REVIEW
Siri D. S. Noordermeer1 & Marjolein Luman1& Jaap Oosterlaan1
Received: 28 August 2015 /Accepted: 18 December 2015 /Published online: 5 February 2016# The Author(s) 2016. This article is published with open access at Springerlink.com
Abstract Oppositional defiant disorder (ODD) and conductdisorder (CD) are common behavioural disorders in childhoodand adolescence and are associated with brain abnormalities.This systematic review and meta-analysis investigates struc-tural (sMRI) and functional MRI (fMRI) findings in individ-uals with ODD/CD with and without attention-deficit hyper-activity disorder (ADHD). Online databases were searched forcontrolled studies, resulting in 12 sMRI and 17 fMRI studies.In line with current models on ODD/CD, studies were classi-fied in hot and cool executive functioning (EF). Both themeta-analytic and narrative reviews showed evidence ofsmaller brain structures and lower brain activity in individualswith ODD/CD in mainly hot EF-related areas: bilateral amyg-dala, bilateral insula, right striatum, left medial/superior fron-tal gyrus, and left precuneus. Evidence was present in bothstructural and functional studies, and irrespective of the pres-ence of ADHD comorbidity. There is strong evidence thatabnormalities in the amygdala are specific for ODD/CD ascompared to ADHD, and correlational studies further supportthe association between abnormalities in the amygdala andODD/CD symptoms. Besides the left precuneus, there wasno evidence for abnormalities in typical cool EF related struc-tures, such as the cerebellum and dorsolateral prefrontal cor-tex. Resulting areas are associated with emotion-processing,error-monitoring, problem-solving and self-control; areas as-sociated with neurocognitive and behavioural deficits impli-cated in ODD/CD. Our findings confirm the involvement of
hot, and to a smaller extent cool, EF associated brain areas inODD/CD, and support an integrated model for ODD/CD (e.g.Blair, Development and Psychopathology, 17(3), 865-891,2005).
Keywords ODD .CD .ADHD .StructuralMRI . FunctionalMRI . ALEmeta-analysis
Introduction
In recent years, magnetic resonance imaging (MRI) studieshave provided insight into the underlying brain mechanismsof disruptive behaviour disorders, including oppositional de-fiant disorder (ODD) and conduct disorder (CD). This reviewwill integrate and discuss studies using structural (sMRI) andfunctional MRI (fMRI) in these disorders. ODD and CD aredevelopmental disorders that are among the most commonlydiagnosed mental health conditions in childhood (Hamiltonand Armando 2008; Loeber et al. 2009). Community samplesshow a prevalence rate for ODD ranging between 2 and 14 %and for CD ranging between 2 and 16 % (Boylan et al. 2007;Loeber et al. 2000). Both disorders are more prevalent in boysthan in girls with ratio’s ranging from 3:1 to 9:1 (Loeber et al.2000). ODD is defined by a frequent and persistent pattern ofirritable and angry mood, vindictiveness and developmentallyinappropriate, negativistic, defiant, and disobedient behaviourtoward authority figures (American Psychiatric Association2013). CD is characterised by a persistent pattern of multipleantisocial behaviours during childhood and adolescence, in-cluding fighting, bullying, stealing, vandalism, and lying forpersonal gain (American Psychiatric Association 2013).Depending on whether the individual was younger or olderthan 10 years at the time of symptom onset, there is a
* Siri D. S. [email protected]
1 Faculty of Behavioural and Movement Sciences, ClinicalNeuropsychology Section, VU University Amsterdam, Van derBoechorststraat 1, 1081 BTAmsterdam, The Netherlands
Neuropsychol Rev (2016) 26:44–72DOI 10.1007/s11065-015-9315-8
A Systematic Review and Meta-analysis of Neuroimagingin Oppositional Defiant Disorder (ODD) and Conduct Disorder(CD) Taking Attention-Deficit Hyperactivity Disorder (ADHD)Into Account
differentiation between childhood-onset CD and adolescent-onset CD (American Psychiatric Association 2013).
Although both disorders have some distinct characteristics,the general consensus is that ODD and CD are highly corre-lated expressions of psychopathology. Till the emergence ofthe DSM-5, ODD has been coined as a milder version of CDas emphasised by the hierarchical rule stipulated in the DSM-IV stating that a diagnosis of ODD is precluded when CD ispresent, due to very high levels of ODD features in individualswith CD and the precursory role of ODD for CD. This pre-cursory role of ODD for the later development of CD is sup-ported by a quadrupled risk for the development of CD inindividuals with ODD (Burke et al. 2002; Loeber et al.2009; Rowe et al. 2010). In addition to the high levels ofODD features in CD, comorbidity rates of ODD are as highas 45 % in children with CD, and in clinical samples theserates increase to up to 96 % (Loeber et al. 2009; Rowe et al.2002). ODD and CD share risk factors in both the psychoso-cial domain, such as poverty and social disadvantage, and thefamily domain, such as a history of criminality in biologicalparents (Rowe et al. 2002; Burke et al. 2002). An extensivereview on the heritability of ODD and CD shows heritabilityrates of 61 and 74 %, respectively, with 50 % of the reportedgenes being associated with both disorders (Coolidge et al.2000; Lahey and Waldman 2012). The presence of eitherODD or CD predicts poor future outcomes, including com-promised psychiatric, family and social functioning, as well asan increased risk for adverse life events: e.g., peer rejection,criminal behaviour and incarceration at a young age (Burke etal. 2002; Hamilton and Armando 2008; Loeber et al. 2000).When ODD or CD persists, individuals are at a heightenedrisk for anxiety disorders and depression. Furthermore, persis-tence of childhood ODD or CD into adulthood results in adiagnosis of antisocial personality disorder (APD), which isin turn related to high rates of domestic violence, unemploy-ment and homelessness (Loeber et al. 2008; Kimonis andFrick 2010). APD can only be diagnosed when there is ahistory of some symptoms of CD and the transition fromCD to APD occurs in around 54 % of individuals with CD(Fairchild et al. 2013b; American Psychiatric Association2013). Treatment of ODD and CD is generally not specificto either disorder, and a combination of interventions that aimat multiple domains tends to bemore successful than treating asingular domain. In the current review, ODD and CD will betreated as representing one dimension of psychopathology,because of the similarities in many domains includingaetiology, phenotypical manifestation, correlated features, aswell as treatment.We will refer to this dimension as ODD/CD.
Several explanatory models of ODD/CD focus specificallyon neurocognitive impairments, which are thought to be relat-ed to abnormalities in underlying brain mechanisms.Neurocognitive impairments that are associated with ODD/CD include lower IQ, deficiencies in inhibitory control,
abnormalities in emotional processing and social cognition,and abnormalities in reinforcement processing. Most of theexplanatory models emphasise a deficit in so-called executivefunctioning (EF). EF is the sum of neurocognitive processesthat maintain an appropriate problem-solving set to attain agoal (Pennington and Ozonoff 1996; Willcutt et al. 2005). Awell-known distinction in EF is that between hot and cool EF.Hot EF is characterised by motivational and affective aspectsof cognitive processing, such as reinforcement learning, affec-tive decision-making and emotional processing (Anderson etal. 2008; Blair and Lee 2013; Kerr and Zelazo 2004; Zelazoand Carlson 2012). Brain areas that are reported to be impor-tant for hot EF include the amygdala, anterior cingulate cortex,insula and orbitofrontal cortex (Crowe and Blair 2008;Prencipe et al. 2011; Rubia 2011). In contrast, cool EF refersto goal-directed and problem-solving behaviours, as well asself-regulation, not involving motivational or affective as-pects. Cool EF encompasses functions using diverse abilitiessuch as inhibition, working memory, planning, flexibility, andthe ability to creatively generate solutions for problems(Sarkar et al. 2013; Diamond 2013). Brain areas reported tobe central to cool EF include the dorsolateral prefrontal cortexand the cerebellum (Prencipe et al. 2011; Rubia 2011; Yangand Raine 2009; Sterzer and Stadler 2009). This distinctionbetween hot and cool EF provides a framework to study un-derlying brain mechanisms of observed behavioural andneurocognitive abnormalities in ODD/CD. This knowledgecan help test theoretical models on ODD/CD through clarify-ing involvement of brain areas central to those models. Thisimportant information on theoretical model building can inturn help to further advance the field, by yielding supportingor opposing evidence for the involvement of brain areas.
An altered reinforcement system, and thus a hot EF problem,was proposed in early models on ODD/CD by Quay (1965,1993) and by Newman and Kosson (1986). Indeed, a recentextensive review in antisocial individuals showed altered sen-sitivity to reward and punishment and processing of these con-tingencies (Byrd et al. 2014). Specifically, antisocial individualsshow an increased affinity for immediate reward over delayedreward, an insensitivity to punishment, and increased reward-seeking (Byrd et al. 2014). This altered reinforcement sensitiv-ity has been related to problems in social cognition in individ-uals with ODD/CD. For example, these individuals tend toshow a preference for more aggressive reactions in social situ-ations, which might be due to their unsparing surge for rewardand decreased punishment sensitivity (Burke et al. 2002;Loeber et al. 2008; Quay and Hogan 1999; Rubia 2011). Inaddition, abnormalities in emotional processing have been re-ported repeatedly in studies with ODD/CD samples (Byrd et al.2014), including reduced levels of empathy and deficits in therecognition of emotional expressions (Blair 2013). Both abnor-malities in reinforcement sensitivity and emotional processingin ODD/CD have been related to abnormalities in hot EF brain
Neuropsychol Rev (2016) 26:44–72 45
areas, such as the amygdala and the striatum (Crowe and Blair2008; Prencipe et al. 2011).
Another important explanatory model, that is more fittingwith a cool EF deficit, was proposed by Moffitt (1993), whodistinguished between adolescent-limited ODD/CD, and themore severe, life-course persistent ODD/CD (Moffitt 1993).According to that model, adolescent-limited ODD/CD ismerely a stage in development during which adaptive socialbehaviour is tested and learned, while life-course persistentODD/CD is thought to arise of an interplay between a difficultand under-controlled temperament and adverse environmentalfactors. This under-controlled temperament is thought to bepromoted by a deficit in cool EF, including difficulties ininhibition and self-control (Burke et al. 2002; Loeber et al.2008; Moffitt 1993). Evidence from neurocognitive studiesgenerally points toward a range of abnormalities in cool EFin ODD/CD, such as low IQ, inefficiencies in problem solv-ing, and less than optimal inhibitory control (Oosterlaan et al.1998; Burke et al. 2002; Loeber et al. 2008; Quay and Hogan1999, but see Van Goozen et al. 2004), and have been relatedto abnormalities in the dorsolateral prefrontal cortex and cer-ebellum (Yang and Raine 2009; Prencipe et al. 2011).
Finally, one of the most influential explanatory models atpresent is proposed by Blair, who suggests that individualswith ODD/CD demonstrate impairments in two separate cir-cuits associated with hot and cool EF (Blair 2005). Accordingto Blair (2005), the first compromised circuit is involved inemotional processing and regulation and is responsible for anincrease in antisocial behaviour. The key component of brainareas underlying this mainly hot EF circuit is supposedly theamygdala. The second compromised circuit is involved inresponse inhibition and is responsible for loss of temper andexaggerated aggressive responses in individuals with ODD/CD. The key component of brain areas underlying this mainlycool EF circuit is supposedly the ventrolateral frontal cortex.
Support for the explanatory models on ODD/CD is wellestablished in behavioural studies. However, knowledgeabout these models in terms of structural and functional neu-roimaging is incomplete. While a review of the current struc-tural and functional neuroimaging literature would be an idealway to provide neurobiological evidence to confirm or reject amodel, such a review has not been conducted so far. Thecurrent review aims to fill this gap with the goal to enhanceinsight into the underlying mechanisms of ODD/CD and ad-ditionally test the plausibility of hot and cool EF models inODD/CD in terms of neural mechanisms.
A highly comorbid condition of ODD/CD is attention-deficit hyperactivity disorder (ADHD), one of the most com-monly diagnosed disruptive behaviour disorders in children.The percentage of individuals diagnosed with ODD/CD thatadditionally qualify for a comorbid ADHD diagnosis rangesup to 35 %, and up to 50 % of children with disruptive behav-iours show symptoms of both ADHD and ODD/CD (Loeber
et al. 2000; Anderson and Kiehl 2013; Connor et al. 2010;Waschbusch 2002). In children and adolescents with ODD/CD and comorbid ADHD, the prognosis, including the risk todevelop anxiety and depressive disorders and antisocial per-sonality disorder, is considerably worse than when only ODD/CD or only ADHD is present (Dolan and Lennox 2013;Loeber et al. 2000). In addition, this comorbid group showsan earlier age of symptom onset, exhibits more physical ag-gression and delinquency, shows significantly higher ODD,CD and ADHD symptom severity, and shows more functionalimpairments than a group with any of these diagnoses in sin-gularity (Anderson and Kiehl 2012; Loeber et al. 2000;Waschbusch 2002). This highlights the importance of clarify-ing the specificity of abnormalities associated with ODD/CDand contrasting ODD/CD-only with ODD/CD+ADHD, whenstudying these disorders.
Although previous reviews have addressed structural andfunctional brain abnormalities in ODD/CD (Matthys et al.2013; Rubia 2011), these reviews were neither systematic re-views, nor focussed exclusively on diagnostic groups ofODD/CD, nor were set out to investigate the relative contri-bution of ADHD. To address structural and functional brainanatomical aspects of ODD/CD, the current comprehensivereview includes a complete systematic narrative review aswell as meta-analyses using ALE (Eickhoff et al. 2009,2012; Laird et al. 2005) of the available structural and func-tional imaging studies. The quantitative approach increasesthe precision and the power of reported results compared toa purely qualitative review, and indicates which brain areas aremost robustly implicated in ODD/CD. However, since anALE meta-analysis does not allow inclusion of studiesreporting on non-significant group differences, the narrativereview provides the necessary balance in interpreting the find-ings. The narrative review complied with the standards of asystematic review, performing a literature search based on adetailed plan and search strategy, and had the goal of reducingbias by identifying, appraising and synthesising all relevantstudies on this topic. In addition, we investigated the specific-ity of brain correlates. Firstly, we investigated specificity bydiscussing results from studies using ODD/CD-only samplesseparately from results from studies using ODD/CD+ADHDsamples. Secondly, we compared results of samples includingindividuals with ODD/CD (with and without comorbidADHD) to samples including individuals with ADHD-only.In addition, we reported associations between abnormalities ininvestigated structures and ODD/CD related symptoms.Knowledge about the specificity of reported abnormalitiesmay help to clarify the heterogeneity in studies on ODD/CD.
This review is divided in two sections. The first sectiondeals with structural findings, integrating findings for all re-ported brain areas; the second section deals with functionalfindings, and is divided in a part describing hot EF and asubsequent part describing cool EF, based on the assessed
46 Neuropsychol Rev (2016) 26:44–72
tasks and contrasts. Each section starts with a quantitativemeta-analysis, for which an activation likelihood estimation(ALE) meta-analysis was performed. ALE is a technique thatis used to identify significant anatomical locations for whicheffects are consistent, and that is robust to publication bias (fora detailed description of ALE see: Eickhoff et al. 2009; Lairdet al. 2005; Eickhoff et al. 2012). This quantitative meta-analysis is followed by a narrative review. Then, both thesections on structural and functional findings are integratedto provide a complete overview of all involved brain areasand to assess the evidence for abnormalities in terms of hotand cool EF related brain areas. We expected to find abnor-malities in brain structure and function of individuals withODD/CD subserving both hot EF (i.e., the amygdala, anteriorcingulate cortex, insula, and orbitofrontal cortex (Rubia 2011;Prencipe et al. 2011; Crowe and Blair 2008)) and cool EF (i.e.the dorsolateral prefrontal cortex and cerebellum (Rubia 2011;Prencipe et al. 2011; Yang and Raine 2009; Sterzer and Stadler2009)), which would be in line with an integrated model suchas the model proposed by Blair.
