To what extent does account for aggression associated with ADHD

symptoms? An experience sampling study

Aja Murray1, Jennifer Lavoie2, Tom Booth1, Manuel Eisner2,3, Denis Ribeaud3

1Department of Psychology, University of Edinburgh

2Institute of Criminology, University of Cambridge

3Jacobs Center for Productive Youth Development, University of Zurich Abstract

Previous research has suggested that aggression is associated with ADHD symptoms and that this may partly reflect problems with emotional regulation. However, previous tests of this hypothesis have yielded inconsistent results and have focused on childhood. In this study we examined the role of emotional dysregulation in the association between ADHD symptoms and aggression in adulthood using experience-sampling derived measures of emotional dysregulation as it occurs in the context of daily life. Data came from the D2M study, a sub-study of the longitudinal z-proso study. Using structural equation modelling, we found that ADHD symptoms were associated with both emotional lability and aggression, but emotional lability did not mediate the ADHD-aggression association.

Results suggest that other factors, such as those specifically related to behavioural dysregulation, may be more important for explaining the elevated levels of aggression in ADHD. Aggression commonly occurs with ADHD symptoms, where it represents a major source of impairment. In childhood it represents one of the most common reasons for referral to mental health services, can undermine the formation of healthy peer relationships, and can interfere with academic functioning (e.g., see Saylor & Amann, 2016 for a review). Towards and during emerging adulthood, defined as the period between 18 and 25 years, both aggression and ADHD symptoms decline on average; however, many individuals continue to experience symptoms into adulthood, putting them at risk of lifelong psychosocial impairment (e.g., Döpfner et al., 2015; Von Polier et al., 2012).

Emerging adulthood is often considered a critical period with respect to aggression. It is at this stage that individuals are afforded opportunities to adopt new prosocial roles (e.g., gaining employment and/or entering long-term intimate partnerships) that promote desistence from antisocial activities, including aggression, whilst also undergoing psychological maturation to attain improved perspective-taking, responsibility, and temperance (Monahan et al., 2013; Wensveen et al., 2017).

However, individuals, such as those with ADHD symptoms, who have difficulties in adopting these new roles and/or who continue to experience difficulties in domains such as impulse control may be particularly vulnerable to a persistent pattern of aggression into adulthood. Thus, there is considerable in identifying the mechanisms that put individuals with ADHD symptoms at risk of aggression such that these can be targeted in interventions. This may be especially important during critical periods such as emerging adulthood. In this study, we therefore examined a potential malleable factor that can be targeted in interventions and tested the hypothesis that emotional dysregulation mediates the association between ADHD symptoms and aggression in emerging adulthood.

It has been proposed that individuals who experience elevated ADHD symptoms may be particularly prone to certain types of aggression (Bennett et al., 2004; Murray et al., 2016). In particular, a distinction can be made between reactive forms of aggression, which are impulsive, and emotionally ‘hot’ and proactive forms of aggression, which are cold, calculated and instrumental acts

(Raine et al., 2006). Previous research suggests that while reactive and proactive aggression are highly correlated, they do emerge as separate factors in factor analyses and have partially distinct patterns of correlates, developmental trajectories, and treatment responses (e.g., Cui et al., 2016; Hubbard et al.,

2010; Polman et al., 2007; Raine et al., 2006; Vitaro et al., 1998).

Reactive aggression has been argued to be particularly linked to ADHD, the rationale being that under-controlled responses to provocation that define reactive aggression reflect exactly the kind of self-regulation difficulties that are assumed to be core to ADHD (Bennett et al., 2004; Murray et al., 2016; Saylor & Amann, 2016). However, several studies have shown that while ADHD symptoms are indeed strongly associated with reactive aggression, they also correlate with proactive aggression

(Bennett et al., 2004; Murray et al., 2016, 2018). One proposed explanation for the association with proactive aggression is that it reflects an indirect association via peer deviancy training. That is, the

ADHD-proactive aggression association is hypothesised to reflect the fact that individuals with

ADHD symptoms tend to be rejected by normative peers and consequently gravitate towards anti- social peers who may model proactive aggression (Bennett et al., 2004). However, it is not inconceivable that there is also a direct effect of ADHD symptoms on proactive aggression. For example, utilising aggression may be perceived as a quicker route to achieving a desired end for individuals with ADHD and attendant poor delay tolerance and a diminished ability to persist with longer term but more socially acceptable strategies.

Though previous discussions have not always distinguished reactive and proactive functions of aggression, a prominent idea regarding the link between ADHD symptoms and aggression is that emotional dysregulation plays an important role. Emotional dysregulation is commonly associated with ADHD symptoms, where it exceeds general population levels for both children and adults

(Donfrancesco et al., 2015; Reimherr et al., 2005) and is associated with a range of functional impairments (Anastopoulos et al., 2011). Though current diagnostic criteria do not recognise it as such, it has been argued by some to represent a potential core symptom of ADHD, alongside behavioural dysregulation, which is currently recognised among the core symptoms (American

Psychiatric Association, 2013; Faraone et al., 2019; Shaw et al., 2015). Emotional dysregulation can be conceptualised as having multiple facets and possible instantiations (see Faraone et al., 2019 for a review); however, Slaughter et al. (2019) proposed that the lability of negative may be a particularly relevant for explaining the link between ADHD and aggression. Slaughter et al. (2019) hypothesised a particular link with reactive aggression, arguing that the link with proactive aggression, which they note is not -based, may be related to other factors.

