Identifying early childhood risk factors of later functional impairment can inform prevention intervention efforts during children’s early years, which represent a period of increased neuroplasticity (Wakschlag et al., 2019). Increasingly, the National Institute of Mental Health’s Research Domain Criteria (RDoC) perspective encourages a focus on dimensional risk that spans traditional diagnostic categories, can be measured at multiple levels, and explains functional impairment (Insel et al., 2010). For example, while one might characterize neurodevelopmental disorders as discrete diagnostic categories [i.e., attention-deficit/hyperactivity disorder (ADHD); autism spectrum disorder (ASD)], children with neurodevelopmental disorders may be more dimensionally characterized as having challenges with impulsivity, hyperactivity, social interaction, and/or emotion dysregulation.

Emotional dysregulation is a powerful predictor of mental-health-related functional impairment (e.g., Mazefsky et al., 2013; Graziano and Garcia, 2016). Emotional regulation is the ability to change or calm one’s emotional state and involves the dual development of both bottom-up approach/avoidance reactivity and top-down executive control (Zelazo and Carlson, 2012; Grabell et al., 2017; Wakschlag et al., 2018). Some emotion regulation ability may be inexorably linked to trait-based emotion reactivity, whereas other emotion strategies are cognitive in nature and are more malleable (Jahromi et al., 2012). Both ADHD and ASD are associated with increased emotion regulation challenges (Shaw et al., 2014; Conner et al., 2021; Day et al., 2022), but the mechanisms by which this is the case are unclear. In ADHD, reduced top-down executive control may contribute to increased emotion dysregulation (Shaw et al., 2014). In ASD, decreased emotion regulation may mechanistically related to ASD characteristics in part due to reduced brain activation during cognitive appraisal of emotive faces (Richey et al., 2015).

Frontal alpha EEG asymmetry is one long-studied physiological index of emotional reactivity within the RDoC’s Arousal and Regulatory Systems domain (Wheeler et al., 1993; Coan and Allen, 2004; Gatzke-Kopp et al., 2014). Brain activity in the alpha frequency band reflects individual differences in awareness and attention (Hanslmayr et al., 2011), and may play a role in both emotion reactivity and regulation (Wheeler et al., 1993). Frontal alpha asymmetry (FAA) is operationalized as the relative difference in frontal alpha activation between left and right hemispheres and is thought to reflect approach (left) and avoidance (right) impulses. FAA measured during a resting state is viewed as a temperament or “trait” measure of approach/avoidance emotional reactivity. Resting FAA predicts clinically significant individual differences in internalizing and externalizing symptomatology in children (Baving et al., 2002; Blackhart et al., 2006; Gatzke-Kopp et al., 2014).

However, FAA may more accurately index the top-down executive control aspect of emotion regulation than bottom-up emotion reactivity. The asymmetric inhibition model (Grimshaw and Carmel, 2014) explains the relation between FAA and the executive function-dependent aspects of emotion regulation. They cite EEG studies using source localization procedures to highlight that FAA may most directly reflect the activity of the dorsolateral prefrontal cortex (dl-PFC). The dl-PFC is centrally involved in executive functioning, including behavioral inhibition, and is active during the cognitive appraisal of emotion. According to the model, the left dl-PFC acts to inhibit avoidance-based motivational urges as they distract from one’s intentional goals, while the right dl-PFC acts to inhibit approach-based motivation as they distract from one’s intentional goals. In this manner, relatively lower left (and therefore, higher right) FAA reflects poorer executive control of internalizing emotions like sadness and anxiety, while relatively lower right (and therefore, higher left) FAA reflects poorer executive control of externalizing, approach-oriented emotions like anger and joy. There is ample literature to reflect this left-externalizing vs. right-internalizing FAA pattern (see Grimshaw and Carmel, 2014 for a review).

