Large-Scale Functional Brain Network Architecture Changes Associated With Trauma-Related Dissociation.

Area of Science:

• Neuroscience
• Psychiatry
• Machine Learning

Background:

• Dissociative experiences are common after trauma but poorly understood.
• Current assessment relies on self-report, limiting clinical utility.
• Objective, brain-based measures are needed to assess dissociation severity.

Purpose of the Study:

• To test the sensitivity and robustness of a brain-based measure for dissociation severity.
• To explore the potential of functional MRI and machine learning for objective assessment.
• To investigate neural correlates of trauma-related dissociation.

Main Methods:

• Functional MRI scans from 65 women with childhood abuse and PTSD.
• Intrinsic network connectivity analysis using a novel machine learning technique.
• Assessment of continuous measures of trauma-related dissociation.

Main Results:

• Machine learning models moderately estimated dissociation severity, independent of trauma and PTSD severity.
• Key connectivity involved default mode and frontoparietal networks.
• Conventional group-based parcellation failed to estimate dissociation levels.

Conclusions:

• Network connectivity can estimate trauma-related dissociative symptoms.
• Between-network brain connectivity may offer unbiased biomarkers for dissociation.
• This approach advances understanding of dissociation's neural mechanisms.