PFC pre- and perinatal development is highly sensitive to perturbations

During the prenatal period, developing cortical neurons are sensitive to a range of endogenous and exogenous factors, including prenatal drug exposures (e.g., cannabis, nicotine, and cocaine) and maternal stress, which have been found to impact cortical morphology (El Marroun et al., Reference El Marroun, Tiemeier, Franken, Jaddoe, van der Lugt and Verhulst2015; Weinstock, Reference Weinstock2008) and disrupt connectivity in frontal circuitry (Salzwedel et al., Reference Salzwedel, Grewen, Vachet, Gerig, Lin and Gao2015) during the postnatal period. In addition, the PFC undergoes a wave of neurogenesis and neural migration during late gestation, coinciding with peak perinatal vulnerabilities related to in utero complications, premature birth, and delivery complications (e.g., fetal distress and hypoxia).

The PFC's protracted developmental course provides a wide window for environmental factors to influence development

Like some other cortical gray matter regions, PFC structural development is characterized by an inverted U-shaped volume by age trajectory, such that volume generally increases during childhood, peaks in middle adolescence, and then declines into adulthood (Giedd, Reference Giedd2004; Lenroot et al., Reference Lenroot, Gogtay, Greenstein, Wells, Wallace and Clasen2007). Although research cannot precisely characterize the molecular basis of such changes in humans, the volumetric changes are hypothesized to reflect an initial overproduction of dendritic spines and proliferation of neural circuits, followed by the initiation and gradual escalation of synaptic pruning in middle adolescence that serves to enhance neural efficiency (Lenroot & Giedd, Reference Lenroot and Giedd2006). In contrast to gray matter, white matter development follows a more linear increase, during childhood and adolescence, which in conjunction with pruning processes increases the efficiency and stability of firing patterns in the PFC (Hagmann et al., Reference Hagmann, Sporns, Madan, Cammoun, Pienaar and Wedeen2010).

At the cellular level, when compared to other cortical regions, the PFC shows a pronounced plateau through childhood, followed by a gradual decrease in synapses during adolescence and young adulthood (Lenroot & Giedd, Reference Lenroot and Giedd2006). The PFC has one of the longest periods of developmental maturation of any brain region, and it is generally one of the last to complete neuronal/synaptic formation and elimination (Casey et al., Reference Casey, Jones, Levita, Libby, Pattwell and Ruberry2010; Elston, Oga, & Fujita, Reference Elston, Oga and Fujita2009; Luciana, Reference Luciana2013). It is generally assumed that brain regions that are more plastic over prolonged developmental periods are more susceptible to environmental factors (e.g., drugs, stress, social experiences, and hormones), which can alter the developmental course (Kolb & Gibb, Reference Kolb and Gibb2011). Thus, the protracted maturational course of the PFC broadens its window of vulnerability to environmental perturbations and enhances the likelihood it will play a role in deviations from the modal course of behavioral development.

The timing, rate, and slope of neuromaturational changes have significant implications for functioning

The pacing of PFC maturation appears to be particularly important to functioning, and volume changes are likely reflections of important neuromaturational processes (e.g., synaptic pruning and myelination). The timing of the onset and completion of these cellular processes appears to have measurable functional consequences. Longitudinal studies increasingly suggest that the developmental trajectory of the PFC, and most other brain regions, can be more predictive of outcome than its absolute volume and/or thickness. For example, in a cohort-sequential study design, adolescents with superior intellectual abilities exhibited greater peak thickness around puberty, followed by a greater cortical thinning into adulthood (Shaw et al., Reference Shaw, Greenstein, Lerch, Clasen, Lenroot and Gogtay2006). This is consistent with more recent research that has identified a positive relationship between the pace of prefrontal cortical thinning and cognitive control abilities (Tamnes et al., Reference Tamnes, Østby, Fjell, Westlye, Due-Tønnessen and Walhovd2010, Reference Tamnes, Walhovd, Grydeland, Holland, Østby and Dale2013), as well as emotion regulation capacities (Vijayakumar et al., Reference Vijayakumar, Whittle, Yücel, Dennison, Simmons and Allen2014). When these maturational processes are off pace, as indexed by morphological features by chronological age, adverse functional consequences are more likely. For example, when compared to controls, children with attention-deficit/hyperactivity disorder (ADHD) show a considerable lag in the neurodevelopment of the medial PFC (Shaw et al., Reference Shaw, Malek, Watson, Sharp, Evans and Greenstein2012), and among boys with conduct disorder the typical reductions in PFC gray matter are not observed (De Brito et al., Reference De Brito, Mechelli, Wilke, Laurens, Jones and Barker2009). Accordingly, interpretations of individual or group differences in PFC volume must be appropriately contextualized, because volume differences resulting from regressive changes (e.g., atrophy) and progressive changes (e.g., neuron proliferation and synaptogenesis) can have different functional implications depending on the developmental stage.

The developmental trajectory of the PFC in relation to other brain regions creates shifting balances in key circuits

Across brain regions, development occurs in a heterogeneous fashion, with regions varying in the onset and pace of development, and this has been posited to result in shifting balances in the relative influence of various regions and circuits on behavior. Even within the PFC, subregions differ in their maturational course, with gray matter volumes reaching adult levels earliest in the OFC, followed by the ventrolateral PFC, and the dlPFC (Walhovd, Tamnes, & Fjell, Reference Walhovd, Tamnes and Fjell2014). PFC maturation lags behind midbrain, limbic, and striatal regions, such that the disparity between the developmental trajectories among these regions is at its peak during adolescence (Mills, Goddings, Clasen, Giedd, & Blackmore, Reference Mills, Goddings, Clasen, Giedd and Blakemore2014). This “developmental mismatch” is hypothesized to have transient functional consequences, reflected in normative adolescent changes in mood and reward tendencies, as well as the expression or amplification of vulnerabilities for certain psychopathologies (Casey et al., Reference Casey, Jones and Hare2008; Van Leijenhorst et al., Reference Van Leijenhorst, Moor, de Macks, Rombouts, Westenberg and Crone2010). Thus, development of the PFC should always be contextualized within the whole brain, because its functional expression and maturational course is intimately tied to other brain regions.

Corticolimbic circuit

The corticolimbic circuit is composed of the amygdala, medial PFC, and lateral PFC, which act in concert to regulate affect, arousal, and vigilance to positive and negative valenced stimuli (Kim et al., Reference Kim, Loucks, Palmer, Brown, Solomon and Marchante2011). Neuroimaging studies using diffusion tensor imaging and functional magnetic resonance imaging (fMRI) have consistently documented both structural and functional connectivity between these regions consistent with reciprocal connections, in which the PFC frequently exerts top-down inhibitory control of amygdalar activity and reactivity (Kim et al., Reference Kim, Loucks, Palmer, Brown, Solomon and Marchante2011; Swartz & Monk, Reference Swartz, Monk, Anderson and Pine2014).

