{"title":"建立中文阅读中的眼动控制模型。","authors":"Yanping Liu, Lili Yu, Erik D Reichle","doi":"10.3758/s13423-024-02570-9","DOIUrl":null,"url":null,"abstract":"<p><p>The Chinese writing system has several features that make it markedly different from the alphabetic systems that have most often been examined in reading research, including the fact that individual words consist of various uniformly sized, box-shaped characters whose boundaries are not clearly demarcated (e.g., by blank spaces). These features raise the question: How do readers of Chinese \"know\" where to move their eyes for the purpose of efficiently segmenting and/or identifying words? To answer this question, we used the E-Z Reader model of eye-movement control in reading to run an 'experiment' involving a series of simulations in which two saccade-targeting assumptions (i.e., directing the eyes towards default targets vs. adjusting saccade length as a function of parafoveal processing difficulty) were factorially manipulated with three word-segmentation heuristics (i.e., ideal-observer knowledge of word boundaries vs. probabilistic guessing vs. familiarity-based segmentation) to examine which combination of assumptions provide the best quantitative account of eye-movement control during the reading of Chinese. Based on these simulations, we conclude the best account is one in which readers use relative differences in the familiarity of groups of parafoveal characters to dynamically adjust the lengths of saccades in a manner that affords efficient word identification. We discuss the broader theoretical implications of these conclusions for models of Chinese reading and for models of reading more generally.</p>","PeriodicalId":20763,"journal":{"name":"Psychonomic Bulletin & Review","volume":null,"pages":null},"PeriodicalIF":3.2000,"publicationDate":"2024-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Towards a model of eye-movement control in Chinese reading.\",\"authors\":\"Yanping Liu, Lili Yu, Erik D Reichle\",\"doi\":\"10.3758/s13423-024-02570-9\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>The Chinese writing system has several features that make it markedly different from the alphabetic systems that have most often been examined in reading research, including the fact that individual words consist of various uniformly sized, box-shaped characters whose boundaries are not clearly demarcated (e.g., by blank spaces). These features raise the question: How do readers of Chinese \\\"know\\\" where to move their eyes for the purpose of efficiently segmenting and/or identifying words? To answer this question, we used the E-Z Reader model of eye-movement control in reading to run an 'experiment' involving a series of simulations in which two saccade-targeting assumptions (i.e., directing the eyes towards default targets vs. adjusting saccade length as a function of parafoveal processing difficulty) were factorially manipulated with three word-segmentation heuristics (i.e., ideal-observer knowledge of word boundaries vs. probabilistic guessing vs. familiarity-based segmentation) to examine which combination of assumptions provide the best quantitative account of eye-movement control during the reading of Chinese. Based on these simulations, we conclude the best account is one in which readers use relative differences in the familiarity of groups of parafoveal characters to dynamically adjust the lengths of saccades in a manner that affords efficient word identification. We discuss the broader theoretical implications of these conclusions for models of Chinese reading and for models of reading more generally.</p>\",\"PeriodicalId\":20763,\"journal\":{\"name\":\"Psychonomic Bulletin & Review\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.2000,\"publicationDate\":\"2024-09-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Psychonomic Bulletin & Review\",\"FirstCategoryId\":\"102\",\"ListUrlMain\":\"https://doi.org/10.3758/s13423-024-02570-9\",\"RegionNum\":3,\"RegionCategory\":\"心理学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"PSYCHOLOGY, EXPERIMENTAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Psychonomic Bulletin & Review","FirstCategoryId":"102","ListUrlMain":"https://doi.org/10.3758/s13423-024-02570-9","RegionNum":3,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PSYCHOLOGY, EXPERIMENTAL","Score":null,"Total":0}
引用次数: 0
摘要
中文书写系统有几个特点,使其与阅读研究中最常考察的字母系统明显不同,其中包括单个词由各种大小一致的方块字组成,其边界没有明确划分(如空白)。这些特点提出了一个问题:中文读者如何 "知道 "自己的视线应向何处移动,从而有效地分割和/或识别词语?为了回答这个问题,我们利用 E-Z Reader 阅读中的眼动控制模型进行了一系列模拟实验,在这些实验中,两种囊状视线瞄准假设(即将视线指向默认目标与根据视网膜旁处理难度调整囊状视线长度)与三种单词分割启发式(即、理想观察者对单词边界的了解 vs. 概率猜测 vs. 基于熟悉度的分词),以考察哪种假设组合能最好地定量解释中文阅读过程中的眼动控制。在这些模拟的基础上,我们得出结论:最佳的解释是读者利用视网膜旁字符组熟悉程度的相对差异来动态调整眼动的长度,从而实现高效的单词识别。我们将讨论这些结论对中文阅读模型和一般阅读模型的广泛理论意义。
Towards a model of eye-movement control in Chinese reading.
The Chinese writing system has several features that make it markedly different from the alphabetic systems that have most often been examined in reading research, including the fact that individual words consist of various uniformly sized, box-shaped characters whose boundaries are not clearly demarcated (e.g., by blank spaces). These features raise the question: How do readers of Chinese "know" where to move their eyes for the purpose of efficiently segmenting and/or identifying words? To answer this question, we used the E-Z Reader model of eye-movement control in reading to run an 'experiment' involving a series of simulations in which two saccade-targeting assumptions (i.e., directing the eyes towards default targets vs. adjusting saccade length as a function of parafoveal processing difficulty) were factorially manipulated with three word-segmentation heuristics (i.e., ideal-observer knowledge of word boundaries vs. probabilistic guessing vs. familiarity-based segmentation) to examine which combination of assumptions provide the best quantitative account of eye-movement control during the reading of Chinese. Based on these simulations, we conclude the best account is one in which readers use relative differences in the familiarity of groups of parafoveal characters to dynamically adjust the lengths of saccades in a manner that affords efficient word identification. We discuss the broader theoretical implications of these conclusions for models of Chinese reading and for models of reading more generally.
期刊介绍:
The journal provides coverage spanning a broad spectrum of topics in all areas of experimental psychology. The journal is primarily dedicated to the publication of theory and review articles and brief reports of outstanding experimental work. Areas of coverage include cognitive psychology broadly construed, including but not limited to action, perception, & attention, language, learning & memory, reasoning & decision making, and social cognition. We welcome submissions that approach these issues from a variety of perspectives such as behavioral measurements, comparative psychology, development, evolutionary psychology, genetics, neuroscience, and quantitative/computational modeling. We particularly encourage integrative research that crosses traditional content and methodological boundaries.