Applying the evolutionary theory of behavior dynamics to model the subtypes of automatically reinforced self‐injurious behavior

Abstract

The delineation of the subtypes of automatically reinforced self‐injurious behavior improved the utility of functional analysis results in predicting treatment efficacy. However, the mechanisms underlying subtype differences remain unclear and difficult to study in clinical populations. Morris and McDowell (2021) attempted to elucidate subtype differences by developing and evaluating models of the subtypes within the evolutionary theory of behavior dynamics. In the current study, we applied techniques from precision medicine to further evaluate the models developed by Morris and McDowell. This evaluation highlighted shortcomings of the existing models and suggested ways they could be improved. Thus, we conducted more extended modeling within the framework of precision medicine to identify models that were more quantitatively similar to available clinical data. Improved models that more closely approximate clinical data were identified. The implications of these models for research, practice, and further applications of the evolutionary theory of behavior dynamics are discussed.

Application of the evolutionary theory of behavior dynamics to severe challenging behavior

“…As discussed, the current demonstration study only manipulated ETBD parameters to mirror contingencies and contexts to examine how well it would generate outcomes that corresponded to what is observed among clinical cases. It would be possible to increase variation among AOs by manipulating certain ETBD parameters if one wished to generate outcomes that matched those with clinical cases (as is discussed below, studies by Morris & Lucia, 2023, and Morris & McDowell, 2021, provide excellent demonstrations of how to develop models for specific functional response classes).…”

“…Our approach to modeling clinical procedures was informed by previous studies using the ETBD to model the behavior of live organisms in laboratory settings (see McDowell, 2019) and challenging behavior (Morris & Lucia, 2023; Morris & McDowell, 2021). These methods are important to enumerate to support research applying the ETBD to challenging behavior.…”

“…Morris and Lucia (2023) recently extended this work by further testing and refining the models developed by Morris and McDowell (2021). Rather than deriving the models for Subtypes 1 and 2 based on functional analysis outcomes of AOs, they derived the models using an established predictive behavioral marker for clinical cases that describes the optimal cutoff point where the level of differentiation in the functional analysis most accurately predicts a positive response to treatment using reinforcement alone (Hagopian et al, 2018).…”

“…Study 3 was a systematic replication of Morris and McDowell's (2021) study modeling Subtypes 1 and 2 automatically maintained SIB. Additional research is needed to further examine the generality of the alternative models more recently described by Morris and Lucia (2023).…”

“…Reinforcement alone was effective for AOs modeling Subtype 1, but for AOs modeling Subtype 2, substantial reductions in SIB were achieved only following the addition of high-magnitude punishment in conjunction with continuous alternative reinforcement. Morris and Lucia (2023) recently extended this work by further testing and refining the models developed by Morris and McDowell (2021). Rather than deriving the models for Subtypes 1 and 2 based on functional analysis outcomes of AOs, they derived the models using an established predictive behavioral marker for clinical cases that describes the optimal cutoff point where the level of differentiation in the functional analysis most accurately predicts a positive response to treatment using reinforcement alone .…”