ANS

The ANS, which comprises the sympathetic and the parasympathetic branches, regulates several vital functions on a moment-to-moment basis. Although there is a long tradition of research of the ANS in children, adolescents, and adults with antisocial and aggressive behavior (for a meta-analysis, see Lorber, Reference Lorber2004), poor electrodermal fear conditioning as a risk factor for later aggressive behavior and criminality has only recently been demonstrated in young children (Gao et al., Reference Gao, Raine, Venables, Dawson and Mednick2010a, Reference Gao, Raine, Venables, Dawson and Mednick2010b).

In a prospective study, fear conditioning using electrodermal responsivity was assessed in children at ages 3, 4, 5, 6, and 8. It was shown that poor fear conditioning from ages 3 to 8 years is associated with aggression at age 8 (Gao et al., Reference Gao, Raine, Venables, Dawson and Mednick2010a). Furthermore, it appeared that poor fear conditioning at age 3 predisposes an individual to crime at age 23 (Gao et al., Reference Gao, Raine, Venables, Dawson and Mednick2010b).

Besides this important series of studies that demonstrate poor fear conditioning in young children as a risk factor for developing aggressive and criminal behavior, psychopathy-prone adolescent boys relative to healthy controls have also been found to display reduced electrodermal activity in anticipation of and in response to an aversive stimulus. However, no differences were found between antisocial nonpsychopathic boys and antisocial psychopathy-prone boys. Thus, the antisocial component of psychopathy may be associated with skin conductance hyporesponsivity (Fung et al., Reference Fung, Raine, Loeber, Lynam, Steinhauer and Venables2005). In another study, children and adolescents with CD comorbid with ADHD showed low skin conductance responses to aversive stimuli as well as to (positive and negative) emotional stimuli and to neutral pictures, compared to children and adolescents with ADHD only and healthy controls (Herpertz et al., Reference Herpertz, Mueller, Qunaibi, Lichterveld and Konrad2005). The authors concluded that this general autonomic hyporeactivity may reflect a deficit in associative information-processing systems that normally produce adaptive cognitive–emotional reactions.

Basal skin conductance activity has also been considered to be a marker of punishment sensitivity, although it is much less specific than skin conductance reactivity to aversive stimuli (for a critique, see Fowles, Reference Fowles1980). In Lorber's (Reference Lorber2004) meta-analysis, it was found that children with conduct problems demonstrate lower basal skin conductance activity than controls (Cohen d = −0.30). Consistent, lower basal skin conductance levels were found in preschool children with aggressive behavior as compared to nonaggressive preschoolers (Posthumus, Böcker, Raaijmakers, van Engeland, & Matthys, Reference Posthumus, Böcker, Raaijmakers, van Engeland and Matthys2009) and in school-aged children with DBDs half of whom also had comorbid ADHD as compared to healthy controls (van Goozen, Matthys, Cohen-Kettenis, Buitelaar, & van Engeland, Reference van Goozen, Matthys, Cohen-Kettenis, Buitelaar and van Engeland2000). In addition, in a follow-up study of 7- to 12-year-old children treated for DBDs, lower basal skin conductance activity was a predictor of externalizing problems and of maintenance of DBDs in adolescence (van Bokhoven, Matthys, van Goozen, & van Engeland, Reference van Bokhoven, Matthys, van Goozen and van Engeland2005).

HPA axis

In addition to the ANS, responsiveness to aversive stimuli and stress is mediated by a neural circuit network comprising the HPA axis and the amygdala (LeDoux, Reference LeDoux2002). Cortisol secretion by the adrenal cortex is controlled by adrenocorticotropic hormone released from the pituitary, which is regulated by corticotrophin-releasing hormone from the hypothalamus. Corticotrophin-releasing hormone is released in response to stress and subsequent activation of the amygdala and prefrontal cortex. Reduced cortisol reactivity to stress has been found in children with DBDs, many of whom had comorbid ADHD (van Goozen et al., Reference van Goozen, Matthys, Cohen-Kettenis, Gispen-de Wied, Wiegant and van Engeland1998, Reference van Goozen, Matthys, Cohen-Kettenis, Buitelaar and van Engeland2000). Low cortisol responsivity during stress appears to be specific to DBDs. In a study that examined children with DBDs, children with ADHD, and healthy controls, only children with DBDs showed a blunted cortisol response (Snoek, van Goozen, Matthys, Buitelaar, & van Engeland, Reference van Goozen, Snoek, Matthys, van Rossum and van Engeland2004).

Amygdala

The amygdala is involved in the allocation of subjective value to stimuli in the internal and external environment (Balleine & Killcross, Reference Balleine and Killcross2006; Cardinal, Parkinson, Hall, & Everitt, Reference Cardinal, Parkinson, Hall and Everitt2002; Phelps & LeDoux, Reference Phelps and LeDoux2005). The amygdala has been widely implicated in learning and expressing the association of certain undesirable behaviors with punishment, as shown by studies on Pavlovian fear conditioning and passive avoidance learning (Phelps & LeDoux, Reference Phelps and LeDoux2005). Impaired functioning of the amygdala associated with decreased aversive stimulus-reinforcement associations is thought to be characteristic of psychopathic individuals (Blair, Reference Blair2007).

A number of studies on structure and function of the amygdala have been performed in children and adolescents with CD or conduct problems with or without psychopathic characteristics. In adolescents with early-onset CD, most of whom were comorbid with ADHD, reduced gray matter volumes were found in a variety of brain regions, including the amygdala, relative to healthy controls. Regression analyses indicated that CD symptoms were primarily correlated with gray matter reductions in limbic brain structures such as the amygdala and prefrontal cortex (Huebner et al., Reference Huebner, Vloet, Marx, Konrad, Fink and Herpertz2008). In addition, reduced gray matter volumes in the left amygdala have been found in adolescents with CD relative to healthy controls (Sterzer, Stadler, Poustka, & Kleinschmidt, Reference Sterzer, Stadler, Poustka and Kleinschmidt2007). In contrast, one study failed to find structural deviances of the amygdala in boys with CU traits (De Brito et al., Reference De Brito, Mechelli, Wilke, Laurens, Jones and Barker2009). Although the boys in this study had conduct problems, the presence of ODD or CD was not assessed. Thus, the negative findings could have been due to the presence of a less severe form of psychopathology in these boys.

