Acquisition of matching - to - sample performances in severe retardation: learning by exclusion.

Immediate discriminative control by spoken words was examined in a profoundly retarded, mute young man. The subject learned to select food items from a two-choice display when the food names were spoken, in a matching-to-sample format. Then a new food was presented as one choice, and a novel name was the sample. The subject selected the new food in response to the first presentation of the novel name. It appeared that the subject rejected or "excluded" the trained food which was the other alternative. This "exclusion" performance was obtained errorlessly with fourteen of sixteen new foods and novel names. Subsequent test suggested that the subject had learned the relations between all the new foods and novel names, after a history of reinforced exclusion trials. The method also proved successful with pictures. Prerequisite baseline training may have been critical, particularly differential auditory control by previously trained names. The procedures systematically replicated previous research on the exclusion paradigm, and suggested a potentially errorless, efficient teaching method for individuals without functional language.