Disentangling Maintenance and Retrieval Processes in the Testing Effect: The Role of Free-Recall Practice

09:00 - 10:30

When and Why Instruction Works: Mechanisms and Boundary Conditions of Interventions in the Science of Learning

Room: HSZ - N5

Chair/s:

Bernhard Pastötter, Tanja C Roembke

This symposium surveys diverse instructional learning approaches and their boundary conditions across tasks, materials, and learner groups. Five talks address the benefits of production, interleaving, pretesting, backward testing (retrieval practice), and forward testing for learning, retention, and transfer, with an emphasis on examining and discussing the cognitive mechanisms of these effects. Tanja Roembke will talk about the production effect, which describes better memory for items spoken aloud than read silently. The presented research investigates to what extent semantic spreading activation contributes to the production effect. David Shanks will talk about the interleaving effect and improved inductive learning when exemplars are mixed rather than blocked. The talk will analyze why learners (children and adults) often misjudge blocking as superior and will describe an intervention that reverses this metacognitive illusion, leading to more advantageous scheduling choices. Oliver Kliegl will talk about the pretesting effect, which describes improved long-term retention when learners attempt answers before study, even when initial guesses are incorrect. The talk will demonstrate that the magnitude of this benefit increases with the number of guesses, with generalization across weakly related word pairs, prose passages, and age groups. Simone Malejka will talk about the (backward) testing effect, which describes superior long-term retention after retrieval practice compared to restudy. She will compare restudy and free-recall practice and use cognitive modeling to separate maintenance from retrieval contributions, specifying conditions under which retrieval practice yields the largest benefits. Bernhard Pastötter will talk about the forward testing effect, which describes enhanced learning of new material following interim tests on earlier material. He examines how divided attention affects forward-testing benefits for word lists and text passages, delineating boundary conditions under which interim tests enhance subsequent learning. Collectively, the symposium addresses five central instructional learning techniques and distills both theoretical (mechanistic) conclusions and practical implications for real-world learning and instruction.

Submission 571

Disentangling Maintenance and Retrieval Processes in the Testing Effect: The Role of Free-Recall Practice

SymposiumTalk-04

Presented by: Simone Malejka

Simone Malejka ¹, Edgar Erdfelder ², Christoph Stahl ¹

¹ University of Cologne, Germany

² University of Mannheim, Germany

Research on test-based learning typically compares restudying with cued-recall practice, showing that testing without feedback reliably enhances long-term retention—a phenomenon known as the testing effect. However, the cognitive mechanisms underlying this effect remain debated. To disentangle maintenance and retrieval processes, we apply a measurement model that estimates their independent contributions to performance. This approach involves two key modifications to the standard testing-effect paradigm: (a) extending the final test to include both cued-recall and free-recall components and (b) administering these tests both immediately and after a delay. We implemented this design to investigate whether free-recall practice produces a testing effect and, if it does, which underlying memory parameters are positively influenced. Notably, free recall has been used far less frequently than cued recall in testing-effect research and, to our knowledge, free recall of paired associates—as opposed to single items—has not previously been examined. By fitting the proposed model to the data, we assess whether free-recall testing (relative to restudying) primarily affects the maintenance of encoded information across the retention interval or the retrieval of that information at final test.

Bookmark