Findings from a functional magnetic resonance imaging (fMRI) study suggest left amygdala hyporeactivity to negative emotional stimuli in children and adolescents aged 9 to 15 years with CD. However, it should be noted that the majority of participants was also diagnosed with ADHD and the amygdala hyporeactivity was only found in those subjects with low anxiety levels (Sterzer, Stadler, Krebs, Kleinschmidt, & Poustka, Reference Sterzer, Stadler, Krebs, Kleinschmidt and Poustka2005). Likewise, in an fMRI study, DBD children and adolescents (aged 10–17 years) with CU traits, many of whom also had comorbid ADHD, were found to have reduced amygdala responsiveness during the presentation of fearful facial expressions in comparison to healthy controls and youth with ADHD. It is interesting that this study's functional connectivity analyses demonstrated greater correlations between the amygdala and ventromedial prefrontal cortex in healthy controls and youth with ADHD relative to those with DBDs and CU traits (Marsh et al., Reference Marsh, Finger, Mitchell, Reid, Sims and Kosson2008). In another fMRI study, boys with conduct problems and elevated levels of CU traits who also had ADHD symptoms showed less right amygdala activity to fearful faces compared with healthy controls; these differences remained after controlling for ADHD symptoms (Jones, Laurens, Herba, & Viding, Reference Jones, Laurens, Herba and Viding2009). Thus, these studies reveal evidence of deficits in amygdala function in children and adolescents with DBDs or conduct problems with or without psychopathic or CU traits.

In contrast, enhanced activity of the amygdala has also been observed. CD adolescents, most of whom also had comorbid ADHD relative to healthy controls, showed enhanced left amygdala activation in response to negative pictures as compared to neutral pictures, an effect that was not observed in a patient control group with ADHD only (Herpertz et al., Reference Herpertz, Huebner, Marx, Vloet, Fink and Stoecker2008). Likewise, when perceiving others in pain, enhanced activation in the left amygdala was observed in adolescents with CD, most of whom were comorbid with ADHD, compared with healthy controls. It is important that amygdala activation was correlated with adolescents’ sadism scores. It may be that enhanced activation of the amygdala in CD adolescents elicited by viewing pain in others reflects enjoyment and excitement as the amygdala is involved in not only the processing of negative affect but also the processing of positive affect, as was found in a study with a small sample size (Decety, Michalska, Akitsuki, & Lahey, Reference Decety, Michalska, Akitsuki and Lahey2009).

Some studies with indirect measures of amygdala function are also relevant to mention here. Children with psychopathic characteristics but without DBDs showed selective impairments in the recognition of sad and fearful facial expressions (Blair, Colledge, Murray, et al., Reference Blair, Colledge, Murray and Mitchell2001). Likewise, adolescents with many CU traits but without DBDs showed reduced attention to other people's eyes during a facial emotion task relative to adolescents with few CU traits, thus accounting for their problems with fear recognition (Dadds, El Masry, Wimalaweera, & Guastella, Reference Dadds, El Masry, Wimalaweera and Guastella2008). Several studies have also been conducted using the startle reflex. In one study it was shown that children with DBDs, the majority of whom also had comorbid ADHD relative to healthy controls, had a blunted response to auditory stimuli that normally elicit a startle reflex. Furthermore, there was a negative correlation between delinquency in the children with DBDs and their startle responses while viewing unpleasant pictures (van Goozen, Snoek, et al., Reference Snoek, van Goozen, Matthys and van Engeland2004). In a second study, attenuation of the eye-blink startle reflex was found both in youth with early-onset CD and in youth with adolescent-onset CD relative to healthy controls. ADHD symptoms did not affect the startle reflex (Fairchild, van Goozen, Stollery, & Goodyer, Reference Fairchild, van Goozen, Stollery and Goodyer2008).

Neurotransmitters

On the neurochemical level, studies examining serotonergic (5-HT) and noradrenergic neurotransmission in the central nervous system are particularly relevant. Studies have revealed that 5-HT neurotransmission has, among other functions, been implicated in the sensitivity to punishment and aversive signals (Cools, Roberts, & Robbins, Reference Cools, Roberts and Robbins2008). Acute tryptophan depletion, thought to result in decreased 5-HT functioning, has been shown to selectively inhibit the recognition of fearful facial expressions (Harmer, Rogers, Timbridge, Cowen, & Goodwin, Reference Harmer, Rogers, Timbridge, Cowen and Goodwin2003). The effect of tryptophan depletion on fear recognition was later shown to occur only in carriers of the short allele of the promotor region (i.e., the section that regulates the transcription process) of the 5-HT transporter (5-HTT) gene (Marsh et al., Reference Marsh, Finger, Buzas, Soliman, Richell and Vythilingham2006) and to depend on individual threat sensitivity (Cools et al., Reference Cools, Calder, Lawrence, Clark, Bullmore and Robbins2005). This may also explain why a subsequent study failed to find an effect of acute tryptophan depletion on fear recognition (aan het Rot, Coupland, Boivin, Benkelfat, & Young, Reference aan het Rot, Coupland, Boivin, Benkelfat and Young2010).

5-HT has specifically been associated with aggression (Nelson & Trainer, Reference Nelson and Trainer2007; Siever, Reference Siever2008). Genetic studies of the 5-HT system in individuals with aggressive behavior have demonstrated an association between a polymorphism in the promoter region (i.e., the section that regulates the transcription process) of the 5-HTT gene and aggression in two studies with children, that is in a general population sample (Haberstick, Smolen, & Hewitt, Reference Haberstick, Smolen and Hewitt2006) and in a clinical sample of children with a DBD the majority of whom had comorbid ADHD (Beitchman et al., Reference Beitchman, Baldassarra, Mik, De Luca, King and Bender2006). This polymorphism has also been found in one study with adolescents with CD (Sakai et al., Reference Sakai, Young, Stallings, Timberlake, Smolen and Steler2006). However, 5-HTT polymorphisms have also been associated with numerous other forms of psychopathology including depression that may relate more generally to susceptibility to negative affect (Caspi, Hariri, Holmes, Uher, & Moffitt, Reference Caspi, Hariri, Holmes, Uher and Moffitt2010).

Various measures can be used to assess 5-HT activity. These include measurement of the principal 5-HT metabolite, 5-hydroindoleacetic acid (5-HIAA), in the cerebrospinal fluid or in plasma; measurements of 5-HT in whole blood; pharmacological challenge studies using a drug or a dietary manipulation that is known to affect central nervous system 5-HT functioning; and measurements of 5-HT receptor activity in blood platelets. In general, studies on neurotransmitters in ODD and CD have been confounded by ADHD comorbidity. Furthermore, studies have often focused on correlations between measures of neurotransmitter levels and aggression or conduct problem measures. We here pay specific, but not exclusive, attention to studies that compare DBD children with healthy controls and refer to the review by van Goozen et al. (Reference van Goozen, Fairchild, Snoek and Harold2007) for additional information. According to a meta-analysis in adults, there is an inverse relationship between cerebrospinal fluid 5-HT and antisocial behavior (Cohen effect size, d = −0.45; Moore, Scarpa, & Raine, Reference Moore, Scarpa and Raine2002). An inverse relationship between 5-HIAA and aggressive behavior has also been found in the cerebrospinal fluid of children and adolescents with DBDs and/or ADHD (Kruesi et al., Reference Kruesi, Rapaport, Hamburger, Hibbs, Potter and Lenane1990) and in the plasma of children with DBDs with or without ADHD (van Goozen, Matthys, Cohen-Kettenis, Westenberg, & van Engeland, Reference van Goozen, Matthys, Cohen-Kettenis, Westenberg and van Engeland1999). In the latter study, no differences were found in plasma 5-HIAA between DBD subgroups with or without ADHD. However, the results from studies investigating whole-blood 5-HT in children and adolescents with CD, most of whom were comorbid with ADHD, appear contradictory: no difference with healthy controls was found in one study (Rogeness, Hernandez, Macedo, & Mitchell, Reference Rogeness, Hernandez, Macedo and Mitchell1982) whereas in another study a positive relation was found between whole-blood 5-HT and aggression (Unis et al., Reference Unis, Cook, Vincent, Gjerde, Perry and Mason1997).

