Reward learning may offer novel therapeutic target in OCD

Using a sensory preconditioning paradigm to determine the extent to which individuals were using past experiences to guide their behavior, researchers found that those with obsessive-compulsive disorder were less likely to generalize from rewards than those with social anxiety disorder and healthy controls.

“It has been speculated that the cycle of obsessions and compulsions in OCD may in a sense hijack the dopaminergic reward system, leading to insensitivity to external rewards and incentives, and preventing the adaptive pursuit of rewards in the environment,” Nina Rouhani, graduate student at the Princeton University Neuroscience Institute, and colleagues wrote. “These findings strongly suggest that individuals with OCD might generalize differentially from reward and losses. [Social anxiety disorder] has also been associated with deficient goal-directed learning of reward, although loss learning was not assessed.”

Researchers examined whether 57 individuals with OCD, social anxiety disorder and healthy comparisons generalized differently from rewards vs. losses. Including participants with social anxiety disorder allowed the investigators to see if differential performance on this task compared to healthy controls was specific to OCD or transdiagnostic across patients with anxiety.

Participants completed a sensory preconditioning paradigm that evaluated stimulus-response learning and generalization from monetary rewards, losses and neutral outcomes. The task consisted of three phases: an association phase, where participants learned to link pairs of images through repeated exposure to the pairs (ie, the generalized stimulus and conditioned stimulus); a reward phase, where the conditioned stimuli were paired with reward, loss or a neutral outcome; and a final decision phase, where they chose between pairs of the stimuli without any reinforcement.

Rouhani and colleagues observed no differences between individuals with OCD, generalized anxiety disorder and healthy comparisons in the direct learning of stimuli with their outcomes.

During the task, all participants demonstrated intact stimulus-response learning by choosing rewarding options and avoiding negative ones; however, participants with OCD were less likely to generalize from rewards than those with social anxiety disorder or healthy comparisons. The researchers did not observe this impairment for losses.

In addition, the results showed that greater deficits in reward generalization were associated with severity of threat estimation measured via a subscale of the Obsessive Beliefs Questionnaire within OCD and across all groups.

“Our findings together with prior studies describing deficits in reward processing in OCD suggest reward learning as a novel therapeutic target in OCD,” the investigators wrote. “Our findings also suggest that the interplay among reward processing, threat estimation, and habit-like behavior is complex and warrants careful study in both healthy control and clinical populations using validated paradigms that can tap all three domains in the same subjects.” – by Savannah Demko

Disclosure: Rouhani reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.