Nate Kornell, Ph.D.
Postdoctoral scholar, UCLA department of psychology.Contact: nkornell@ucla.edu
Additional contact information can be found on my CV.
Research overview
My research centers on understanding the functional architecture of human learning and memory. I am particularly interested in two interrelated issues: What do people believe about their own learning and memory, and how can we optimize learning, particularly in educational settings? The unifying theme of my research is comparing optimal study strategies with people's actual study strategies--that is, I compare what works to what people think works. I have investigated these issues--namely, metacognition, study decisions, and conditions that optimize learning--in various populations, including college-age and older adults, children, and rhesus monkeys. Brief descriptions of some ongoing and recent projects are provided below.Misunderstanding memory
The dynamics of memory are complex and frequently unintuitive. Still, some features seem straightforward; for example, it seems obvious that repeated studying increases learning. When I asked participants in an experiment to predict the likelihood that they would remember a series of word pairs after studying them between 1-4 times, however, they predicted little or no difference, despite large differences in actual learning. This underestimation of learning ability reflects a fundamental misunderstanding of how memory works.Spacing and inductive learning: studying paintings
Learning about a new concept or category by observing examples--that is, inductive learning--happens constantly, from a baby learning a new word to a doctor classifying x-rays. We investigated people's ability to learn an artist's style by viewing their paintings (Kornell & Bjork, 2008a). Although spacing study trials over time virtually always enhances memory, we expected massing study trials to enhance inductive learning by allowing the studier to notice similarities in an artist's paintings. To our surprise, induction profited from spacing, not massing. The participants suffered from the same illusion as we did, however, and believed that massing had been more effective than spacing, even after taking the test. I have shown similar results with three-year-old children (Vlach, Sandhofer, & Kornell, in press) and older adults (in collaboration with Alan Castel). Download Poster: Categorizing paintings: The spacing effect in inductive learning (2007) (Best Poster, APA division 3, 2007.)
The basis of self-regulated study
A review of research on self-regulated study (see Son & Kornell, 2008) suggests that people try to study in a way that will optimize learning. In some circumstances, however, students seem to do the opposite. When I asked participants whether they wanted to learn by being tested or via presentation trials, they chose testing. But when asked which they thought would enhance their learning most, they said presentation. People appear to test themselves to diagnose their learning, without realizing that self-testing also enhances learning.Download Poster: Self-testing: A metacognitive disconnect between memory monitoring and study choice (2006)
Studying flashcards
Students face a wide variety of decisions when they study (Kornell & Bjork, 2007). One common real-world decision is whether to "drop" (i.e., put aside and stop studying) flashcards. We found that that students learned less when they were allowed to drop flashcards than when dropping was not allowed (Kornell & Bjork, 2008b). The students seemed to believe that if one can recall an item now, studying it again soon is pointless. Another decision students face is how many flashcards to study at once. Students often break a set of flashcards into small stacks. I have found that, because of the benefits of spaced learning, large stacks of flashcards produce more learning than do small stacks of flashcards. Even "cramming" with a small stack is less effective than daily study of a large stack. Yet my participants were under the illusion that small stacks were more effective. Download Poster: On the illusory benefits of easy learning: Studying small stacks of flashcards (2007)
Applying cognition research in the classroom
Throughout graduate school, I worked with middle-school students at a poorly-performing school in New York City's South Bronx. We created computer programs that implemented a number of effective study techniques simultaneously (e.g., spacing, testing, feedback), and found that doing so was tremendously effective in fostering long-term learning (Metcalfe, Kornell, & Son, 2007). We then turned to more specific investigations of which principles were effective (Metcalfe & Kornell, 2007). Region of Proximal Learning
Janet Metcalfe and I have proposed the Region of Proximal Learning (RPL) model of study time allocation (Metcalfe, 2002; Metcalfe & Kornell, 2003, 2005; Kornell & Metcalfe, 2006a). Our model predicts that when people do not have unlimited time to study, they will choose to study information that is close to being learned--often the easiest unlearned information--rather than studying the most difficult information. According to the model and the evidence, such choices are adaptive (Kornell & Metcalfe, 2006a). Once an item is in the process of being studied, the model makes a different prediction: People persist until the rate of learning drops too low to be efficient, and often persist longest on the most difficult items (Metcalfe & Kornell, 2005).Animal cognition
My research with animals mirrors my research with humans (Kornell, in press). Monkeys show a generation effect, which is similar to the generation and testing effects that enhance learning in humans (Kornell & Terrace, 2007). Monkeys also have metacognitive abilities that resemble human metacognition--they can make accurate confidence judgments about their memories, and request information when uncertain (Kornell, Son, & Terrace, 2007; Son & Kornell, 2005). Download Poster: Information seeking in rhesus macaques (2004)
Download Poster: Confidence judgments by rhesus macaques on a serial memory task (2003)
Last modified February 17, 2009
