Concept Mapping Improves Metacomprehension Accuracy Among 7<sup>th</sup> Graders
نویسندگان
چکیده
Two experiments explored concept map construction as a useful intervention to improve metacomprehension accuracy among 7 grade students. In the first experiment, metacomprehension was marginally better for a concept mapping group than for a rereading group. In the second experiment, metacomprehension accuracy was significantly greater for a concept mapping group than for a control group—a group of students who were given already constructed concept maps had accuracy between these two groups. In both experiments, control groups had poor metacomprehension accuracy. That is, they performed worse on tests they predicted better performance and performed better on tests they predicted worse performance. Although constructing concept maps did not produce the same high level of accurate monitoring previously reported in the literature, it still reduced the illusion of knowing. Models of self-regulated learning suggest that learners monitor their progress toward a goal and use this information to regulate their study (e.g., Ariel, Dunlosky, & Bailey, 2009; Metcalfe & Kornell, 2003; Nelson & Narens, 1990; Thiede & Dunlosky, 1999; Winne & Hadwin, 1998). Accurate monitoring identifies which information is welllearned and which information requires additional study. The influence of accurate monitoring on learning has been empirically supported across a variety of domains (for a review, see Dunlosky, Hertzog, Kennedy, & Thiede, 2005). More relevant to the present research, accurate monitoring and effective regulation of study has lead to better comprehension of texts (Thiede, Anderson, & Therriault, 2003). Therefore, it is important to find ways to improve metacomprehension accuracy (the accuracy with which one monitors his or her own comprehension of text). To understand how one might improve metacomprehension accuracy, it is important to combine theories of metacognitive monitoring and comprehension (Rawson, Dunlosky, & Thiede, 2000; Weaver, 1990; Wiley, Griffin & Thiede, 2005). The cue-utilization model of metacognitive monitoring (Koriat, 1997) and the construction-integration model of comprehension (Kintsch, 1994, 1998) provide a framework for understanding techniques for improving metacomprehension accuracy (Rawson et al., 2000). Consider the processes involved in judging one’s comprehension of texts. After reading, a person is asked to judge his or her comprehension of a text. According to the cue-utilization framework of metacognitive monitoring (Koriat, 1997), the metacomprehension judgment may be based on a number of cues, such as how easily the text was processed during reading (Dunlosky & Rawson, 2005; Rawson & Dunlosky, 2002), how successfully the material was retrieved at the time of the judgment (Baker & Dunlosky, 2006; Benjamin, Bjork, & Schwartz, 1998; Morris, 1990), the familiarity with the domain of the text (Glenberg & Epstein, 1987; Glenberg, Sanocki, Epstein & Morris, 1987; Maki & Serra, 1992), or global characteristics of texts such as length or difficulty (Weaver & Bryant, 1995). Metacomprehension accuracy will tend to increase as the cues that are used to make a judgment more highly correlate with performance on a test of comprehension (for empirical evidence linking metacomprehension accuracy and cue use see Thiede, Griffin, Wiley, & Anderson, 2010). The construction-integration model (Kintsch, 1994, 1998) suggests different cues that may be available for metacomprehension judgments. According to this model, a reader creates multiple representations of a text as he or she reads. For instance, the reader constructs a representation of the surface level (i.e., the exact words), a textbase level (i.e., the meaning of sentences), and the situation-model level (i.e., connections between ideas contained in the 2 NOTICE: This is the author’s version of a work accepted for publication by Elsevier. Changes resulting from the publishing process, including peer review, editing, corrections, structural formatting and other quality control mechanisms, may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. The definitive version has been published in Learning and Instruction, Volume 22, Issue 4, 2012. DOI: 10.1016/j.learninstruc.2011.10.007 text, and the connection between these ideas and prior knowledge). A well-constructed situation model integrates the ideas contained in a text and allows the reader to form a causal model and inferences implied by the text. When tests of comprehension assess the quality of the situation model of a text (Kintsch, 1994; McNamara, Kintsch, Songer, & Kintsch, 1996), metacomprehension accuracy should increase if readers use cues based on their situation model to judge their comprehension (for a detailed discussion of designing texts and tests for assessing the situation model see Wiley et al., 2005). In fact, many of the techniques shown to improve metacomprehension accuracy arguably focus readers on their situation