Association for Behavior Analysis International

When more than one option is available, why do people sometimes fail to select the alternative that is in their own long term best interests? This symposium will explore a range of contexts in which people often make non-optimal choices by gambling, persisting in losing endeavours and procrastinating instead of working towards valued academic goals. Talks will cover experimental, translational and applied research aimed at understanding and addressing the drivers of non-optimal choice.

Understanding the decision making processes involved in student procrastination could lead to the development of interventions for this common problem that improve student learning outcomes. Delay discounting refers to the fact that reinforcers lose their value if you have to wait for them. Procrastination may be caused by the fact that reinforcers for studying are delayed, however no task measuring delay discounting of academic outcomes currently exists. We developed a measure of academic discounting modelled on tasks successfully used in the discounting literature. Participants were first year psychology students. We piloted two versions of the academic discounting task and identified the superior version; all participants showed systematic discounting. In general, large delayed rewards are discounted less steeply than small delayed rewards (the magnitude effect). We found that the magnitude effect also exists in academic discounting; participants discounted a not important assignment more steeply than an important assignment. The overall results of these experiments show that delayed rewards are an important contributor to student procrastination

The sunk-cost effect describes a suboptimal tendency to continue investing in a losing course of action due to a prior investment, and despite a low expected payoff. Following recent demonstrations of the sunk-cost effect in non-human animals, we presented pigeons with discrete trials separated by 1-s inter-trial intervals. Each trial arranged either a short or long fixed-interval (FI) schedule and ended after a single food reinforcer. Short and long trials were not differentially signaled and pecks to a concurrently-available escape key terminated the current trial and initiated the next trial. Therefore, a sunk-time effect can be indexed by a tendency to persist through long trials, rather than escape after the duration of the short FI. The pattern of escaping shifted when we varied the initial investment and future payoff by manipulating the duration and probability of the short FI, respectively. Furthermore, when exposed to extended conditions, three of six pigeons reliably persisted while the other three reliably escaped after the short Fl. These two stable patterns of responding demonstrate the importance of obtained over arranged contingencies and suggest a key role of reinforcement history on sunk-cost decisions.

Slot machines are the primary mode of gambling for problem gamblers in New Zealand, yet are played by fewer people than other gambling activities like the national lottery or scratch tickets. The structural characteristics of gambling activities are an important factor in determining gambling behaviour. One feature of slot machines that has received little experimental analysis is the bonus feature or free spins that are commonly central to the design of slot machines. The current study used an experimental design with slot-machine simulations and hypothetical money, to investigate whether a free-spins feature increased the persistence of slot-machine gambling in the face of disruption. Preliminary datashow that participants rate of gambling is higher on a free-spins machine compared to a control machine without free spins. This procedure has the potential to answer more general questions about behavioural momentum in humans.

Treatment seeking disordered gamblers (n=11) were counselled using cognitive behavioural therapy (total sample n=44). Treatment consisted of 12-weeks of counselling where participants could attend up to three sessions per week. In addition, participants were able to earn study credit that was transferrable into gift vouchers redeemable at local businesses. Study credit was earned by treatment attendance, evidence of gambling abstinence based on their financial records, and corroboration from significant others of gambling abstinence. Study credit rates monotonically increased when participants attended consecutive sessions and/or provided consecutive evidence of gambling abstinence. Participants also completed as series of questionnaires and assessments including behavioural measures of inhibition and impulsivity (i.e., Stroop, delay-discounting). Gamblers nearing completion have attended approximately 20 sessions. This compares with typical out-patient gambling counselling retention rates by Alberta Health Services of approximately four sessions. Further, participants that have attended multiple sessions per week earned more gift-cards and had the longest number of abstinent weeks. This preliminary data suggests that contingency management can successfully be applied to the treatment of disordered gambling and that mass like behavioural momentum effects can be induced when multiple treatment sessions and the opportunity to earn credit are available.