Conduct Problems in children: Refining the neurocognitive phenotype

Rarely a day goes by without a news story of violence and young people. Less serious antisocial behaviour is also commonplace and has an impact on everyone's life. For many childhood antisocial behaviour is just a phase which they grow out of; for others it is the beginning of a lifelong pattern of aggression, poor relationships and mental health problems.

Why do some children follow this path into adulthood while others do not? Are all children with antisocial behaviour the same or can we identify different types of children at risk? We need to answer these questions if we want to help those youngsters who are most vulnerable. For a minority, aggressive and antisocial behaviour can be cold and calculated, with little or no empathy for the victim. Aggressive and antisocial behaviour can also be hot-headed and impulsive, for example when someone 'loses their cool' during an argument.

Our research focuses on looking at whether the way a child deals with emotional signals in their environment is different depending on whether they are more prone to calculated or hot-headed antisocial behaviour. For example, recent brain imaging research by our group suggests that a tiny almond-shaped part of the brain, the amygdala, is important when human beings take in the emotional signals around them. We have shown that when children view pictures of other people looking scared, amygdala is under-active in those children whose antisocial behaviour is cold and calculated. This could mean that instead of their amygdala telling them to stop their aggressive behaviour, it is telling them nothing at all. We have also shown that amygdala is over-active in those children whose antisocial behaviour is hot-headed and impulsive. This could mean that these children's brains are 'trigger happy', which makes them vulnerable to react aggressively to anything they see as a threat. These recent findings from our group are important in helping us understand children with antisocial behaviour better, but we need to know more.

Do these children process other people's emotions differently, because they experience their own emotions differently too? Can we get the cold and calculated children to feel more for other people by focusing their attention on other people's discomfort? Can we get the hot-headed and impulsive children to be less reactive to other people's emotions? We plan to give children with behavioural problems a number of tasks, both in the brain scanner and outside it, to find out answers to these questions. Finding out more will help us to better understand how different types of children with antisocial behaviour see the world around them. The more we know, the more effectively we can develop interventions that take into account specific problems a child may have.

While this is a basic science research project, with the results being published in specialist, peer-reviewed journals, we will also seek to translate the clinical implications of this research to those who work with children in order to promote more adaptive patterns of behaviour. We are well placed to do this as one of us is a clinician and both of us work with educational psychologists and teachers.

The proposed research aims to delineate psychological and neural processes that are compromised in children with conduct problems. Specifically, we will compare children with conduct problems and high callous-unemotional traits (N=35), children with conduct problems and low callous-unemotional traits (N=35), and ability/SES/sex matched typically developing control children (N=35) on a range of behavioural and neuroimaging tasks. We will assess how these children process their own emotions, how own emotion processing relates to processing of other people's emotions / empathy, and whether attention / familiarity can modulate emotion / empathy processing in the two subgroups with conduct problems.

Data Analysis:
i. Behavioural data: Analyses will be conducted using SPSS software with in-house statistical consultation where appropriate. Standard parametric analyses (e.g. ANOVA) will be used to analyse the behavioural data. Mean scores, measures of standard error and effect sizes will be calculated and a predefined procedure for removal of outliers will be implemented for any reaction time data.
ii. fMRI data: Functional MRI images will be analyzed using Statistical parametric mapping as implemented in SPM8 software. After realignment, normalization and smoothing of the images, each experimental condition will be modelled separately using a synthetic hemodynamic response (HRF) function. For each fMRI task the General Linear Model (GLM) will be used to compare experimental conditions by group (CP/high CU traits, CP/low CU traits and TD) using GLM Flex (http://nmr.mgh.harvard.edu/harvardagingbrain/People/AaronSchultz/GLM_Flex.html). Further analyses will explore whether brain activity to own emotions correlates with brain activity to other people's emotions / empathy across tasks. The role of attention / familiarity in modulating emotional / empathic response to others will also be investigated.

The proposed research has scope to contribute to nation's health and wealth by informing policy and practice in ways that lead to the reduction of the mental health burden and incarceration, as well as increase the economic productivity of its citizens.

We plan to build on the already successful impact strategy of our research unit. The following groups will benefit from this research:

1) Clinicians and practitioners who work with children with conduct problems. Relevant professions include psychiatrists, clinical psychologists, educational psychologists and psychotherapists. In the short to medium term our research will help inform a clinical model of emotion processing biases in different subtypes of children with conduct problems and the degree to which these biases may be open to modification. The PI and Co-I recently organised a conference in London attended by over 250 clinicians (October 2011 "Childhood Disorders: Neuroscience and Intervention") that paired presentations from both leading academics and expert practitioners and included a session on conduct problems and their treatment.

In the longer-term we hope that gains from this project will contribute to improved clinical assessment of children with conduct problems and more tailored clinical approaches to intervention and treatment, in this group, particularly in relation to those with and without CU traits. Short briefing downloadable documents will be created for clinicians and practitioners that will be uploaded onto our UCL Unit website (www.drru-research.org) in order to summarise the potential implications of our research (and that of others in the field) for intervention. We will also continue to deliver clinical workshops and lectures.

2) Children's charities including a variety of voluntary sector organisations that have a direct interest in advancing our understanding of conduct problems in children. These include organisations such as Kids Company and Barnardo's. The PI and Co-I have strong links with Kids Company and contributed to conferences and media events. Engagement with charities will help support their case for intervention: many of these charities lobby government and public to increase provision for children presenting with conduct problems and their families.

3) Policy makers including the Department of Children, Schools and Families, The Department of Health, and the Department of Justice. The PI has previously contributed to policy seminars organised by the Department of Health and Prime Minister's Strategy Unit, and was recently invited to take part in a policy meeting organised by the Royal Society on Neuroscience and the Law. Findings from the proposed study will be well placed to inform future policy and government priorities in relation to interventions for children with conduct problems. We will continue our active engagement with policy makers and policy iniatives throughout the grant period.

4) Carers those adults who look after children with conduct problems (including parents, foster carers and adoptive parents, as well as kinship carers) often find it a challenging to understand and manage their child's emotional and behavioural problems. A fuller understanding of the specific difficulties, can make them easier to manage. The PI has contributed in an advisory / Co-I roles to recent school and clinic based intervention trials for children with different subtypes of conduct problems. The Co-I treats children with conduct problems in his clinical role and has authored a treatment manual for conduct problems. We will continue to engage with carers and communicate our research in ways that will help carers manage the parenting challenges they face.

5) Individuals with conduct problems. Ultimately the primary beneficiaries should be those who have conduct problems. This will be realised by achieving impact in relation to the above.