|
| Recent Advances in Preference and Reinforcement Procedures |
| Monday, May 30, 2005 |
| 3:00 PM–4:20 PM |
| Stevens 1 (Lower Level) |
| Area: DDA; Domain: Applied Research |
| Chair: Eileen M. Roscoe (New England Center for Children) |
| CE Instructor: Eileen M. Roscoe, Ph.D. |
| Abstract: A great deal of research has demonstrated the utility of systematic preference and reinforcer assessments. The present series of studies detail refinements to current methods by evaluating the effects of including different types or quality of stimuli. The first study compared the effects of substituting tokens for high- and low-preference edibles both before and after token training. In the second study, preference and reinforcer efficacy for different magnitudes of one or two functional reinforcers (attention or escape) were assessed. The third study compared the effects of choice of task order during three conditions, a participant-selected condition, an experimenter-selected condition that was not yoked to the participant-selected condition, and an experimenter-selected condition that was yoked to the participant-selected condition. In the final paper, the role of reinforcer potency for correct responses was evaluated when task interspersal procedures were and were not in effect. Collectively, these studies extend the literature on preference assessments by examining the conditions under which novel and qualitatively different items may predict reinforcer effects. |
| |
| Assessment and Training of Tokens Utilizing Preference Assessment Methodology |
| ELISA M. HEGG (New England Center for Children), Daniel Gould (New England Center for Children), Myrna E. Libby (New England Center for Children), Richard B. Graff (New England Center for Children) |
| Abstract: Two adolescents participated in several phases of preference and reinforcer assessments. First, multiple-stimulus preference assessments were conducted with edibles, and preference hierarchies were generated. Novel tokens (without backup reinforcers) were substituted for the most- and least-preferred edibles, and preference assessments were again conducted; tokens (without backup reinforcers) ranked as the least preferred stimuli. Tokens (without backup reinforcers) did not function as reinforcers on subsequent reinforcer assessments. Next, token training paired one novel token with the most-preferred edible; the other token was paired with the least-preferred edible. Following token training, another preference assessment was conducted with the tokens substituted for the most- and least-preferred edible; however, when tokens were selected, they were now exchanged for the back-up reinforcer. The token paired with the most-preferred edible ranked highest in the preference hierarchy, while the token paired with the least-preferred edible ranked lowest. In the final reinforcer assessment, high-preference foods and tokens paired with high-preference foods both generated high rates of responding, while low-preference foods and tokens paired with low-preference foods both generated lower rates of responding. Interobserver agreement data were collected in at least 38% of preference and reinforcer assessment across participants, and was at least 97% for all assessments. |
| |
| The Importance of Reinforcement Magnitude: An Examination of Preference and Reinforcer Efficacy |
| NICOLE M. TROSCLAIR-LASSERRE (Louisiana State University), Dorothea C. Lerman (Louisiana State University), Tiffany Kodak (Louisiana State University) |
| Abstract: Reinforcers that maintain problem behavior often are an integral part of treatment with differential reinforcement. However, the magnitude of the reinforcer (e.g., duration of attention) has varied widely across studies and seemed to be chosen arbitrarily. Relatively little is known about children’s preference for different magnitudes of functional reinforcers or the relationship between magnitude and treatment effectiveness. The purpose of this study was to evaluate the basic relation between reinforcer magnitude, preference, and efficacy. Preference for different magnitudes of one or two functional reinforcers (e.g., attention, escape) was evaluated with two children who engaged in problem behavior to identify values that were and were not associated with a relative preference. Next, the relation between preference and reinforcer efficacy was evaluated via progressive ratio schedules. Results indicated that children may show a preference for different magnitudes of reinforcement and that preference may be a predictor of reinforcer efficacy. |
| |
| Effects of Choice of Task Sequence in Individuals With Developmental Disabilities in Public Schools |
| SHERRY STAYER SMELTZER (New England Center for Children), Richard B. Graff (New England Center for Children), Myrna E. Libby (New England Center for Children), William H. Ahearn (New England Center for Children) |
| Abstract: Although researchers have examined the effects of choice on responding, there is limited research on the effects choice of task order on responding. This study examined the effect of choice of task order on the on-task behavior, duration, and challenging behaviors in 2 children with autism and one diagnosed with Fragile X syndrome. Three low-preference tasks were identified for each child (using brief multiple-stimulus without replacement preference assessments) and then utilized in an experimental design that included alternating treatments and concurrent operants phases. During single operant phases, when participants selected the task order, all participants showed higher on-task behavior, decreased total duration, and decreased rates of challenging behavior compared to when the experimenter selected the task order. However, when the task order in the experimenter-selected sessions was yoked to the order selected by the participant, little difference in responding was noted. When given the ability to choose between two conditions in the concurrent operants phase, participants preferred the condition in which they could choose the task order. Interobserver agreement data (IOA) were collected in approximately 33% of sessions across participants; mean IOA was 96% across participants. |
| |
| The Role of Reinforcer Preference in the Effectiveness of Task Interspersal Procedures |
| VALERIE M. VOLKERT (Louisiana State University), Dorothea C. Lerman (Louisiana State University), Nicole M. Trosclair-Lasserre (Louisiana State University), Laura R. Addison (Louisiana State University), Tiffany Kodak (Louisiana State University) |
| Abstract: Past research has demonstrated that interspersing known (i.e., maintenance) tasks with unknown (i.e., acquisition) tasks facilitates learning under certain conditions. However, little is known about factors that influence the effectiveness of this popular treatment strategy. In the current investigation, we evaluated the effects of reinforcer potency on interspersal procedures while teaching object labels to three children diagnosed with autism or with autistic-like features. The potency of the reinforcer delivered for correct responses on trials with unknown and/or known object labels was manipulated, and performance with and without the interspersal technique was compared. Performance was only enhanced under the interspersal condition when either (a) less preferred reinforcers (e.g., praise only) were delivered for all correct responses, or (b) more preferred reinforcers (e.g., praise plus food) were provided for performance on known trials than on unknown trials. This latter finding is inconsistent with recommendations for using the interspersal technique. Interoberserver agreement coefficients exceeded 80% or 90% for all participants. |
|
| |