Thrailkill Eric A, Elste Noah, Thorpe Catherine R, Bouton Mark E
University of Vermont, Department of Psychological Science.
J Exp Psychol Anim Learn Cogn. 2025 Apr;51(2):73-91. doi: 10.1037/xan0000395.
Habits are important in everyday life and are thought to be involved in several human behavioral pathologies, including addictions. Experiments with rats suggest that habit, as indexed by insensitivity of an instrumental response to separate devaluation of its outcome, develops with extended practice. Motivated behavior often involves a sequence or chain of behaviors (Rs), with each cued by a different discriminative stimulus (S). We therefore examined performance of a two-response discriminated heterogeneous behavior chain (R1-R2) in which R1 and R2 were occasioned by different Ss and were both required to earn a reinforcer. We further asked whether extended training decreases the sensitivity of R1 to the extinction of R2, which is known to decrease R1 and is analogous to an outcome devaluation effect. In Experiment 1 with rats, R1 was sensitive to extinction of R2 after moderate but not extended training, suggesting the development of habit. In Experiment 2, human participants learned three R1-R2 chains before one "R2" was extinguished. Extinction of R2 specifically decreased performance of the R1 that had been associated with it, but extended training did not reduce this effect. Based on findings in the nonhuman literature, Experiment 3 then had human participants learn only one R1-R2 chain before R2 was extinguished. Under these conditions, R1 became insensitive to extinction of R2 after extended training, consistent with the idea that habit can develop in a laboratory experiment with humans. The findings are discussed relative to difficulties demonstrating habits in humans. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
习惯在日常生活中很重要,并且被认为与多种人类行为病理学有关,包括成瘾行为。对大鼠的实验表明,习惯(通过工具性反应对其结果单独贬值的不敏感性来衡量)会随着长期练习而形成。有动机的行为通常涉及一系列行为(Rs),每个行为都由不同的辨别性刺激(S)引发。因此,我们研究了一种双反应辨别异质行为链(R1-R2)的表现,其中R1和R2由不同的S引发,且两者都需要获得强化物才能实现。我们还进一步探讨了长期训练是否会降低R1对R2消退的敏感性,已知R2的消退会降低R1,这类似于结果贬值效应。在大鼠实验1中,经过适度但非长期训练后,R1对R2的消退敏感,这表明习惯正在形成。在实验2中,人类参与者在一个“R2”被消退之前学习了三条R1-R2链。R2的消退特别降低了与之相关的R1的表现,但长期训练并没有减少这种影响。基于非人类文献中的研究结果,实验3让人类参与者在R2被消退之前只学习一条R1-R2链。在这些条件下,经过长期训练后,R1对R2的消退变得不敏感,这与习惯可以在人类实验室实验中形成的观点一致。我们将结合在人类中证明习惯的困难来讨论这些发现。(PsycInfo数据库记录(c)2025美国心理学会,保留所有权利)