Developmental research findings suggest that maturation of the corticolimbic circuit underlies the changes in emotional reactivity and regulation observed throughout childhood and adolescence (e.g., Nelson, Lau, & Jarcho, Reference Nelson, Lau and Jarcho2014; Silvers et al., Reference Silvers, McRae, Gabrieli, Gross, Remy and Ochsner2012; Tottenham, Hare, & Casey, Reference Tottenham, Hare and Casey2011). Earlier in development, subcortical circuitry appears to dominate emotional responses, as evidenced by heightened magnitudes of activity in subcortical regions in response to emotional cues (e.g., Gee et al., Reference Gee, Humphreys, Flannery, Goff, Telzer and Shapiro2013). As individuals continue through development, this subcortical dominance is increasingly modulated by PFC regulatory engagement, which is associated with enhanced capacities to incorporate emotional information into appropriate action (Heller & Casey, Reference Heller and Casey2016). To date, only a few studies have looked at developmental changes in functional connectivity within the corticolimbic circuit. In a cross-sectional fMRI study spanning ages 4 to 22, Gee et al. (Reference Gee, Humphreys, Flannery, Goff, Telzer and Shapiro2013) found a shift from positive (i.e., cofiring) to negative prefrontal–amygdalar connectivity during the transition to adolescence. This connectivity shift was inferred from functional connectivity, which suggests an increasingly reciprocal relationship between the PFC and amygdala, consistent with more efficient top-down modulation. Further, the shift in functional connectivity was paralleled by a steady decline in amygdala reactivity.

There is some evidence from tract-tracing studies conducted on rodents that amygdala to PFC projections emerge earlier than PFC to amygdala projections (Bouwmeester, Smits, & Van Ree, Reference Bouwmeester, Smits and Van Ree2002), and that this early bottom-up signaling may facilitate connectivity and augment development of connections between these regions. This is consistent with findings at the whole brain level that increased positive connectivity early in life may be necessary to establish a connection that later evolves into a regulatory circuit (Gee et al., Reference Gee, Humphreys, Flannery, Goff, Telzer and Shapiro2013). This also raises the important possibility that early developmental patterns, which are often viewed as “immature” (e.g., immature PFC regulatory capacities), may actually be instrumental in establishing healthy neural functioning later in development. Disruptions or differences in subcortical reactivity early in life will likely have significant consequences for the later developing PFC and emotion regulation processes. As circuitry becomes fine-tuned, repeated activation patterns in this network as a result of specific early experiences are likely to become stronger and more automatic.

Variations in corticolimbic circuit functioning appear to be associated with individual differences in threat detection, affective responsiveness, emotion regulation, and fear extinction (Swarz & Monk, Reference Swartz, Monk, Anderson and Pine2014), as well as the magnitude, duration, and habituation of emotional experiences (Heller & Casey, Reference Heller and Casey2016). At the more extreme ranges, these differences can subserve symptoms of irritability, aggression (Coccaro, McCloskey, Fitzgerald, & Phan, Reference Coccaro, McCloskey, Fitzgerald and Phan2007; Buckholtz & Meyer-Lindenberg, Reference Buckholtz and Meyer-Lindenberg2008; Hoptman et al., Reference Hoptman, D'Angelo, Catalano, Mauro, Shehzad and Kelly2010), anxiety (e.g., Monk et al., Reference Monk, Telzer, Mogg, Bradley, Mai and Louro2008), and emotional lability (see Churchwell, Morris, Heurtelou, & Kesner, Reference Churchwell, Morris, Heurtelou and Kesner2009).

Frontostriatal circuit

The frontostriatal circuit is composed of the lateral PFC, ventromedial PFC, OFC, amygdala, and striatum. Across various tasks, it has been consistently shown that interactions between the PFC and striatum are central to accomplishing goal-directed behavioral regulation (Belleine, Delgado, & Hikoshaka, Reference Belleine, Delgado and Hikosaka2007), suppressing impulses (Miller & Cohen, Reference Miller and Cohen2001), regulating of positive affect (Heller et al., Reference Heller, Johnstone, Shackman, Light, Peterson and Kolden2009), and integrating reward association with behavioral outputs (Pasupathy & Miller, Reference Pasupathy and Miller2005). Current neuroanatomical models emphasize important functional distinctions between the dorsal and ventral portions of the circuit. The dorsal portion, which includes reciprocal connections between the dorsal striatum (caudate nucleus and putamen) and OFC, mediates habit-and stimulus-response learning (Packard & Knowlton, Reference Packard and Knowlton2002), supporting self-regulatory functions, movement, and behavioral control. Meanwhile, the ventral portion, which includes connections between the ventral striatum (nucleus accumbens and olfactory tubercle) and prefrontal regions (anterior cingulate cortex and vmPFC) mediates reward, drive, and motivation, supporting reward-processing functions (Haber & Knutson, Reference Haber and Knutson2010). As a result, different components of the frontostriatal network play key roles in self-regulation, reward processing, and positive affect regulation.

Findings from fMRI studies indicate that functional changes within the dorsal frontostriatal circuits in healthy youth underlie age-related improvements in the capacity for self-regulation, cognitive control, and inhibitory control (Marsh, Maia, & Peterson, Reference Marsh, Maia and Peterson2009; Marsh et al., Reference Marsh, Zhu, Schultz, Quackenbush, Royal and Skudlarski2006; Rubia et al., Reference Rubia, Smith, Woolley, Nosarti, Heyman and Taylor2006). When compared to adults, children recruit distinct and often larger, more diffuse frontostriatal regions when performing cognitive control tasks (Casey, Tottenham, & Fossella, Reference Casey, Tottenham and Fossella2002; Luna et al., Reference Luna, Thulborn, Munoz, Merriam, Garver and Minshew2001; Bunge, Dudukovic, Thomason, Vaidya, & Gabrieli, Reference Bunge, Dudukovic, Thomason, Vaidya and Gabrieli2002), which is hypothesized to reflect greater effort to regulate subcortical structures. Developmental fMRI studies of reward processing suggest that adolescents differ from adults in the balance between PFC activity versus the motivation/emotional responses of risk and reward based in the striatum and amygdala (Casey et al., Reference Casey, Jones and Hare2008; Ernst & Fudge, Reference Ernst and Fudge2009). For example, fMRI studies of incentive processing have found that adolescents typically show greater ventral striatal activation and weaker prefrontal activation in comparison to adults (Bjork, Smith, Chen, & Hommer, Reference Bjork, Smith, Chen and Hommer2010; Cauffman et al., Reference Cauffman, Shulman, Steinberg, Claus, Banich and Graham2010; Padmanabhan, Geier, Ordaz, Teslovich, & Luna, Reference Padmanabhan, Geier, Ordaz, Teslovich and Luna2011; van Leijenhorst et al., Reference Van Leijenhorst, Moor, de Macks, Rombouts, Westenberg and Crone2010). It appears that the resulting imbalance between reward load and cognitive control, and the limited ability of PFC resources to sufficiently regulate reward responses, during adolescence may lead to a transient increase in reward-related behaviors during adolescence. This normative increase in striatal reactivity to reward during adolescence is likely to be particularly problematic in individuals with compromised PFC resources. In addition to top-down regulation of striatal activity, human and animal work also suggests that the PFC can activate the striatum and modulate reward-related firing in the striatum in a top-down manner (Heller et al., Reference Heller, Johnstone, Shackman, Light, Peterson and Kolden2009). Thus, the PFC can both up- and downregulate activity in the striatum.

