On the Complexity of Discounting, Choice Situations, and People

Although steep delay discounting is associated with various behavioral problems, perhaps most prominently substance abuse, we argue that it is best not conceived of as a character flaw such as impulsivity. Such a view, although part of a centuries-old tradition, does not distinguish between actions whose outcomes involve gains and losses, or between delayed outcomes and probabilistic outcomes, nor does it acknowledge that how steeply an individual discounts one of these kinds of outcome appears to be independent of how steeply they discount other kinds. Therefore, we advocate an approach that does not require making judgments about the character of the individuals involved. We show that when drug-dependent individuals are compared with controls, a substantial number of the drug-dependent individuals discount delayed monetary rewards less steeply than the average (median) member of the control group. Moreover, a substantial number of the controls discount more steeply than the average drug-dependent individual. Finally, we point out that many everyday choice situations differ from those studied in most discounting experiments in that they involve both gains and losses as well as qualitatively different outcomes that may be both delayed and probabilistic. Past research on discounting that focused on simple choice situations has provided a solid foundation, but research on more complicated situations is now needed. We believe that the principles revealed by such research will both inform the choices of treatment providers and improve our understanding of the complicated decisions that people face every day.