Association for Behavior Analysis International

A typically developing 2 year old would be saying sentences with 2 to 4 words. A 4 year old would have a vocabulary of over 1000 words and be able to put together sentences of 4 or 5 words (Mannheim, 2015) . Significant proportions of children with autism do not acquire functional speech and remain non vocal even until age 9 or later. The studies in this symposium demosntrate effectiveness of specific behavioral technologies, namely sign mand training with prompts with and without delays and intraverbal training in inducing first instances of speech and in increasing vocal responding in children with autism. An additional study assessed tact-mand transfer in 7 early learners with autism, established this abstract relation in participants for whom it was initially absent, and concluded that functional independence (absence of tact-mand transfer) may often be a temporary aspect of an early learner's repertoire.

The current study spanning 5 years examines the effectiveness of sign mand training with vocal prompts and intraverbal training in inducing first instances of speech in non-vocal children on the autism spectrum. These technologies leverage the power of motivating operations. A total of 91 non-vocal children between the ages of 1.5 to 13 years participated in the study that uses several multiple baseline design sub-studies across subjects. The interventions were carried out in 6 centers across 4 cities in India. 81.6% of children below the age of 6 years became vocal and the success rate was 87% in children between 6-13 years. Vocals emerged as mands, echoics and intraverbals in 23%, 24% and 21% respectively of the 75 who acquired vocals. 64% of the children acquired their first vocal within 180 days. Both the technologies were effective in inducing first instances of speech in 82.4% of the participants. The time to vocalization and the stimulus conditions under which they emerged are explored.

Some reports estimate that approximately one-third to one-half of individuals with autism do not use speech functionally (National Research Council, 2001). The high incidence of non-vocal persons with autism has lead parents, educators, and clinicians to search for procedures that may facilitate the development of vocal responding and potentially lead to effective vocal verbal behavior repertoires. The purpose of this study was to determine the effect of manual sign mand training combined with time delay and vocal prompting procedures on the production of vocal responses in non-vocal children with developmental disabilities. A multiple baseline design across participants was used to verify the effectiveness of this intervention. All participants showed increases in vocal production following the implementation of the independent variables.

The present study replicated and extended Hall & Sundberg (1987) by first assessing tact-mand transfer in 7 early learners with autism (4 vocal, 3 signing plus vocal), ranging from 3.6-5.8 years of age. All participants met VB MAPP entry criteria of tacting at least 4 items and manding at least 4 visible, but 0 missing items via the interrupted chain procedure. For each participant, the study targeted 3 tact-mand response forms for each of 3 chains of behavior (9 total) and probed mands after all tacts were trained to criterion. If 0/9 untrained mands emerged, a participant was considered nave with respect to tact-mand transfer and received mand training on targets previously trained as tacts--a version of multiple exemplar instruction. After each mand was trained, remaining untrained mands were probed to assess the point of transfer. Results showed that untrained mands emerged after direct mand training on 4-5 targets with 2 participants, 1 target with 3 participants and 0 targets with 2 participants. Findings from participants requiring multiple exemplar instruction replicate those of Hall & Sundberg, and suggest that functional independence (absence of tact-mand transfer) is a temporary aspect of the learners repertoire rather than a static entity.

A variety of technologies such as Echoic training, Shaping, Antecedent Rapid Motor Imitation sequence, Stimulus Stimulus Pairing (SSP) procedures and Mand training with and without time delay continue to be studied for their effectiveness in increasing vocalizations in children with autism. Intraverbal fill-in training with rhymes, fun and contextual fill-ins builds anticipation and excitement creating opportunities for vocals to be emitted under the control of specific verbal stimuli. In the current study spanning 5 years involving simultaneous administration of sign mand training with vocal prompts and intraverbal fill in training to children with no vocal-verbal repertoire, the latter was found valuable in inducing first instances of speech in non-vocal children with autism. Of the 91 children, each serving as a single subject, first instances of speech were induced successfully in 75 and of these, 51 had at least one vocal emerge during intraverbal training. 16 participants had vocals emerge predominantly under intraverbal fill- in training. Inter observer agreement was taken for every specific vocal acquired for every participant and was at 100% throughout the study. The role of motivating operations in intraverbal fill-in training and comparisons with vocals emerging as echoics and mands during implementation of intervention package are explored