Many studies have highlighted the potential of oxytocin (OT) to enhance facial affect recognition in healthy humans. However, inconsistencies have emerged with regard to the influence of OT on the recognition of specific emotional expressions (happy, angry, fear, surprise, disgust, and sadness). In this study, we conducted a meta-analysis of seven studies comprising 381 research participants (71 females) examining responses to the basic emotion types to assess whether OT enhances the recognition of emotion from human faces and whether this was influenced by the emotion expression and exposure time of the face. Results showed that intranasal OT administration enhances emotion recognition of faces overall, with a Hedges g effect size of 0.29. When analysis was restricted to facial expression types, significant effects of OT on recognition accuracy were specifically found for the recognition of happy and fear faces. We also found that effect sizes increased to moderate when exposure time of the photograph was restricted to early phase recognition (< 300 ms) for happy and angry faces, or later phase recognition for fear faces (> 300 ms). The results of the meta-analysis further suggest that OT has potential as a treatment to improve the recognition of emotion in faces, allowing individuals to improve their insight into the intentions, desires, and mental states of others.