Methods
Study Selection
This review included all empirical studies that met the follow-ing inclusion criteria: (1) the study reported on functional orstructural magnetic resonance imaging results, comparing (a)individuals with ODD/CD with or without comorbid ADHDto control subjects, and if included, to individuals withADHD-only, or (b) individuals with ODD/CD withoutADHD to ODD/CD individuals with ADHD. The controlgroup of each study was carefully checked on reported psy-chiatric disorders, and when a study reported on the presenceof any psychiatric disorder in participants of the control group,that study was excluded from the current study. This resultedin the exclusion of two studies. (2) Diagnosis of the partici-pants had to be based onDSM-III, DSM-IVor DSM-5 criteria.(3) The study had to be published in a peer-reviewed Englishlanguage journal. No limits were set on the ages of partici-pants. All relevant studies published up till June 2015 wereincorporated.
The databases PubMed, EMBASE, PsycInfo and Web ofScience were searched, using the search terms ODD, CD,disruptive behavioural disorder, disruptive behaviour,externalising behavioural disorder, externalising behaviour,MRI, neuroimaging, and equivalent MeSH terms.Furthermore, reference lists of selected studies and reviewswere checked for additional relevant studies. A total of 576studies were initially retrieved and screened, after which atotal of 67 studies remained that fulfilled inclusion criteriabased on screening of the title and abstract. These 67 studies
were further assessed for eligibility using the full text of thestudy, resulting in 29 studies that met inclusion criteria andwere incorporated in the present review; see Fig. 1 for the flowdiagram of included studies. The 29 studies selected for re-view included a total of 1278 individuals, including 713 pa-tients and 565 controls.
In this review, we distinguished the following samples: (1)ODD/CD-only, including only individuals with ODD/CDwithout comorbid ADHD, (2) ODD/CD-mixed, includingboth individuals with ODD/CD-only and individuals withODD/CD+ADHD, and (3) ODD/CD+ADHD, including onlyindividuals classifying for both ODD/CD and ADHD. Weincluded a total of 10 ODD/CD-only studies (7 structural, 3functional), 14 ODD/CD-mixed studies (9 structural, 5 func-tional), and 5 ODD/CD+ADHD studies (1 structural, 4 func-tional). By differentiating between these three samples, it waspossible to discern whether an abnormality was related toODD/CD and not confounded by comorbid ADHD, whichwould be suggested when an abnormality is found in anODD/CD-only sample. Alternatively, when an abnormalityis found in an ODD/CD-mixed or ODD/CD+ADHD sample,it may be possible that the abnormality is also related to co-morbid ADHD. Therefore, results are discussed separately forODD/CD-only samples versus controls and for ODD/CD-mixed and ODD/CD+ADHD samples versus controls.When included in a study, results from additional comparisonswith an ADHD-only sample were provided.
The section on structural neuroimaging findings reports allbrain areas, not distinguishing between hot and cool EF. Thesection on functional neuroimaging findings is divided in twoparts: the first part is focused on hot EF and the second part isfocused on cool EF. Classification of studies as hot or coolwas done in line with the literature. Two authors independent-ly inspected the task characteristics and the reported contrastsfor each study and judged these as assessing either hot or coolEF. In case of disagreement, the third author was consulted,resulting in full agreement on all studies. One study (Rubia etal. 2009b) targeted both hot and cool EF and we thereforediscussed this study in both parts of the fMRI results. Afterpresentation of the meta-analytic and narrative results of thestructural and functional studies, findings are integrated forstructural and functional studies in order to provide an over-view of all relevant literature and resulting conclusions interms of hot and cool EF related brain areas.
Meta-analysis: Activation Likelihood Estimation
For the quantitative meta-analysis, an activation likelihoodestimation (ALE) meta-analysis was performed using theBrainmap GingerALE software package (Eickhoff et al.2009, 2012; Laird et al. 2005). For the ALE meta-analysis tobe reliable, the minimum number of studies to be included isfive (www.brainmap.org). The algorithm used by GingerALE
Neuropsychol Rev (2016) 26:44–72 47
applies a random-effects approach to identify anatomical lo-cations for which effects are observed most consistently, andwhich renders it robust for the possible effects of publicationbias (Fox et al. 1998). Analyses were performed separately forthe structural and the functional studies, and for the functionalstudies these were performed separately for studies on hot andon cool EF. Studies eligible for inclusion in the meta-analysiswere additionally required to report x/y/z coordinates for clus-ters showing group differences in volume (structural MRI) oractivity (functional MRI) in either Montreal NeurologicalInstitute (MNI) or Talairach space.
All coordinates originally reported in MNI space were nor-malised to Talairach space using Lancaster’s Transform; co-ordinates which were already in Talairach space were convert-ed back to MNI coordinates and subsequently normalised toTalairach space using Lancaster’s Transform, to account fordivergent analyses procedures and minimise differences incoordinates between studies (Lancaster et al. 2007; Laird etal. 2010). The 2.3.1 version of GingerALE (Eickhoff et al.2009) was applied in the current study and used the coordi-nates of the reported voxels of each study, referring to theareas that showed a group difference, as a probability distri-bution to create an ALE distribution map (Turkeltaub et al.2002; Eickhoff et al. 2009, 2012). Results across studies wereaggregated by GingerALE, and modelled activation maps
were generated by calculating the probability that a particularvoxel was activated as the union of probabilities for that voxelacross studies. The contribution of each study to the meta-analytic result was weighted using the study’s sample size,by widening the Gaussian distribution for a voxel with smallersamples to compensate for spatial uncertainty. Meta-analyticmaps were generated by combining all the modelled activa-tions maps, and were subsequently corrected for multiplecomparisons using Family Wise Error correction. This en-sured that differences between studies in terms of the numberof areas showing significant group differences (e.g., due toapplying a lower statistical threshold), did not influence thecombined ALE map (Turkeltaub et al. 2012). This combinedALEmap was then compared to a map from a null distributionwith the same number of foci, but now randomly placedthroughout the grey matter of the brain. After this, the finalALE map was thresholded at p< .05 using a False DiscoveryRate (FDR) correction for multiple comparisons, and a mini-mum cluster size of 100 mm3 (see www.brainmap.org/ale/).
Both studies using a region of interest (ROI) approach, thusa hypothesis driven pre-selection of specific brain areas, andstudies using a whole brain analyses (WBA) approach, thusstudying the entire brain, were included in this review. To testthe possibility that studies using an ROI approach might biasmeta-analytic results due to less stringent statistical criteria,
Fig. 1 PRISMA flow diagramshowing the process of studyselection
48 Neuropsychol Rev (2016) 26:44–72
the resulting map of all studies (using both WBA and ROIapproach) was compared to the resulting map from studieswith a WBA approach. When differences between the mapswere present, results from the map including only the WBAstudies were reported. The ALE map was overlaid onto aTalairach anatomical template for visualisation purposes, andthe areas reported in our Results section refer to the locationsof the extrema.
Results
Results of the reviewed studies are discussed in two mainsections with the first section summarising the structural neu-roimaging findings and the second section summarising thefunctional neuroimaging findings. Both sections start with ageneral overview of all included studies, after which the re-sults from the meta-analysis are reported, followed by theresults from the narrative review. The narrative parts containa summary of the main findings for ODD/CD-only groupsfirst since those are not biased by comorbid ADHD, followedby the main findings in ODD/CD-mixed and ODD/CD+ADHD groups, and ends with the correlational findings andthe specificity of findings regarding ODD/CD.
After presentation of the meta-analytic and narrative resultsof the structural and functional studies, structural and func-tional findings are integrated in order to provide an overviewof all relevant literature and resulting conclusions in terms ofhot and cool EF related brain areas. Since an ALE meta-analysis cannot include studies that reported on non-significant group differences, but rather identifies anatomicallocations for which effects are observed most consistently,findings from the ALE meta-analysis are combined with theresults of the narrative review to provide a comprehensive andbalanced overview of available evidence.
Structural Neuroimaging Findings
General Overview Table 1 shows a total of 12 studies thatinvestigated structural differences in individuals with ODD/CD-only (three studies), ODD/CD-mixed (five studies), orODD/CD+ADHD (four studies) compared to controls. All12 studies usedWBA to investigate regional brain differences,while three studies additionally used an ROI approach(Fairchild et al. 2011, 2013a; Sterzer et al. 2007). Two of thesestudies also reported on results of contrasts between ODD/CD(with and without comorbid ADHD) and ADHD-only. In ad-dition, seven studies investigated the association betweenbrain region volumes and symptom counts of ODD/CD(Sterzer et al. 2007; Huebner et al. 2008; Fahim et al. 2011;Fairchild et al. 2011, 2013a; Stevens and Haney-Caron 2012;Michalska et al. 2015).
All studies investigated children/adolescents. For informa-tion regarding group and study characteristics of the includedstudies, see Table 1. Reported differences in brain structuresare bilateral, unless reported to be either left or right sided.Results are reported from WBA approaches, unless specifiedto be the result of ROI based analyses, see Table 1.
Structural ALE Meta-analysis Eight structural studies, allusing aWBA approach, reported coordinates and were includ-ed in the meta-analysis (studies marked with a in Table 1). Dueto the limited number of studies in ODD/CD-only (three stud-ies) it was not possible to perform an ALE meta-analysis onstudies comparing ODD/CD-only groups and controls (mini-mum number of studies required is five). Hence, the ALEmeta-analysis was performed on all ODD/CD samples, there-with including both individuals with and without comorbidADHD. Total sample sizes for all included studies rangedbetween 24 and 90 individuals, adding up to a total of 415individuals, of which 267 were patients and 148 were controls(age range 8–21 years). Five studies used a full male sample,two studies used a sample consisting of 71 % males and onestudy used a full female sample. These studies provided a totalof 58 foci of grey matter volume abnormalities in individualswith ODD/CD with and without comorbid ADHD.
The ALE analysis revealed four significant clusters of al-tered grey matter volumes that differed between patients andcontrols (see Fig. 2 for visualisation). The largest cluster(760 mm3) was located in the left amygdala, with three fociinside this cluster. A second cluster (456 mm3) was found inthe left insula, containing three foci. A third cluster (352 mm3)was located in the left medial/superior frontal gyrus, contain-ing two foci. The fourth cluster (216 mm3) was located in theright insula, containing two foci.
Narrative Review Table 1 provides an overview of investi-gated samples and approach (WBA or ROI) and summarizesthe findings of the 12 studies reporting on sMRI. Total samplesizes for all included studies ranged between 20 and 111 indi-viduals, adding up to a total of 643 individuals, including 334patients and 309 controls (age range 6–21 years). The samplesconsisted predominantly of males, with five studies using afull male sample, six studies using a sample largely consistingof males (ranging from 48 to 90%), and one study using a fullfemale sample.
ODD/CD-only versus controls (3 studies). For total greymatter volume, one of three studies reported a reduction in theODD/CD-only group. In terms of regional grey matter vol-umes, abnormalities in the insula (left and bilateral), amygdala(left and bilateral), the cingulate cortex (left and paracentral),the inferior frontal gyrus (left and right), and the leftdorsomedial prefrontal cortex were repetitively reported (twoout of three studies).
Neuropsychol Rev (2016) 26:44–72 49
Tab
le1
Summaryof
studycharacteristicsandresults
forstructuralim
aging
Study
Total
sample
size
(%male/
female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Resultsof
comparisons
betweenODD/CD-onlyandcontrolg
roup,
betweenODD/CD+ADHDandcontrolg
roup,and
betweenODD/CD-
mixed
andcontrolg
roup.
(brain
regionsreported
show
reducedvolumes
inthepatient
group,
unless
otherw
isereported)
ODD/
CD
-only
ODD/
CD+
ADHD
ODD/CD-
mixed
ADHD-
only
Controls
Amygdala
Insula
Prefrontal
cortex
Other
1.Fahim
etal.2011a
,d47
c
(100
%male)
22N/A
N/A
N/A
25Patients:8.4
(0.10)
Controls:8.4
(0.07)
Wholebrain(V
BM,
corticalthickness)
Correlationalanalyses
nsBilateral
Left dorsom
edial
ODD/CD-onlyversus
controls:
VBM:L
eftm
edialfrontalcortex
/claustrum,right
inferior
frontal
cortex/in
ferior
parietalcortex
Corticalthickness:leftinsula,left
cingulate,leftanterior
cingulate,
leftmedialfrontal,leftrectal/
orbitofrontal,leftuncus,left
precuneus,rightm
iddlefrontal,
rightsuperiortemporal,right
posteriorcingulate
ODD/CD-onlyversus
controls:
VBM:N
egativecorrelation
betweenrightsuperiortemporal
cortex,leftsuperiorfrontal
gyrus,righto
ccipitalcortexand
leftprecuneusvolumeand
ODD/CDsymptom
severity
2.Fairchild
etal.2011a
,d90
(100
%male)
63N/A
N/A
N/A
2716–21
ROI;am
ygdala,insula,
anterior
cingulate
cortex,orbitofrontal
cortex
(VBM)
Wholebrain
(VBM)
Correlationalanalyses
Bilateral
(ROI)
Left(ROI)
Left dorsom
edial
ODD/CD-onlyversus
controls:
Bilateralcaudate,leftfusiform
gyrus,leftinferior
and
superior
occipitalcortex
ODD/CD-onlyversus
controls:
Negativecorrelationbetween
rightinsulaandCDsymptom
s
3.Stevensand
Haney-Caron
2012
a,d
72(71%
male)
24N/A
N/A
2424
12–18
Wholebrain(V
BM)
Correlationalanalyses
Left
nsns
ODD/CD-onlyversus
controls:
Leftinferiorfrontal,right
inferior/m
iddlefrontal,right
parahippocam
pal/fusiform
,paracentralcingulate
ODD/CD-onlyversus
controls:
Positiv
ecorrelationbetween
bilateralamygdala,bilateral
temporalcortexandright
lateralo
rbito
frontalv
olum
eandODD/CDsymptom
severity
4.Bussing
etal.2002
31 (74% male)
N/A
12N/A
N/A
198–12
Wholebrain(V
BM)
nsns
nsODD/CD+ADHDversus
controls:
Leftand
totalp
osterior
superior
verm
is,leftand
totalp
osterior
inferior
verm
is
5.Kruesi
etal.2004
20b,c
(90% male)
N/A
10N/A
N/A
10Patients:16.1
(3.6)
Controls:
15.9
(3.2)
Wholebrain(V
BM)
nsns
nsODD/CD+ADHDversus
controls:
Right
temporallobe
6.Sterzer
etal.2007a
,d24 (100
%male)
N/A
N/A
12 (5ODD/
CD-
only)
N/A
12Patients:12.8
(0.49)
Controls:12.5
(0.45)
ROI:am
ygdala,anterior
insula,anteriorcingulate
cortex,orbitofrontal
cortex
(VBM)
Left (ROI)
Bilateral
(ROI)
nsns ODD/CD-m
ixed
versus
controls:
Negativecorrelationbetweenleft
amygdalaandbilateralinsula
50 Neuropsychol Rev (2016) 26:44–72
Tab
le1
(contin
ued)
Study
Total
sample
size
(%male/
female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Resultsof
comparisons
betweenODD/CD-onlyandcontrolg
roup,
betweenODD/CD+ADHDandcontrolg
roup,and
betweenODD/CD-
mixed
andcontrolg
roup.