Variations in frontostriatal development can manifest as individual differences in behavioral flexibility, reinforcement learning, decision making, approach motivation, hedonic responses, positive affect regulation, and salience attribution (Kehagia, Murray, & Robbins, Reference Kehagia, Murray and Robbins2010). At the ends of the distributions of these dimensions, individual differences can manifest as symptoms of impulsivity, compulsivity, and hedonic capacity (e g., anhedonia) that likely underpin atypical approach and avoidance behaviors observed across internalizing and externalizing disorders. For example, it has been hypothesized that reduced reward salience provokes a compensatory increase in reward-seeking behavior (Robbins & Everitt, Reference Robbins and Everitt1999), which in combination with reduced executive functioning abilities results in impulsivity (Marsh et al., Reference Marsh, Maia and Peterson2009). In contrast, anhedonia and depressive symptoms have been associated with patterns of reduced striatal activation in combination with heightened mPFC activation (Forbes & Dahl, Reference Forbes and Dahl2012). This pattern may reflect blunted reward sensitivity (Martin-Soelch, Reference Martin-Soelch2009) and/or mPFC overregulation that dampens sustained engagement of positive affect and reward responsivity (Feldman, Joorman, & Johnson, Reference Feldman, Joormann and Johnson2008), as well as increasing engagement in excessive self-processing (e.g., negative, self-focused rumination; Forbes & Dahl, Reference Forbes and Dahl2012). Meanwhile, symptoms of compulsivity, commonly observed in obsessive–compulsive disorder, substance use, and eating disorders, have been associated with hyperactivity of dorsal frontostriatal circuits in which perseverative behaviors and habits are maintained by maladaptive stimulus-response learning (Marsh et al., Reference Marsh, Maia and Peterson2009).

Of course, dysfunction within these circuits does not necessarily stem solely from the PFC. Variations in the activity of limbic and striatal regions may similarly compromise the developing circuit and overwhelm PFC resources. However, independent of the initial locus of disruption, the functional consequences at the circuit level can be similar. Perturbing connectivity in either the corticolimbic or the frontostriatal circuits would likely disrupt a range of mood, reward, and regulatory processes that are not disorder specific, thus increasing susceptibility to multiple clinical disorders.

DA

DA is a key neuromodulator of the PFC and provides both inhibitory and excitatory influences that help facilitate neural information processing and exchange (Puig, Rose, Schmidt, & Freund, Reference Puig, Rose, Schmidt and Freund2014). Research in both humans and animals has consistently found that DA levels in the PFC influence cognitive performance on a range of tasks, including working memory, planning, and attention (Arnsten, Reference Arnsten2011; Puig et al., Reference Puig, Rose, Schmidt and Freund2014). These findings generally support an inverted-U relation, such that low and high DA levels impair performance (Vijayraghavan, Wang, Birnbaum, Williams, & Arnsten, Reference Vijayraghavan, Wang, Birnbaum, Williams and Arnsten2007), although research increasingly suggests that dose-response patterns vary by cognitive domain/task (Floresco, Reference Floresco2013). Among the psychological disorders in which altered DA is strongly implicated (e.g., psychotic disorders and ADHD), cognitive deficits are typically a feature. In addition to influencing cognitive performance, DA has modulatory influences on synaptic plasticity in the PFC, and is thought to influence important adolescent neuromaturational processes such as prefrontal pruning (Goto, Yang, & Otani, Reference Goto, Yang and Otani2010). Given the important integrative role the PFC plays with respect to afferents from subcortical and other cortical regions, changes in the synaptic plasticity mechanisms in the PFC have the potential for widespread effects on multimodal networks throughout the brain. Finally, due to unique characteristics of the DA system in the PFC, the region is reliant on secondary mechanisms (e.g., catechol-O-methyltransferase [COMT]) to inactive extracellular DA (Diamond, Reference Diamond2007). As a result, genetic polymorphisms and other factors affecting DA levels are hypothesized to have greater impacts on PFC function, relative to other brain regions (Diamond, Reference Diamond2007; Malhotra et al., Reference Malhotra, Kestler, Mazzanti, Bates, Goldberg and Goldman2002).

