Don't Forget to Elaborate

Explicit instruction and retrieval practice are fundamental components of learning, but students need more than that to acquire deep knowledge.

---

The science of learning, or cognitive science, is having a moment—or, let’s hope, more than that. A top-down and bottom-up movement in England and parts of Australia is infusing scientifically grounded principles into classroom teaching. The enthusiasm seems to be spreading elsewhere—including the U.S., albeit slowly. Most of the interest has focused on explicit instruction and retrieval practice.

That’s a hugely important and positive development. For too long, teachers and prospective teachers have been assured that it isn’t necessary—and may even be harmful—to explicitly convey factual information to students, ensure they’ve retained it in long-term memory, and provide them with practice in retrieving it from long-term memory so that the information will be available to them in the future. Scientific studies, in contrast, have yielded abundant evidence that these teaching strategies are hugely beneficial, especially for the most vulnerable students.

But as I’ve read the recent wave of books and articles on cognitive science aimed at teachers, I’ve had the sense that something important is often missing, or at least under-emphasized: the need to have students engage in reflection or analysis based on the information they’ve retained.

I’ve seen this happening in practice too. Teachers in classrooms I visited in Australia were doing an amazing job of implementing explicit teaching and retrieval practice, often with students who are far behind where they should be. These teachers kept up such a brisk pace that kids didn’t have a chance to let their attention wander. The teachers were expert in providing brief, clear explanations of concepts or definitions of words and quickly asking students to repeat the information back, sometimes orally and sometimes by means of answers written and displayed on individual whiteboards. These classrooms bristled with energy.

What I didn’t see, for the most part, were questions that required students to make new connections or add their own ideas.

The emphasis on explicit instruction and retrieval practice is understandable. Those are the components of teaching that the education system has long ignored or rejected. Teachers have been encouraged to go straight for higher-order cognitive activities—analysis, synthesis, and the like—without laying the necessary groundwork. As cognitive science tells us, that doesn’t work.

But laying the groundwork without building on it doesn’t work either. And if teachers get a message that explicit instruction and retrieval practice are all students need, that may be all they provide.

Good ideas in education are sometimes taken to extremes, and that may be what is happening with retrieval practice. “It is now,” education researcher Carl Hendrick has argued, “the darling of the new evidence-informed movement, the miracle treatment that promises deeper learning simply by asking students to bring information to mind.”¹

The Power of Elaboration

Deep learning is complex—far more complex than retrieval practice—and there’s no simple universal formula. But I would argue that one critical component is what cognitive scientists call “elaboration.”

Elaboration simply means adding something to information you’ve absorbed. That could consist of making a connection between different texts or between a text and a student’s pre-existing knowledge or opinions. It could be answering questions like how, why, or what’s an example. (Answering those kinds of questions is sometimes called “elaborative interrogation.”)

“We want students to engage in deep, meaningful, thoughtful processing,” cognitive scientist Jeffrey Karpicke has told me. “And that is what we really mean when we’re talking about elaboration.”²

Studies have found that having students engage in tasks involving elaboration not only improves their retention of information—as retrieval practice does—but also deepens their comprehension and makes long-term retention more likely.³

A 2024 study illustrates the difference between the effects of retrieval practice and elaboration. Researchers had undergraduates watch a recorded lecture on cognitive load theory and then engage in different follow-up activities. One group simply watched the video a second time. Other groups engaged in different kinds of retrieval practice, including one group that had to recall examples from the lecture. Another had to think up their own examples either supporting or contradicting the lecture content, thus engaging in both retrieval practice and elaboration. Yet another group engaged in elaboration without retrieval practice—meaning they had access to the video while coming up with their examples.

The students who simply rewatched the video retained the least information, as would be expected from previous retrieval practice studies. And only the two groups who engaged in elaboration—by coming up with their own examples—achieved deeper comprehension.

