Is emotion regulation a transdiagnostic factor underlying both internalizing and externalizing conditions or does it mediate transitions between these conditions?

Some of the articles reviewed above indicate that emotion regulation might be a transdiagnostic factor associated with the development of internalizing and internalizing conditions (Abela & Hankin, Reference Abela and Hankin2011; Aldao et al., Reference Aldao, Nolen-Hoeksema and Schweizer2010; Bierman et al., Reference Bierman, Kalvin and Heinrichs2015; Halligan et al., Reference Halligan, Cooper, Fearon, Wheeler, Crosby and Murray2013; Hilt et al., Reference Hilt, Armstrong and Essex2015; McLaughlin et al., Reference McLaughlin, Hatzenbuehler, Mennin and Nolen-Hoeksema2011, Reference McLaughlin, Aldao, Wisco and Hilt2014; Oldehinkel et al., Reference Oldehinkel, Verhulst and Ormel2008; Spasojević & Alloy, Reference Spasojević and Alloy2001), although the relative strength of the associations remains to be determined. Other studies found that it mediated transitions between one type of condition and the other one (McLaughlin et al., Reference McLaughlin, Aldao, Wisco and Hilt2014; Nolen-Hoeksema et al., Reference Nolen-Hoeksema, Stice, Wade and Bohon2007). Identifying whether specific emotion regulation processes play a stronger role in the transition from health to psychopathology or from one form of psychopathology to another one is critical for prevention and treatment programs. Forms of emotion regulation that are associated with the former would be targeted in large-scale prevention efforts aimed at the general population (e.g., entire school system, retirement homes, community centers, and work-place programs). Conversely, those more strongly linked with transitions from one form of psychopathology to another one would be targeted in treatment and relapse prevention efforts with those already suffering from a given condition (or perhaps even at high risk for it). Overall, reaching this level of specificity in our understanding of emotion regulation and internalizing and externalizing conditions has great potential for the development of more efficient and cost-effective prevention and treatment efforts, and ultimately, the reduction of suffering.

In addition, it would be important to adopt a more nuanced approach by examining emotion regulation in relation to specific forms of internalizing and externalizing disorder. A number of candidates emerge from the literature. First, it is important to examine the “internalizing pathway” to alcohol abuse, whereby alcohol consumption initially serves the function of regulating internalizing emotions, such as anxiety and depression, and over time becomes reinforced and problematic (e.g., Hussong, Jones, Stein, Baucom, & Boeding, Reference Hussong, Jones, Stein, Baucom and Boeding2011; Kashdan, Ferssizidis, Collins, & Muraven, Reference Kashdan, Ferssizidis, Collins and Muraven2010; Kushner, Sher, & Beitman, Reference Kushner, Sher and Beitman1990; Shadur & Lejuez, Reference Shadur and Lejuez2015; Sher & Grekin, Reference Sher, Grekin and Gross2007). Second, it will be fruitful to explore the link between anxiety and ADHD (e.g., Falk, Lee, & Chorpita, Reference Falk, Lee and Chorpita2015; Lee, Falk, & Aguirre, Reference Lee, Falk and Aguirre2012; Mennin, Biederman, Mick, & Faraone, Reference Mennin, Biederman, Mick and Faraone2000; Schatz & Rostain, Reference Schatz and Rostain2006; Van den Bergh & Marcoen, Reference Van den Bergh and Marcoen2004). In a related vein, it will be important to examine the relationship between anxiety and disruptive behavior disorders, such as oppositional defiant disorder and conduct disorder, especially given the strong commonalities between these disorders in issues with emotion regulation (e.g., Bubier & Drabick, Reference Bubier and Drabick2009; Drabick, Ollendick, & Bubier, Reference Drabick, Ollendick and Bubier2010; Forehand, Jones, & Parent, Reference Forehand, Jones and Parent2013; Fraire & Ollendick, Reference Fraire and Ollendick2013). Third, given the risk of fatality in both depression and substance use disorders, it will be critical for future research to examine the role of emotion regulation in the link between these disorders as well as transitions between them (e.g., Brière, Rohde, Seeley, Klein, & Lewinsohn, Reference Brière, Rohde, Seeley, Klein and Lewinsohn2014; Deykin, Levy, & Wells, Reference Deykin, Levy and Wells1987; Levy & Deykin, Reference Levy and Deykin1989; Lewinsohn, Gotlib, & Seeley, Reference Lewinsohn, Gotlib and Seeley1995).

Which are critical emotion regulation processes for understanding the temporal comorbidity between internalizing and externalizing conditions?

Our survey of the literature revealed that the majority of existing research has focused on trait level self-reports of the use of rumination and perceived regulation skills, and of a particular biomarker of emotion regulation (RSA). Nevertheless, emotion regulation is a complex construct that entails a wide range of targets (e.g., positive and negative emotions), strategies (e.g., reappraisal and avoidance), abilities (e.g., focus on a task in the face of intense emotions), goals (e.g., increased and decreased intensity or duration), and outcomes (e.g., subjective feelings, facial expressivity, motivated behavior, physiological reactivity, and neurobiological activity; Mauss & Robinson, Reference Mauss and Robinson2009). That is, it unfolds in myriad ways that might have different functional relations with symptom development (e.g., Cicchetti et al., Reference Cicchetti, Ackerman and Izard1995; Cole et al., Reference Cole, Michel and Teti1994; Thompson, Reference Thompson1994). Therefore, it is critical to adopt a multiprocess approach to the assessment of emotion regulation that entails focusing on multiple processes simultaneously.