Challenge studies with fenfluramine in children with ADHD and aggressive behavior, using the prolactin response as an indirect measure of brain 5-HT functioning, have shown mixed results (Halperin et al., Reference Halperin, Sharma, Siever, Schwartz, Matier and Wornell1994, Reference Halperin, Newcorn, Schwartz, Sharma, Siever and Koda1997). No clear relationship between prolactin response and aggression in children with or without ADHD was found (Schultz et al., Reference Schultz, Newcorn, McKay, Himelstein, Koda and Siever2001). However, a longitudinal study demonstrated that lower prolactin responsivity to fenfluramine (indicating lower 5-HT functioning) in 9-year-old children with ADHD predicts the emergence of antisocial personality disorder in early adulthood (Flory, Newcorn, Miller, Harty, & Halperin, Reference Flory, Newcorn, Miller, Harty and Halperin2007). Furthermore, a challenge study with sumatriptan, a selective 5-HT_1B/1D receptor agonist, demonstrated that the growth hormone response was enhanced in children with ODD. There were no differences on baseline and peak growth hormone levels between ODD subgroups with and without ADHD comorbidity (Snoek et al., Reference Snoek, van Goozen, Matthys, Sigling, Koppeschaar and Westenberg2002). These findings suggest that the sensitivity of 5-HT_1B/1D receptors is increased in ODD. Because 5-HT_1B/1D receptors can function as autoreceptors providing negative feedback on 5-HT release mechanisms (Hoyer, Hannon, & Martin, Reference Hoyer, Hannon and Martin2002), their hypersensitivity is likely to contribute to lower 5-HT activity in ODD.

Platelet membranes that show similarities to pre- and postsynaptic membranes of 5-HT neurons in the central nervous system have been used to estimate central 5-HT functioning. An inverse correlation has been shown between platelet markers of 5-HT function and aggression in children and adolescents with CD, the majority of whom were also diagnosed with comorbid ADHD (Birmaher et al., Reference Birmaher, Stanley, Greenhill, Twomey, Gavrilescu and Rabinovich1990). However, no difference in 5-HT function was found between children and adolescents with DBDs and/or ADHD and controls (Stoff et al., Reference Stoff, Ieni, Friedman, Bridger, Pollock and Vitiello1991).

Taken together, there is support for an inverse relationship between 5-HT measures and aggressive behavior in children and adolescents with ODD and CD, although this relationship is less clear than in adults. The relationship between altered 5-HT function and ODD or CD is supported by reports that risperidon and lithium, which both affect 5-HT function, reduce disruptive behavior in the DBDs (Findling et al., Reference Findling, McNamara, Branicky, Schluchter, Lemon and Blumer2000; Malone, Delaney, Luebbert, Cater, & Campbell, Reference Malone, Delaney, Luebbert, Cater and Campbell2000). 5-HT neurotransmission has been implicated in sensitivity to punishment and aversive signals (Cools et al., Reference Cools, Roberts and Robbins2008). Hence, the reduction in 5-HT function that accompanies ODD and CD may contribute to the altered impact that punishment and aversive learning have on behavior in ODD and CD. Paradoxically, though, acute depletion of 5-HT appears to enhance the neural and behavioral responsiveness to punishment (Cools et al., Reference Cools, Roberts and Robbins2008). Therefore, if altered function of 5-HT directly contributes to the behavioral deficits in ODD and CD, it is likely to be the consequence of long-lasting reductions in 5-HT function that result in a series of adaptations in this system that contribute to the reduced punishment sensitivity in ODD and CD.

Noradrenergic neurotransmission may also play an important role in behavioral arousal associated with punishment (Berridge & Waterhouse, Reference Berridge and Waterhouse2003). If signals associated with punishment do not lead to the noradrenergically driven increase of attention and change in emotional state, these signals become less meaningful. Levels of the metabolite 3-methoxy-4-hydroxyphenylglycol in the cerebrospinal fluid, as well as dopamine-β-hydroxylase (the enzyme that converts dopamine into noradrenaline) in plasma, have been used as an indirect way to measure central noradrenergic activity. Kruesi et al. (Reference Kruesi, Rapaport, Hamburger, Hibbs, Potter and Lenane1990) found a negative correlation between 3-methoxy-4-hydroxyphenylglycol in cerebrospinal fluid and lifetime history of aggression in children and adolescents with DBDs or ADHD. Similarly, low plasma dopamine-β-hydroxylase has been shown in children and adolescents with CD (Rogeness, Javors, Maas, & Macedo, Reference Rogeness, Javors, Maas and Macedo1990), although this finding was not replicated in a study examining adolescents with CD, many of whom also had comorbid ADHD (Pliszka, Rogeness, Renner, Sherman, & Broussard, Reference Pliszka, Rogeness, Renner, Sherman and Broussard1988). In sum, there is some evidence to suggest decreased noradrenergic functioning in the DBDs.

Interim summary of punishment processing

Studies on electrodermal fear conditioning, cortisol reactivity to stress, serotonergic and noradrenergic neurotransmission, and, although less consistent, the amygdala function indicate that reduced sensitivity to aversive cues and punishment plays a role in the development of ODD and CD. Specifically, learning to refrain from inappropriate behaviors based on aversive conditioning, that is, making the association between inappropriate behavior and punishment, seems to play a role in the development of the ODD and CD from early childhood on. According to Blair (Reference Blair2004), this dysfunctional moral socialization is specifically associated with instrumental antisocial behavior. There is, however, no research in children and adolescents that has demonstrated the specificity of the relation between impaired sensitivity to aversive/punishment cues and instrumental/proactive antisocial behavior or aggression.

ANS

The pre-ejection period (PEP) of the heart, which is an index of sympathetic nervous system activity, is considered to be a peripheral marker of reward sensitivity (Beauchaine, Reference Beachaine2001). PEP nonreactivity to monetary incentives has been shown in preschoolers with ODD and ADHD relative to controls (Crowell et al., Reference Crowell, Beauchaine, Gatzke-Kopp, Sylvers, Mead and Chipman-Chacon2006), in male school children and adolescents with DBDs (Beauchaine, Gatzke-Kopp, & Mead, Reference Beauchaine, Gatzke-Kopp and Mead2007), and in aggressive boys with conduct problems (Beauchaine, Hong, & Marsh, Reference Beauchaine, Hong and Marsh2008). These results point toward reduced reward sensitivity in boys with conduct problems.

