Dissertation Research: Statistical Models of Hypertext Comprehension

超文本被认为是造成阅读理解困难的原因。然而，在一定条件下，具有适当语义组织的超文本有助于理解和记忆。为了解决这些问题，本研究将涉及来自两组的超文本读者：(i)神经科学新手心理学本科专家，（ii）心理学新手神经科学本科专家。读者将遍历与三个不同内容域相关联的三个网站，以便在这些网站上生成内容域的摘要。此外，一些网站将以线性顺序格式向读者呈现材料，而另一些网站将以超文本格式向读者呈现材料。一半的超文本格式将是“完全连接的”（即，读者将能够从网站的任何一点遍历到任何其他点），而另一半的超文本格式将是“语义组织的”（即，读者将被允许做出自己的遍历决策，但他们的决策选择将仅限于在语义相关的网页之间遍历）。因此，相对于线性文本环境和网格-超文本环境，专家们期望语义理论驱动的超文本环境能够提高导航和记忆性能。对于不熟悉说明文语义结构的新手来说，在线性文本环境中导航和回忆信息应该是最有效的。本研究将尝试探讨先前定义的领域知识和内容呈现因素如何影响网站浏览策略和网站摘要的组织。经典数据分析将用于识别这些定性现象。此外，知识有向图贡献分析（一种以理论为导向的统计鲁棒分类回归时间序列分析）将用于导航模式和学生反应的顺序结构，以便根据知识组织的特定语义网络理论对数据进行更详细的定量理解。该项目的长期目标是支持我们对超文本环境中人类理解的科学理解的进步。提高对超文本理解的理解不仅会促进对人类认知的一般理解，而且会促进对设计更容易理解的网站的条件的理解。这个项目将为认知心理学家、网页设计师和人机交互专家提供新的方法来观察内容呈现和领域知识对导航模式和生产数据的影响。作为博士论文研究改进奖，该奖项还将提供支持，使有前途的学生建立一个强大的独立研究生涯。

Hypertext has been attributed to reading comprehension difficulties. Yet under certain conditions, hypertexts with the appropriate semantic organization have been found to facilitate comprehension and recall. To address these issues, this study will involve hypertext readers from two groups: (i) undergraduate experts in Psychology who are Neuroscience novices, and (ii) undergraduate experts in Neuroscience who are Psychology novices. The readers will traverse three web sites associated with three different content domains for the purposes of generating summaries of the content domain on those websites. In addition, some web sites will present material to readers in a linear sequential format while other web sites will present material to readers in a hypertext format. Half of the hypertext formats will be "fully connected" (i.e., a reader will be able to traverse from any point in the web site to any other point), while the other half of the hypertext formats will be "semantically organized" (i.e., readers will be allowed to make their own traversal decisions but their decision choices will be restricted to traversing between semantically related web pages). Thus, improved navigation and memory performance for the semantic theory driven hypertext environment relative to the linear text environment and meshed-hypertext environment is expected for experts. Novices, unfamiliar with semantic structure of the expository hypertexts, should be most effective at navigating and recalling information presented within a linear text environment. This study will attempt to investigate how the previously defined factors of domain knowledge and content presentation influence web site traversal strategies and the organization of web site summaries. Classical data analysis will be used to identify these qualitative phenomena. In addition, Knowledge Digraph Contribution analysis (a theoretically-oriented statistically-robust categorical regression time-series analysis) of sequential structure in navigation patterns and student responses will be used to obtain a more detailed quantitative understanding of the data in terms of specific semantic network theories of knowledge organization. The long-term objective of this project is to support advancements in our scientific understanding of human comprehension in the hypertext environment. An improved understanding of hypertext comprehension will not only advance general understanding of human cognition but also will advance understanding of the conditions for designing web sites which are easier to understand. This project will provide cognitive psychologists, web designers, and Human-Computer Interaction specialists with new ways of observing effects of content presentation and domain knowledge upon navigation patterns and production data. As a Doctoral Dissertation Research Improvement award, this award also will provide support to enable a promising student to establish a strong independent research career.