(brain
regionsreported
show
reducedvolumes
inthepatient
group,
unless
otherw
isereported)
ODD/
CD
-only
ODD/
CD+
ADHD
ODD/CD-
mixed
ADHD-
only
Controls
Amygdala
Insula
Prefrontal
cortex
Other
Wholebrain(V
BM)
Correlationalanalyses
volumeandODD/CDsymptom
severity
(ROI)
7.McA
lonan
etal.2007a
,d59
a,c
(100
%male)
N/A
28N/A
N/A
316–13
Wholebrain(V
BM)
nsns
nsODD/CD+ADHDversus
controls:
Midlinecerebellum,right
globus
pallidus,rightm
iddlefrontal
gyrus,rightsuperiorfrontal
gyrus,rightp
recuneus,left
inferior
parietalgyrus,left
superior
occipitalg
yrus
8.Huebner
etal.2008a
,d46
c
(100
%male)
N/A
N/A
23 (6ODD/
CD-
only)
N/A
2312–17
Wholebrain(V
BM)
Correlationalanalyses
Left
nsns
ODD/CD-m
ixed
versus
controls:
Greymatter,bilateralinferior
temporallobes,left
hippocam
pus,leftorbitofrontal
gyrus
Increasedbilateralcerebellar
volumein
ODD/CD-m
ixed
group
ODD/CD-m
ixed
versus
controls:
Negativecorrelationbetween
bilateralamygdalavolumeand
ODD/CDsymptom
severity
9.Sasayama
etal.2010a
,d35 (71% male)
N/A
10N/A
817
6–16
Wholebrain(V
BM)
Left
nsns
ODD/CD+ADHDversus
controls:
Bilateraltem
porallobes,bilateral
occipitallobes
10.F
airchild
etal.2013a
,d42
b,e (100
%female)
N/A
N/A
22 (203O
DD/
CD-
only)
N/A
2014–20
ROI;am
ygdala,anterior
insula,striatum,anterior
cingulatecortex,
orbitofrontalcortex
(VBM)
Wholebrain
(VBM)
Correlationalanalyses
nsBilateral
(ROI)
Right dorsolateral
ODD/CD-m
ixed
versus
controls:
Right
(ventral)striatum
(ROI),
righto
rbito
frontalcortex(ROI),
leftprecentralgyrus,rightm
id-
occipitalcortex,rightinferior
frontalg
yrus,leftp
recuneus,
middletemporalg
yrus
ODD/CD-m
ixed
versus
controls:
Negativecorrelationbetween
bilateralinsulavolumeand
psychopathictraits(ROI),
negativecorrelationbetween
both
bilateralinsulaandleft
striatum
volumeandcallous-
unem
otionaltraits
(ROI),
positivecorrelationbetween
bilateralm
iddle/superior
orbitofrontalcortexand
callous-unemotionaltraits
(ROI).
11.H
ummer
etal.2015
66c
1419
33N/A
3313–17
Wholebrain
(VBM)
nsns
nsns
Neuropsychol Rev (2016) 26:44–72 51
Tab
le1
(contin
ued)
Study
Total
sample
size
(%male/
female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Resultsof
comparisons
betweenODD/CD-onlyandcontrolg
roup,
betweenODD/CD+ADHDandcontrolg
roup,and
betweenODD/CD-
mixed
andcontrolg
roup.
(brain
regionsreported
show
reducedvolumes
inthepatient
group,
unless
otherw
isereported)
ODD/
CD
-only
ODD/
CD+
ADHD
ODD/CD-
mixed
ADHD-
only
Controls
Amygdala
Insula
Prefrontal
cortex
Other
(73% male)
(14ODD/
CD-
only)
12.M
ichalska
etal.2015
111b
(48% male)
N/A
43N/A
N/A
689–11
Wholebrain
(VBM)
Correlationalanalyses
nsns
nsns ODD/CD+ADHDversus
controls:
Negativecorrelationbetweenleft
superior
temporalsulcus
volumeandCDsymptom
s
Groupsdidnotdifferinterm
sof
gender,IQ,orsocioeconom
icstatus,unlessstated
otherw
ise.ADHDAttentionDeficitHyperactiv
ityDisorder;CDConductDisorder;N/A
notapplicable;nsno
significant
results
arereported
forthisregion;O
DDOppositionalDefiant
Disorder;ROIregion
ofinterest;V
BM
voxel-basedmorphom
etry
aIncluded
nodataon
medication-useor
didnotadjustfor
effectsof
medication-use
bGroupsdiffered
onIQ
,orincluded
nodataon
IQcParticipantswith
held
stim
ulantm
edicationfor>24
hpriorto
scanning
dResultsincluded
inALEanalysis
eGroupsdiffered
onsocio-econom
icstatus
52 Neuropsychol Rev (2016) 26:44–72
ODD/CD-mixed and ODD/CD+ADHD (9 studies). Two ofthe four studies investigating total grey matter in ODD/CD-mixed and ODD/CD+ADHD samples reported smaller vol-umes in the diagnostic groups, compared to controls. In ac-cordance with results from the studies in the ODD/CD-onlygroup, three of the nine studies in ODD/CD-mixed and ODD/CD+ADHD groups provided support for abnormalities in theleft amygdala. Additionally, abnormal volumes were reportedin the ODD/CD-mixed and ODD/CD+ADHD groups forstructures that were not found for ODD/CD-only groups,which might indicate that these abnormalities reflect the pres-ence of comorbid ADHD rather than ODD/CD. These struc-tures with smaller volumes were the temporal lobe (four out ofnine studies: one left, one right, two bilateral) and the cerebel-lum (three out of nine studies: one left, two bilateral).
Correlational findings and specificity. A total of sevenstudies investigated relations between grey matter abnormali-ties and ODD/CD symptoms. Three out of seven studies re-ported associations for the amygdala, insula and temporal cor-tex. The three studies that reported associations between theamygdala and ODD/CD symptoms, showed either a positiveassociation (thus a larger volume related to more ODD/CDsymptoms) for the amygdala in an ODD/CD-only sample, ornegative associations (thus a smaller volume related to moreODD/CD symptoms) for the bilateral or left amygdala inODD/CD-mixed samples. For the insula, all three studies re-ported negative associations, one between the right insula andODD/CD symptoms in an ODD/CD-only sample and twobetween the insula and ODD/CD symptoms in ODD/CD-mixed samples. For the three studies that found associationsbetween the superior temporal cortex and ODD/CD symp-toms, one study reported a positive association for the superior
temporal cortex in an ODD/CD-only sample, while the othertwo studies reported a negative association for the right supe-rior temporal cortex in an ODD/CD-only sample and for theleft superior temporal cortex in an ODD/CD+ADHD sample.
In terms of specificity of abnormalities for ODD/CD ascompared to ADHD, one study reported grey matter reduc-tions in the amygdala and frontal gyrus to be specific forODD/CD-only, since these were not present in the ADHD-only group. In addition, another study looked at disorder spec-ificity by covarying for ODD/CD in an ODD/CD-mixedgroup and reported that the grey matter reductions in theamygdala turned non-significant, suggesting that these greymatter reductions were specific for ODD/CD.
Structural Summary Combining the results from the meta-analysis and the narrative review, structural neuroimagingstudies most consistently implicated abnormalities of the leftamygdala, insula and left frontal gyrus in ODD/CD.
Functional Neuroimaging Findings
General Overview Tables 2 and 3 shows a total of 17 studiesthat investigated brain activity correlates in individuals withODD/CD-only (seven studies), ODD/CD-mixed (nine studies),or ODD/CD+ADHD (one study) compared with typically de-veloping individuals. Of the 17 included studies, 14 conducted aWBA approach to investigate regional brain differences, ofwhich 5 also conducted ROI analyses (S.F. White et al. 2013;Finger et al. 2011; Herpertz et al. 2008; Passamonti et al. 2010;Cohn et al. 2014). In three studies only ROI analyses werereported (Gatzke-Kopp et al. 2009; Marsh et al. 2008; Sterzeret al. 2005). Seven studies investigated the association between
Fig. 2 Results of the structuralALE meta-analysis showing theclusters in left amygdala, insula,and left medial/superior frontalgyrus (pcorrected < 0.05)superimposed on a structural scanin Talairach space. Top row: cor-onal view, bottom row: axial view.L Left, R Right
Neuropsychol Rev (2016) 26:44–72 53
Tab
le2
Summaryof
studycharacteristicsandresults
forfunctio
nalimaging–Hot
EF
Study
Totalsam
ple
(%male/female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Task
andconditions
Resultsof
comparisons
betweenODD/
CD-onlyandcontrolg
roup,between
ODD/CD+ADHDandcontrolg
roup,
andbetweenODD/CD-m
ixed
and
controlg
roup.(brainregionsreported
show
reducedvolumes
inthepatient
group,unless
otherw
isereported)
ODD/
CD-
only
ODD/
CD+
ADHD
ODD/CD
-mixed
ADHD-
only
Controls
13.R
ubia
etal.
2009ba
,b
48c,d (100
%male)
14N/A
N/A
1816
9–16
Wholebrain
Contin
uous
Perform
ance
Task
-Rew
ard
-Non-rew
ard
ODD/CD-onlyversus
controls:
Behavioural:N
osignificantg
roup
differences,allg
roupsshow
edequally
enhanced
performance
asa
resultof
reward(versusnon-reward
trials)
Rew
ard>non-reward:
reducedactiv
ityin
rightlateralandmedial
orbitofrontalcortex
14.K
alnin
etal.2011
44c,d,e
(59%
male)
22N/A
N/A
N/A
2213–17
Wholebrain
EmotionalS
troop
task
-Violent
words
-Non-violent
words
ODD/CD-onlyversus
controls:
Behavioural:N
osignificantg
roup
differences
Nosignificantg
roup
differences
15.M
arsh
etal.2013a
35c,d,f
(66%
male)
14N/A
N/A
N/A
2110–17
Wholebrain
Correlatio
nal
analyses
Viewingof
pictures
with
threepain
intensities
(severe/
mo-derate/none)
during
two
conditions:pain
appliedto
subject
him/herselfor
tosomeone
else
ODD/CD-onlyversus
controls:
Behavioural:N
osignificantg
roup
differences
Group
differences
forthemaineffectof
pain
view
ing:
reducedactiv
ityin
left
medialfrontalgyrus,rostralanterior
cingulatecortex,right
putamen
Other’spain
<ow
npain:reduced
activ
ityin
leftam
ygdala/uncus,left
superior
frontalg
yrus,right
insula
ODD/CD-onlyversus
controls:
Negativecorrelationbetweenleft
amygdalaactiv
ationandaggressive
behaviour,negativ
ecorrelation
betweenleftanterior
cingulatecortex
activ
ationandaggressive
behaviour
16.S
terzer
etal.2005
27d,f (100
%male)
N/A
N/A
13 (5ODD/
CD-only)
N/A
149–15
ROI:am
ygdala,
hippocam
pus,
orbitofrontal
cortex,anterior
cingulate
cortex
Correlatio
nal
analyses
Passiveview
ingtask
ofneutraland
negativ
epictures.
ODD/CD-m
ixed
versus
controls:
Behavioural:low
ered
arousalratings
fornegativ
epictures
andlowered
valenceandarousalratings
for
neutralp
icturesin
theCD-m
ixed
groupcomparedto
controls
Negative>neutralp
ictures:reduced
activ
ityin
leftam
ygdala(ROI),right
dorsalanterior
cingulate(ROI)
54 Neuropsychol Rev (2016) 26:44–72
Tab
le2
(contin
ued)
Study
Totalsam
ple
(%male/female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Task
andconditions
Resultsof
comparisons
betweenODD/
CD-onlyandcontrolg
roup,between
ODD/CD+ADHDandcontrolg
roup,
andbetweenODD/CD-m
ixed
and
controlg
roup.(brainregionsreported
show
reducedvolumes
inthepatient
group,unless
otherw
isereported)
ODD/
CD-
only
ODD/
CD+
ADHD
ODD/CD
-mixed
ADHD-
only
Controls
ODD/CD-m
ixed
versus
controls:
Negativecorrelationbetweenboth
left
amygdalaandrightanteriorcingulate
cortex
activ
ationandaggressive
behaviour(negativeversus
neutral
contrast)
17.H
erpertz
etal.2008a
57e
(100
%male)
N/A
N/A
22 (6ODD/
CD-only)
1322
12–17
ROI:am
ygdala,
anterior
cingulate
cortex,
orbitofrontal/
medialfrontal
cortex,insula
Wholebrain
Passiveview
ingtask
Includingpositiv
e,negativ
e,neutral
pictures
ODD/CD-m
ixed
versus
controls:
Behavioural:C
D-m
ixed
evaluated
negativ
eandpositiv
epictures
asless
arousing,and
positiv
epictures
asless
pleasant
comparedto
controls
Negative>neutralp
ictures:increased
activ
ityin
leftam
ygdala(ROI)
18.F
inger
etal.2008a
42c,d,e(67%
male)
N/A
N/A
14 (4ODD/
CD-only)
1414
10–17
Wholebrain
Rew
ardedreversal
learning
Task
-Pu
nished
reversal
errors
-Correctrewarded
responses
ODD/CD-m
ixed
versus
controls:
Behavioural:N
osignificantg
roup
differences
Group
differences
foroverallresponse
type
(rew
ardandpunishment):
increasedactiv
ityin
leftprecuneus,
rightsuperiorfrontalg
yrus
Punishedreversal
errors
>correct
rewardedresponses:increased
activ
ityin
ventromedialp
refrontal
cortex,right
caudate
19.M
arsh
etal.2008
36e
(58%
male)
N/A
N/A
12 (5ODD/
CD-only)
1212
10–17
ROI:am
ygdala
Correlatio
nal
analyses
Implicitprocessing
task:categorise
gender
offearful,
angry,andneutral
faces
ODD/CD-m
ixed
versus
controls:
Behavioural:N
osignificantg
roup
differences
Fearful
>neutralp
ictures:reduced
activ
ityin
bilateralamygdala(ROI)
ODD/CD-m
ixed
versus
controls:
Negativecorrelationbetween
connectiv
itybetweenam
ygdalaand
rightv
entrom
edialp
refrontalcortex,
andODD/CDsymptom
severity
20.G
atzke-
Kopp
etal.2009
30d,e
(100
%male)
N/A
N/A
19 (3ODD/
CD-only)
N/A
1112–16
ROI:anterior
cingulate
cortex,
Monetaryincentive
task
-Rew
ard
ODD/CD-m
ixed
versus
controls:
Behavioural:N
osignificantg
roup
differences
Neuropsychol Rev (2016) 26:44–72 55
Tab
le2
(contin
ued)
Study
Totalsam
ple
(%male/female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Task
andconditions
Resultsof
comparisons
betweenODD/
CD-onlyandcontrolg
roup,between
ODD/CD+ADHDandcontrolg
roup,
andbetweenODD/CD-m
ixed
and
controlg
roup.(brainregionsreported
show
reducedvolumes
inthepatient
group,unless
otherw
isereported)
ODD/
CD-
only
ODD/
CD+
ADHD
ODD/CD
-mixed
ADHD-
only
Controls
caudate,
putamen
-Non-rew
ard
Non-rew
ard>reward:
reducedactiv
ityin
bilateralanteriorcingulate(ROI)
increasedactiv
ityin
bilateralstriatum
(caudate)(ROI)
21.P
assamonti
etal.2010a
75 (100
%male)
N/A
N/A
52 (43ODD/
CD-only)
N/A
2316–21
ROI: ventromedial
prefrontal
cortex,
amygdala,
insula,
orbitofrontal
cortex
Wholebrain
Correlatio
nal
analyses
Implicitprocessing
task:categorise
gender
ofangry,
sadandneutral
faces
ODD/CD-m
ixed
versus
controls:
Behavioural:N
osignificantg
roup
differences
Angry
>neutralp
ictures:reduced
activ
ityin
bilateralamygdala(ROI),
leftinsula(ROI),right
ventromedial
prefrontalcortex
(ROI),b
ilateral
dorsolateralprefrontalcortex,
bilaterald
orsomedialp
refrontal
cortex,bilateralo
rbito
frontalcortex
(ROI),right
inferior
parietalcortex,
bilateralinferiortemporalg
yrus,left
fusiform
gyrus,rightm
iddle
temporalg
yrus,right
superior
temporalsulcus/gyrus,bilateral
thalam
us,leftp
utam
en,left
cerebellu
mSad>neutralp
ictures:reducedactiv
ityin
bilateralamygdala(ROI),
ventromedialp
refrontalcortex
(ROI),leftd
orsolateralp
refrontal
cortex,putam
en,right
cerebellu
m,
bilateralsuperiortemporalsulcus/
gyrus
ODD/CD-m
ixed
versus
controls:
Negativecorrelationbetweenright
amygdalaandCDsymptom
s(ROI),
negativ
ecorrelationbetween
ventromedialp
refrontalcortexand
CDsymptom
s(ROI),negative
correlationbetweenleftinsulaand
CDsymptom
s(ROI)
22.F
inger
etal.2011a
30f
(60%
male)
N/A
N/A
15N/A
15Patients:
14,1(1,8)
Passiveavoiding
learning
task
ODD/CD-m
ixed
versus
controls:
56 Neuropsychol Rev (2016) 26:44–72
Tab
le2
(contin
ued)
Study
Totalsam
ple
(%male/female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Task
andconditions
Resultsof
comparisons
betweenODD/
CD-onlyandcontrolg
roup,between
ODD/CD+ADHDandcontrolg
roup,
andbetweenODD/CD-m
ixed
and
controlg
roup.(brainregionsreported
show
reducedvolumes
inthepatient
group,unless
otherw
isereported)
ODD/
CD-
only
ODD/
CD+
ADHD
ODD/CD
-mixed
ADHD-
only
Controls
(5ODD/
CD-only)
Controls:
13,2
(1,1)
ROI:am
ygdala,
orbitofrontal
cortex,striatum
Wholebrain
-Rew
ardedcorrecth
its-Pu
nished
commission
errors
Behavioural:O
DD/CD-m
ixed
made
morecommission
errorsthan
controlsduring
thelatelearning
phase(allsevenblocks
afterthefirst
block).
