Evidence From Children with Autism that Derived Relational Responding is a Generalized Operant

We conducted an empirical examination of derived relational responding as a generalized operant and concurrently evaluated the validity and efficacy of program items contained in the Promoting the Emergence of Advanced Knowledge - Equivalence (PEAK-E) curriculum. A first study utilized a multiple-baseline across-skills experimental arrangement to determine the efficacy of equivalence-based instruction guided by PEAK-E, replicated across 11 children with autism. A total of 33 individualized skills were taught, and the subsequent emergence of untrained relations was tested throughout the investigation. The mastery criterion was achieved for 29 of the 33 instructional targets. Additionally, for 3 participants, results were again replicated with a novel set of stimuli. A second study evaluated the degree to which multiple-exemplar equivalence-based instruction led to the emergence of derived relational responding as a generalized operant. The organized nature of the PEAK curriculum allowed the impact on derived relational responding to be compared to that produced by earlier PEAK models that are focused on the direct training of traditional verbal operants. PEAK-E instruction was introduced in a multiple-baseline design across two participants, with a third staying in a training baseline throughout. Increases in derived relational responding using novel, untrained stimuli were only observed when multiple-exemplar equivalence-based instruction was introduced. Taken together, these results provide support for derived relational responding as a generalized operant and demonstrate the utility of conducting larger scale evaluations of higher order behavioral phenomena in single-case experimental arrangements.