Classroom learning typically involves the restructuring of existing knowledge (Mason & Gava, 2005). One instructional tool that has been found to be effective with enriching the potentials of learning is refutational texts. Refutational texts are designed to activate students’ prior knowledge by directly stating misconceptions about a topic, refuting those misconceptions, then presenting the scientific explanation as a plausible alternative (Hynd, 2003). Indeed, a meta-analysis conducted by Guzzetti, Snyder, Glass, and Gamas (1993) found that refutational texts are more effective than expository texts in promoting conceptual change.

Researchers have offered possible explanations for the refutational text effect, such as the design of the text itself-- misconceptions are elicited then refuted (Chinn & Brewer, 1993; Hynd, 2003). A second view is that refutational texts include persuasive messages (Dole & Sinatra, 1998; Hynd, 2003). What is missing from these studies is an examination of how and why the refutation text effect occurs.

Several international scholars were invited to report empirical research or make a theoretical contribution to exploring the nature of the refutation text effect. In the first presentation, Irini Skopeliti and Stella Vosniadou (University of Athens) look at the nature of the information provided in refutational texts. Next, Lucia Mason and Alessia Schiocchet (University of Padova) examine the effect of individual characteristics on text processing and conceptual change. The third presentation by Suzanne Broughton, Gale Sinatra, and Ralph Reynolds (University of Nevada, Las Vegas, USA) explores the role of attention and background knowledge in processing refutational text information. Finally, Mirjamaija Mikkila-Erdmann, Erkki Anto, and Marjaana Penttinen (University of Turku) use eye-movement methodology to explore text processing and conceptual change. The discussants (Clark Chinn, Rutgers University, Erno Lehtinen, University of Turku) will provide comments on the papers, including insights for future directions for this line of research.

Aim. Recent research has highlighted the intricacy of the process of conceptual change and the delicate interactions of the multiple factors that shape it (Murphy & Mason, 2006; Sinatra & Mason, in press). This presentation is aimed at discussing the findings of two studies on the interplay of instructional and individual variables to extend our understanding of the complex dynamics underlying the process of knowledge restructuring. Specifically, we examined the role of refutational texts in relation to learner epistemic beliefs and interest to try to shed light on why and how this type of effect of text occurs.   

Theoretical framework. Most of the new knowledge students learn in school comes from text reading, especially when abstract concepts are acquired. Therefore, the texts presented to students are an essential factor in conceptual change learning. A particular type of text, the refutational, has been investigated by researchers interested in knowledge restructuring. These texts directly state students’ alternative conceptions about a topic, refute them, and present scientific conceptions as viable alternatives (Alvermann & Hynd, 1989; Hynd, 2003). It has been documented that refutational texts are more effective than traditional expository texts (Diakidoy, Kendeou, & Ioannides, 2003; Guzzetti, Snyder, Glass, & Gamas, 1993; Hynd, 2003; Mikkilä-Erdman, 2002) in producing conceptual change, as they elicit dissatisfaction with the reader’s current conceptions, explain the scientific concept clearly, make it credible, unambiguous, and in multiple forms (Chinn, 1993; Chinn & Malhotra, 2004). In order to understand better what is behind the effectiveness of refutational text, we examined it in relation to two learner characteristics. Following the warming trend (Sinatra, 2005), inspired by Pintrich, Marx, and Boyle’s (1993) seminal article, as learner characteristics we took into consideration two motivational factors considered as possible resources for conceptual change: epistemic beliefs and interest. Regarding epistemic beliefs (Hofer & Pintrich, 1997), it has been documented that only more constructivist representations, that is, beliefs in complex, hypothetical, and evolving knowledge promote students’ recognition that what they already know does not match a new conception (Mason, 1993). It has also been revealed that interest has an energizing role in cognitive processing, as it stimulates attention arousal, positive emotional reactions, effort, and willingness to persist in the task. High interest therefore sustains deep cognitive processing (Alexander & Murphy, 1998; Flowerday, Schraw, & Stevens, 2004; Hidi, 1990, 2001; Schiefele, 1991), which is beneficial to knowledge restructuring.

Research questions and hypotheses. Relying on the findings from these different lines of research, we carried out two studies. The first was guided by the research question: Is refutational text more effective for students with more constructivist epistemic beliefs, who are more likely to recognize a conflict in their conceptual structures? For the second study we asked if topic interest would also interact with epistemic beliefs and refutational text in the process of knowledge restructuring. For the first study we hypothesized that reading a refutational text would be more beneficial for students with more constructivist beliefs about knowledge. For the second study we hypothesized that students with more constructivist epistemic beliefs and high topic interest would benefit most from the refutational text.

Method. The first study involved 110  8^th graders and the topic was natural selection and animal evolution, while the second involved 94 5^th graders studying the topic of light and vision.

In both studies epistemic beliefs were measured by means of self-reported questionnaires, as well as students being assigned to two reading conditions. In one condition they read a refutational text, while participants in the other read a traditional text. Within each reading condition students had more or less advanced beliefs about the nature of scientific knowledge. In the second study they also had high or low topic interest. In both studies repeated measures ANCOVAs were performed (reading comprehension skill was considered as covariate).

 The purpose of this study is to examine the process of conceptual change while reading science text. Thirty sixth graders read a text concerning photosynthesis. Both eye tracking and video material were used together with written pre- and post-tests. This study showed tendencies that a concept called look from time functions as an indicator of cognitive conflict while reading science text. Thus if the reader experiences cognitive conflict, he might try to solve the problem by leaving the critical part of the text, reading previous text, coming back to the critical part and then continuing his reading. In this study new methodological and theoretical aspects were tested. This line of study can be beneficial for developing science texts for school purposes.

Keywords: cognitive conflict, conceptual change, eye tracking, text comprehension

Tracing the process of conceptual change in reading science text

 Mirjamaija Mikkilä-Erdmann, Erkki Anto & Marjaana Penttinen Department of Teacher Education, University of Turku, Finland

  Introduction

 The purpose of this study is to examine the process of conceptual change while reading science text. Theoretical framework is based on following research traditions; text comprehension (e.g. Kintsch 1988; Gilabert, Martinez & Vidal-Abarca 2005), conceptual change research (e.g. Limón & Mason 2002) and eye movement research (e.g. Rayner 1998; Hyönä, Lorch & Rinck 2003).

According to the Construction-Integration -model (Kintsch & van Dijk 1983; Kintsch 1988) the reader forms a text base and a situation model in a continuous cycle while reading. These are constructed through propositions activated by the text. The propositions are interpreted through the reader's knowledge base (i.e. prior knowledge).

It is likely that cognitive conflict, the collision of naive and scientific conceptions, can be experienced in different levels of comprehension. Based on the CI -model, cognitive conflict on a propositional level can be placed in the comparison of a newly formed, naively interpreted proposition and the text base, through which the situation model will be formed.