A multiprocess approach has two clear advantages. First, it will allow investigators to include multiple emotion regulation processes in the same statistical models and, thus, identify which aspects of emotion regulation are more central to the development of internalizing versus externalizing conditions. That is, this approach might facilitate a more nuanced understanding of which aspects of emotion regulation might be transdiagnostic risk factors versus those that might have specificity to certain conditions. Second, by assessing multiple forms of emotion regulation, investigators open the door to the possibility of understanding how different emotion regulation processes might influence one another. For example, recent work suggests that the use of maladaptive emotion regulation strategies (e.g., avoidance, rumination, and suppression) moderates the association between the use of adaptive strategies (e.g., acceptance and reappraisal) and symptoms of depression, anxiety, and alcohol abuse (e.g., Aldao, Jazaieri, Goldin, & Gross, Reference Aldao, Jazaieri, Goldin and Gross2014; Aldao & Nolen-Hoeksema, Reference Aldao and Nolen-Hoeksema2012; Conklin et al., Reference Conklin, Cassiello-Robbins, Brake, Sauer-Zavala, Farchione and Cirauloin press). In a similar vein, a number of studies have begun to derive profiles of emotion regulation repertoires and to link them to mental health functioning (e.g., Dixon-Gordon et al., Reference Dixon-Gordon, Aldao and De Los Reyesin press; Hollenstein, Granic, Stoolmiller, & Snyder, Reference Hollenstein, Granic, Stoolmiller and Snyder2004). Further, it will be important to combine the assessment of emotion regulation strategies and abilities (e.g., Tull & Aldao, Reference Tull and Aldao2015).

Investigators can adopt this multiprocess approach in a number of ways. First, as we mentioned above, the majority of the work on the habitual use of regulation strategies has focused on depressive rumination. This strategy, albeit a very important one, is only one of many that people have in their repertoires (e.g., Aldao & Dixon-Gordon, Reference Aldao and Dixon-Gordon2014). In this vein, it will be important to assess the habitual use of additional putatively maladaptive strategies, such as avoidance, suppression, and worry. Similarly, it will be useful to assess the chronic use of putatively adaptive regulation strategies, such as cognitive reappraisal and acceptance (e.g., Aldao et al., Reference Aldao, Nolen-Hoeksema and Schweizer2010) because it is possible that different strategies have different associations with the development of internalizing and externalizing symptoms. Cross-sectional meta-analytic research on adults and children suggests that maladaptive strategies have a moderate effect size in relation to internalizing conditions (e.g., anxiety and depression; d = 0.42), but a small effect size in relation to externalizing ones (e.g., substance abuse and eating disorders, d = 0.25). However, the chronic use of adaptive strategies has small effect sizes in relation to both types of psychopathology (ds = –0.23 and –0.11, respectively; see Aldao et al., Reference Aldao, Nolen-Hoeksema and Schweizer2010). However, it is important to keep in mind that the sample sizes for the externalizing cells in this study were considerably smaller than for the internalizing conditions (k = 30 vs. 211), reflecting an asymmetry in the study of emotion regulation in relation to internalizing versus externalizing conditions. In all, it will be important to examine how different strategies might simultaneously predict the development of internalizing and externalizing conditions as well as mediate the transitions between these disorders.

Second, it will be useful to assess variations of a given strategy or abilities that might be particularly relevant to certain contexts. A clear example is that of depressive rumination (Nolen-Hoeksema, Reference Nolen-Hoeksema1991) and angry rumination (Sukhodolsky, Golub, & Cromwell, Reference Sukhodolsky, Golub and Cromwell2001). Whereas the former focuses on questioning one's actions and capabilities to deal with stressors, the latter entails perseverating on feelings of having been wronged, slighted, and hurt. Therefore, it is possible that depressive rumination might be more strongly associated with the development of internalizing conditions (Nolen-Hoeksema et al., Reference Nolen-Hoeksema, Larson and Grayson1999), whereas angry rumination might be more strongly linked to the development of externalizing problems (Peled & Moretti, Reference Peled and Moretti2007). Testing these strategies simultaneously would be of great value for developing a more nuanced understanding of the functional relationship between rumination and different forms of psychopathology. More broadly, a similar approach can be beneficial by parsing out different types of negative perseverative thinking, including postevent processing, obsessive thinking, and worry.

Third, it might be helpful to anchor trait-level questionnaires of habitual use of strategies and regulation abilities to specific stressors. For example, investigators could ask participants to rate the extent to which they engaged in rumination in response to having being victimized, having felt embarrassed, having underperformed in school or work, or having had romantic setbacks. Moreover, they could assess whether a given regulation ability (e.g., “When I am upset, it takes me a long time to feel better” from the Difficulties in Emotion Regulation Scale; Gratz & Roemer, Reference Gratz and Roemer2004) is more likely to manifest in response to certain types of stressors. This approach would allow investigators to begin to parse out how the use of regulation strategies and abilities in certain situations might pose different risks for internalizing versus externalizing conditions. In a similar vein, investigators could focus on the regulation of internalizing and externalizing emotions (along the lines of the tasks involving mirror tracing and listening to arguments used by Hinnant and El-Sheikh, Reference Hinnant and El-Sheikh2009, Reference Hinnant and El-Sheikh2013). In all cases, the use of experience sampling methodologies could play a key role in facilitating more precise estimates of the link between stressors, regulation, and outcomes (e.g., Silk, Steinberg, & Morris, Reference Silk, Steinberg and Morris2003).