Overallresponse
type
(rew
ardand
punishment):reduced
activ
ityin
right
amygdala(ROI),bilateralsuperior
frontalg
yrus,leftinsula,rightm
edial
frontalg
yrus,leftm
iddlefrontal
gyrus,leftsuperior
parietallobule,
leftsuperior
temporalg
yrus,left
lingualgyrus,rightfusiform
gyrus,
leftcaudate(ROI),right
thalam
usRew
ardedcorrecth
its>punished
commission
errors:reducedactiv
ityin
righto
rbito
frontalcortex(ROI),
leftmiddlefrontalg
yrus,
parahippocam
palg
yrus
23.W
hite
etal.2013a
38c,d,f,g(Patients:
82%
maleControls:
56%
male)
N/A
N/A
20 (16ODD/C
D-on-
ly)
N/A
18Patients:
15,2(2,0)
Controls:
14,9
(2,2)
ROI:am
ygdala,
ventromedial
prefrontal
cortex,caudate
Wholebrain
Passiveavoiding
learning
task
-Rew
ard
-Pu
nishment
ODD/CD-m
ixed
versus
controls:
Behavioural:a
smallerproportio
nof
the
ODD/CD-m
ixed
groupthan
ofthe
controlsshow
edan
association
betweenexpected
valueandchoice
behaviour
Receiving
rewarding
feedback
modulated
bypredictio
nerror:
reducedactiv
ityin
leftcaudate(ROI)
Receiving
punishingfeedback
modulated
bypredictio
nerror:
increasedactiv
ityin
leftcaudate
(ROI)
24.W
hite
etal.2014a
30d,f(Patients:73
%maleControls:
66%
male)
N/A
N/A
15 (8ODD/
CD-only)
N/A
1510–17
Wholebrain
Reinforcement
learning
task
-Rew
ard(appetitive
image)
-Pu
nishment
(aversiveim
age)
ODD/CD-m
ixed
versus
controls:
Behavioural:O
DD/CD-m
ixed
were
less
likelyto
avoidphysicalthreat,
butn
otcontam
inationthreat,stim
uli
than
controls
Neuropsychol Rev (2016) 26:44–72 57
Tab
le2
(contin
ued)
Study
Totalsam
ple
(%male/female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Task
andconditions
Resultsof
comparisons
betweenODD/
CD-onlyandcontrolg
roup,between
ODD/CD+ADHDandcontrolg
roup,
andbetweenODD/CD-m
ixed
and
controlg
roup.(brainregionsreported
show
reducedvolumes
inthepatient
group,unless
otherw
isereported)
ODD/
CD-
only
ODD/
CD+
ADHD
ODD/CD
-mixed
ADHD-
only
Controls
Receiving
rewarding
feedback
modulated
bypredictio
nerror:
reducedactiv
ityin
rightinferior
parietalcortex
Receiving
punishingfeedback
modulated
bypredictio
nerror:
increasedactiv
ityin
rightinferior
parietalcortex
25.C
ohn
etal.2014a
68 (79%
male)
N/A
N/A
45 (18ODD/C
D-on-
ly)
N/A
2317,7(1,6)
ROI:am
ygdala,
ventral
striatum
,medial
prefrontal
cortex
Wholebrain
Correlatio
nal
analyses
Monetaryincentive
delaytask
-Rew
ard
-Neutral
-Loss
ODD/CD-m
ixed
versus
controls:
Behavioural:N
osignificantg
roup
differences
Rew
ardhit>
rewardmiss(rew
ard-
feedback):reducedactiv
ityin
right
ventralstriatum
(ROI)
Loss
miss>hit(loss-feedback):
increasedactiv
ityin
rightamygdala
(ROI)
ODD/CD-m
ixed
versus
controls:
Negativecorrelationbetweenleft
amygdalaandCUtraits(ROI)
Nocorrelationbetweenpsychopathic
traitsandneuralresponses
Participantsdidnotd
ifferin
term
sof
gender,IQ,orsocioeconomicstatus,unlessstated
otherw
ise.The
italic
text
describesthecontrastforwhich
theresults
arereported.
ADHDAttentionDeficitHyperactiv
ityDisorder;CDConductDisorder;N/A
notapplicable;O
DDOppositionalDefiant
Disorder;ROIregion
ofinterest
aResultsincluded
inALE
bStudyincluded
inboth
hotand
cool
EFnarrativereview
sections
cGroupsdiffered
ongender
dGroupsdiffered
onIQ
measures
eParticipantswith
held
stim
ulantm
edicationfor>24
hpriorto
scanning
procedure,or
didnotu
semedication
fParticipantsdidnotw
ithheld
stim
ulantm
edication,butadditionalanalysisdidnotshowan
effectof
medicationuseon
outcom
emeasures
gGroupsdiffered
onsocio-econom
icstatus
58 Neuropsychol Rev (2016) 26:44–72
Tab
le3
Summaryof
Study
CharacteristicsandResultsforFunctionalImaging–CoolE
F
Study
Totalsam
ple
(%male/female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Task
andconditions
Resultsof
comparisons
between
ODD/CD-onlyandcontrolg
roup,
betweenODD/CD+ADHDand
controlg
roup,and
betweenODD/
CD-m
ixed
andcontrolg
roup.(brain
regionsreported
show
reduced
volumes
inthepatient
group,unless
otherw
isereported)
ODD/CD-
only
ODD/CD+
ADHD
ODD/CD-
mixed
ADHD-
only
Controls
26.R
ubia
etal.2008
48–53
(100
%male)
13–14
N/A
N/A
14–20
16–20
9–17
Wholebrain
Stop
Task
-Failedstop
-Successfulstop
-Go
ODD/CD-onlyversus
controls:
Behavioural:N
osignificantg
roup
differences
Failedstops>go
trials:reduced
activ
ityin
rightp
osterior
cingulategyrus,rightp
recuneus,
leftparietalcortex
Go>successful
stop
trials:reduced
activ
ityin
anterior
cingulate
gyrus,insula,caudate,putam
en,
thalam
us,leftsuperiortemporal
cortex,premotor
cortex
27.R
ubia
etal.2009a
Sim
onTask
-Successfulcongruent
-Successfulincongruent
-Successfulo
ddball
Behavioural:,ODD/CDmademore
errorscomparedto
controls,no
differencesbetweentheADHD-
only
groupandcontrols
Successful
incongruent>
successful
oddballtrials(interference
inhibitio
n):reducedactiv
ityin
rightm
iddleandsuperior
temporallobe,rightp
recuneus
13.R
ubia
etal.2009b
aContin
uous
Performance
-Non-rew
ard
-Non-target
Behavioural:N
osignificantg
roup
differences
Non-rew
ard>non-targettrials
(sustained
attention):reduced
activ
ityin
rightinsula,right
hippocam
pus,rightanterior
cingulatecortex,cerebellum,right
thalam
us,lefto
ccipitalg
yrus,left
posteriorcingulate,leftprecuneus
28.R
ubia
etal.2010
Correlatio
nalanalyses
(2outo
f4studies)
Switching
Task
-Switch
-Repeat
Behavioural:N
osignificantg
roup
differences,buto
verall
performance
was
poorer
during
thesw
itchtrialsthan
repeattrials
Group
differences
foroveralltask
performance:reducedactiv
ityin
rightinferiorparietallobe,right
Neuropsychol Rev (2016) 26:44–72 59
Tab
le3
(contin
ued)
Study
Totalsam
ple
(%male/female)
Num
berof
subjects(specified
pergroup)
Age
(years,
rangeor
M(SD))
Analysismethod
Task
andconditions
Resultsof
comparisons
between
ODD/CD-onlyandcontrolg
roup,
betweenODD/CD+ADHDand
controlg
roup,and
betweenODD/
CD-m
ixed
andcontrolg
roup.(brain
regionsreported
show
reduced
volumes
inthepatient
group,unless
otherw
isereported)
ODD/CD-
only
ODD/CD+
ADHD
ODD/CD-
mixed
ADHD-
only
Controls
precentralgyrus,left-superior
temporal/inferiorparietalcortex,
leftprecuneus,cuneus
ODD/CD-onlyversus
controls:
Negativecorrelationbetween
dorsolateralprefrontalcortex
and
conductp
roblem
s(1
study)
29.Z
huetal.2014
21b
(100
%male)
11N/A
N/A
N/A
1010–12
Wholebrain
GoS
topTask
-Com
pletetask
ODD/CD-onlyversus
controls:
Behavioural:O
DD/CD-onlyshow
edhigher
errorrateduring
response
inhibitio
nandalonger
stop
latency
Group
differences
foroveralltask
performance:reducedactiv
ityin
rightinferiorfrontalg
yrus,
increasedactiv
ityin
dorsolateral
partsof
bilateralinferiorfrontal
frontalg
yrus,leftm
iddlefrontal
gyrus,rightsuperiorfrontalg
yrus
Participantsdidnotd
ifferin
term
sof
gender,IQ,orsocioeconomicstatus,unlessstated
otherw
ise.The
italic
text
describesthecontrastforwhich
theresults
arereported.
ADHDAttentionDeficitHyperactiv
ityDisorder;CDConductDisorder;N/A
notapplicable;O
DDOppositionalDefiant
Disorder;ROIregion
ofinterest
aStudyincluded
inboth
hotand
cool
EFnarrativereview
sections
bParticipantswith
held
stim
ulantm
edicationfor>24
hpriorto
scanning
procedure,or
didnotu
semedication
60 Neuropsychol Rev (2016) 26:44–72
activity in brain areas and ODD/CD symptom counts (A.A.Marsh et al. 2013; Sterzer et al. 2005; Passamonti et al. 2010;Cohn et al. 2014; Finger et al. 2008; Rubia et al. 2008, 2010).
We reviewed the studies using the framework of hot and coolEF in ODD/CD, which resulted in 13 studies in the hot EFsection and 5 studies in the cool EF section, of which 1 studyappeared in both sections (Rubia et al. 2009b). The 13 studies inthe hot EF section included groups with ODD/CD-only (threestudies), ODD/CD-mixed (nine studies) or ODD/CD+ADHD(one study). In all studies, these groups were compared with atypically developing group and in four studies these groupswere additionally compared with an ADHD-only group(Finger et al. 2008; Herpertz et al. 2008; Marsh et al. 2008;Rubia et al. 2009b). The five studies in the cool EF section onlyincluded individuals with ODD/CD-only that were compared toa typically developing group and in four of the studies the ODD/CD-only groupwas additionally comparedwith anADHD-onlygroup (Rubia et al. 2008, 2009a, b, 2010). Because four of thefive cool EF studies were published by the same research groupand included largely overlapping samples (Rubia et al. 2008,2009a, b, 2010), we did not perform an ALE meta-analysis oncool EF fMRI studies in ODD/CD as at least five independentsamples are required for this analysis (www.brainmap.org).
All studies investigated children/adolescents. For informa-tion regarding group characteristics of the included samples,see Tables 2 and 3. Reported differences in brain structures arebilateral, unless reported to be either left or right sided. Resultsare reported fromWBA approaches, unless specified to be theresult of ROI based analyses, see Tables 2 and 3.
Hot EF – ALE Meta-analysis Thirteen functional neuroim-aging studies of hot EF were initially included in the meta-
analysis. Due to the limited number of studies in ODD/CD-only (three studies) it was not possible to perform an ALEmeta-analysis on studies comparing ODD/CD-only groupsand controls, thus the ALE meta-analysis was performed onall ODD/CD samples, therewith including both individualswith and without comorbid ADHD.
There was a difference between the resulting maps of theALE analysis with and without the ROI-only studies.Therefore, we reported the results from the map that includedonly WBA approach-based studies given the unbiased results.Total sample sizes for all included studies ranged between 30and 75 individuals, adding up to a total of 423 subjects, of which256were patients and 167were controls (age range 9–21 years).Three studies assessed a full male sample and six studiesassessed a sample largely consisting of males (ranging from56 to 79 %). The studies reported on different tasks associatedwith hot EF functions: three studies assessed processing of con-tingencies, two studies assessed passive viewing of emotionalpictures, two studies assessed passive avoidance learning, onestudy assessed active viewing and rating of painful situations,and one study assessed implicit emotional processing.
The included studies provided a total of 68 foci of areasshowing altered activity in individuals with ODD/CD withand without comorbid ADHD. The ALE analysis revealedfive significant clusters representing areas with altered activityin patients withODD/CD-only or ODD/CD+ADHD that weremost consistently reported across the studies (see Fig. 3 forvisualisation). The largest cluster (416 mm3) was located inthe right globus pallidus, with two foci inside this cluster. Thesecond (328 mm3) and third (256 mm3) clusters were locatedin the right and left amygdala, and contained three and twofoci, respectively. The fourth cluster (208mm3) was present in
Fig. 3 Results of the functionalALE meta-analysis of hot EFshowing the clusters in rightglobus pallidus, bilateralamygdala, left caudate, and leftfusiform gyrus (pcorrected < 0.05)superimposed on a structural scanin Talairach space. Top row:coronal view, bottom row: axialview. L Left, R Right
Neuropsychol Rev (2016) 26:44–72 61
the left caudate, while the fifth cluster (200 mm3) was locatedin the left fusiform gyrus, both contained two foci.
Hot EF – Narrative Review Tables 2 and 3 provides anoverview of investigated samples and approach (WBA orROI) and summarizes the findings of the 13 included studiesreporting on hot EF. Sample sizes ranged between 27 and 75individuals, adding up to a total of 560 individuals, including334 patients and 226 controls (age range 9–21 years). Thesamples consisted predominantly of males, with five studiesusing a full male sample, and the other eight studies using asample largely consisting of males (ranging from 56 to 82 %).The studies reported on different tasks associated with hot EFfunctions: five studies assessed processing of contingencies,three studies assessed passive viewing of emotional pictures,two studies assessed passive avoidance learning, one studyassessed active viewing and rating of painful situations, onestudy assessed implicit emotional processing, and one studyassessed interference control with emotional stimuli. Of the 13studies, 10 used a WBA approach and 3 only an ROI ap-proach, resulting in different numbers of studies per investi-gated structure.
ODD/CD-only versus controls (3 studies). Two of the threestudies reported abnormalities for ODD/CD-only compared tocontrols, of which one study reported lower activity in theright striatum, left amygdala, right insula, anterior cingulatecortex, and left medial and superior frontal gyrus, and theother study reported lower activity in the right lateral andmedial orbitofrontal cortex.
ODD/CD-mixed and ODD/CD+ADHD (10 studies).Studies assessing ODD/CD-mixed and ODD/CD+ADHDgroups replicated findings of abnormal function of the stria-tum (six out of eight studies: three left, two right, one bilater-al), including the caudate and putamen, and the amygdala (sixout of nine studies: two left, two right, two bilateral).Additionally, functional abnormalities were reported inODD/CD-mixed groups for the parietal cortex (three out ofseven studies: two right, one left), which was not found forODD/CD-only groups. This might indicate that the abnormalactivity in the parietal cortex reflects the presence of comorbidADHD rather than ODD/CD.