model while judging comprehension. We next provide a brief review of the metacomprehension literature. Initial metacomprehension work began in the 1980s (e.g., Glenberg & Epstien, 1985; Maki & Berry, 1984). Glenberg and Epstein (1985) developed a paradigm in which participants read 16 texts and rate their level of comprehension for each. Participants then completed a comprehension test for each text. Metacomprehension accuracy was operationalized as the intra-individual correlation between a participant’s metacomprehension rating and his or her test performance computed across the texts. Although poor metacomprehension accuracy reported in early studies was in part due to poor measures of comprehension (Weaver, 1990), much of the work in this area suggested college students are not adept at monitoring their own comprehension. Maki (1998) found that the mean correlation between predicted and actual test performance (metacomprehension accuracy) across the studies conducted in her laboratory was just .27. Dunlosky and Lipko (2007) reviewed the metacomprehension literature and found the mean accuracy across all the reviewed studies was no better. Recently, interventions have been developed to improve metacomprehension accuracy. For instance, Thiede and Anderson (2003) had college students write summaries of texts after reading but prior to judging comprehension. When summaries were written immediately after reading, the summaries contained more details from the text, which may have focused participants on the surfaces features of a text when judging comprehension, and metacomprehension accuracy did not improve. By contrast, when summaries were written after a delay encouraged readers to focus on the situation model because the surface features decays over time, and metacomprehension accuracy improved (see also Anderson & Thiede, 2008). Generating a list of five keywords that captured the essence of a text, instead of writing a summary, produced a similar boost in metacomprehension accuracy (Thiede et al., 2003). Again the timing of keyword generation was critical to improving metacomprehension accuracy. Thiede, Dunlosky, Griffin, and Wiley (2005) showed that the delay between reading and generating keywords was critical to improving metacomprehension accuracy. Again with an eye toward focusing participants on the situation model of a text when judging comprehension, Griffin, Wiley, and Thiede (2008) had participants self-explain the connects between the ideas in a text as they read. Self-explaining, which facilitates construction of a situation model (Chi, DeLeeuw, Chiu, & Lavancher, 1994), increased the salience of cues related to the situation model at the time of judging comprehension, and improved metacomprehension accuracy. Thiede, Griffin, Wiley, and Anderson (2010) had college students create concept maps of texts prior to judging comprehension. This too was hypothesized to increase the salience of cues related to the situation model at the time of judging comprehension, and improved metacomprehension accuracy. Although a delay was necessary before keyword generation (i.e., allow degradation of surface information and ensure that participants focused on their situation models), concept map construction could occur immediately. Concept map construction is an intervention specifically targeting the situation model. Thiede, Wiley and Griffin (2011) showed that providing instructions about what kind of test to expect and giving practice tests focused college students on appropriate cues and produced more accurate monitoring (see also Thomas & McDaniel, 2007). Two additional advantages arise from assessing the validity of concept map construction as an intervention with a youner population. First, as keyword generation requires a delay to be effective, metacomprehension accuracy is arguably due to transfer-appropriate-monitoring. Specifically, delayed keyword generation improved metacomprehension accuracy because the participants is employing the same long-term memory retrieval process during the rating and test phase. Finding an improvement in metacomprehension accuracy with concept map construction provides additional support that the critical contributing factor to metacomprehension accuracy is whether the situation model of the text is being encoded. A second benefit is that concept map construction is more practical than delay-based interventions. Teachers would have an easier time integrating concept map construction into their standard pedagogical practice than some practice that requires students to pause for several minutes prior to engaging in the next phase of text processing. For a review of the metacomprehension literature and theory related to improving 3 NOTICE: This is the author’s version of a work accepted for publication by Elsevier. Changes resulting from the publishing process, including peer review, editing, corrections, structural formatting and other quality control mechanisms, may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. The