The Role of Task Models and Evaluation Processes in Comprehending Multiple Texts

• In Event: 49.029 - Competing and Complementary Models of Multiple Source Use: Theoretical and Empirical Advancements

Abstract

Purpose

The goal of this presentation is to examine some empirical predictions drawn from the Multiple Documents – Task-Based Relevance Assessment and Content Extraction (MD-TRACE, Rouet & Britt, 2011) model, a framework aimed at describing the processes and resources involved in the comprehension of multiple documents.

Theoretical Frame

In the first part of the presentation, we introduce some of the core features of the MD-TRACE framework, and we articulate key assumptions that arise from the framework regarding how lay readers handle multiple text comprehension situations and potential sources of difficulties. MD-TRACE assumes that readers form a task model based on their interpretation of explicit task instructions as well as other, more implicit contextual and pragmatic cues. The (subjective) task model then drives a series of processes whereby readers scan, evaluate and select texts and text passages to be read; comprehend, evaluate and integrate the informational contents of the texts; and form and update a so-called "task product." In addition, MD-TRACE includes several regulation mechanisms whereby readers may reconsider their selection of information resources or even their task model as a function of the outcomes.

Data Sources and Results

Based on recent studies conducted with secondary and higher education students, we provide evidence for the link between the quality of readers' task model and the evaluative processes involved in multiple text comprehension. We report evidence that students vary in their interpretation of reading task demands as they apply to complex multi-text environments such as the Web. We also found evidence that task model variations influence students' selection of texts, their actual engagement with the texts, and their integration of text information into a task product.

Scholarly Contributions

We conclude with a discussion of the ways in which the MD-TRACE framework advances our understanding of purposeful reading of multiple texts. We contend that MD-TRACE elicits evaluation and integration processes not accounted for by more traditional reading comprehension frameworks. Finally, we consider some avenues for future research into multiple text comprehension.

Authors

• Jean-Francois Rouet, CNRS and University of Poitiers

• Anne Britt, Northern Illinois University