Does curiosity drive learning?
In recent years, researchers have demonstrated that curiosity — long thought to help motivate learning — is also associated with better learning outcomes. In a paper published in 2009, researchers found that people were more likely to recall the answers to questions they were especially curious about. People’s curiosity levels were the highest when they were especially uncertain about whether their answer was right or wrong. Additional research published in 2014 confirms that people are better able to remember information they are curious about, as well as information learned during high-curiosity states.
In light of these findings, buzz within the education world around interventions to boost curiosity has increased. Researchers, however, don’t yet understand the mechanics of the relationship between curiosity and learning outcomes, nor are they certain of the direction of causality between the two factors.
In a recent paper released in May 2019, Celeste Kidd, an assistant professor of psychology at the University of California at Berkeley, and coauthor Shirlene Wade attempted to answer this crucial question. They used a classic paradigm for studying curiosity – trivia questions. In the first stage, they asked the 114 participants 100 trivia questions. Participants were asked to provide the following: “their best guess” of the correct answer to the question; an estimate of how close they thought their guess was to the correct answer (i.e., their perception of their own knowledge); a yes or no response as to whether they believed their guess was correct; and a ranking of how curious they were about the correct answer to the question.
Research assistants rated the accuracy of each response, providing an objective measure of how close participants’ answers were to the correct response. Participants were then shown the answer to the trivia question and asked to rate how surprised they were by the answer. After an intermediate buffering phase, participants were tested again on the trivia questions.
“It’s relevant to consider what learners know about what they don’t know, because it’s relevant for their motivation.”
The researchers found that the participants’ perceptions of their own prior knowledge — as opposed to the objective measure of prior knowledge provided by research assistants — predicted their curiosity. Learning outcomes, by contrast, were predicted by both curiosity and objective measures of prior knowledge. This finding leads the researchers to conclude that the drivers of curiosity and learning are different.
This offers new insights into learners’ behaviors. “The most important implication … is that we need to rethink what we mean when we talk about classic ideas in education like readiness to learn,” Kidd says. “When we talk about readiness … there’s a lot of focus on the objective truth of the matter. But what this research suggests is it’s also important to consider the meta-cognitive aspects. It’s relevant to consider what learners know about what they don’t know, because it’s relevant for their motivation.”
Kidd and Wade note that their findings support education philosophies, such as those advanced by Maria Montessori, that argue for presenting learners with materials and concepts that are slightly more advanced than their current level of competence. Kidd also notes that overconfidence may help learners, in some situations.
“If you are constantly on the verge of thinking that you know something, that is where we think learning might happen.”
“When you are starting off learning something very new, overestimating your own competence could be beneficial in that it keeps you motivated,” Kidd says. “If you are constantly on the verge of thinking that you know something, that is where we think learning might happen.”