The work of our Unit aims to better understand why and how mental health problems emerge, and the nature of risk and resilience in childhood. There is a need to better understand the risk mechanisms underlying common childhood disorders. But we should not only be interested in risk, it is equally important to understand what contributes to a positive outcome in the face of early adversity.

Research into developmental risk and resilience has scope to inform professionals and policy makers in ways that will improve lives of children and families. Our work adopts an interdisciplinary research strategy that combines cognitive experimental, twin-model-fitting, fMRI, MRI and genotyping methods and can be thought of within two main themes.

Resilience and childhood maltreatment:

It is now well established that childhood maltreatment is a key risk factor for poor outcomes later in life – from an increased risk of mental health problems, to lower economic productivity. Yet many of these effects are not immediately evident nor do they characterise all children who experience maltreatment. This suggests that for some children experiences of early adversity ‘get under the skin’ in ways that can embed long-term vulnerability. How and why does this happen? In a recent paper, we have introduced the concept of “Latent Vulnerability”. According to this theory, children adapt to early chaotic, unpredictable or violent home or community settings in ways that help them survive and cope at both the psychological and biological levels. However, we suggest that these adaptations come at a high price: they can carry a lifelong increased risk of mental health problems. Children adapt to ‘fit’ atypical and disturbed environments in ways that may not be helpful when they go out into the world to make friends, learn and develop as adults.

For example, we have shown that exposure to physical abuse and domestic violence leads to heightened reactivity in ‘threat’ related circuitry, including the amygdala and anterior insula. This may represent a form of ‘adaptation’ to early risk environments, which may confer short-term functional advantages for the child. However, heightened responsiveness of such brain structures to threat could also constitute a latent neural risk factor that predisposes to later mental health problems. It may, for example, make it more difficult for individuals to cope with future stressors, which could be experienced as overwhelming. There is research in support of this notion. We know from other studies of adults that high levels of amygdala reactivity to threat stimuli increase risk of mental health symptoms if individuals are exposed to future stressors such as combat or stressful life events in the community.

Such heightened vigilance to threat is only one possible neurocognitive mechanism that may confer latent vulnerability. We are currently investigating a number of other possible mechanisms, including risk and reward processing and autobiographical memory.

Over the long term, a more refined understanding of the mechanisms associated with Latent Vulnerability may enable the identification of a subgroup of children exposed to maltreatment who are at particularly high risk for psychopathology. One of our longer-term aims is to improve our approach to prevention, and improve our understanding of how we might intervene to prevent disorders before they emerge.

Development of antisocial behaviour:

Conduct problems (CP), in other words antisocial behaviour in children and adolescents, incur considerable individual and societal costs and constitute the most common reason for referral to child mental health and educational services. In the longer term, they are associated with incarceration, and poor mental and physical health outcomes.

Our programme of research has focused on studying the aetiology and neurocognitive processes associated with different types of CP. Specifically, we have been interested in the role of callous-unemotional (CU) traits in defining distinct subgroups of children with CP. Our work using twin study, experimental, and neuroimaging methodologies has provided converging evidence that high levels of CU traits delineate a sub-group of children with CP who appear distinct from their peers with low levels of CU traits in aetiology and neurocognitive vulnerability.

Our research was the first to demonstrate that CP with high levels of CU traits are strongly heritable, whereas CP with low levels of CU traits are not. This does not mean that children with high levels of CU traits are destined to become antisocial, but it does indicate that they have a genetic vulnerability. However, such vulnerability is not destiny. Knowing how vulnerable children develop over time and what protective factors can influence the direction of the development, will be important if we want to help these children and their families. Such help is likely to involve behavioural interventions for the children and the families, which take into account both the specific ways in which the child sees the world around them, as well parenting and peer relationships.

Much of our current work is concentrated on understanding how children with different types of CP see the world around them. We have demonstrated that those with CP and high CU traits have difficulty in resonating with other people’s emotions. Although they can understand what other people think, they tend to feel less empathy for other people’s distress. By contrast children with CP and low levels of CU traits appear, if anything, over reactive to emotional stimuli and often have trouble regulating their emotions. We can also see this at the neural level. Conduct problems with high CU traits are associated with lower brain activity in emotion processing areas (e.g. amygdala) compared with typically developing children. Those who have CP and low levels of CU traits have the opposite pattern of brain activity; they look ‘overactive’ compared to typically developing children.

Recently we have become equally interested in how children with CP process positive affect. Through this research we hope to better understand why some children with CP do not show the typical need to form close bonds and please other people.

Our work has highlighted that children with CP are not all the same. This means that clinicians and educators may have to consider some different tactics for children with high vs. low levels of CU traits. As we proceed systematically with our work, we hope to gain an ever more nuanced picture of what puts children at risk and what protects them from developing antisocial behaviour. Current interventions work for many children, but hopefully, as we come to better understand the risk and resilience factors at play, we can become better still in helping the children with CP and their families.

Funding Sources

Our research is currently funded by the ESRC, MRC, Royal Society, and the Anna Freud Centre.