Association for Behavior Analysis International

Opioid Use Disorder (OUD) remains a significant public health challenge. In recent years, communities nationwide have been impacted by an influx of illicitly manufactured fentanyl – a synthetic opioid 50 times more potent than heroin and one increasingly associated with overdose fatalities. Identifying the behavioral processes underlying OUD is important component of informing effective treatment and prevention strategies. To this end, behavioral economic demand blends microeconomics and operant behavior analysis to understand the behavioral processes underlying drug consumption against increasing prices. Hypothetical purchase tasks have received increasing attention for their efficiency in assessing demand across different commodities. Compared to other non-medical opioids, however, little is known about demand in the context of fentanyl use. The current studies sought to address this gap by developing a novel fentanyl purchase task. In Study 1, volunteers in inpatient substance use treatment (n = 405) completed a heroin/fentanyl purchase task via Trac9 Informatics, a commercial treatment outcomes tool. Data were adequately described by quantitative demand modeling. Using a modified version of the task, Study 2 assesses demand in a similar sample of substance use treatment volunteers. Data will be discussed in terms of quantitative modeling fits and variables associated with systematic and non-systematic responding.

Delay discounting (DD), or the subjective devaluation of rewards as a function of their delay, is a useful construct when studying decisions involving a tradeoff between immediate and delayed outcomes (e.g., substance use, gambling, wearing seatbelts, eating). An implicit assumption not explicitly stated when scoring binary choice DD tasks is that there is a constant degree of certainty in preferences for outcomes, but this assumption is likely incorrect. We explored ambivalence in DD by leveraging the date-delay effect, where DD decreases when delayed outcomes emphasize calendar dates rather than delays in time. Participants (n=90, Mage 35.3, 75% male, 87% White) completed a Monetary Choice Questionnaire with temporal or date delays. To obtain ambivalence scores, participants used a sliding scale on each trial indicating their degree of relative preference for immediate outcomes at the far left (0) or delayed outcomes at the far right (100). We examined ambivalence scores on trials immediately proximal to the calculated individual-level discount rate. Contrasts of ambivalence scores showed no differences between delay and date conditions, but revealed significantly higher ambivalence on trials associated with preference for delayed outcomes (all p <.05). This research informs understanding of risk behaviors that may impact long-term health and safety.

Recent approaches in behavioral economics have used marijuana purchase tasks (MPTs) to understand driving after cannabis use (DACU). One factor that may influence DACU is the latency between cannabis use and having to drive. Participants (n = 167) completed 4 MPTs in the context of different cannabis use and driving latencies (i.e., no driving; 20-min, 1-hr, 6-hrs). We found a significant main effect of latency on all cannabis demand indices (all ps <.001). We also found significant differences in demand from no driving to the 20-minute latency condition for all observed indices (ps <.001). Additionally, significant decreases were also found for all observed demand indices when examining the comparisons between the typical situation and the 1-hour latency (all ps <.001, d range = 0.30 – 1.08) and between the typical situation and the 6-hour latency (p range = <.001 - .015, d range = 0.09 – 0.34). Notably, demand significantly increased across conditions as the latency between smoking and driving increased when the differences between the three driving MPTs were examined. These results suggest that cannabis demand is sensitive to specific parameters surrounding driving contingencies.

Behavioral economics posits that alcohol use is influenced by contextual variables and may therefore be impacted by the availability of alternative commodities. Behavioral economic purchase tasks isolate demand for a commodity alone (single-commodity) or when concurrently available with other commodities (cross-commodity). In cross-commodity purchase tasks, participants make choices across multiple concurrently available commodities, allowing researchers to pinpoint consumption patterns across commodities. While promoting substance-free alternative activities is a promising avenue for behavioral interventions, maximizing intervention efficacy requires examining the impact of alternatives on alcohol demand. We developed a novel cross-commodity purchase task adaptation in which alcohol was concurrently available with alternative activities of varying preference. Adults who currently consume alcohol were crowdsourced from Prolific (n=158) and were asked to rank-order their preferences for a list of different activities. The highest, second-highest, and lowest preferred activities were incorporated into three separate alcohol vs. activity cross-commodity purchase tasks. Results showed the availability of any activity resulted in statistically significant reductions in demand for alcohol. Systematic differences in hypothetical engagement in the alternative activity were observed as a function of preference rank. Implications for the utility of commodity purchase tasks in recovery efforts for hazardous alcohol use are discussed.