On the validity of data produced by isolated and synthesized contingencies during the functional analysis of problem behavior

Abstract

Functional analyses are intended to identify the reinforcers maintaining problem behavior in order to inform clinicians' selection of interventions. Traditionally, these analyses have exposed problem behavior to multiple, isolated reinforcement contingencies and in doing so, have ruled in and ruled out potential sources of reinforcement. Recently, some functional analysis models have forgone testing individual reinforcement contingencies and instead exposed problem behavior to 2 or more reinforcers simultaneously in a single, synthesized reinforcement contingency. The current review applies assessment analytics to these approaches to consider their relative sensitivity, specificity, discriminant validity, and predictive validity to yield practice recommendations and to nominate areas of future research.

Integration of experimental functional analysis procedural advancements: Progressing from brief to extended experimental analyses

“…The participants' continuation through at least Phase 3 also sheds light on the extent to which the automatic reinforcement screener (Phase 1) and brief assessment results (Phase 2) are predictive of more extended analyses, and speak to relevant core psychometric assessment properties of functional analysis procedures. Tiger and Effertz (2020) applied traditional assessment analytics (sensitivity, specificity, discriminant validity, and predictive validity) to different functional analysis methodologies. A complete analysis of these validity indices is outside the scope of this article, but a cursory overview of Table 3 provides some insight for clinicians.…”

An evaluation of trial‐based functional analyses of inappropriate mealtime behavior

“…For children whose functional analyses indicated different treatments, the comparison of treatments provided more information as to whether the trial‐based and extended functional analyses resulted in false‐positive or false‐negative findings. When a functional analysis identifies a maintaining variable that may not function as a reinforcer, this is referred to as a false‐positive finding (Tiger & Effertz, 2020). Callie's solids trial‐based functional analysis may have produced a false‐positive finding by identifying an attention function, as escape extinction with and without attention extinction resulted in decreases in inappropriate mealtime behavior.…”

“…When a functional analysis does not identify a maintaining variable when, in fact, a maintaining variable exists, this is referred to as a false‐negative finding (Tiger & Effertz, 2020). Hank's liquids trial‐based functional analysis produced a false‐negative finding, as no functions were identified, but escape and attention extinction were necessary to produce clinically significant changes in inappropriate mealtime behavior and acceptance.…”

“…Previous studies have either compared the functional analyses (e.g., Bloom et al, 2011; LaRue et al, 2010) or only implemented the treatment suggested by the trial‐based functional analysis (e.g., Hodges et al, 2018; Lambert et al, 2012). Even though the extended functional analysis is used regularly and has empirical support (e.g., Bachmeyer et al, 2019), there is discussion as to whether the extended functional analysis should be treated as the “gold standard” assessment from which to determine the validity of other assessments (Tiger & Effertz, 2020). Ultimately, the usefulness of a functional analysis is determined by its unique ability to indicate a treatment that produces the desired changes in the target behavior.…”

Further comparison of isolated and synthesized contingencies in functional analyses

“…It is common for researchers to use the isolated FA as the gold standard for comparison when evaluating other functional assessments, such as the SCA (Tiger & Effertz, 2020). Several such comparative evaluations of isolated FAs and SCAs have indicated a propensity for the SCA to produce false positive results.…”