Studies on the interaction of amount and delay of reinforcement in choice.

本研究探讨了并发和连续选择情境下强化量和延迟的相互作用。在并发选择情境中，大鼠被给予不同强化量和延迟的选择。首先，延迟的比率恒定为3:2，而它们的绝对值是变化的。较短的延迟以1粒食物丸结束，而较长的延迟以3粒食物丸结束。随着延迟强化的绝对值的增加，对延迟较长但较大的强化物的偏好增加。其次，在并行链计划的两种不同终端链接计划下，一种选择的食物颗粒数量不同，而另一种选择的食物颗粒数量不变（1粒）。大鼠在较短的末端连接（5秒）下的反应率低于强化物数量比，而在较长的末端连接（20秒）上的反应率高于强化物数量比。在连续选择情境下，大鼠和鸽子暴露于具有三种不同状态（搜索、选择和处理状态）的觅食计划中。较短（或较小）的强化物在较长（或较大）强化物之前较长的处理时间更容易被接受。所有的结果都与选择模型的预测不一致，假设强化的数量和延迟的简单乘法相互作用。

The present study investigated the interaction of amount and delay of reinforcement in concurrent and successive choice situations. In concurrent choice situations, rats were given a choice between alternatives differing in amount and delay of reinforcement. First, the ratio of the delays was constant at 3:2 while their absolute values were varied. The shorter delay terminated with 1 food pellet whereas the longer delay terminated with 3 food pellets. As the absolute values of the delays to reinforcement increased, preference for the longer-delayed but larger reinforcer increased. Second, the number of food pellets was varied in one alternative while it was constant (1 pellet) in the other alternative under two different terminal link schedules of a concurrent-chains schedule. Rats undermatched the response ratios to reinforcer amount ratios with the shorter terminal links (5 sec) whereas they overmatched with the longer terminal links (20 sec). In successive choice situations, rats and pigeons were exposed to a foraging schedule characterized by three different states (search, choice and handling states). The shorter duration (or smaller) reinforcer was more likely to be accepted the longer the handling time preceding the longer duration (or larger) reinforcer. All ofthe results are inconsistent with the predictions derived from the choice model assuming a simple multiplicative interaction of amount and delay of reinforcement.