Development of brain mechanisms underlying threat bias: relations with childhood social reticence and adolescent anxiety

• In Event: Poster Session 08
  In Poster Session: PS 08 Topic - Developmental Psychopathology

Abstract

Social reticence during early childhood predicts the development of anxiety symptoms, typically emerging during adolescence. However, not all socially reticent children develop anxiety symptoms. Therefore, it is essential to determine which factors moderate the relation between social reticence and pathological anxiety. Here, we focused on one such potential moderator: neurodevelopmental processes associated with threat bias. Previous research showed that amygdala-prefrontal cortex functional connectivity when attending to threat stimuli moderated relations between early-childhood temperament and anxiety symptoms at 10 and 13 years. The current study extends this research by assessing neural correlates of threat bias across adolescence (10-16 years) and testing relations with childhood social reticence and concurrent anxiety symptoms.

This study is part of a longitudinal study of temperament. Children from two samples were included: one selected on high negative, high positive or non-extreme temperaments at four months (n=291), and one randomly recruited from the community at age 2 (n=384). Social reticence was measured when children were 2, 3, 4, 5 and 7 years old using behavioral observations and parent-report questionnaires; children were divided into high and low social reticence groups (mean split). Self-reported and parent-reported anxiety symptoms were measured when children were 10, 13 and 16 years old. Threat bias was measured concurrently using the dot-probe task; children were presented with pairs of faces (angry-neutral, happy-neutral, or neutral-neutral) and responded to the location of a probe replacing one of the faces. Attention bias variability, measuring within-session temporal variability in attention allocation, served as the behavioral measure. Useable fMRI data were collected during the task for 73 children at 10 years of age (M=10.64 years, SD=0.47, 43 girls), 92 children at 13 years (M=13.25 years, SD=0.75, 47 girls) and 78 children at 16 years (M=16.27 years, SD=0.67, 38 girls). All available data were included. 51 children had data at 2 time points, 20 children had data at 3 time points

Linear mixed-effects models were used to test the effects of social reticence, concurrent anxiety symptoms, task condition, time, and their interactions on attention bias variability and fMRI data, including subject as random effect. For fMRI data, we focused on amygdala functional connectivity assessed with a psychophysiological interaction (PPI) analysis. There were no significant effects for attention bias variability. For fMRI, there was a significant three-way interaction between social reticence, concurrent anxiety and time for functional connectivity between the left amygdala and the right dorsolateral prefrontal cortex (dlPFC; -36.2, -43.8, 16.2; 34 voxels; p<0.001 FDR). Decomposition of this interaction (see figure) indicated that children with high social reticence scores showed a negative relation between anxiety symptoms and amygdala-dlPFC connectivity at 13 and 16 years. In contrast, children with low social reticence scores showed a positive relation between anxiety symptoms and amygdala-dlPFC connectivity at all time points. This finding was not specific to condition, which might suggest that this is a more general effect of processing emotional faces. This study extends previous work, identifying distinct neurodevelopmental pathways to anxiety for children showing high versus low social reticence during childhood.

Authors

• Anita Harrewijn, National Institute of Mental Health
  Presenting Author

• Sonia Ruiz, National Institutes of Health
  Non-Presenting Author

• Rany Abend, National Institutes of Health
  Non-Presenting Author

• Simone Haller, National Institutes of Health
  Non-Presenting Author

• Anni Subar, University of Denver
  Non-Presenting Author

• Caroline Swetlitz, Boston University
  Non-Presenting Author

• Emilio Alejandro Valadez, University of Maryland - College Park
  Non-Presenting Author

• Melissa Brotman, National Institutes of Health
  Non-Presenting Author

• Andrea Chronis-Tuscano, University of Maryland, College Park
  Non-Presenting Author

• Ellen Leibenluft, National Institutes of Health
  Non-Presenting Author

• Nathan A Fox, University of Maryland - College Park
  Non-Presenting Author

• Daniel S. Pine, National Institutes of Health
  Non-Presenting Author