Abstract
Aberrant reward processing has been predominantly associated with depressive disorders, with evidence that pre-treatment abnormalities in striatal reward responsiveness relates to treatment outcomes. Emerging research also implicates reward processing differences in anxiety disorders, particularly generalized anxiety disorder (GAD). The current study examined whether pre-treatment reward- and loss-related neural activity predicts symptom improvement with behavioral activation (BA) and exposure therapy (EXP) for GAD. In this randomized clinical trial (ClinicalTrials.gov NCT02807480) conducted from 2016 to 2021, treatment-seeking adults with GAD completed the monetary incentive delay task during functional magnetic resonance imaging pre-treatment, then were randomized to 10-session EXP or BA. The primary outcome measure was the GAD-7. Of 101 participants consented, 69 completed treatment, the 46 completers with quality imaging data were included in analyses (22 EXP, 24 BA; mean 32.7 years, 10.9 % male). A priori region-of-interest analysis revealed that greater left caudate activity during loss receipt predicted greater symptom improvement in EXP, and did not relate to symptom change in BA (F(1, 428)= 5.24, p = 0.023), though this was not significant after correction for multiple comparisons. Whole-brain analysis further identified that greater activity during reward receipt in left frontoparietal regions and anterior insula / ventrolateral prefrontal cortex was associated with better outcomes in BA and worse outcomes in EXP. These findings highlight the role of reward and loss reactivity in GAD treatment. In particular, patients with elevated reactivity to reward salience may benefit most from BA or other reward-focused treatments. Future clinical trials are warranted to further elucidate reward-related predictors of anxiety treatment.