

Ethnoracial composition for youth was 42% White, 40.7% Black, 3.7% Hispanic/Latino, 0.6% American Indian/Alaskan Native, and 16.7% multiracial.

2

Ethnoracial composition for parents was 53.7% White, 39.5% Black, 1.9% Hispanic/Latino, 0.6% Asian, and 6.2% multiracial.

3

Participants completing both assessments were compared to those with missing data (i.e., those who only completed the first assessment) on all demographic covariates (age, sex, minority status, receipt of public assistance). Participants with complete data did not differ from those who only completed the first assessment.

4

To ensure findings were not biased by reading ability, we also reran our models including verbal IQ (PPVT-4; Dunn & Dunn, 2007) as a covariate. All direct and indirect effects remained unchanged.

5

We also examined direct and indirect paths within a three-factor model to assess potential informant specific associations between non-supportive factors and emotion dysregulation and reactive aggression. However, model fit was poor (χ2(349) = 554.456, p < .05; CFI = .76; TLI = .71; RMSEA = .06) and modification indices suggested this was due to high correlations between the non-supportive latent indicators, hence our retention of the higher order model.

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10802-020-00754-0.

Compliance with Ethical Standards

The authors declare that they have no conflict of interest.

References

  1. Belsky J, & Pluess M. (2009). Beyond diathesis stress: differential susceptibility to environmental influences. Psychological bulletin, 135(6), 885. [DOI] [PubMed] [Google Scholar]
  2. Bentler PM (1990). Comparative fit indexes in structural models. Psychological bulletin, 107, 238–246. 10.1037/0033-2909.107.2.238. [DOI] [PubMed] [Google Scholar]
  3. Berkowitz L. (1989). Frustration-aggression hypothesis: examination and reformulation. Psychological bulletin, 106(1), 59. [DOI] [PubMed] [Google Scholar]
  4. Branje S. (2018). Development of parent–adolescent relationships: Conflict interactions as a mechanism of change. Child Development Perspectives, 12(3), 171–176. [Google Scholar]
  5. Buckholdt KE, Parra GR, & Jobe-Shields L. (2014). Intergenerational transmission of emotion dysregulation through parental invalidation of emotions: Implications for adolescent internalizing and externalizing behaviors. Journal of child and family studies, 23(2), 324–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Card N, & Little TD (2006). Proactive and reactive aggression in childhood and adolescence: A meta-analysis of differential relations with psychosocial adjustment. International Journal of Behavioral Development, 30, 466–480. 10.1111/j.1467-8624.2008.01184.x. [DOI] [Google Scholar]
  7. Chang L, Schwartz D, Dodge KA, & McBride-Chang C. (2003). Harsh parenting in relation to child emotion regulation and aggression. J Fam Psychol, 17(4), 598–606. Retrieved from 14640808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chaplin TM, Cole PM, & Zahn-Waxler C. (2005). Parental socialization of emotion expression: gender differences and relations to child adjustment. Emotion, 5(1), 80. [DOI] [PubMed] [Google Scholar]
  9. Dixon-Gordon KL, Whalen DJ, Scott LN, Cummins ND, & Stepp SD (2016). The main and interactive effects of maternal interpersonal emotion regulation and negative affect on adolescent girls’ borderline personality disorder symptoms. Cognitive therapy and research, 40(3), 381–393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dodge KA, & Coie JD (1987). Social-information-processing factors in reactive and proactive aggression in children’s peer groups. J Pers Soc Psychol, 53(6), 1146–1158. Retrieved from 3694454. [DOI] [PubMed] [Google Scholar]
  11. Dunn LM, & Dunn DM (2007). PPVT-4: Peabody picture vocabulary test. Pearson Assessments. [Google Scholar]
  12. Eisenberg N. (2020). Findings, issues, and new directions for research on emotion socialization. Developmental psychology, 56(3), 664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Eisenberg N, Cumberland A, & Spinrad TL (1998). Parental socialization of emotion. Psychological inquiry, 9(4), 241–273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Eisenberg N, Fabes RA, & Murphy BC (1996). Parents’ reactions to children’s negative emotions: Relations to children’s social competence and comforting behavior. Child Development, 67(5), 2227–2247. [PubMed] [Google Scholar]
