|
| DEV |
| Sunday, May 28, 2017 |
| 12:00 PM–3:00 PM |
| Convention Center, Exhibit Hall D |
| Chair: Per Holth (Oslo and Akershus University College of Applied Sciences) |
|
| 102. Using Inactive Reinforcers To Increase Active Behavior |
| Area: DEV; Domain: Applied Research |
| REBECCA DUNCAN (Ball State University), Valdeep Saini (Upstate Medical University) |
| Discussant: R. Douglas Greer (Columbia University Teachers College and Graduate School of Arts and Sciences) |
| Abstract: Previous research evaluated the preliminary efficacy of prize-based contingencies to increase activity levels of adults (Donlin, 2014). Although those studies examined the use of experimenter-mediated tangible reinforcers, no studies have examined self-managed, naturally-occurring reinforcers. Moreover, the extent to which reinforcers associated with inactivity could maintain an active behavior. The purpose of the current study was to assess the use of self-managed naturally-occurring leisure reinforcers (e.g., watching television) to increase active behavior (i.e., steps throughout the day). A changing-criterion design was used to evaluate the effects of inactive reinforcers on daily steps with two female participants in their late 20’s. Daily, inactive reinforcement (i.e., preferred television shows) was provided contingent on meeting the criterion step established for each day. A larger, monetary reinforcer was provided at the end of the study if participants met the criterion for at least 80% of sessions. The results indicate that daily reinforcement was effective in some phases, but not all, suggesting that the use of inactive reinforcers to increase active behavior had only moderate success. These results further suggest that programmed reinforcers may not have controlled behavior across all phases and that variability may have partly been due to the influence of extraneous variables. |
| |
|
| 103. Comparison of a Stimulus Avoidance Assessment and a Concurrent Chains Assessment in Treatment Selection |
| Area: DEV; Domain: Applied Research |
| NICOLE CONNOR MOORE (The Ivymount School), Megan B. Boucher (The Ivymount School) |
| Discussant: R. Douglas Greer (Columbia University Teachers College and Graduate School of Arts and Sciences) |
| Abstract: Challenges often arise when reinforcement-based interventions are ineffective at decreasing challenging behavior to clinically significant rates. The current study examines the correspondence between a stimulus avoidance assessment and a concurrent chains assessment. The participant of the current study was diagnosed with Autism Spectrum Disorder and engaged in inappropriate vocalizations that interfered with learning. Interobserver agreement was collected on avoidant behavior, inappropriate vocalizations, and choice for 100% of trials during the concurrent chains assessment. During the stimulus avoidance assessment, the participant was briefly exposed to five potential punishers. Data were collected on avoidant behavior and generated to create a hierarchy of avoidance. A concurrent chains preference assessment was conducted to identify a hierarchy of preference for potentially aversive procedures. The data from these assessments were compared to determine if rates of avoidant behavior corresponded with choice for procedures. Results indicated correspondence between the two assessments; procedures associated with higher rates of avoidant behavior were chosen less frequently. Additionally, a punisher assessment was run to determine if procedures identified function at punishers for target behaviors. Preliminary data indicate that the procedures chosen most frequently were least effective as punishers. This suggests that a concurrent chains assessment may be useful for determining putative punishers, while limiting exposure to potentially aversive procedures and incorporating choice. |
| |
|
| 104. Using High-Probability Instructional Sequence to Decrease Latency to the First Bite
and Overall Meal Duration in a Child Diagnosed with a Pediatric Feeding Disorder |
| Area: DEV; Domain: Applied Research |
| CHELSEA PAULA (Clinic 4 Kidz), Meeta R. Patel (Clinic 4 Kidz) |
| Discussant: R. Douglas Greer (Columbia University Teachers College and Graduate School of Arts and Sciences) |
| Abstract: High-probability (high-p) instructional sequence may set the pace for bite acceptance and allow meal durations to be within a consistent time frame. When bite acceptance is delayed, often times overall meal duration is increased and consuming large volume is difficult. This study was conducted to observe the effects of using a pacifier (a high-p response) on latency to the first bite with a 6-year-old girl diagnosed with cerebral palsy, global developmental delay, autism, and failure to thrive. Upon being admitted to an intensive home-based interdisciplinary feeding program, Kathy was 100% gastrostomy (G-) tube dependent and had severe oral aversion. She would suck on a pacifier and chew on a chewy tube. However, she had severe refusal to any tools placed near her mouth by a feeder. A reversal design was used to evaluate the use of the pacifier on the latency to the first bite. The high-probability instructional sequence began with an empty pacifier 1x and then followed with a liquid dipped pacifier 2x prior to a spoon presentation of a liquid. Escape extinction (EE) was also implemented throughout the analyses but EE was ineffective at decreasing latency to the first bite during baseline; therefore, the need for the high-p instructional sequence. During baseline latency to the first bite averaged 24 minutes and meal durations averaged 27 minutes for an average bite number of 1 bite. The pacifier was presented during treatment and latency to the first bite decreased to 18 seconds and meal duration decreased to 5 minutes for an average bite number of 12 bites. Once latency to the first bite increased, we were eventually able to increase volume and decrease tube feedings. These data are discussed in relation to behavioral momentum. |
| |
|
| 105. Intelligent Control with Hierarchical Stacked Neural Networks |
| Area: DEV; Domain: Theory |
| MICHAEL LAMPORT COMMONS (Harvard Medical School), Sofia Leite (University of Porto ), Sarthak Giri (Dare Association), Saranya Ramakrishnan (Harvard T.H. Chan School of Public Health) |
| Discussant: R. Douglas Greer (Columbia University Teachers College and Graduate School of Arts and Sciences) |