Correlational findings and specificity. Five studies investi-gated relations between activity of specific structures andODD/CD related symptoms. All five studies reported negativeassociations (thus a smaller volume related to more ODD/CDsymptoms) between amygdala activity and ODD/CD relatedsymptoms, one for an ODD/CD-only sample (left) and fourfor ODD/CD-mixed samples (two left, one right, one bilater-al). Furthermore, two of the five studies reported negativeassociations between left and right anterior cingulate cortexactivity and ODD/CD related symptoms, of which oneassessed an ODD/CD-only sample and the other assessed anODD/CD-mixed sample, respectively.
In terms of specificity, four studies investigated disorderspecificity of the abnormalities compared to ADHD-onlygroups. Of these four studies, one study reported the loweractivity in the orbitofrontal cortex in an ODD/CD-only groupto be disorder specific, while the other study reported loweractivity in the amygdala in an ODD/CD-mixed group to bedisorder specific.
Hot EF – Summary Combining the results from the meta-analysis and the narrative review, hot EF functional neuroim-aging studies most consistently implicated abnormalities ofthe amygdala and insula. Even though the meta-analysis im-plicated the left fusiform gyrus in ODD/CD, the narrativereviews did not show support for abnormalities in this areain ODD/CD.
Cool EF – Narrative Review Tables 2 and 3 provides anoverview of investigated samples and approach (WBA orROI) and summarizes the findings of the five included studiesreporting on cool EF. Sample sizes ranged between 21 and 53individuals, with all studies assessing full male samples (agerang 9–17 years). The studies reported on different tasks as-sociated with cool EF functions: one study assessed inhibitorycontrol, one study assessed interference control, one studyassessed attention allocation, one study assessed continuousperformance, and one study assessed cognitive flexibility.
ODD/CD-only versus controls (5 studies). Four of the fivestudies reported less activation in the precuneus of ODD/CD-only groups, with two studies reporting the left precuneus, onethe bilateral precuneus, and one the right precuneus.Furthermore, two of the five studies reported lower activityof the right insula, right anterior cingulate and right posteriorcingulate cortex in ODD/CD-only.
Correlational findings and specificity. Two studies investi-gated relations between activity of specific structures andODD/CD related symptoms, of which one reported a negativeassociation between dorsolateral prefrontal cortex and ODD/CD related symptoms and the other reported no associations.In terms of specificity, four studies investigated disorder spec-ificity of the abnormalities for ODD/CD as compared toADHD-only. One of these four studies reported abnormal ac-tivity in the insula, anterior cingulate cortex, cerebellum andhippocampus to be specific for ODD/CD-only, while anotherstudy reported specificity for lower activity in the left parietal-temporal cluster and the right parietal lobe.
Cool EF – Summary The number of available studies intocool EF did not allow a meta-analysis to be conducted. Basedon the narrative review, abnormalities in the precuneus weremost consistently implicated in ODD/CD.
62 Neuropsychol Rev (2016) 26:44–72
Integration of Quantitative and Qualitative ReviewFindings
In order to determine which hot and cool EF related brainareas are implicated in ODD/CD, findings from both sMRIand fMRI meta-analyses as well as from the narrative reviewsare integrated. Since the meta-analyses provide the strongestevidence for the convergence of findings, the main findings ofthe meta-analyses are discussed first. Second, findings fromODD/CD-only studies (10) are discussed, since these are notbiased by comorbid ADHD, followed by findings from theODD/CD-mixed and ODD/CD+ADHD studies (19). Finally,all these findings are integrated, resulting in an overall con-clusion for neuroanatomical abnormalities related to ODD/CD.
Table 4 provides an overview and integration of all report-ed abnormalities. For the overall conclusion, priority was giv-en to results from the ALE meta-analyses, since these providethe most objective and strong evidence for abnormalities of acertain structure. Qualifications used in the overall conclusionencompassed ‘strong evidence’, ‘some evidence’, ‘weak evi-dence’ and ‘no evidence’. For the qualification of ‘strong ev-idence’, the structure had to be reported in both of the ALEmeta-analyses. Alternatively, the structure had to be reportedin either one of the ALEmeta-analyses, while being supportedby at least half of the studies included in the narrative reviews(structural and functional) that investigated that structure. Forthe qualifications of ‘some evidence’ and ‘weak evidence’ atleast a quarter of the studies included in the narrative review(structural and functional) that investigated that structure hadto report abnormalities, while for ‘some evidence’ the reportedabnormality needed additional support of one of the ALEmeta-analyses. Finally, when less than a quarter of the studiesthat investigated that structure reported abnormalities, it wasconcluded that there was ‘no evidence’ for involvement of thatstructure in ODD/CD.
structures: the amygdala (left and bilateral, respectively).Other hot EF related areas were only implicated in the struc-tural (insula, left frontal gyrus) or functional meta-analyses(striatum, left fusiform gyrus).
Narrative Review: ODD/CD-Only Versus Controls (10Studies) Integration of the structural and functional findingsimplicated abnormalities in several hot EF related structuresfor ODD/CD, since these structures were reported in morethan a quarter of the studies: the amygdala, right striatum,insula, and left frontal gyrus (see Table 4). For a range ofstructures there was only weak or no evidence for abnormal-ities in ODD/CD (see Table 4 for an overview).
Overall Conclusion and Specificity of Findings Integrationof the structural and functional findings across the ODD/CD-only, ODD/CD-mixed and ODD/CD+ADHD samples pro-vided strong evidence for abnormalities in structure and func-tion of two hot EF related structures: the amygdala and thestriatum (see Table 4). Additionally, there was some evidencefor structural and functional abnormalities in two other hot EFrelated structures: the insula and the left frontal gyrus (seeTable 4).
Further support for involvement of abnormalities of someof these structures in ODD/CD was established by comparingthe results from studies using ODD/CD-only samples with theresults from studies using ODD/CD-mixed and ODD/CD+ADHD samples to see if results were mainly driven byODD/CD, and reviewing the correlational and specificityfindings. The involvement of the amygdala in ODD/CD wasfurther supported by the finding that abnormalities were main-ly driven by the studies including ODD/CD-only or ODD/CD-mixed groups, and to a lesser extent by studies includingODD/CD+ADHD groups. Hence, it seems that abnormalitiesare related to ODD/CD and not to comorbid ADHD.Furthermore, there was strong evidence from studies on spec-ificity of abnormalities in the amygdala for ODD/CD as com-pared to ADHD-only, and there was strong evidence for anassociation between both amygdala structure and function andODD/CD related symptoms. Similar evidence was reportedfor the insula, where again the results were mainly driven bystudies including ODD/CD-only or ODD/CD-mixed groups,rather than by studies including ODD/CD+ADHD groups.Also, some evidence for an association between both insulastructure and function and ODD/CD related symptoms waspresent. For the striatum there was some evidence for an as-sociation between abnormalities in this area and ODD/CDrelated symptoms. For the other areas (frontal and fusiformgyrus) there was no further supporting evidence from studiesinvestigating specificity by comparing anODD/CD-only sam-ple with an ADHD-only sample, nor did correlational ap-proaches provide evidence for associations between abnor-malities in these areas and ODD/CD symptoms.
Neuropsychol Rev (2016) 26:44–72 63
Meta-analyses Meta-analytic results implicated both struc-tural and functional abnormalities for one of the hot EF related
Narrative Review: ODD/CD-Mixed and ODD/CD+ADHD (19 Studies) Overall, findings from the ODD/CD-mixed and ODD/CD+ADHD groups again showed ab-normalities in several hot EF related structures for ODD/CD:the amygdala, right striatum, insula and left frontal gyrus (seeTable 4). For the amygdala and right striatum at least half ofthe structural and functional studies investigating those struc-tures reported abnormalities, and for the insula and left frontalgyrus more than a quarter of the structural and functionalstudies investigating those structures reported abnormalities.Again, there was only weak or no evidence for involvement ofa range of structures (see Table 4 for an overview).
Tab
le4
Overviewof
AllReportedStructuresandMeta-analyticandNarrativ
eReviewFindings
ALEMeta-Analysis
Abnormalities
reported
forthestudiesincludingsamples
ofODD/CD-onlyversus
Controls
Abnormalities
reported
forthestudiesincludingsamples
ofODD/CD-
mixed
andODD/CD+ADHDversus
Controls
Conclusion
ODD/CD
versus
controls
Structural
Functio
nal
Structural
Functional
Structural+
unctional
Significant
associations
bStructural
Functio
nal
Structural+
functio
nal
Significant
associations
e
Hot
Coola
Hot
Coola
(n=8)
(n=9)
(n=3,all
WBA)
(n=3,all
WBA)
(n=5,all
WBA)
(n=11,all
WBA)
(n=5)
(n=9,all
WBA)c,d
(nvaries
7WBA;3
ROI)
d(n=0,all
WBA)
(nvaries)d
(nvaries)d
Totalg
reymatter
NA
NA
13NA
NA
1NA
3/4
3,5,8
NA
NA
3/4
NA
Someevidence
f
Amygdala
Left
Bilateral
22,3
115
ns3
23,15
36,8,9
6/9
16,17,19,21,22,25
NA
9/1
85/8
6,7,8,19,25
Strong
evidence
Striatum
gns
Bilateral
1211
521
3,26
4ns
31,7,10
6/8
18,20,21,22,23,25
NA
9/1
71/5
10
Strong
evidence
Insula
Bilateral
ns21
,211
521
3,26
512
26,10
2/6
21,22
NA
4/1
53/6
6,10,21
Someevidence
Frontalg
yrus
Left
ns21
,311
512
94
1127
,10
2/7
18,22
NA
4/1
6ns
/5So
meevidence
Fusiform
gyrus
nsLeft
22,3
nsns
2ns
ns2/7
21,22
NA
2/1
6ns
/5Weakevidence
Temporalcortex
nsns
11ns
226,27
321
,345
,8,9,10
2/7
21,22
NA
6/1
61/5
12
Weakevidence
(Pre)cuneus
nsns
11ns
413,26,27,28
511
27,10
1/7
18
NA
3/1
6ns
/5Weakevidence
Parietalcortex
nsns
11ns
226,28
3ns
17
3/7
21,22,24
NA
4/1
6ns
/5Weakevidence
Anteriorcingulate
nsns
nsns
213,26
211
521
,33/6
15,16,20
NA
5/1
61/7
16
Weakevidence
Orbito
frontalcortex
nsns
1111
3ns
2ns
28,10
2/8
21,22
NA
4/1
7ns
/7Noevidence
Prefrontalcortex
nsns
21,2
nsns
212
811
02/7
18,21
NA
3/1
62/6
19,21
Noevidence
Cerebellum
nsns
nsns
113
1ns
34,7,8
1/7
21
NA
4/1
6ns
/5Noevidence
Occipitalcortex
nsns
12ns
113
2ns
37,8,10
ns/7
NA
3/1
6ns
/5Noevidence
Cingulate
nsns
21,3
ns21
3,26
4ns
115
ns/7
NA
1/1
6ns
/5Noevidence
Hippocampus
nsns
nsns
113
1ns
181/8
22
NA
2/1
7ns
/5Noevidence
Pre-centralgyrus
nsns
nsns
128
1ns
110
ns/7
NA
1/1
6ns
/5Noevidence
Thalamus
nsns
nsns
213,26
2ns
ns1/7
21
NA
1/1
6ns
/5Noevidence
Lingualgyrus
nsns
nsns
nsns
nsns
1/7
22
NA
1/1
6ns
/5Noevidence
Uncus
nsns
11ns
ns1
nsns
ns/7
NA
1/1
6ns
/5Noevidence
Prem
otor
cortex
nsns
nsns
113
1ns
nsns
/7NA
1/1
6ns
/5Noevidence
Firstcolum
nwith
results
show
sthemeta-analyticalresults,secondcolumnshow
sresults
from
studiesusingODD/CD-onlysamples,third
columnshow
sresults
from
studiesusingODD/CD-m
ixed
and
ODD/CD+ADHDsamples,lastcolum
nshow
soverallconclusion.Fo
rthe
second
andthirdcolumns
thenumbersinthecolumns
refertothenumbersof
studiesshow
ingsignificantresultsforthisstructure,
numbers
insuperscriptreferto
thereview
edstudiesin
Table1(structuralstudies)
orTables
2and3(functionalstudies).Fo
rtheoverallconclusion
oninvolvem
entof
structures
inODD/CD,three
qualifications
wereused:strongevidence,som
eevidence,w
eakevidence,and
noevidence.T
hequalificationstrong
evidence
was
used
ifastructurewas
foundinvolved
inODD/CD(1)in
both
meta-
analyses
OR(2)eith
eroneof
themeta-analyses
ANDby
halformoreof
thestudiesinthenarrativereview
.The
qualificationsomeevidence
was
used
ifastructurewas
foundinvolved
inODD/CDineither
oneof
themeta-analyses
ANDby
aquarteror
moreof
thestudiesinthenarrativereview
.The
qualificationweakevidence
was
used
ifastructurewas
notfound
involved
inODD/CDineitheroneof
the
meta-analysis,butwas
foundinaquartero
rmoreof
thestudiesinthenarrativereview
.The
qualificationno
evidence
was
used
ifastructurewas
notfound
involved
inODD/CDineitheroneof
themeta-
analysis,andadditio
nally
was
foundby
less
than
aquarterof
thestudiesin
thenarrativereview
.ADHD
AttentionDeficitHyperactiv
ityDisorder;CD
Conduct
Disorder;NAnotapplicable;ns
not
significant;ODDOppositionalDefiant
Disorder;ROIregion
ofinterest;W
BAwholebrainanalysis
aFo
urof
thefive
studiespertainedto
largelyoverlappingsamples
bTw
ostructural,one
hotE
Fandtwocool
EFstudies
cFo
rthestudiesusingODD/CD-m
ixed
andODD/CD+ADHDsamples,onlyfour
ofthestructuralstudiesinvestigated
totalg
reymatter
dResultsaregivenaccordingto
theform
at‘num
berof
studiesreportingsignificantfindingsforthisstructure/totalnumberof
studiesreportingon
thisstructure’
eSixstructural,fivehotE
Fandtwocool
EFstudies
fThe
qualification‘som
eevidence’was
givenalthough
therewereno
meta-analyticalfindings,since
ameta-analysiswas
notp
ossiblefortotalg
reymatter
gIncludingcaudate,putamen,globuspallidus
64 Neuropsychol Rev (2016) 26:44–72
Finally, driven by results from the cool EF section usingonly ODD/CD-only samples, some evidence was found forabnormalities in a cool EF related structure in ODD/CD: theleft precuneus. However, studies that compared ODD/CD-only samples with ADHD-only samples did not provide evi-dence for these abnormalities to be specific for ODD/CD, nordid studies that assessed associations provide evidence forprecuneus volume to be related to ODD/CD symptoms.Taken together, our results show strong evidence for involve-ment of the amygdala and the striatum, some evidence forinvolvement of the insula and the frontal gyrus, and weakevidence for involvement of the precuneus in ODD/CD.
Discussion
We reviewed 29 structural and functional neuroimaging stud-ies into hot and cool EF in ODD/CD (with and withoutADHD) and performed two meta-analyses on subsets of thesestudies. We hypothesized that, compared to controls, individ-uals with ODD/CD (with and without ADHD) would showboth structural and functional brain abnormalities in areas re-lated to hot and cool EF. In addition, we investigated thespecificity of these abnormalities by discussing results sepa-rately for studies including an ODD/CD-only group and re-sults from studies including anODD/CD-mixed or ODD/CD+ADHD group, using a narrative approach. Furthermore, indi-viduals with ODD/CD were compared to individuals withADHD. The results confirmed impairments in structure andfunction in most of the hypothesised hot EF related structures(i.e. amygdala, insula, and anterior cingulate), and to a lesserextent results confirmed such impairments in cool EF relatedstructures (i.e. dorsolateral prefrontal cortex and its subcorticalconnections, including the cerebellum). Impairments weremainly present as reductions in volume or reductions in activ-ity of the structures. Some of these abnormalities were report-ed to be specific to ODD/CD when compared to ADHD-only,or showed associations with ODD/CD symptom levels.
