Students Are Not Always the Best Judges of How They Learn

“I learn better when I can just look back at my notes.”

“I learn better when the teacher tells me exactly what is going to be on the test.”

“I’m a visual learner.”

“I don’t learn from quizzes.”

“I understand it better when I just reread it.”

“I need you to explain it to me first.”

Teachers hear versions of these all the time. And most of the time, I believe the student.

I believe that rereading feels better to them and looking at the notes feels easier. Some prefer the teacher to walk them through the answer. A color-coded study guide can make them feel organized.

The problem is that feeling like you are learning and actually being able to retrieve, explain, and use the information later are not always the same thing.

A student can be completely sincere about how they think they learn best and still be wrong about what will help them remember it next Thursday.

That sounds harsh until you realize adults do the exact same thing. I can read something three times and feel extremely familiar with it. Then somebody asks me to explain it without looking, and suddenly the confidence level changes.

The information looked familiar on the page. That familiarity felt like knowing.

Psychologists have studied this problem for years. It falls under metacognition, which is basically our ability to judge and manage our own thinking and learning. Human beings are not terrible at it, but we are not perfectly accurate either.

One of the traps is fluency. When something feels easy to process, we tend to think we know it better. The words look familiar. The teacher’s explanation makes sense while we are listening. The answer is obvious when it is sitting directly underneath the question. The notes are easy to reread.

Everything feels good.

Then the notes close.

That is where learning gets a little rude.

Research on study choices helps explain why a strategy can feel unhelpful even when it improves later performance.

A 2024 study by Onan and colleagues published in Learning and Instruction looked at interleaved practice, where learners mix different types of problems or categories instead of practicing one type repeatedly before moving to the next. In that study, interleaving produced stronger learning, but participants perceived it as more effortful. They felt that they were learning more when the material was blocked into easier chunks, even though the more difficult interleaved condition produced better learning.

What interests me as a teacher is how differently participants judged the two approaches. Their experience was real, but it did not accurately predict their learning.

I see versions of that in high school classrooms all the time. Ask students whether they would rather reread their notes or close the notebook and try to answer five questions from memory.

Rereading usually wins the popularity contest.

I understand why.

Rereading is forgiving. The answer is right there.

Retrieval is much less polite. Retrieval tells you what you actually know.

There can be an uncomfortable few seconds where the student stares at the question and realizes the information is not immediately available. That feeling can easily become, “I’m bad at this,” “I don’t know this,” or “This way doesn’t work for me.”

Sometimes the effort is part of trying to retrieve the answer. It can also reveal a gap that needs another explanation or corrective feedback. Difficulty alone does not tell me which it is.

Agarwal, Nunes, and Blunt’s 2021 review of classroom research found benefits from retrieval practice across a range of subjects and education levels. That does not mean every retrieval task works equally well for every learner. I still need to consider what students have been taught, the questions I ask, and the feedback they receive.

That nuance matters.

I am not arguing that every learning strategy works the same way for every child.

I teach actual teenagers, not laboratory averages.

Students differ in prior knowledge, language background, reading proficiency, attention, and experience with school. Disabilities and working-memory demands matter too.

But there is a leap education sometimes makes that I am becoming less comfortable with.

We move from students are different to therefore the student already knows exactly which instructional method will produce the strongest learning.

Those are two very different claims.

Students know a tremendous amount about themselves. They can tell me that they cannot hear well from the back of the room. They can tell me that the directions are confusing. They can tell me they need more processing time. They can tell me a paragraph contains vocabulary they do not understand. A multilingual student can tell me the language load is getting in the way of showing what they know. A student with a disability may have legally required accommodations that remove a genuine access barrier.

I take those things seriously.

Access is not a popularity contest.

But “this helps me access the learning” and “this is the learning method I prefer” are not automatically the same statement.

That becomes especially important when we start talking about learning styles.

I still hear students say, “I’m a visual learner.”

Sometimes adults say it about children.

“She’s very visual.”

“He’s a hands-on learner.”

“He just doesn’t learn by listening.”