In addition, heart rate has also been linked to reward processing (Fowles, Reference Fowles1980), whereby low resting heart rate may reflect reduced reactivity to rewarding stimuli associated with an unpleasant state. In support of this interpretation, preschool boys who chose to watch videotapes of intense anger had lower basal heart rates than boys who chose to watch videotapes of mild anger. Thus, boys with a lower heart rate level might have sought a higher level of stimulation to experience a pleasant emotional state (El-Sheikh, Ballard, & Cummings, Reference El-Sheikh, Ballard and Cummings1994). A longitudinal general population study found that sensation seeking in boys at age 13.5 and 16 mediated the relation between low resting heart rate at age 11 and rule breaking at age 16 (Sijtsema et al., Reference Sijtsema, Veenstra, Lindenberg, van Roon, Verhulst and Ormel2010). This relation was not mediated by behavioral disinhibition. This study therefore shows that sensation seeking mediates the relationship between heart rate and rule breaking only in adolescence but not in preadolescence. The study also supports stimulation-seeking theory, which states that rule breaking in adolescence serves to alleviate the unpleasant state of stimulus hyporeactivity associated with low resting heart rate, to experience a pleasant level of emotional stimulation.

School-age children with DBDs with and without ADHD have been found to have a lower resting heart rate than healthy controls. There were no differences in resting heart rate between subgroups with or without ADHD comorbidity (van Goozen et al., Reference van Goozen, Matthys, Cohen-Kettenis, Buitelaar and van Engeland2000). Moreover, over the years many studies have demonstrated lower resting heart rate in children and adolescents with antisocial behavior. In a meta-analysis by Ortiz and Raine (Reference Ortiz and Raine2004), the effect size (Cohen d) for resting heart rate in children and adolescents with antisocial behavior was −0.44. Lorber (Reference Lorber2004) has subsequently conducted a meta-analysis of heart rate effects in groups of children with more specifically defined forms of antisocial behavior. The effect size of resting heart rate in children with aggressive behavior was −0.51, whereas in children with conduct problems it was −0.34.

Amygdala

Although the amygdala has been widely implicated in negative emotions such as fear, an increasing amount of evidence indicates that the amygdala is associated with valence detection rather than just fear and punishment (Sander, Grafman, & Zalla, Reference Sander, Grafman and Zalla2003). Furthermore, it is well established that the amygdala is involved in positive emotions and appetitive learning as well (Balleine & Killcross, Reference Balleine and Killcross2006; Cardinal et al., Reference Cardinal, Parkinson, Hall and Everitt2002). Therefore, a dysfunctional amygdala may likely result in disturbances in positive emotions and stimulus-reinforcement learning. Thus, reduced gray matter volumes in the amygdala in adolescents with CD discussed above (Huebner et al., Reference Huebner, Vloet, Marx, Konrad, Fink and Herpertz2008; Sterzer et al., Reference Sterzer, Stadler, Poustka and Kleinschmidt2007) may also result in altered reward processing.

Cerebral cortex

Cues and actions associated with reward are processed in the orbitofrontal cortex (Rolls, Reference Rolls2004). Blair (Reference Blair2004) has argued that orbitofrontal cortex dysfunctioning is involved in the modulation of reactive aggression. Specifically, as the orbitofrontal cortex is involved in the computations of expectation of reward and violations of expected reward result in frustration, orbitofrontal dysfunction may be associated with reactive aggression. Evidence for orbitofrontal dysfunction in CD comes from an fMRI study that assessed brain activation during a continuous performance task measuring sustained attention and the effects of reward on performance. Children and adolescents with CD without ADHD showed underactivation in the right orbitofrontal cortex during the reward condition relative to healthy comparison subjects and children and adolescents with ADHD without CD (Rubia, Smith, et al., Reference Rubia, Smith, Halari, Matsukara, Mohammad and Taylor2009). This hyposensitivity to reward is consistent with the psychophysiological studies discussed above (Beauchaine et al., Reference Beauchaine, Gatzke-Kopp and Mead2007; Beauchaine et al., Reference Beauchaine, Hong and Marsh2008; Crowell et al., Reference Crowell, Beauchaine, Gatzke-Kopp, Sylvers, Mead and Chipman-Chacon2006), as well as with the study by Herpertz et al. (Reference Herpertz, Mueller, Qunaibi, Lichterveld and Konrad2005) discussed earlier that showed a generalized deficit in autonomic responding. Deficient functioning of the orbitofrontal cortex may result in compromised stimulus-reinforcement learning in CD (Rubia, Reference Rubia2011). Abnormalities in reward computations that are mediated by the orbitofrontal cortex may lead to enhanced frustration and facilitate reactive aggression (Rubia, Smith, et al., Reference Rubia, Smith, Halari, Matsukara, Mohammad and Taylor2009), as suggested by Blair (Reference Blair2004). This specific association with reactive aggression and related uncontrolled disruptive symptoms, however, has not been explicitly studied in children and adolescents with CD.

Finally, in the study by Rubia, Smith, et al. (Reference Rubia, Smith, Halari, Matsukara, Mohammad and Taylor2009), CD individuals showed decreased activation in paralimbic regions of the insula, hippocampus, and anterior cingulate cortex and cerebellum during the sustained attention condition. These findings suggest that CD is characterized by reduced brain activity in paralimbic regions that contribute to attention networks through their role in motivation (Rubia, Smith, et al., Reference Rubia, Smith, Halari, Matsukara, Mohammad and Taylor2009). This is especially relevant with respect to the possible role of dopamine in ODD and CD.

Neurotransmitters

Lower activity of brain dopamine systems could result in a reduced salience of positive emotional stimuli in the environment, or a lack of motivation to exert effort to obtain rewards (Berridge, Reference Berridge2007). Low motivation to obtain natural rewards associated with unpleasant affect may incite individuals to search unnaturally strong reinforcers such as drugs. A recent study showed that psychopathic traits (assessed in a community sample) predicted an increased nucleus accumbens dopamine response to amphetamine (Buckholtz et al., Reference Buckholtz, Treadway, Cowan, Woodward, Benning and Li2010). Although seemingly at odds with the view of hypoactive dopamine systems, it is very well possible that low baseline dopamine activity is associated with adaptive changes such as reduced sensitivity of D2 autoreceptors, which result in an increased dopamine response when this system is pharmacologically challenged with psychostimulant drugs.