In terms of hot EF related neuroanatomical correlates, sev-eral structures were consistently reported to be associated withODD/CD (with and without ADHD) compared to controls(see Table 4 for an overview). This pattern of findingsemerged irrespective of a structural or functional approachof the study. For two areas the combination of the narrativereview and meta-analyses of both structural and functionalMRI studies provided strong evidence of abnormalities inODD/CD. These areas were the bilateral amygdala and theright striatum, including the caudate, putamen, and globuspallidus. In addition to these two areas, some evidence wasprovided for abnormalities of the bilateral insula in ODD/CDgroups. For all three areas, the majority of studies reportedreductions in volume and activity in ODD/CD with and with-out ADHD groups compared to controls. Furthermore, for all
three areas abnormalities appeared to be mainly driven bystudies including ODD/CD-only or ODD/CD-mixed groups,and to a lesser extent by studies including ODD/CD+ADHDgroups. Hence, it seems that ODD/CD may drive these abnor-malities, rather than ADHD. For the amygdala, this claim issupported by strong evidence for the abnormalities to be spe-cific for ODD/CD when compared to ADHD-only and by asignificant association between abnormalities in the amygdalaand ODD/CD symptoms in both ODD/CD-only and ODD/CD-mixed groups. For the insula there was some evidencefor specificity for ODD/CD compared to ADHD and someevidence for an association with ODD/CD symptoms.However, for the right striatum there was neither evidencefor specificity for ODD/CD nor for an association betweenODD/CD related symptoms.
The finding of abnormalities in brain areas related to hot EFfor ODD/CD is in line with the abnormalities observed inODD/CD in terms of performance on hot EF related tasks.The amygdala, striatum, insula, and frontal gyrus are impli-cated in emotion processing (Schumann et al. 2011; Andersonand Kiehl 2012), reinforcement processing (Di Martino et al.2008; Helie et al. 2013; Jessup and O’Doherty 2011), empathy(Gu et al. 2012, 2013; Lamm and Singer 2010), and introspec-tion (Passamonti et al. 2012; Goldberg et al. 2006) in typicallydeveloping controls, respectively. There is robust support forabnormalities in emotion processing, altered reinforcementsensitivity, deficits in empathy and abnormalities in self-control in individuals with ODD/CD (Blair 2013; Byrd et al.2014; Burke et al. 2002).
In terms of cool EF related neuroanatomical correlates, fewsignificant findings emerged. Partly, this was due to the limit-ed number of functional neuroimaging studies into cool EF inindividuals with ODD/CD. Almost all available functionalstudies were performed within the same sample, precludinga meta-analysis of these findings and restricting our approachto a narrative review of studies. For the left frontal gyrus someevidence for abnormalities in ODD/CDwas found. For the leftprecuneus weak evidence was found, however this structurewas predominantly reported in studies using ODD/CD-onlygroups, suggesting involvement of this structure in ODD/CD.Future studies should confirm this claim. Surprisingly, therewas no evidence for involvement of abnormalities of moretypical cool EF structures, such as the dorsolateral prefrontalcortex or the cerebellum in ODD/CD. The lack of robustcerebellum-related results in ODD/CD may be due to abnor-malities in the cerebellum being more strongly related to co-morbid ADHD, than to ODD/CD itself. Some support for thisinterpretation was found, since structural and functional ab-normalities in the cerebellum were reported mainly in studiesincluding ODD/CD+ADHD and ODD/CD-mixed groups,and not so much in studies including ODD/CD-only groups.
The absence of a significant relationship between brainareas related to cool EF and ODD/CD is in line with the
Neuropsychol Rev (2016) 26:44–72 65
inconsistent findings on tasks assessing cool EF in this group.ODD/CD seems to be associated with specific inhibitory-related abnormalities (Willoughby et al. 2011; Giedd et al.2010; Hobson et al. 2011), rather than with more general coolEF abnormalities. This is emphasised by results from studiesin other cool EF related domains, such as cognitive flexibilityand working memory, that showed these domains to be intactin ODD/CD and to be associated with comorbid ADHD in-stead of ODD/CD (Dawel et al. 2012; Giedd et al. 2010;Hobson et al. 2011; Prencipe et al. 2011).
Although we have suggested that abnormalities in ODD/CD are mainly associated with hot EF, most of the reportedhot EF brain areas are also involved in cool EF (see Fig. 4 forvisualisation), implying an integrated model of abnormalitiesin both hot and cool EF, rather than a strict separation betweenthese two. This is in line with models that propose abnormal-ities in both hot and cool EF circuits to underlie the behav-ioural characteristics of ODD/CD, such as the model by Blair(2005). The connectivity between the structures we found tobe involved in ODD/CD (amygdala, insula, striatum, medial/frontal gyrus, precuneus) has been extensively studied, andstudies show that all these areas appear to be interconnected,albeit each to a different degree, and are found to subserveseveral different functions. For example, it is well establishedthat the amygdala, as part of the limbic system, is critical forhot EF, such as the processing of emotional information.Moreover, it also modulates behavioural responses subse-quent to emotionally salient information, with the regulatoryprocess being regarded as a cool EF related process(Schumann et al. 2011; Anderson and Kiehl 2012). The insulais also involved in emotional information processing, and isnecessary for empathic pain perception and compassion,hence referring to hot EF, but also for the integration of
emotional and cognitive processes, thus referring to cool EF(Gu et al. 2012, 2013; Lamm and Singer 2010). The striatum,specifically the dorsal part (mainly caudate and putamen), isassociated with inhibitory functions as well as with reinforce-ment learning, and thus referring to both cool and hot EFrelated processes, respectively (Di Martino et al. 2008; Helieet al. 2013; Jessup and O’Doherty 2011). The medial/superiorfrontal gyrus is associated with response inhibition and othercool EF functions, but also with introspection, which can beseen as a more hot EF related process (Passamonti et al. 2012;Goldberg et al. 2006). The precuneus, finally, is an area that isinvolved in reflective, self-related processing and awareness,incorporating both hot and cool EF processes (S. Zhang and Li2012). Overall, it shows that our findings of abnormalities inthe five abovementioned structures may lead to impairmentsthat show striking parallels with the behavioural characteris-tics of individuals with ODD/CD, such as their difficulties inlearning socially acceptable behaviour, and their tendency tocontribute hostile intentions to peers in socially ambiguoussituations (Byrd et al. 2014; Blair 2013; Burke et al. 2002).This supports an integrated model of hot and cool EFabnormalities in ODD/CD, for which the model by Blair(2005) is currently the most comprehensive and encompassesa full description of both circuits.
Our results regarding abnormalities in brain structure andfunction in ODD/CD are in line with models on brain func-tioning of other externalising disorders for which ODD andCD are risk-factors, such as antisocial personality disorder(ASPD) or psychopathy (American Psychiatric Association2013). ASPD and psychopathy are characterised by symp-toms of emotional detachment and impulsive behaviour, andthe disregard for, or violation of, rights of others. The overlapin neurocognitive and behavioural abnormalities betweenASPD, psychopathy and ODD/CD, suggests that these disor-ders may share an underlying aetiology. A recently proposedneurocognitive model of psychopathy includes the paralimbicsystem as a whole, comprising both the frontal lobe and thetemporal lobe (including the amygdala). A review on psy-chopathy reported that most consistently implicated areaswere the amygdala, anterior and posterior cingulate cortices,orbitofrontal cortex and adjacent (para)limbic structures (Yangand Raine 2009). Furthermore, these brain areas are also im-plicated in ASPD (Kiehl 2006; Blair 2010). Since we foundsome of these areas to be involved in ODD/CD, a sharedaetiology with ASPD and psychopathy seems plausible.
Research Agenda
The goal of the current review was to aggregate and integratefindings from structural and functional studies on brain ana-tomical aspects of ODD/CD, to provide a complete overviewof brain abnormalities associated with ODD/CD. During thisprocess we encountered several issues in the field. Firstly, in
Fig. 4 Overview of the connectivity of areas involved in ODD/CD de-scribed in this review, based on connectivity studies in typically develop-ing individuals (see main text). Green arrows indicate excitatory connec-tivity, red arrows indicate inhibitory connectivity, and arrows in bothgreen and red indicate both excitatory and inhibitor connectivity. Redboxed areas exhibit mainly hot EF, blue boxed areas exhibit mainly coolEF, and areas boxed in both red and blue exhibit a combination of hot andcool EF
66 Neuropsychol Rev (2016) 26:44–72
our meta-analysis of hot EF functional studies, different re-sults were obtained when the analyses were conducted withand without inclusion of ROI-based studies, suggesting thatROI-based studies may bias conclusions. We therefore re-stricted our hot EF meta-analysis to WBA approach-basedstudies, given their unbiased results. Future studies investigat-ing brain abnormalities associated with ODD/CD should notonly include ROI analyses, but would benefit from reporting(additional) whole brain analyses, rendering the results eligi-ble to be included in a meta-analysis. In addition, because ofthe limited amount of studies into cool EF and the overlap insubjects in these studies, we were not able to perform a meta-analysis on these data; further research is needed in order to beable to draw firm conclusions on the possible involvement ofcool EF abnormalities in ODD/CD, particularly since evi-dence from behavioural studies into cool EF impairments inODD/CD is inconclusive.
Secondly, in terms of sample characteristics, an importantfuture direction is to study the role of callous-unemotional(CU) traits or psychopathic traits in individuals with ODD/CD. It has been reported that children with CD and high levelsof CU traits show greater abnormalities in reward sensitivity,lower responsivity to treatment by parenting strategies, andare more resistant to psychosocial intervention than those withCD and few CU traits (Byrd et al. 2014). Another importantissue in terms of the investigated samples is that most of thestudies focussed on males, limiting generalization of currentfindings. This is emphasised by the differing pathways ofbrain development in boys and girls (Haney-Caron et al.2014), and the fact that boys and girls seem to differ in theneural characteristics of ODD/CD (Hyatt et al. 2012). Forexample, when comparing male and female samples withODD/CD-mixed, gender differences emerged for the insula(Fairchild et al. 2013a). Moreover, since substantial develop-mental changes take place during the transition from child-hood to adulthood (Giedd et al. 2009), studies should takedevelopmental patterns into account. This can be achievedby adhering to smaller age ranges when including participants,but also by matching on age or using large samples that allowdevelopmental patterns to be studied. In the majority of theincluded studies, participant ages ranged between 10 and17 years, which cover a substantial time of brain development.Furthermore, since ODD and CD are childhood disorders andare referred to as APD when persisting into adulthood, ourfindings may not be generalizable to older individuals.
Thirdly, at a statistical level, future studies should be moreconsistent in statistical corrections and thresholding. A signif-icant proportion of the studies did not correct for multiplecomparisons or Type 1 errors, nor applied clusterthresholding. Of these, most studies using a WBA approachdid report Family Wise Error or False Discovery Ratecorrected results, except for three structural and four function-al studies (Bussing et al. 2002; Kruesi et al. 2004; Fairchild et
al. 2013a; Marsh et al. 2013; Passamonti et al. 2010; White etal. 2013, 2014). However, most of these studies did set aminimum number of >10 contiguous voxels and lowered thestatistical threshold to p< .005 (Marsh et al. 2013; White et al.2013, 2014) or p< .001 (Passamonti et al. 2012; Fairchild etal. 2013a), leaving only two studies that reported at an uncor-rected statistical threshold of p< .05. This implies that theresults of both the meta-analyses and narrative reviews arein general reliable, but should be interpreted with some cau-tion, since especially in WBA-based studies, these statisticalmethods are of importance (Chumbley and Friston 2009;Chumbley et al. 2010; Nichols and Hayasaka 2003).
Fourthly, future studies should look at other imaging tech-niques, such as DTI, to improve our knowledge on structuraland functional brain abnormalities associated with ODD/CD.Up to date, eight studies have used DTI to study abnormalitiesin pathways in ODD/CD. Although there is relatively consis-tent evidence for involvement of the uncinate fasciculus(Wang et al. 2012; Finger et al. 2012; Haney-Caron et al.2014; Passamonti et al. 2012; Sarkar et al. 2013; Zhang etal. 2014a), studies largely differed in terms of the white matterstructures implicated in ODD/CD. Furthermore, while somestudies report heightened DTI measures in ODD/CD(Passamonti et al. 2012; Sarkar et al. 2013; Zhang et al.2014a, 2014b), other studies report lowered DTI measures inODD/CD (Haney-Caron et al. 2014; Wang et al. 2012), or nodifferences at all (Finger et al. 2012). In addition to integratingfindings of DTI studies, future studies could look at otherimaging techniques, such as resting-state fMRI and combinedsMRI-fMRI studies.
Finally, to advance our understanding of ODD and CD andthe specificity of neural abnormalities future studies shouldassess well-defined and sufficiently sized samples. For exam-ple, the ideal study should at least compare ODD/CD-only,ADHD-only, comorbid ODD/CD+ADHD, and typically de-veloping control groups with sufficient sample size, to inves-tigate the possible confounding effect of comorbid ADHD.Furthermore, since there appears to be a difference in under-lying mechanisms for childhood-onset and adolescent-limitedODD/CD (Moffitt 1993), future studies should differentiatebetween these types of ODD/CD. Moreover, studies shouldinclude separate groups of individuals with pure ODD andpure CD, to investigate whether these disorders are indeedtwo ends of a spectrum or rather two separate disorders, sincethis is not completely evident from studies performed so far(Matthys et al. 2013; Rowe et al. 2010). A strict separation ofthe two disorders was not possible at the time of the currentreview, given that until the release of the DSM-5, a diagnosisof ODDwas precluded when CDwas present. Therefore moststudies into CD did not investigate or report on ODD, possiblylimiting the generalization of our findings to ODD-only andCD-only samples. In terms of ADHD comorbidity, most stud-ies that reported on ODD/CD did not investigate (or report on)
Neuropsychol Rev (2016) 26:44–72 67
possible comorbidity with ADHD, hence, studies investigat-ing ODD/CD-only samples are sparse. For this reason, wecould not perform a meta-analysis on studies that comparedgroups with ODD/CD-only to groups with ADHD-only. Also,it would be of great interest to investigate whether ADHDcomorbidity adds abnormalities associated with ADHD tothose already present due to ODD/CD, or that ODD/CD withcomorbid ADHD is a genuinely separate disorder.Furthermore, in the current review we did not examine theimpact of other common comorbidities such as anxiety ordepression. This is especially important because these comor-bidities are associated with abnormalities in some of the samebrain areas that appear to be implicated in ODD/CD (Sterzer etal. 2005). Especially the amygdala is related to both anxietyand depressive disorder, albeit in an opposite direction, show-ing higher volumes, instead of the lower volumes, to be asso-ciated with ODD/CD (Machado-de-Sousa et al. 2014; Younget al. 2014). Furthermore, not all studies investigated (or re-ported on) (past) medication use, even though in almost allstudies, MRI scanning was done in individuals free of medi-cation at the time of testing (Rubia et al. 2013). Current andpast mediation use should be incorporated in future studies,since this may have long-term effects on brain structure andfunction (Hart et al. 2013; Rubia et al. 2014).
Conclusion
We reviewed 29 studies on structural and functional neuroim-aging in samples with ODD/CD with and without comorbidADHD, and performed meta-analyses on a subset of eightstructural studies and nine hot EF functional studies. Despitesome limitations and heterogeneity amongst studies, resultsindicated that individuals with ODD/CD primarily show ab-normalities in the bilateral amygdala, bilateral insula, rightstriatum and left medial/superior frontal gyrus as well as theleft precuneus. Evidence of involvement of these areas waspresent in both structural and functional studies, and irrespec-tive of whether the study included individuals with ODD/CD-only or with ODD/CD with comorbid ADHD. Our resultsshow strong evidence of specificity for abnormalities in theamygdala for ODD/CD as compared to ADHD, and correla-tional studies further support the association between abnor-malities in the amygdala and ODD/CD symptoms. Besidesthe left precuneus that was revealed by the narrative reviewof cool EF, there was no evidence for abnormalities in typicalcool EF related structures, such as the cerebellum and dorso-lateral prefrontal cortex. Our findings confirm the involve-ment of hot, and to a smaller extent cool EF associated brainareas in ODD/CD, and support an integrated model of bothhot and cool deficits for ODD/CD (e.g. Blair 2005). The areasfound associated with ODD/CD are involved in emotion-pro-cessing, error monitoring, self-control, and empathic and
social behaviour. It is precisely these functions that are im-paired in children and adolescents with ODD/CD, and thatresult in difficulties learning socially accepted behavioursand reactions, attributing hostile intentions to others, and pre-ferring aggressive solutions to social dilemmas.