Negative emotional lability can be defined as the tendency to experience sharp, intense shifts of negative emotion (e.g., , , ) that are not commensurate with situation or developmental stage (e.g., Posner et al., 2014). Though useful questionnaire measures of emotional lability have been developed, negative emotional lability is arguably best operationalised using experience sampling designs that can capture changes in emotions from moment-to-moment in the course of daily life (Factor et al., 2014; Skirrow et al., 2014; Van Liefferinge et al., 2018; Walerius et al., 2016, 2018). As they are completed during or shortly after a relevant event (e.g., which could include the experience of an affective state), experience sampling measures of emotional lability have greater ecological validity and can help overcome issues such as retrospective recall bias that questionnaire measures (Rosen & Factor, 2015). Thus, studies that have used experience sampling designs are especially valuable in assessing a mediating role of emotional dysregulation in the association between ADHD symptoms and aggression (e.g., Leaberry et al., 2017; Rosen & Factor,

2015; Slaughter et al., 2019).

While Slaughter et al. (2019) did not directly test whether experience sampling derived negative emotional lability mediated the association between ADHD symptoms and reactive aggression, all three variables were significantly correlated in their study of n=96 children (53 of whom had ADHD), which is suggestive of at least a partial mediating effect. ADHD symptoms and negative emotional lability were also correlated with proactive aggression, suggesting that ostensibly non-emotion driven forms of aggression are also related – directly or indirectly – to emotional lability.

Other studies have provided additional preliminary evidence for a potential mediating role of negative emotional lability in the association between ADHD symptoms and aggression. Rosen & Factor,

(2015) found that experience sampling derived emotional lability (positive affect and negative affect were not distinguished) based on parent reports was significantly associated with proactive aggression

(r=.62) in a sample of n=27 children with ADHD. However, others have found only partial evidence. Rosen et al. (2015) found in a sample of n=102 children that while experience sampling derived emotional lability was significantly correlated with proactive aggression, it was not correlated with

ADHD diagnostic status nor reactive aggression. There thus remains a need for further clarification of the role of negative emotional lability in aggression associated with ADHD symptoms.

While there have been no direct experience sampling-based studies examining the relations between ADHD symptoms, emotional lability, and aggression in emerging adults, available evidence is generally consistent with the view that emotional lability may remain a potentially important factor in aggression associated with ADHD in adulthood. Skirrow et al. (2014) used an experience sampling design to show that adults with ADHD (n=41) tended to show heightened intensity and instability of and , and greater intensity of anger compared with controls (n=47), which could conceivably act as antecedent factors to aggression.

The goal of the current study was to build on these studies and to employ an experience sampling based methodology to evaluate the role of emotional dysregulation, specifically, emotional lability in the co-occurrence of ADHD symptoms and aggression. Further, while previous studies have tended to focus on this issue in childhood, we evaluated whether the hypothesised role of emotional dysregulation holds in emerging adulthood. To avoid the potential inflation of ADHD- aggression-emotional dysregulation co-occurrence estimates due to Berkson’s bias i.e., the over- estimation of the extent of symptom co-occurrence in clinically ascertained samples (e.g., Pearce &

Richiardi, 2014) and reflecting the fact that both ADHD symptoms and aggression vary meaningfully both above and below clinical cut-offs, we used a community-ascertained rather than a clinical sample. We hypothesised that negative emotional lability would significantly mediate the association between ADHD symptoms and aggression, irrespective of the form or function of the aggression.

Specifically, we hypothesised that higher levels of ADHD symptoms would be associated with greater negative emotional lability which would, in turn lead to higher levels of both reactive and proactive aggression.

Methods Ethics

Ethical approval was obtained from the Ethics Committee from the Faculty of Arts and Social

Sciences of the University of Zurich (ECFASSUZH). Participants provided informed consent before participating. For the experience sampling component of the study, the study information sheet instructed participants not to answer prompts during times where it would not be safe to do so (e.g., while driving).

Participants

Participants were from the D2M (‘decades-to-minutes’) study which is an experience sampling study embedded within the 13-year longitudinal z-proso cohort study (M. Eisner & Ribeaud,

2007; N. L. Eisner et al., 2018). The D2M study was designed to investigate the links between long- term developmental processes and the short-term day-to-day processes underpinning aggression.

