The instability of graded structure: implications for the nature of concepts

The instability of graded structure: implications for the nature of concepts
复制标题

分级结构的不稳定性:对概念本质的影响

DOI:
--
复制
发表时间:
1987
期刊:
影响因子:
--
通讯作者:
L. Barsalou
L. Barsalou
中科院分区:
--
文献类型:
--
作者:
L. Barsalou

文献摘要

被引文献

相似文献

在简要介绍分级结构之后,本章回顾了实证研究结果,表明类别的分级结构是不稳定的,在不同背景下变化很大。讨论了这些发现对分类理论的影响,并得出结论:分级结构并不代表类别的不变结构特征。相反,有人提出,分级结构的稳定性之所以会出现,是因为不同的概念在不同的场合暂时代表工作记忆中的同一类别。概念不是作为静态单元从记忆中检索来表示类别,而是源于一个高度灵活的过程,该过程从长期记忆中检索通用和情景信息,以在工作记忆中构建临时概念。因为这个概念构建过程受到目标、背景和近期经验的高度限制,所以很少为一个类别构建相同的概念。提出了概念构建理论,以及该理论与梦的关系(Foulkes 1985)、概念组合(Hampton in press a,b;Smith & Osherson 1984)、范例理论(Brooks 1978;Jacoby & Brooks .1984;Media & Schaeffer 1978)、规范理论(Kaimeman &: Miller, 1986)以及并行分布式处理。 (McClelland & Rumelhart 1986;Rumelhart & McCleland,1986)进行了讨论。
After a brief introduction to graded structure, this chapter reviews empirical findings showing that the graded structure of a category is unstable, varying widely across contexts. Implications of these findings for theories of categorization are discussed, and it is concluded that graded structures do not represent invariant structural characteristics of Categories. Instead it is proposed that in stability in graded structure occurs because different concept temporarily represent the same categoy in working memory on different occasions.Rather than being retrieved as static units from memory to represent categories, concepts originate in a highly flexible process that retrieves generic and episodic information from long-term memory to construct temporary concepts in Work-ing memorу. Because this concept construction process is highly constrained by goals, context, and jecent experience, the same concept is rarely if ever constructed for a category. A theory of concept construction is presented, and the relations of this theory to dreaming (Foulkes 1985), conceptual combination (Hampton in press a,b; Smith & Osherson 1984), exemplar theories (Brooks 1978; Jacoby & Brooks .1984; Media & Schaeffer 1978), norm theory (Kaimeman &: Miller, 1986), and parallel distributed processing (McClelland & Rumelhart 1986; Rumelhart & McClelland, 1986) are discussed.