Open Access This article is distributed under the terms of the CreativeCommons At t r ibut ion 4 .0 In te rna t ional License (h t tp : / /creativecommons.org/licenses/by/4.0/), which permits unrestricted use,distribution, and reproduction in any medium, provided you give appro-priate credit to the original author(s) and the source, provide a link to theCreative Commons license, and indicate if changes were made.
References
American Psychiatric Association (2013)Diagnostic and statistical man-ual of mental disorders: DSM-5, 5th edn. Arlington: AmericanPsychiatric Publishing, Inc.
Anderson, N. E., & Kiehl, K. A. (2012). The psychopath magnetized:insights from brain imaging. Trends in Cognitive Sciences, 16(1),52–60. doi:10.1016/j.tics.2011.11.008.
Anderson, N. E., & Kiehl, K. A. (2013). Psychopathy: developmentalperspectives and their implications for treatment. RestorativeNeurology and Neuroscience. doi:10.3233/rnn-139001.
Anderson, V. A., Jacobs, R., & Anderson, P. J. (2008). Executive func-tions and frontal lobes: A lifespan perspective. Hove: PsychologyPress.
Blair, R. J. (2005). Applying a cognitive neuroscience perspective to thedisorder of psychopathy.Development and Psychopathology, 17(3),865–891. doi:10.1017/s0954579405050418.
Blair, R. J. (2010). Neuroimaging of psychopathy and antisocial behav-ior: a targeted review. Current Psychiatry Reports, 12(1), 76–82.doi:10.1007/s11920-009-0086-x.
Blair, R. J. (2013). The neurobiology of psychopathic traits in youths.Nature Reviews Neuroscience, 14(11), 786–799. doi:10.1038/nrn3577.
Blair, R. J., & Lee, T. M. (2013). The social cognitive neuroscience ofaggression, violence, and psychopathy. Social Neuroscience, 8(2),108–111. doi:10.1080/17470919.2012.757869.
Boylan, K., Vaillancourt, T., Boyle, M., & Szatmari, P. (2007).Comorbidity of internalizing disorders in children with oppositionaldefiant disorder. European Child and Adolescent Psychiatry, 16(8),484–494. doi:10.1007/s00787-007-0624-1.
Burke, J. D., Loeber, R., & Birmaher, B. (2002). Oppositional defiantdisorder and conduct disorder: a review of the past 10 years, part II.Journal of the American Academy of Child and AdolescentPsychiatry, 41(11), 1275–1293. doi:10.1097/00004583-200211000-00009.
Bussing, R., Grudnik, J., Mason, D., Wasiak, M., & Leonard, C. (2002).ADHD and conduct disorder: anMRI study in a community sample.The World Journal of Biological Psychiatry, 3(4), 216–220.
Byrd, A. L., Loeber, R., & Pardini, D. A. (2014). Antisocial behavior,psychopathic features and abnormalities in reward and punishmentprocessing in youth. Clinical Child and Family Psychology Review,17(2), 125–156. doi:10.1007/s10567-013-0159-6.
Chumbley, J. R., & Friston, K. J. (2009). False discovery rate revisited:FDR and topological inference using Gaussian random fields.NeuroImage, 44(1), 62–70. doi:10.1016/j.neuroimage.2008.05.021.
Chumbley, J. R., Worsley, K., Flandin, G., & Friston, K. (2010).Topological FDR for neuroimaging. NeuroImage, 49(4), 3057–3064. doi:10.1016/j.neuroimage.2009.10.090.
68 Neuropsychol Rev (2016) 26:44–72
Cohn, M. D., Veltman, D. J., Pape, L. E., van Lith, K., Vermeiren, R. R.,van den Brink, W., et al. (2014). Incentive processing in persistentdisruptive behavior and psychopathic traits: a functional magneticresonance imaging study in adolescents. Biological Psychiatry. doi:10.1016/j.biopsych.2014.08.017.
Connor, D. F., Steeber, J., &McBurnett, K. (2010). A review of attention-deficit/hyperactivity disorder complicated by symptoms of opposi-tional defiant disorder or conduct disorder. Journal ofDevelopmental and Behavioral Pediatrics, 31(5), 427–440. doi:10.1097/DBP.0b013e3181e121bd.
Coolidge, F. L., Thede, L. L., & Young, S. E. (2000). Heritability and thecomorbidity of attention deficit hyperactivity disorder with behav-ioral disorders and executive function deficits: a preliminary inves-tigation. Developmental Neuropsychology, 17(3), 273–287. doi:10.1207/s15326942dn1703_1.
Crowe, S. L., & Blair, R. J. (2008). The development of antisocial be-havior: what can we learn from functional neuroimaging studies?Development and Psychopathology, 20(4), 1145–1159. doi:10.1017/s0954579408000540.
Dawel, A., O’Kearney, R., McKone, E., & Palermo, R. (2012). Not justfear and sadness: meta-analytic evidence of pervasive emotion rec-ognition deficits for facial and vocal expressions in psychopathy.Neuroscience and Biobehavioral Reviews, 36(10), 2288–2304.doi:10.1016/j.neubiorev.2012.08.006.
Di Martino, A., Scheres, A., Margulies, D. S., Kelly, A. M., Uddin, L. Q.,Shehzad, Z., et al. (2008). Functional connectivity of human stria-tum: a resting state FMRI study. Cerebral Cortex, 18(12), 2735–2747. doi:10.1093/cercor/bhn041.
Diamond, A. (2013). Executive functions. Annual Review of Psychology,64, 135–168. doi:10.1146/annurev-psych-113011-143750.
Dolan, M., & Lennox, C. (2013). Cool and hot executive function inconduct-disordered adolescents with and without co-morbid atten-tion deficit hyperactivity disorder: relationships with externalizingbehaviours. Psychological Medicine, 1–10. doi:10.1017/s0033291712003078.
Eickhoff, S. B., Laird, A. R., Grefkes, C., Wang, L. E., Zilles, K., & Fox,P. T. (2009). Coordinate-based activation likelihood estimationmeta-analysis of neuroimaging data: a random-effects approachbased on empirical estimates of spatial uncertainty. Human BrainMapping, 30(9), 2907–2926. doi:10.1002/hbm.20718.
Eickhoff, S. B., Bzdok, D., Laird, A. R., Kurth, F., & Fox, P. T. (2012).Activation likelihood estimation meta-analysis revisited.NeuroImage, 59(3), 2349–2361. doi:10.1016/j.neuroimage.2011.09.017.
Fahim, C., He, Y., Yoon, U., Chen, J., Evans, A., & Perusse, D. (2011).Neuroanatomy of childhood disruptive behavior disorders.Aggressive Behavior, 37(4), 326–337. doi:10.1002/ab.20396.
Fairchild, G., Passamonti, L., Hurford, G., Hagan, C. C., von demHagen,E. A., van Goozen, S. H., et al. (2011). Brain structure abnormalitiesin early-onset and adolescent-onset conduct disorder. The AmericanJournal of Psychiatry, 168(6), 624–633. doi:10.1176/appi.ajp.2010.10081184.
Fairchild, G., Hagan, C. C.,Walsh, N. D., Passamonti, L., Calder, A. J., &Goodyer, I. M. (2013a). Brain structure abnormalities in adolescentgirls with conduct disorder. Journal of Child Psychology andPsychiatry, 54(1), 86–95. doi:10.1111/j.1469-7610.2012.02617.x.
Fairchild, G., van Goozen, S. H., Calder, A. J., & Goodyer, I. M. (2013b).Research review: evaluating and reformulating the developmentaltaxonomic theory of antisocial behaviour. Journal of ChildPsychology and Psychiatry, 54(9), 924–940. doi:10.1111/jcpp.12102.
Finger, E. C., Marsh, A. A., Mitchell, D. G., Reid, M. E., Sims, C.,Budhani, S., et al. (2008). Abnormal ventromedial prefrontal cortexfunction in children with psychopathic traits during reversal learn-ing. Archives of General Psychiatry, 65(5), 586–594.
Finger, E. C., Marsh, A. A., Blair, K. S., Reid, M. E., Sims, C., Ng, P., etal. (2011). Disrupted reinforcement signaling in the orbitofrontalcortex and caudate in youths with conduct disorder or oppositionaldefiant disorder and a high level of psychopathic traits. TheAmerican Journal of Psychiatry, 168(2), 152–162. doi:10.1176/appi.ajp.2010.10010129.
Finger, E. C., Marsh, A., Blair, K. S., Majestic, C., Evangelou, I., Gupta,K., et al. (2012). Impaired functional but preserved structural con-nectivity in limbic white matter tracts in youth with conduct disorderor oppositional defiant disorder plus psychopathic traits. PsychiatryResearch, 202(3), 239–244. doi:10.1016/j.pscychresns.2011.11.002.
Fox, P. T., Parsons, L. M., & Lancaster, J. L. (1998). Beyond the singlestudy: function/location metanalysis in cognitive neuroimaging.Current Opinion in Neurobiology, 8(2), 178–187.
Gatzke-Kopp, L. M., Beauchaine, T. P., Shannon, K. E., Chipman, J.,Fleming, A. P., Crowell, S. E., et al. (2009). Neurological correlatesof reward responding in adolescents with and without externalizingbehavior disorders. Journal of Abnormal Psychology, 118(1), 203–213. doi:10.1037/a0014378.
Giedd, J. N., Lalonde, F. M., Celano, M. J., White, S. L., Wallace, G. L.,Lee, N. R., et al. (2009). Anatomical brain magnetic resonance im-aging of typically developing children and adolescents. Journal ofthe American Academy of Child and Adolescent Psychiatry, 48(5),465–470. doi:10.1097/CHI.0b013e31819f2715.
Giedd, J. N., Stockman, M., Weddle, C., Liverpool, M., Alexander-Bloch, A., Wallace, G. L., et al. (2010). Anatomic magnetic reso-nance imaging of the developing child and adolescent brain andeffects of genetic variation. Neuropsychology Review, 20(4), 349–361. doi:10.1007/s11065-010-9151-9.
Goldberg, I. I., Harel, M., & Malach, R. (2006). When the brain loses itsself: prefrontal inactivation during sensorimotor processing.Neuron,50(2), 329–339. doi:10.1016/j.neuron.2006.03.015.
Gu, X., Gao, Z., Wang, X., Liu, X., Knight, R. T., Hof, P. R., et al. (2012).Anterior insular cortex is necessary for empathetic pain perception.Brain, 135(Pt 9), 2726–2735. doi:10.1093/brain/aws199.
Gu, X., Hof, P. R., Friston, K. J., & Fan, J. (2013). Anterior insular cortexand emotional awareness. The Journal of Comparative Neurology.doi:10.1002/cne.23368.
Hamilton, S. S., & Armando, J. (2008). Oppositional defiant disorder.American Family Physician, 78(7), 861–866.
Haney-Caron, E., Caprihan, A., & Stevens, M. C. (2014). DTI-measuredwhite matter abnormalities in adolescents with Conduct Disorder.Journal of Psychiatric Research, 48(1), 111–120. doi:10.1016/j.jpsychires.2013.09.015.
Hart, H., Radua, J., Nakao, T., Mataix-Cols, D., & Rubia, K. (2013).Meta-analysis of functional magnetic resonance imaging studies ofinhibition and attention in attention-deficit/hyperactivity disorder:exploring task-specific, stimulant medication, and age effects.JAMA Psychiatry, 70(2), 185–198. doi:10.1001/jamapsychiatry.2013.277.
Helie, S., Chakravarthy, S., & Moustafa, A. A. (2013). Exploring thecognitive and motor functions of the basal ganglia: an integrativereview of computational cognitive neuroscience models. Frontiersin Computational Neuroscience, 7, 174. doi:10.3389/fncom.2013.00174.
Herpertz, S. C., Huebner, T., Marx, I., Vloet, T. D., Fink, G. R., Stoecker,T., et al. (2008). Emotional processing in male adolescents withchildhood-onset conduct disorder. Journal of Child Psychologyand Psychiatry, 49(7), 781–791. doi:10.1111/j.1469-7610.2008.01905.x.
Hobson, C. W., Scott, S., & Rubia, K. (2011). Investigation of cool andhot executive function in ODD/CD independently of ADHD.Journal of Child Psychology and Psychiatry, 52(10), 1035–1043.doi:10.1111/j.1469-7610.2011.02454.x.
Neuropsychol Rev (2016) 26:44–72 69
Huebner, T., Vloet, T. D., Marx, I., Konrad, K., Fink, G. R., Herpertz, S.C., et al. (2008). Morphometric brain abnormalities in boys withconduct disorder. Journal of the American Academy of Child andAdolescent Psychiatry, 47(5), 540–547. doi:10.1097/CHI.0b013e3181676545.
Hummer, T. A., Wang, Y., Kronenberger, W. G., Dunn, D. W., &Mathews, V. P. (2015). The relationship of brain structure to ageand executive functioning in adolescent disruptive behavior disor-der. Psychiatry Research, 231(3), 210–217. doi:10.1016/j.pscychresns.2014.11.009.
Hyatt, C. J., Haney-Caron, E., & Stevens, M. C. (2012). Cortical thick-ness and folding deficits in conduct-disordered adolescents.Biological Psychiatry, 72(3), 207–214. doi:10.1016/j.biopsych.2011.11.017.
Jessup, R. K., & O’Doherty, J. P. (2011). Human dorsal striatal activityduring choice discriminates reinforcement learning behavior fromthe gambler’s fallacy. The Journal of Neuroscience, 31(17), 6296–6304. doi:10.1523/jneurosci.6421-10.2011.
Kalnin, A. J., Edwards, C. R., Wang, Y., Kronenberger, W. G., Hummer,T. A., Mosier, K. M., et al. (2011). The interacting role of mediaviolence exposure and aggressive-disruptive behavior in adolescentbrain activation during an emotional Stroop task. PsychiatryResearch, 192(1), 12–19. doi:10.1016/j.pscychresns.2010.11.005.
Kerr, A., & Zelazo, P. D. (2004). Development of Bhot^ executive func-tion: the children’s gambling task. Brain and Cognition, 55(1), 148–157. doi:10.1016/s0278-2626(03)00275-6.
Kiehl, K. A. (2006). A cognitive neuroscience perspective on psychopa-thy: evidence for paralimbic system dysfunction. PsychiatryResearch, 142(2-3), 107–128. doi:10.1016/j.psychres.2005.09.013.
Kimonis, E. R., & Frick, P. J. (2010). Oppositional defiant disorder andconduct disorder grown-up. Journal of Developmental andBehavioral Pediatrics, 31(3), 244–254. doi:10.1097/DBP.0b013e3181d3d320.
Kruesi, M. J., Casanova, M. F., Mannheim, G., & Johnson-Bilder, A.(2004). Reduced temporal lobe volume in early onset conduct dis-order. Psychiatry Research, 132(1), 1–11. doi:10.1016/j.pscychresns.2004.07.002.
Lahey, B. B., & Waldman, I. D. (2012). Annual research review: pheno-typic and causal structure of conduct disorder in the broader contextof prevalent forms of psychopathology. Journal of Child Psychologyand Psychiatry, 53(5), 536–557. doi:10.1111/j.1469-7610.2011.02509.x.
Laird, A. R., Fox, P. M., Price, C. J., Glahn, D. C., Uecker, A. M.,Lancaster, J. L., et al. (2005). ALE meta-analysis: controlling thefalse discovery rate and performing statistical contrasts. HumanBrain Mapping, 25(1), 155–164. doi:10.1002/hbm.20136.