Genetic studies have found aggression and ODD and CD symptoms to be associated with variations in the structure of the gene encoding the dopamine D4 receptor (DRD4). The DRD4 gene was associated with maternal report of problems with aggression at age 4 (Schmidt, Fox, Rubin, Hu, & Hamer, Reference Schmidt, Fox, Rubin, Hu and Hamer2002), as well as with ADHD with comorbid conduct problems (Holmes et al., Reference Holmes, Payton, Barrett, Harrington, McGuffin and Owen2002). Moreover, genetic studies of catechol-O-methyltransferase (COMT) and monoamine oxidase A (MAOA) show an association between lower levels of bioavailable catecholamines in the brain and the DBDs. COMT is an enzyme that metabolizes dopamine, adrenaline, and noradrenaline into inactive forms. The short form of COMT is responsible for the substitution of methionine (met) for valine (val). Carriers of the short isoform of the valine allele have significantly more COMT activity than carriers of the methionine allele. Therefore, the val/val genotype is argued to more rapidly inactivate catecholamines as compared to the val/met genotype. Thus, the val/val genotype may be associated with lower levels of bioavailable catecholamines in the brain, including dopamine. Together with the evidence suggesting that COMT plays an important role in prefrontal cortex function (Winterer & Goldman, Reference Winterer and Goldman2003), a link between COMT and early antisocial behavior has been proposed (Thapar et al., Reference Thapar, Langley, Fowler, Rice, Turic and Whittinger2005). The first evidence in support of the COMT–DBD link was reported by Thapar et al. (Reference Thapar, Langley, Fowler, Rice, Turic and Whittinger2005), who found a relation between antisocial behavior in ADHD and the presence of the val/val genotype. Likewise, Caspi et al. (Reference Caspi, Langley, Milne, Moffitt, O'Donovan and Owen2008) have shown that ADHD subjects with the val/val genotype displayed more symptoms of CD, were more aggressive, and were more likely to be convicted of criminal offenses than methionine carriers. In addition, evidence was found for involvement of the valine variant in ADHD comorbid with ODD (Qian et al., Reference Qian, Liu, Wang, Yang, Guan and Faraone2009). These studies suggest a relationship between lowered catecholaminergic activity, including dopamine, and ODD or CD.

Although a polymorphism in the promoter region of the gene encoding MAOA conferring low levels of MAOA expression (resulting in high levels of dopamine) has been found to be associated with antisocial behavior (albeit in interaction with environmental risks; Caspi et al., Reference Caspi, McClay, Moffitt, Mill, Martin and Craig2002; Foley et al., Reference Foley, Eaves, Wormley, Silberg, Maes and Kuhn2004), high MAOA activity (resulting in low levels of dopamine) has also been found in children with externalizing and aggressive behavior (Lee et al., Reference Lee, Jin, Hong, Park, Kim and Yim2004; van der Vegt et al., Reference van der Vegt, Oostra, Arias-Vàsquez, van der Ende, Verhulst and Tiemeier2009). In sum, results of studies on MAOA in ODD and CD are inconsistent.

To study dopamine functioning, measurements of homovanillic acid (HVA), the metabolite of dopamine, have been used. No correlation of HVA in cerebrospinal fluid and aggression has been demonstated in children and adolescents with DBDs and/or ADHD (Kruesi et al., Reference Kruesi, Rapaport, Hamburger, Hibbs, Potter and Lenane1990). In addition, no correlation between HVA in plasma and conduct problems in hospitalized children was found (Rogeness, Javors, Maas, Macedo, & Fischer, Reference Rogeness, Javors, Maas, Macedo and Fischer1987). In contrast, low HVA plasma levels in children with DBDs with or without ADHD were found; plasma HVA levels between DBD subgroups with or without ADHD did not differ (van Goozen et al., Reference van Goozen, Matthys, Cohen-Kettenis, Westenberg and van Engeland1999). In sum, based on these studies there is limited evidence of decreased dopaminergic functioning in children and adolescents with ODD and CD.

Studies of the effectiveness of psychostimulants, which enhance dopaminergic and noradrenergic neurotransmission by blocking the reuptake and/or enhancing the release of these neurotransmitters (Fone & Nutt, Reference Fone and Nutt2005), may give insight into the neurochemical mechanisms of ODD and CD. These studies have been conducted primarily in ADHD, but in view of the substantial comorbidity of the DBDs with ADHD (Angold, Costello, & Erkanli, Reference Angold, Costello and Erkanli1999), it is no surprise that many studies have investigated the effect of psychostimulant drugs, such as methylphenidate and amphetamine, in subjects with ADHD with or without DBD or DBD symptoms. The effect of psychostimulants on disruptive behaviors has been examined in a number of studies and in one meta-analysis.

In an important early study on the effect of methylphenidate on disruptive behaviors in children with disruptive, inattention, and hyperactivity symptoms, methylphenidate was an effective treatment for many children with disruptive behavior (Taylor et al., Reference Taylor, Schachar, Thorley, Wieselberg, Everitt and Rutter1987). In this study, hyperactivity was the predictor of the effectiveness of methylphenidate. Specifically, it was hyperactivity rather than defiance that predicted the degree to which defiance was reduced. In another study, the efficacy of methylphenidate was investigated in children and adolescents with CD; two-thirds of the participants also met criteria of ADHD (Klein et al., Reference Klein, Abikoff, Klass, Ganales, Seese and Pollack1997). Methylphenidate was shown to reduce antisocial behavior. This effect was independent of severity of the children's initial ADHD symptoms, in contrast to the findings of the study by Taylor et al. (Reference Taylor, Schachar, Thorley, Wieselberg, Everitt and Rutter1987). Thus, the Klein et al. study (Reference Klein, Abikoff, Klass, Ganales, Seese and Pollack1997) shows an independent influence of methylphenidate on antisocial behavior.

Connor, Glatt, Lopez, Jackson, and Melloni (Reference Connor, Glatt, Lopez, Jackson and Melloni2002) have conducted a meta-analysis of the effect size for psychostimulants on overt and covert aggression-related behaviors in children and adolescents with ADHD. Examples of overt aggression are physical assault, malicious teasing of others, and temper tantrums, whereas examples of covert aggression are lying, stealing, and vandalism. In 24 of the 28 reviewed studies, ADHD was the primary diagnosis. Of the 24 studies in which ADHD was the primary disorder, 17 studies had CD or ODD as comorbid diagnoses. In two studies CD was the primary diagnosis with ADHD as a comorbid disorder, and in two studies mental retardation was the primary diagnosis with ADHD and CD as comorbid disorders. The overall weighted mean effect size was 0.84 for overt aggression and 0.69 for covert aggression. However, in studies in which ADHD was the primary diagnosis, increased prevalence of either ODD or CD led to diminished effect sizes for overt aggression. Thus, although psychostimulants affected aggression in children with ADHD, ODD and CD appeared to moderate the effect.