Does social context make a difference?

Our review of the literature suggests that social processes (e.g., peer victimization and delinquent peers) might be important moderators of the link between emotion regulation and the development of internalizing/externalizing symptoms (e.g., Bierman et al., Reference Bierman, Kalvin and Heinrichs2015; Buehler et al., Reference Buehler, Lange and Franck2007; Hilt et al., Reference Hilt, Armstrong and Essex2015). Social context is one of the primary moderators outlined in the developmental psychopathology frameworks of Beauchaine and McNulty (Reference Beauchaine and McNulty2013) and Nolen-Hoeksema and Watkins (Reference Nolen-Hoeksema and Watkins2011). This notion is aligned with a growing literature documenting the crucial role that social processes play in emotion regulation (e.g., Aldao, Reference Aldao2013; Christensen et al., in press; Hofmann, Reference Hofmann2014; Marroquín, Reference Marroquín2011; Shallcross, Frazier, & Anders, Reference Shallcross, Frazier and Anders2014; Troy, Shallcross, & Mauss, Reference Troy, Shallcross and Mauss2013) and with decades of research in developmental psychopathology emphasizing the role that caregivers, siblings, and teachers play in the development of emotion regulation skills (e.g., Cole et al., Reference Cole, Michel and Teti1994; Gee et al., Reference Gee, Gabard-Durnam, Telzer, Humphreys, Goff and Shapiro2014; Gunnar & Donzella, Reference Gunnar and Donzella2002; Hofer, Reference Hofer1994; McCoy & Masters, Reference McCoy and Masters1985). For these reasons, it can be extremely valuable to systematically model social processes in the study of the association between emotion regulation and the development of internalizing and externalizing conditions. Below, we provide a series of suggestions for how to do so.

The first and most straightforward way to incorporate social processes is by examining regulation in response to social stressors (e.g., peer victimization, aggression, divorce, and job performance). A second method is to examine the extent to which individuals recruit others to regulate their emotions (and also help others regulate theirs). This is known as interpersonal emotion regulation, a rapidly growing area of research that used to be primarily confined to infants and younger children (e.g., Cole et al., Reference Cole, Michel and Teti1994; Hofmann, Reference Hofmann2014; Marroquín, Reference Marroquín2011; Zaki & Williams, Reference Zaki and Williams2013). For example, when an adolescent is sad and disappointed about a fight with her boyfriend, she might end up ruminating about these emotions by herself or with a friend. The latter is referred to as corumination and has been linked to increases in symptoms of depression in adolescent girls (e.g., Rose, Reference Rose2002; Rose, Carlson, & Waller, Reference Rose, Carlson and Waller2007). Thus, studying patterns of coregulation might enhance our understanding of the development of internalizing and externalizing conditions.

In this respect, it would be important to examine whether the use of a given regulation strategy in isolation confers greater risk for, or protection against, internalizing and externalizing conditions than its joint use with other people. That is, is rumination by oneself more or less strongly associated with the development of psychopathology than corumination, or does drinking by oneself to regulate one's anxiety confer more or less of a risk than drinking with others to achieve anxiety reduction? Does self-regulation versus coregulation confer different protection against developing internalizing versus externalizing symptoms (a crucial point)?

In the future, a more fine-grained approach could entail distinguishing whether the consequences of coregulation vary as a function of specific social context. For example, coruminating with individuals who are experiencing the same stressors (e.g., coworkers experiencing problems with their bosses) and therefore have less of a “big picture” perspective might be more problematic than with individuals who are experiencing different stressors and might therefore offer a more distanced perspective (Kross, Ayduk, & Mischel, Reference Kross, Ayduk and Mischel2005; Smith & Rose, Reference Smith and Rose2011).

Though it may appear overwhelming to capture interpersonal emotion regulation, a rich amount of information can be collected by adopting the multiple-informant method that is at the cornerstone of developmental psychopathology (e.g., De Los Reyes, Reference De Los Reyes2013; De Los Reyes et al., Reference De Los Reyes, Augenstein, Wang, Thomas, Drabick and Burgers2015; De Los Reyes & Kazdin, Reference De Los Reyes and Kazdin2005; Hunsley & Mash, Reference Hunsley and Mash2007). For example, investigators can assess patterns of interpersonal regulation by asking participants to complete self-report and other-report versions of the use of emotion regulation strategies and abilities. Moreover, with the availability of myriad tools for online data collection, investigators can more easily reach out to friends, family members, significant others, and/or coworkers of participants to request information about their coregulation with the participants. These multiple-informant assessments, when integrated with independent assessments of the social processes described previously, can provide a rich, comprehensive picture of the interplay between social contexts and individual differences in displays of emotion regulation strategies (see De Los Reyes, Bunnell, et al., Reference De Los Reyes, Bunnell and Beidel2013; De Los Reyes et al., Reference De Los Reyes, Henry, Tolan and Wakschlag2009; Mischel & Shoda, Reference Mischel and Shoda1995).

Are there crucial differences across different developmental stages?

