| Abstract: |
Twenty single-subject investigations on early language interventions were reviewed. Treatments included direct reinforcement of verbalization and various training procedures including mand-model and time delay, and have been generally successful. Quantitative synthesis procedures utilizing a percentage of nonoverlapping data outcome metric suggested that specific generalization training procedures were associated with stronger outcomes than were "train and hope" methods (i.e., procedures without specific generalization training) or assessments of spontaneous generalization. Implications for training and future research are discussed. |
