Moment-by-Moment Processing and Knowledge Revision

• In Event: Reducing the Impact of Misconceptions

Abstract

Readers’ prior knowledge can conflict with concepts in text, which can impede learning. For example, the term fitness conjures up thoughts of physical education, running, and overall health for many students. However, fitness has a specific meaning in biology: the number of offspring an organism has that survive to adulthood. A student who encounters fitness in a biology text may be unaware that it has a different meaning in biology than other contexts, and may not develop a conceptual understanding of fitness that is consistent with its meaning in biology.
One way to help students change or update their conceptual knowledge is to have them read a refutational text. A refutational text is written to explicitly acknowledge an alternative conception, directly refute it, and provide a more satisfactory explanation (Dole, 2000; Hynd, 2001).
However, research on the efficacy of refutational texts to promote conceptual change has been mixed. Although some research has shown positive effects of these texts on retention of science knowledge and inference-level performance, other studies have reported null results. These mixed findings indicate that conceptual change from refutational texts is not a simple or straightforward process. Rather, this learning process likely depends on a number of other factors, and in particular, the cognitive processes readers use while reading.
Researchers in text processing have attempted to measure moment-by-moment processes that occur during reading of refutation text and link processes to knowledge revision. Two primary means of measuring moment-by-moment reading processes include the use of verbal protocols (e.g., think-alouds, typed reports) and measures of attention allocation (e.g., reading time, eye-tracking). Both types of measures provide useful information about moment-by-moment processes but they provide different types of information and have different advantages.
Measures of attention provide strong evidence of bottom-up processes, which is crucial for detecting discrepancies between prior knowledge and text-based information (i.e., co-activation), shown by a slowdown in reading when such a discrepancy occurs. An advantage of measures of attention allocation over verbal protocols is that they can less intrusive or disruptive of normal reading processes. However, an advantage of verbal protocols over measures of attention is that they can provide insights into the types of moment-by-moment processes readers use as they progress through a text. In particular, they can provide insights into how readers attempt to address discrepancies between prior knowledge and text information. The predominant approach to evaluating verbal protocols is to parse verbal reports into clauses, and then categorize the clauses using pre-existing schemes. An alternative approach is to assess (1) the cohesion of students’ text responses, (2) the extent to which the cohesion of students’ moment-by-moment processes vary as a function of text type (e.g., refutational vs. non-refutational) or task instructions (e.g., think-aloud vs. self-explain), and (3) whether the cohesion of students’ moment-by-moment processes predict post-reading performance on a conceptual knowledge test. This approach can be used to identify the specific strategic processes that contribute to knowledge revision once a discrepancy has been identified. This approach will be illustrated using experimental data.

Authors

• Matthew T. McCrudden, Pennsylvania State University

• Laura K. Allen, Arizona State University

• Danielle S. McNamara, Arizona State University