Neurobiological correlates of externalizing and prosocial behavior in school-age children

This thesis describes a series of studies on the neurobiological correlates of externalizing and prosocial behavior in six-to ten-year old children. Chapter 1 provides an outline and describes the background and aims of our work. The studies described in this thesis are embedded in the Generation R study, a prospective cohort from fetal life onwards in Rotterdam, the Netherlands. We describe both structural (chapter 2, 3, and 6) and functional neuroimaging studies (chapter 4 and 5) on the association between externalizing and prosocial behavior and the brain, and examine behavior both from a trait-like perspective (chapter 2 and 3) as well as from a state-like perspective (chapter 4 and 5).

In chapter 2 and 3, we explored the neuroanatomical correlates of aggressive and prosocial behavior in six-to-nine year old children. Aggressive behavior was associated with smaller amygdala, but not smaller hippocampal volume. Moreover, aggression was associated with a thinner cortex in sensorimotor regions, and widespread decreased right hemisphere gyrification. Finally, aggressive behavior was associated with cortical thickness in regions that are part of the default mode network (superior and middle frontal gyrus, precuneus, posterior cingulate cortex), with positive associations in girls and negative associations in boys. Prosocial behavior was related to a thicker cortex in a cluster that included part of the left superior frontal and rostral middle frontal cortex. Moreover, the association between prosocial behavior and cortical thickness was different for boys and girls in similar regions as reported in relation to aggressive behavior (superior and middle frontal cortex, cuneus, precuneus).

In chapter 4, we aimed to differentiate between honest children, children who showed typical lie-telling behavior and children who lied more persistently, and we examined what demographic, cognitive, social and neurobiological factors are associated with lie-telling behavior. To this aim, we observed lying in a condition with low perceived lie-detectability and a condition with high perceived lie-detectability. Children who were honest in both conditions were referred to as honest or persistently honest. Children who were dishonest in one condition only -the largest group- were referred to as typical lie-tellers (chapter 4) or low lie-detectability lie-tellers (chapter 5). Participants who lied in both the low liedetectability and the high lie-detectability condition were considered persistent lie-tellers. Of the 163 children, 75% lied in the low lie-detectability condition, compared to 34% in high-risk condition. Persistent lie-tellers could be discriminated from other children based on gender (more boys), lower age, lower IQ, less effortful control, and lower educated mothers. Compared to honest children and persistent lie-tellers, typical lie-tellers were more likely to be girls and to come from families with higher incomes. The neurobiological data mimicked our behavioral results in showing similar neural patterns in honest children and typical lie-tellers, while persistent lie-tellers differed from typical lie-tellers and honest children in showing less brain activation of the anterior cingulate cortex and frontal pole during the task. These findings suggest that persistent lie-tellers engage in less conflict monitoring and error detection during lie-telling.

In chapter 5 we examined whether the neurobiological correlates of lie-telling and honesty are dependent upon perceived lie-detectability by comparing neural activation in the high lie-detectability condition with the low lie-detectability condition for each group separately. Compared to low perceived lie-detectability, high perceived lie-detectability was related to increased activation in regions implicated in social cognition (persistently honest children), autonomic control (low lie-detectability lie-tellers), and decision making, error detection or conflict monitoring (persistent lie-tellers). While persistently honest as well as persistently dishonest children showed similar behavior in the low lie-detectability condition and high lie-detectability condition, in both groups of children the high perceived lie-detectability condition was associated with an increase in brain activation compared to the low perceived lie-detectability condition. Thus, situational characteristics (i.e. liedetectability) affect brain activation patterns of honest and dishonest behavior.

In chapter 6, we describe the prospective relation between mothers’ and fathers’ sensitive caregiving in early childhood and brain morphology later in childhood. As quality of parenting has been associated with child externalizing behavior, examining the association between parental sensitivity and brain morphology may provide information on mechanisms through which parenting influences child behavior. More parental sensitivity in early childhood was associated with larger total brain volume and gray matter volume at 8 years, controlling for infant head size. Moreover, sensitivity was associated with a thicker cortex of the bilateral precentral, and middle frontal gyri. The results illustrate the important role of parents in child brain development.

In chapter 7, the main findings of the studies reported in this thesis are summarized. Moreover, the chapter describes methodological considerations, and practical implications along with suggestions for future research.