|
| Effects of Reinforcement History: From Laboratory to Application |
| Tuesday, August 14, 2007 |
| 9:30 AM–10:50 AM |
| L2 Room 2 |
| Area: EAB; Domain: Experimental Analysis |
| Chair: Claire C St. Peter (West Virginia University) |
| Discussant: Carlos A. Bruner (National University of Mexico) |
| CE Instructor: Claire C St. Peter, Ph.D. |
| Abstract: Recently, Critchfield (2006) urged behavior analysts to submit symposia geared toward diverse audiences as one way of maintaining the continuity of the field. We have attempted to answer his call by developing a symposium focused on reinforcement history, but approaching the topic from diverse perspectives. We therefore included a review of the use of nave animals in studies published in the Journal of the Experimental Analysis of Behavior (Lattal & Okouchi), a research study examining effects of the order of delays in delay-discounting procedures (Anderson, Diller, & Slezack), and a research study examining the effects of exposure to reinforcement schedules commonly used in applied behavior analysis during baseline and treatment (St. Peter Pipkin & Vollmer). In all, we examine the potential effects of reinforcement history in nonhuman animal, human operant, and applied research in the hopes of building bridges across seldom-connected disciplines. Reinforcement history has important implications for all areas of behavior analysis; we hope to illuminate a portion of these in our presentations. |
| |
| To Be or Not to Be Naïve: On the Behavioral Histories of Subjects in Experiments. |
| KENNON A. LATTAL (West Virginia University), Hiroto Okouchi (Osaka Kyoiku University) |
| Abstract: We reviewed all of the experimental studies involving nonhuman animals published in the Journal of the Experimental Analysis of Behavior between 1958 and 2005 to determine the extent to which experimenters used subjects that were or were not experimentally naïve. Over the years, the use of naive animals has declined in favor of animals with a prior history of use in behavioral experiments. There are differences in species, with monkeys and pigeons being more likely to have had a prior history of experimentation. Using naïve or non-naïve animals also depended on the nature of the experiment, in particular whether the experiment was a part of a series of experiments on a given topic. These outcomes are discussed in relation to Skinner’s suggestion in a JEAB paper on the future of behavior analysis that the importance of using naïve animals in experiments is exaggerated. |
| |
| History of Delayed Presentation Affects Delay-Discounting Function. |
| KAREN G. ANDERSON (West Virginia University), James W. Diller (West Virginia University), Jonathan M. Slezak (James Madison University) |
| Abstract: As the delay to the presentation of a relatively large reinforcer is increased, the choice for that outcome (the self-controlled option) may be reduced below the choice for an alternative smaller, more immediate reinforcer (the impulsive option). The rate at which the value of the larger reinforcer is discounted as its delay increases has been shown to vary across individuals. Higher rates of delay discounting have been correlated with various behavioral disorders, e.g., substance abuse, gambling, violence. To examine a role for behavioral history in influencing differential rates of discounting, rats were exposed to a choice between one food pellet delivered immediately and three food pellets delivered after different delays. In one study, history with a fixed, ascending order of delay presentation systematically decreased larger-reinforcer choices. In subsequent studies, the order (within and across sessions) in which the various delays to the larger reinforcer were presented affected later delay-discounting functions. Together, these studies have implications for interpreting data from delay-discounting paradigms and may suggest ways to affect impulsive choice. A better understanding of the historical variables that determine impulsive/self-control choice may yield better prevention and treatment strategies for many behavioral disorders. |
| |
| Effects of Recent Reinforcement History on Responding During Random-Ratio Schedules. |
| CLAIRE C ST. PETER (West Virginia University), Timothy R. Vollmer (University of Florida) |
| Abstract: Recent reinforcement history may affect responding during subsequent reinforcement schedules, but these types of “history effects” have not been thoroughly examined in relation to applied issues. For example, we do not yet know the effects histories with the schedules commonly used in applied research on subsequent responding. We examined the effects of typical baseline (concurrent FR1/EXT) and full treatment (concurrent EXT/FR1) histories on responding during equal random ratio schedule (concurrent RR2/RR2) in a reversal design, using both human operant and applied methods. A computer analog for problem and appropriate behavior was used in experiments I and II, with nonclinical human adults as participants. Experiment I evaluated possible history effects with brief exposures to each type of reinforcement schedule. In experiment II, phases were conducted to stability to assess the potential long-term impact of history and to determine if effects observed in experiment I were merely transition states. Experiment III involved a replication of experiment II with a child with disabilities who engaged in aggression in a school setting. In all three experiments, recent reinforcement history affected responding during the random ratio schedules. We conclude by discussing implications of history for application and possible avenues for further examination of history effects. |
|
| |