Lasting Learning in Physics by Constructive Retrieval

• Applicants: Professorin Dr. Claudia von Aufschnaiter; Professor Dr. Alexander Renkl
• Subject Area: Developmental and Educational Psychology ; General and Domain-Specific Teaching and Learning
• Term: since 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 450142163

In physics instruction, students should understand the content so that they can use their knowledge to solve new problems (transfer). In addition, they should remember this knowledge for longer periods (retention). Ideally, teachers should foster the attainment of both goals (transfer and retention) simultaneously to prepare for future learning (e.g., to make conceptual knowledge about the connection between force and motion in linear motion usable for later learning about rotational motion). Despite the relevance of addressing the two goals simultaneously, their attainment is usually investiga¬ted in separate research fields. For example, knowledge consolidation (retention) is studied in research on retrieval practice and learning for understanding and transfer is analyzed in research on generative learning. In this project, we combine retrieval practice and generative learning (here: focus on self-explanation and on example comparison) in physics instruction to test whether we can thereby foster lasting learning outcomes (i.e., delay of eight weeks) with respect to factual retention, transfer, and preparation for future learning. More specifically, we test the assumption that just the combination of both retrieval demands and prompting generative learning activities—also called constructive retrieval—leads to lasting learning outcomes. We address the following main research questions: (A) Can lasting learning outcomes be fostered by combining retrieval demands and self-explanation prompting? Is this effect mediated by both mental effort and self-explanation quality? (Exp. 1). (B) Do students make better use of learning tasks that demand retrieval (i.e., investing mental effort) and of learning tasks that prompt self-explanations (i.e., providing good self-explanations) when learners are informed about the rationale of these learning tasks (principle of informed training)? Does this better use mediate better lasting learning outcomes? (Exp. 2). (C) Are the effects of combining retrieval and generative learning activities on lasting learning outcomes moderated by the complexity of the prompted generative learning activity (prompting self-explanation vs. prompting example comparison; the latter being more complex for students)? (Exp. 3). (D) Are the effects of our instructional procedures (i.e., retrieval demands, self-explanation prompting, example comparison prompting, and informed training) moderated by learners’ motivational goal orientations? (Exp. 1-3). In addition, we replicate an experiment of a partner project within the present Research Unit (Exp. 4). We conduct our field experiments in mechanics instruction at "Gymnasiums" (11th grade), teaching important school-relevant knowledge. Overall, we aim to gain insights about how to optimize learning by combining instructional procedures from different research fields (i.e., retrieval practice and generative learning).

• DFG Programme: Research Units
• Subproject of: FOR 5254: Lastig Learning: Cognitive mechanisms and effective instructional implementation
• International Connection: USA
• Co-Investigators: Professor Dr. Alexander Eitel; Dr. Tino Endres; Professor Dr. Andreas Vorholzer
• Cooperation Partner: Professorin Dr. Shana Carpenter