- At baseline, anxious youth exhibited lower density within the central executive network and between central executive and default mode networks than low or no anxiety youth.
- Exposure-focused CBT selectively increased connectivity between the central executive and salience networks, consistent with enhanced top down control of threat signals.
- Relaxation mentorship training increased within central executive network density, indicating dissociable neural effects with implications for treatment personalisation.
J Am Acad Child Adolesc Psychiatry. 2026 Nov 5:S0890-8567(26)02216-1. doi: 10.1016/j.jaac.2026.10.024. Online ahead of print.
ABSTRACT
OBJECTIVE: Exposure is considered the most active element of cognitive behavioral therapy (CBT) for pediatric anxiety. Yet many youth do not respond to CBT, and exposure is the component patients most often find difficult to tolerate. Cognitive control and its supporting neural substrates may enhance exposure efficacy by helping youth remain engaged with feared stimuli despite urges to avoid. However, little work has identified how exposure-based CBT modulates intrinsic connectivity of tripartite networks that support cognitive control (e.g., central executive [CEN], salience [SN], and default mode networks [DMN]), limiting efforts to develop strategies to improve and personalize treatment. Progress may be limited by individual variability in network connectivity in anxiety, which may obscure treatment-related effects in group-averaged analyses. This randomized clinical trial (RCT) leverages person-specific network modeling to test how exposure versus relaxation components of CBT influence resting-state connectivity of cognitive control networks in pediatric anxiety.
METHOD: Youth aged 7-18 years with an anxiety disorder (N = 100) or low/no anxiety (L/NA; N = 35) completed resting-state fMRI scans at baseline. Patients were randomized to exposure-focused CBT (EF-CBT) or relaxation mentorship training (RMT). Resting-state connectivity was reassessed following treatment (or after a commensurate interval in L/NA youth) in 112 participants. Changes in within-CEN, CEN-SN, and CEN-DMN density were examined using Group Iterative Multiple Model Estimation, which yields sparse, person-specific networks capturing both shared and individual connectivity structure.
RESULTS: At baseline, anxious youth exhibited lower density within-CEN and between CEN-DMN than L/NA youth. Treatment effects differed by intervention: EF-CBT selectively increased CEN-SN density, whereas RMT increased within-CEN density.
CONCLUSION: These findings demonstrate dissociable effects of exposure and relaxation on cognitive control network organization in pediatric anxiety. These results indicate that exposure-focused CBT is uniquely associated with increased connectivity between control and salience systems, consistent with a mechanism supporting top-down control of threat-related signals during exposure. Person-specific measures of tripartite network connectivity may inform future efforts to identify candidate neural targets for optimizing and personalizing treatment.
PMID:42988353 | DOI:10.1016/j.jaac.2026.10.024
Share Evidence Blueprint
Save to Google Notes
Search Google Scholar
Save as PDF
⭐ My Revision List
