Suppr超能文献

整合指代典型事件序列的词汇。

Integrating words that refer to typical sequences of events.

作者信息

Khalkhali Saman, Wammes Jeffrey, McRae Ken

机构信息

Department of Psychology, University of Western Ontario, London, Ontario, Canada.

出版信息

Can J Exp Psychol. 2012 Jun;66(2):106-14. doi: 10.1037/a0027369.

Abstract

The organisation and processing of event concepts in semantic memory is an important issue in language processing and memory research. The present study tested whether pairs of words denoting events that can plausibly occur in sequence (marinate-grill) generate expectancies for a target that denotes a subsequently occurring event (chew). In Experiment 1, two events that tend to occur sequentially primed the third. In Experiment 2, the individual primes (i.e., marinate and grill separately) did not prime their related event targets. Experiments 1 and 2 used a lexical-decision task on the target. Therefore, information from both primes must be integrated to sufficiently activate knowledge of the subsequently occurring target. This is the first study to demonstrate priming among words denoting sequentially occurring events. In Experiment 3, a relatedness decision task, processing of these event triplets was facilitated when the first two event words were presented in a temporally correct order compared with when their order was reversed. These findings are not predicted by spreading activation theory and cannot be simulated by corpus-based models that do not include order-sensitive measures. We interpret the results as evidence for the role of situation models and the use of world knowledge during online language comprehension, even in the absence of sentential contexts.

摘要

语义记忆中事件概念的组织和加工是语言加工和记忆研究中的一个重要问题。本研究测试了表示可能按顺序发生的事件的词对(腌制-烤制)是否会对表示随后发生事件的目标词(咀嚼)产生预期。在实验1中,两个倾向于相继发生的事件启动了第三个事件。在实验2中,单个启动词(即分别为腌制和烤制)并未启动与其相关的事件目标词。实验1和实验2对目标词使用了词汇判断任务。因此,必须整合来自两个启动词的信息,才能充分激活对随后出现的目标词的认知。这是第一项证明表示相继发生事件的词之间存在启动效应的研究。在实验3中,采用了相关性判断任务,与两个事件词顺序颠倒时相比,当两个事件词按时间正确顺序呈现时,对这些事件三元组的加工更容易。扩散激活理论无法预测这些发现,不包括顺序敏感测量的基于语料库的模型也无法模拟这些发现。我们将这些结果解释为情境模型的作用以及在线语言理解过程中世界知识运用的证据,即使在没有句子语境的情况下也是如此。

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