The studies we found on emotion regulation as a process related to the development of internalizing and externalizing symptoms primarily focused on children and adolescents. Consequently, much remains to be done in terms of elucidating how this process might play a role in the development of internalizing and externalizing conditions over the course of the lifespan and in relation to critical developmental periods, such as transition to elementary school, adolescence, college, marriage, parenting, divorce, menopause, and cognitive decline, among others. Adopting this expanded developmental psychopathology approach is particularly important in light of a growing literature documenting marked differences in emotion regulation as a function of aging. This work tends to find that older adults regulate their emotions more effectively than younger adults (e.g., Isaacowitz & Blanchard-Fields, Reference Isaacowitz and Blanchard-Fields2012; Livingstone & Isaacowitz, Reference Livingstone and Isaacowitz2015; Nolen-Hoeksema & Aldao, Reference Nolen-Hoeksema and Aldao2011; Prakash, Whitmoyer, Aldao, & Schirda, Reference Prakash, Whitmoyer, Aldao and Schirda2015; Pruzan & Isaacowitz, Reference Pruzan and Isaacowitz2006; Yeung, Wong, & Lok, Reference Yeung, Wong and Lok2011; Zhang, Ersner-Hershfield, & Fung, Reference Zhang, Ersner-Hershfield and Fung2010). According to socioemotional selectivity theory, these differences might stem from older adults being more selective when building their social networks (e.g., Carstensen, Isaacowitz, & Charles, Reference Carstensen, Isaacowitz and Charles1999; Carstensen et al., Reference Carstensen, Turan, Scheibe, Ram, Ersner-Hershfield and Samanez-Larkin2011; Sims, Hogan, & Carstensen, Reference Sims, Hogan and Carstensen2015). What remains to be understood is how these, and other, changes in emotion regulation over time might result in differential patterns of internalizing and externalizing symptoms.

Moreover, elucidating the neurodevelopmental mechanisms underlying age-related changes in emotion regulation and their association with risk for the onset of internalizing and externalizing disorders will be critical to informing early intervention and prevention approaches. Given the dynamic changes that occur in frontolimbic circuitry across the life span, mechanisms of illness and treatment are likely to differ depending on developmental stage (Lee et al., Reference Lee, Heimer, Giedd, Lein, Šestan and Weinberger2014). Research that seeks to identify the similarities and differences in neurobiological mechanisms characterizing different psychiatric disorders will play a critical role in better understanding emotion regulation as a transdiagnostic factor.

Do developmental changes in brain circuitry matter?

In our review of the literature, we found that, despite growing enthusiasm for identifying the neurobiological underpinnings of emotion regulation and psychopathology (e.g., Aldao & De Los Reyes, Reference Aldao and De Los Reyes2015; Beauchaine & Thayer, Reference Beauchaine and Thayer2015; Etkin et al., Reference Etkin, Büchel and Gross2015; Hinnant & El-Sheikh, Reference Hinnant and El-Sheikh2009, Reference Hinnant and El-Sheikh2013; Ochsner & Gross, Reference Ochsner and Gross2005; Oldehinkel et al., Reference Oldehinkel, Verhulst and Ormel2008; Pang & Beauchaine, Reference Pang and Beauchaine2013; Patrick & Hajcak, Reference Patrick and Hajcak2016), much remains to be understood regarding how developmental changes in brain circuitry might play a role in the association between emotion regulation and the development of internalizing and externalizing conditions. Thus, below we provide an overview of developmental research on brain circuitry and provide suggestions for future work.

Interactions between subcortical limbic and cortical prefrontal regions are fundamental to the processing and regulation of emotional reactivity (Banks, Eddy, Angstadt, Nathan, & Phan, Reference Banks, Eddy, Angstadt, Nathan and Phan2007; Kim, Somerville, Johnstone, Alexander, & Whalen, Reference Kim, Somerville, Johnstone, Alexander and Whalen2003; Lieberman et al., Reference Lieberman, Eisenberger, Crockett, Tom, Pfeifer and Way2007; Ochsner, Bunge, Gross, & Gabrieli, Reference Ochsner, Bunge, Gross and Gabrieli2002). Emotion regulation and the connections between cortical and subcortical brain regions undergo dynamic changes across the lifespan, which are likely to have important implications for the development of psychiatric disorders. Interactions between bottom-up, subcortical regions supporting emotional reactivity and top-down, cortical regions underlying regulatory control are central to emotion regulation. In general, subcortical regions, such as the amygdala and ventral striatum, facilitate emotional reactivity and motivational processes that may be more automatic and tendency related, whereas top-down control mediated by prefrontal regions may be more involved in the volitional regulation of emotion (see Beauchaine, Reference Beauchaine2015a).

