Delay Discounting of Different Outcomes in a Sample of American Indian and Non-indian College Students

Delay discounting occurs when an individual prefers a lesser amount of an outcome that is available immediately, rather than waiting for the full amount. The present study was a preliminary investigation into delay discounting in a yet unstudied population, American Indians (AIs). AI college students completed a delay-discounting task that consisted of fi ve different outcomes (e.g., money, retirement income, obtaining the ideal body image). An equal-sized group of Caucasian respondents was then matched to the AI sample in terms of sex, age, and grade point average. Results demonstrated that AI and Caucasian respondents sometimes differed in how they discounted certain outcomes, suggesting that the value of these outcomes may differ across ethnicities. Further, the AI participants displayed different rates of discounting across the different outcomes, indicating that those outcomes may hold different values for the AI respondents. The potential value of delay discounting in understanding cross-cultural and intra-cultural differences is discussed and, because of the preliminary nature of the present study, a call for additional research is made. It is not unusual for people to choose a lesser amount of something in exchange for getting that something immediately, rather than having to wait to get the full amount. For instance, when individuals win the lottery, it is common for them to choose to receive a smaller “lump sum” of cash immediately, rather than having their full winnings paid out in installments across many years (e.g., Baker, Johnson, & Bickel, 2003). This decision sometimes makes intuitive sense because the future is uncertain. In the case of the lottery, one must depend on the institution responsible for paying out the money over time having the continued ability to pay. Further, one’s own future is never certain; by waiting, one runs the risk of not surviving long enough to collect the full amount. The study of how and why people make decisions between getting a certain amount of an outcome now versus waiting for a greater amount of that outcome later is known as the study of temporal or delay discounting (e.g., see Critchfi eld & Kollins, 2001, or Madden & Bickel, 2010, for reviews). Delay discounting research has its roots in early delay-of-gratifi cation work (Rotter, American Indian and Alaska Native Mental Health Research Copyright: Centers for American Indian and Alaska Native Health Colorado School of Public Health/University of Colorado Anschutz Medical Campus (www.ucdenver.edu/caianh) 44 VOLUME 17, NUMBER 2 1954). Although the concepts are related, delay discounting involves making multiple choices across a series of delays (see Madden & Bickel, 2010). Delay of gratifi cation, on the other hand, often involves a single, dichotomous choice. In terms of delay discounting, how much the subjective value of an outcome decreases as the full amount is increasingly delayed determines the “rate” at which the individual discounts (i.e., a single rate of discounting is calculated across the multiple delays). The rate of discounting varies systematically as a function of certain factors. For instance, the greater the value of the outcome, the less discounting is observed (e.g., Chapman, 1996; Smith & Hantula, 2008), a fi nding known as the magnitude effect. Take the situation in which someone owed you $10 but could not pay you the full amount for two weeks. In such a situation, you might be willing to accept $9 today rather than waiting two weeks. On the other hand, if someone owed you $10,000, you would be unlikely to accept $9,000 today rather than waiting two weeks for the full amount. Thus, over the same delay, the value of the smaller amount has been discounted by at least 10% whereas the value of the larger amount has not. The study of delay discounting has received a great deal of research interest over the past decade because the rate at which people discount future outcomes has been shown to vary as a function of group membership, psychological characteristics, and/or experience. Such fi ndings are of interest because they provide insight into the decision-making process that helps defi ne those variables, as well as raising the issue of whether those variables change as a function of delay discounting or whether the ways in which people discount delayed outcomes changes as a function of those variables. For example, individuals who are pathological gamblers tend to discount hypothetical monetary rewards more steeply than do individuals who do not gamble pathologically (e.g., Dixon, Marley, & Jacobs, 2003; see Petry, 2005, for a review), raising the question of whether increased discounting leads to pathological behavior, pathology alters discounting, or there are other factors that lead to both. Ostaszewski (1996) showed that extraverted, highly impulsive respondents displayed greater rates of discounting than did their counterparts. Do changes in discounting lead one to be more extraverted? Logue and Anderson (2001) found that experienced university administrators discounted future budgetary amounts more steeply than did less experienced administrators, a fi nding the authors suggested indicated that experienced administrators have learned not to trust promises of future budget