Dynamic causal modeling of fMRI time series in response to fearful faces

Dynamic causal modeling was used to quantify effective connectivity among limbic regions within each brain hemisphere. fMRI data obtained from seven healthy, normal subjects belonging to the low trait anxiety group during passive viewing of emotionally-valent facial stimuli was used for connectivity analyses. A fully connected and reciprocal model consisting of the amygdala, hippocampus and prefrontal cortex regions was defined and intrinsic connections between each of the components of the model were estimated for both the left and right hemispheres. We hypothesized that the strongest components within the model will be inhibitory in nature. No evidence supporting predominantly inhibitory mechanism was observed in our preliminary modeling results. Our results indicate the existence of distinct effective connectivity patterns between the two hemispheres.