Participants were initially identified as being eligible for the main z-proso cohort study when entering the first grade in 2004 (the main z-proso cohort is a school-year based cohort) and have since been invited to participate in up to 8 main waves of data collection until the most recent wave at age 20 in

2018. The main z-proso sample has been found to be reasonably representative of the underlying same-aged population (Eisner et al., 2018). Participants in the current study (n=260; 100 males; median age 20) represented a convenience sample from the main cohort. The only inclusion criterion was that participants were required to have access to a smartphone running iOS or android operating systems in order to be able to download the smartphone application through which the experience sampling survey was administered. As this is estimated to cover 99.6% of all smartphones (Thai &

Page-Gould, 2017), this was not thought to have been a significant sample biasing factor. The D2M sample do not appear to differ from the main cohort on their levels of ADHD symptoms

(t(434.9)=1.20, p=.17); however, they score slightly lower on aggression (t(516.7)=-2.92, p=.004).

The D2M sample also has a higher proportion of females than the main cohort (which has an approximately equal gender split).

Measures ADHD symptoms

ADHD symptoms were measured using an adapted self-reported version of the Social

Behavior Questionnaire (SBQ; Tremblay et al., 1991), administered as part of the age 20 main wave of z-proso. There were nine items measuring inattention, hyperactivity, and impulsivity symptoms, including restlessness, being easily distracted, having difficulties concentrating, acting without thinking, forgetfulness, difficulties paying attention, being hectic and fidgety, inability to settle to anything for long, and internal of restlessness. Responses were recorded on a five-point

Likert-type scale from never to very often.

The psychometric properties of versions of the SBQ have been examined in previous studies in the current and in other independent samples. These studies have provided support for the reliability, structural validity, and criterion validity of the its scores (Murray, Eisner, Obsuth, et al.,

2017; Murray, Eisner, & Ribeaud, 2017; Murray, Obsuth, et al., 2017; Tremblay et al., 1991). In the current sample, omega reliability (omega total; McDonald, 1999) for the ADHD symptoms scale was .87.

Aggression

Aggression was also measured by an adapted version of the SBQ administered at the age 20 main z-proso wave. There were 15 items measuring aggression, covering different forms and functions of aggression. Specifically, there were four items each measuring reactive, proactive and indirect aggression, and three items measuring physical aggression. Reactive aggression items referred to behaving aggressively when teased, when insulted, when something has been taken from the respondent, and when the respondent didn’t get something they wanted. Proactive aggression items referred to trying to scare others into doing something; bossing others around; humiliating others; and threatening others to get something. The indirect aggression items referred to talking badly about someone behind their back; inciting others to dislike another; actively excluding someone in a social situation; and sharing the secrets of another person when annoyed by that person. The physical aggression items referred to violently attacking another person; hitting, biting, or kicking; and engaging in a brawl. Responses were recorded on a five-point scale from never to very often. In the current sample, omega reliability for the reactive aggression scale was .65, for the proactive scale it was .65, for indirect aggression it was .67, and for physical aggression it was .76.

Emotional lability

Emotional lability was measured in the experience sampling component of D2M. Affective state at each prompt was measured using a slightly abbreviated and adapted version of the negative affect schedule of the Positive Affect Negative Affect Schedule Expanded (PANAS-X; Watson &

Clark, 1999). An abbreviated version was used to reduce participant burden. Data were collected on the following affective states: afraid, scared, hostile, guilty, ashamed, distressed, and stressed, with responses recorded on a five-point Likert-type scale from Very slightly or not at all to Extremely.

Mean squared successive differences (MSSDs; Von Neumann et al., 1941) were then calculated for each of the emotion items separately, to obtain a measure of the variability in emotion for each individual over the experience sampling period (Jahng et al., 2008).

Procedure

The D2M study from which the data were drawn has two linked components: a traditional questionnaire component and an experience sampling component. Participants completed questionnaires during the main age 20 data collection wave of the ongoing z-proso study. In addition to the above-described measures, these questionnaires included measures on topics such as substance use, stress, social exclusion, delinquency, internalising problems, psychosis-like symptoms, prosociality, in-keeping with the substantive focus of z-proso on psychosocial development.

D2M participants completed experience sampling measures as part of an optional add-on study taking place around the same time as the main age 20 data collection wave. Participants downloaded an application on their smartphones, through which survey questions were delivered four times a day over a 14-day period. Survey questions were delivered at quasi-random intervals between the hours of 10am and 10pm. Participants were prompted to complete the survey questions via notifications from the experience sampling application. As well as the above-described emotion measures, participants also reported on their current context (who they were with and what they were doing), their stress, aggressive behaviour (and associated provocations), and substance use.

Participants received a reward for participation that was proportional to their level of compliance.

They could receive a maximum of 50CHF (~50USD) if they completed at least 70% of the measures on both weeks 1 and 2 of the study.