Findings on the normative maturation of these emotion generation and regulation systems, and the connections between them, can provide an important reference with which to compare deviations from typical development in individuals who develop internalizing and externalizing disorders. Risk for psychiatric disorders increases during development, and many psychiatric disorders have been conceptualized through a neurodevelopmental framework (e.g., Lee et al., Reference Lee, Heimer, Giedd, Lein, Šestan and Weinberger2014; Pine, Cohen, Gurley, Brook, & Ma, Reference Pine, Cohen, Gurley, Brook and Ma1998). Moreover, dynamic changes in this circuitry may help to explain the influence of environmental and genetic factors on the onset of certain disorders at specific developmental windows (e.g., Burghy et al., Reference Burghy, Stodola, Ruttle, Molloy, Armstrong and Oler2012; reviewed in Gee & Casey, Reference Gee and Casey2015; Gee et al., Reference Gee, Fetcho, Jing, Li, Glatt and Drysdale2016). For example, altered cortical–subcortical interactions may be a common possible pathway through which early-life stress and maltreatment influence both internalizing and externalizing behaviors (Burghy et al., Reference Burghy, Stodola, Ruttle, Molloy, Armstrong and Oler2012; Gee, Gabard-Durnam, et al., Reference Gee, Gabard-Durnam, Flannery, Goff, Humphreys and Telzer2013; Hanson et al., Reference Hanson, Chung, Avants, Shirtcliff, Gee and Davidson2010; Herringa et al., Reference Herringa, Birn, Ruttle, Burghy, Stodola and Davidson2013). Understanding the nature of disruptions in the connectivity between bottom-up and top-down regions is also likely to have important implications for the treatment of psychopathology during development.

Subcortical regions involved in emotional reactivity tend to mature earlier in normative development relative to more protracted development of cortical regions. The amygdala undergoes rapid change early in development (reviewed in Tottenham & Sheridan, Reference Tottenham and Sheridan2009). Children show heightened amygdala reactivity to fearful faces and other emotional stimuli, with reactivity typically decreasing following childhood (e.g., Decety, Norman, Berntson, & Cacioppo, Reference Decety, Norman, Berntson and Cacioppo2012; Gee, Humphreys, et al., Reference Gee, Gabard-Durnam, Flannery, Goff, Humphreys and Telzer2013; Silvers, Weber, Wager, & Ochsner, Reference Silvers, Weber, Wager and Ochsner2014; Swartz, Williamson, & Hariri, Reference Swartz, Williamson and Hariri2014; Vink, Derks, Hoogendam, Hillegers, & Kahn, Reference Vink, Derks, Hoogendam, Hillegers and Kahn2014), a finding that parallels decreases in normative fears that occur early in life (Gee, Humphreys, et al., Reference Gee, Humphreys, Flannery, Goff, Telzer and Shapiro2013). Activation in the ventral striatum related to motivational cues has been shown to peak during adolescence (e.g., Galvan et al., Reference Galvan, Hare, Parra, Penn, Voss and Glover2006; Somerville, Hare, & Casey, Reference Somerville, Hare and Casey2011). By contrast, the prefrontal cortex (PFC) undergoes more protracted development into young adulthood (e.g., Chareyron, Lavenex, Amaral, & Lavenex, Reference Chareyron, Lavenex, Amaral and Lavenex2012; Lenroot & Giedd, Reference Lenroot and Giedd2006; Machado & Bachevalier, Reference Machado and Bachevalier2003; Payne, Machado, Bliwise, & Bachevalier, Reference Payne, Machado, Bliwise and Bachevalier2010). The connections between these subcortical regions and the PFC also show protracted maturation, with these developmental shifts underlying dynamic behavioral changes in regulation during childhood and adolescence (see Casey, Galván, & Somerville, Reference Casey, Galván and Somerville2016).

Connections between the amygdala and medial PFC that support effective emotion regulation in adults change substantially across the course of childhood and adolescence (e.g., Decety et al., Reference Decety, Norman, Berntson and Cacioppo2012; Gabard-Durnam et al., Reference Gabard-Durnam, Flannery, Goff, Gee, Humphreys and Telzer2014; Gee, Humphreys, et al., Reference Gee, Humphreys, Flannery, Goff, Telzer and Shapiro2013; Lebel et al., Reference Lebel, Gee, Camicioli, Wieler, Martin and Beaulieu2012; Perlman & Pelphrey, Reference Perlman and Pelphrey2011; Swartz et al., Reference Swartz, Williamson and Hariri2014; Vink et al., Reference Vink, Derks, Hoogendam, Hillegers and Kahn2014), consistent with behavioral changes in emotion regulation during development (e.g., Bunge, Dudukovic, Thomason, Vaidya, & Gabrieli, Reference Bunge, Dudukovic, Thomason, Vaidya and Gabrieli2002; Tottenham, Hare, & Casey, Reference Tottenham, Hare and Casey2011). For example, the regulatory inverse pattern of frontoamygdala functional connectivity that is associated with effective emotion regulation in adults appears to emerge around the transition from childhood to adolescence (Gee, Humphreys, et al., Reference Gee, Gabard-Durnam, Flannery, Goff, Humphreys and Telzer2013). Prior to adolescence, external sources of emotion regulation, such as parents, have been shown to regulate children's behavior via modulation of frontoamygdala connectivity (Gee et al., Reference Gee, Gabard-Durnam, Telzer, Humphreys, Goff and Shapiro2014). During adolescence, inverse frontoamygdala functional connectivity has been associated with greater amygdala habituation, suggesting a similar function of this circuitry by adolescence (Hare et al., Reference Hare, Tottenham, Galvan, Voss, Glover and Casey2008). Similarly, stronger inverse connectivity and medial PFC recruitment correspond to improved reappraisal success with age (McRae et al., Reference McRae, Gross, Weber, Robertson, Sokol-Hessner and Ray2012; Silvers et al., Reference Silvers, Weber, Wager and Ochsner2014). Given the role of the medial PFC in regulating amygdala reactivity, evidence of stronger inverse coupling and reduced amygdala reactivity with age may provide a neurobiological basis for developmental improvements in regulation of negative emotion and normative anxiety.

