Understanding eye movements in skilled readers and novice readers: A behavioral and computational approach

当我们阅读时，感觉我们一次可以看到很多单词，但实际上我们一次只能清楚地看到一两个单词，足以阅读。熟练的阅读者非常擅长快速有效地移动眼睛，看到一个单词，然后转到下一个单词。这个项目是关于读者如何学会进行这些快速有效的眼球运动。世界各地和美国的许多人都在努力成为熟练的读者。这个项目将调查年轻读者如何成为熟练的读者，以及为了成为熟练的读者，一个人必须获得什么知识。这种理解将有助于教师和临床医生制定更好的策略来帮助那些有阅读困难的人。在规划眼球运动时，读者会考虑许多信息来源，包括单词长度或单词中字母组合等视觉属性。一个未被重视的信息来源是语言——读者根据自己的语言经验对单词和短语之间的联系的了解。这个项目将调查熟练读者用来引导眼球运动的语言信息，以及熟练读者和不熟练读者的不同之处。熟练的成年读者和年轻的青少年读者在阅读复杂程度和可预见性（读者预测下一个单词的能力）不同的句子时，会测量他们的眼球运动。然后，一个计算模型将模拟和预测阅读过程中的眼球运动，以更好地了解语言因素和个人因素（如阅读经验）如何影响读者的眼球运动。这项工作将为健康阅读发展的技能和知识提供信息，这对于开发补救措施来帮助那些阅读困难的个人是必要的。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

When we read, it feels like we can see many words at once, but in fact we can only see a word or two at a time clearly enough to read. Skilled readers are very good at quickly and efficiently moving their eyes to see a word and move on to the next. This project is about how readers learn to make these quick and efficient eye movements. Many individuals around the world and in the United States struggle to become skilled readers. This project will investigate how young readers become skilled readers and what knowledge someone must gain in order to become a skilled reader. This understanding will help teachers and clinicians develop better strategies to help individuals who struggle to read.Readers consider many sources of information when planning eye movements, including visual properties like word length or what letters tend to appear together in words. One underappreciated source of information is linguistic—a reader’s knowledge of what words and phrases tend to follow each other, based on their language experience. This project will investigate the linguistic information that skilled readers use to guide eye movements and how skilled and less skilled readers are different. Skilled adult readers and younger adolescent readers will read sentences that vary in complexity and predictability (how well a reader is able to anticipate what words are coming next) while their eye movements are measured. Then, a computational model will model and predict eye movements during reading to better understand how linguistic factors and individual factors, like reading experience, affect how readers move their eyes. This work will be informative of the skills and knowledge underlie healthy reading development, which is necessary to develop remediations to help individuals who struggle to read.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.