4 - (SYM 4) Effect of Oxytocin on Neural Signals of Vicarious Extinction Learning in Social Anxiety Disorder: Gender and Clinical Moderator

Vicarious fear extinction, the reduction of learned fear through observation of a non-fearful model, plays a key role in social learning and fear regulation. This study investigated how oxytocin impacts neural activity during vicarious extinction, and moderators of its effects. We focused on brain regions involved in fear processing and regulation, including the anterior cingulate cortex (ACC), ventromedial prefrontal cortex (vmPFC), insula, hippocampus, and amygdala. Participants with social anxiety disorder (n=51) and healthy controls (n=53) received 24 IU intranasal oxytocin or placebo prior to observing a demonstrator undergoing extinction learning, where some previously conditioned threat cues were still paired with an aversive outcome (CSR), while others were no longer paired and signaled safety (CSS). Vicarious extinction was followed by a reinstatement procedure to test the return of conditioned fear. Extinction scores were calculated from fMRI activation differences between CSR and CSS, using contrast estimates (vs CS-) for each region of interest.

During vicarious extinction under placebo, gender significantly moderated activation in the ACC (𝛽 = .953, p = .011) and marginally so in the vmPFC to CSS vs. CSR (𝛽 = -.676, p = .096). Women showed CSS >CSR activation in the ACC, whereas men showed no difference. Men showed CSS >CSR activation in the vmPFC, while women showed no difference. Oxytocin reduced CSR vs. CSS differences in the ACC (by decreasing CSS activation, similar trends in the insula) and vmPFC (by increasing CSR activation), particularly in controls but not in patients. During reinstatement under placebo, controls (particularly female controls) showed CSR >CSS activation in the right amygdala (𝛽 = -1.017, p = .017). Men showed marginal effects of CSS >CSR activation in the hippocampus (𝛽 = -.773, p = .057) and insula (𝛽 = -.723, p = .072). Oxytocin altered these patterns, reversing CSR-CSS differences in male controls within the vmPFC, insula, and hippocampus, while reducing amygdala activation differences in female controls.

Findings suggest that oxytocin may shift how the brain responds to threat and safety learned through others, with different effects based on gender and clinical status. Oxytocin reduced CSR-CSS differences in healthy controls, consistent with previous reports that oxytocin may impair fear extinction in humans. Further whole-brain analyses are needed to identify sensitive neural markers of vicarious extinction in patients, which could inform targeted interventions for anxiety disorders.