Can we ever research a topic too much? Can we reach a point where, well, there’s nothing really more to say about teaching better and learning more?
Perhaps, for instance, we’ve reached peak retrieval practice.
Blog readers – and conference attendees – know that actively recalling information results in greater learning than simple review.
For example: rather reminding my students of yesterday’s discussion of the Harlem Renaissance, I can ask them to write down the key details from memory. When they make the mental effort to remember, they learn more.
This blog and many authors have written about this topic at length. What more is there to say?
I recently found a study that reminds us: there’s always more to say. If we want to combine teaching experience with researcher insight, we need to take time to get the details just right.
Here’s the story.
A Problem, a Solution, Another Problem
One problem with retrieval practice: it takes time.
I ask the question.
The students write their answers to the question.
I check their answers.
The minutes tick by.
Wouldn’t it be great if we could skip a few steps. How about this abbreviated version:
I ask the question.
The students think about their answers to the question.
I move on.
If my students truly think about the answers, then they’ll get the retrieval practice benefit in much less time.
This solution, however, creates its own problem.
If my students don’t write anything down, how can I know they actually think about the answers? Couldn’t they just nod and look earnest?
After all, what’s their motivation to do the thinking?
Let’s Check
A respected research team in this field has explored this set of problems, and their potential solutions.
In a recently published study, Megan Sumeracki and Johanny Castillo wanted to see if that first problem exists.
They had college students read a short passage. Some wrote answers to review questions; some were instructed to think about answers to those questions.
What happened a few days later?
Sure enough, the students who just thought about (but did not write down) answers were relatively confident that they’d remember information. (That is: they were more confident than those who wrote answers down.)
However, the thinkers actually remembered less than the writers.
Sure enough, as we predicted, students don’t always follow instructions to think about answers.
In other words: when I solve the first problem (retrieval practice takes time) by asking students simply to think, I create a second problem (students don’t really think).
How do we solve this conundrum? Can I solve BOTH problems?
Despair Not
Sumeracki and Castillo had an idea.
They repeated the “think about it” strategy, but this time with an additional ingredient: cold calling.
That is: they asked students to try retrieval practice by thinking about the answer. AND then they cold-called one student at random. (That is, they called on one student who hadn’t raised a hand.)
The researchers hoped to communicate this message: when told to think about the answer, students really should think about it – because they might actually have to answer the question.
What did they find?
Sure enough: students who thought about the answer now remembered as much as the students who wrote down their answers – presumably because they really did the thinking.
This two-part strategy – “retrieval practice by thinking” plus “cold calling” – takes less time AND produces the learning benefits of retrieving.
Problem solved!
One More Problem?
Some readers will have noticed that I raced past a potential controversy.
Truthfully, people do worry about cold calling.
Teachers worry that it creates a hostile, punitive environment. One grad school professor told me that cold calling ramps up stress, and stress destroys the hippocampus, so cold calling is malpractice.
Honestly, we don’t have lots and lots of research here.
One study I’ve found pushes back on the “ramps up stress” narrative. Others support that narrative.
And, as far as I know, we just don’t have good research in K-12 classrooms.
My own instincts say: yes, cold calling can be done badly. But, anything can be done badly. The key point is that cold calling can be done well.
If we create a classroom environment where making mistakes is an entirely normal part of the class routine – an environment that Doug Lemov calls a “culture of error” – then the potential stress of cold calling shouldn’t be a problem.
But, until we have actual research in many different classrooms, I can’t make that recommendation too emphatically.
The Sumeracki and Castillo strategy strikes me as a sensible solution to a real problem. More research on cold calling will make it more persuasive still.
Sumeracki, M. A., & Castillo, J. (2022). Covert and overt retrieval practice in the classroom. Translational Issues in Psychological Science.
