Abstract
The purpose of the present study was to examine whether children with autism could acquire and differentiate appropriate vocal requests in situations after training with the Picture Exchange Communication System (PECS; Bondy & Frost, 2002). With a multiple-baseline design, PECS training was given to 2 children with autism who were at the 1-word utterance level in their language development. The visual prompt fading procedure involved a gradual removal of the pictures. In baseline before the first intervention, the children did not make correct vocal requests in each situation. After training with multiple cues, including visual (PECS) and auditory (vocal imitation) stimuli, they could correctly say "give me", "take it", "teach me", and "look at". In generalization tests, the children produced the correct vocal requests for novel objects, situations, and persons. The purpose of the second intervention was to examine whether PECS would be effective for teaching 2-word utterances, using matrix training. After the training, both children could make correct sequential picture selections using the PECS strip and made correct vocal requests. In the matrix probes, both children produced correct vocal responses to untrained items. In the generalization probe, the children made correct vocal requests for novel objects, situations, and persons. These results suggest that PECS was an effective way to teach children with autism who were at the 1-word utterance level to extend the function and structure of their vocal requests.
