Association for Behavior Analysis International

Responses that serve a communicative function are critical in promoting the autonomy of an individual and enabling his or her access to reinforcers. Throughout the course of an individuals life, communication responses typically begin as basic vocalizations, such as echoics, and over time evolve to more complex types of responses, such as intraverbals. This symposium will review current research on instructional procedures used to establish verbal behavior repertoires for children with autism, with such tactics including acoustical markers, relational frame training, the use of multiple exemplars, transfer of stimulus control, and fluency timings. Results of these studies demonstrate how verbal operant responses can be strengthened through explicit training, and how they can also emerge in the absence of direct training, as seen in generalization probes and tests of derived relational responding.

An acoustical marker contingent on target behavior has been shown to be an effective procedure for increasing adaptive behavior in children with autism (Khowaja, Malmquist, & Eshleman, 2011; Maendler, Eshleman, & Cihon, 2009; Morien, Eshleman, & Malmquist, 2010). The current study examined the use of a contingent acoustical marker on teaching new words (i.e., echoics) to children with autism with a limited vocal verbal repertoire. This study used a shaping procedure facilitated by an acoustical marker, which provided more immediate consequences for the behavior than typical reinforcers such as praise and tangible items. Two children diagnosed with autism ages 3 and 7 participated in this study. The data suggested that the shaping procedure facilitated by the use of the acoustical marker increased echoic vocalizations, including a full word for one participant and several successive approximations for both participants.

This study evaluated the effects of a multiple exemplar training package with a fluency component on intraverbal empathy responses of four children with autism. This study extended the research from Schrandt, Townsend, and Poulson (2009). The experimenter presented a variety of role-play scenarios related to three different response categories (sad/hurt, happy/excited, and frustrated). Participants were initially provided with an echoic prompt for the correct empathetic response. For each response type, participants were trained to accuracy criterion, followed by training to a fluency criterion. Results indicated that all participants responded at accurate as well as fluent levels not only in training sessions, but also in discrimination sessions and during retention, endurance, stability, and application probes.