  15. Eisenberg N, Fabes RA, Shepard SA, Guthrie IK, Murphy BC, & Reiser M. (1999). Parental reactions to children’s negative emotions: Longitudinal relations to quality of children’s social functioning. Child Development, 70(2), 513–534. [DOI] [PubMed] [Google Scholar]
  16. Fonagy P, Luyten P, & Allison E. (2015). Epistemic Petrification and the Restoration of Epistemic Trust: A New Conceptualization of Borderline Personality Disorder and Its Psychosocial Treatment. Journal of Personality Disorders, 29(5), 575–609. 10.1521/pedi.2015.29.5.575. [DOI] [PubMed] [Google Scholar]
  17. Gonzales NA, Cauce AM, & Mason CA (1996). Interobserver agreement in the assessment of parental behavior and parent-adolescent conflict: African American mothers, daughters, and independent observers. Child Development, 67(4), 1483–1498. [PubMed] [Google Scholar]
  18. Gottman JM (1997). Katz, & Hooven. How families communicate emotionally. Lawrence Erlbaum Associates: Meta-emotion. [Google Scholar]
  19. Granic I, Hollenstein T, Dishion TJ, & Patterson GR (2003). Longitudinal analysis of flexibility and reorganization in early adolescence: a dynamic systems study of family interactions. Developmental psychology, 39(3), 606–617. Retrieved from . [DOI] [PubMed] [Google Scholar]
  20. Gray JA (1987). The psychology of fear and stress. Cambridge, England: Cambridge University Press. [Google Scholar]
  21. Guyer AE, Silk JS, & Nelson EE (2016). The neurobiology of the emotional adolescent: From the inside out. Neuroscience & Biobehavioral Reviews, 70, 74–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Harty SC, Miller CJ, Newcorn JH, & Halperin JM (2009). Adolescents with childhood ADHD and comorbid disruptive behavior disorders: aggression, anger, and hostility. Child psychiatry and human development, 40(1), 85–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hastings PD, Klimes-Dougan B, Kendziora KT, Brand A, & Zahn-Waxler C. (2014). Regulating sadness and fear from outside and within: Mothers’ emotion socialization and adolescents’ parasympathetic regulation predict the development of internalizing difficulties. Development and psychopathology, 26(4pt2), 1369–1384. [DOI] [PubMed] [Google Scholar]
  24. Houltberg BJ, Sheffield Morris A, Cui L, Henry CS, & Criss MM (2016). The Role of Youth Anger in Explaining Links Between Parenting and Early Adolescent Prosocial and Antisocial Behavior. The Journal of Early Adolescence, 36(3), 297–318. 10.1177/0272431614562834. [DOI] [Google Scholar]
  25. Hu L-T, & Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 424–453. 10.1080/10705519909540118. [DOI] [Google Scholar]
  26. Hubbard JA, Smithmyer CM, Ramsden SR, Parker EH, Flanagan KD, Dearing KF, Simons RF (2002). Observational, physiological, and self-report measures of children’s anger: Relations to reactive versus proactive aggression. Child Development, 73(4), 1101–1118. Retrieved from . [DOI] [PubMed] [Google Scholar]
  27. Jones SC, & Neblett EW (2017). Future directions in research on racism-related stress and racial-ethnic protective factors for Black youth. Journal of Clinical Child & Adolescent Psychology, 46(5), 754–766. [DOI] [PubMed] [Google Scholar]
  28. Katz LF, Shortt JW, Allen NB, Davis B, Hunter E, Leve C, & Sheeber L. (2014). Parental Emotion Socialization in Clinically Depressed Adolescents: Enhancing and Dampening Positive Affect. Journal of Abnormal Child Psychology, 42(2), 205–215. 10.1007/s10802-013-9784-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kaufman JS, & Cooper RS (2001). Commentary: considerations for use of racial/ethnic classification in etiologic research. American journal of epidemiology, 154(4), 291–298. [DOI] [PubMed] [Google Scholar]
  30. Kazdin AE (2003). Psychotherapy for children and adolescents. Annual Review of Psychology, 54, 253–276. Retrieved from 12185210. [DOI] [PubMed] [Google Scholar]
  31. Klimes-Dougan B, Brand AE, Zahn-Waxler C, Usher B, Hastings PD, Kendziora K, & Garside RB (2007). Parental emotion socialization in adolescence: Differences in sex, age and problem status. Social Development, 16(2), 326–342. 10.1111/j.1467-9507.2007.00387.x. [DOI] [Google Scholar]
