Instruction in auditory-visual conditional discriminations for young children with autism spectrum disorder is typically based on either a conditional-only or a simple-conditional method. In this study, we evaluated a modified simple-conditional method in which we removed the steps for which visual comparisons were presented in isolation. We compared this modified simple-conditional method with the conditional-only method when teaching auditory-visual conditional discriminations to six young Chinese children with autism spectrum disorder. We included two efficiency measures: total sessions and time to mastery. Our results replicated the findings of previous research in that both methods were efficacious for all but one participant. Although efficacy outcomes were similar across methods, the conditional-only method was more efficient across participants according to sessions and time to mastery. Thus, the results add to support for the use of the conditional-only method to teach auditory-visual conditional discriminations.