People absolutely have preferences. I have preferences too. Some students like diagrams. Some like hearing an explanation. Some would rather read it themselves. Some want to touch, move, build, draw, talk, or write.

Preference is real.

The stronger claim is that a student learns better when instruction is matched to a fixed visual, auditory, or kinesthetic “learning style.”

The research has never given that idea the support its popularity would suggest.

Clinton-Lisell and Litzinger’s 2024 meta-analysis examined 21 studies containing 101 effect sizes and 1,712 participants. It found a small average benefit from matching instruction to learning styles. But only 26% of learning-outcome measures showed the crossover pattern needed to support the matching theory. With benefits small and inconsistent, and much of the evidence low in quality, the authors did not recommend widespread implementation.

Don Affognon’s 2026 conceptual analysis also questions the evidence and reasoning behind modality-based learning styles. It is an analysis of the theory and existing evidence, rather than a new classroom experiment.

I find that uncomfortable because education helped teach people this.

Students did not invent “visual learner” by themselves.

Adults handed them the label.

Then some students grew up believing there were entire categories of learning they simply were not built to do.

“I’m not an auditory learner.”

“I’m not a reading person.”

“I need everything hands-on.”

That can become limiting very quickly.

History sometimes needs a map. Sometimes it needs text. Sometimes students need to listen to somebody explain a complicated sequence of events. Sometimes they need to write. Sometimes they need to analyze an image.

Sometimes they need to retrieve information with absolutely nothing sitting in front of them.

The content should help determine the method.

If we are studying territorial change, a map makes sense. If we are studying a political argument, students probably need to read the argument. If I want to know whether they remember something, eventually I have to remove the answer from their line of sight.

I cannot guarantee long-term learning by permanently keeping the information visible.

That is where student preference gets especially tricky.

Students often prefer strategies that make the immediate task easier.

Adults do too.

Ease is persuasive.

The interleaving research raises a problem I recognize in teaching: learners can interpret the extra effort a strategy requires as a sign that it is not helping.

Read that from the perspective of a classroom teacher.

The mental effort involved in learning can make students believe they are learning less.

That explains a lot.

Sometimes a student leaves a lesson where the teacher explained everything beautifully and thinks, “That was easy. I totally understand it.”

Maybe they do.

Maybe the teacher did most of the cognitive work.

Those are not always distinguishable until the student has to do something independently.

I can explain the causes of the French Revolution in a way that makes complete sense while students listen. Heads nod. Nobody looks confused. They may genuinely understand what I am saying.

Then I can ask them tomorrow, “What were two conditions that made France unstable before the Revolution?”

Now we find out what remained.

Those are different moments.

Understanding something while someone else is explaining it is valuable. Being able to retrieve and reconstruct that understanding later is another stage of learning.

That is one reason I care so much about recall in my classroom.

I want students to bring information back because they need knowledge available when they are trying to think with it.

You cannot analyze much with knowledge that disappears the second the slideshow closes.

This is one place where CLSM has changed how I approach instruction. I want the classroom to be structurally predictable because I do not want students wasting limited mental energy figuring out where everything is or what routine we are using today.

I want language to be accessible because vocabulary should not unnecessarily block the history.

I want cognitive load managed because working memory is limited.

And I keep bringing information back because learning needs somewhere to go after class ends.

The structure can feel comfortable.

The thinking does not always need to.

That has probably become one of the biggest differences in how I teach.

I am much less impressed by how smooth a lesson feels.

I care more about what students can do afterward.

A lesson can look successful because it is smooth, entertaining, or easy. I still need to check what students learned.

Students can be engaged for ninety minutes and remember almost nothing tomorrow. They can also complain about a five-minute retrieval activity and remember more because of it.

Engagement matters. Student experience matters. Enjoyment matters. I want kids to like being in my classroom.

I am simply not willing to use enjoyment as the sole measurement of whether something worked.

The same goes for student feedback.

I think schools should listen to students. I ask mine questions. I watch what frustrates them. I notice when something is confusing. I change things.

Students see parts of the classroom that teachers cannot see from the front of the room.

I consider student feedback alongside my professional judgment, assessment evidence, and research on learning.

