M. Anne Britt

The Role of Epistemic Beliefs in the Comprehension of Multiple Expository Texts: Toward an Integrated Model

In present-day knowledge societies, competent reading involves the integration of information from multiple sources into a coherent, meaningful representation of a topic, issue, or situation. This article reviews research and theory concerning the comprehension of multiple textual resources, focusing especially on linkages recently established between dimensions of epistemic beliefs and multiple-text comprehension. Moreover, a proposed model incorporates epistemic beliefs into a theoretical framework for explaining multiple-text comprehension, specifying how and why different epistemic belief dimensions may be linked to the comprehension and integration of multiple texts. Also discussed is the need for further research concerning mediational mechanisms, causality, and generalizability.

Readers’ use of source information in text comprehension

In two experiments, we examined the role of discrepancy on readers’ text processing of and memory for the sources of brief news reports. Each story included two assertions that were attributed to different sources. We manipulated whether the second assertion was either discrepant or consistent with the first assertion. On the basis of the discrepancy-induced source comprehension (D-ISC) assumption, we predicted that discrepant stories would promote deeper processing and better memory for the sources conveying the messages, as compared to consistent stories. As predicted, readers mentioned more sources in summaries of discrepant stories, recalled more sources, made more fixations, and displayed longer gaze times in source areas when reading discrepant than when reading consistent stories. In Experiment 2, we found enhanced memory for source–content links for discrepant stories even when intersentential connectors were absent, and regardless of the reading goals. Discussion was focused on discrepancies as one mechanism by which readers are prompted to encode source–content links more deeply, as a method of integrating disparate pieces of information into a coherent mental representation of a text.

FACILITATING TEXTUAL INTEGRATION WITH MACRO-STRUCTURE FOCUSING TASKS

Students are often asked to integrate information derived from reading multiple documents into a consistent story or model. Based on models of comprehending individual text, we predict that the structure and accessibility of earlier texts should influence ones ability to integrate a new text with previously learned material. In two experiments, we examined the facilitating effect of macro-structure focusing tasks on the integration of information from two texts. Experiment 1 found that having participants summarize a text before reading a subsequent text enhanced integration. Experiment 2 found that both answering macro-level questions and reading instructions to integrate resulted in a more integrated final representation.

Scientific Literacy: The Role of Goal-Directed Reading and Evaluation in Understanding Scientific Information

In this article, we examine the mental processes and representations that are required of laypersons when learning about science issues from texts. We begin by defining scientific literacy as the ability to understand and critically evaluate scientific content in order to achieve ones goals. We then present 3 challenges of learning from science texts: the intrinsic complexity of science phenomena, the need to coordinate multiple documents of various types, and the rhetorical structure of the texts themselves. Because scientific information focuses on models, theories, explanations, and evidence, we focus on how explanatory and argumentative texts are processed. Then we examine 2 components of executive control in reading—goal-directed guidance and evaluation of content—that readers can acquire and adopt to deal with these challenges. Finally, we discuss 3 implications that these theories and empirical findings have for interventions intended to improve laypersons’ understanding of scientific information.

Do students’ beliefs about knowledge and knowing predict their judgement of texts’ trustworthiness?

We examined whether epistemic beliefs predict students’ evaluation of documents. Undergraduates read two texts on climate change. Participants judged the trustworthiness of each text and then indicated the criteria for their rating. We found that readers who believe strongly in relying on personal interpretations rather than on authorities trusted both documents less and used the document’s content or their own opinion as criteria for judging trustworthiness. We also found that readers who believe that knowledge claims should be critically evaluated through logic and rules rated the science text as more trustworthy and used the criteria of their own opinion, author and content more than readers who believe in relying on their own experiences. These effects hold true while controlling for readers’ prior knowledge and text comprehensibility.

