The components of discrete-trial teaching (DTT) may be individualized to each learner during instruction (e.g., the type of prompts used). However, there is limited research on the relative efficiency and effectiveness of these different prompt types. In addition, the learner’s preference for how they are taught is not always considered. The present study compared relative effectiveness of three prompt types (i.e., a gesture, modelling, physical guidance) to a no-prompt control condition during a receptive identification task with three boys with autism. One participant met the mastery criterion first in the model prompt condition, and two participants in the physical prompt condition. All participants selected the physical prompt during a concurrent-chains preference assessment. In addition, all participants completed a chained task using the most effective prompt type.