Advancing understanding of affect labeling with dynamic causal modeling.
Level 4 - case-series / case-control
Observational mechanistic fMRI neuroimaging study in healthy volunteers
PubMed 23774393 · doi:10.1016/j.neuroimage.2013.06.025
What was done
Dynamic Causal Modeling (DCM) for fMRI was applied to 45 healthy participants during a facial affect labeling task. Researchers modeled effective connectivity across four brain regions, including the right ventrolateral prefrontal cortex (vlPFC), Broca's area, and the amygdala. Sixty-four models were evaluated per subject using family-level Bayesian Model Selection and Bayesian Model Averaging to identify driving inputs and modulatory connectivity changes induced by affect labeling.
What was found
Family-level inference identified a winning family of 32 models. Bayesian Model Averaging revealed that affect labeling exerted a dampening influence on the amygdala originating from Broca's area and more strongly from the right vlPFC. Aside from model counts (64 tested, 32 in winning family), the abstract provides no exact numerical values, effect sizes, or confidence intervals for the connectivity parameters.
Why it matters
This study provides directional effective connectivity evidence supporting the role of prefrontal-to-subcortical pathways, specifically right vlPFC and Broca's area to amygdala, in incidental emotion regulation through affect labeling.
Limits
The study evaluated a small sample of 45 healthy subjects, limiting direct generalization to clinical populations. Modeling was constrained to four predefined brain regions, omitting other potentially relevant nodes. The abstract lacks quantitative connectivity values, variance estimates, and behavioral outcome data.