Implementing programs at less than full procedural integrity may be detrimental to the desired outcomes, but not all errors in implementing a procedure are likely to be equal (Carroll, Kodak, & Fisher, 2013). The purpose of the current study was to evaluate the effects of programmed errors on the behavior of persons with ASDs of implementation of reinforcement, prompting, and error correction on skill acquisition during a Match-to-Sample (MTS) task. Using an adapted alternating treatments design, 16 neuro-typical adults were taught 3 similar sets of stimuli. For each participant, 1 set was taught with full integrity using reinforcement for correct responses, an error correction procedure for incorrect responses, and a point that was systematically faded across sessions using a time delay. Another set served as the control and was taught without prompting, reinforcement, or error correction. The third set served as the test condition and varied across participants in order to systematically assess a range of potential errors. Errors tested included errors of reinforcement (omission or commission) with or without prompting and error correction procedures. Acquisition did not occur in the control condition until full integrity was implemented. Errors of omission in reinforcement had no effect on the rate of skill acquisition, whereas errors of commission in reinforcement did negatively impact the rate of acquisition for 5 of 7 participants. Overall, there were more incorrect responses in the test condition than in the full integrity condition. These findings suggest that MTS are robust teaching procedures for neuro-typical adults, but that errors of commission in reinforcement may be particularly detrimental to acquisition. These data were presented at ABAI 2016.
Interagency Autism Coordinating Committee (IACC)
Autism Research Database (AFD)
