Search

Psychopathy is a personality disorder characterised by distinctive interpersonal and affective abnormalities combined with recurring disruptive and antisocial behaviours (Hare & Neumann, 2009). It is associated with violence, criminal versatility, and increased rates of recidivism (Hare & Neumann, 2008; Hare & Neumann, 2009; Salekin et al., 1997). Previous structural MRI studies have identified several volumetric and surface-based cortical and subcortical brain differences in individuals with psychopathy compared to control groups (Johanson et al., 2020; Howner et al., 2012; Ly et al., 2012; Yang et al., 2009a; Yang et al., 2009b; Yang et al., 2011). Furthermore, studies examining associations between structural brain alterations and psychopathy Factor 1 (indexing interpersonal and affective traits such as egocentricity, deception, low empathy and manipulation) and Factor 2 (indexing impulsive and antisocial traits such as social deviance, criminality and impulsivity) have noted that while there are certain structural differences associated with both factors, the two factors of psychopathy might be also associated with distinct structural correlates (Contreras-Rodríguez et al., 2015; Cope et al., 2012; De Brito et al., 2021b; Howner et al., 2012; Leutgeb et al., 2015; Miglin et al., 2022; Miskovich et al., 2018; Yang et al., 2009). Finally, a small body of work also reveals that among individuals with a diagnosis of antisocial personality disorder (ASPD), individuals who additionally present with psychopathy (ASPD+P), might show unique abnormalities compared to those without psychopathy (ASPD-P) (De Brito et al., 2021; Gregory et al., 2012, 2015; Johanson et al., 2020). While neuroimaging investigations have sought to characterise the neural differences that underlie psychopathy, there still appears to be some variability in findings across studies and diverging patterns of associations between psychopathy factors and cortical structure (e.g., De Brito et al., 2021a, 2021b; Poeppl et al., 2019). Ambiguity in the current literature may be due to methodological factors such as heterogenous image acquisition techniques, analytical approaches, sample characteristics (e.g., community vs clinical/forensic), and participant characteristics. Furthermore, most published research in the psychopathy literature has focused on male adults from forensic or clinical populations. Given that such individuals are often difficult to access and recruit, previous studies have suffered from small sample sizes and have been unable to consider sex as a variable within the research design or test for sex-by-diagnosis interactions. Overall, we argue that a consistent picture of the brain morphometric differences associated with psychopathy is lacking. Therefore, a comprehensive, large-scale analysis in relation to psychopathy could provide a more accurate account of the neuroanatomical correlates of the disorder. Utilising existing individual-level brain imaging data (such as that assembled by the ENIGMA consortium), some of which may not have been previously published as surface-based morphometry studies, provides an opportunity to overcome the weaknesses of existing research and provides greater population generalisability owing to the large sample size and increase in statistical power. Furthermore, this approach offers the opportunity to consider individual differences in more detail. It allows for the investigation of psychopathy as a dimension and consideration of how the different psychopathy factors may be related to brain structure across the full range of disorder severity. This study will use datasets provided to the ENIGMA-Antisocial behavior working group to examine the cortical and subcortical correlates of psychopathy and take a mega-analytical approach which allows for a more flexible and comprehensive analysis of the data (Boedhoe et al., 2019). Main Research Questions: 1: Do adults with psychopathy show consistent and reliable differences in cortical thickness, surface area and/or subcortical volume compared to control groups without psychopathy? 2: In comparison to a healthy control group, can we identify structural differences that are unique to antisocial personality disorder with psychopathy, and which are common to both antisocial personality disorder with/without psychopathy? 3: How are the two factors of psychopathy associated with cortical thickness, surface area and subcortical volume? 4: Do these structural differences occur across the full spectrum of disorder severity (i.e., across community and/as well as forensic/clinical samples)? 5: Is the relationship between psychopathy and cortical thickness, surface area or subcortical volume different in male versus female participants?

