An Examination of the Neural Correlates of Emotion Acceptance Versus Worry in Generalized Anxiety Disorder
Abstract: Worry in generalized anxiety disorder (GAD) has been conceptualized as an emotion regulation strategy whereby distress triggered by uncertain outcomes is temporarily avoided through internal verbalizations and associated reductions in autonomic arousal. However, worry is ineffective, as subsequent exposure to worry topics elicits continued distress and generalizes to new topics. One potential mechanism by which worry may perpetuate distress is by preventing the development of greater tolerance of internal distress. By contrast, acceptance as an emotion regulation strategy is associated with reduced distress and increased willingness to re-engage in previously distressing tasks. Acceptance may exert its influence by keeping patients in contact with internal distress and preventing avoidance, allowing for repeated exposure and the subsequent development of greater distress tolerance. The present study used functional magnetic resonance imaging to explore neural mechanisms of worry and acceptance in a sample of 21 women meeting DSM-IV criteria for GAD. During scanning, participants read personally relevant worry statements, and were presented with randomized instructions to observe and accept their reactions (Accept), worry as usual (Worry), or suppress reactions (Suppress). Participants were then randomized to receive either brief training in acceptance-based regulation strategies, or no training. Pre-training, the Worry condition was associated with increased limbic (amygdala, insula, hippocampus) and ventromedial prefrontal cortex (vmPFC) activation relative to both the Accept and Suppress conditions, whereas the Accept condition was associated with greater dorsomedial PFC (dmPFC) activation, and the Suppress condition was associated with greater dorsolateral PFC activation. Following training in emotion acceptance, the Training group evidenced significant reductions in amygdala and hippocampus activation in the Worry condition. These results were not found in the No-Training group. Additionally, decreased activation in the Worry and Accept conditions was found in the Training group relative to No-Training in the subgenual anterior cingulate, a vmPFC region implicated in extinction learning. Decreased activation in this region was significantly correlated with participant ratings of lower subjective distress and greater regulation success. Results of this study suggest potential mechanisms by which emotion acceptance may exert effects in the regulation of distress are through neural pathways implicated in both cognitive control and extinction learning.