Uncovering the "Hidden Lives" of Learners

A book by the late New Zealand-based researcher Graham Nuthall provides a rare and important window into what and how students learn.

---

Here’s part of an interview Nuthall’s team recorded with a New Zealand middle-school student named Ann, a year after her class had done a unit on New York City. The interviewer asked Ann about a test question: “New York is made up of (a) three states, (b) four suburbs, (c) five countries, (d) five boroughs, (e) I don’t know.”

Ann: I thought it was made up of five states … I know it’s not made up of five countries. But it could be made up of three states.

Interviewer: Right. Explain to me how you learned that.

Ann: Um, well, I don’t think it’s five countries. I don’t know why, but it couldn’t be five countries, ‘cause one country is America … And, oh, it could be five boroughs. But I don’t think so … I don’t know …

Interviewer: Were you taught about it in class?

Ann: Um, yes. I think we were actually. Well, that’s all I could remember, five states. I think … I just, I don’t know why, but I thought it was five states.

Interviewer: Right. And you’ve got a feeling it came up in class? Can you remember anything else? … Can you remember the names of the states?

Ann: Yeah, I think it’s actually three boroughs. It must be because it’s not three countries, ‘cause I remember that Manhattan was one of them. But I can’t remember any other ones.

Clearly, Ann hadn’t retained much of what had been covered.

In their classroom-based research, Nuthall and his colleagues found that students are also frequently misled by their own assumptions, the misconceptions offered by their peers, and their lack of attention.

A confident but low-performing student named Tui, for example, fails to grasp much that goes on in class because he’s often singing to himself or relying on what he thinks he already knows. Although the teacher and other students discuss the way magnetic force determines how a compass works, for example, Tui sticks to his belief that the key factor is the wind, whispering to himself, “No, the wind’s pulling it.”

Teachers may wonder how they can reach students like Ann and Tui—and why so many students don’t seem to have learned what they’ve been taught. Nuthall’s book, called The Hidden Lives of Learners, can provide many of the answers.

Nuthall died in 2004, and the book—which drew on four decades of his research, both in New Zealand and the U.S.—was published posthumously in 2007. But it’s unlikely that much of what he describes has changed in the last 20 years, so his insights still have relevance.

Instead of focusing on what teachers were doing, as virtually all other education researchers have, Nuthall and his colleagues focused on what students were doing and thinking, using microphones that recorded their self-talk and peer interactions, along with researchers’ observations and interviews.

Nuthall’s work isn’t as well-known as it should be, at least in the U.S. I won’t attempt to provide a complete summary of his findings (more detailed accounts appear in an article in researchED magazine from 2019 and in a blog post by Tom Sherrington from 2020). I recommend reading the book itself, since it’s written in an accessible style and is only about 160 pages. I’ll just home in on a few of the many points that piqued my interest and sparked some thoughts.

The Three-Times Rule

One finding gleaned from several of Nuthall’s studies is that students need at least three different encounters with “the complete set of information” necessary to understand a concept in order to retain it in long-term memory. This three-times rule could predict what students would learn with an accuracy rate of 80 to 85 percent.

That doesn’t mean teachers should just repeat the same information but rather that aspects of it should be presented in different ways and contexts over a period of time, perhaps by breaking a big question or concept into smaller ones. (Of course, if you break down a concept into smaller ones, more than three encounters may be needed for students to get enough exposure to “the complete set of information” necessary to understand the big concept).

As Andrew Watson has noted, this discovery seems to have anticipated the “spacing effect,” a finding from cognitive science that students learn more when they spread out study of a specific concept over time rather than “massing” their practice.

The three-times rule also helps explain why a literacy curriculum that focuses primarily on comprehension skills is less effective than one that focuses on building knowledge. In a skills-focused curriculum, students are exposed to random content in a fleeting, superficial way—the solar system today, reptiles tomorrow—because the idea is that what they’re really learning are supposedly abstract skills like “determining the author’s purpose.” (In fact, reading comprehension, like learning in general, is highly dependent on background knowledge.) With that approach, kids are unlikely to retain whatever new information they’re exposed to because they encounter it only once.

An effective knowledge-building curriculum sticks with a topic for several weeks or more, giving kids repeated exposure to the same content and vocabulary in different contexts. That repeated, spaced exposure helps them retain the information in long-term memory, boosting reading comprehension and further learning.

All Students Learn at the Same Rate

“Our studies have shown,” Nuthall writes, “that we need to rethink the commonly held belief that high-ability students learn differently, or more quickly, or more efficiently than low-ability students.” This is a corollary to the three-times rule: everyone needs at least three encounters with a new concept to learn it, including the smart kids.

