In Brief

Think you'll know later what you think you know now? Think again. People tend to overestimate their knowledge of a topic while studying it and often cannot recall studied information when tested later, according to a study in March's Journal of Experimental Psychology: Learning, Memory and Cognition (Vol. 31, No. 2).

Study authors Asher Koriat, PhD, of Israel's University of Haifa, and Robert Bjork, PhD, of the University of California, Los Angeles, examined how students can avoid what Koriat calls the "curse of knowledge"--when they judge that they've learned an answer but fail to discount the fact that the answer is in front of their face.

They don't seem to realize that the answer won't be in front of them on test day, he says. Students could retrieve an answer were it in long-term memory, but, the researchers posit, the curse of knowledge makes them think they've already got the answer down cold.

In one experiment, 16 Haifa undergraduates studied 72 word pairs on a computer for about four seconds each. The researchers evenly divided the 72 word pairs into three categories. Unrelated pairs served as a control. A priori words were those in which the second word was a very likely response to the first word through free association (such as "cheese" to the word "cheddar"). A posteriori words were those in which the words seemed related when placed together, but the second word, according to free-association norms, was very unlikely to come to mind when the first word was shown alone (such as "citizen" and "tax").

Participants judged the probability that they would later be able to recall the second word of a pair when prompted with the first word. The first word in each pair appeared onscreen during the test phase, and participants had six seconds to name its mate.

Participants accurately predicted their recall scores for the 24 a priori words but vastly overestimated their performance on the a posteriori words.

The findings suggest that people are susceptible to what the authors term "foresight bias"--namely, people's tendency to overestimate their understanding and knowledge of a topic when they see an "answer" presented along with a "question."

"In monitoring our knowledge, we rely on a gut feeling that reflects information processing taking place in the here and now," Koriat explains. "Doing so, however, creates an inherent conflict. To learn, information must be in front of you. You must study that information but somehow also discount that information when measuring your own knowledge. To predict self-knowledge accurately, a learner needs to adopt the perspective of the examinee."

So how can students truly know that they know something?

"Self-testing can be much better than simple rehearsal because it improves both memory and metamemory," says Koriat, noting that self-testing not only enhances long-term retention but also improves one's ability to monitor one's own mastery of the material. "Come back later, cover the answer and test yourself," he adds. "That's the best way to overcome foresight bias."

--M. GREER