If you’ve heard about retrieval practice, you’ve probably heard this mantra:
Retrieval practice helps everyone learn everything.
That is:
- we have research with younger students and older students
- we have research in math and in ELA
- we have research with neurotypical children and atypical learners
Yes: retrieval practice helps with procedural knowledge as well as factual knowledge. Yes, it promotes transfer. Yes, it helps in actual classrooms, not just psychology labs.
Despite all this good news, reasonable folks might ask a skeptical question: “don’t the benefits of retrieval practice have some limits?” Hard experience has taught us that NOTHING is good all the time. So, with retrieval practice like everything else, we should ask: “what’s the catch?”
In this post, I’ll explain:
- An important LIMIT to the benefits of retrieval practice, and
- Important limits to that limit
Remembering Causes Forgetting?
Retrieval practice basics sound simple enough. When I want my students to review, I have two options:
- They could look back over the material they learned. They could, say, review their notes, or reread the textbook chapter. Let’s call this strategy “simple review.”
- They could try to remember the material they learned. They could, say, ask a friend to quiz them on their notes, or write the important highlights from the textbook chapter. We call this strategy “retrieval practice.”
As noted above, we have dozens of studies showing that retrieval practice results in better learning than simple review. In a sentence: “retrieving consistently supports long-term learning.”
Let’s say I have a list of twenty equations, or dates, or vocabulary words. If I really want to remember ten of the twenty, I can use retrieval practice as my study strategy. Voila — I’ll remember them better.
However, this strategy has an important catch.
- Retrieval practice makes it more likely that I’ll remember those ten words that I retrieved.
- But, retrieval makes it less likely that I’ll remember the ten words I didn’t practice.
Retrieving improves memory for retrieved words, but it interferes with memory of un-retrieved words. Yes: retrieval induces forgetting. This surprising memory phenomenon — “retrieval-induced forgetting” — is an important limit to the benefits of retrieval practice.
If I go back to my twenty-item list, I should remember this essential point: if I want to remember all of them, I have to retrieve all of them. If I retrieve some and skip others, those omissions will harm my recall.
Do the Limits Have Limits?
Because retrieval-induced forgetting (RIF) complicates the good news about retrieval practice (RP), we should be as clear as we can be about these trade-offs. To give our students the best study advice possible, we want to maximize the benefits and minimize the harm. With that goal in mind, what advice should we give?
I’ve found a study that explores this question in impressive detail. Dr. Marie Carroll and her team wanted to know:
- Does RIF happen with realistic classroom materials, not just the word lists typically used in psychology studies?
- Does anything reduce the likelihood of RIF, including
- The student’s background knowledge and expertise?
- The amount of time between retrieval and attempted recall?
- The clarity and coherence of the initial explanation?
- Does the kind of final test matter?
Because this blog focuses on research-informed conclusions, I typically try to give a substantial overview of a study’s methodology. In this case, however, RIF itself is complicated to measure — and this study investigates multiple different variables. For that reason, I’ll summarize very briefly.
Carroll and her team had over 100 college students read passages describing patients with different psychological conditions.
- Some students already knew a lot in this field; they were, relatively speaking, “experts.” Others didn’t know much; we can call them “beginners.”
- Some took a test 15 minutes after retrieval; others waited a day.
- Some read organized and coherent passages; others read a jumble of sentences.
- Some took short answer and essay tests; others took multiple choice tests.
When Carroll’s team crunched all that data, what did they find?
The Envelope Please
For the most part, this study provides us with lots of reassurance about encouraging retrieval practice.
The bad news:
- RIF hindered memory for novices taking short answer or essay tests right after retrieving. But that’s the only combination where RIF created problems.
In other words, the good news:
- RIF did not harm recall for relative experts.
- It did not harm recall 24 hours after retrieval practice — even for novices.
- It did not harm recall on multiple-choice quizzes — even for novices.
Teachers should also take note: the clarity of the initial passage didn’t protect students against RIF. We shouldn’t assume that if we teach really well, then RIF won’t be a problem — at least for novices who will take tests immediately after they review.
TL;DR
So, what is the best study advice for students?
- Retrieval practice helps students remember and use all kinds of knowledge.
- Beginners should be careful exactly how they use this strategy.
- Specifically, if they plan to retrieve right before a test, they should retrieve ALL the most important ideas and procedures.
- If they can’t retrieve everything, then simple review is probably a better approach (again, right before the test).
- This research team emphasizes that their study isn’t the final word on the subject. We need further research on the limits of RIF — and that research might update this advice.
In brief: yes, retrieval practice does have limits. But those limits seem to be quite narrow. In most cases for most students, retrieval beats review.
Carroll, M., Campbell-Ratcliffe, J., Murnane, H., & Perfect, T. (2007). Retrieval-induced forgetting in educational contexts: Monitoring, expertise, text integration, and test format. European Journal of Cognitive Psychology, 19(4-5), 580-606.