Analysis

Structural equation modelling was used to examine the relations between ADHD symptoms, negative emotional lability, and different types of aggression. Each construct was specified as a latent factor or set of latent factors. We did not have independent data on which to develop our measurement models, therefore, we used a combination of exploratory factor analyses and preliminary confirmatory factor analyses in the same data to help develop optimal measurement models. Our measurement models will thus require replication in future studies. ADHD was specified as a single latent variable based on preliminary exploratory factor analyses that suggested that the inattention and hyperactivity/impulsivity items did not clearly separate out into distinct factors. This lack of distinct inattention and hyperactivity/impulsivity factors based on self-reports has previously been found in the current sample (Murray et al., 2018) and likely reflects the relative brevity of the scale.

Emotional lability was specified as a single latent variable; however, correlated residuals were included between three sets of items: scared and afraid; guilty and ashamed; and hostile and upset.

This was based on modification indices and expected parameter changes indicating these as local areas of mis-fit. It seemed reasonable to include these additional parameters in the model as the affective states to which the correlations pertained appeared conceptually similar and could, therefore, plausibly covary in their MSSDs over and above their covariation due to a common dependence on general emotional lability. Aggression was specified as a set of correlated factors: indirect aggression, reactive aggression, proactive aggression, and physical aggression. These groupings were based on the questionnaire design. We also tested a single-factor model and bi-factor model for aggression, but the above-described oblique model was judged most suitable because it was more parsimonious than the bi-factor model, while still allowing different forms/functions of aggression to be differentiated. All latent factors were scaled by fixing the loading of a reference indicator to 1. Each aggression latent variable was regressed on both the emotional lability latent variable and the ADHD symptoms latent variable. The emotional lability latent variable was also regressed on the ADHD symptoms latent variable. The regression of aggression on ADHD symptoms allowed the estimation of the direct effect of ADHD symptoms on aggression. The indirect effects (via emotional lability) were estimated by multiplying the coefficient for the regression of emotional lability on ADHD symptoms by the coefficient for the regression of the relevant aggression variable on emotional lability. The statistical significances of the indirect effects were based on bootstrapped standard errors, using 1000 bootstrapped samples.

Models were fit using lavaan (Rosseel, 2012) in R statistical software and using full information maximum likelihood estimation (FIML). The use of FIML also addressed item-level missingness and provided unbiased parameter estimates provided that data are missing at random

(MAR; Rubin, 1976), meaning that missingness is at random conditional on the predictors included in the model.

Results

On balance, the structural equation model had acceptable fit, with CFI and TLI values falling short of ideal minima but good RMSEA and SRMR values (CFI=.88, TLI= .87, RMSEA=.059,

SRMR=.068). Results from this model are summarised in Figure 1, which provides the standardised parameter estimates for the structural part of the structural equation model only. Significant paths (at p<.05) are indicated by solid lines while non-significant paths are indicated with dashed lines. The latent aggression factor correlations (also not shown in Figure 1), were as follows: Reactive aggression correlated with indirect aggression at r=.49 (p<.001), with proactive aggression at r=.71

(p<.001), and with physical aggression at r=.47 (p<.011); indirect aggression correlated with proactive aggression at r=.83 (p<.001), and with physical aggression at .15 (p=.21); proactive and physical aggression correlated at r=.67 (p=.006). The measurement model components of the model are provided in Tables 1-3. All loadings were >.3 on the standardised scale. Results suggested that while there was a significant effect of ADHD symptoms on both emotional lability and all four types of aggression, there was no significant effect of emotional lability on any type of aggression. There were also no significant indirect effects of ADHD symptoms on any type of aggression via emotional lability. The unstandardized parameters for the indirect effects were for reactive aggression: b= 0.05 (95% CI= -0.02 to 0.17), for indirect aggression: b=-0.01 (95%CI= -

0.07 to 0.06), for proactive aggression: b= 0.00 (95% CI= -0.04 to 0.06), and for physical aggression: b= -0.01 (95% CI= -0.05 to 0.03).

Discussion

Aggression commonly co-occurs with ADHD symptoms where it can adversely affect psychosocial functioning across a range of domains (e.g., Bacchini et al., 2008; Saylor & Amann,

2016). It has previously been proposed that negative affective lability may partly account for the elevated rates of aggression observed in individuals with ADHD symptoms (Slaughter et al., 2019). In this study, using measures of emotional lability based on real life emotional functioning, we fit a structural equation model to test the hypothesis that emotional lability mediates the association between ADHD symptoms and aggression. We found that ADHD symptoms had a significant effect on emotional lability as well as reactive, proactive, indirect, and physical aggression. However, emotional lability did not mediate the association between ADHD symptoms and aggression.

Our study replicates numerous others that have identified a link between ADHD symptoms and aggression on the one hand (Bennett et al., 2004; Murray et al., 2016) and ADHD symptoms and emotional dysregulation on the other (see Faraone et al., 2019 and Shaw et al., 2015 for reviews). It also adds to the small but growing literature demonstrating associations between ADHD symptoms and manifestations of emotional dysregulation as they occur in the context of daily life, captured through experience sampling methodologies (Miguelez-Fernandez et al., 2018; Skirrow et al., 2014;

Van Liefferinge et al., 2018). However, only a handful of studies have attempted to simultaneously examine ADHD symptoms, aggression, and ‘real life’ emotional dysregulation (Rosen et al., 2015;

Rosen & Factor, 2015; Slaughter et al., 2019) and none have directly evaluated whether the latter mediates the association between the former two constructs. Our results thus add important evidence that while ADHD symptoms are related to both negative emotional lability and aggression, the former association does not explain the latter, based on our findings within this particular sample.

