|On the Efficiency, Complexity, and Safety of Functional Analyses of Problem Behavior|
|Saturday, May 26, 2018|
|4:00 PM–5:50 PM |
|Manchester Grand Hyatt, Grand Hall D|
|Area: AUT/DDA; Domain: Service Delivery|
|Chair: Holly Gover (Western New England University)|
|Discussant: Richard G. Smith (University of North Texas)|
|CE Instructor: Richard G. Smith, M.S.|
Obstacles behavior analysts cite as reasons not to conduct functional analyses include the amount of time, complexity, and safety of functional analyses, and the potential inability to address multiple topographies of problem behavior at once. The current symposium will address these and related issues through evaluations of (a) components of functional analyses, (b) the utility of different models of functional analysis, (c) social acceptability of consultant-supported functional analyses, and (d) procedures for promoting safety and addressing multiple topographies of problem behavior.
|Instruction Level: Intermediate|
|Keyword(s): consulting, functional analysis, problem behavior, social acceptability|
|Target Audience: |
The target audience for this symposium is researchers and practitioners interested in the assessment of problem behavior. The ideal audience has a background in behavior analysis or works with individuals with developmental disabilities.
|Learning Objectives: 1. The first presentation will help participants to describe components necessary to conduct a successful functional analysis of problem behavior 2. Participants will learn to assess the utility of different models of functional analyses based on efficiency and data-based decisions 3. The third presentation will help participants learn about a consulting model to teach paraprofessionals a functional analysis method 4. The fourth presentation will teach participants how to conduct an expedited extinction analysis of problem behavior when confronted with the task of analyzing multiple topographies|
A Review of Trends in Efficiency and Implementation Components of Published Functional Analyses (1965-2016)
|RACHEL METRAS (Student), Joshua Jessel (Queens College), Gregory P. Hanley (Western New England University), Mahshid Ghaemmaghami (University of the Pacific)|
Since the publication of Iwata et al. 1982/1994, functional assessment has evolved to become considered best practice in the treatment of problem behavior for individuals with intellectual and developmental disabilities (Fischer et al., 2016; Pelios, Morren, Tesch, & Axelrod, 1999). Though legal mandates requiring access to functional assessment are further evidence of the broader cultural importance of function-based behavioral intervention (Individuals with Disabilities Education Improvement Act, 2004), surveys of practicing behavior analysts conducted over the last 10 years (Love, Carr, Almason, & Petursdottir, 2009; Oliver, Pratt, & Normand, 2015; Roscoe et al., 2015) reveal most practitioners are choosing to run descriptive assessments over more rigorous experimental analyses in their own practices. To investigate this issue, we review trends of published functional analyses for efficiency and number of components necessary for a successful analysis.
A Data-Based Decision-Making Model in the Selection of Functional Analysis Procedures
|JAMES MOORE (University of Southern Mississippi; Canopy Children's Solutions), Breanna Newborne (Canopy Children's Solutions), Hayden Rizer (University of Southern Mississippi), Meleah Ackley (University of Southern Mississippi), Heather Whipple (University of Southern Mississippi; Kennedy-Krieger Institute)|
Hanley et al. (2003) outlined the practical issue of efficiency in the use of functional analysis procedures. Recently, fervent debate has arisen regarding the use of standardized, synthesized, and other functional analysis. At times, standard functional analysis procedures yield zero-rates of problem behavior, perhaps because they fail to capture the true establishing operation, which may involve a combination of multiple stimulus events. In the current study, indirect descriptive assessments in the form of open-ended interviews were conducted and compared with standard functional analysis procedures as described by Iwata et al. (1982/1994). These results were then compared with functional analysis procedures derived from the results of the open-ended interview. Results suggested that in some cases, standard functional analysis procedures that yield zero-rates of responding were not congruent with the establishing operations suggested by interviewees. Revised analysis, including concurrent operant arrangements and synthesized contingencies, produced clear rates of responding under conditions described by caregivers and teachers during interviews. Finally, a trial-based synthesized contingency analysis was piloted that successfully determined behavioral function in under 30 minutes of session time, on average. Results are discussed in terms of individualized assessment, and using descriptive analyses to inform the design of relevant functional analysis conditions.
Consultant-Supported Functional Analysis in Educational Settings
|CORY WHELAN (May Institute; Western New England University ), Gregory P. Hanley (Western New England University)|
Practitioners report an almost exclusive reliance on indirect and descriptive assessments when conducting functional behavior assessments (FBA) of severe problem behavior (SPB) in school and residential settings, despite the absence of evidence supporting their utility when not complimented with a functional analysis. Practitioners also report that they do not conduct functional analyses due to concerns with safety, amount of time required, and lack of necessary resources. This project describes a collaborative approach for conducting effective functional analysis of SPB in educational settings. The process involved the collaborative conduct of a particular type of functional analysis referred to as an Interview-Informed Synthesized Contingency Analysis (IISCA) for students with SPB. With the consult of a board certified behavior analyst, the participants conducted interviews, designed conditions, ran sessions, and collected and analyzed data. All analyses yielded differentiated outcomes safely and quickly. Survey results showed that the process was considered sufficiently safe, fast, and effective by the practitioners for use in their educational settings.
An Evaluation of Progressive Extinction to Assess Response Class Membership of Multiple Topographies of Problem Behavior
|CHRISTINE A. WARNER (Western New England University; New England Center for Children), Gregory P. Hanley (Western New England University), Mahshid Ghaemmaghami (University of the Pacific), Holly Gover (Western New England University), Robin K. Landa (Western New England University), Adithyan Rajaraman (Western New England University), Jessica Slaton (Nashoba Learning Group), Kelsey Ruppel (Western New England University)|
Persons with autism often engage in multiple topographies of problem behavior. Conducting functional analyses of each form as recommended by Hanley, Iwata, and McCord (2003) may be too time consuming. As an alternative, we progressively applied extinction in test conditions that were differentiated from their control conditions to determine response class membership of multiple topographies of problem behavior. During interview-informed and synthesized test conditions, all reported problem behaviors were initially reinforced. Progressive extinction based on the procedures described by Magee and Ellis (2000) was then implemented during which problem behavior types were sequentially placed on extinction for four participants. Expedited or brief extinction analyses were conducted with the remaining participants in which all but the most concerning topography of problem behavior were placed on extinction. Results showed that all topographies of problem behavior that were reported to co-occur, including the most concerning topographies, were evoked and maintained by the same contingencies across all nine participants. We highlight the conditions under which a full or expedited extinction analysis should be considered when functionally analyzing multiple topographies of problem behavior.