| Approximately 50% of children with autism fail to develop spoken language skills (Rutter, 1978). When children with autism do learn to speak, they fail to use their language for the purpose of communication (e.g., Landry & Loveland, 1989; Stone & Caro-Martinez, 1990). In addition, children with autism have difficulty generalizing and maintaining their acquired language across contexts and time (Howlin, 1981). Existing research provides support for milieu teaching as a viable intervention program for facilitating the acquisition and generalization of functional language of children who are in the early stages of language development (e.g., Warren, 1992; Warren & Gazdag, 1990). Nevertheless, studies involving children with autism are limited in both quantity and scope. To further knowledge about the intervention effects associated with milieu teaching for children with autism, more research is needed. The purpose of this study was to examine the effects of milieu teaching procedures on the acquisition, generalization, and maintenance of early functional language of three young children with autism. Milieu teaching procedures included model, mand, and time delay. Early functional spoken language targeted in this study included one-, two-, and three-term utterances representing a semantic category of action (e.g., "Eat"), a semantic relationship of action + "please" (e.g., "Open please"), and a semantic relationship of action + object + "please" (e.g., "Open jar please") that were used to make requests. A single-subject, multiple probe design across participants was employed to investigate the functional relationship between the milieu teaching procedures and early functional spoken language. The results of this study indicated that all three children with autism learned and used their language targets to request food or assistance from the experimenter as a result of milieu teaching procedures. Additionally, they generalized acquired language skills to their teacher or teaching assistant. Their language skills were maintained one week after intervention was terminated. The results of this present study, limitations, future research, and implications are discussed. |
