COMPREHENSION OF WORDS AND PICTURES

文字和图片的理解

• 批准号: 2247623
• 负责人: MARY C POTTER
• 金额: $ 17.04万
• 依托单位: MASSACHUSETTS INSTITUTE OF TECHNOLOGY
• 依托单位国家: 美国
• 财政年份: 1992
• 资助国家: 美国
• 起止时间: 1992-08-01 至 1997-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2247623
• 关键词: attention; biological information processing; cognition; concept; human subject; perception; psycholinguistics; reading; short term memory; speech; stimulus /response; attention; biological information processing; cognition; concept; human subject; perception; psycholinguistics; reading; short term memory; speech; stimulus /response

A major theoretical question in cognition is how contextual information is brought to bear in identifying spoken and written words and objects. We propose to investigate the interaction of context and stimulus information in perception and immediate memory for words at four sequential points in processing, each reflected in a particular paradigm. These are: (1) an early stage of perception, sensitive to repetition blindness; (2) a somewhat later stage involving selection of one perceived word from several possible word candidates activated by that stimulus, when nonword conversion may occur; (3) a subsequent stage at which, if two words have been perceived at or near the same time and are competing for a single slot in a sentence, attentional priority is allocated to one, as in double-word selection; and (4) a short-term memory phase in which a conceptual representation of a sentence or sentence fragment tunes the meaning of words, leading to primed lexical intrusions in short-term recall. These four stages or levels serve as an initial framework for testing a modular interactive model of word identification. Three questions addressed in the proposed research are the time course of context effects, and in particular the time period over which a stimulus remains malleable and subject to reinterpretation after it has been presented; what contextual factors a given level is sensitive to; and whether more standard on-line tasks, lexical decision and naming, provide converging evidence for the model we are developing. Nineteen studies are proposed, some of which use pictured objects to investigate differences between conceptual and lexical processing. The long-range scientific significance of the work derives from the central role that written and spoken language plays in human cognition. The practical significance of the proposed work lies in the importance of understanding in detail the process of reading and language comprehension, in order to place the teaching of reading and the use of written communication on a rational scientific foundation.

认知中的一个主要理论问题是上下文信息的方式 带来识别口语和书面单词和对象的责备。 我们 建议研究上下文和刺激信息的相互作用 在感知和立即记忆单词在四个顺序点中的单词 处理，每个都反映在特定的范式中。 这些是：（1） 感知的早期阶段，对重复失明敏感； （2）a 在某种程度上，涉及从几个中选择一个感知单词的阶段 当非单词时，可能被该刺激激活的可能单词候选者 可能会发生转换； （3）随后的阶段，如果有两个单词 在同一时间或附近被感知，正在争夺一个插槽 在一句话中，注意力优先级分配给一个，例如双词 选择； （4）一个短期记忆阶段，其中一个概念 句子或句子片段的表示含义 单词，导致短期召回中启动词汇入侵。 这些 四个阶段或级别是测试模块化的初始框架 单词识别的交互模型。 拟议的研究中解决的三个问题是时间课程 上下文效果，尤其是刺激的时期 保持延展性，并在其后被重新诠释 提出；给定级别的哪些上下文因素对；和 是否有更多标准的在线任务，词汇决定和命名，提供 我们正在开发的模型的融合证据。 十九个研究是 提出的，其中一些使用图中的对象研究差异 在概念和词汇处理之间。 工作的远程科学意义来自中央 在人类认知中扮演和口语扮演的角色。 这 拟议工作的实际意义在于重要性 详细了解阅读和语言理解的过程， 为了放置阅读教学和书面的使用 理性科学基础上的沟通。