Does CBT change the brain? A systematic review of neuroimaging RCTs on anxiety disorders, depression and obsessive-compulsive disorder
Abstract
Cognitive behavioral therapy (CBT) is effective for anxiety disorders, depression and obsessive-compulsive disorder (OCD), but the neurobiological mechanisms of change remain unclear. This systematic review synthesized neuroimaging evidence from randomized controlled trials (RCTs) to examine whether CBT can be considered a neurobiological intervention. Following PRISMA guidelines (PROSPERO: CRD420251239465), PubMed, Scopus and PsycInfo were searched for RCTs of CBT in adults with anxiety disorders, depression or OCD-spectrum disorders reporting pre- and post-treatment neuroimaging. Risk of bias was assessed with RoB 2, and findings were synthesized narratively. Seventeen RCTs were included (OCD = 5, anxiety disorders = 6, depression = 6), using task-based fMRI (n = 8), resting-state fMRI (n = 6), structural MRI (n = 2) or PET (n = 1). Changes after CBT involved emotional reactivity regions, prefrontal and frontoparietal control systems and their connectivity, with disorder-specific patterns: reduced threat reactivity in anxiety disorders, altered cognitive control networks in OCD, and altered mood-related, self-referential and reward connectivity in depression. Neural and clinical change did not always coincide, and comparisons with pharmacotherapy were mixed. Five studies were at high risk of bias. These findings provide preliminary support for CBT as a neurobiological intervention, although small samples and heterogeneous protocols limit conclusions.
Related
- Whole-exome sequencing in individuals with obsessive–compulsive disorder and chronic tic disorders identifies 36 large-effect risk genes
- Autism Spectrum Disorder, Attention‐Deficit/Hyperactivity Disorder, Tic Disorder, and Obsessive‐Compulsive Disorder in Individuals With Eating Disorders and Their Siblings: A Nationwide Finnish Cohort Study
- Systematic Review and Meta-Analysis of Cognitive Bias Modification for Interpretation in Obsessive-Compulsive Disorder
- Reduced cortical thickness in obsessive-compulsive disorder and candidate molecular correlates
- Intensive cognitive behavioral therapy for obsessive-compulsive disorder: a meta-analysis of randomized controlled trials
BibTeX
@misc{kokka2026does,
title = {{Does CBT change the brain? A systematic review of neuroimaging RCTs on anxiety disorders, depression and obsessive-compulsive disorder}},
author = {Ioulia Kokka and Iraklis Mourikis},
howpublished = {PsyArXiv},
year = {2026},
month = nov,
doi = {10.31234/osf.io/fuqjb_v1},
url = {https://osf.io/preprints/psyarxiv/fuqjb_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Does CBT change the brain? A systematic review of neuroimaging RCTs on anxiety disorders, depression and obsessive-compulsive disorder
AU - Ioulia Kokka
AU - Iraklis Mourikis
PY - 2026
DA - 2026/11/09
PB - PsyArXiv
DO - 10.31234/osf.io/fuqjb_v1
UR - https://osf.io/preprints/psyarxiv/fuqjb_v1/
AB - Cognitive behavioral therapy (CBT) is effective for anxiety disorders, depression and obsessive-compulsive disorder (OCD), but the neurobiological mechanisms of change remain unclear. This systematic review synthesized neuroimaging evidence from randomized controlled trials (RCTs) to examine whether CBT can be considered a neurobiological intervention. Following PRISMA guidelines (PROSPERO: CRD420251239465), PubMed, Scopus and PsycInfo were searched for RCTs of CBT in adults with anxiety disorders, depression or OCD-spectrum disorders reporting pre- and post-treatment neuroimaging. Risk of bias was assessed with RoB 2, and findings were synthesized narratively. Seventeen RCTs were included (OCD = 5, anxiety disorders = 6, depression = 6), using task-based fMRI (n = 8), resting-state fMRI (n = 6), structural MRI (n = 2) or PET (n = 1). Changes after CBT involved emotional reactivity regions, prefrontal and frontoparietal control systems and their connectivity, with disorder-specific patterns: reduced threat reactivity in anxiety disorders, altered cognitive control networks in OCD, and altered mood-related, self-referential and reward connectivity in depression. Neural and clinical change did not always coincide, and comparisons with pharmacotherapy were mixed. Five studies were at high risk of bias. These findings provide preliminary support for CBT as a neurobiological intervention, although small samples and heterogeneous protocols limit conclusions.
ER -