  32. Kline RB (2005). Principles and Practice of Structural Equation Modeling (Guilford, New York: ). In. [Google Scholar]
  33. Koopman J, Howe M, Hollenbeck JR, & Sin H-P (2015). Small sample mediation testing: Misplaced confidence in bootstrapped confidence intervals. Journal of Applied Psychology, 100(1), 194. [DOI] [PubMed] [Google Scholar]
  34. Koopman J, Howe M, Hollenbeck JR, & Sin H-P (2015). Small sample mediation testing: Misplaced confidence in bootstrapped confidence intervals. Journal of Applied Psychology, 100(1), 194–202. 10.1037/a0036635. [DOI] [PubMed] [Google Scholar]
  35. Lahey BB, Miller TL, Gordon RA, & Riley AW (1999). Developmental epidemiology of the disruptive behavior disorders. In Handbook of disruptive behavior disorders (pp. 23–48): Springer. [Google Scholar]
  36. Laursen B, & Collins WA (2009). Parent—child relationships during adolescence. Handbook of adolescent psychology. [Google Scholar]
  37. Li C-H (2016). Confirmatory factor analysis with ordinal data: Comparing robust maximum likelihood and diagonally weighted least squares. Behavior research methods, 48(3), 936–949. [DOI] [PubMed] [Google Scholar]
  38. Lindahl KM, & Malik NM (2000). The system for coding interactions and family functioning. In Family observational coding systems (pp. 91–106): Psychology Press. [Google Scholar]
  39. Loeber R, & Hay D. (1997). Key issues in the development of aggression and violence from childhood to early adulthood. Annual Review of Psychology, 48(1), 371–410. [DOI] [PubMed] [Google Scholar]
  40. Lougheed JP, Benson L, Cole PM, & Ram N. (2019). Multilevel survival analysis: Studying the timing of children’s recurring behaviors. Developmental Psychology, 55(1), 53–65. 10.1037/dev0000619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Magai C. (1996). Emotions as a child. Unpublished manuscript, Long Island University, Brooklyn. [Google Scholar]
  42. Marsee MA, Weems CF, & Taylor LK (2008). Exploring the association between aggression and anxiety in youth: A look at aggressive subtypes, gender, and social cognition. Journal of child and family studies, 17(1), 154–168. [Google Scholar]
  43. Maurizi LK, Gershoff ET, & Aber JL (2012). Item-Level Discordance in Parent and Adolescent Reports of Parenting Behavior and Its Implications for Adolescents’ Mental Health and Relationships with Their Parents. Journal of Youth and Adolescence, 41(8), 1035–1052. 10.1007/s10964-011-9741-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. McDonald RP, & Ho MR (2002). Principles and practice in reporting structural equation analyses. Psychological Methods, 7(1), 64–82. 10.1037/1082-989x.7.1.64. [DOI] [PubMed] [Google Scholar]
  45. Moed A, Gershoff ET, Eisenberg N, Hofer C, Losoya S, Spinrad TL, & Liew J. (2015). Parent-Adolescent Conflict as Sequences of Reciprocal Negative Emotion: Links with Conflict Resolution and Adolescents’ Behavior Problems. Journal of Youth and Adolescence, 44(8), 1607–1622. 10.1007/s10964-014-0209-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Morey LC (2007). Personality Assessment Inventory (PAI): Psychological Assessment Resources. [Google Scholar]
  47. Morris AS, Criss MM, Silk JS, & Houltberg BJ (2017). The impact of parenting on emotion regulation during childhood and adolescence. Child Development Perspectives, 11(4), 233–238. [Google Scholar]
  48. Morris AS, Silk JS, Steinberg L, Myers SS, & Robinson LR (2007). The role of the family context in the development of emotion regulation. Social Development, 16(2), 361–388. 10.1111/j.1467-9507.2007.00389.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Muthén BO, & Muthén LK (2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthen and Muthen. [Google Scholar]
  50. Ostrov JM, & Houston RJ (2008). The Utility of Forms and Functions of Aggression in Emerging Adulthood: Association with Personality Disorder Symptomatology. Journal of Youth and Adolescence, 37(9), 1147–1158. [Google Scholar]
  51. Patterson GR (1982). Coercive family process (Vol. 3): Castalia Publishing Company. [Google Scholar]
  52. Pinquart M. (2017). Associations of parenting dimensions and styles with externalizing problems of children and adolescents: An updated meta-analysis. Developmental psychology, 53(5), 873. [DOI] [PubMed] [Google Scholar]