Serotonin

Current research provides evidence for serotonin influences on the cellular processes supporting early brain development (Daubert & Condron, Reference Daubert and Condron2010), cognitive functioning (Cowen & Sherwood, Reference Cowen and Sherwood2013; Schmitt, Wingen, Ramaekers, Evers, & Riedel, Reference Schmitt, Wingen, Ramaekers, Evers and Riedel2006), corticolimbic activity (Hariri, Drabant, & Weinberger, Reference Hariri, Drabant and Weinberger2006; Volman et al., Reference Volman, Verhagen, den Ouden, Fernández, Rijpkema and Franke2013), and synaptic plasticity (Lesch & Waider, Reference Lesch and Waider2012). For example, serotonin depletion has also been linked to impairments in a number of cognitive functions, including reversal learning (Clarke et al., Reference Clarke, Walker, Crofts, Dalley, Robbins and Roberts2005), attentional and cognitive flexibility (Lapiz-Bluhm, Soto-Piña, Hensler, & Morilak, Reference Lapiz-Bluhm, Soto-Piña, Hensler and Morilak2009), working memory (Cano-Colino, Almeida, Gomez-Cabrero, Artigas, & Compte, Reference Cano-Colino, Almeida, Gomez-Cabrero, Artigas and Compte2014; Enge, Fleischhauer, Lesch, Reif, & Strobel, Reference Enge, Fleischhauer, Lesch, Reif and Strobel2011), and behavioral flexibility (Roberts, Reference Roberts2011), as well as impulsivity (Dalley & Rosier, Reference Dalley and Roiser2012). The serotonin system has also been associated with variation in neural reactivity within corticolimbic circuitry to emotionally salient cues. For example, lower levels of serotonin (as indexed by either 5-HT genotype or 5-hydroxytryptamine receptor 2A [5-HTR2A] density) have been associated with weaker mPFC modulation of amygdalar responses to emotional cues (Fisher et al., Reference Fisher, Meltzer, Price, Coleman, Ziolko and Becker2009; Volman et al., Reference Volman, Verhagen, den Ouden, Fernández, Rijpkema and Franke2013). There is also accumulating evidence that, in addition to an immediate agnoism, certain classes of serotonin receptors (e.g., 5-HTR2A) also lead to activation of pathways involved in induction of synaptic plasticity, particularly through stimulation of brain-derived neurotrophic factor (BDNF). While more work is needed to elucidate this relationship, there is preliminary evidence that serotonin activity in the PFC may influence structural remodeling related to synaptic plasticity (Aznar & Klein, Reference Aznar and Klein2013). More work is needed to clarify the role of serotonin in the PFC, particularly given that serotonin signaling is altered in many forms of psychopathology and many psychotropic drugs target serotonin receptors in the PFC.

Glutamate

The medial PFC (mPFC) is rich in glutamatergic innervations and has been identified as a key site for the mediation of the behavioral effects of glutamate (Diamond, Reference Diamond, Stuss and Knight2002). In both human and animal studies, mPFC-task related activity is altered when glutamatergic agents are administered (e.g., Jackson, Homayoung, & Moghaddam, Reference Jackson, Homayoun and Moghaddam2004). Findings suggest that within the mPFC, glutamate receptors, both AMPA and NMDA, are necessary for set shifting, and that NMDA receptor hypofunction impairs the capacity to modify existing memory stores and diminished the inhibitory control of PFC neurons (Homayoun & Moghaddam, Reference Homayoun and Moghaddam2007). Glutamate concentration in the mPFC has also been found to predict rest-state, cortical–subcortical functional connectivity (Duncan et al., Reference Duncan, Wiebking, Tiret, Marjanska, Hayes and Lyttleton2013), highlighting an important role for glutamate in cortical–subcortical communication.

Neurohormones (adrenal and gonadal)

Research has also documented an influence of adrenal and gonadal hormones, such as cortisol, testosterone, and estrogen, on the expression of genes that govern PFC development and function (Brouwer et al., Reference Brouwer, Koenis, Schnack, van Baal, van Soelen and Boomsma2015; Peper & Dahl, Reference Peper and Dahl2013; Trotman et al., Reference Trotman, Holtzman, Ryan, Shapiro, MacDonald and Goulding2013). Through epigenetic processes, hormones can trigger transcription and protein synthesis of genes involved in neurodevelopment, which in turn influence neuronal structure and function, including neuronal growth, neurotransmitter synthesis and reuptake, and receptor density and sensitivity (Zhang & Ho, Reference Zhang and Ho2011). Thus, changes within these neurohormone systems during certain developmental periods, such as in utero and during puberty, are likely to have significant influences on genetically governed neurodevelopmental processes in the PFC, among other regions. There is also significant interaction and potentiation among neurohormones and neurotransmitter systems (e.g., cortisol and DA, estrogen and DA), which can lead to cascading biological effects (see Sinclair, Purves-Tyson, Allen, & Weickert, Reference Sinclair, Purves-Tyson, Allen and Weickert2014).

Parenting

Parenting styles provide the foundational social context in which children develop and refine regulatory abilities through observational learning, modeling, and social referencing (Morris, Silk, Steinberg, Myers, & Robinson, Reference Morris, Silk, Steinberg, Myers and Robinson2007). Among parenting factors, there is a large literature documenting the impact that parenting styles, such as emotional responsiveness, warmth, acceptance, criticism, validation, and control, have on a child's emotional development, internalizing problems, aggression, and depression (Bernier, Carlson, & Whipple, Reference Bernier, Carlson and Whipple2010; Bilsky et al., Reference Bilsky, Cole, Dukewich, Martin, Sinclair and Tran2013; Buckholdt, Parra, & Jobe-Shields, Reference Buckholdt, Parra and Jobe-Shields2014; Crowell et al., Reference Crowell, Baucom, McCauley, Potapova, Fitelson and Barth2013). These associations are likely to be complex and characterized by parent-driven effects, child-driven or evocative effects, and recursive processes.

Researchers have speculated that maladaptive parenting styles (e.g., invalidating, controlling, or unresponsive) may increase risk for psychopathology through compromised development of emotion regulation and attentional control in early childhood (Bernier et al., Reference Bernier, Carlson and Whipple2010). More specifically, maladaptive parenting strategies tend to limit opportunities for effective coping, increase dependence on adults, promote coercive interactions, and reinforce avoidance and aggressive regulatory strategies (Crowell et al., Reference Crowell, Baucom, McCauley, Potapova, Fitelson and Barth2013; Eisenberg et al., Reference Eisenberg, Losoya, Fabes, Guthrie, Reiser and Murphy2001; Suveg, Shaffer, Morelen, & Thomassin, Reference Suveg, Shaffer, Morelen and Thomassin2011). In contrast, multiple studies document the association between warm-sensitive and validating parenting and emerging youth emotion regulation (see review by Calkins & Marcovitch, Reference Calkins, Marcovitch, Calkins and Bell2010). These parenting styles provide external soothing and help children cope with frightening or frustrating situations in a way that models self-regulation strategies, which children internalize to develop effective emotion regulation strategies.

At a mechanistic level, parenting effects may operate, in part, through emotion regulation and EF, both of which implicate PFC involvement. This is supported by research that has found that warm-sensitive parenting is associated with child attention set-shifting skills in toddlers (Berneir et al., Reference Bernier, Carlson and Whipple2010) and executive functioning skills in preschool (Hughes, Ensor, Wilson, & Graham, Reference Hughes, Ensor, Wilson and Graham2009). In addition, harsh parenting, low maternal warmth, and maternal negative affect measured in at-risk groups have all been associated with atypical brain structure (e.g., larger regional volumes) and functional engagement (e.g., increased activation) of the PFC, among other brain regions (i.e., amygdala and striatum; Tan et al., Reference Tan, Lee, Dahl, Nelson, Stroud and Siegle2014; Taylor, Eisenberger, Saxbe, Lehman, & Lieberman, Reference Taylor, Eisenberger, Saxbe, Lehman and Lieberman2006; Whittle et al., Reference Whittle, Simmons, Dennison, Vijayakumar, Schwartz and Yap2014; Yap, Allen, & Ladouceur, Reference Yap, Allen and Ladouceur2008). For example, Taylor et al. (Reference Taylor, Eisenberger, Saxbe, Lehman and Lieberman2006) found that children from “risky” families (as indexed from a questionnaire assessing multiple aspects of parenting style and family environment) have different patterns of neural activation to tasks assessing reactivity and response to emotional stimuli, compared to children from “nonrisky” families. In the labeling task, the at-risk group was distinguished by increased amygdala activation and a strong, positive correlation between amygdala activation and right ventrolateral PFC, suggesting that these youth did not recruit the right ventrolateral effectively for regulating amygdala responses to threatening stimuli.

