Linking descriptive and experimental analyses in the treatment of bizarre speech.

Descriptive and experimental methods were used to analyze the environmental determinants of an adult's bizarre speech. A descriptive analysis of behavior under natural conditions indicated that bizarre vocalizations occurred most often in the presence of task-related demands and in the absence of adult attention. Further, bizarre speech occurring during tasks was followed frequently by the cessation of task demands by staff or the subject's voluntary disengagement from task-related activities; bizarre speech observed during noninteractional periods (i.e., in the absence of adult attention) was frequently followed by staff attention. The escape and attention hypotheses were tested under analogue conditions. Results of the experimental analysis supported only the attention hypothesis; that is, bizarre speech appeared to function as an attention-producing behavior. The functional analysis data were used to select two different yet functionally equivalent treatments. The first treatment provided the subject with noncontingent scheduled attention. The second intervention taught the subject social language skills in the form of initiation and expansion statements. Both interventions were effective in suppressing maladapted speech. Advantages of linking descriptive and experimental analyses are discussed.