|
| Towards a Better Understanding of the Efficiency and Validity of Different Functional Analysis Formats |
| Tuesday, May 31, 2016 |
| 2:00 PM–3:50 PM |
| Crystal Ballroom A, Hyatt Regency, Green West |
| Area: AUT/PRA; Domain: Translational |
| Chair: Gregory P. Hanley (Western New England University) |
| Discussant: Wayne W. Fisher (Munroe-Meyer Institute, University of Nebraska Medical Center) |
| CE Instructor: Gregory P. Hanley, Ph.D. |
| Abstract: Functional analysis (Iwata, Dorsey, Slifer, Bauman, & Richman, 1982/1994) was initially developed as part of an assessment process for enhancing treatment of severe problem behavior. Researchers have since developed different formats for increasing the efficiency of the analysis while retaining the methodological features that allow one to demonstrate control over problem behavior by a suspected reinforcement contingency. Via literature review and several new empirical analyses, the presenters will address the important topics of efficiency and validity of different functional analysis formats. |
| Keyword(s): functional analysis, interview informed, synthesized contingency, validity |
| |
Trial-Based Assessment to Inform Treatment of Elopement and Flopping When Walking Near or Leaving a Playground |
| Melissa Bowen (Munroe-Meyer Institute, University of Nebraska Medical Center), Wayne W. Fisher (Munroe-Meyer Institute, University of Nebraska Medical Center), ANDREA CLEMENTS (Munroe-Meyer Institute, University of Nebraska Medical Center) |
| Abstract: Trial-based functional analyses (FA) have been successful in determining the variables maintaining problem behavior (Bloom, Iwata, Fritz, Roscoe & Carreau, 2011; Lambert & Bloom, 2010), and the results have corresponded to traditional FAs (Larue et al., 2010). The current study utilized a trial-based FA to assess problem behavior when participants walked near and left a playground. During a pre-assessment, three boys between the ages of 3 to 7 years old with autism eloped to access a playground. When we attempted to remove them from the playground, they flopped to the ground. We then conducted an analysis in which trials alternated between the child being led past a playground, the child being prompted to leave the playground following brief access, and a control trial in which the child had free access to the playground. Results showed increased elopement when we brought each child near the playground and elevated levels of flopping when we prompted him to leave the playground. For all participants, two functional communication responses were taught to functionally replace the child�s elopement and flopping. Elopement and flopping both reduced to zero levels in their respective test conditions. For two participants, treatment was successfully extended to teach the child to tolerate when the functional communication response would not be honored. |
| |
| On the Efficiency of and Control Shown by Different Functional Analysis Formats: A Literature Review |
| JOSHUA JESSEL (Western New England University), Gregory P. Hanley (Western New England University), Mahshid Ghaemmaghami (Western New England University) |
| Abstract: Because functional analysis was developed as a clinical tool to enhance treatment effects of severe problem behavior (Iwata, Dorsey, Slifer, Bauman, & Richman, 1982/1994), researchers have since developed multiple formats geared towards increasing the efficiency of the analysis while retaining the capability to demonstrate control over problem behavior by a reinforcement contingency. We conducted a literature review of all published and differentiated functional analyses to determine the (a) procedures emphasized across different functional analysis formats, (b) average amount of time each format required to produce a differentiated result, and (c) level of control over problem behavior provided by each format. Results showed that the interview-informed, synthesized-contingency analysis (IISCA, Hanley, Jin, Vanselow, & Hanratty, 2014) yielded the most control while requiring the least amount of time to conduct. The methods unique to the IISCA that are likely responsible for the enhanced speed and control afforded by this analysis format will be discussed. |
| |
| Preliminary Comparisons of the Convergent and Divergent Outcomes of Synthesized- and Individual-Reinforcement Contingencies During Functional Analysis |
| AMANDA ZANGRILLO (Munroe-Meyer Institute, University of Nebraska Medical Center), Wayne W. Fisher (Munroe-Meyer Institute, University of Nebraska Medical Center), Brian D. Greer (University of Nebraska Medical Center), Patrick Romani (The University of Iowa), Todd M. Owen (University of Nebraska Medical Center) |
| Abstract: Researchers typically modify individual functional-analysis (FA) conditions only following initially inconclusive FAs (Hanley, Iwata, & McCord, 2003). Hanley, Jin, Vanselow, and Hanratty (2014) introduced a radical departure from this conventional practice by using an open-ended interview in combination with brief, informal observations to develop an efficient functional analysis with a single, synthesized test condition and a single control condition. In the test condition, they delivered multiple reinforcers (e.g., attention, escape) simultaneously, as a synthesized contingency, following each occurrence of problem behavior; in the control condition, they delivered those same reinforcers continuously on a response-independent basis. In the current investigation, we compared the results of this synthesized FA with a more traditional FA in which we evaluated each putative reinforcer individually. The synthesized FA produced false-positive outcomes for four of five consecutive participants. We discuss the implications of these findings relative to developing accurate and efficient functional analyses. |
| |
| Relying on Effective Action to Determine the Validity of Synthesized- and Individual-Reinforcement Contingencies During Functional Analysis |
| JESSICA SLATON (Nashoba Learning Group), Gregory P. Hanley (Western New England University), Kate Raftery (Nashoba Learning Group) |
| Abstract: Hanley, Jin, Vanselow, and Hanratty (2014) described a functional analysis model in which reinforcement contingencies identified via open-ended interviews with caregivers were combined in a single-test analysis. This interview-informed synthesized contingency analysis (IISCA) was shown to provide an effective baseline from which to develop socially-validated treatments. However, the contingency synthesis prohibits an understanding of whether problem behavior is maintained by the interaction of contingencies or by one or more of the individual contingencies. We therefore compared the results of IISCAs to the results standard analyses (Iwata et al., 1982/1994) and the results of treatments derived from both to determine the relative merits of synthesizing or isolating contingencies of reinforcement in analyses of problem behavior. For five children, the IISCA yielded differentiated results whereas the standard analysis did not; these data illustrate the importance of searching for interactions rather than or in addition to main effects of contingencies. Both analysis types were differentiated for four other children. Differential reinforcement-based treatments were effective for two of the four when designed from the standard FA. By contrast, treatments were effective for all children when designed from the IISCA. The relative efficacy, efficiency, and treatment utility of the IISCA and standard functional analysis will be discussed. |
|
| |