Peterson Jennifer R, Hill Catherine C, Kirkpatrick Kimberly
Department of Psychological Sciences, Kansas State University.
J Exp Anal Behav. 2015 Jan;103(1):166-79. doi: 10.1002/jeab.124. Epub 2014 Dec 9.
Impulsive choice is typically measured by presenting smaller-sooner (SS) versus larger-later (LL) rewards, with biases towards the SS indicating impulsivity. The current study tested rats on different impulsive choice procedures with LL delay manipulations to assess same-form and alternate-form test-retest reliability. In the systematic-GE procedure (Green & Estle, 2003), the LL delay increased after several sessions of training; in the systematic-ER procedure (Evenden & Ryan, 1996), the delay increased within each session; and in the adjusting-M procedure (Mazur, 1987), the delay changed after each block of trials within a session based on each rat's choices in the previous block. In addition to measuring choice behavior, we also assessed temporal tracking of the LL delays using the median times of responding during LL trials. The two systematic procedures yielded similar results in both choice and temporal tracking measures following extensive training, whereas the adjusting procedure resulted in relatively more impulsive choices and poorer temporal tracking. Overall, the three procedures produced acceptable same form test-retest reliability over time, but the adjusting procedure did not show significant alternate form test-retest reliability with the other two procedures. The results suggest that systematic procedures may supply better measurements of impulsive choice in rats.
冲动选择通常通过呈现较小即时（SS）与较大延迟（LL）奖励来衡量，对SS的偏好表明冲动性。当前研究在不同的冲动选择程序中对大鼠进行测试，通过操纵LL延迟来评估同形式和交替形式的重测信度。在系统-GE程序（Green & Estle，2003）中，经过几次训练后LL延迟增加；在系统-ER程序（Evenden & Ryan，1996）中，延迟在每次训练中增加；在调整-M程序（Mazur，1987）中，根据每只大鼠在前一个试验块中的选择，延迟在每次试验块后发生变化。除了测量选择行为外，我们还使用LL试验期间反应的中位数时间来评估对LL延迟的时间跟踪。经过广泛训练后，两种系统程序在选择和时间跟踪测量方面产生了相似的结果，而调整程序导致相对更多的冲动选择和较差的时间跟踪。总体而言，随着时间的推移，这三种程序产生了可接受的同形式重测信度，但调整程序与其他两种程序相比，未显示出显著的交替形式重测信度。结果表明，系统程序可能为大鼠的冲动选择提供更好的测量方法。
