Between-Network Brain Architecture Indicative of Trauma-Related Dissociative Symptoms

brain MRI
brain MRI
Researchers tested for continuous measures of trauma-related dissociation using the Multidimensional Inventory of Dissociation. Connectivity estimates were derived with a novel machine learning technique using individually defined homologous functional regions for each participant.

Researchers from Harvard Medical School observed that between-network brain connectivity may allow for estimating severity of trauma-related dissociative symptoms. These findings, from an intrinsic network connectivity analysis, were published in the American Journal of Psychiatry.

Women (N=65) seeking treatment at McLean Hospital were recruited for this study. All participants had a history of childhood maltreatment, posttraumatic stress disorder (PTSD), and dissociative symptoms. Participants underwent magnetic resonance imaging scans while performing a series of tasks and were assessed by the Multidimensional Inventory of Dissociation score.

The severity of dissociation symptoms among the participants was observed to be significantly correlated with brain connections involving the frontoparietal control and default mode networks.

With a support vector machine for regression algorithm approach controlling for PTSD and childhood trauma, the association between predicted and observed severe dissociation scores was found to be significant (z=2.52; P <.006) and group-level connections were less correlated with symptom scores compared with individualized regions (t=8.82; P <.001).

Between-network connections were observed to contribute more to patient symptoms, especially involving visual, frontoparietal control, and default mode networks. These between-network connections among patients were reduced overall by an average of 5.15% compared with atlas-defined networks. Despite the reduced connections, they better predicted patient-specific symptoms.

The size of functional regions was predictive for symptoms scores (r=0.442; P =.018), in which dissociation severity scores were negatively correlated with ventral attention network size (r=-0.277; P =.031) and positively correlated with network size (r=0.271; P =.035).

Dissociation severity scores combined with region sizes were found to be the superior predictive models (r=0.549; P <.001).

This study was limited by the fact that participants were on various psychiatric medications, which may have altered some brain connectivity patterns. This study did not confirm these findings by an independent cohort.

These results provided biological evidence of trauma-related dissociation independent of PTSD and childhood trauma, giving legitimacy to perceived psychiatric symptoms, and indicated that unique patient connectivity networks may allow for better patient diagnosis and risk stratification.

Disclosure: Multiple authors declared affiliations with industry. Please refer to the original article for a full list of disclosures.


Lebois LAM, Li M, Baker JT, et al. Large-scale functional brain network architecture changes associated with trauma-related dissociation [published online September 25, 2020]. Am J Psychiatry. doi: 10.1176/appi.ajp.2020.19060647.