Instructed knowledge shapes feedback-driven aversive learning in striatum and orbitofrontal cortex, but not the amygdala

Description: This includes all maps from Atlas et al., 2016. The experiment was an aversive reversal learning task (fear conditioning with reversals) in which participants viewed images (angry faces from the Ekman set) and one stimulus was paired with a shock on 30% of trials. There were 3 reversals across the task. One group (n = 30, Instructed Group) was informed about contingencies before learning & prior to each reversal, whereas a second group (n = 40, Uninstructed Group) learned only through experience. Neuroimaging analyses focused on correlations with dynamic expected value (EV) calculated based on an adapted Rescorla-Wagner learning model with an additional parameter to measure the effects of instructed reversals (see Atlas et al). This model was fit to skin conductance from learners in either the Instructed group (n=20) or the Uninstructed Group (n = 20) to generate Instructed and Feedback-driven EV. We used the best fits from the models fit to each group to generate parametric modulators for fMRI analyses and modeled EV on unreinforced (no shock) trials in our first level analyses and compared within and across groups at second level. We also used task-based fMRI to look at trials surrounding reversal within the Instructed Group to identify regions that update immediately with instruction and those that continue to respond to previous contingencies, and how well reversals correlated with responses to instructions in the DLPFC region that showed greater activation in the Instructed Group across all trials.

Related article: http://doi.org/10.7554/eLife.15192