definitive version has been published in Learning and Instruction, Volume 22, Issue 4, 2012. DOI: 10.1016/j.learninstruc.2011.10.007 metacomprehension accuracy by focusing readers on their situation model see Thiede, Griffin, Wiley, and Redford (2009). Many of the techniques that have been shown to improve metacomprehension accuracy arguably focus readers on their situation model while judging comprehension. One such technique, which was used in the present investigations, is instructing participants to construct concept maps of texts prior to judging comprehension. As a concept map is a graphic representation of the underlying structure of the text, the act of constructing a concept map helps readers form connections among concepts in a text (Weinstein & Mayer, 1986). Put differently, concept mapping can help readers form an external representation of a situation model for a text, which can be particularly helpful for less-able readers (Stensvold & Wilson, 1990; for a review and meta-analysis on effectiveness of concept maps with low-ability learners, see Nesbit & Adesope, 2006). Moreover, constructing concept maps should increase the salience of cues related to the situation model when judging comprehension. Thiede et al. (2010) showed that constructing concept maps improved metacomprehension accuracy with college students enrolled in remedial reading courses, but this technique has not been evaluated with younger readers. We know very little about younger readers’ ability to monitor comprehension. To date, we know of only one other study examining metacomprehension with younger readers. de Bruin, Thiede, Camp and Redford (2011) showed the generating keywords improved metacomprehension accuracy for 6, and 7 graders, but had no effect on accuracy for 4 graders. Given these results, one might expect concept mapping to improve metacomprehension accuracy. Nonetheless, it is important to evaluate whether concept mapping improves metacomprehension accuracy for adolescents. Although concept mapping has been shown to improve metacomprehension accuracy for some college students, there are some critical differences between college students and 7 graders. First, college students have relatively more experience reading expository texts; whereas, younger students are only just beginning to learn how to learn from expository texts. Reading instruction in early grades is generally confined to the understanding of narrative texts, even though students are expected to eventually read for understanding from informational expository texts within particular subject-matter areas. Students have much less familiarity with the structures and types of nonnarrative texts (Duke, 2000; Pearson & Duke, 2002; Venezsky, 2000). Thus, younger students may not be sensitive to the unique demands of constructing a causal situation model from informational expository texts, which in turn may affect their access to and selection of cues that they use to judge their comprehension. Second, even if lack of experience with expository texts does not affect metacomprehension accuracy, younger students may lack the cognitive skills or capacity required to accurately monitor learning while reading complex expository texts. Studies have shown that children are less sophisticated in related cognitive domains, such as memory regulation (e.g., Roderer & Roebers, 2009), uncertainty confidence monitoring (e.g., Roebers, von der Linden, & Howie, 2007), judgments of learning (e.g., Roebers, von der Linden, Schneider & Howie, 2007), comprehension (Markman, 1977, 1979), working memory (e.g., Wollman, Eylon, & Lawson, 1979), and attention (Pasual-Leone, 1994). Therefore, younger students may not possess the capacity required to monitor comprehension during reading. If less experience with expository texts inhibit 7 graders’ ability to attend to appropriate cues for judging comprehension or 7 graders lack the metacognitive sophistication to use cues to accurately assess their own comprehension, concept mapping will not affect metacomprehension accuracy. By contrast, if the improved metacomprehension accuracy for 7 graders who generated keywords (de Bruin et al., 2011) is evidence that younger readers can monitor their comprehension of expository texts, then concept mapping should improve metacomprehension accuracy. We expect concept mapping to improve metacomprehension accuracy for two reasons (Hypothesis 1). First, concept mapping eases the working memory requirements of text comprehension. Instead of maintaining earlier content as they try to understand later portions of a text, they are able to see the text’s content unfold visually as they construct they construct their maps. This enables more resources to be devoted to gauging their level of text comprehension when they revisit their maps rather than having resources divided between metacognitive monitoring and comprehending the text (Rawson et al., 2000). The research cited above that indicates the various limitations observed in