  53. Polman H, Orobio de Castro B, Koops W, van Boxtel HW, & Merk WW (2007). A meta-analysis of the distinction between reactive and proactive aggression in children and adolescents. Journal of abnormal child psychology, 35(4), 522–535. Retrieved from 17340178. [DOI] [PubMed] [Google Scholar]
  54. Ramakrishnan JL, Garside RB, Labella MH, & Klimes-Dougan B. (2019). Parent Socialization of Positive and Negative Emotions: Implications for Emotional Functioning, Life Satisfaction, and Distress. Journal of Child and Family Studies, 28(12), 3455–3466. 10.1007/s10826-019-01528-z. [DOI] [Google Scholar]
  55. Ramsden SR, & Hubbard JA (2002). Family expressiveness and parental emotion coaching: Their role in children’s emotion regulation and aggression. Journal of abnormal child psychology, 30(6), 657–667. [DOI] [PubMed] [Google Scholar]
  56. Richeson JA, & Sommers SR (2016). Toward a social psychology of race and race relations for the twenty-first century. Annual Review of Psychology, 67, 439–463. [DOI] [PubMed] [Google Scholar]
  57. Sameroff A. (1975). Transactional models in early social relations. Human development, 18(1–2), 65–79. [Google Scholar]
  58. Sanders W, Zeman J, Poon J, & Miller R. (2015). Child Regulation of Negative Emotions and Depressive Symptoms: The Moderating Role of Parental Emotion Socialization. Journal of Child and Family Studies, 24(2), 402–415. 10.1007/s10826-013-9850-y. [DOI] [Google Scholar]
  59. Scarpa A, Haden SC, & Tanaka A. (2010). Being hot-tempered: Autonomic, emotional, and behavioral distinctions between childhood reactive and proactive aggression. Biological psychology, 84(3), 488–496. [DOI] [PubMed] [Google Scholar]
  60. Scott LN, Stepp SD, & Pilkonis PA (2014). Prospective associations between features of borderline personality disorder, emotion dysregulation, and aggression. Personality Disorders: Theory, Research, and Treatment, 5(3), 278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Spear LP (2009). Heightened stress responsivity and emotional reactivity during pubertal maturation: Implications for psychopathology. Development and psychopathology, 21(01), 87–97. 10.1017/s0954579409000066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Steinberg L, & Morris AS (2001). Adolescent development. Journal of Cognitive Education and Psychology, 2(1), 55–87. 10.1891/194589501787383444. [DOI] [Google Scholar]
  63. Steinberg L, & Silk JS (2002). Parenting adolescents. [Google Scholar]
  64. Thomas SA, Wilson T, Jain A, Deros DE, Um M, Hurwitz J, & Dunn EJ (2017). Toward Developing Laboratory-Based Parent-Adolescent Conflict Discussion Tasks that Consistently Elicit Adolescent Conflict-Related Stress Responses: Support from Physiology and Observed Behavior. Journal of child and family studies, 26(12), 3288–3302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Trull TJ, & Ebner-Priemer U. (2013). Ambulatory assessment. Annual review of clinical psychology, 9, 151–176. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Trull TJ, Solhan MB, Tragesser SL, Jahng S, Wood PK, Piasecki TM, & Watson D. (2008). Affective instability: Measuring a core feature of borderline personality disorder with ecological momentary assessment. Journal of abnormal psychology, 117(3), 647. [DOI] [PubMed] [Google Scholar]
  67. Tucker LR, & Lewis C. (1973). The reliability coefficient for maximum likelihood factor analysis. Psychometika, 38, 1–10. 10.1007/BF02291170. [DOI] [Google Scholar]
  68. Vanwoerden S, Hofmans J, & De Clercq B. (2020). Reciprocal effects between daily situational perceptions and borderline personality symptoms in young adulthood: the role of childhood parenting experiences. Psychological medicine, 1–11. [DOI] [PubMed] [Google Scholar]
  69. Vitaro F, Brendgen M, & Barker ED (2006). Subtypes of aggressive behaviors: A developmental perspective. International Journal of Behavioral Development, 30(1), 12–19. [Google Scholar]
  70. Walton A, & Flouri E. (2010). Contextual risk, maternal parenting and adolescent externalizing behaviour problems: The role of emotion regulation. Child: Care, Health and Development, 36(2), 275–284. 10.1111/j.1365-2214.2009.01065.x [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Materials
NIHMS1754486-supplement-Supplemental_Materials.docx (45.1KB, docx)

RESOURCES