How Does Retrieval Improve New Learning?
Cover image by Dharmendra Rai from Pixabay
By Althea Need Kaminske
While we talk about the benefits of retrieval practice a lot here at the Learning Scientists, we usually talk about the benefits of retrieval practice for already learned information. However, retrieval practice has also been shown to be beneficial for learning new information. That is, retrieving already-learned material not only helps you to remember that material (1), it also helps you to learn new material (2). While this effect (called the forward testing effect or test potentiated learning or retrieval potentiated learning) has been well established (2), what is less well-understood is how retrieval facilitates new learning.
In a review of studies on retrieval potentiated learning, Chan, Meissner, and Davis (2018) four theoretical perspectives on how retrieval improves new learning: resource theories, metacognitive theories, context theories, and integration theories.
Resource Theories
Resource theories state that retrieval is beneficial for new learning because it frees up cognitive resources. Retrieval acts as a way for you to set aside already learned material and focus on the to-be-learned material. Pretend you are asked to remember two lists of words. According to this theory, you will remember more from the second list of words if you have been tested over the first list because you’re not still thinking about that first list. Chan et al. (2018) that this is consistent with the literature on expressive writing that people have fewer intrusive thoughts during a learning task if they are able to write about them before hand (see this post from Cindy about expressive writing an math anxiety).
Research on mind-wandering is also consistent with this set of theories. Szpunar and colleagues have found that testing between video lectures reduced or changed mind wandering during subsequent lectures (3, 4). Interestingly, in one study when participants were tested intermittently, they reported more lecture-relevant mind wandering than participants who were not tested (4).
Metacognitive Theories
Theories in this category theorize that retrieval practice helps learners to optimize their encoding strategies. That is, after taking a test on information they might change the way they approach learning new information. One way this may happen is that retrieval practice may change how much time or effort learners spend on items that they think are more difficult or more important to learn (2)
Context Theories
Another reason why retrieval might benefit future learning is due to updating context. Contextual information refers to any information that is not necessarily directly related to the to-be-remembered information, but that does provide information about the context (e.g. time or order information - knowing that an item was presented towards the beginning or end of a session.). According to these theories, people are better able to learn new information because old information has been associated with context (2). For example, practicing retrieval on one list of information may may make it more likely that you associate items on that list with “things that happened in the first list” and therefore will be less likely to have those interfere when you try to remember “thing that happened on the second list”.
Integration Theories
A final group of theories suggests that retrieval may benefit new learning because it helps to improve the accessibility of the old information, making it easier to integrate new information (2). These theories are similar to the concept of priming where ideas that are similar to previously presented material are recognized faster. In other words, if you have been learning about (and practicing retrieval on!) tree you may be better able to learn and remember willow because it’s a related topic.
Evidence
A meta-analysis of studies on retrieval potentiated learning revealed mixed support for each of these classes of theories (2). Resource theories had substantial support from the available evidence, however these theories did not fully account for all cases where we find retrieval potentiated learning. Metacognitive theories, somewhat surprisingly (to me, at least), had little support from the available evidence. Context theories also had little support from the available evidence. Finally, integration theories, had strong support from the available evidence, though, again, they did not fully account for all of the data.
What does this mean for educators?
First, there is strong evidence that not only does retrieval practice improve memory for the information that is being retrieved, but it also facilitates learning of new information. Neat! By better understanding how and why that happens, we can better understand when and how to use retrieval practice to improve learning. Researchers are still investigating how and why that happens, but there are some initial ideas. So far, we have strong evidence that retrieval practice may benefit new learning by freeing up cognitive resources and improving integration of new learning.
References
Roediger, H. L., & Karpicke, J. D. (2006). The power of testing memory: Basic research and implications for educational practice. Perspectives on Psychological Science, 1(3), 181-210. https://doi.org/10.1111/j.1745-6916.2006.00012.x
Chan, J. C. K., Meissner, C. A., & Davis, S. D. (2018). Retrieval potentiates new learning: A theoretical and meta-analytic review. Psychological Bulletin, 144(11), 1111–1146. https://doi.org/10.1037/bul0000166
Szpunar, K. K., Khan, N. Y., & Schacter, D. L. (2013). Interpolated memory tests reduce mind wandering and improve learning of online lectures. PNAS Proceedings of the National Academy of Sciences of the United States of America, 110(16), 6313–6317. https://doi.org/10.1073/pnas.1221764110
Jing, H G., Szpunar, K. K., & Shacter, D. L. (2016). Interpolated testing influences focused attention and improves integration of information during a video-recorded lecture. Journal of Experimental Psychology: Applied, 22, 305-318. http://dx.doi.org/10.1037/xap0000087
