Incarcerated youth with history of childhood trauma are characterized by altered neural activation during a Go/NoGo fMRI task

Background Childhood trauma has been linked to abnormal neurobiological functioning during error-processing and response inhibition tasks. However, it remains unclear whether trauma subtypes (e.g., directly experienced vs. observed trauma) are associated with distinct neurobiological alterations. To test this question, the present study examined whether childhood trauma subtypes, assessed using the Trauma Checklist (TCL) 2.0, were differentially associated with abnormal neurobiological processing during a response inhibition task (the Go/NoGo task) among high-risk incarcerated juvenile males ( n = 225). Methods Analyses included a general linear model (GLM) with a priori regions of interest (ROI) and a whole-brain data-driven independent component analysis (ICA). ICA functional measures included inter- and intra-network connectivity and power spectra. We tested the association of childhood trauma as indexed by TCL Factor 1 (experienced trauma) and TCL Factor 2 (community trauma) with these functional measures. Results GLM analyses indicated that TCL Factor 2 was associated with increased error-processing activation in the anterior cingulate cortex (ACC) and insula, while TCL Factor 1 was associated with increased response inhibition activity in the secondary visual cortex. ICA analyses revealed more widespread effects during error-processing and response inhibition. TCL Factor 2 scores were associated with increased intra-network connectivity in the ACC, and increased high-frequency power spectra/decreased intra-network connectivity in cognitive control, auditory, visual, and cerebellar domains. TCL Factor 1 scores were associated with increased high-frequency power spectra and intra-network connectivity in subcortical regions. Conclusion Overall, our results suggest that childhood trauma among incarcerated juvenile males is associated with increased error-processing activity and abnormal intra-network dynamics across multiple networks including predominately paralimbic regions. Trauma subtypes displayed primarily divergent effects. These effects align with previous findings in community samples. Finally, our results provide additional support for childhood adversity-induced hypervigilance when processing errors that may be compensatory in nature, though do not rule out alternative explanations such as neural inefficiency, increased cognitive effort, or alternative processing strategies.