An evaluation of preference for video and in vivo modeling.

We assessed preference for video or in vivo modeling using a concurrent-chains arrangement with 3 children with autism. The two modeling conditions produced similar acquisition rates and no differential selection (i.e., preference) for all 3 participants.