Investigating Instruction for Improving Revision of Argumentative Essays

Students are expected to come into the current college classroom already possessing certain skills including the ability to write at the appropriate academic level regardless of discipline and the ability to create well-structured arguments. Research indicates, however, that most students entering college are underprepared in both areas. One strategy that may help students write at a more academic level is teaching students to focus on spending their time on revision. In the current study, we examine two potential sources of difficulty in the revision of argumentative essays: a poorly developed argument schema and a poorly developed global revision task schema. We created and tested the effectiveness of two written tutorials designed to provide college students information to saturate their knowledge base as well as provide them with procedural tasks to complete. We found that without instruction, students focused their revisions on making local wording changes that did not qualitatively improve their essays. An argument tutorial helped students make higher level global changes, include more argument content, and improve the structure of the essay. A global revision tutorial also helped students make more substantive structural changes. Thus, both tutorials helped students improve their revisions, and the tutorials were completed independently by the students successfully.

Constructing representations of arguments

Three experiments were conducted to test whether presentation order affects the reading and later recall of simple two-clause arguments. Participants read arguments in a claim-first order or in a reason-first order. Three experiments found that arguments were read faster when claims preceded reasons and this effect was independent of whether the reason began with a subordinating conjunction. Shorter reading times were observed for claims when they occurred in the initial position. Claims were also recalled better than reasons and claim-first arguments were recalled more accurately than reason-first arguments. Experiments 3a and 3b showed that readers identified claims by the presence of markers (e.g., modals and qualifiers) and that arguments with modals are read more quickly and recalled better in a claim-first order. These results suggest that readers use a claim-centered argument schema to guide the processing of persuasive prose.

The Effects of Hedges in Persuasive Arguments A Nuanced Analysis of Language

Drawing together research on persuasion and text comprehension, two experiments test the effects of hedge placement (Experiment 1) and hedge type (Experiment 2) on attitudes, source evaluations, and perceptions of argument strength. Participants read an editorial in support of implementing comprehensive exams at their university. Experiment 1 shows that hedges placed on data statements (and not interpretation statements) lead to negative perceptions of the policy, source, and argument. This is especially pronounced on source evaluations among individuals with more scientific training. Experiment 2 reveals that colloquial, but not professional, hedges placed on interpretation statements lead to more negative evaluations relative to no hedges. Data related to perceptions of the source are moderated by individual differences in scientific reasoning. This research suggests that hedges describing data statements and/or that use colloquial language can, but do not always, undermine persuasive attempts.

The efficacy of a Web-based counterargument tutor.

In two experiments, we developed and tested an interactive Web-based tutor to help students identify and evaluate counterarguments. In Experiment 1, we determined the extent to which high- and low-argumentationability participants were able to identify counterarguments. We tested the effectiveness of having participants read didactic text regarding counterarguments and highlight claims. Both preparations had some positive effects that were often limited to high-ability participants. The Web-based intervention included interactive exercises on identifying and using counterarguments. Web-based presentation was state driven, using a Java Server Pages page. As participants progressively identified argument elements, the page changed display state and presented feedback by checking what the user clicked against elements that we had coded in XML beforehand. Instructions and feedback strings were indexed by state, so that changing state selected new text to display. In Experiment 2, the tutor was effective in teaching participants to identify counterarguments, recognize responses, and determine whether counterarguments were rebutted, dismissed, or conceded.

Writing to Learn from Multiple-Source Inquiry Activities in History

Inquiry writing activities seek to engage students in a similar context by asking them to use multiple documents, like historians do, to reason about how different peoples motives and actions; the cultural context including social, political and economic factors; and the occurrence of other events, all interact to make History happen. As much of the attention in learning from multiple documents in History has focused on improving sourcing skills, this chapter focuses on the other two main skills underlying learning from document-based inquiry activities: contextualization and corroboration/integration. It prompts chronological thinking using timeline activities, prompt integration using different writing prompts, and examines the effects that both of these manipulations have on student learning about History from document-based writing activities. The chapter also moves the investigation of learning from multiple-document inquiry activities using the Panama Unit into even younger student populations. Keywords: chronological thinking; contextualization; corroboration; cultural context; integration; multiple-document inquiry activities; Panama Unit