SPECIFIC AIMS

This research proposal aims to identify mechanisms explaining why early adversity, namely child maltreatment, is consistently associated with an elevated risk for mental disorder 1-7. While there has been much emphasis on the long-term effects of adversity, including how genes and adversity jointly shape risk 8, there is limited research on how timing of adversity shapes trajectories of development and risk for internalizing symptoms.

The overarching concept we seek to test is that there are "sensitive periods" in development 9-11, or windows of time in the lifespan when the developing brain is particularly vulnerable or sensitive to experience, including adversity. Sensitive periods are characterized by phases of high plasticity and rapid brain development in a given domain 12; 13. First articulated by Hubel and Wiesel 14, a robust literature in animals 15; 16 and now in humans has identified sensitive periods in visual 17 and auditory system processes, including language development 18-22. Several genes (e.g., GAD1, GAD2, GABA1 alpha subunit, and BDNF) and environmental exposures (e.g., exercise, psychotropic medications) have been shown in animal studies to manipulate sensitive periods, either triggering their opening or closing 23. Importantly, recent literature also suggests sensitive periods are not fixed or limited solely to the early years of life and brain development. Instead, sensitive periods can be reopened throughout the lifecourse, even in adulthood, as a result of genetically-mediated factors and environments that "remove the brakes on plasticity" 24; 25.

Currently, we lack an in-depth understanding of the time-dependent effects of adversity on the Research Domain Criteria (RDoC) functional domains 26-29. This is especially true for social system processes, including attachment, social cognition, and social communication, which are likely precursors of internalizing symptoms. Although understudied, extant research suggests there are sensitive periods in attachment 30 and social cognition 31. Child abuse and neglect are known to affect social system processing, including perception of emotional expression32. However, it is unclear whether these effects are the same for all maltreated children or whether there is individual variation, depending on the age when children are first exposed. To that end, early evidence from epidemiological studies now suggests the effects of child maltreatment on internalizing symptoms differ depending on age at first exposure, which is suggestive of a sensitive period 33-35.

Based on prior research and our preliminary studies, the main hypothesis we are testing is that disruptions in social system processes explain the elevated risk of internalizing symptoms among children exposed to maltreatment and that the magnitude of this relationship will depend on the developmental timing of exposure to adversity and what sets of genes are expressed at that point in time.

HYPOTHESES

This proposal will test the following aims/hypotheses:

Aim 1: Explore the time-dependent effect of maltreatment on social domains/internalizing symptoms

Hypothesis 1: Children exposed to maltreatment during a sensitive period (age 3-5) will have lower functioning in social system domains and higher internalizing symptoms when compared to both non-exposed children and children exposed outside of the sensitive period, even after adjusting for maltreatment features (e.g., duration).

Aim 2: Examine the association between age at first onset of exposure to child maltreatment, social function, and internalizing symptoms in adolescents

Hypothesis 2: Social functioning will partially mediate or explain the association between age at first onset of exposure to child maltreatment and internalizing symptoms.

Aim 3: Investigate whether developmental genes modify the association between age at first exposure to child maltreatment and both social domain functioning and internalizing symptoms

Hypothesis 3: Variants in genes shown in animal studies to regulate sensitive periods, and sets of genes shown in humans to be differentially expressed across the lifespan, will modify the effect of adversity timing on domains of social functioning and internalizing symptoms.

The expected outcome of this proposal will be the development of a more nuanced approach to defining sensitive periods and the time-dependent effects of adversity on domains relevant to RDoC. This knowledge will have major impacts on the field, including greater insights into the biology of and mechanisms explaining susceptibility to internalizing symptoms and lead to the development of novel preventions and treatments.

EXPOSURE VARIABLES

Caregiver Reported Childhood Adversities: At yearly intervals between ages 1 and 8, caregivers provided information about whether the child was exposed to the different adversities in the past 12 to 18 months, including being taken into some form of foster care; physically hurt by someone else; sexually abused; separated from mother; and separated from father. This was included in the section entitled "Upsetting Events."

Child Reported Childhood Adversity: At age 16 children were asked to report whether they had been exposed to 23 different types of stressors since age 12.

Documented Reports of Child Maltreatment: We are also seeking information from the 329 children who were invested by social services for suspected maltreatment prior to age 6; this includes 162 children who were placed on local child protection registries as having confirmed cases of maltreatment.

MEDIATOR VARIABLES

Diagnostic Assessment of Non-Verbal Accuracy (DANVA): We are requesting data from the DANVA, which provides information about the offspring's ability to accurately detect facial displays of emotion. This task was administered at age 8.

Emotion Triangles Task: The child's ability to attribute emotional states to other individuals, a component of theory of mind, was assessed at age 13, using the emotion triangle task.

Separation Anxiety, as measured by the Development and Wellbeing Assessment (DAWBA), was asssessed at ages 7, 10, 13, 15.

Social and Communication Disorders Checklist: We are requesting data from this measure, which taps domains of social cognition. It appears this measure was administered to children at ages 7 and 15 (and possibly 10 and 13).

OUTCOME VARIABLES

We are proposing to examine the following outcome variables: (1) Rutter parent scale for preschool children; (2) Strengths and difficulties questionnaire; (3) Development and Wellbeing Assessment; (4) Short Mood and Feelings Questionnaire. For all of these outcomes, we will examine symptoms of mood and anxiety disorders (as a continuous measure) as well as extremes in the symptom distribution or diagnoses (from DAWBA).

EFFECT MODIFIERS

We are proposing to examine sets of genes as effect modifiers of the association between maltreatment and both social system processing and mood and anxiety disorders. Please see Appendix 1B for further details on the genetic information we are requesting.

COVARIATES

We would also be interested in receiving basic demographic variables to use as covariates in our analysis. This includes information about: age, gender, race/ethnicity, socioeconomic status.