心理语言学
语义相似性
计算机科学
固定(群体遗传学)
语义学(计算机科学)
意义(存在)
自然语言处理
语义记忆
约束(计算机辅助设计)
阅读(过程)
心理学
词(群论)
人工智能
认知心理学
语言学
认知
人口
数学
神经科学
几何学
人口学
程序设计语言
社会学
心理治疗师
哲学
作者
Nan Li,Dongxia Sun,Suiping Wang
标识
DOI:10.1007/s11145-022-10350-x
摘要
In natural reading, the processing of words in fixation is influenced by semantic information obtained through preview (i.e., the semantic preview effect). Previous studies have confirmed that two types of semantic information exhibit the semantic preview effect: semantic association, which is reflected by the semantic relationship between preview words and target words, and semantic integration, which is reflected by the plausibility of preview words in sentences. This study examined whether and how these two types of semantic preview information interact to influence readers' processing of words in the fovea. We referenced high constraint sentences, in which contextual information strongly limits the possible meanings, and the meaning of the target word can be activated before the target word becomes fixated. Thus, the meaning of the target word can be obtained at least as early as when the pretarget word becomes fixated. Therefore, by creating a high constraint context, the reader can obtain the meaning of the upcoming word both through preview and through preactivation within the same preview time window. We tested the preview effect of semantic relatedness when preview words were implausible (Experiment 1) and plausible (Experiment 2). Readers’ eye movements were measured. The results showed that the preview effect (shortened fixation duration) of semantic relatedness appeared only when the preview word was plausible. This finding suggests that readers can use semantic information from different sources within the same preview time window and that message-level representations play an immediate and pivotal role in this process.
科研通智能强力驱动
Strongly Powered by AbleSci AI