Although cortical–subcortical interactions support emotion regulation and deviations in related neurodevelopment are likely to contribute to the emergence of psychopathology in development, much remains unknown about the specific nature of these deviations for internalizing and externalizing disorders. One possibility is that impairments in specific circuitry might mediate core dimensional processes, such as emotion regulation, which confer risk for psychopathology in general rather than for specific disorders (Buckholtz & Meyer-Lindenberg, Reference Buckholtz and Meyer-Lindenberg2012), yet there appears to be some distinction between the primary neural mechanisms of regulation in internalizing versus externalizing disorders (see Beauchaine, Reference Beauchaine2015a). In general, the regulation of internalizing symptoms like anxiety appears to occur through top-down lateral and medial prefrontal control of amygdala reactivity (e.g., Lieberman et al., Reference Lieberman, Eisenberger, Crockett, Tom, Pfeifer and Way2007; Ochsner et al., Reference Ochsner, Bunge, Gross and Gabrieli2002), whereas the regulation of externalizing symptoms like impulsivity involves dorsolateral prefrontal and orbitofrontal modulation of striatal activity (e.g., Dalley, Mar, Economidou, & Robbins, Reference Dalley, Mar, Economidou and Robbins2008; Heatherton & Wagner, Reference Heatherton and Wagner2011).

Consistent with its role in the modulation of negative emotion, frontoamygdala circuitry has been strongly implicated in internalizing disorders (e.g., Rauch, Shin, & Wright, Reference Rauch, Shin and Wright2003). Neuroimaging involving the regulation of negative emotion in healthy adults has generally underscored the role of lateral regions of the PFC (e.g., ventrolateral and dorsolateral PFC), in addition to medial regions (e.g., ventromedial PFC) that are central to fear reduction processes such as extinction (e.g., Delgado, Nearing, LeDoux, & Phelps, Reference Delgado, Nearing, LeDoux and Phelps2008; Erk et al., Reference Erk, Mikschl, Stier, Ciaramidaro, Gapp and Weber2010; Goldin, McRae, Ramel, & Gross, Reference Goldin, McRae, Ramel and Gross2008; Lieberman et al., Reference Lieberman, Eisenberger, Crockett, Tom, Pfeifer and Way2007; Ochsner et al., Reference Ochsner, Bunge, Gross and Gabrieli2002; Phelps, Delgado, Nearing, & LeDoux, Reference Phelps, Delgado, Nearing and LeDoux2004; Wager, Davidson, Hughes, Lindquist, & Ochsner, Reference Wager, Davidson, Hughes, Lindquist and Ochsner2008). These same regions have been implicated in pediatric anxiety (e.g., De Bellis et al., Reference De Bellis, Casey, Dahl, Birmaher, Williamson and Thomas2000; Guyer et al., Reference Guyer, Monk, McClure-Tone, Nelson, Roberson-Nay and Adler2008; Monk et al., Reference Monk, Telzer, Mogg, Bradley, Mai and Louro2008; Roy et al., Reference Roy, Fudge, Kelly, Perry, Daniele and Carlisi2013) and depression (e.g., Gaffrey et al., Reference Gaffrey, Luby, Belden, Hirshberg, Volsch and Barch2011; Hulvershorn, Cullen, & Anand, Reference Hulvershorn, Cullen and Anand2011; Luking et al., Reference Luking, Repovs, Belden, Gaffrey, Botteron and Luby2011; Yang et al., Reference Yang, Simmons, Matthews, Tapert, Frank and Max2010). For example, youth with generalized anxiety disorder had weaker inverse connectivity between the ventrolateral PFC and the amygdala during a task involving masked angry faces, consistent with weaker regulatory control (Monk et al., Reference Monk, Telzer, Mogg, Bradley, Mai and Louro2008). However, few studies have examined frontoamygdala circuitry in pediatric psychiatric populations during tasks of emotion regulation, and much remains unknown about the specific associations between risk for emotion regulation difficulties and neural circuitry abnormalities prior to adulthood.

