Oxytocin Enhances Social-Emotional Reciprocity in Autism

We evaluated whether oxytocin improves social-emotional reciprocity in children and adolescents with autism spectrum disorder (ASD) by conducting a secondary, hypothesis-driven reanalysis of the SOARS-B trial, the largest randomized clinical trial of intranasal oxytocin to date involving over 272 youth. We used a machine learning approach to construct data-driven composite outcome measures maximally sensitive to treatment from the Aberrant Behavior Checklist modified Social Withdrawal (ABC-mSW) and Social Responsiveness Scale-2 Emotion Recognition (SRS2-ER) subscales. Permutation testing rigorously controlled the Type I error rate under outcome learning. Oxytocin significantly increased emotional responsiveness in the learned ABC-mSW composite (mean Cohen’s d = −0.25, p = 0.014) and improved social-emotional reciprocity in the learned SRS2-ER composite (mean d = −0.43, p = 0.022) compared to placebo. Both effects replicated in the open-label phase, with prior placebo participants also improving on the ABC-mSW composite (mean d = −0.30, p = 0.039) and the SRS2-ER composite (mean d = −0.22, p = 0.041) after crossing over to oxytocin. We thus provide robust evidence that oxytocin enhances social-emotional reciprocity in ASD – a rare finding given that most pharmacologic trials for core ASD symptoms are negative. The findings also highlight the value of outcome learning for detecting nuanced treatment effects and support oxytocin’s potential as a targeted intervention in ASD. Prospective, preregistered trials should replicate these effects to strengthen the evidence base.