Stimulus selectivity of drug purchase tasks: A preliminary study evaluating alcohol and cigarette demand

The use of drug purchase tasks to measure drug demand in human behavioral pharmacology and addiction research has proliferated in recent years. Few studies have systematically evaluated the stimulus selectivity of drug purchase tasks to demonstrate that demand metrics are specific to valuation of or demand for the commodity under study. Stimulus selectivity is broadly defined for this purpose as a condition under which a specific stimulus input or target (e.g., alcohol, cigarettes) is the primary determinant of behavior (e.g., demand). The overall goal of the present study was to evaluate the stimulus selectivity of drug purchase tasks. Participants were sampled from the Amazon.com's crowdsourcing platform Mechanical Turk. Participants completed either alcohol and soda purchase tasks (Experiment 1; N = 139) or cigarette and chocolate purchase tasks (Experiment 2; N = 46), and demand metrics were compared to self-reported use behaviors. Demand metrics for alcohol and soda were closely associated with commodity-similar (e.g., alcohol demand and weekly alcohol use) but not commodity-different (e.g., alcohol demand and weekly soda use) variables. A similar pattern was observed for cigarette and chocolate demand, but selectivity was not as consistent as for alcohol and soda. Collectively, we observed robust selectivity for alcohol and soda purchase tasks and modest selectivity for cigarette and chocolate purchase tasks. These preliminary outcomes suggest that demand metrics adequately reflect the specific commodity under study and support the continued use of purchase tasks in substance use research.