Interestingly, there was no significant difference between those two groups. The students who engaged in elaboration only—the ones who had access to the video during their task—retained just as much information as those who had engaged in both elaboration and retrieval practice. Maybe, the researchers hypothesized, that’s because when you’re elaborating, you’re simultaneously retrieving information.

Should Only Experts Elaborate?

The caveat is that you can only elaborate successfully on material you know something about. How much do you need to know? Or, to put that another way, at what point in the learning process does elaboration become beneficial?

You might see the answer as developmental: as kids get older, they naturally become better able to elaborate. Cognitive scientists, I suspect, would provide a different answer. Children do develop more analytical capacity as they mature, but if learners know little about a topic—no matter their age—they aren’t able to engage in any higher-order thinking about it, including elaboration. Conversely, if they’ve acquired expertise on a topic, they might be able to elaborate on it even if they’re in elementary school.

But do you need to be an expert on a topic in order to benefit from elaboration? That seems to be the view of one prominent cognitive scientist I’ve spoken with—John Sweller, the father of cognitive load theory. If you know a topic “really, really well,” he told me, then you’ll elaborate on it more or less naturally. Elaboration, he suggested, is something that can’t really be taught.

True, he acknowledged, students who have only just learned something about a complex topic like the presidency of Abraham Lincoln might be able to deepen their understanding by writing an essay on it. But in his view, it would be more efficient to have those students read a model essay produced by a Lincoln expert and use it as the basis for their own essay.⁴

That approach is akin to showing students “worked examples” of mathematical problems, which has been found more effective than having them try to figure out solutions without such examples. When learners are new to a topic, the cognitive load involved in figuring out solutions without much guidance is so great that it interferes with learning. I agree the same thing can certainly happen when students are asked to write essays about topics they don’t know much about, especially if they haven’t been explicitly taught how to write.

Still, and with all due respect to Sweller, from what I’ve seen in classrooms I’m convinced you don’t need to wait until students are experts to make elaborative tasks worthwhile. There are ways to modulate cognitive load so that elaboration becomes a reasonably efficient way to deepen students’ knowledge—and perhaps create experts in the process.

When it doesn’t consist of just copying information, writing is an inherently elaborative activity, even at the sentence level. Rather than asking students to write an entire essay on a complex topic, teachers can ask students to construct a few sentences on one aspect of that topic, providing carefully crafted sentence stems that focus thinking and modulate cognitive load. Students who have learned something about Lincoln, for example, might get a stem like this: “Lincoln is considered a great president, but __________________________________________.”

Teachers can modulate cognitive load by tailoring a writing task to suit their students’ knowledge of the topic and writing skills. Some students might be guided to engage in elaborative activities orally, lightening cognitive load. Others could be taught to construct a linear outline for a paragraph or essay, which makes the process of writing at length less overwhelming.

Incorporating Elaboration into Instruction

Of course, many advocates of the science of learning do stress the benefits of elaboration and provide ways of weaving it into classroom instruction. In a recent collection of essays titled The Value of a Knowledge-Rich Curriculum, secondary history teacher Harry Hudson explains that a “knowledge-rich” history lesson would include guiding students through a text and ensuring they have understood and absorbed the information. But, he adds, the teacher would also intersperse the reading with questions that “encourage them to think beyond the text itself. Why might this have happened? What might be the consequence of that? How does this link to what was studied last week?”

Similarly, the Four-Question Method framework for teaching history starts with having students narrate a historical event they’ve learned about, which involves retrieval practice, but it doesn’t stop there. The other four questions—after “what happened”—all require elaboration. Students need to discuss what the historical actors were thinking, perhaps based on primary sources; why the event happened in a particular time and place; and what they themselves think about some aspect of it.

Elaboration for Everyone

As I’ve suggested, elaboration shouldn’t be reserved just for older students. In one second-grade class, I saw multiple students’ hands shoot up when the teacher asked them what was unique about Ancient Greek civilization as compared to other ancient civilizations they had learned about. When she called on one boy, he answered, “Something unique was that they weren’t near a river and they didn’t have any fertile soil, so it was difficult for them to farm.” That was elaboration.⁵

Some educators may believe that students who have learning challenges or large gaps in skills and background knowledge need a lengthy period of retrieval practice before they’re ready to engage in elaboration.⁶ But I’ve seen elaborative activities work well in schools that serve many such students, especially when they’re embodied in a method of explicit, manageable writing instruction.

