|BIG SIG Symposium: Advances in the Behavior Analysis of Gambling
|Saturday, May 29, 2010
|3:00 PM–4:20 PM
|Lone Star Ballroom Salon E (Grand Hyatt)
|Area: EAB/CBM; Domain: Experimental Analysis
|Chair: Simon Dymond (Swansea University)
|CE Instructor: Robert Kohlenberg, Ph.D.PhD
|Abstract: This symposium from the Behaviorists Interested in Gambling Special Interest Group (BIG SIG) of the Association for Behavior Analysis International brings together behavior analytic research on gambling behavior. The first presentation describes how escape scores on a subscale of the Gambling Functional Assessment predict video poker play. The second presentation describes the findings of research on the "near miss" effect in slot machine gamblers. The third presentation presents the findings of experiments designed to understand derived transfer of response allocation in slot machine gambling. Finally, the fourth presentation presents data on the often assumed correlation between delay discounting performance and learning on the popular Iowa Gambling Task.
|Do escape scores on the GFA predict video poker play in the laboratory?
|JEFFREY N. WEATHERLY (University of North Dakota), Kevin Montes (University of North Dakota), Danielle Christopher (University of North Dakota)
|Abstract: The Gambling Functional Assessment (GFA; Dixon & Johnson, 2007) is a paper-pencil measure intended to help identifying the maintaining consequences for gambling behavior. Subsequent research has suggested that, along with potentially identifying such consequences, one category of the GFA (i.e., Escape) may also identify gamblers displaying pathology. The present study recruited 40 individuals to play video poker. Before doing so, they completed a questionnaire on their gambling history, the GFA, and a delay-discounting task. The hypothesis was that those scoring high in the Escape category of the GFA would play more hands, bet more credits, and make more mistakes when playing poker than those scoring low. Only some of these predictions were correct. However, Escape scores did just as well, and in some cases better, at predicting gambling behavior than did the gambling history questionnaire and the discounting measure, both of which should have been accurate predictors according to the literature.
|Formal and Functional Investigation and Manipulation of the “Near-Miss” Effect in Gamblers
|BECKY L. NASTALLY (Southern Illinois University), Mark R. Dixon (Southern Illinois University)
|Abstract: The present study experimentally investigated the potential for recreational and pathological gamblers to respond as if certain types of losing slot machine outcomes were actually closer to a win than others. Following demonstration of such behavior, termed the “near-miss effect” in the gambling literature, the present study sought to disrupt the near-miss effect via the exposure of participants to a brief conditional discrimination training and testing procedure. Subsequent performance of participants showed a decrease in the near-miss effect as measured by self report, or topographically, and through decreased response allocation toward a simulated slot machine programmed with a high density of near miss outcomes. The implications of the data for a verbally as well as functionally based behavioral explanation of gambling are presented.
|Derived Transfer of Response Allocation and Outcome Ratings in a Simulated Slot Machine Task
|SIMON DYMOND (Swansea University), Kate Mills (Swansea University), Amanda Cox (Swansea University), Victoria Crocker (Swansea University), Joanne Griffiths (Swansea University), Alice E. Hoon (Swansea University)
|Abstract: A defining feature of a contemporary behavior analytic account of gambling is that gambling related stimuli may obtain their functions based, at least in part, on participation in derived relations. In this way, gambling may be considered a verbal event. The present study describes the findings of three experiments designed to test this. In all experiments, non-problem gamblers were first trained and tested for the formation of 2, 3-member equivalence relations (A1-B1-C1; A2-B2-C2). Participants were then exposed to two simulated slot machines labeled with members of the relational network (B1 and B2, respectively). Slot machine B1 was programmed with a low payout probability (0.2) and slot machine B2 with a high payout probability (0.8). In Experiment 1, transfer to C1 and C2 was tested with a forced choice procedure, in Experiment 2 with all slot machine spins under extinction, and in Experiment 3 with machines of matched probabilities (0.5). Self-report ratings of the likelihood of winning were also obtained. Findings demonstrate derived transfer of response allocation and self-report ratings in accordance with equivalence relations, and highlight the utility of approaching gambling as a derived, verbal event.
|Is There a Correlation Between the Iowa Gambling Task and Delay Discounting?
|FRANK D. BUONO (Southern Illinois University Carbondale), Mark R. Dixon (Southern Illinois University)
|Abstract: The purpose of this paper was to identify if there is a correlation between performance on the Iowa gambling task (IGT; Bechara et al, 1992) and delay discounting by pathological gamblers. Participants were asked to complete both assessments in a counterbalanced delivery. Computerized versions of the tasks were constructed using Microsoft Visual Basic that allowed for the collection of response allocations as well as time-based measures by each subject. Results indicated that the two assessments share a fair degree of linear relationship with each other suggesting that convergent validity exists between the two assessments. However, neither predicted actual gambling performance on a subsequent slot machine task.