Abstract Some but not other studies on oxytocin for schizophrenia, particularly those using a higher dose, indicate that oxytocin improves negative symptoms of schizophrenia. We performed an add-on randomized controlled trial to examine the effect of high dose oxytocin, social skills training, and their combination in the treatment of negative symptoms and social dysfunction in schizophrenia. Fifty-one subjects with schizophrenia were randomized, employing a two-by-two design: intranasal oxytocin (24 IU X3/day) or placebo, and social skills training or supportive psychotherapy, for three weeks. The primary outcome was the difference in the total score from baseline to end-of-study of a semi-structured interview which assessed patients’ social interactions in three scenarios: sharing a positive experience, sharing a conflict, and giving support when the experimenter shared a conflict. The interactions were scored using the Coding Interactive Behavior Manual (CIB), clinical symptoms were assessed with the PANSS. No significant difference was found between groups in the total CIB or PANSS scores. The majority of comparisons in the different social interactions between oxytocin and placebo, and between social skills training vs. supportive psychotherapy, were also non-significant. Social skills training reduced blunted affect and gaze. In post-hoc analyses of the support interaction, oxytocin improved synchrony and decreased tension, while in the positive interaction it improved positive affect and avoidance. None of these findings remained significant when controlling for multiple comparisons. In conclusion, oxytocin did not influence participants’ social behavior, and was not effective in improving the symptoms of schizophrenia.