increases. Alternatively, are administrators inherently conservative? As noted above, fi nding that rates of delay discounting vary across groups is interesting, but the results are correlational in nature. What is potentially more interesting is the fi nding that measures of delay discounting may predict actual behavior in certain situations (e.g., Weatherly, Marino, Ferraro, & Slagle, 2008). If that is the case, then knowing how an individual discounts delayed outcomes could potentially be used in a preventative fashion by allowing therapists or American Indian and Alaska Native Mental Health Research Copyright: Centers for American Indian and Alaska Native Health Colorado School of Public Health/University of Colorado Anschutz Medical Campus (www.ucdenver.edu/caianh) DELAY DISCOUNTING 45 counselors to address issues before they become problematic for the individual. That is, knowing how a student discounts a certain outcome may inform school counselors as to whether the student may be at risk for engaging in problem behavior (e.g., experimenting with drugs or sex). Furthermore, measures of delay discounting could be used as dependent measures that might indicate whether or not a particular treatment is working. For example, if delay discounting contributes to pathological gambling, then a successful gambling treatment should produce changes in the individual’s rate of discounting. To our knowledge, no investigations have yet attempted to study delay discounting in AIs, although some (e.g., Granzberg, 1973) have attempted to study delay of gratifi cation. Studying delay discounting among AIs would seem inherently warranted, if not socially compelling, for several reasons. One reason is theoretical. That is, does delay discounting vary as function of culture or ethnicity? Although very few studies have attempted to address this question, results from one cross-cultural study (Du, Green, & Myerson, 2002) would suggest it does. Du et al. found differences in the rate that American, Chinese, and Japanese graduate students temporally discounted hypothetical monetary amounts. Finding differences in delay discounting across different cultures and ethnicities would be of interest because the study of those differences could potentially provide a metric of the values held by different cultures. A second reason for studying delay discounting in AIs is a practical one. AIs differ from the non-AI majority population in regards to a number of (mental-) health-related factors, including substance abuse, pathological gambling, and psychopathology (e.g., McDonald & Chaney, 2003; Wardman, el-Gueblay, & Hodgins, 2001). Given that research has shown that delay discounting differs as a function of disorders such as pathological gambling (e.g., Dixon et al., 2003) and substance abuse (Petry, 2001; Petry & Casarella, 1999), one might expect to fi nd that AIs differ in their delay discounting relative to their majority-population counterparts. On the other hand, if similar rates of delay discounting were to be found between AIs and non Indians, then such results would suggest that the higher rates of disordered behavior among AI populations are likely the outcome of other factors not directly related to decision-making characteristics (e.g., socioeconomic status, educational opportunities, degree of acculturation). A third reason to pursue the study of delay discounting in AI populations is that it may ultimately shed light on cultural differences between tribes. Although the full pursuit of this reason was beyond the scope of the present preliminary investigation, it may be the case that tribes in more impoverished areas may show differences in discounting of monetary outcomes relative to tribes in less impoverished areas. Similarly, differences may be found in how members of different tribes discount outcomes related to their own health, personal well-being, or government policies. Thus, knowing how members of certain tribes discount delayed outcomes could be informative for American Indian and Alaska Native Mental Health Research Copyright: Centers for American Indian and Alaska Native Health Colorado School of Public Health/University of Colorado Anschutz Medical Campus (www.ucdenver.edu/caianh) 46 VOLUME 17, NUMBER 2 policy makers at both the local and national levels. For example, if one knew how tribal members discounted particular outcomes (e.g., the building of a new fi re station), then policy makers would have a better idea about how much the members might be willing to pay now to get those outcomes at some date in the future. The present study was a preliminary investigation into measuring delay discounting in a sample of AI university students. Two different groups of AI students completed discounting tasks on a different set of fi ve outcomes and the results were then compared to an equal-sized group of Caucasian students matched to the AI sample in terms of sex, age, and grade point average. Two issues were of particular interest in this preliminary investigation. First, would AI and Caucasian students differ in how they discounted the different outcomes? Second, what differences would be observed within the AI samples in how the respondents discounted the different types of outcomes? Because of the exploratory nature of this study, we did not have a priori hypotheses about the answers to these questions.