Our results suggest that mechanisms other than those reflected by negative affective lability reflects likely account for the ADHD-aggression association. One possibility is that emotional dysregulation does play a role in aggression, but not the aspects of emotional dysregulation that manifest as negative emotional lability. There remains no clear consensus on an appropriate mechanistic model for emotional dysregulation in ADHD, but it is clear that there could be very many processes involved, including in the selection of provoking situations, production of affective responses, and management of already-generated affective states (Faraone et al., 2019; Posner et al.,

2014; Shaw et al., 2015). Further research is required to better delineate and operationalise these processes in terms of measurable phenomena, and doing so would help clarify both the nature of the emotional dysregulation impairment in ADHD and its role in common ADHD sequelae (Faraone et al., 2019; Shaw et al., 2015). Experience sampling methodologies hold promise in this area because they can be potentially used to estimate and distinguish various aspects of emotional functioning that could map to different emotional dysregulation profiles, such as overall tendency to experience negative affect; the strength of coupling between affective stimuli and emotional responses; complexity of affective states (e.g., extent of co-occurrence of different emotions, such as anger and ); and the variability of affective states over different time lags in the context of real life (see e.g.,

Mejía et al., 2014). They are thus likely to provide an important more ecologically valid complement to lab-based and questionnaire-based measures of emotional dysregulation. However, there also remains work to be done in establishing what experience-sampling derived indices of emotional lability reflect. For example, individuals with ADHD symptoms could manifest emotional lability because they tend to select themselves into provoking situations at a high rate where their affective responses are actually appropriate to situation (Skirrow et al., 2014), because they have difficulties in controlling the generation of emotions, or because they have difficulties in managing existing affective states (e.g., Faraone et al., 2019). Another possibility is that is that it is behavioural dysregulation rather than emotional dysregulation that is most important for explaining the prevalence of aggression in ADHD. Unlike emotional dysregulation, which is currently considered only an associated feature of ADHD, behavioural dysregulation (e.g., acting on impulse) is considered a core feature of ADHD in contemporary diagnostic models (American Psychiatric Association, 2013). However, emotion may still play an important role. For example, individuals with high levels of ADHD symptoms may experience a stronger coupling between a negative affective states and aggressive behaviour owing to difficulties in resisting the urge to ‘lash out’, even if the affective states themselves are quite proportional to the situation. This possibility could also be tested using an experience sampling design, by evaluating whether the strength of coupling between negative affective states and aggression is stronger among individuals with ADHD symptoms and, in turn, whether this mediates their association with trait measures of aggression.

Emotional lability itself may be more important for emotional problems such as and , which are also commonly associated with ADHD symptoms (Jarrett & Ollendick, 2008;

Roy et al., 2015). Consistent with this, previous research in the current sample found that emotional dysregulation partially mediated the relation between ADHD symptoms and internalising problems

(Murray et al., 2020) and other research has pointed to similar findings (Anastopoulos et al., 2011;

Leaberry et al., 2017; Rosen & Factor, 2015). However, as with the literature on aggression, there are currently few studies that have tested this hypothesis using measures of emotional functioning in daily life (e.g., through experience sampling), therefore, further work in this area is required to understand which ADHD sequelae emotional dysregulation contributes to, and how.

Limitations

It is important to consider the limitations of the current study. First, we were restricted to testing cross-sectional mediation because only one burst of experience sampling was available. There are currently very few studies that have longitudinal experience sampling data; however, as more such data becomes available, the current findings could be replicated using longitudinal mediation analysis

(Maxwell & Cole, 2007). Second, though the z-proso sample provides a good representation of underlying same-aged population (N. L. Eisner et al., 2018), the D2M sub-sample was a convenience sample from z-proso and is, therefore, not itself representative of that population. However, we were able to test how the sub-sample differed from the main sample in terms of both ADHD and aggression, and given that the differences were minor (non-significant for ADHD), we suspect that this has not biased results substantially. Nevertheless, future replications in more representative samples will be valuable for assessing the generalisability of results. Finally, our factor analyses suggested that our ADHD items did not separate out into separate ADHD dimensions (e.g., inattention versus hyperactivity versus impulsivity), therefore, we were not able to evaluate whether different subdimensions of ADHD are particularly associated with emotional dysregulation and/or aggression.

Previous studies suggest that those with hyperactive symptoms alone or in combination with inattention are especially likely to show both (Martel, 2009).