Moreover, disruptions in frontoamygdala neurodevelopment are not specific to internalizing disorders, because they have also been observed in children and adolescents with ADHD (e.g., Hulvershorn et al., Reference Hulvershorn, Mennes, Castellanos, Di Martino, Milham and Hummer2014; Posner et al., Reference Posner, Nagel, Maia, Mechling, Oh and Wang2011), bipolar disorder (e.g., Passarotti, Ellis, Wegbreit, Stevens, & Pavuluri, Reference Passarotti, Ellis, Wegbreit, Stevens and Pavuluri2012; Pfeifer, Welge, Strakowski, Adler, & Delbello, Reference Pfeifer, Welge, Strakowski, Adler and Delbello2008; Rich et al., Reference Rich, Fromm, Berghorst, Dickstein, Brotman and Pine2008), psychotic symptoms (e.g., Gee et al., Reference Gee, Karlsgodt, van Erp, Bearden, Lieberman and Belger2012; Wolf et al., Reference Wolf, Satterthwaite, Calkins, Ruparel, Elliott and Hopson2015), and callous–unemotional traits (e.g., Marsh et al., Reference Marsh, Finger, Mitchell, Reid, Sims and Kosson2008). While much of the research on frontoamygdala circuitry in adults with psychiatric disorders has focused on anxiety (reviewed in Rauch et al., Reference Rauch, Shin and Wright2003) and depression (e.g., Siegle, Thompson, Carter, Steinhauer, & Thase, Reference Siegle, Thompson, Carter, Steinhauer and Thase2007; reviewed in Heller, Reference Heller2016), altered frontoamygdala circuitry has also been observed in adults with bipolar disorder (e.g., Townsend et al., Reference Townsend, Torrisi, Lieberman, Sugar, Bookheimer and Altshuler2013; Wang et al., Reference Wang, Kalmar, He, Jackowski, Chepenik and Edmiston2009), schizophrenia (e.g., Anticevic, Van Snellenberg, & Barch, Reference Anticevic, Van Snellenberg and Barch2012; Fakra, Salgado-Pineda, Delaveau, Hariri, & Blin, Reference Fakra, Salgado-Pineda, Delaveau, Hariri and Blin2008; Taylor et al., Reference Taylor, Kang, Brege, Tso, Hosanagar and Johnson2012), borderline personality disorder (e.g., New et al., Reference New, Hazlett, Buchsbaum, Goodman, Mitelman and Newmark2007; Silbersweig et al., Reference Silbersweig, Clarkin, Goldstein, Kernberg, Tuescher and Levy2007), psychopathy (e.g., Blair, Reference Blair2008), and ADHD (e.g., Maier et al., Reference Maier, Szalkowski, Kamphausen, Feige, Perlov and Kalisch2014; Plichta et al., Reference Plichta, Vasic, Wolf, Lesch, Brummer and Jacob2009; Tajima-Pozo et al., Reference Tajima-Pozo, Yus, Ruiz-Manrique, Lewczuk, Arrazola and Montañes-Rada2016). Thus, future research is needed to better understand the nature of frontoamygdala changes and transdiagnostic risk associated with deficits in emotion regulation across many internalizing and externalizing disorders.

By contrast, frontostriatal circuitry is more strongly associated with regulatory difficulties in externalizing disorders. Disrupted interactions between frontal regions (e.g., dorsolateral PFC and orbitofrontal cortex) and the ventral striatum have been observed in adults with substance abuse disorders (e.g., Goldstein & Volkow, Reference Goldstein and Volkow2011; Koob & Le Moal, Reference Koob and Le Moal2001), ADHD (e.g., Casey et al., Reference Casey, Epstein, Buhle, Liston, Davidson and Tonev2007), and psychopathy (e.g., Glenn & Yang, Reference Glenn and Yang2012). Similarly, frontostriatal circuitry has been implicated in children and adolescents with externalizing disorders such as conduct disorder (e.g., De Brito et al., Reference De Brito, Mechelli, Wilke, Laurens, Jones and Barker2009), ADHD (e.g., Shaw et al., Reference Shaw, Malek, Watson, Sharp, Evans and Greenstein2012), ADHD (e.g., Plichta & Scheres, Reference Plichta and Scheres2014), and psychopathic traits (e.g., Blair, Reference Blair2013). However, just as frontoamygdala circuitry is not uniquely implicated in internalizing disorders, alterations in frontostriatal circuitry are not specific to externalizing disorders. For example, neuroimaging investigations of altered reward processing and positive affect in depression have led to an increased focus on frontostriatal interactions in depression (Heller, Reference Heller2016).

Thus, it remains unknown the extent to which the neurobiological disruptions underlying deficits in emotion regulation might be similar or distinct across internalizing and externalizing disorders. Consistent with the association between individual differences in normative anxiety and frontoamygdala connectivity (e.g., Hare et al., Reference Hare, Tottenham, Galvan, Voss, Glover and Casey2008; Pezawas et al., Reference Pezawas, Meyer-Lindenberg, Drabant, Verchinski, Munoz and Kolachana2005), frontoamygdala circuitry is disrupted in pathological anxiety (reviewed in Kim et al., Reference Kim, Loucks, Palmer, Brown, Solomon and Marchante2011). Specifically, weaker frontoamygdala connectivity, amygdala hyperreactivity, and prefrontal hypoactivity appear to play central roles in the pathophysiology of anxiety disorders (e.g., Shin et al., Reference Shin, Wright, Cannistraro, Wedig, McMullin and Martis2005; reviewed in Rauch et al., Reference Rauch, Shin and Wright2003). Findings on frontoamygdala alterations are more variable across other disorders, and it may be that the specific nature of emotion regulation and related neural mechanisms differ by symptom type. For example, emerging evidence suggests that, unlike in anxiety disorders, the core mechanism underlying depression may be a failure to upregulate and sustain positive emotion via frontostriatal circuitry (e.g., Heller et al., Reference Heller, Johnstone, Shackman, Light, Peterson and Kolden2009). Similarly, the nature of neural abnormalities in schizophrenia is markedly different from many other psychiatric disorders, because patients with schizophrenia exhibit hypoactive amygdala activation (Taylor et al., Reference Taylor, Kang, Brege, Tso, Hosanagar and Johnson2012). Thus, although similar circuitry is implicated, neurobiological alterations may relate to excessive or insufficient emotion depending on the form of psychopathology. Moreover, it may be that deficits in emotion regulation differentiate some patients from others with the same disorder (e.g., Hulvershorn et al., Reference Hulvershorn, Mennes, Castellanos, Di Martino, Milham and Hummer2014; Karalunas et al., Reference Karalunas, Fair, Musser, Aykes, Iyer and Nigg2014). Future research will be essential to delineating the commonalities and differences in the neurobiological substrates of emotion regulation and transdiagnostic risk across internalizing and externalizing disorders.