With respect to the behavioral mechanism of action of psychostimulants, methylphenidate has been shown to exert positive effects on cognitive performance by increasing task salience (Volkow et al., Reference Volkow, Wang, Fowler, Telang, Maynard and Logan2004). Because incentive salience or “wanting” is particularly influenced by dopaminergic neurotransmission (Berridge, Reference Berridge2007), an increase in incentive salience by psychostimulants might explain why parents of children and adolescents with DBDs treated with methylphenidate report improvement in their children's attention and motivation to comply and otherwise to engage more positively in social interactions. It may be that methylphenidate improves function of paralimbic regions in children and adolescents with ODD and CD because these regions have been shown to play a role in attention networks in CD (Rubia, Smith, et al., Reference Rubia, Smith, Halari, Matsukara, Mohammad and Taylor2009). Taken together, there is some evidence of decreased dopaminergic functioning associated with ODD and CD based on genetic studies, studies of the metabolite of dopamine, and pharmacological studies.

Interim summary of reward processing

Studies on the ANS suggest hyposensitivity to reward, and decreased dopaminergic functioning may be associated with reduced salience of environmental stimuli in ODD and CD. These factors make adolescents more vulnerable for stimulation/thrill/sensation seeking or reward seeking, which may manifest in antisocial and delinquent behavior. Healthy individuals will usually not percieve these forms of behavior as pleasurable, but they may be experienced as positive in ODD and CD individuals because they improve an unpleasant state of stimulus deprivation, which may manifest in boredom. In addition, reward seeking may make adolescents with ODD or CD vulnerable for substance abuse. Substances of abuse have much larger rewarding effects than natural rewards such as food and sex, and they enhance the incentive salience of associated stimuli in the environment through their effects on dopaminergic neurotransmission (Berridge, Reference Berridge2007; Fone & Nutt, Reference Fone and Nutt2005). Reduced reward sensitivity and deficient functioning of the orbitofrontal cortex may also result in compromised stimulus-reinforcement learning of appropriate behaviors in children and adolescents with ODD or CD. Finally, orbitofrontal cortex dysfunctioning in CD may be associated with reactive aggression owing to abnormalities in computations of expectation of reward.

EFs

According to a meta-analysis of studies in school-aged children, adolescents, and adults with externalizing disorders, the average mean effect size (Cohen d) of a variety of EF measures for the groups with antisocial behavior (criminality, delinquency, CD, psychopathy, antisocial personality disorder) was 0.62, whereas the effect size for CD was 0.36 (Morgan & Lilienfield, Reference Morgan and Lilienfeld2000). The possible role of ADHD comorbidity in EF impairments in CD was not examined in this study. However, another meta-analysis revealed that EF deficits in CD are likely due to the presence of comorbid ADHD (Pennington & Ozonoff, Reference Pennington and Ozonoff1996). In contrast, a meta-analysis showed that deficits in response inhibition were found not only in elementary school children with ADHD but also in children with DBD without comorbid ADHD (Oosterlaan, Logan, & Sergeant, Reference Oosterlaan, Logan and Sergeant1998). Likewise, impairments in inhibition have been found in preschool children with aggressive behavior when compared with nonaggressive preschoolers, and these impairments were maintained after controlling for attention problems (Raaijmakers et al., Reference Raaijmakers, Smidts, Sergeant, Maassen, Posthumus and van Engeland2008).

The results of studies in DBDs are less inconsistent when motivational factors are involved in EFs. In this respect, the distinction has been made between “cool” cognitive EFs mediated by lateral inferior and dorsolateral frontostriatal and frontoparietal networks and “hot” EFs that involve motivation and are mediated by orbitofrontal and medial prefrontal networks and underlying limbic structures including the amygdala (Rubia, Reference Rubia2011). Impairments in hot EFs may result in perseverative behavior, which involves the inability to show adaptive behavioral responses to changes in the environment. For example, the tendency to continue a previously rewarded response that is now punished, in other words, response perseveration, has been demonstrated in children and adolescents with DBDs (Daugherty & Quay, Reference Daugherty and Quay1991; Matthys, van Goozen, de Vries, Cohen-Kettenis, & van Engeland, Reference Matthys, van Goozen, de Vries, Cohen-Kettenis and van Engeland1998; Matthys, van Goozen, Snoek, & van Engeland, Reference Matthys, van Goozen, Snoek and van Engeland2004; Shapiro, Quay, Hogan, & Schwartz, Reference Shapiro, Quay, Hogan and Schwartz1988; van Goozen et al., Reference van Goozen, Cohen-Kettenis, Snoek, Matthys, Swaab-Barneveld and van Engeland2004) and in children with psychopathic tendencies (O'Brien & Frick, Reference O'Brien and Frick1996). In one study, boys with ODD showed not only response perseveration but also decreased punishment sensitivity (Matthys et al., Reference Matthys, van Goozen, Snoek and van Engeland2004). Thus, children with DBDs seem to have difficulties in stopping their ongoing behavior in response to punishment cues. However, it cannot be excluded that decreased reward sensitivity in these children induces an increased search for rewards in order to achieve a pleasant affective state. Response perseveration is thought to be the result of impaired functioning of the orbitofrontal cortex (Schoenbaum, Roesch, Stalnaker, & Takahashi, Reference Schoenbaum, Roesch, Stalnaker and Takahashi2009).

Blair (Reference Blair2004) suggested that dysfunctions of the orbitofrontal cortex in individuals with antisocial behavior may be associated with impaired decision making. To assess impaired decision making in patients with damage to the ventromedial prefrontal cortex (consisting of the medial prefrontal cortex and the orbitofrontal cortex), the Iowa gambling task was developed by Bechara, Damasio, Damasio, and Anderson (Reference Bechara, Damasio, Damasio and Anderson1994). This decision-making task simulates real-life decisions involving reward, punishment, and uncertainty of outcomes. Specifically, in this task it is not possible to calculate the net gains and losses when making decisions. Instead, one needs to develop an estimate or intuition of which choices are risky and which ones are profitable in the long run. Therefore, this task is supposed to assess affective or intuitive decision making. Impaired functioning on this task and similar tasks, indicating favoring immediate rewards despite long-term punishments, has been shown in young adults with psychopathic characteristics (van Honk, Hermans, Putman, Montagne, & Schutter, Reference van Honk, Hermans, Putman, Montagne and Schutter2002); children and adolescents with psychopathic tendencies (Blair, Colledge, & Mitchell, Reference Blair, Colledge and Mitchell2001), adolescents with ADHD or CD (Ernst et al., Reference Ernst, Grant, London, Contoreggi, Kimes and Spurgeon2003); children with ODD, most of whom were comorbid with ADHD (Luman, Sergeant, Knol, & Oosterlaan, Reference Luman, Sergeant, Knol and Oosterlaan2010); adolescents with CD, some of whom had comorbid ADHD (Fairchild et al., Reference Fairchild, van Goozen, Stollery, Aitken, Savage and Moore2009); and adolescents with both DBDs and substance dependence (Schutter, van Bokhoven, Vanderschuren, Lochman, & Matthys, Reference Schutter, van Bokhoven, Vanderschuren, Lochman and Matthys2011).

