WEST LAFAYETTE, Ind. — Students learn more from taking tests than they do by studying, according to new research from a Purdue University cognitive psychologist and memory expert.
“That’s contrary to our conventional way of thinking; learning only takes place when studying, additional studying increases learning, and testing just assesses what we know,” said Jeffrey D. Karpicke, an assistant professor of psychological sciences.
“Now we have some pretty powerful evidence that during a test there are cognitive processes happening that actually promote learning. Testing is not just an assessment of what you studied. The act of retrieving information actually improves memory because you are practicing a skill. And that’s the exact same skill you are going to need to retrieve that information again and again.”
Karpicke and Henry L. Roediger III, the James S. McDonnell Distinguished University Professor of Psychology at Washington University in St. Louis, co-authored an article that appears Thursday (Feb. 14) in Science. The article is titled “The Critical Importance of Retrieval for Learning.” The researchers, who have made similar findings in earlier research, compared how college students learned by studying and testing on Swahili-English vocabulary words in four different conditions. The students learned the vocabulary words in a computerized flashcard format on the first day and took a final test a week later. Forty students participated in the study and 10 students were tested in each group.
In the first group, students studied all the words and were repeatedly tested on all, even after they could correctly recall the words. In the second group, once a word was recalled it was dropped from studying, but continued to be repeatedly tested. Students correctly recalled about 80 percent of the word pairs on the final test in the two situations that emphasized repeated testing.
In two other situations, once words could be recalled they were dropped from repeated testing. Those participants recalled only 36 percent and 33 percent of the pairs correctly.
“Repeated testing caused a dramatic difference,” Karpicke said. “Related to that, we showed that once you could recall something easily from memory, further study of it produced no additional measurable learning.”
These methods of studying and testing are similar to how students learn from a lecture or textbook material, Karpicke said.
“More research needs to be done, but I am confident these findings have wide application across a variety of educational materials,” he said.
The researchers also asked the students to predict the percent of the pairs they would recall in one week.
“They didn’t have any insight when it came to how well they were learning,” he said. “That is surprising considering these are students who have been learning and studying much of their lives, and it is really amazing that they were not able to accurately predict their abilities.”
Currently, Karpicke is assessing students’ preferences for methods of studying and how students can be encouraged to utilize more effective strategies.
Karpicke also shares his research results with his students when he administers tests.
“Testing is really most effective when students implement it on their own, which can be done simply with flash cards,” he said. “But our research points to a really effective way to use flash cards: Even if you can recall a new piece of information, you should continue practicing recalling it, rather than dropping it from practice. It’s crucial that students develop such testing skills and tools to use on their own because so much learning takes place outside of the classroom.”
This study was supported by a Collaborative Activity Grant of the James S. McDonnell Foundation. Purdue’s Department of Psychological Sciences is housed in the College of Liberal Arts.
