In support of our primary hypothesis, the PTSD group exhibited lower rACC activation to trauma-unrelated interference, as compared to the TENP group. These results are consistent with those of Kim and colleagues [13] and suggest that diminished rACC function may reflect a more general abnormality in processing emotional material in PTSD.

We did not find any significant between-group differences on response times or error rates. It is possible that rACC dysfunction in PTSD is not behaviorally apparent until stimuli are more salient (trauma-related). Indeed, a recent meta-analysis has suggested that trauma-unrelated emotional word stimuli do not elicit greater behavioral interference in PTSD participants, as compared to controls [28]. However, previous studies that used trauma-related stimuli also failed to find significant behavioral impairment despite showing diminished rACC activation in PTSD [6, 7]. It should be noted, however, that the latter study measured only accuracy (not response times) and the former study did report greater group differences in behavioral interference than the current study. Thus, it remains possible that behavioral evidence of rACC dysfunction is stronger when the stimuli are more salient. Regardless, the lack of behavioral differences in the present study suggests that our neuroimaging results cannot be attributed to group differences in behavioral performance.

Consistent with previous research, we found a trend toward a negative correlation between total symptom severity and activation in the rACC proper [11–13]. We also found significant negative correlations between total symptom severity and activation of structures adjacent to the rACC: the MFG and the dACC. Previous studies have revealed negative correlations between symptom severity and MFG activation (e.g., [29, 30]). The observed dACC cluster lies 6mm away from the rACC boundary as defined herein. Given the smoothing kernel (7 mm), this cluster could reflect function of the rACC. Importantly, we also found that only re-experiencing symptom severity (CAPS-B) was significantly negatively correlated with activation in rACC proper. That the degree of diminished rACC activation during emotional interference may reflect the severity of re-experiencing symptoms per se makes sense given that emotional interference tasks were originally intended to tap into the cognitive processes underlying re-experiencing symptoms of PTSD [3].

If the rACC deficit in PTSD is more general (i.e., not specific to trauma-related material), then rACC activation to emotional interference should correlate negatively with PTSD symptom severity regardless of whether the emotional stimuli are trauma-related. Consistent with this hypothesis, Kim et al. [13] found that rACC activation to trauma-unrelated emotional stimuli is negatively correlated with PTSD symptom severity. In addition, we found a similar negative correlation in the current study between rACC activation and CAPS-B re-experiencing symptoms. Unfortunately, correlations between brain activation and symptom severity were not reported in other previous neuroimaging studies of emotional interference in PTSD [6, 7, 14].

In summary, our findings suggest that relatively diminished rACC function in PTSD may reflect a more generalized abnormality that can be observed even when emotional stimuli are unrelated to trauma. This finding also has implications for the planning of future neuroimaging studies that examine rACC function in groups with diverse trauma histories. Individually tailoring stimuli to match each participant’s traumatic event may not be necessary as emotional interference tasks need not contain trauma-related stimuli in order to reveal rACC abnormalities in PTSD.