| Abstract: The use of hierarchical-stacked-neural networks will help computers learn through processing information and make complex decisions in a manner that simulates cognitive development in the human brain. This invention models the ordered stages that the brain moves through during development that allow it to perform increasingly complex actions at higher stages of development. In this developmental process, actions performed at a particular stage of development are created by ordering, combining, and transforming the actions performed in the immediately preceding stage. As a result of this process, at each stage of development more complex actions can be performed than those performed at the immediately preceding stage. Prior-art neural networks, in contrast to the present invention, are not modeled on the cognitive development of the human brain. They employ simple models of both biological systems and the physiological structure of the brain to process information and perform tasks. When prior-art, architecturally distinct neural networks are linked together to form hierarchies, the complexity of the actions performed in consecutive neural networks does not increase at higher levels in a hierarchy. Actions performed in lower level networks in the hierarchy are not systematically ordered, combined, and transformed to create higher-stage actions in higher-level networks in the hierarchy in the manner that the human brain uses during learning and development. As a result, prior-art neural networks, whether or not hierarchical, cannot perform many of the complex tasks that humans perform easily. |
| |
|
| 106. An Evaluation of Deferred Time-Out to Treat Noncompliance in the Classroom Setting |
| Area: DEV; Domain: Applied Research |
| JENNIFER S. KAZMERSKI (Baylor College of Medicine; Texas Children’s Hospital
), Jessica Buzenski (East Carolina University; St. Jude Children's Research Hospital; University of Tennessee Health Science Center), Ryan Ford (East Carolina University; Munroe Meyer Institute; University of Nebraska Medical Center) |
| Discussant: R. Douglas Greer (Columbia University Teachers College and Graduate School of Arts and Sciences) |
| Abstract: Initial studies of deferred time-out strategies have been conducted using single case design in a clinical setting. The current study will extend the results of previous evaluations of this new procedure by determining functional applications of deferred time-out , adding to the standard behavioral strategy of time-out. Deferred time-out was previously applied in a clinical setting with families seeking treatment but will now be implemented in the classroom setting with teachers and students. Deferred time-out was developed to increase compliant behavior in children who were not responsive to the traditional time-out procedure (Warzak & Floress, 2009). In the original study, deferred time-out was described as a process initiated when a child was resistant to time-out in the traditional form and a time-out training procedure was needed. The results from the initial research indicate that the deferred time-out strategy reduces time-out latency without the need for put-backs or other physical means in attempt to gain time-out compliance. As well as being more time consuming and less effective, these more physical strategies to force time-out compliance are much less favorable to caregivers than a solution such as deferred time-out that does not require other controversial means (Kazdin, 1980). In the school setting, similar concerns are present and strategies are needed to increase the feasibility of time-out in the classroom (Rathvon, 2012; Cowan and Sheridan, 2003). |
| |
|
| 107. A Case Study in Bullying Behavior: Functional Analysisand Treatment of Bullying Behavior in a Preschool Aged Girl |
| Area: DEV; Domain: Applied Research |
| LINDSAY M. KNAPP (St. Cloud State University), Julie A. Ackerlund Brandt (The Chicago School of Professional Psychology ) |
| Discussant: R. Douglas Greer (Columbia University Teachers College and Graduate School of Arts and Sciences) |
| Abstract: Bullying is the most-commonly cited form of violence in schools (Batsche, 1997). There are upwards of 30% of school children reporting they are a victim of bullying (Nansel et al., 2001; Swearer & Espelage, 2004). Previous researchers have focused on Positive Behavior Support-based procedures implemented using a school-wide system (Ross & Horner, 2009). However, this is not a function-based procedure; therefore, it may be more efficient and have greater reductions in bullying if there was a function-based procedure. One purpose of the current project was to complete a Functional Behavior Assessment (FBA) to assess the function of bullying in a typically developing preschool child. The results were that bullying was maintained by access to tangible items and escape from rules and demands. The second purpose was to use Functional Communication Training (FCT) to decrease bullying. FCT has been an effective intervention for teaching skills to individuals as an alternative for engaging in problematic behavior (Tiger et al., 2008). By teaching replacement behaviors, there was a decrease in bullying throughout the day. Implications of these results will also be discussed. |
| |
|
| 108. An Evaluation of the Effects of Choice Arrangements on Skill Acquisition for Typically Developing Preschool Children |
| Area: DEV; Domain: Applied Research |
| LINDSAY M. KNAPP (St. Cloud State University), Mackenzie Schroeder (St. Cloud State University), Julie A. Ackerlund Brandt (The Chicago School of Professional Psychology ) |
| Discussant: R. Douglas Greer (Columbia University Teachers College and Graduate School of Arts and Sciences) |
| Abstract: Providing choice opportunities has been successful for increasing appropriate behavior and decreasing inappropriate behavior; however, the arrangement of how choices are provided varies. In some cases, the choice is provided before the session begins, and in other cases, the choices are provided within the session (e.g., Smith, Iwata, & Shore, 1995; Graff & Libby, 1999). There may be benefits to both arrangements. For example, choices within the session may provide access to high preferred items based on momentary changes in preferences. In this case, a within-session choice arrangement would be a more effective arrangement than a pre-session choice arrangement. In the current study, we replicated and extended previous research by (a) determining the efficacy of different choice arrangements on skill acquisition. Results were that the majority of participants acquired skill across multiple conditions; however, for two of the four participants, the most skills were acquired during the pre-session choice condition. Possible limitations and implications of these results will also be discussed. |
| |
| |