This investigation examined the effects of computer-based video self-modeling on the receptive understanding of six prepositions by three students with a diagnosis of moderate intellectual disabilities. Using a multiple probe design across three sets of prepositions, video captions were paired with photographs on the computer in order to simulate and teach prepositional concepts. When students selected a target photograph on the computer screen, the corresponding self-modeling video segment was activated. Results indicate an increase in the number of photographs correctly selected with the computer-based program and that students were able to generalize receptive understanding of concepts to the activities "positioning self" and "positioning objects" measured during probe conditions. Implications are discussed with respect to teaching generalized use of specific language concepts using computer-based video instruction. (Contains 2 tables and 3 figures.)