Association for Behavior Analysis International

Behavior analysis can be considered a science of learning. Not only does the field address how learning occurs, but it also seeks to improve teaching technologies. To make learning more efficient, the goal should be on increasing rate of acquiring individual targets and increasing generative learning. In the literature, the focus is often on explicitly taught skills (e.g. Kodak et al, 2016; Sidman, 1994). However, generative learning, the ability to demonstrate responses that have not been explicitly taught or reinforced prior, is a crucial component of efficient learning (e.g., Critchfield & Twyman, 2014). Howard, Sparkman, Cohen, Green, and Stanislaw (2005) investigated the effects of having children with an autism spectrum disorder in an intensive behavior based preschool program compared to an electric program and a control, non-intensive program. In their discussion, they point out that in order for a child with a developmental delay to close the gap with neurotypical peers, an intervention must produce learning rates that are faster than the typical peer, which was seen in the intensive behavior-based program. Thus, it is imperative to have teaching procedures which are both effective and efficient (Albarran & Sandbank, 2018) to reach this learning rate.

The target audience is both newer BCBAs as well as those who have been in the field for a long time and may continue to use certain procedures because that is what they are used to doing. It provides both an overview and more in depth view into the various methods to improve efficiency in learning.

Procedures developed to provide corrective feedback are part of many types of instructional programming (Englemann, 1988; Binder & Watkins 1990). A correction procedure is essential to the formation of operants and literature has provided some considerations that should be made in regard to correction procedures. Within the literature a large degree of variations is used within similar procedures for example reinforcement delivery; number of repetitions; errorless learning vs. error corrections (Worsdell, 2005; Cowley, Green, & Braunling-McMorrow, 1992; Koegel & Egel, 1979, Carr & Kologinsky, 1983; Wheeler & Sulzer, 1970, and Carey & Bucher, 1983; Remington & Clarke, 1983). These variations regarding error correction procedures have resulted in numerous studies with inconclusive findings (Smith et al., 2006; Turan, Moroz, & Croteau, 2012; Rodgers and Iwata, 1991). The purpose of this study included gaining a better understanding of how and why practitioners currently select error correction procedures for their students and how they may affect student progress. In this study, ten questions were presented to BCBAs using an electronic survey. Survey data and the importance of the findings will be discussed.

This study added to existing research on the use of error correction procedures when teaching children with direct instruction. Previous research has shown that different error corrections can lead to increased numbers of correct responses or objectives met depending on the individual (Turan, Moroz and Croteau, 2011; Smith et al., 2006; Carroll et al., 2015). Recent research has also suggested that different error corrections may be successful depending on the types of academic program being taught or level of a pupil’s verbal behavior (Gautreux et al., 2017). This experiment was conducted with two pupils with autism spectrum disorders aged 10 and 11, using an alternating treatments design that tested programs across listener selection, listener production, speaker and visual matching topographies. The error correction procedures utilized were variations sourced from the existing literature. As well as looking at the impact of the treatments on the individual’s responses to academic programs (created for the purposes of the study), the experiment also recorded data on the amount of inappropriate behavior that was emitted during each treatment.

Equivalence-based instructional (EBI) technology has been shown to be highly effective in teaching relations among arbitrary stimuli (Sidman, 1994). The present study examined the use of EBI to train caregivers a five-relation stimulus class consisting of the following elements: a) labels of social functions of behavior, b) descriptions of antecedent events, c) descriptions of consequence events, d) vignettes with both antecedent and consequence events, and e) function-based responses (Albright, Schnell, Reeve, & Sidener, 2016; Fienup, Covey, & Crithchfield, 2010). Initial results demonstrated a functional relation between teaching the first set of relations (A:B:C) and a significant increase within untrained relations (B:C:D:E). These results must be viewed cautiously as responding in the pre-test of the second teaching set (A:D:E) was higher than expected. This supports previous experience with these stimuli as part of the formation of the untrained class merger (B-C-D-E) relations. Demonstration of an effective EBI intervention could lead to the development of computer-based training that could assist caregivers in acquiring more efficient function-based responses to problem behavior.