Conclusions

Our results confirm that ADHD symptoms are associated with both emotional dysregulation and reactive, proactive, indirect and physical aggression. Emotional dysregulation did not, however, mediate the association between ADHD symptoms and aggression, suggesting that the association is due to other mechanisms. A potential fruitful avenue of research in the future would be to use an experience sampling design to examine whether the coupling between negative affect states and aggression is stronger among individuals with ADHD symptoms. References

Anastopoulos, A. D., Smith, T. F., Garrett, M. E., Morrissey-Kane, E., Schatz, N. K., Sommer, J. L.,

Kollins, S. H., & Ashley-Koch, A. (2011). Self-regulation of emotion, functional impairment,

and comorbidity among childrenwith AD/HD. Journal of Attention Disorders, 15(7), 583–

592.

Association, A. P. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®).

American Psychiatric Pub.

Bacchini, D., Affuso, G., & Trotta, T. (2008). Temperament, ADHD and peer relations among

schoolchildren: The mediating role of school bullying. Aggressive Behavior: Official Journal

of the International Society for Research on Aggression, 34(5), 447–459.

Bennett, D. S., Pitale, M., Vora, V., & Rheingold, A. A. (2004). Reactive vs. proactive antisocial

behavior: Differential correlates of child ADHD symptoms? Journal of Attention Disorders,

7(4), 197–204.

Cui, L., Colasante, T., Malti, T., Ribeaud, D., & Eisner, M. P. (2016). Dual trajectories of reactive and

proactive aggression from mid-childhood to early adolescence: Relations to sensation

seeking, risk taking, and moral reasoning. Journal of Abnormal Child Psychology, 44(4),

663–675.

Donfrancesco, R., Innocenzi, M., Marano, A., & Biederman, J. (2015). Deficient Emotional Self-

Regulation in ADHD Assessed Using a Unique Profile of the Child Behavior Checklist

(CBCL) Replication in an Italian Study. Journal of Attention Disorders, 19(10), 895–900.

Döpfner, M., Hautmann, C., Görtz-Dorten, A., Klasen, F., Ravens-Sieberer, U., & Group, B. S.

(2015). Long-term course of ADHD symptoms from childhood to early adulthood in a

community sample. European Child & Adolescent Psychiatry, 24(6), 665–673.

Eisner, M., & Ribeaud, D. (2007). Conducting a criminological survey in a culturally diverse context:

Lessons from the Zurich Project on the Social Development of Children. European Journal

of Criminology, 4(3), 271–298. Eisner, N. L., Murray, A. L., Eisner, M., & Ribeaud, D. (2018). A practical guide to the analysis of

non-response and attrition in longitudinal research using a real data example. International

Journal of Behavioral Development, 0165025418797004.

Factor, P. I., Reyes, R. A., & Rosen, P. J. (2014). Emotional impulsivity in children with ADHD

associated with comorbid—Not ADHD—symptomatology. Journal of Psychopathology and

Behavioral Assessment, 36(4), 530–541.

Faraone, S. V., Rostain, A. L., Blader, J., Busch, B., Childress, A. C., Connor, D. F., & Newcorn, J.

H. (2019). Practitioner Review: Emotional dysregulation in attention-deficit/hyperactivity

disorder–implications for clinical recognition and intervention. Journal of Child Psychology

and Psychiatry, 60(2), 133–150.

Hubbard, J. A., McAuliffe, M. D., Morrow, M. T., & Romano, L. J. (2010). Reactive and proactive

aggression in childhood and adolescence: Precursors, outcomes, processes, experiences, and

measurement. Journal of Personality, 78(1), 95–118.

Jahng, S., Wood, P. K., & Trull, T. J. (2008). Analysis of affective instability in ecological

momentary assessment: Indices using successive difference and group comparison via

multilevel modeling. Psychological Methods, 13(4), 354.

Jarrett, M. A., & Ollendick, T. H. (2008). A conceptual review of the comorbidity of attention-deficit/

hyperactivity disorder and anxiety: Implications for future research and practice. Clinical

Psychology Review, 28(7), 1266–1280.

Leaberry, K. D., Rosen, P. J., Fogleman, N. D., Walerius, D. M., & Slaughter, K. E. (2017).

Comorbid internalizing and externalizing disorders predict lability of negative emotions

among children With ADHD. Journal of Attention Disorders, 1087054717734647.

Martel, M. M. (2009). Research Review: A new perspective on attention-deficit/hyperactivity

disorder: emotion dysregulation and trait models. Journal of Child Psychology and

Psychiatry, 50(9), 1042–1051.

Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal mediation.

Psychological Methods, 12(1), 23.

McDonald, R. P. (1999). Test theory: A unified treatment. Mejía, S., Hooker, K., Ram, N., Pham, T., & Metoyer, R. (2014). Capturing intraindividual variation

and covariation constructs: Using multiple time-scales to assess construct reliability and

construct stability. Research in Human Development, 11(2), 91–107.