Laird, A. R., Robinson, J. L., McMillan, K.M., Tordesillas-Gutierrez, D.,Moran, S. T., Gonzales, S. M., et al. (2010). Comparison of thedisparity between Talairach and MNI coordinates in functional neu-roimaging data: validation of the Lancaster transform. NeuroImage,51(2), 677–683. doi:10.1016/j.neuroimage.2010.02.048.
Lamm, C., & Singer, T. (2010). The role of anterior insular cortex insocial emotions. Brain Structure and Function, 214(5–6), 579–591. doi:10.1007/s00429-010-0251-3.
Lancaster, J. L., Tordesillas-Gutierrez, D., Martinez, M., Salinas, F.,Evans, A., Zilles, K., et al. (2007). Bias between MNI andTalairach coordinates analyzed using the ICBM-152 brain template.Human Brain Mapping, 28(11), 1194–1205. doi:10.1002/hbm.20345.
Loeber, R., Burke, J. D., Lahey, B. B., Winters, A., & Zera, M. (2000).Oppositional defiant and conduct disorder: a review of the past 10years, part I. Journal of the American Academy of Child andAdolescent Psychiatry, 39(12), 1468–1484. doi:10.1097/00004583-200012000-00007.
Loeber, R., Slot, N. W., Van der Laan, P., & Hoeve, M. (2008).Tomorrow’s criminals. Farnham: Ashgate Publishing Limited.
Loeber, R., Burke, J., & Pardini, D. A. (2009). Perspectives on opposi-tional defiant disorder, conduct disorder, and psychopathic features.Journal of Child Psychology and Psychiatry, 50(1–2), 133–142. doi:10.1111/j.1469-7610.2008.02011.x. Research Support, N.I.H.,Extramural Review.
Machado-de-Sousa, J. P., Osorio Fde, L., Jackowski, A. P., Bressan, R.A., Chagas, M. H., Torro-Alves, N., et al. (2014). Increasedamygdalar and hippocampal volumes in young adults with socialanxiety. PLoS One, 9(2), e88523. doi:10.1371/journal.pone.0088523.
Marsh, A. A., Finger, E. C., Mitchell, D. G., Reid, M. E., Sims, C.,Kosson, D. S., et al. (2008). Reduced amygdala response to fearfulexpressions in children and adolescents with callous-unemotionaltraits and disruptive behavior disorders. The American Journal ofPsychiatry, 165(6), 712–720. doi:10.1176/appi.ajp.2007.07071145.
Marsh, A. A., Finger, E. C., Fowler, K. A., Adalio, C. J., Jurkowitz, I. T.N., Schechter, J. C., et al. (2013). Empathic responsiveness in amyg-dala and anterior cingulate cortex in youths with psychopathic traits.Journal of Child Psychology and Psychiatry, 54(8), 900–910. doi:10.1111/jcpp.12063.
Matthys, W., Vanderschuren, L. J., & Schutter, D. J. (2013). The neuro-biology of oppositional defiant disorder and conduct disorder: al-tered functioning in three mental domains. Development andP s y c h o p a t h o l o g y , 2 5 ( 1 ) , 1 9 3 –2 07 . d o i : 1 0 . 1 0 1 7 /s0954579412000272.
McAlonan, G. M., Cheung, V., Cheung, C., Chua, S. E., Murphy, D. G.,Suckling, J., et al. (2007). Mapping brain structure in attentiondeficit-hyperactivity disorder: a voxel-based MRI study of regionalgrey and white matter volume. Psychiatry Research, 154(2), 171–180. doi:10.1016/j.pscychresns.2006.09.006.
Michalska, K. J., Decety, J., Zeffiro, T. A., & Lahey, B. B. (2015).Association of regional gray matter volumes in the brain with dis-ruptive behavior disorders in male and female children.NeuroImage, 7, 252–257. doi:10.1016/j.nicl.2014.12.012.
Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent an-tisocial behavior: a developmental taxonomy.Psychological Review,100(4), 674–701.
Newman, J. P., Kosson, D. S. (1986). Passive avoidance learning inpsychopathic and nonpsychopathic offenders. Journal ofAbnormal Psychology, 95, 252–256
Nichols, T., & Hayasaka, S. (2003). Controlling the familywise error ratein functional neuroimaging: a comparative review. StatisticalMethods in Medical Research, 12(5), 419–446.
Oosterlaan, J., Logan, G. D., & Sergeant, J. A. (1998). Response inhibi-tion in AD/HD, CD, comorbid AD/HD+CD, anxious, and controlchildren: a meta-analysis of studies with the stop task. Journal ofChild Psychology and Psychiatry, 39(3), 411–425.
Passamonti, L., Fairchild, G., Goodyer, I. M., Hurford, G., Hagan, C. C.,Rowe, J. B., et al. (2010). Neural abnormalities in early-onset andadolescence-onset conduct disorder. Archives of General Psychiatry,67(7), 729–738. doi:10.1001/archgenpsychiatry.2010.75.
Passamonti, L., Fairchild, G., Fornito, A., Goodyer, I. M., Nimmo-Smith,I., Hagan, C. C., et al. (2012). Abnormal anatomical connectivitybetween the amygdala and orbitofrontal cortex in conduct disorder.PLoS One, 7(11), e48789. doi:10.1371/journal.pone.0048789.
Pennington, B. F., & Ozonoff, S. (1996). Executive functions and devel-opmental psychopathology. Journal of Child Psychology andPsychiatry, 37(1), 51–87.
Prencipe, A., Kesek, A., Cohen, J., Lamm, C., Lewis, M. D., & Zelazo, P.D. (2011). Development of hot and cool executive function duringthe transition to adolescence. Journal of Experimental ChildPsychology, 108(3), 621–637. doi:10.1016/j.jecp.2010.09.008.
Quay, H. C. (1965). Psychopathic Personality as PathologicalStimulation-seeking. The American Journal of Psychiatry, 122180–183
70 Neuropsychol Rev (2016) 26:44–72
Quay, H. C. (1993). The psychobiology of undersocialized aggressiveconduct disorder: A theoretical perspective. Development andPsychopathology, 5(1–2), 165–180
Quay, H. C., & Hogan, A. E. (1999). Handbook of disruptive behaviourdisorders. US: Springer.
Rowe, R., Maughan, B., Pickles, A., Costello, E. J., & Angold, A. (2002).The relationship between DSM-IVoppositional defiant disorder andconduct disorder: findings from the Great Smoky Mountains Study.Journal of Child Psychology and Psychiatry, 43(3), 365–373.
Rowe, R., Costello, E. J., Angold, A., Copeland, W. E., & Maughan, B.(2010). Developmental pathways in oppositional defiant disorderand conduct disorder. Journal of Abnormal Psychology, 119(4),726–738. doi:10.1037/a0020798.
Rubia, K. (2011). BCool^ inferior frontostriatal dysfunction in attention-deficit/hyperactivity disorder versus Bhot^ ventromedialorbitofrontal-limbic dysfunction in conduct disorder: a review.Biological Psychiatry, 69(12), e69–e87. doi:10.1016/j.biopsych.2010.09.023.
Rubia, K., Halari, R., Smith, A. B., Mohammed, M., Scott, S.,Giampietro, V., et al. (2008). Dissociated functional brain abnormal-ities of inhibition in boys with pure conduct disorder and in boyswith pure attention deficit hyperactivity disorder. The AmericanJournal of Psychiatry, 165(7), 889–897. doi:10.1176/appi.ajp.2008.07071084.
Rubia, K., Halari, R., Smith, A. B., Mohammad, M., Scott, S., &Brammer, M. J. (2009a). Shared and disorder-specific prefrontalabnormalities in boys with pure attention-deficit/hyperactivity dis-order compared to boys with pure CD during interference inhibitionand attention allocation. Journal of Child Psychology andPsychiatry, 50(6), 669–678. doi:10.1111/j.1469-7610.2008.02022.x.
Rubia, K., Smith, A. B., Halari, R., Matsukura, F., Mohammad, M.,Taylor, E., et al. (2009b). Disorder-specific dissociation oforbitofrontal dysfunction in boys with pure conduct disorder duringreward and ventrolateral prefrontal dysfunction in boys with pureADHD during sustained attention. The American Journal ofPsychiatry, 166(1), 83–94. doi:10.1176/appi.ajp.2008.08020212.
Rubia, K., Halari, R., Cubillo, A., Mohammad, A. M., Scott, S., &Brammer, M. (2010). Disorder-specific inferior prefrontalhypofunction in boys with pure attention-deficit/hyperactivity disor-der compared to boys with pure conduct disorder during cognitiveflexibility. Human Brain Mapping, 31(12), 1823–1833. doi:10.1002/hbm.20975.
Rubia, K., Alegria, A. A., Cubillo, A. I., Smith, A. B., Brammer, M. J., &Radua, J. (2013). Effects of stimulants on brain function in attention-deficit/hyperactivity disorder: a systematic review and meta-analy-sis. Biological Psychiatry. doi:10.1016/j.biopsych.2013.10.016.
Rubia, K., Alegria, A., & Brinson, H. (2014). Imaging the ADHD brain:disorder-specificity, medication effects and clinical translation.Expert Review of Neurotherapeutics, 14(5), 519–538. doi:10.1586/14737175.2014.907526.
Sarkar, S., Craig, M. C., Catani, M., Dell’Acqua, F., Fahy, T., Deeley, Q.,et al. (2013). Frontotemporal white-matter microstructural abnor-malities in adolescents with conduct disorder: a diffusion tensorimaging study. Psychological Medicine, 43(2), 401–411. doi:10.1017/s003329171200116x.
Sasayama, D., Hayashida, A., Yamasue, H., Harada, Y., Kaneko, T.,Kasai, K., et al. (2010). Neuroanatomical correlates of attention-deficit-hyperactivity disorder accounting for comorbid oppositionaldefiant disorder and conduct disorder. Psychiatry and ClinicalNeurosciences, 64(4), 394–402. doi:10.1111/j.1440-1819.2010.02102.x.
Schumann, C. M., Bauman, M. D., & Amaral, D. G. (2011). Abnormalstructure or function of the amygdala is a common component ofneurodevelopmental disorders. Neuropsychologia, 49(4), 745–759.doi:10.1016/j.neuropsychologia.2010.09.028.
Sterzer, P., & Stadler, C. (2009). Neuroimaging of aggressive and violentbehaviour in children and adolescents. [Review]. Frontiers inBehavioral Neuroscience, 3, doi:10.3389/neuro.08.035.2009.
Sterzer, P., Stadler, C., Krebs, A., Kleinschmidt, A., & Poustka, F. (2005).Abnormal neural responses to emotional visual stimuli in adoles-cents with conduct disorder. Biological Psychiatry, 57(1), 7–15. doi:10.1016/j.biopsych.2004.10.008.
Sterzer, P., Stadler, C., Poustka, F., & Kleinschmidt, A. (2007). A struc-tural neural deficit in adolescents with conduct disorder and its as-sociation with lack of empathy. NeuroImage, 37(1), 335–342. doi:10.1016/j.neuroimage.2007.04.043.
Stevens, M. C., & Haney-Caron, E. (2012). Comparison of brain volumeabnormalities between ADHD and conduct disorder in adolescence.Journal of Psychiatry and Neuroscience, 37(6), 389–398.
Turkeltaub, P. E., Eden, G. F., Jones, K. M., & Zeffiro, T. A. (2002).Meta-analysis of the functional neuroanatomy of single-word read-ing: method and validation. NeuroImage, 16(3 Pt 1), 765–780.
Turkeltaub, P. E., Eickhoff, S. B., Laird, A. R., Fox, M., Wiener, M., &Fox, P. (2012). Minimizing within-experiment and within-groupeffects in Activation Likelihood Estimation meta-analyses. HumanBrain Mapping, 33(1), 1–13. doi:10.1002/hbm.21186.
Van Goozen, S. H. M., Cohen-Kettenis, P. T., Snoek, H., Matthys, W.,Swaab-Barneveld, H., & Van Engeland, H., et al. (2004). Executivefunctioning in children: a comparison of hospitalised ODD andODD/ADHD children and normal controls. Journal of ChildPsychology and Psychiatry, 45(2), 284–292 doi:10.1111/j.1469-7610.2004.00220.x.
Wang, Y., Horst, K. K., Kronenberger, W. G., Hummer, T. A., Mosier, K.M., Kalnin, A. J., et al. (2012). White matter abnormalities associ-ated with disruptive behavior disorder in adolescents with and with-out attention-deficit/hyperactivity disorder. Psychiatry Research:Neuroimaging, 202(3), 245–251. Article.
Waschbusch, D. A. (2002). A meta-analytic examination of comorbidhyperactive-impulsive-attention problems and conduct problems.Psychological Bulletin, 128(1), 118–150. doi:10.1037/0033-2909.128.1.118.
White, S. F., Pope, K., Sinclair, S., Fowler, K. A., Brislin, S. J., Williams,W. C., et al. (2013). Disrupted expected value and prediction errorsignaling in youths with disruptive behavior disorders during a pas-sive avoidance task. The American Journal of Psychiatry, 170(3),315–323. doi:10.1176/appi.ajp.2012.12060840.
White, S. F., Fowler, K. A., Sinclair, S., Schechter, J. C., Majestic, C. M.,Pine, D. S., et al. (2014). Disrupted expected value signaling inyouth with disruptive behavior disorders to environmental rein-forcers. Journal of the American Academy of Child andAdolescent Psychiatry, 53(5), 579–588. doi:10.1016/j.jaac.2013.12.023. e579.
Willcutt, E. G., Doyle, A. E., Nigg, J. T., Faraone, S. V., & Pennington, B.F. (2005). Validity of the executive function theory of attention-deficit/hyperactivity disorder: a meta-analytic review. BiologicalPsychiatry, 57(11), 1336–1346. doi:10.1016/j.biopsych.2005.02.006.
Willoughby, M., Kupersmidt, J., Voegler-Lee, M., & Bryant, D. (2011).Contributions of hot and cool self-regulation to preschool disruptivebehavior and academic achievement . DevelopmentalNeuropsychology, 36(2), 162–180. doi:10.1080/87565641.2010.549980.
Yang, Y., & Raine, A. (2009). Prefrontal structural and functional brainimaging findings in antisocial, violent, and psychopathic individ-uals: a meta-analysis. Psychiatry Research, 174(2), 81–88. doi:10.1016/j.pscychresns.2009.03.012.
Young, K. D., Zotev, V., Phillips, R., Misaki, M., Yuan, H., Drevets, W.C., et al. (2014). Real-time FMRI neurofeedback training of amyg-dala activity in patients with major depressive disorder. PLoS One,9(2), e88785. doi:10.1371/journal.pone.0088785.
Neuropsychol Rev (2016) 26:44–72 71
Zelazo, P. D., & Carlson, S.M. (2012). Hot and cool executive function inchildhood and adolescence: development and plasticity. ChildDevelopment Perspectives, 6(4), 354–360. doi:10.1111/j.1750-8606.2012.00246.x.
Zhang, S., & Li, C. S. (2012). Functional connectivity mapping of thehuman precuneus by resting state fMRI. NeuroImage, 59(4), 3548–3562. doi:10.1016/j.neuroimage.2011.11.023.
Zhang, J., Gao, J., Shi, H., Huang, B., Wang, X., Situ, W., et al. (2014a).Sex differences of uncinate fasciculus structural connectivity in in-dividuals with conduct disorder. BioMed Research International,2014, 673165. doi:10.1155/2014/673165.
Zhang, J., Zhu, X., Wang, X., Gao, J., Shi, H., Huang, B., et al. (2014b).Increased structural connectivity in corpus callosum in adolescentmales with conduct disorder. Journal of the American Academy ofChild and Adolescent Psychiatry, 53(4), 466–475. doi:10.1016/j.jaac.2013.12.015. e461.
Zhu, Y., Ying, K., Wang, J., Su, L., Chen, J., Lin, F., et al. (2014).Differences in functional activity between boys with pure opposi-tional defiant disorder and controls during a response inhibition task:a preliminary study. Brain Imaging and Behavior, 8(4), 588–597.doi:10.1007/s11682-013-9275-7.
72 Neuropsychol Rev (2016) 26:44–72