About 20 years ago, Roediger was running an experiment on how images help people remember. He separated his subjects into three groups and asked each group to try to memorize 60 pictures. The first group just studied the pictures for 20 minutes. The second studied them for most of that time, but was asked to recall the pictures once during the session. But Roediger tested the third group on the pictures three times over the 20 minutes.
When Roediger tested the three groups on the pictures a week later, there were huge differences in how much they each remembered. The first group, which had just studied the whole time, remembered 16 of the 60 pictures. The second group did a little better. But the third group, the ones he had tested over and over, did great. They remembered 32 pictures — twice as many as the first group.
This phenomenon — testing yourself on an idea or concept to help you remember it — is called the "testing effect" or "retrieval practice." People have known about the idea for centuries. Sir Francis Bacon mentioned it, as did the psychologist William James. In 350 BCE, Aristotle wrote that “exercise in repeatedly recalling a thing strengthens the memory.”
But the testing effect had been mostly overlooked in recent years. "What psychologists interested in learning and memory have always emphasized is the acquisition part. The taking [information] in and getting it into memory," Roediger said.
Laypeople — and even experts — tend to think of human memory as a box to be packed with information.
"What people neglected and didn't think about was the 'getting it out' part," Roediger said. "We don't get information into memory just to have it sit there. We get it in to be able to use it later. … And the actual act of retrieving the information over and over, that's what makes it retrievable when you need it."
Why does retrieval, or quizzing, slow forgetting and help us remember?
"It's a good question, and we don't know the answer to it," said Roediger's colleague Mark McDaniel.
One theory is that the act of retrieving information from the vastness of our memory systems poses a challenge to the brain, and retrieval practices that act: in effect, greasing the wheels of memory.
Another theory is that information goes into our brains attached to context. The texture of the book page that we flip as we read; the hum of the air conditioner in the background; the taste of the chips we're snacking on as we study: these all become part of a stored memory.
"Memory is dynamic, and it keeps changing," McDaniel said. "And retrieval helps it change."
Every time a memory is retrieved, it becomes connected to new sensations and contexts. "The more things you have it connected to, the easier it is to pull it out, because you have lots of different ideas that can lead you to that particular material," McDaniel said. "And the things you retrieve get more accessible later on, and the things you don't retrieve get pushed into the background and become harder to retrieve next time."
Studying using these methods requires a lot of work. Rather than just reading, students have to create little quizzes for themselves. It can feel uncomfortable and inefficient. "This is a difficult way to study," McDaniel admitted. "I think most people want learning to be easy and effortless. They want a magic bullet for it. And learning is not easy and effortless. It takes work, and it takes effort and time and dedication."
"I didn't know how to study," said Sydney Baranovitz, a student studying occupational therapy at Georgia Regents University who credits "retrieval practice" for saving her medical career. "I had the ability; I just didn't know what to focus on. The issue with learning is, no one ever sits down and teaches you [how to study]."
Mark McDaniel agrees. "One of the gaps or problems in the educational system is that no one ever helps a student figure out how to learn, and yet that's the primary challenge a student is faced with. You’ve got to assist them with how to do that. And that's where I think we're failing somewhat."
TEST OFTEN FOR BETTER RESULTS
It’s not just that many students are never taught how to study. It’s also that many classes, especially in higher education, are set up to encourage bad study habits.
Andrew Sobel is a professor of international studies at Washington University in St. Louis. He used to teach a freshman introduction to political science class. He structured it in the traditional way, with daily lectures, a midterm exam and a final.
Then he heard Roddy Roediger give a presentation on the testing effect, and Sobel realized that his students were studying in exactly the wrong way, by rereading their notes the night before his two exams.
A vastly better model, Sobel thought, would be one where he essentially forced his students to retrieve knowledge over and over again throughout the course.
So, every semester, instead of two exams, he started giving his students nine quizzes. All these little tests would count for a grade, but they would also, Sobel hoped, be a tool for learning.
At first, Sobel says, his students hated the quizzes. But he was shocked when he realized that by the end of the semester, his students were writing answers to his questions that were comparable to those of his upper division students. "That had never happened before," Sobel said. "And so the only thing that can explain that, the only thing that varied in there, was the testing structure."
Sobel has tried to talk to his colleagues about the results he was seeing with quizzing, but he says most of them aren't interested in switching from a few exams to multiple quizzes. “University faculty are considered very smart, but are also very conservative,” Sobel said. “We don’t like to change our ways.”
"I've always said there was kind of a conspiracy between students and faculty," Roediger said. "Faculty hate making up and grading tests. Students hate taking them. So we pretend they're not very important, and we don't give them. … [Our lab] is arguing for more testing, not less — not standardized tests, but tests that help kids learn."