Trying and Failing Enhances Learning According to Research
WILLIAMSTOWN, Mass. - Sometimes, coming up with the wrong answer to a question can help you come up with the right one later on.Nate Kornell, assistant professor of psychology at Williams College, has found that we are more likely to remember new facts if we are first asked to produce them ourselves -- even if we get them wrong.
Psychologists have known for years that "testing" students about information is more effective than just presenting it. That is, at least, when students get it right. Kornell wondered, "If successful tests enhance learning, do unsuccessful tests impede learning -- or do they also enhance learning?"
To find out, Kornell and two colleagues set up six experiments. In the first two, the researchers asked subjects to answer trivia questions. To make sure the subjects got them wrong, half the questions were fictional, with fictional answers. ("Who is the bouncy and egotistical friend of Kenny Peters?") The rest of the questions were nonfictional, thrown in to keep the subjects from suspecting anything.
The questions were presented in two different ways. In some cases, subjects were simply shown the question and answer together. In the other cases, they were asked the question and given eight seconds to come up with the correct fictional answer. After failing to do so, they were shown both the question and answer.
Then the subjects were tested to see how well they recalled the new information. As it turned out, they were more likely to remember the answer if they had tried and failed to produce it on their own than if they had only studied it. Then the researchers ran the experiment again, changing the length of the time subjects had to study the questions. They found that "unsuccessful retrieval attempts were just as effective as studying the answer."
In the next four experiments, researchers used weak word associations instead of fictional trivia questions. The word pairings, like "olive-branch" and "train-caboose," allowed subjects to make semantic connections between the cue and target. When subjects were "tested" and then provided with feedback, they learned more than if they spent the same amount of time studying the cue and target words together.
From the six experiments, Kornell and his co-authors concluded that "unsuccessful tests are helpful, not hurtful," as long as feedback is provided. Their findings suggest "educators and learners should introduce challenges into learning situations, including using tests as learning events, even if doing so increases initial error rates."
Kornell, Matthew Jensen Hays, and Robert A. Bjork published their findings in the article Unsuccessful Retrieval Attempts Enhance Subsequent Learning, which appeared in the Journal of Experimental Psychology: Learning, Memory, and Cognition.
Kornell joined the Williams faculty this fall. He is teaching the course "Optimizing Learning and Memory" and will teach "Cognitive Psychology" in the spring. His research focuses on human learning and memory, education, metacognition and self-regulated learning, and memory in nonhuman animals.
His other articles include The Pretesting Effect: Do Unsuccessful Retrieval Attempts Enhance Learning? (2009) and Learning Concepts and Categories: Is Spacing the "Enemy of Induction"? (2008). His writing has appeared in Psychological Science and the Journal of Experimental Psychology, among others.
Kornell is the authors of chapters in the forth-coming books "The Encyclopedia of Applied Animal Behaviour & Welfare" and "Encyclopedia of the Mind."
Kornell received his B.A. from Reed College. He received his Ph.D. in psychology from Columbia University in 2005. In 2007, the Association for Psychological Science named him a "rising star."