In fact, in such situations, elaborative activities could be more effective than retrieval practice, especially for more complex concepts. Retrieval practice only works if students can successfully retrieve information, and the 2024 study I described above found that elaboration alone resulted in just as much retention of information as a combination of elaboration and retrieval practice. So if students aren’t likely to be able to recall a lot of information from a text, it might be more effective to have them elaborate on the text while being able to consult it.

Enabling all students to engage in elaboration is important for at least two reasons. One is strategic. As energetic as the lessons I observed in Australia were, they were largely about having students absorb and repeat back factual information. Classrooms that look like that play into the hands of those who reject science-informed instruction as nothing more than “rote memorization.” To ensure that the green shoots of the science of learning movement continue to grow, schools in the vanguard need to demonstrate that following cognitive science involves more than that.

The other perhaps more fundamental reason is that elaborative activities are what all students need to realize their full academic and cognitive potential. Of course, to make those activities effective, teachers need support in determining whether students have a sufficient base of information before being asked to elaborate, and whether a given elaborative task is too difficult.

Ideally, they’ll have a curriculum that provides guidance, along with professional learning opportunities, coaching, and assessments tied to that curriculum. But whether they have such a curriculum or not, one relatively simple way to ensure that students are engaging in both retrieval practice and elaboration is to have teachers embed a coherent method of writing instruction in whatever content they’re teaching. The Writing Revolution is one such method. (I’m the co-author of a book that describes the method.)

When I visited Monroe, Louisiana, a high-poverty school district that has adopted that method across subject areas, I heard from teachers at all grade levels that writing instruction had enhanced not only their students’ writing but also their learning ability, reading comprehension—and thinking. That’s what can happen when students are given the opportunity not only to retain and retrieve information but also to process it in a deep but manageable way.

1. In describing several “lethal mutations” of retrieval practice, Hendrick suggests it works only for discrete items like word pairs and definitions—not for interpretation or analysis of complex material. There appears to be a dispute among cognitive scientists on that point, one that I don’t feel qualified to wade into. A recent study argues that retrieval practice doesn’t work for the kind of lengthy, complex texts that college students are expected to read, apparently confirming a limitation on its usefulness first put forward by VanGog and Sweller in 2015. Karpicke and Auer, however, have argued that studies show retrieval practice is “an effective way to improve meaningful learning of complex materials. In any event, I think it’s clear that—whether because the material to be retrieved is too complex, or the task of retrieval alone is insufficient to deepen comprehension—retrieval practice has its limits.

2. Another way of looking at this might be the distinction between “performance” and “learning,” as discussed by Paul Kirschner, Carl Hendrick, and Jim Heal in this article in the current issue of American Educator, which is an excerpt from their book Instructional Illusions.

3. It’s true that there are more studies showing the effectiveness of retrieval practice than of elaboration, but that may be because elaboration is more complicated and therefore harder to study.

4. Actually, Sweller suggested that students might be given essays by several different Lincoln experts, expressing different points of view, as worked examples—not just one. Then, presumably, they would need to synthesize those points of view, or at least decide which one they agreed with, in order to write their own essay. That seems to me to be a far more complex endeavor than imitating the example of just one model essay, with a corresponding increase in cognitive load. I’m not sure it’s all that different from what we ask students to do when they write a research paper.

5. Of course, in order to enable students to elaborate, you need to provide them with a curriculum that is rich in content. The second-graders I was observing had been getting a knowledge-building curriculum since kindergarten. Most elementary curricula, which focus largely on reading comprehension skills, don’t provide enough content to allow students to do much elaborating.

6. The one school I visited in Australia where I did see students engaging in elaboration, by means of writing activities, served a higher-income community.