Miguelez-Fernandez, C., de Leon, S. J., Baltasar-Tello, I., Peñuelas-Calvo, I., Barrigon, M. L.,

Capdevila, A. S., Delgado-Gómez, D., Baca-García, E., & Carballo, J. J. (2018). Evaluating

attention-deficit/hyperactivity disorder using ecological momentary assessment: A systematic

review. ADHD Attention Deficit and Hyperactivity Disorders, 10(4), 247–265.

Monahan, K. C., Steinberg, L., Cauffman, E., & Mulvey, E. P. (2013). Psychosocial (im) maturity

from adolescence to early adulthood: Distinguishing between adolescence-limited and

persisting antisocial behavior. Development and Psychopathology, 25(4pt1), 1093–1105.

Murray, A. L., Booth, T., Obsuth, I., Zirk-Sadowski, J., Eisner, M., & Ribeaud, D. (2018). Testing the

exacerbation and attenuation hypotheses of the role of anxiety in the relation between ADHD

and reactive/proactive aggression: A 10-year longitudinal study. Psychiatry Research, 269,

585–592.

Murray, A. L., Eisner, M., Obsuth, I., & Ribeaud, D. (2017). Situating violent ideations within the

landscape of mental health: Associations between violent ideations and dimensions of mental

health. Psychiatry Research, 249, 70–77.

Murray, A. L., Eisner, M., & Ribeaud, D. (2017). Can the Social Behavior Questionnaire help meet

the need for dimensional, transdiagnostic measures of childhood and adolescent

psychopathology? European Journal of Psychological Assessment.

Murray, A. L., Obsuth, I., Eisner, M., & Ribeaud, D. (2017). Evaluating longitudinal invariance in

dimensions of mental health across adolescence: An analysis of the Social Behavior

Questionnaire. Assessment, 1073191117721741.

Murray, A. L., Obsuth, I., Zirk-Sadowski, J., Ribeaud, D., & Eisner, M. (2016). Developmental

relations between ADHD symptoms and reactive versus proactive aggression across

childhood and adolescence. Journal of Attention Disorders, 1087054716666323.

Pearce, N., & Richiardi, L. (2014). Commentary: Three worlds collide: Berkson’s bias, selection bias

and collider bias. International Journal of Epidemiology, 43(2), 521–524. Polman, H., de Castro, B. O., Koops, W., van Boxtel, H. W., & Merk, W. W. (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.

Posner, J., Kass, E., & Hulvershorn, L. (2014). Using stimulants to treat ADHD-related emotional

lability. Current Psychiatry Reports, 16(10), 478.

Raine, A., Dodge, K., Loeber, R., Gatzke-Kopp, L., Lynam, D., Reynolds, C., Stouthamer-Loeber,

M., & Liu, J. (2006). The reactive–proactive aggression questionnaire: Differential correlates

of reactive and proactive aggression in adolescent boys. Aggressive Behavior: Official

Journal of the International Society for Research on Aggression, 32(2), 159–171.

Reimherr, F. W., Marchant, B. K., Strong, R. E., Hedges, D. W., Adler, L., Spencer, T. J., West, S.

A., & Soni, P. (2005). Emotional dysregulation in adult ADHD and response to atomoxetine.

Biological Psychiatry, 58(2), 125–131.

Rosen, P. J., & Factor, P. I. (2015). Emotional impulsivity and emotional and behavioral difficulties

among children with ADHD: An ecological momentary assessment study. Journal of

Attention Disorders, 19(9), 779–793.

Rosen, P. J., Walerius, D. M., Fogleman, N. D., & Factor, P. I. (2015). The association of emotional

lability and emotional and behavioral difficulties among children with and without ADHD.

ADHD Attention Deficit and Hyperactivity Disorders, 7(4), 281–294.

Rosseel, Y. (2012). Lavaan: An R package for structural equation modeling and more. Version 0.5–12

(BETA). Journal of Statistical Software, 48(2), 1–36.

Roy, A., Hartman, C. A., Veenstra, R., & Oldehinkel, A. J. (2015). Peer dislike and victimisation in

pathways from ADHD symptoms to depression. European Child & Adolescent Psychiatry,

24(8), 887–895.

Rubin, D. B. (1976). Inference and missing data. Biometrika, 63(3), 581–592.

Saylor, K. E., & Amann, B. H. (2016). Impulsive aggression as a comorbidity of

attention-deficit/hyperactivity disorder in children and adolescents. Journal of Child and

Adolescent Psychopharmacology, 26(1), 19–25. Shaw, P., Stringaris, A., Nigg, J., & Leibenluft, E. (2015). Emotion dysregulation in attention deficit

hyperactivity disorder. Focus, 14(1), 127–144.

Skirrow, C., Ebner-Priemer, U., Reinhard, I., Malliaris, Y., Kuntsi, J., & Asherson, P. (2014).

Everyday emotional experience of adults with attention deficit hyperactivity disorder:

Evidence for reactive and endogenous emotional lability. Psychological Medicine, 44(16),

3571–3583.

