Tracking the Impact of Childhood Adversity on Brain Development
Poster No:
664
Submission Type:
Abstract Submission
Authors:
Katharina Brosch1, Lisa Wiersch1, Erynn Christensen1, Elvisha Dhamala1
Institutions:
1Feinstein Institutes for Medical Research, Manhasset, NY
First Author:
Co-Author(s):
Introduction:
Adverse childhood experiences are established risk factors for the subsequent development of psychopathology. While specific types of adversity, such as neighborhood threat or physical and sexual abuse have been linked to unique brain structural changes, the impact of these traumatic experiences on neurodevelopment remains poorly understood. Here, we investigated the association between different adversity factors and cortical thickness in a large, longitudinal sample of children assessed at ages 9-10, 11-12, and 13-14 years. This allowed us to examine brain regions particularly sensitive to these adverse experiences across critical developmental stages.
Methods:
Data from the Adolescent Brain Cognitive Development (ABCD) study were used to establish associations between adverse childhood experiences and cortical thickness. Six adversity factors, previously identified in a factor analysis by Orendain et al. (2023), were considered: 1) Physical and Sexual Violence, 2) Neighborhood Threat, 3) Prenatal Substance Exposure, 4) Household Dysfunction, 5) Scarcity, and 6) Parental Psychopathology. Factors were calculated from 47 items using 14 different questionnaires. Further, socioeconomic status was included as an additional factor. Data were analyzed at baseline (age 9-10, N=6885; f=3268), two-year follow-up (age 11-12, N=5636; f=2609), and four-year follow-up (age 13-14, N=2179; f=1016). We trained brain-based predictive models using linear ridge regression algorithms and k-fold cross validation to predict adversity factors at the three time points using regional measures of cortical thickness (Desikan-Killiany atlas). Models were trained in the entire sample, and sex-specific models were trained to capture sex-specific associations between cortical thickness and adversity. P-values were adjusted for multiple comparisons using False-Discovery Rate.
Results:
At baseline, the Neighborhood Threat-trained model was significantly associated with cortical thickness in the entire sample (r=0.14, padj.< .001), and generalized to the association between Scarcity and cortical thickness (r=0.12, padj.< .001). The Scarcity-trained model was significantly associated with cortical thickness (r=0.15, padj.< .001), and in turn generalized to Neighborhood Threat (r=0.12, padj.< .001). The Socioeconomic Status-trained model was significantly associated with cortical thickness (r=0.28, padj.< .001), but the model did not generalize to any other risk factor. At 2-year follow-up, the Neighborhood Threat-trained model (r=0.11, padj.< .001), and the Socioeconomic Status-trained model (r=0.27, padj.< .001) were again significantly associated with cortical thickness but did not generalize to other factors. At 4-year follow-up the Socioeconomic Status-trained model (r=0.24, padj.< .001) was significantly associated with cortical thickness but did not generalize to other factors.
·Figure 1: Tested models at three different time-points.
Conclusions:
Our findings shed important light on the effect of specific adversity at different developmental stages on cortical thickness. We demonstrate that Neighborhood Threat and Scarcity are significantly associated with cortical thickness measures at 9-10 years, and models trained on either, also generalize to the other. We demonstrate that while Socioeconomic Status, another potential risk or protective factor, is significantly associated with cortical thickness at ages 9-10, 11-12, and 13-14, the trained model does not generalize to other adversity factors. This suggests that Socioeconomic Status alone is insufficient for characterizing effects on cortical thickness, highlighting the need to consider additional environmental factors such as Neighborhood Threat and Scarcity. Our findings highlight age-specific effects of adversity on neurodevelopment and may inform personalized interventions to support resilience during critical periods in childhood.
Emotion, Motivation and Social Neuroscience:
Social Neuroscience Other 1
Lifespan Development:
Early life, Adolescence, Aging 2
Keywords:
Development
Machine Learning
Psychiatric
Trauma
Other - ACE
1|2Indicates the priority used for review