|Evaluations of Practices Used in Behavior Analytic Interventions for Children with Autism
|Sunday, May 25, 2008
|1:30 PM–2:50 PM
|Area: AUT; Domain: Applied Research
|Chair: Adel C. Najdowski (Center for Autism and Related Disorders, Inc.)
|Discussant: James E. Carr (Western Michigan University)
|CE Instructor: Adel C. Najdowski, Ph.D.
The papers in this symposium touch upon two important topics in behavior analytic intervention for children with autism. The first is related to practices used for data collection during discrete trials instruction and the second is related to practices that lead to generalization of skills taught during intervention. The first presentation in this symposium will show results from an evaluation of continuous data collection versus first-trial data collection during discrete trials instruction. The second presentation will show an assessment of generalization of pure tacts in the natural environment as a result of teaching impure tacts during discrete trials instruction. The third presentation will show a comparison of using free operant to restricted operant procedures on generalization of skills. From the first presentation, attendees of this symposium will learn which of the evaluated methods of data collection appear to be more accurate. From the second two presentations, attendees will learn whether generalization of skills is more or less likely given the evaluated practices.
|An Examination of Data Collection Methods in an Early Intensive Behavioral Intervention Program for Children with Autism.
|ADEL C. NAJDOWSKI (Center for Autism and Related Disorders, Inc.), Ryan Bergstrom (Center for Autism and Related Disorders, Inc.), Vardui Chilingaryan (Center for Autism and Related Disorders, Inc.), Susie Balasanyan (Center for Autism and Related Disorders, Inc.), Barbara C. Aguilar (University of Nevada, Reno)
|Abstract: There is a growing debate regarding the frequency with which data needs to be collected during discrete trials instruction (DTI) for children with autism. Cummings and Carr (in press) compared continuous (trial-by-trial) to intermittent (first trial only) data collection methods and found overall that targets mastered via the intermittent data collection method were mastered in slightly fewer sessions, but that targets mastered via the continuous data collection method were slightly better maintained. To keep true to the intermittent data collection method, the authors did not continue to collect data on the remainder of the trials conducted in the intermittent condition sessions as this could compromise quality of therapy. This study used continuous versus intermittent data to make decisions of mastery, but still collected data on all trials to determine if data on the first trial versus all trials would lead to similar or different conclusions about a child’s performance. Results showed that both methods predicted mastery at similar times and that both were correct as responding was maintained.
|Assessing Generalization of Discrete Trial Impure Tact Training to Pure Tacts in the Natural Environment.
|VARDUI CHILINGARYAN (Center for Autism and Related Disorders, Inc.), Jonathan J. Tarbox (Center for Autism and Related Disorders, Inc.), Romelea Manucal (Center for Autism and Related Disorders, Inc.), Ellen Kong (Center for Autism and Related Disorders, Inc.)
|Abstract: A common criticism of discrete trial training (DTT) is that it may not produce generalization to more natural settings. Natural environment training (NET) occurs in the natural environment from the start and is said to produce better generalization. However, some have suggested that, particularly in the case of verbal behavior, the operants which are taught in DTT may be functionally distinct from those taught in NET. That is, the controlling variables for verbal behavior in unstructured natural environments are not necessarily the same as those for verbal behavior occurring in the context of DTT. In this study, we used a standard DTT training format to teach two children with autism tacts of picture cards and assessed for generalization to unstructured probe sessions in their homes. Generalization to “pure” tacts in the unstructured natural setting did not occur and NET instruction was required to establish pure tacts there. In a second study, we taught a third child with autism to tact everyday items in their home environment and assessed for generalization to tacting picture cards in a DTT setting and again did not find generalization. Results appear to suggest that generalization should not be expected between DTT and unstructured settings, and vice versa, and that tacting may have to be directly established in each respective setting.
|A Comparison of Free Operant and Restricted Operant Procedures on Generalization of Academic Tasks with Young Children with Autism.
|RACHEL S. F. TARBOX (Center for Autism and Related Disorders, Inc.), Jonathan J. Tarbox (Center for Autism and Related Disorders, Inc.), Amy Kenzer (Center for Autism and Related Disorders, Inc.)
|Abstract: Initially, ABA programs for children with autism utilized only Discrete Trial Teaching (DTT). However, ABA programs continue to evolve, placing greater emphasis on the generalization and spontaneity of skills learned, and it has been suggested that generalization is difficult to acquire for skills that have been taught via DTT. One approach that has recently garnered attention is fluency instruction. As opposed to DTT, fluency instruction is a free-operant teaching procedure that utilizes a rate measure of behavior. Proponents of this approach suggest that free operant procedures promote the development of enduring behavioral repertoires. However, there is limited empirical evidence to support the use of a DTT versus a fluency approach with respect to the generalization of skills. The purpose of the current investigation was to compare the effects of free operant and restricted operant procedures on the generalization of academic skills with children with autism.