|
| Evaluation of Idiosyncratic Variables During Functional Analyses |
| Sunday, May 30, 2004 |
| 10:30 AM–11:50 AM |
| Beacon B |
| Area: DDA/AUT; Domain: Applied Research |
| Chair: Eileen M. Roscoe (New England Center for Children) |
| Abstract: Learning Objectives
Identify situations when additional manipulations need to be conducted during a functional analysis
Determine methods for manipulating various idiosyncratic antecedent and consequent events during a functional analyses
Describe the competing-items assessment used by Bowman et al. |
| |
| The Use of Competing Stimuli to Clarify Undifferentiated Responding during a Functional Analysis |
| STEPHANIE A.C. KUHN (Kennedy Krieger Institute), Lynn G. Bowman (Johns Hopkins University School of Medicine), Jennifer Lynne Bruzek (Kennedy Krieger Institute), Lisa M. Toole (Kennedy Krieger Institute), David E. Kuhn (Kennedy Krieger Institute), Vanessa Rodriguez-Catter (The Ohio State University) |
| Abstract: Several studies have demonstrated the utility of using competing stimuli in reducing behavior maintained by automatic reinforcement (Piazza et al., 1998; Piazza et al., 2000). When results of functional analyses have demonstrated that problem behavior is maintained by automatic reinforcement and competing items are evaluated, the items have typically been evaluated in alone or ignore conditions. It is possible that competing items may compete with problem behavior in some environmental conditions (e.g., low attention) but not in others (e.g., tangible). In order to investigate this possibility, competing stimulus assessments were conducted for several children following functional analyses that suggested problem behavior was maintained by automatic reinforcement. Next, the effects of the items that were identified as competing stimuli during ignore or alone conditions were assessed across several conditions (e.g., attention, escape, tangible). Results suggest that competing items identified for behavior maintained by automatic reinforcement may not compete with problem behavior across all environmental situations. |
| |
| Examination of Idiosyncratic Antecedent and Consequent Variables in Functional Analysis |
| EILEEN M. ROSCOE (New England Center for Children), Griffin Rooker (New England Center for Children), Carly Moher Eby (New England Center for Children) |
| Abstract: Functional analyses (FAs) may produce undifferentiated outcomes when idiosyncratic antecedent and/or consequent variables affect behavior. In addition, FA outcomes may be inaccurate when multiple consequences are obtained during a single condition. In the present study, we examined antecedent and/or consequent variables during FAs of two participants’ problem behavior. For participant 1, a standard FA indicated his aggression was maintained by escape. However, extended analysis that isolated these two consequent events indicated that the participants’ aggression was maintained by access to social-positive reinforcement (flopping) and not by social-negative reinforcement (escape). For participant 2, a standard FA did not suggest a clear function for her disruption (i.e., high levels of disruption occurred during both demand and play conditions). A subsequent component analysis, of antecedent variables (proximity of therapist, presence vs. absence of toys, and presence vs. absence of attention) present during the play condition, was conducted to determine which variable was associated with disruption. Results of this analysis suggested a modified play condition that, when incorporated in a subsequent FA, helped determine a clear function. IOA was collected during at least 33% of the sessions with a score of at least 90%. |
| |
| Evaluation and Treatment of Destructive Behavior Maintained by Access to Idiosyncratic Reinforcement Variables |
| ALYSON N. HOVANETZ (Marcus Institute), Henry S. Roane (Marcus Institute), Nathan Call (Marcus Institute), Michael E. Kelley (Marcus Institute), Camille Grier (Marcus Institute) |
| Abstract: Functional analysis (FA) procedures as described by Iwata et al. (1982/1994) has been demonstrated to be effective at determining functional relationships between problem behavior and environmental consequences. However, the test conditions included in many FAs may not address idiosyncratic functions of problem behavior. In the current study, the results of a FA conducted with a 6-year-old boy diagnosed with autism and cerebral palsy indicated that higher rates of self-injurious behavior (finger biting) occurred in tangible and toy play conditions relative to ignore, attention, and demand conditions. A component analysis was conducted to determine what component of the leisure item available in these two conditions (i.e., a video) evoked finger biting. Conditions were developed to evaluate which specific properties of the video were associated with finger biting. The removal of the item resulted in a decrease in video engagement and finger biting, suggesting that the two may have functioned as complementary reinforcers. A treatment analysis was developed in which finger biting was decreased while the video was present. Reliability data were collected on over 33% of sessions and averaged over 90% for all dependent measures. Results are discussed in terms of analyzing interactions between the consumption of multiple reinforcers during FAs. |
| |
| Combining Establishing Operations as a Means of Further Assessing Functional Analysis Results |
| NATHAN CALL (Marcus Institute), David P. Wacker (University of Iowa), Joel Eric Ringdahl (University of Iowa), Eric Boelter (University of Iowa) |
| Abstract: Some individuals assessed via functional analyses do not exhibit problem behavior during assessment, especially when evaluations are brief. When no problem behavior is observed during a functional analysis, the result can be a false negative, or Type-II error. A potential reason for false negative findings is that the establishing operations (EOs) manipulated within the test conditions of a functional analysis may be insufficient to motivate problem behavior. Most functional analyses conducted to date have introduced a single EO in each test condition. It seems possible that some functions have been missed because of a failure to consider the influence of multiple EOs in the natural environment. In the current study, test conditions were added to functional analyses in which two or more of the EOs typically manipulated in functional analysis test conditions were combined within a single condition. Four of five participants either engaged in problem behavior exclusively in the combined EO test conditions, or engaged in higher and more stable levels of problem behavior during the combined EO test conditions as compared to single EO test conditions. Results suggest that combinations of establishing operations within test conditions may be useful when target behavior does not occur during a typical functional analysis. |
|
| |