Developing a Deeper Understanding of the Cognitive Processes that Drive Multiple Text Comprehension

Project: Research project

Project Details


Developing a Deeper Understanding of the Cognitive Processes that Drive Multiple Text Comprehension Developing a Deeper Understanding of the Cognitive Processes that Drive Multiple Text Comprehension Developing a Deeper Understanding the Cognitive Processes that Drive Multiple Document Comprehension Cognition and Student Learning. Goal 1: Exploratory Purpose: There has been an increased interest students processing of multiple documents. This focus is in part due to the explosion of information and texts readily available in our modern, technology-driven society. As such, the ability to construct an integrated and coherent representation from information in multiple documents is an essential skill for school and the workforce. Having a better understanding of the processes involved in this complex task will help us to develop the kinds of instruction needed to foster multiple document (MD) comprehension skills. In this Goal 1 Exploration project, our objective is to examine the extent that MD comprehension is supported by intratextual (withing a single text) and intertextual (between-text) support MD comprehension and to asses the factors that affect students ability to generate these intertextual inferences. This study will explore the impact of two types of strategy training. While the current literature on MD comprehension focuses on intervention that promote attending to source (e.g., Britt, Perfetti, & van Dyke, 2000), an alternative approach involves strategy training, such as self explanation training, that directly targets the skills needed to integrate information within a texts (McNamara, 2004). As such, we will investigate the benefits of strategy training for MD comprehension skills and how these outcomes compare to more traditional source evaluation training. We will also investigate effects of student individual differences (prior knowledge, reading skill, writing skill) as potential moderating variables. Sample: The project will consist of three studies, all including high school students. Studies 1 (n = 105) and 2 (n = 97) will occur in a laboratory setting, where as Study 3 will be a small scale study in classroom settings (16 classrooms with expected 20-25 per classroom). Design: All studies involve a manipulation of whether students are in conditions that emphasize source evaluation, self-explanation, or control (of which the nature depends on the study) conditions. Key Measures: All studies will have students write integrative essays based on MDs. Studies 1 and 2 will also involve short answer questions that specifically target intertextual inferences. Study 1 will involve having students produce verbal protocols while reading MDs. Computer-based algorithms will be developed and applied to the analysis of the protocols and essays to derive measures of intra-and intertextual inference, and the research team has extensive experience developing and applying these tools to student constructed responses. Data Analytic Strategy: The analyses will depend on studies and research questions. The studies will involve applications of ANCOVA, HLM, and Growth modeling (which is a form of multi-level modeling). Expected outcomes: This project will lead to: 1) an understanding of how to successfully use strategy training to support MD comprehension and in particular intetextual inferences; 2) an understanding of how individual differences moderate the impact of strategy instruction; 3) an understand of the what is needed to successfully develop a classroom or computer based implementation of strategy training to support MD comprehension. Research Team: The team will be led by PI Danielle McNamara (Arizona State University) and co-PIs Laura Allen (Mississippi State University) and Joe Magliano (Northern Illinois University).
Effective start/end date9/1/188/31/23


  • US Department of Education (DOEd): $1,399,466.00


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.