childrens’ cognition is countered by research that shows the effectiveness of interventions to strengthen various cognitive domains such as problem solving (Jitendra, Star, 4 NOTICE: This is the author’s version of a work accepted for publication by Elsevier. Changes resulting from the publishing process, including peer review, editing, corrections, structural formatting and other quality control mechanisms, may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. The definitive version has been published in Learning and Instruction, Volume 22, Issue 4, 2012. DOI: 10.1016/j.learninstruc.2011.10.007 Rodriguez, Lindell, & Someki, 2011), reading comprehension (Sporer, Brunstein, & Kieschke, 2009), and working memory (Lee, Lu, & Ko, 2007). Second, we expect concept mapping to improve metacomprehension accuracy as it increases the salience of cues related to the situation model, which should improve metacomprehension accuracy. Experiment 1 In the first experiment, we asked students to construct concept maps as they read a set of three texts. We expected that concept map construction would encourage situation model development as well as ease the cognitive demands required to monitor comprehension. If concept maps allow participants to access and select cues based in their situation models for judging comprehension, then constructing concept maps should improve metacomprehension accuracy. Method Participants and Design Fifty-nine 7 grade students (ages 12 – 13) from a local junior high school participated in this experiment. Of the 59, 33 were female and 26 were male; 45 were Caucasian, 14 were Hispanic or African-American. The school serves a low SES population, with 70% of students eligible for free or reduced lunch. All participants were treated in accord with APA ethical standards. Group assignment (concept mapping versus rereading/control) was done randomly between participants. We used a reread condition in order to equate the amount of time participants in the two groups worked with the texts. Of the 59 participants, thirty-eight participants were in the concept map construction condition and 21 participants were in the reread condition. We over assigned participants to the concept map construction group to increase the statistical power in examining the relation between the characteristics (content) of concept maps and metacomprehension judgment, test performance, and metacomprehension accuracy.
منابع مشابه
Generating keywords improves metacomprehension and self-regulation in elementary and middle school children.
The ability to monitor understanding of texts, usually referred to as metacomprehension accuracy, is typically quite poor in adult learners; however, recently interventions have been developed to improve accuracy. In two experiments, we evaluated whether generating delayed keywords prior to judging comprehension improved metacomprehension accuracy for children. For sixth and seventh graders, me...
متن کاملElementary School Experience with Comprehension Testing May Influence Metacomprehension Accuracy Among Seventh and Eighth Graders
We explored whether exposure to different kinds of comprehension tests during elementary years influenced metacomprehension accuracy among 7 th and 8 th graders. This research was conducted in a kindergarten through eighth grade charter school with an expeditionary learning curriculum. In literacy instruction, teachers emphasize reading for meaning and inference building, and they regularly ass...
متن کاملParenchymal Pressure Inconsistency in Different Brain Areas After Kaolin Injection into the Subarachnoid Space of Neonatal Rats.
AIM To describe the relationship between the parenchymal pressure changes and the development of hydrocephalus in kaolininjected neonatal rats according to cerebral regions and time intervals of developing hydrocephalus. MATERIAL AND METHODS Neonatal rats aged 2 to 3 days were examined in 5 groups as kaolin frontal "K-F", kaolin parietal "KP", saline frontal "SF-F", saline parietal "SF-P" and...
متن کاملA cryptic new species of Indigo Snake (genus Drymarchon) from the Florida Platform of the United States.
Indigo Snakes (genus Drymarchon) occur from northern Argentina northward into to the United States, where they inhabit southern Texas and disjunct populations in Mississippi, Florida and Georgia. Based on allopatry and morphological differences Collins (1991) hypothesized that the two United States taxa-the Western Indigo Snake, D. melanurus erebennus (Cope, 1860), and the Eastern Indigo Snake,...
متن کاملBiocompatibility of Plastic Clip in Neurocranium - Experimental Study on Dogs.
AIM A potential advantage of the use of the plastic clips in neurosurgery is their property of causing fewer artifacts than titanium clips as assessed by computed tomography and magnetic resonance scans. The biocompatibility of plastic clips was demonstrated in the peritoneal cavity, but their behavior in the neurocranium is not known. MATERIAL AND METHODS Twelve aggressive stray dogs designa...
متن کامل