Of course, some students do learn more than others, but that’s largely because they start out with more knowledge that’s relevant to what they’re learning. In addition, Nuthall writes, some students are able to “create more learning experiences for themselves” because of “their cultural backgrounds, their interests and motivations, and their skills in managing their social relationships.”

This may sound counter-intuitive—it certainly seems that some students are simply smarter or quicker than others—but it echoes a more recent study finding, to the surprise of the researchers, that all students learn at basically the same rate. That study, which was done at Carnegie Mellon, found that seven “learning opportunities” were necessary for any learner to master a concept. But like Nuthall, the researchers concluded that the key difference among learners was the amount of background knowledge they started with. (The seven-vs.-three discrepancy may stem from a difference in measurement; the Carnegie Mellon researchers appear to have broken down concepts into smaller chunks than Nuthall did.)

These findings suggest that if we can build students’ background knowledge, we can significantly boost their achievement—and Nuthall’s findings are even more encouraging than the Carnegie Mellon ones. The instructional software used in the Carnegie Mellon study incorporated practices that are backed by cognitive science but rarely used in classrooms: frequent low-stakes tests, immediate feedback on errors, repeated opportunities to master tasks. Nuthall’s research suggests that teachers don’t need to use all those practices for their instruction to work (although it would undoubtedly help). They do, however, need to ensure that all students get enough repeated engagement with the same concept.

Students’ Prior Knowledge and Misconceptions

Nuthall found that most students already know 40 to 50 percent of what the teacher is going to teach—but what they know differs dramatically from student to student, along with their levels of knowledge. Perhaps 20 percent already know everything that will be taught, another 50 percent know something about it, and another 20 percent have little or no relevant knowledge.

Because background knowledge determines so much of what a student will learn, teachers need to have an idea of what individual kids already know about a topic—and what misconceptions they have—before introducing it. In Tui’s case, the teacher was presumably unaware of his pre-existing belief that the wind caused the needle of a compass to move, a belief that persisted in the face of information to the contrary.

As the book reveals, encouraging students to work out explanations for themselves often only exacerbates such misunderstandings—especially in a small-group situation where a dominating but misinformed student like Tui can wield influence. And teachers are often unaware of what students are failing to understand. “Evolving misunderstandings are never easy to identify,” Nuthall writes, “especially when the students talk as though they understand.”

One way to uncover misconceptions is simply to ask students what they think they already know about a topic. I saw this approach in a first-grade class I observed that was learning about Ancient Egypt. What do you know about mummies, the teacher asked?

Among the typical student responses were these: “They chase you … They walk like they’re crazy … They kidnap you.”

“You all have a lot of ideas about mummies,” the teacher observed mildly, before proceeding to read them a nonfiction book about mummies. Afterwards, she was able to tell them that the kinds of things they had learned on TV about mummies—that they come back to life—were made up. “We’re studying about what really happened,” she said.

Another way of uncovering hidden misunderstandings is through writing activities (which are also a great way to give students the repeated practice they need to retain new concepts and vocabulary). In a third-grade class I observed that was reading Charlotte’s Web, the teacher asked students to finish the sentence, “Wilbur is different from the other animals on the farm, but …”

These students had been working with the conjunction but for months, and they could easily recite a definition: it shows a change of direction. Nevertheless, several students struggled to complete the sentence in a way that made sense. One girl, for example, wrote “Wilbur is different from the other animals on the farm, but he is a different animal.”

Class discussion might not have been enough to uncover this misunderstanding. But thanks to the writing activity, the teacher was able to recognize and address the problem before it was too late, helping students deepen both their understanding of how to use but and their understanding of the story.

The Limits of Randomized Controlled Trials

The “gold standard” in evaluating the effectiveness of pedagogical approaches has become the randomized controlled trial. The idea, borrowed from medicine, is that if you try out an intervention on two groups that are as similar as possible, you can then conclude that whatever difference you see is caused by the intervention—because that’s basically the only relevant difference between the two groups.

Nuthall points out that this approach doesn’t translate well to the education context. The nature of a drug doesn’t change depending on who administers it. But, Nuthall says—drawing on his experience as a researcher—teaching methods do. “Every teacher adapts and modifies so-called methods,” he writes. “Research shows that teachers who believe they are using different methods may be doing essentially the same things, and teachers who believe they are using the same method may be doing quite different things.”

Perhaps that doesn’t mean we should completely give up on the idea of randomized controlled trials in education. But we should at least be cautious in drawing conclusions from their findings. In conjunction with well-established principles from cognitive science—like the critical role of prior knowledge in learning—rigorous and deeply observational studies like Nuthall’s may ultimately be more illuminating. We need more of them.