Studies in which coin-like tokens have been exchanged for food (e.g., Cowles, 1937; Kelleher, 1958; Silberberg, Bauman & Hursh, 1993; Wolfe, 1936) show that token use has properties uncharacteristic of other conditioned-reinforcement effects with which it is so often compared. In particular, when tokens are given or are earned by apes or monkeys, they seem more sensitive to the long-term consequences of their behavior. This fact is reflected in their capacity to bridge long delays between working for tokens and exchanging them for food, and their greater willingness to trade greater immediacy of food and higher rates of food delivery for larger amounts of food per day that are delivered at lower rates and with longer delays. Unlike other conditioned reinforcers, tokens seem to function for apes and monkeys as money does in the human economy, not only as a medium of exchange, but also as a store of value. The implications of this store-of-value role are broad. One of the ideas entertained in the proposal is that the apparently large differences in time horizon between humans and other animals may reflect not just a species difference, but also the effects of money and money use on human behavior. This possibility is tested directly in five experiments. Depending on the study, monkeys are either given or earn tokens that can be exchanged for food either immediately or after a delay. Their performances are typically compared with those in similar studies in which a panel press (rather than a token deposit) serves as the operant. The experimental effects addressed include the role of tokens in risky choice, foraging behavior, self control and conditioned reinforcement. This range of experiments permits asking two questions. Does token-based behavior differ from that seen when performances are maintained by other types of conditioned reinforcers? How are any differences obtained related to operant phenomena where behavioral economic analyses have already been applied with some success (e.g., self control, risky choice)? The results obtained could accommodate the attempts of economically minded behavior analysts to develop models capable of explaining human and non-human behavior in terms of economic principles.