Impaired functioning of the orbitofrontal cortex (Rubia, Smith, et al., Reference Rubia, Smith, Halari, Matsukara, Mohammad and Taylor2009) may underlie compromised affective decision making in ODD and CD. The orbitofrontal cortex is involved in controlling reward-related and punishment-related behavior (Rolls, Reference Rolls2004). Specifically, the orbitofrontal cortex is responsible for calculating the value of reward outcome, including the assessment of trade-offs, determining how well the outcome satisfies current needs, and comparing the outcome with other potential reward outcomes (Schoenbaum et al., Reference Schoenbaum, Roesch, Stalnaker and Takahashi2009). Thus, affective decision making in which quick decisions in everyday situations are made on the basis of intuition, regarding which solution is risky and which is profitable in the long run, may be compromised in ODD and CD. This type of deficit also occurs in other individuals who decide against their best interest and have difficulty learning from their mistakes, such as those with substance use disorders (Bartzokis et al., Reference Bartzokis, Lu, Beckson, Rapaport, Grant and Wiseman2000; Bechara et al., Reference Bechara, Dolan, Denburg, Hindes, Anderson and Nathan2001; Grant, Contoreggi, & London, Reference Grant, Contoreggi and London2000; Mazas, Finn, & Steinmetz, Reference Mazas, Finn and Steinmetz2000).

Cerebral cortex

With respect to possible structural deficits of the cerebral cortex, in a region of interest (ROI) study right temporal lobe and right temporal gray matter volumes were significantly reduced in children and adolescents with early-onset CD, the majority of who also had comorbid ADHD compared to controls. Prefrontal cortical volumes in these subjects were 16% smaller than controls, but the difference did not reach statistical significance (Kruesi, Casanova, Mannheim, & Johnson-Bilder, Reference Kruesi, Casanova, Mannheim and Johnson-Bilder2004). In a study with early-onset adolescents with CD (most of whom were comorbid with ADHD), whole brain volume analysis showed reduced gray matter volumes in the left orbitofrontal region, bilaterally in the temporal lobes, and in the amygdala and hippocampus on the left side compared with healthy controls; the mean total gray matter volume was 6% smaller in the clinical group. Regression analyses indicated that CD symptoms were correlated primarily with gray matter reductions in limbic brain structures, including the amygdala and the prefrontal areas, whereas hyperactive/impulsive symptoms were correlated with gray matter abnormalities in the frontoparietal and temporal cortices (Huebner et al., Reference Huebner, Vloet, Marx, Konrad, Fink and Herpertz2008). In a community sample of adolescents with conduct problems, ADHD problems, and CU traits, whole brain volumetric analysis showed increased gray matter in the medial orbitofrontal cortex and the anterior cingulate cortex relative to their healthy twins, controlling for ADHD symptoms. It is interesting that the norm-typical negative correlation between age and cortical thickness was not found in the adolescents with conduct problems and CU traits, suggesting a delay of normal brain maturation in these brain structures (De Brito et al., Reference De Brito, Mechelli, Wilke, Laurens, Jones and Barker2009). In an ROI study, reduced gray matter volumes bilaterally in the anterior insula and in the left amygdala were found in adolescents with CD relative to healthy controls. Aggressive behavior appeared to be the strongest predictor for gray matter volume in the left and right insula, whereas attention problems were the strongest predictor for gray matter volume in the left amygdala (Sterzer et al., Reference Sterzer, Stadler, Poustka and Kleinschmidt2007). Finally, in an ROI study, structural deficits of the cerebellum have been found not only in ADHD but also in CD: children with ADHD only and those with comorbid ADHD and CD both differed from controls in the volume of the left and total posterior superior and inferior lobes of the vermis (Bussing, Grudnik, Mason, Wasiak, & Leonard, Reference Bussing, Grudnik, Mason, Wasiak and Leonard2002). Vermal abnormalities are in agreement with the view that the cerebellum is involved in affective and cognitive processes (Schutter & van Honk, Reference Schutter and van Honk2005).

With respect to functional impairments, the anterior cingulate cortex, which is involved in emotion processing and social functioning, has been studied in children and adolescents with DBDs. In an fMRI study by Stadler et al. (Reference Stadler, Sterzer, Schmeck, Krebs, Kleinschmidt and Poustka2007), children aged 9 to 15 years with CD, the majority of who also had comorbid ADHD, viewed negative pictures and showed a deactivation in the dorsal part of the anterior cingulate cortex, that is, the part involved in the cognitive control of emotional behavior, as compared to healthy controls. This abnormal suppression of neural activity may result in a failure to cognitively control emotional behavior. Likewise, abnormal right anterior cingulate cortex activation during the presentation of images with negative valence was shown in children and adolescents aged 9 to 15 years with CD, most of whom also had comorbid ADHD relative to healthy controls (Sterzer et al., Reference Sterzer, Stadler, Krebs, Kleinschmidt and Poustka2005). In addition, in another fMRI study, children and adolescents with CD and children and adolescents with ADHD performed an inhibition task and showed reduced activation in the posterior cingulate compared to healthy controls. Children and adolescents with CD showed reduced activation in temporal–parietal regions during failed inhibition when compared with the other groups. Because participants obtained feedback about their inhibition failures, the results suggest that performance-monitoring networks are dysfunctional in CD when compared to ADHD and healthy controls (Rubia et al., Reference Rubia, Halari, Smith, Mohammed, Scott and Giampetro2008). Consistent with this notion is the observation that adolescents with psychopathic traits and ODD or CD, the majority of who also had comorbid ADHD, showed abnormal responses of the ventromedial prefrontal cortex during punished errors in a reversal learning task, as compared to adolescents with ADHD and healthy controls (Finger et al., Reference Finger, Marsh, Mitchell, Reid, Sims and Budhani2008). According to Rubia et al. (Reference Rubia, Halari, Smith, Mohammed, Scott and Giampetro2008), this could mean that adolescents with CD care less about their mistakes than adolescents with ADHD and healthy controls, which is in line with evidence that children with DBDs are undermotivated and respond less to negative feedback than controls (Matthys et al., Reference Matthys, van Goozen, Snoek and van Engeland2004; van Goozen, Cohen-Kettenis, et al., Reference van Goozen, Snoek, Matthys, van Rossum and van Engeland2004). As already discussed in the section on reward processing, lower responsiveness to reward outcome information within the orbitofrontal cortex has been shown in children and adolescents with CD (Rubia, Smith, et al., Reference Rubia, Smith, Halari, Matsukara, Mohammad and Taylor2009). In line with this, less orbitofrontal responsiveness to stimulus-reinforcement exposure and rewards while performing a passive avoidance learning task was shown in adolescents with DBDs and psychopathic traits relative to comparison youths. Thus DBD adolescents with psychopathic traits seem to be less able to represent reward expectance values resulting in a higher vulnerability to impaired decision making (Finger et al., Reference Finger, Marsh, Blair, Reid, Sims and Ng2011). During interference inhibition and attention allocation, boys with CD only compared to controls showed reduced activation in the right middle and superior temporal and parietal regions; impaired functioning of these areas possibly accounts for attentional deficits in CD, causing more errors (Rubia, Halari, et al., Reference Rubia, Smith, Halari, Matsukara, Mohammad and Taylor2009). Finally, during cognitive flexibility, when compared with control boys and ADHD-only boys, no disorder-specific brain underactivation was observed in boys with CD only (Rubia et al., Reference Rubia, Halari, Cubillo, Smith, Mohammed and Scott2010).