While most research has focused on how parenting affects the child's regulatory development, bidirectional influences are almost certainly at play. Children who are more emotionally dysregulated or inattentive may be more likely to elicit maladaptive parenting responses (e.g., criticism or control) as a result of increasing parental frustration and/or perceived need for external management. For example, Eisenberg, Spinrad, and Eggum (Reference Eisenberg, Spinrad and Eggum2010) found that when toddlers were less self-regulated at 30 months, their mothers used more directive parenting strategies. Through mutual reinforcement, a child with biological vulnerabilities for negative emotional reactivity and poor attention may pull for harsh and invalidating parenting, which in turn fosters coercive parent–child reciprocities during early childhood. As a result, ineffective parenting practices may compound biological vulnerabilities and potentially lead to the development of more serious deficits. While outside of the scope of this review, it is important to consider that parenting styles are often associated with a range of larger social contextual factors (e.g., ethnicity, community, and social economic status), and thus may be a mediator in relationships commonly documented between larger social factors and psychopathology.

Peers

Peer relationships provide another foundational context for development of important skills related to learning social norms, developing, social cognitive abilities, and perspective taking (Rubin & Ross, Reference Rubin and Ross1982). Multiple studies have found that childhood experiences of peer rejection influence the pathways of externalizing and internalizing symptoms (Arseneault, Bowes, & Shakoor, Reference Arseneault, Bowes and Shakoor2010; Brendgen et al., Reference Brendgen, Boivin, Vitaro, Bukowski, Dionne and Tremblay2008; Flanagan, Erath, & Bierman, Reference Flanagan, Erath and Bierman2008; Trentacosta & Shaw, Reference Trentacosta and Shaw2009), including depression and anxiety (Pedersen, Vitaro, Barker, & Borge, Reference Pedersen, Vitaro, Barker and Borge2007), antisocial behavior (Laird, Jordan, Dodge, Pettit, & Bates, Reference Laird, Jordan, Dodge, Pettit and Bates2001), and substance abuse (Fite, Colder, Lochman, & Wells, Reference Fite, Colder, Lochman and Wells2007).

Peer rejection and negative peer experiences deprive youth of important social experiences, opportunities to develop and refine social skills, and sources of support (Rubin, Bukowski, & Parker, Reference Rubin, Bukowski, Parker, Damon and Lerner2006). These experiences also lend themselves to feelings of resentment and cognitive beliefs related to exclusion and lack of belonging (Ladd, Reference Ladd2006). Over time these processes can reinforce restriction of social activities, increased relationship problems, and distress. In addition, decreased opportunities to join prosocial peer groups increases the risk that rejected peers will engage in more deviant behavior and peer groups (Fite et al., Reference Fite, Colder, Lochman and Wells2007). As a result, compounded social skills deficits, restricted social activities, and relationship problems can translate into more global impairments in functioning and psychiatric symptomatology over time.

Individual differences in executive functioning, emotion regulation, and reward processes also likely play a role in exacerbating or ameliorating children's risk for poor peer relationships. Problems managing behavior, inhibiting aggression and impulses, and regulating emotions have significant implications for a child's status in a peer group. A number of recent studies have demonstrated an association between self-regulation and peer relations during both childhood and adolescence (Farley & Kim-Spoon, Reference Farley and Kim-Spoon2014; Holmes, Kim-Spoon, & Deater-Deckard, Reference Holmes, Kim-Spoon and Deater-Deckard2015; Stenseng, Belsky, Skalicka, & Wichstrøm, Reference Stenseng, Belsky, Skalicka and Wichstrøm2015). From a relatively young age, children learn to avoid social interactions with peers who are more extreme in their emotional displays (e.g., anger or fearfulness; Ladd, Reference Ladd2006). Consequently, youth with more extreme disruptive or inhibited tendencies experience greater peer rejection and fewer positive relationships (Hasenfratz, Benish-Weisman, Steinberg, & Knafo-Noam, Reference Hasenfratz, Benish-Weisman, Steinberg and Knafo-Noam2015). Consistent with this, children with attention difficulties have been found to have fewer friendships (Hoza et al., Reference Hoza, Gerdes, Mrug, Hinshaw, Bukowski and Gold2005) and less stable relationships than typically developing peers (Blachman & Hinshaw, Reference Blachman and Hinshaw2002). Meanwhile, withdrawn children have been found to avoid initiating peer interactions (Rubin, Coplan, & Bowker, Reference Rubin, Coplan and Bowker2009).

The interaction between behavioral propensities endowed by PFC dysfunction (e.g., poor emotion regulation, hyperactivity, poor attention, and behavioral inhibition) and peer rejection can amplify existing vulnerabilities and propagate new ones. For example, although externalizing symptoms typically elicit initial peer rejection (Mrug, Hoza, & Gerdes, Reference Mrug, Hoza and Gerdes2001), peer rejection further aggravates the poor outcomes these children experience. Mrug and McCay (Reference Mrug and McCay2013) found that in ADHD youth followed after treatment, peer rejection predicted delinquency, anxiety, and global impairment at 6 years and global impairment at 8 years after baseline. Similarly, in anxious/withdrawn children, initial avoidance facilitates a pattern of restricted social engagement and peer exclusion that subsequently reinforces decreased engagement and motivation for future interactions and amplifies internalizing distress.