How important is it to assess and model information from multiple informants?

Much like symptom expression, emotion regulation capabilities too change over the course of the life span. Thus, a developmental psychopathology approach encompassing emotion regulation ought to take into account the larger context in which assessments are conducted (e.g., biology, culture, environment, and social processes). In this respect, one of the most robust findings in this field is that these variations in assessment produce substantial inconsistencies in findings regarding psychopathology and its risk and protective factors (Achenbach, Reference Achenbach2011; De Los Reyes, Reference De Los Reyes2013; De Los Reyes et al., Reference De Los Reyes, Augenstein, Wang, Thomas, Drabick and Burgers2015). Recently, we developed a theoretical model to account for these variations (De Los Reyes & Aldao, Reference De Los Reyes and Aldao2015; De Los Reyes, Thomas, et al., Reference De Los Reyes, Thomas, Goodman and Kundey2013).

Specifically, the operations triad model (OTM) accounts for variations among the outcomes of multiple-informant assessments by distinguishing three different kinds of possible outcomes from these assessments. In one instance, two or more informants (e.g., parents and teachers) provide corresponding reports of a target individual's mental health (e.g., child displays externalizing difficulties across reports), thus signaling that the target individual displays the assessed concern consistently across contexts observed by the informants (e.g., home and school). These assessments yield outcomes that reflect converging operations (see Garner et al., Reference Garner, Hake and Eriksen1956). Two other instances reflect circumstances in which informants provide inconsistent reports of a target's mental health (e.g., child displays externalizing difficulties based on teacher report and not parent report). When inconsistencies reflect diverging operations, the reports meaningfully point to variations in the assessed concerns across contexts (e.g., child displays difficulties at school and not home). When they reflect compensating operations, mundane, methodological factors exist to parsimoniously explain the inconsistencies (e.g., different item content across informants’ reports). Overall, the OTM presents a hypothesis-driven, theory-guided approach to making sense of consistencies and inconsistencies in the outcomes of multiple-informant mental health assessments.

Recent empirical work supports the OTM, and indicates that variations in measurement do not necessarily pose barriers to our basic understanding of psychopathology. Instead, these variations may inform our understanding of individual differences in mental health functioning (e.g., De Los Reyes, Thomas, et al., Reference De Los Reyes, Thomas, Goodman and Kundey2013). For example, in assessments of disruptive behavior among young children, clinical assessments often incorporate parent and teacher reports, and these reports commonly disagree in estimates of disruptive behavior. In recent work involving parent and teacher reports and independent observations of young children interacting with various adult authority figures (i.e., parents vs. unfamiliar clinical examiners), researchers profiled young children as to whether they behaved disruptively within controlled laboratory interactions with their parents, clinical examiners, neither of these interactions, or both of them (De Los Reyes et al., Reference De Los Reyes, Henry, Tolan and Wakschlag2009). Consistent with diverging operations, children who behaved disruptively with parents and not clinical examiners were also those children who were rated as disruptive by parents and not teachers. Further, children who behaved disruptively with clinical examiners and not parents were also those children who were rated as disruptive by teachers and not parents, and children who behaved disruptively with both clinical examiners and parents were rated as disruptive by teachers and parents. Effects consistent with the OTM have also been observed in assessments of adolescents (De Los Reyes, Alfano, Lau, Augenstein, & Borelli, Reference De Los Reyes, Alfano, Lau, Augenstein and Borelli2016) and adults (De Los Reyes, Bunnell, et al., Reference De Los Reyes, Bunnell and Beidel2013).

The work reviewed previously may inform longitudinal research on the role of deficits in emotion regulation and vulnerability to psychopathology. For example, we can posit that adolescents for whom multiple informants’ reports reflect converging operations in displays of high degrees of internalizing psychopathology are at particular risk for pervasive emotion regulation difficulties (e.g., parents, teachers, and peers), relative to adolescents for whom only one informant suggests such psychopathology (i.e., diverging operations that reflect context-specific emotion regulation difficulties). Later in adolescence and into emerging adulthood, the pervasive adolescents may manifest a particularly high risk for maladaptive strategies for coping with their mental health concerns, namely, risk-taking behaviors (e.g., substance use and risky sexual behavior), thus resulting in development of a comorbid internalizing–externalizing clinical presentation. Conversely, the context-specific adolescents might not manifest comorbid externalizing difficulties, but may nonetheless display emotion regulation strategies that maintain internalizing concerns persistently across development. With these examples, a combination of multiple-informant assessments of mental health concerns, and laboratory tasks designed to assess emotion regulation difficulties, might provide a particularly robust paradigm to assess the development of comorbidity in psychopathology, in ways similar to the cross-sectional mental health research described previously (e.g., De Los Reyes et al., Reference De Los Reyes, Henry, Tolan and Wakschlag2009, Reference De Los Reyes, Alfano, Lau, Augenstein and Borelli2016; De Los Reyes, Bunnell, et al., Reference De Los Reyes, Bunnell and Beidel2013). These paradigms and their implications for emotion regulation research, merit further study.