Promoting text comprehension through refutation, visualization and consolidation: A triad approach to facilitate learning from texts

Journal articleResearchPeer reviewed

Publication data


ByJohannes Kröger, Hendrik Härtig, Jan Retelsdorf, Sascha Bernholt
Original languageEnglish
Published inInternational Journal of Science and Mathematics Education
Pages27
Editor (Publisher)Springer
ISSN1573-1774, 1571-0068
DOI/Linkhttps://doi.org/10.1007/s10763-025-10608-0 (Open Access)
Publication statusPublished advanced online – 08.2025

Detailed verbal descriptions of scientific concepts in the form of texts are indispensable in science education. However, students find it difficult to use text as a productive resource in science. It is therefore of great interest to enable students to use science texts for knowledge acquisition. A crucial factor impeding science learning (from texts) are misconceptions. To encourage active engagement with these misconceptions, refutation texts have been established as a promising tool. Unfortunately, poor readers usually do not benefit from these texts. Hence, graphic organizers in combination with refutation texts can be considered to scaffold those readers. To further support consolidation of newly learned information into the long-term memory, a variety of instructional tasks can be implemented. Yet, many specific interactions of these measures with person characteristics as well as their combination are still unknown. The present study is aimed to shed light on possible interactions and effects of textual and instructional measures. Therefore, a repeated measures design was employed focusing on students’ learning gains after reading different content representations as well as after processing a set of consolidation tasks. Surprisingly, we found that students (N = 326) performed equally well in knowledge tests regardless of the content representation presented and did not further benefit from additional tasks. Furthermore, students did not seem to benefit differently from content representations with regards to different covariates assessed. However, structural equation modelling revealed potential differences in the activation of cognitive processes between the groups, suggesting the need for further research.