Peer experiences become more potent during adolescence. During this time it becomes more common to spend time with peers, place greater value on peers' approval, and to become more concerned about peers' acceptance (Albert, Chein, & Steinberg, Reference Albert, Chein and Steinberg2013; Brechwald & Prinstein, Reference Brechwald and Prinstein2011). This shift is driven, in part, by a suite of normative developmental changes in brain circuitry, neurotransmitters, motivational systems, sexual development, and cultural influences that increase the affective salience of socially related events (Casey & Jones, Reference Casey and Jones2010; Steinberg, Reference Steinberg2010). The combination of these biological changes and a more challenging social landscape increases opportunities and sensitivity to both positive and negative peer experiences. Social support also offers an important buffer to experiences of stress that accompany adolescence, and can provide corrective feedback to depressogenic cognitions that are likely to increase at this age (Panzarella, Alloy, & Whitehouse, Reference Panzarella, Alloy and Whitehouse2006). In adolescence, peer rejection has been associated with social avoidance, and higher levels of anxiety, depression, suicidality, excessive risk taking, and substance abuse (Rubin et al., Reference Rubin, Bukowski, Parker, Damon and Lerner2006).

Cannabis

Accumulating evidence suggests that the PFC is particularly vulnerable to cannabis during the early adolescent period, when cortical circuits are being refined (Rubino & Parolaro, Reference Rubino and Parolaro2008). The PFC is enriched with the CB1 subtype of cannabinoid receptor, which increases in density during adolescence. The repetitive use of cannabis during adolescence is hypothesized to blunt the normal developmental increase in CB1 and consequently disrupt PFC maturational processes (Hirvonen et al., Reference Hirvonen, Goodwin, Li, Terry, Zoghbi and Morse2012; Rubino & Parolaro, Reference Rubino and Parolaro2008). Multiple studies have found that early cannabis users (defined as before 16–18 years) show executive dysfunction (Gruber, Sagar, Dahlgren, Racine, & Lukas, Reference Gruber, Sagar, Dahlgren, Racine and Lukas2012; Hanson, Winward, et al., Reference Hanson, Winward, Schweinsburg, Medina, Brown and Tapert2010; Lisdahl & Price, Reference Lisdahl and Price2012), structural alterations (Churchwell, Lopez-Larson, & Yurgelun-Todd, Reference Churchwell, Lopez-Larson and Yurgelun-Todd2010; Gruber, Sagar, Dahlgren, Racine, & Lukas, Reference Gruber, Silveri, Dahlgren and Yurgelun-Todd2011; Lopez-Larson et al., Reference Lopez-Larson, Bogorodzki, Rogowska, McGlade, King and Terry2011; Shollenbarger, Pice, Wieser, & Lisdahl, Reference Shollenbarger, Price, Wieser and Lisdahl2015), and abnormal brain activation (Eldreth, Matochik, Cadet, & Bolla, Reference Eldreth, Matochik, Cadet and Bolla2004; Jager, Block, Luijten, & Ramsey, Reference Jager, Block, Luijten and Ramsey2010) in the PFC relative to noncannabis users. The functional consequences associated with cannabis use (e.g., sleep deprivation and lower grades) may further potentiate the negative consequences of use and confer additional environmental risk.

Alcohol use

Neuroimaging studies of chronic excessive alcohol use consistently reveal structural and functional impairments in the PFC as well as various other brain regions (Bava & Tapert, Reference Bava and Tapert2010; Squeglia, Jacobus, & Tapert, Reference Squeglia, Jacobus and Tapert2014). A smaller number of studies focused on more limited alcohol exposure in adolescents and young adulthood find that alcohol use is associated with reduced prefrontal volumes (De Bellis et al., Reference De Bellis, Narasimhan, Thatcher, Keshavan, Soloff and Clark2005; Medina et al., Reference Medina, McQueeny, Nagel, Hanson, Schweinsburg and Tapert2008), alterations in white matter integrity (De Bellis et al., Reference De Bellis, Van Voorhees, Hooper, Gibler, Nelson and Hege2008; Jacobus et al., Reference Jacobus, McQueeny, Bava, Schweinsburg, Frank and Yang2009; McQueeny et al., Reference McQueeny, Schweinsburg, Schweinsburg, Jacobus, Bava and Frank2009), and alterations in PFC response during certain tasks (Caldwell et al., Reference Caldwell, Schweinsburg, Nagel, Barlett, Brown and Tapert2005; Schweinsburg, McQueeny, Nagel, Eyler, & Tapert, Reference Schweinsburg, McQueeny, Nagel, Eyler and Tapert2010). The effects on white matter integrity are observed even in adolescents who engage in infrequent or subtle binge drinking (Squeglia et al., Reference Squeglia, Jacobus and Tapert2009). However, the direction of the deviation from healthy controls is debated, with inconsistent findings potentially being the result of differences in methodology.

In terms of neuropsychological correlates of adolescent alcohol use, the most consistent findings appear to be impaired executive functioning, especially working memory and attention (Hartley, Elsabagh, & File, Reference Hartley, Elsabagh and File2004; Townshend & Duka, Reference Townshend and Duka2005; Brown & Tapert, Reference Brown and Tapert2004), which appears to remain stably impaired even if abstinence occurs in younger adulthood (Lyvers, Reference Lyvers2000). This suggests that adolescent alcohol use is associated with irreversible changes to the PFC, likely due to the vulnerable nature of protracted PFC maturation. Both alcohol and cannabis have the potential to put youth at heightened risk of acute harm (e.g., accidents) and increased exposure to stressful experiences (e.g., physical fights, unwanted and/or risky sexual encounters, and legal problems).

Taken together, it is clear that PFC mechanisms operate across multiple levels of analysis and are involved in developmental processes that likely give rise to psychopathology. Individual differences in PFC functions that fall on the extreme ends of the spectrum not only underlie maladaptive behaviors but also increase vulnerability to further environmental risk through evocative processes. These transactions lead to an unfortunate cycle in which PFC dysfunction may be further compounded.

Developmental moderators

There are a number of normal developmental processes that likely facilitate age-related differences in PFC mechanisms of symptom expression. For example, clinical phenomena are likely to vary with cognitive maturation, particularly in regard to behavioral responses to negative emotionality and anxious distress. For example, the enhanced cognitive capacities (e.g., rumination, abstract thinking, hypothetical thinking, and future expectances; Steinberg, Reference Steinberg2005) that accompany PFC maturation, and become operative during the adolescent transition, will change how youth experience, cope with, and express negative emotions. While deficits in emotion regulation and executive functioning are more likely to manifest behaviorally (e.g., aggression, tantrum, or defiance) in response to negative mood states in early childhood, these same underlying deficits may assume a more internalizing presentation as development proceeds and new cognitive capacities put youth at greater risk for mood disorders. This is hypothesized to be a contributing factor to the increase of depression rates during adolescence. The cognitive abilities to generate future expectations, engage in hypothetical thinking, and anticipate distal rewards may provide the basis for hopelessness, which is not typically present in younger children.

