In this project we investigated whether and when students can make accurate judgements of their own understanding of scientific texts.
This funded project ran from 2004-7.
We are following up this work with a new project on "Improving Metacomprehension and Self-regulated Learning from Scientific Texts" which is funded from 2007-11.

Principal Investigators:
Keith Thiede, Educational Psychology, UIC (Now at Boise State)
Jennifer Wiley, Psychology, UIC

Project Contributors:
John Dunlosky, Psychology, Kent State University
Thomas Griffin, Post-Doctoral Fellow, Psychology, UIC (Now Assistant Research Professor at UIC, and co-PI on new project).

Graduate Students: Travis Ricks and Ben Jee are interested in the effects of domain knowledge on cognitive processing have both assisted on projects related to this grant. UIC undergraduate Robert Ariel also assisted on this grant.  He started the PhD Program in Cognitive Psychology at Kent State in Fall 2007.

Overview:
Reading is a basic skill that many consider a fundamental building block to all other learning.  Thus, it’s not surprising that improving literacy levels in America has been a cornerstone of the educational policy for the last two Presidential administrations.  Furthermore, states are universally including standards or benchmarks to guarantee that students attain at least a minimal level of reading proficiency. Reading proficiency includes both competent decoding skills as well as comprehension skills.  There have recently been strong advances made in the very important goal of promoting basic reading instruction with evidence from empirical studies. However, comprehension skill is also critical to the ability to learn from text.  This research program builds on previous work by Thiede and his colleagues (Rawson, Dunlosky, Thiede, 2000; Thiede, 1999; Thiede & Anderson, 2003; Thiede, Anderson, & Therriault, 2003) and empirically tests methods of improving reading comprehension by improving self-regulated learning.

Many models of self-regulated learning describe learning as an interaction between metacognitive monitoring and control -- also called regulation of study (e.g., Nelson & Narens, 1990; Thiede & Dunlosky, 1999; Winne & Butler, 1997).  For instance, consider a student preparing for an upcoming exam.  As the student studies, she monitors her progress toward the goal of mastering the material.  If her monitoring indicates that she has not yet mastered the material, she will likely restudy the material until her monitoring suggests that the material has been mastered, at which time she will stop studying [see Son & Metcalfe (2000) and Thiede & Dunlosky (1999) for situations in which people do not restudy under-learned material (e.g., when under strict time constraints)]. Thus, accurate metacognitive monitoring is critical to effective self-regulation of study behaviors (Winne & Perry, 2000).  If a person is not able to accurately differentiate well-learned material from less-learned material, he or she could waste time studying material that is already well learned or worse fail to restudy material that has not yet been adequately learned.  Therefore, it is important to find ways to improve the accuracy of metacognitive monitoring.

Our current work is based on findings of Thiede and his colleagues who have found that re-reading, summarizing a text, or generating keywords can lead to more accurate metacomprehension of expository text, especially when these activities are done at a delay following a first reading.  In work done under this project, we explored the mechanisms that may be leading to these improvements, and other new ways of improving the accuracy of comprehension monitoring during reading. We have explored the construct of metacomprehension and how it differs from metamemory, as learning from expository science text requires readers to monitor their understanding of text, which may be very different than monitoring their memory for text (Wiley, Griffin & Thiede, 2005).  We are interested in which interventions may specifically improve readers' accuracy in their metacomprehension judgements, and improve their ability to judge whether they have understood the scientific phenomena described by a text.

These successful techniques that can improve students metacomprehension accuracy are:

1. Generating Keywords of Texts after a Delay.
In a series of studies, we found that the delay between reading a text and generating keywords is critical (and showed that other lags are not important players in the "delayed keyword" effect (Thiede, Dunlosky, Griffin & Wiley, 2005).

2. Generating Summaries of Texts after a Delay.
In a similar line of research, we found that having students generate a summary for texts after a delay between reading a text and generating the summary, metacomprehension accuracy is also improved . (Thiede & Anderson, 2003; Thiede, Anderson & Therriault, 2003)

3. Encouraging students to Re-read and Self-explain Texts.
Re-reading can improve metacomprehension accuracy under some conditions, but self-explanation tasks are a more robust intervention (Griffin, Wiley & Thiede, in press). We are also interested in individual differences and how these may interact with interventions that improve metacomprehension accuracy.  In particular, we have found that both comprehension skill and working memory capacity can be related to metacomprehension accuracy (Griffin, Wiley & Thiede, in press).  We have also found that topic knowledge can improve the absolute accuracy of comprehension judgments (Jee, Wiley & Griffin, 2006).