Sometimes students tell me something I need to change.

Sometimes they tell me something that teaches me what they still need to learn how to do.

Those require very different responses.

A student may say, “I learn better when I can use my notes on the quiz.”

Maybe what they mean is, “I perform better on this quiz when the answers remain available.”

That is useful information.

It just answers a different question.

If the purpose of the quiz is to see what they can retrieve independently, keeping the notes open changes what I am measuring.

A student may tell me, “I learn better when you tell me the answer.”

Of course that feels better.

Being told removes uncertainty.

It also removes the retrieval attempt, the productive mistake, and the need to decide what to try next.

A student may tell me, “I study better when I reread everything.”

And rereading can absolutely be part of studying.

But rereading creates familiarity very quickly.

Recognition starts impersonating memory.

Then the student walks into the assessment furious because, “I studied for two hours.”

I believe them.

Time spent studying and quality of studying are not identical.

A 2025 study by Seban, Šikl, and Urban of 479 university students found that rereading, note-taking, and highlighting were among the strategies students commonly used and considered effective, while stronger strategies such as retrieval practice and distributed practice were used much less frequently. Only 17.7% reported that they had learned study strategies through formal educational settings.

That study involved university students in Slovakia, so I would not pretend those exact percentages describe American high school students.

The broader finding is still worth noticing.

Students are often expected to manage their own learning before anyone has actually taught them much about how learning works.

That part bothers me.

We tell teenagers, “Study.”

Then get annoyed when “study” means staring at Quizlet for forty minutes.

Maybe we should teach studying with the same seriousness we teach content.

Close the notes and see what you remember. Check it, fix what was wrong, and come back tomorrow.

Mix old material with new material. Explain it without reading it. Use the vocabulary and connect it to something you learned last week.

That is less glamorous than a learning-style inventory.

It is probably more useful.

Students also need to learn what productive difficulty feels like.

That may be one of the most important pieces.

Difficulty is not always evidence that instruction is bad.

Confusion is not automatically productive either.

There is a difference between a student struggling because the directions are poorly written and a student struggling because retrieving the answer requires effort.

There is a difference between inaccessible language and challenging thinking.

There is a difference between cognitive overload and cognitive work.

I spend a lot of time trying to separate those.

If the barrier is unnecessary, I want to reduce it.

If the thinking is the point, I am much less interested in removing it.

That sometimes means I have to tolerate students not liking something immediately.

Teachers are under tremendous pressure to make learning engaging, personalized, enjoyable, relevant, accessible, and student-centered.

Most of those goals are worth pursuing.

But somewhere in that conversation, we need room for the possibility that students may not always choose the strategy that produces the strongest learning.

That is normal.

They are learning how to learn too.

Adults are not perfectly calibrated either.

I still catch myself rereading something and thinking I know it because it looks familiar. Then I close it and discover my brain has apparently retained three nouns and a general vibe.

Students need practice checking how closely their confidence matches what they can recall.

Researchers call that calibration.

In plain English, it means getting better at answering one very important question:

What do I actually know?

That is a skill.

Teachers can help build it.

Have students predict how they will do before a quiz. Let them retrieve information before reopening the notes. Compare what they thought they knew with what they could actually recall. Show them the difference between recognition and retrieval. Let them experience a harder strategy, then look at what they remembered later.

And tell them why we are doing it.

The 2024 interleaving study is encouraging on that point. Researchers found that teaching learners about the misconception surrounding effort and learning increased their willingness to use interleaving. Some of those changes remained a week later and even transferred to a new task.

Students can learn to become better judges of their own learning.

That is very different from assuming they already are.

I want students eventually making intelligent decisions about how they learn.

I want independence.

But independence does not develop by handing a fourteen-year-old complete control of instruction and hoping intuition takes care of the rest.

Part of education is giving students better intuition.

Students need opportunities to try a strategy, check what they remember later, and use that evidence when deciding how to study. I want them to understand why something can feel difficult and still be useful, while knowing they can tell me when the directions, language, or task itself are getting in the way.

Originally published on LinkedIn on November 4, 2026.