The potency and availability of certain risk factors also change across development. For example, peer rejection and poor social competence will likely have a stronger effect during the adolescent transition as interpersonal relationships become highly valued and biological systems become more attuned to and motivated by social experiences (Albert et al., Reference Albert, Chein and Steinberg2013). It is possible that the effects of negative peer relationships during adolescence may confer additional internalizing distress on an individual whose presentation had previously been primarily externalizing in nature. With age, opportunities to join deviant peer groups, engage in drug use, and experience significant relationship stress (i.e., friendship turmoil and romantic breakups) also accrue. The advent of intimate and romantic relationships, as well as the increasingly complexity of social groups in early adolescence, introduces new opportunities for rejection and loss. Pubertal development and sexual exploration, particularly among females, similarly affords new opportunities for negative and high-risk sexual experiences.

The predominant expression of symptoms at different ages is often informed by normative development changes and challenges. For example, nonclinical childhood anxiety is typically predominated by fears of separation and specific objects/situations (e.g., animals) in early childhood, by fears about danger and death in middle childhood, and by fears about social competence and performance in adolescence (Weems, Reference Weems2008). This shift parallels the age-related patterns frequently observed across clinical anxiety diagnoses. Thus, developmental changes or “comorbidities” in anxiety diagnoses likely represent continuity of a common vulnerability that is shaped by normal maturational trends, rather than the development of discrete disorders. Finally, transient developmental changes in the biological stress and reward systems certainly affect all adolescents, but for those with preexisting vulnerabilities, this period of instability may be the time in which tendencies toward internalizing, externalizing, and psychotic symptoms become full-blown disorders. For these vulnerable youth, normative adolescent changes may further instantiate long-term alterations in PFC-related functions that would otherwise be temporary and more time limited.

Contextual moderators

In addition to developmental moderators, there are also important contextual moderators that will influence how, if, and to what degree symptoms are expressed. The characteristics and concerns presented by the current context will likely bias symptoms expression. For example, an unstable, threatening environment, or one filled with unpredictable stress, will be more likely to exacerbate and maintain anxiety symptoms by generating emotions of fear and presenting concerns related to safety and consistency. Meanwhile, a social context that is characterized by recent experiences of loss, rejection, and failure may be more likely to manifest as depression, by generating instances of sad affect and concerns of worth and belonging. Thus, while the underlying biological mechanisms may be the same, the emotion and concerns presented by the context will modulate the expression of vulnerability at that time point (Nolen-Hoeksema & Watkins, Reference Nolen-Hoeksema and Watkins2011). With regard to comorbidity, it is possible that contexts that include both characteristics will be more likely to give rise to a comorbid symptom presentation. However, social contexts are far from static and can shift across development, presenting an individual with new concerns. For example, the developmental progression that is often observed between and adolescents and depression (Rutter et al., Reference Rutter, Kim-Cohen and Maughan2006) may result from the introduction of new experiences with rejection, loss, and failure that typically accompanies the normal adolescent experience.

Case example

Brandon was born at 36 weeks gestational age, weighing 5 lb, 4 oz. Medical records indicate a complication with the umbilical cord during delivery that likely resulted in mild hypoxia. However, Brandon was otherwise in good condition and his neonatal course was uneventful. Brandon's mother described him as a “fussy and temperamental” infant and noted difficulties self-soothing and sleeping. As a toddler, Brandon's mother reported that while he seemed to be more emotionally reactive and more easily frustrated than her other children, there were no significant concerns.

At the start of first grade Brandon began to show behavioral problems both at school and at home. His teachers reported that he was defiant in the classroom, often stubbornly refusing to read aloud or engage in classroom activities. When corrected or redirected by teachers, he quickly became sullen and angry, and had been observed to break his own pencils and tear up his paper. While he was generally a diligent student and seemed motivated to perform, he tended to have difficulty organizing his work and materials, became easily upset by careless mistakes, found it difficult to stay on task, and often fell behind other students in his pace of work. This continued through the second grade, although Bandon was performing above grade level and interacted well with peers.

Brandon's mother also reported difficulties managing his behavior at home. At first, she attributed this to Brandon's failure to adjust to his parents' recent divorce and subsequent stressors in the home. However, as his behavior escalated and persisted, she became exasperated. While she expected to have “normal” arguments about doing chores and homework, she was surprised that she and Brandon fought even about doing fun things (e.g., going to friends birthday party or a trip to an amusement park). She reported that she had never seen such a “wimpy child” and was frustrated at his inability to “toughen up.” She also noted that Brandon seemed unable to just “let things go” and seemed to get stuck on the littlest things. For example, Brandon liked to spend hours in his room building elaborate Lego structures. However, this activity would inevitably end with frustration and toys being thrown. When his mother came to check on him, she reported that he would scream about how he hated Legos, the world, and even her. Homework was also a battle. Brandon often refused to do his work and would break pencils to irritate his mother. This led to verbal arguments that escalated until both parties gave up and Brandon stormed off and slammed doors.

Brandon had always been a friendly and sociable child, but he began to experience difficulties with peers in the third grade. He was often demanding and insisted that other children do things his way. When they did not, he would either sulk or act out aggressively. He was quick to retaliate when he felt teased or embarrassed. As a result, many of his peers began to distance themselves and avoid him. During most of elementary and middle school, Brandon continued to exhibit aggression, defiance, irritable mood, and worry.

During the transition to high school Brandon became less interested in activities he used to enjoy (e.g., basketball and building model airplanes). His mother reported that he seemed to be constantly “down” and “completely unmotivated.” While previously a motivated, albeit difficult student, his academic performance significantly dropped, and he reported it “was not worth his time.” Brandon told his guidance counselor that the future seemed pretty hopeless for him, and it was better not to waste his time trying when he was likely to be disappointed. Early in his freshman year, Brandon became a peripheral member of a deviant peer group. Brandon and his new friends preferred to spend their time smoking marijuana after school and binge drinking on the weekends in various parks. They also regularly stole things from the local shops and vandalized properties. Brandon's mother reported that Brandon sometimes made suicidal threats in response to punishment, but she was not sure if he was serious or just trying to “get a rise out of her.”

This case example aptly illustrates how heterotypic comorbidity can arise from developmental interactional processes. There are, of course, multiple ways in which this particular symptom presentation could have unfolded. The explanation most consistent with our heuristic suggests that shared PFC-related vulnerabilities underlie both oppositional and anxious syndrome presentations, via contributions of PFC-mediated functions. As described in earlier sections, functional alterations in executive functioning, emotion regulation, and reward processing at the extreme ends of the continua, or when exacerbated by environmental factors, can subserve a host of symptoms that fall across traditional diagnostic categories (e.g., worry, rumination, irritable mood, low positive affect, impulsivity, and aggression). While these individual differences do not begin as psychiatric symptoms, they are exacerbated by experiences of chronic stress (e.g., parent divorce or financial stress), negative parental interactions, poor peer relationships, reinforced escape coping, and academic struggles across development. While DSM criteria indicate both ODD and anxiety diagnoses here, this clinical picture is best understood by considering the common vulnerability processes underlying these symptoms.