4. Concept Mapping while Reading.
A series of studies with low ability readers has shown that they may have particular difficulty with the delayed/generation interventions that have been shown to be effective with high ability readers.  On the other hand, we have found that teaching students to create concept maps of expository texts improves their metacomprehension accuracy (Thiede, Anderson, Griffin & Wiley, under review).

We interpret these results with a cue-utilization framework that assumes that readers need access to valid cues in order to accurately judge their own level of comprehension.  In this case, where tests of comprehension are based in the ability to make inferences and draw connections, the valid cues are those that correspond to a situation model-level representation of the text.  We view these interventions as giving readers better access to such cues, either by focussing their attention on the situation model level through explanation or concept mapping tasks, or via delayed tasks that may allow surface cues to fade, and thus situation model cues to become more salient.  As a result, students are better able to accurately predict their comprehension of the text (Wiley, Thiede, & Griffin, 2007).  In our next grant, we are continuing this theme by exploring how to better teach students what it means to "understand" expository text, and how valid expectations can be instilled via reading instructions, feedback, and practice tests on different topics. We are also extending these interventions into middle school reader samples, as the work until now has investigated effects on college reader samples.

Papers from the project:

Thiede, K. W., Griffin, T. D., Wiley, J., & Redford, J. (forthcoming, expected 2008) Metacognitive Monitoring During and After Reading. To appear in D. J. Hacker, J. Dunlosky & A. C. Graesser (Eds). Handbook of Metacognition in Education. Routledge.

Griffin, T. D., Wiley, J. & Thiede, K. W. (in press) Individual Differences, Rereading, and Self-Explanation: Concurrent Processing and Cue Validity as Constraints on Metacomprehension Accuracy.  Memory & Cognition.

Wiley, J., Thiede, K. W. & Griffin, T. D. (2007). What does it mean to learn from and understand science text? Paper presented at the 2007 Annual Meeting of the American Educational Research Association, Chicago, IL.

Jee, B., Wiley, J & Griffin, T. D. (2006). Expertise and the illusion of comprehension.  Proceedings of the Annual Conference of the Cogntive Science Society.

Thiede, K. W., Dunlosky, J., Griffin, T. D., & Wiley, J. (2005). Understanding the Delayed-Keyword Effect on Metacomprehension Accuracy.  Journal of Experimental Psychology: Learning, Memory, and Cognition. Vol 31(6), 1267-1280.

Wiley, J., Griffin, T. D., & Thiede, K. W. (2005). Putting the comprehension in metacomprehension.  Journal of General Psychology. Vol 132(4), 408-428.

Dunlosky, J. D., Hertzog, C., Kennedy, M. R. F & Thiede, K. W. (2005). The Self-Monitoring Approach For Effective Learning. International Journal of Cognitive Technology. Vol 10(1), 4-11.

Dunlosky, J. & Thiede, K. W. (2004). Causes and constraints of the shift-to-easier-materials effect in the control of study. Memory & Cognition. Vol 32(5), 779-788.

Relevant Previous Work by Thiede and colleagues:

Thiede, K. W. & Anderson, M. C. M. (2003) Summarizing can improve metacomprehension accuracy. Contemporary Educational Psychology. Vol 28(2), 129-160.

Thiede, K. W; Anderson, M. C. M, &  Therriault, D. (2003) Accuracy of metacognitive monitoring affects learning of texts.  Journal of Educational Psychology. Vol 95(1), 66-73.

Rawson, K. A., Dunlosky, J., &  Thiede, K. W. (2000)  The rereading effect: Metacomprehension accuracy improves across reading trials. Memory & Cognition. Vol 28(6), 1004-1010.

This research has been generously funded by Grant R305H030170 to Keith Thiede and Jennifer Wiley from the Cognition and Student Learning Program of the Institute of Education Sciences.