Slaughter, K. E., Leaberry, K. D., Fogleman, N. D., & Rosen, P. J. (2019). Reactive and proactive

aggression in children with and without ADHD and negative emotional lability. Social

Development.

Tandon, M., Tillman, R., Agrawal, A., & Luby, J. (2016). Trajectories of ADHD severity over 10

years from childhood into adulthood. ADHD Attention Deficit and Hyperactivity Disorders,

8(3), 121–130.

Thai, S., & Page-Gould, E. (2017). ExperienceSampler: An open-source scaffold for building

smartphone apps for experience sampling.

Tremblay, R. E., Loeber, R., Gagnon, C., Charlebois, P., Larivee, S., & LeBlanc, M. (1991).

Disruptive boys with stable and unstable high fighting behavior patterns during junior

elementary school. Journal of Abnormal Child Psychology, 19(3), 285–300.

Van Liefferinge, D., Sonuga-Barke, E., Danckaerts, M., Fayn, K., Van Broeck, N., & Van der Oord,

S. (2018). Measuring child and adolescent emotional lability: How do questionnaire-based

ratings relate to experienced and observed emotion in everyday life and experimental

settings? International Journal of Methods in Psychiatric Research, 27(3), e1720.

Vitaro, F., Gendreau, P. L., Tremblay, R. E., & Oligny, P. (1998). Reactive and proactive aggression

differentially predict later conduct problems. The Journal of Child Psychology and

Psychiatry and Allied Disciplines, 39(3), 377–385.

Von Neumann, J., Kent, R. H., Bellinson, H. R., & Hart, B. t. (1941). The mean square successive

difference. The Annals of Mathematical Statistics, 12(2), 153–162.

Von Polier, G. G., Vloet, T. D., & Herpertz-Dahlmann, B. (2012). ADHD and delinquency–a

developmental perspective. Behavioral Sciences & the Law, 30(2), 121–139. Walerius, D. M., Fogleman, N. D., & Rosen, P. J. (2016). The role of ADHD and negative emotional

lability in predicting changes in parenting daily hassles. Journal of Child and Family Studies,

25(7), 2279–2291.

Walerius, D. M., Reyes, R. A., Rosen, P. J., & Factor, P. I. (2018). Functional Impairment Variability

in Children With ADHD Due to Emotional Impulsivity. Journal of Attention Disorders,

22(8), 724–737. https://doi.org/10.1177/1087054714561859

Watson, D., & Clark, L. A. (1999). The PANAS-X: Manual for the positive and negative affect

schedule-expanded form.

Wensveen, M., Palmen, H., Blokland, A., & Meeus, W. (2017). Examining the work–crime

association in emerging adulthood: A longitudinal analysis based on a Dutch population

sample. European Journal of Criminology, 14(4), 467–484. Tables

Table 1: Standardised factor loadings for the ADHD symptoms latent variable

Item Loading SE p restless 0.57 Fixed Fixed easily distracted 0.70 0.19 <.01 difficulties concentrating 0.74 0.20 <.01 acting without thinking 0.49 0.12 <.01 forgetful 0.62 0.21 <.01 inattentive 0.78 0.20 <.01 hectic and fidgety 0.58 0.08 <.01 unable to settle for long 0.69 0.17 <.01 internal restlessness 0.54 0.15 <.01 Table 2: Standardised factor loadings for the negative emotional lability latent variable

Item Loading SE p MSSD afraid .70 Fixed Fixed MSSD scared .74 0.07 <.01 MSSD hostile .57 0.15 <.01 MSSD guilty .66 0.09 <.01 MSSD ashamed .69 0.10 <.01 MSSD upset .71 0.22 <.01 MSSD distressed .89 0.16 <.01 MSSD stressed .72 0.20 <.01 Note. MSSD= mean squared successive difference. Table 3: Standardised factor loadings for the aggression latent variables

Item Estimate SE p Reactive Aggression aggressive when teased 0.60 Fixed Fixed aggressive when insulted 0.73 0.14 <.01 aggressive when something taken 0.30 0.22 .17 gets mad when not getting something 0.62 0.10 <.01 Indirect Aggression says bad things behind others' backs 0.58 Fixed Fixed incites others to dislike another 0.63 0.19 <.01 actively excludes others 0.54 0.25 .01 tells someone’s secrets when mad at that person 0.55 0.22 .01 Proactive Aggression scares others to force them to do something 0.62 Fixed Fixed bosses others around 0.50 0.35 <.01 humiliates others 0.64 0.23 <.01 threatens others to get something 0.47 0.12 .01 Physical Aggression violently attacks 0.63 Fixed Fixed hit, bites, kicks others 0.86 0.30 0.00 engages in brawls 0.63 0.34 0.00 Figure 1: Standardised structural parameters for the structural equation model examining the direct and indirect effects of ADHD symptoms on aggression

Note. Measurement models and residual covariances between aggression factors omitted for visual clarity. Solid lines represent statistically significant paths (p<.05) and dashed lines represent non- significant paths.