Retrieval then generative learning or generate learning then retrieval?

Riding on @P_A_Kirschner coat tails, I decided to have a look at the research paper and Professor Julian Roelle research more widely. There were lots of interesting research topics to explore (quizzing, self-regulation, metacognition, Cognitive Load, worked examples) but for now, I stuck with generative learning tasks and retrieval practice.

Essentially, Roelle et al, (2022) where investigating the ‘best’ way to combine retrieval practice with generative learning (selecting, organising, and integrating information) given that they “serve fundamentally different functions and thus promote learning via different mechanisms.”

Generative learning activities such as organisation and elaborating mainly serve the function of constructing coherent mental representations of the learning content that are well integrated with learners’ prior knowledge.

[Post testing] Retrieval practice mainly serves the function of consolidating the constructed mental representations in memory. In view of these complementing functions, the combination of generative learning and retrieval practice has received increasing attention in recent years.

When learners are engaged in both types of activities – does the sequence matter? Roelle and Nuckles (2019) presented a very similar question in an earlier paper – the benefits of retrieval practice substantially depend on the quality of learners’ mental representations before they engage in retrieval practice.

What did we learn?

• Retrieval practice before generative learning is more effective than vice versa.
• Engaging in retrieval practice first fosters retention.
• Engaging in retrieval practice first reduces subjective cognitive load.
• The retrieval-first effect on extraneous load mediates the effect on retention.

I would have been interested in a third condition. When both retrieval and generative activities are combined, for example, when students are asked to retrieve a piece of information and then asked to elaborate.

For example in the KS3 Romeo and Juliet deck there is a card that asks

Do these conclusions apply when retrieval is post-testing and what about when pre-testing? In the same way as different generative learning strategies are more or less effective with different student cohorts? I would underline that the paper reports retrieval as a consolidation activity and I would argue, it is also a very powerful encoding activity.

Roelle, J., Froese, L., Krebs, R., Obergassel, N., & Waldeyer, J. (2022). Sequence matters! Retrieval practice before generative learning is more effective than the reverse order. Learning and Instruction, 80, 101634.

Roelle, Julian & Nückles, Matthias. (2019). Generative Learning vs. Retrieval Practice in Learning From Text: The Cohesion and Elaboration of the Text Matters. Journal of Educational Psychology. 10.1037/edu0000345.