Failure of anterior cingulate activation and connectivity with the amygdala during implicit regulation of emotional processing in generalized anxiety disorder

Abstract

Objective: Clinical data suggest that abnormalities in the regulation of emotional processing contribute to the pathophysiology of generalized anxiety disorder, yet these abnormalities remain poorly understood at the neurobiological level. The authors recently reported that in healthy volunteers the pregenual anterior cingulate regulates emotional conflict on a trial-by-trial basis by dampening activity in the amygdala. The authors also showed that this process is specific to the regulation of emotional, compared to nonemotional, conflict. Here the authors examined whether this form of noninstructed emotion regulation is perturbed in generalized anxiety disorder.

Method: Seventeen patients with generalized anxiety disorder and 24 healthy comparison subjects underwent functional MRI while performing an emotional conflict task that involved categorizing facial affect while ignoring overlaid affect label words. Behavioral and neural measures were used to compare trial-by-trial changes in conflict regulation.

Results: Comparison subjects effectively regulated emotional conflict from trial to trial, even though they were unaware of having done so. By contrast, patients with generalized anxiety disorder were completely unable to regulate emotional conflict and failed to engage the pregenual anterior cingulate in ways that would dampen amygdalar activity. Moreover, performance and brain activation were correlated with symptoms and could be used to accurately classify the two groups.

Conclusions: These data demonstrate that patients with generalized anxiety disorder show significant deficits in the noninstructed and spontaneous regulation of emotional processing. Conceptualization of anxiety as importantly involving abnormalities in emotion regulation, particularly a type occurring outside of awareness, may open up avenues for novel treatments, such as by targeting the medial prefrontal cortex.

Figure 1

Patients with generalized anxiety disorder fail to adapt to emotional conflict. (A) Example task time course, illustrating the contrasts made to examine adaptation during congruent or incongruent trials. (B) Reaction time difference scores reflecting the overall effect of emotional conflict (incongruent minus congruent trials), the facilitation in reaction times during emotional conflict adaptation (post-incongruent incongruent trials (iI) faster than post-congruent incongruent trials (cI), resulting in a negative reaction time difference score), and similar adaptation on congruent trials (post-congruent congruent trials (cC) faster than post-incongruent congruent trials (iC)). A group difference was found only during adaptation on incongruent trials. Error bars are standard errors of the mean. Two-sample t-test: * p<0.05, one-sample t-test: # p<0.05, ## p<0.01, ### p<0.001. The inset shows reaction times for each condition (see supplemental materials online for a detailed table).

Figure 2

Deficits in dorsomedial prefrontal activation to emotional conflict in patients. (A) Healthy comparison>patient contrast for the incongruent (inc) minus congruent (con) trial difference within the dorsomedial prefrontal region of interest, with each group’s data extracted for the cluster in (B), for both difference scores and individual trial types (inset). Only healthy comparison subjects were found to activate the dorsomedial prefrontal cortex in response to emotional conflict. Error bars are standard errors of the mean. Two-sample t-test: *** p<0.001, one-sample t-test: ## p<0.01.

Figure 3

Inability of patients to activate the pregenual cingulate and modulate dorsomedial prefrontal activity during emotional conflict adaptation. (A) Healthy comparison>patient contrast for the post-incongruent incongruent trial (iI) minus post-congruent incongruent trial (cI) difference within the pregenual cingulate region of interest, with each group’s data extracted for the cluster in (B), for both difference scores and individual trial types (inset). Only healthy comparison subjects were found to activate the pregenual cingulate. (C) Healthy comparison>patient contrast for the post-congruent incongruent trial (cI) minus post-incongruent incongruent trial (iI) difference within the dorsomedial prefrontal region of interest, with each group’s data extracted for the cluster in (D), for both difference scores and individual trial types (inset). Healthy comparison subjects were found to show less dorsomedial prefrontal activity in post-incongruent incongruent trials (hence a positive difference score). By contrast, in patients, there was inappropriately greater activity in the dorsomedial prefrontal cortex in response to post-incongruent incongruent trials (i.e. negative difference scores). Error bars are standard error of the mean. Two-sample t-test: ** p<0.01, one-sample t-test: # p<0.05, ## p<0.01.

Figure 4

Absence of negative pregenual cingulate-amygdalar connectivity during emotional conflict adaptation in patients. (A) Healthy comparison>patient contrast for the psychophysiological interaction functional connectivity analysis between the pregenual cingulate during post-incongruent incongruent trials versus post-congruent incongruent trials (iI-cI) and the amygdala, with each group’s clusters in (B). Only healthy comparison subjects showed robust negative connectivity between the pregenual cingulate and the amygdalae during post-incongruent incongruent trials. Error bars are standard errors of the mean. Two-sample t-test: ** p<0.01, *** p<0.001, one-sample t-test: ## p<0.01, ### p<0.001. Coronal sections are displayed in neurological convention (i.e. left=left).