EFs are thought to heavily rely on optimal functioning of the prefrontal cortex. In this regard, the prefrontal dysfunctions that have been reported in CD are consistent with impairments in EF. Given the anatomical and functional heterogeneity of the prefrontal cortex, a more detailed understanding of prefrontal function in ODD and CD is necessary in order to appreciate how altered function of separate prefrontal subregions contributes to various mental and behavioral aspects of the DBDs. Finally, both neuroimaging and neuropsychological research has been biased in the selection of ROI (structural studies), paradigm selection (fMRI), and use of EF tasks, in that functions such as working memory and planning, and their associated neural systems, have not been systematically investigated in CD and ODD (Rubia, Reference Rubia2011).

The observed structural abnormalities in the cerebral cortex are paralleled by functional deficiencies in information processing. Electrophysiological (EEG) studies have found amplitude reductions of the P300 brain wave in children with DBDs (for a review see Cappadocia, Desrocher, Pepler, & Schroeder, Reference Cappadocia, Desrocher, Pepler and Schroeder2009). The P300 is a cortically generated brain wave that is suggested to reflect online adjustment of mental representations in response to environmental changes. The P300 emerges around 300 ms over frontal and parietal regions in response to a stimulus that is relevant to the subject. It has been proposed that the P300 amplitude reductions represent a neurocognitive deficit of explicitly acknowledging and incorporating novel information in order to effectively guide behavior under changing conditions. The reduced P300 amplitude and its associated neurocognitive deficits could provide an explanation as to why children with DBDs show less flexibility in adjusting their behavior (Bauer & Hesselbrock, Reference Bauer and Hesselbrock1999; Baving, Rellum, Laucht, & Schmidt, Reference Baving, Rellum, Laucht and Schmidt2006).

In addition to externally (exogenous) evoked brain activity, irregularities in the spontaneous (endogenous) generation of neural activity as reflected by the waxing and waning of brain oscillations across different frequency bandwidths have also been observed in various externalizing spectrum disorders. For example, significant reductions in the spontaneous generation of beta (13–30 Hz) oscillations in the EEGs of young adults have been observed in subjects with impulsivity and antisocial personalities (Houston & Stanford, Reference Houston and Stanford2005). These reductions have been interpreted as lowered levels of cortical arousal that set the stage for decreased cognitive control and increased impulsive behavior. In agreement, other studies have found that reduced beta activity relative to theta (4–7 Hz) activity is associated with disadvantageous decision making in healthy adult volunteers (Schutter & van Honk, Reference Schutter and van Honk2005) and impulsive behavior found in patients with ADHD (Barry, Clarke, & Johnstone, Reference Barry, Clarke and Johnstone2003). In sum, electrophysiological studies provide further support for the view that the behavior in DBDs can at least in part be understood in terms of cortical abnormalities and subsequent impairments in attention and EF.

Neurotransmitters

Proper function of the prefrontal cortex strongly depends upon monoaminergic inputs. One particular aspect that is pertinent to EF in ODD and CD is the notion that they modulate mental functions according to an inverted U-shaped function: dopaminergic, noradrenergic, as well as serotonergic activity needs to be at an optimum, whereby both increases and decreases from this optimal level lead to impaired function (Robbins & Arnsten, Reference Robbins and Arnsten2009). Dopaminergic neurotransmisson has been widely implicated in working memory, as well as in impulse control (Chamberlain, Müller, Robbins, & Sahakian, Reference Chamberlain, Müller, Robbins and Sahakian2006; Dalley, Cardinal, & Robbins, Reference Dalley, Cardinal and Robbins2004; Pattij & Vanderschuren, Reference Pattij and Vanderschuren2008; Robbins & Arnsten, Reference Robbins and Arnsten2009). A reduction in dopaminergic activity could therefore well play a role in the EF impairments in ODD and CD. Because it has been shown that psychostimulants activate the medial orbitofrontal cortex and the rostral part of the anterior cingulate cortex (Völm et al., Reference Völm, de Araujo, Cowel, Rolls, Kringelbach and Smith2004), the beneficial effect of methylphenidate on symptoms of ODD and CD in children and adolescents with these disorders and ADHD comorbidity (Turgay, Reference Turgay2009) may be the result of improved functioning of these structures.

In addition, noradrenergic neurotransmission has been implicated in working memory, behavioral inhibition, and the balance between optimizing task performance versus searching alternative behavioral strategies, whereas serotonergic neurotransmission plays an important role in behavioral flexibility (i.e., reversal learning), behavioral inhibition, and decision making (Aston-Jones & Cohen, Reference Aston-Jones and Cohen2005; Chamberlain et al., Reference Chamberlain, Müller, Robbins and Sahakian2006; Dalley et al., Reference Dalley, Cardinal and Robbins2004; Pattij & Vanderschuren, Reference Pattij and Vanderschuren2008; Robbins & Arnsten, Reference Robbins and Arnsten2009; Rogers et al., Reference Rogers, Everitt, Baldacchino, Blackshaw, Swainson and Wynne1999). Therefore, the alterations in the function of these important modulatory neurotransmitters likely play a prominent role in the deficits in various EFs such as behavioral inhibition and decision making in ODD and CD. In this respect it is worth mentioning that atomoxetine, a selective noradrenaline transporter inhibitor, has been found to reduce ODD symptoms in children and adolescents with ADHD with or without ODD comorbidity (Newcorn, Spencer, Biederman, Milton, & Michelson, Reference Newcorn, Spencer, Biederman, Milton and Michelson2005).

Interim summary of cognitive control

DBDs have been associated with structural deficits and impaired functioning of the paralimbic system comprising the orbitofrontal cortex, superior temporal lobes, and cingulate cortices. Likewise, impairments in executive or cognitive control functions have been shown in ODD and CD, especially when motivational factors (reward and punishment) are involved. Thus, compromised cognitive control over emotional behavior in ODD and CD may result in reactive aggression and related uncontrolled disruptive symptoms (e.g., losing temper). Furthermore, compromised processing of both punishment and reward cues in the orbitofrontal cortex leads to impaired representation of information on reinforcement expectations, increasing the risk of deficient decision making. Thus, children and adolescents with ODD or CD choose not to approach objects associated with reward and avoid those associated with punishment but instead choose actions that harm others.