FAA may contribute differentially to cascading clinically impairing challenges with affective regulation in children with different neurodevelopmental disorders. At least 50% of autistic children and 33% of children with ADHD present with co-occurrence of anxiety and depression, including 15–20% with a mood disorder by preschool or elementary school (Salazar et al., 2015; Schendel et al., 2016). Executive function (EF) deficits are highly linked to the behavioral regulation of emotion in both ASD and in all children (Demetriou et al., 2019). EF is a core deficit in ADHD (Seidman, 2006) and linked to emotion dysregulation in these children (Garcia et al., 2020). Autistic children also have less developed executive function (Demetriou et al., 2019), including delay of gratification (Faja and Dawson, 2013), a skill that involves the top-down inhibition of a prepotent response in the “hot” or emotionally valenced context of reward.

There is clear evidence of diagnostic group differences in resting FAA for ADHD, but findings are mixed for autism. Children with ADHD demonstrated greater relative left FAA during a behavioral inhibition task than did NT children (Ellis et al., 2017), which suggests that they have higher approach motivation or less inhibited avoidance motivation than NT children. Similarly, adults with ADHD have greater relative left FAA than NT adults (e.g., Keune et al., 2015). In autism, a study of 6- to 18-month-olds at higher likelihood for ASD by virtue of having an older sibling with ASD found that these infants had relatively greater right FAA than “low likelihood” infants at 6 months, but that this pattern faded by 12 months (Gabard-Durnam et al., 2015). Children aged 9–14 years with ASD and no intellectual impairment were found to have relatively greater left midfrontal activation than typically developing children (Sutton et al., 2005). A similar study of 8- to 15-year-old children with ASD and below average to superior verbal IQ found no diagnostic group differences in the number of children with relative left versus right FAA, measured dichotomously (Burnette et al., 2011). To summarize, group differences in FAA are complicated by measurement (i.e., dichotomous vs. continuous asymmetry; lateral vs. midfrontal) and cognitive ability. Inconsistent findings also potentially strengthen the hypothesis that FAA may better reflect transdiagnostic characteristics than diagnostic categories.

It appears that FAA predicts emotion dysregulation for children in general, but it is unclear whether diagnostic group (autism, ADHD) moderates the strength of this relationship. For all children, emerging evidence supports the relation between relatively greater resting right FAA and broad measurements of later internalizing symptoms (Diego et al., 2006; Smith and Bell, 2010) as well as relatively greater resting left FAA and later externalizing symptoms (Smith and Bell, 2010). There is mixed evidence as to whether the relation between FAA and emotion dysregulation differs for autistic children compared to neurotypical (NT) children or children with ADHD. In a study of autistic adolescents, Schiltz et al. (2018) found that individuals with relatively greater right FAA had greater anxiety symptoms, which is consistent with previous literature on FAA and anxiety in the general population. However, in the two studies of FAA in slightly younger children with ASD and NT development, the relation between asymmetry and internalizing symptoms in ASD is less clear. Specifically, for their ASD samples, both studies found that increased relative left FAA was correlated with greater general anxiety (Sutton et al., 2005) and higher OCD and anger-related symptoms (Burnette et al., 2011), which is opposite what would be expected. Finally, while left FAA may predict behavioral inhibition in ADHD (Ellis et al., 2017), there has been little research on whether FAA predicts EF in emotionally valenced contexts (i.e., “hot” EF) in autism.

Children with ADHD displayed greater left (relative to right) frontal alpha activation (FAA) compared to both autistic and neurotypical children. While children with ADHD displayed greater challenges with “hot” EF (reward contingency learning) than did autistic children, children with co-occurring autism and ADHD were reported to have greater emotion regulation challenges than neurotypical (NT) children and children with autism alone. Higher relative left FAA predicted worse reward contingency learning for all children, but it was not significantly related to parent-reported emotion dysregulation. The strength of these relations did not differ by diagnostic group.

Our hypothesis that children with ADHD would display greater left (relative to right) FAA compared to both autistic and neurotypical children was supported and is consistent with the literature (e.g., Keune et al., 2015; Ellis et al., 2017). However, we did not find an omnibus effect or any differences for autistic individuals compared to other groups. There is additional evidence that neurodevelopmental group differences in FAA may be moderated by both participant characteristics and task context. For example, researchers found a gender difference such that 4- to 8-year-old males with ADHD had relatively greater resting left FAA than NT males, while females with ADHD had relatively greater right FAA than NT females (Baving et al., 1999). Compared to NT children, 7- to 14-year-old children with ADHD had relatively greater left FAA during failed trials of a go/no-go task, a measure of inhibition, but not during other trials, although FAA was not measured at rest (Ellis et al., 2017).