While this review has focused on the assumption that comorbidities arise from shared vulnerability processes, a developmental cascade account is also possible and not mutually exclusive. In this conceptualization, the functional consequences of one syndrome (e.g., ODD) are hypothesized to potentiate the onset of another syndrome (e.g., anxiety). Previous work suggests that negative developmental outcomes from one disorder can influence other functioning domains that subsequently potentiate risk another disorder (Masten & Cicchetti, Reference Masten and Cicchetti2010). In our case example, the child's early externalizing behaviors begin to garner functional consequences in elementary school, a time in which increased demands for regulation are likely to unmask underlying weaknesses. These behavioral tendencies trigger evocative effects in both the home and the school environment. Negative interactions between the child and his mother reinforce his emotional dysregulation and promote avoidance coping methods, in place of more adaptive coping styles. This socialization effect carries over to his peer interactions, which suffer as a result of his behavioral tendencies. In this case, while the initial impetus is externalizing behaviors, the functional consequences that occur across other domains (e.g., school, peers, and parent), increase the risk for serious internalizing distress (Moilanen, Shaw, & Maxwell, Reference Moilanen, Shaw and Maxwell2010; Obradović, Burt, & Masten, Reference Obradović, Burt and Masten2009; Patterson, DeBaryshe, & Ramsey, Reference Patterson, DeBaryshe, Ramsey, Gauvain and Cole1989).

The developmental trajectory outlined by this case example also illustrates the phenomena of heterotypic continuity (i.e., sequential development of different disorders across the life span). The changes in the child's symptom presentation, while still likely derived from the same underlying vulnerability processes, are further compounded and modulated by developmental factors. For example, increased cognitive capacities (e.g., future thinking and rumination) and accruing experiences of social and academic disappointment introduce new experiences of hopelessness. In addition, peer difficulties early in life increase the risk for deviant peer affiliation in adolescence. The combination of negative peer influences (which have high saliency during adolescence), as well as increased freedom and autonomy associated with adolescence, provides new opportunities to engage in criminal acts and substance use. Although substance use may be driven by deviant peer influence or desires to self-medication (likely both), the usage similarly compounds his PFC deficits (see previous sections) and exacerbates symptoms. While this youth's diagnostic history could include at least four different DSM disorders (depending on the date of assessment), this case is best understood by considering the longitudinal transactions that have occurred across development.

Focus on transdiagnostic processes

As the field moves toward conceptualizing disorders based on transdiagnostic neural circuits and processes, rather than collections of symptoms that exceed a clinical threshold, clearer targets for intervention are likely to be revealed. It will be important for these findings to inform treatment development, such that interventions can be tailored to remediating compromised transdiagnostic processes as observed from symptoms. This is likely to not only increase treatment efficacy by targeting the underlying process (rather than the behavioral expression) but also simultaneously address comorbidity, because treatments would be targeting shared vulnerability processes.

Given the importance of the PFC and its related functions addressed in this review, interventions developed to capitalize on skill acquisition in executive functioning and emotion regulation may be particularly fruitful. Diamond and Lee (Reference Diamond and Lee2011) have reviewed a number of approaches with some evidence for improving EFs in early school years, although further research is needed to address their generalizability and efficacy. In addition, inclusion of techniques specifically aimed at increasing emotional awareness and understanding, effectively managing expressions of sadness and of anger, and reducing engagement in rumination and other avoidant coping methods are important targets of youth interventions.

Target contextual factors

The plasticity and sensitivity of the PFC to environmental factors throughout much of development highlights important treatment avenues that should be a critical component of current preventative interventions, regardless of disorder. As stated repeatedly throughout this review, biological vulnerability is not inevitably expressed in maladaptive behavior or clinical disorder. Thus, there is an incredible opportunity to modify environmental risk processes that may prevent the progression of psychopathology. Given the available research on common environmental risk factors, positive parenting training, social skills training, and general stress management should complement current preventative interventions. From a prevention standpoint, these types of interventions have the potential to mitigate the very processes that often exacerbate existing difficulties and “socialize” further dysfunction. Intervening early, and at multiple levels, has the potential to alter the course of development and provide protective buffers for individuals with biological/preexisting vulnerabilities. These same environmental interventions will be similarly effective for youth with internalizing, externalizing, and comorbid presentations, given their increased risk of eliciting negative parenting interactions, reduced social competences, and greater stress. While medication and therapy can target certain levels of analysis effectively (i.e., neurochemical), attention to contextual environmental factors is necessary, though sometimes neglected.

Consider development

It is imperative that interventions take development into account, in terms of both treatment initiation and design. The relative plasticity of younger brains provides opportunities to confer protective long-term changes in regions such as the PFC, particularly when supported by enriched environments. As reviewed above, maladaptive behaviors and PFC dysfunction becomes increasingly entrenched over time, making it more difficult to intervene and potentially impossible to reverse. Thus, it is critical that intervention efforts are initiated earlier in the developmental trajectory. The selection of appropriate treatment approaches should also consider cognitive maturation. While cognitive behavioral therapy is often the gold standard, it is important to consider whether cognitive approaches are commensurate with the youth's current cognitive abilities and whether cognitive vulnerabilities plays a prominent role in the generation of symptoms at that point in development. It is possible that targeting more reward-related processes via behavioral activation approaches in younger children may be more appropriate. Of course, these decisions will likely vary on a case-by-case basis, depending on the constellation of symptoms and history of the child.

Anticipate continuity

Finally, clinical providers should consider and anticipate comorbidity and continuity. While our understanding of the mechanisms underlying these phenomena is currently limited, descriptive reports of diagnostic patterns warn us of what is likely to lie ahead. Thus, it is important to address risks for disorders that are not yet apparent, but may lie further down the continuum, as well as domains of functioning that are likely to be negatively impacted. For example, while treatment of ADHD in young children may focus on reducing hyperactivity and impulsivity via stimulant medication and behavioral parent training, it is important to consider how risk for compromised peer relationships may unfold and potentiate risk for more serious externalizing symptoms, as well as internalizing problems, down the road. Thus, treatments that address social skills early on, before problems occur, may confer a buffer against worsening psychopathology. The same anticipation of continuity could take place for anxious youth, many of whom will go one to experience depression in adolescence (Rutter et al., Reference Rutter, Kim-Cohen and Maughan2006).