When teachers say we want our students to learn, we might also say we want them to remember; after all, if I’ve learned something, I can remember it later on. Sadly and surprisingly, there’s a curious danger to remembering: remembering can cause you to forget.
Yes, you read that right. The wrong kind of remembering causes forgetting.
Imagine the following mental exercise—a mental exercise that resembles many research studies1:
To start, you study a list of words in four different groups—say, Animals (dog, cat), Instruments (guitar, violin), Foods (pizza, steak), and Furniture (sofa, table). After a while, you recall half of the words in two of the groups. For example, in the Animal group, you recall the word “dog” (but not “cat”), and in the Foods group, you recall the word “pizza” (but not “steak”). And you don’t recall any words in the Instrument or Furniture groups.
When I test you on all these words several hours later, there are three logical categories.
First, there are the two groups of words you didn’t recall all: Instruments and Furniture. You’re likely to remember—perhaps—50 % of those words.
Second, there are the words and groups you did recall: the word “dog” in the Animal group, or “pizza” in the Food group. Because you recalled these words, you’re likelier to remember them, so your score will be higher—say, 75%.
Third, there are words that you didn’t recall (“cat,” “steak”) even though you recalled other words in Animal and Food groups.
Take a moment to ask yourself: what percentage of words in this 3rd group are you likely to remember?
Perhaps—because you practiced their groups—you’ll remember them at the 75% level. Or perhaps—because you didn’t practice these specific words—you’ll remember them at the 50% level.
It turns out both answers are wrong. You’ll remember even fewer of those words: say, 40%.
Why? Because practicing some of the words in the Animal and Food categories makes it less likely you’ll remember the un-practiced words. In other words, recalling some of the words prompts you to forget the words you didn’t recall.
The wrong kind of remembering caused you to forget.
In the neuroscience community, there is an active debate about the mechanisms that cause “retrieval-induced forgetting.”2,3 And while that debate is fascinating, it doesn’t really help teachers answer our constant question: “what should teachers do in the classroom with this scientific information?”
I haven’t read any research that addresses this question directly. (More precisely: I don’t remember having read any research that answers it; perhaps I read it, and forgot the source.) But I think the potential dangers of retrieval-induced forgetting (often abbreviated RIF) should shape our practice in very specific ways—in particular, the way we review.
Here’s an example. In yesterday’s class, my students discussed the five ways that the French and Indian War lay the foundation for the American Revolutionary War. To begin today’s class, naturally, I ask my students what conclusions we reached. One student calls out: “The French and Indian War cost a lot of money, and the British government decided to tax the colonies to pay for it. Those taxes helped spark the revolution.” Exactly so. Another student adds to the list: “George Washington gained essential military training and a cross-colony reputation for bravery.” Because we’ve gone over these two key points from yesterday, I assume my students will be prompted to remember the other three. Confident in this assumption, I move on to today’s new topic…
But there’s a problem here. Yesterday, my students got a list of five key points; today, we began class by reviewing two of them. I hoped—in fact, assumed—that my two-item review will help them remember the other three points. However, if the RIF research is true, then my two-item review will in in fact make it less likely that the students will remember the other three items. Because they practiced two of the examples in this group (“ways that one war set the stage for the next”), they are less likely to remember the un-practiced examples in that group.
When I first read this research, and started thinking about my own teaching practice, I realized with increasing alarm how often I review this way. If we studied ten vocabulary words yesterday, I’ll prompt students to recall two or three. If we looked at eight subject-verb agreement rules, I’ll asked them to jot down two, and discuss them with a partner. Of course, teachers must help their students review the material they learn, but if the first review is incomplete, we may very well be reducing—not increasing—the long-term likelihood that our students remember all the information.
In my own teaching, the RIF research has led to this guideline: the first two or three times I go over a topic, I make sure to cover all of the material that is a) conceptually related and b) equally important:
- “Conceptually related”: RIF results from partial review of conceptually related information only; it influences Animal and Food words, not Instrument and Furniture words.1 For this reason, I don’t need to review an entire lesson—just the logically connected pieces of it. When I go over five essentials for a strong topic sentence, I don’t also need to review the highlights of “Young Goodman Brown.” We discussed both topics on the same day, but our discussion of the short story was conceptually distinct from our discussion of effective writing.
- “Equally important”: when we go over all five ways that the French and Indian War led to the Revolutionary War, I don’t need to go through the detailed specifics; they’re not as important as the main concept. If I think of my lesson plan in an outline, I should cover all (or none) of the points on the same level of that outline.
One final danger to consider: student directed review might be especially prone to RIF. If students come up with their own list of key terms to remember, for example, their incomplete list might prompt them to forget the examples they didn’t include. As teachers, we need to find mechanisms to ensure that student generated review covers all equally important information.
Of course, research into RIF continues, and we don’t yet completely understand how and why it happens. For teachers, the key point to keep in mind is this: whenever we prompt our students to review, we must be sure that RIF doesn’t cause them to forget what we want them to remember.
References & Further Reading
- Jonker, T. R., Seli, P., MacLeod, C.M. (2012). Less we forget: Retrieval cues and release from retrieval-induced forgetting. Memory & cognition 40(8), 1236-1245. [Paper]
- Dobler, I.M. & Bäuml, K.T. (2013). Retrieval-induced forgetting: dynamic effects between retrieval and restudy trials when practice is mixed. Memory & cognition 41(4), 547-557. [Paper]
- Mall, J.T. & Morey, C.C. (2013). High working memory capacity predicts less retrieval induced forgetting. PLOSOne 8(9), e52806. [Paper]
- Johansson, M. et al. (2007). When remembering causes forgetting: Electrophysiological correlates of retrieval-induced forgetting. Cerebral Cortex 17(6), 1335-1341. [Paper]