Suppr超能文献

鸽子对匹配和不匹配的灵活学习。

Flexible learning of matching and mismatching by pigeons.

机构信息

University of Kentucky, USA.

University of Kentucky, USA.

出版信息

Behav Processes. 2022 Sep;201:104715. doi: 10.1016/j.beproc.2022.104715. Epub 2022 Jul 25.

Abstract

When pigeons learn a conditional discrimination in which a sample stimulus indicates which of two comparison stimuli is correct Skinner (1950) proposed that they learn a chain involving the sample, the correct comparison stimulus, and the reinforcer. This implies that they do not learn to reject the incorrect comparison stimulus and the sameness relation between the sample and the correct or the incorrect comparison stimulus plays little role in learning. There is, however, considerable evidence that learning to match or mismatch the sample can transfer to novel stimuli. But there is little evidence that the sameness relation facilitates acquisition. In the present research, pigeons were trained on two 0-s delay conditional discriminations: two matching tasks, two mismatching tasks, or one of each. No differences were found in acquisition, suggesting that consistent matching or mismatching does not facilitate acquisition of conditional discriminations. In testing, when either the correct or the incorrect comparison stimulus in each discrimination was replaced with one of the stimuli from the other task, results suggested that when learning both tasks, the pigeons in all three groups had learned both to select the correct stimulus and to reject the incorrect stimulus. It appears that the pigeons may have had learned the tasks based on sample/comparison-stimulus configurations with the sample serving as an occasion setter.

摘要

当鸽子学会了一种条件辨别,其中一个样本刺激表明两个比较刺激中的哪一个是正确的,斯金纳(Skinner, 1950)提出它们学会了一个涉及样本、正确的比较刺激和强化物的链。这意味着它们没有学会拒绝错误的比较刺激,而且样本和正确或错误的比较刺激之间的相同关系在学习中作用不大。然而,有大量证据表明,学习匹配或不匹配样本可以转移到新的刺激上。但是,几乎没有证据表明相同关系有助于获得。在本研究中,鸽子接受了两个 0 延迟条件辨别训练:两个匹配任务、两个不匹配任务或各一个。在获得方面没有发现差异,这表明一致的匹配或不匹配并不有助于条件辨别获得。在测试中,当每个辨别中的正确或错误比较刺激被另一个任务的一个刺激取代时,结果表明,当学习两个任务时,所有三组鸽子都学会了选择正确的刺激和拒绝错误的刺激。似乎鸽子可能已经学会了基于样本/比较刺激配置的任务,其中样本充当了一个引发者。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验