Wogar M A, Bradshaw C M, Szabadi E
Department of Psychiatry, University of Manchester, U.K.
Q J Exp Psychol B. 1992 Jul;45(1):1-13.
Choice between two reinforcers differing in magnitude and delay was investigated in rats using an adjusting-delay discrete-trials schedule in which the two reinforcers were associated with two levers (A and B). The delay to Reinforcer A (the smaller reinforcer) was always 2 sec, whereas the delay to Reinforcer B was varied in accordance with the distribution of choices in successive blocks of trials. In Experiment 1, the mean delay to the large reinforcer during the last 5 of 60 training sessions was greater when the rats were maintained at 80% than when they were maintained at 90% of their free-feeding body weights. In Experiment 2, the delay to the larger reinforcer was greater when the two reinforcers consisted of one and two 45-mg food pellets than when they consisted of three and six pellets. The results are consistent with a model of "self-control" which posits hyperbolic relations between reinforcer value and reinforcer magnitude, and between reinforcer value and delay of reinforcement.
在大鼠中,使用一种调整延迟的离散试验程序,研究了在大小和延迟方面存在差异的两种强化物之间的选择。在该程序中,两种强化物与两个杠杆(A和B)相关联。强化物A(较小的强化物)的延迟始终为2秒,而强化物B的延迟则根据连续试验块中的选择分布而变化。在实验1中,在60次训练的最后5次中,当大鼠维持在其自由进食体重的80%时,到大强化物的平均延迟比维持在90%时更大。在实验2中,当两种强化物分别由1个和2个45毫克食物颗粒组成时,到较大强化物的延迟比由3个和6个颗粒组成时更大。这些结果与一种“自我控制”模型一致,该模型假定强化物价值与强化物大小之间以及强化物价值与强化延迟之间存在双曲线关系。