Suppr超能文献

对大量日常物品进行自由分类时,更多的是基于主题而非分类学。

Free classification of large sets of everyday objects is more thematic than taxonomic.

作者信息

Lawson Rebecca, Chang Franklin, Wills Andy J

机构信息

Department of Experimental Psychology, University of Liverpool, UK.

Department of Experimental Psychology, University of Liverpool, UK.

出版信息

Acta Psychol (Amst). 2017 Jan;172:26-40. doi: 10.1016/j.actpsy.2016.11.001. Epub 2016 Nov 15.

Abstract

Traditionally it has been thought that the overall organisation of categories in the brain is taxonomic. To examine this assumption, we had adults sort 140-150 diverse, familiar objects from different basic-level categories. Almost all the participants (80/81) sorted the objects more thematically than taxonomically. Sorting was only weakly modulated by taxonomic priming, and people still produced many thematically structured clusters when explicitly instructed to sort taxonomically. The first clusters that people produced were rated as having equal taxonomic and thematic structure. However, later clusters were rated as being increasingly thematically organised. A minority of items were consistently clustered taxonomically, but the overall dominance of thematically structured clusters suggests that people know more thematic than taxonomic relations among everyday objects. A final study showed that the semantic relations used to sort a given item in the initial studies predicted the proportion of thematic to taxonomic word associates generated to that item. However, unlike the results of the sorting task, most of these single word associates were related taxonomically. This latter difference between the results of large-scale, free sorting tasks versus single word association tasks suggests that thematic relations may be more numerous, but weaker, than taxonomic associations in our stored conceptual network. Novel statistical and numerical methods for objectively measuring sorting consistency were developed during the course of this investigation, and have been made publicly available.

摘要

传统观点认为,大脑中类别总体组织是分类学式的。为检验这一假设,我们让成年人对140 - 150个来自不同基本层次类别的多样、熟悉的物体进行分类。几乎所有参与者(80/81)对物体的分类更多是基于主题而非分类学。分类仅受到分类学启动的微弱调节,并且当明确被要求按分类学进行分类时,人们仍然产生了许多基于主题结构的聚类。人们产生的第一批聚类在分类学和主题结构方面被评为相当。然而,后来的聚类在主题组织方面的评分越来越高。少数物品始终按分类学聚类,但基于主题结构的聚类总体占主导地位表明,人们对日常物体之间主题关系的了解多于分类学关系。最后一项研究表明,在最初研究中用于对给定物品进行分类的语义关系预测了与该物品相关的主题词与分类学词联想的比例。然而,与分类任务的结果不同,这些单个词联想大多是按分类学相关的。大规模自由分类任务与单个词联想任务结果之间的后一种差异表明,在我们存储的概念网络中,主题关系可能比分类学关联更多,但更弱。在这项研究过程中开发了用于客观测量分类一致性的新颖统计和数值方法,并已公开提供。

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