We found that children’s relative left FAA was related to their executive function in a “hot,” emotionally valenced context. We also found that reward contingency learning scores during the Hungry Donkey task were worse for children with ADHD compared to all other groups, suggesting that this type of hot EF may be an ADHD-specific deficit. However, the relation between greater relative left FAA and hot EF performance held across all children. In other words, relatively greater approach motivation (or a lack of adaptive inhibition to avoidance) relates to reward contingency learning, regardless of neurodevelopmental diagnosis. This suggests that children who are better at inhibiting the extent to which their emotional reactivity affects their behavior may make better long-term decisions “under pressure.” That this finding is not specific to—or stronger for—children with ADHD, but rather extends to all children, suggests one of two things: that the effect of EEG-indexed emotional reactivity on in-the-moment EF behaviors is transdiagnostic, or that its effect is stronger and more detectable in relation to behavioral tasks than to more global or parent-report measures. For example, He et al. (2010) found that infants’ relative left FAA (i.e., increased approach motivation) was associated with increased anger via a behavioral observation measure (rather than parent-report measure). This is one of the few studies in children to demonstrate a link between frontal asymmetry and behaviorally measured action urges in emotionally valenced situations.

We observed differing levels of emotion dysregulation by diagnostic group, but FAA appears to be unrelated to emotion dysregulation in our sample. In contrast, Richey et al. (2015) found that for adults with ASD, activation in the dl-PFC (the same area that is theorized to be indexed by EEG FAA) indexed via fMRI was reduced when viewing emotive faces, compared to neurotypical adults. Therefore, while we found evidence for a neural mechanism (i.e., FAA) by which approach motivation may be related to reward contingency learning, we do not find a neural mechanism for emotion dysregulation. We also do not find group differences in FAA levels or the relation between FAA and our affective variables of interest, suggesting that FAA-based neural mechanisms may be transdiagnostic.

Children with co-occurring autism and ADHD had the highest reported challenges with emotion regulation, pointing toward an additive effect of neurodevelopmental differences on outward signs of emotion regulation. This kind of additive effect has rarely been observed (England-Mason, 2020). That FAA does not significantly predict or underpin these diagnostic differences in emotion dysregulation may point toward diagnosis-specific functional pathways for emotion dysregulation. For example, individuals with ASD but relatively low symptom severity may have increased internalizing or externalizing symptomatology as a result of the effort needed to “compensate” for social communication challenges inherent to the disorder (Livingston et al., 2019). While for children with ADHD, emotional reactivity may be more directly linked with impulsivity-related EF deficits, and some researchers even posit an “irritable subtype” of ADHD, for which emotion regulation challenges are central (Karalunas et al., 2019). More research is needed to explore these possibilities.

This study was one of the first to examine whether children’s relative left or right frontal alpha activation (FAA) was related to their executive function (EF) in a “hot,” emotionally valenced context. We found that for all children, greater relative left FAA, or more approach-oriented motivation, related to worse functioning during a hot EF task. While there has been much more research linking FAA to internalizing and externalizing symptoms, we did not find a relation between FAA and emotion dysregulation-based symptoms in our sample, for all children or for children with ADHD or ASD specifically. Instead of comparing children by diagnostic category, commonly observed variability in functional impairment in children with neurodevelopmental disorders may be better explained by using an RDoC approach to explore dimensional characteristics in approach/avoidance reactivity, executive function, and clinically impairing emotion dysregulation.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: The dataset analyzed for this study can be found in the National Database for Autism Research (NDAR reference number 2030).

The studies involving human participants were reviewed and approved by Boston Children’s Hospital IRB and University of Washington IRB. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

SE conceptualized the study, conducted the statistical analyses, and drafted the manuscript. JF and SF assisted in conceptualizing the study and reviewed intermediate and final drafts of the manuscript. YB processed the EEG data, drafted the EEG methods, and reviewed the final manuscript. RG collected behavioral and EEG data, conducted preliminary analyses, and reviewed the final manuscript. All authors contributed to the article and approved the submitted version.

This study was supported by NICHD R00HD071966 to SF.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.