Conduct disorder (CD) is one of the most common childhood psychiatric disorders and is characterised by a repetitive and pervasive pattern of aggressive or antisocial behaviours, such as physical aggression towards other humans or animals, theft, property damage and rule violations (APA, 2013). The worldwide prevalence of CD is estimated to be 2-2.5%, with a prevalence of 3-4% in boys and 1-2% in girls (Maughan et al., 2004). Overall, CD is a leading cause of referral to child and adolescent mental health services and places an enormous burden on the affected individuals, their families and society (Erksine et al., 2014; Rivenbark et al., 2018). Although neuroimaging studies have reported alterations in brain structure in CD and supported the view that CD is a neurodevelopmental disorder, a major concern of the diagnostic criteria for CD (e.g., initiating physical fights) is that they are entirely based on behavioural symptoms, and can be shared with other disorders (e.g., ADHD, Substance Use Disorder, Tourette's disorder or Bipolar Disorder), and are therefore less informative about the underlying psychological states/cognitions or neural substrates which drive the symptoms (Blair et al., 2005; Fairchild et al., 2019). A comprehensive and thorough understanding of the underlying pathophysiology of CD may help address this question and help refine diagnostic approaches. Neuroimaging studies on CD have shown altered neural responses in tasks involving emotional processing and executive function (Alegria et al., 2016; Noordermeer et al., 2016; Raschle et al., 2015) and reductions in grey matter volume (Noordermeer et al., 2016; Raschle et al., 2015; Rogers & De Brito, 2016) across a number of cortical (e.g., ventromedial prefrontal and insular cortices) and subcortical (e.g., amygdala and basal ganglia) regions critical for emotion processing and regulation, reinforcement-based decision-making, executive function and empathic responding (Fairchild et al., 2019). Although these discoveries marked pivotal advances in our understanding of CD, the evidence base suffers from two important limitations. First, inconsistent findings and a lack of replication are common; demographic and clinical features of the samples, as well as methodological factors, have likely contributed to this. For example, CD is a heterogeneous disorder with subtypes based on the age-of-onset of symptoms (childhood-onset versus adolescence-onset, which is defined by whether severe antisocial behaviours emerge before or after 10 years of age), and the presence or absence of callous-unemotional (CU) traits (encompassed by the ‘with limited prosocial emotions’ specifier in the DSM-5; APA, 2013). However, most primary magnetic resonance imaging (MRI) studies of CD have not examined the influence of these subtypes on their findings. Potential sex differences in the relationship between CD and brain structure might also have contributed to the inconsistent findings (Smaragdi et al., 2017). Most studies of CD have focused only on males and hence it is currently understudied/unclear whether females with CD show similar alterations. Relatedly, studies that have investigated mixed-sex samples were largely underpowered to test for sex-by-diagnosis interactions and may hence have contributed to the inconsistencies observed in the literature. Second, most of the neuroimaging studies had small sample sizes, increasing the risk for false positives and false negatives (Button et al., 2013), as well as resulting in limited power to test differences between CD subtypes or explore heterogeneity. In this context, our understanding of CD-related brain structural alterations could be improved through a large-scale coordinated and harmonised analysis of the vast amounts of existing data to map brain differences in heterogeneous CD patients worldwide. In 2009, the Enhancing NeuroImaging Genetics through Meta-Analyses (ENIGMA) consortium (http://enigma.ini.usc.edu) was established to pool brain imaging data from existing cohorts across the world in order to better understand brain structure, function, and disease, based on meta- and mega-analyses of brain imaging and genetic data (Bearden & Thompson, 2017; Thompson et al., 2020). The ENIGMA-Antisocial Behavior (ASB) working group aims to clarify associations between CD, conduct problems (CPs), psychopathy, or antisocial personality disorder and alterations in brain structure and function by using ENIGMA’s highly powered consortium approach to overcome the limitations of small studies with low power and uncertain reproducibility. Therefore, the proposed study will focus on examining the cortical and subcortical correlates of CD using a mega-analysis (i.e., individual participant data meta-analysis) of structural MRI (sMRI) data, considering the influence of subtypes and testing for possible sex-specific effects. In addition to those with a CD diagnosis, there is a considerable amount of youths demonstrating elevated CP (Ghandour et al., 2019; Deighton et al., 2019), but who have not been assessed with diagnostic instruments and/or have not received a clinical diagnosis of CD. It is of interest if the findings from case-control comparisons of brain structure can be generalised to undiagnosed youths with elevated CP. As youths with a CD diagnosis are often recruited from specialised settings (e.g., clinics, mental health services or youth offending services), studying youths with elevated CP recruited through non-specialised settings would reduce selection bias and increase the generalisability of findings to a broader population (Horga et al., 2014). Neuroimaging studies have demonstrated structural brain alterations in youths with elevated CP, including decreased grey matter volume in the insula, amygdala, frontal and temporal regions (Raschle et al., 2015; Rogers & De Brito, 2016). A longitudinal study also reported altered trajectories of cortical thickness development in the dorsolateral prefrontal cortex and the anterior cingulate cortex (ACC), and volume of the hippocampus in youths showing elevated CP (Oostermeijer et al., 2016). Hence, both CP and CD appear to be related to similar and overlapping MRI findings of lower volume in the prefrontal cortex and limbic system. However, to date, no study has directly investigated if and how the cortical and subcortical alterations in youths with elevated CP overlap with the findings in youths with CD. Comparing youths with CD, elevated CP and TD will help clarify the neural mechanisms underlying antisocial behaviours in the clinical population, and extend the results to a more heterogeneous group of undiagnosed individuals showing elevated CP. Based on the ENIGMA-ASB datasets and using large-scale mega-analytical methods, the proposed study will also include an elevated CP group to identify the shared brain alterations underlying elevated antisocial behaviours as well as the specific alterations limited to youths with CD through the large-scale mega-analysis. Research Questions: (1) Do youths with CD show differences in cortical thickness, surface area and/or subcortical volume compared to TD youths? (2) How are sex, age-of-onset and CU traits related to brain alterations in CD? (3) Can alterations detected in youths with CD be generalised to youths with elevated CP?