Grant allows professor to expand innovative research


Why do geese fly in a V pattern, fish swim in schools and ants march in a line formation?

Arizona State University Professor Michelene Chi says that students tend to give a similar kind of incorrect answer to questions about these common events in the world, as well as to questions about many processes taught in science classes.

Chi joined the Mary Lou Fulton College of Education in August after working three decades at the University of Pittsburgh, where one of her projects was to try to understand why deeply held misconceptions inhibit children’s ability to learn about these and other emergent science processes and how misconceptions can be overcome.

“Dr. Chi’s work at the University of Pittsburgh’s Learning Research and Development Center (LRDC) and the Department of Psychology has had a large impact on the fields of Psychology and Education. She has made seminal contributions to our understanding of human expertise, child development, and learning,” said Timothy Nokes, assistant professor of psychology and learning sciences and policy at Pittsburgh. “Dr. Chi is a formidable intellectual force and colleague and ASU is very lucky to have such an innovating scientist and educator to count among its faculty.”

Chi received her doctorate in cognitive and developmental psychology from Carnegie-Mellon University and focused her research in Pittsburgh on how students learn, particularly in regard to science concepts. Her research has uncovered ways to optimize learning and to capture the structure’s underlying misconceptions.

“Her innovative research has pushed the field of Cognitive Science in novel and important directions through careful experimental investigations of topics such as the representation and organization of expert memory to determining the cognitive processes that underlie deep learning of new concepts,” Nokes said.

Chi has published more than 100 journal articles and book chapters on expertise and learning and has edited three seminal books on those subjects. She has received numerous awards for her research including a Spencer Fellowship, the Boyd R. McCandless Young Scientist Award, and the Chancellor’s Distinguished Research Award, among others. She has been elected as a Fellow to a number of associations including the American Psychological Association, the Association for Psychological Science, the Cognitive Science Society, and the Center for Advanced Study in Behavioral Sciences at Stanford University.

In the move to ASU with her husband, Kurt VanLehn, a professor in ASU’s School of Computing and Informatics, Chi made yet another transition in her career.

“It was time for a change, and it’s a very significant change for me to move from the Department of Psychology at Pitt to the College of Education here,” she explained. “My current research is more oriented toward how students learn, and this issue has a practical application for instruction.”

As an experimental psychologist in Pittsburgh, Chi developed a module to teach middle school students to differentiate their ways of thinking about emergent processes using examples like the school of fish to understand other scientific processes such as natural selection. She plans to couple the project with a tutorial technique she developed for her newest research endeavor in cognitive learning funded by a $500,000 single investigator Major Research Grant from the Spencer Foundation.

In a paper recently published in Cognitive Science, Chi showed that pairs of students can learn to solve physics problems by watching and listening to another student (a tutee) being tutored. The new Spencer-funded project will further investigate how the observing-students can learn the scientific content from overhearing the dialogues between the tutor and the tutee, a practice commonly used by generations of people in various cultures to acquire skills or trades through observation in apprenticeships. While people can learn to weave or forge iron through observing an expert, she said, there has been no evidence showing that they can learn complex skills such as how to troubleshoot or solve a difficult physics problem through modeling because the thinking processes aren’t visible. One could make a model’s or expert’s thinking visible by asking an expert to talk out loud, but novices or students may not grasp the concepts if the experts can’t properly explain what they are doing or if they explain it at too high a level.

“Although there is lots of evidence showing that people can learn physical skills from observing, our study is the first to show that people can learn cognitive skills such as problem solving or troubleshooting by observing and overhearing the dialogue between a novice and an expert,” Chi noted. “I’ve designed a new instructional method that allows students to learn just as well from observing as being tutored. The technique we implemented was very low tech. We simply videotaped a tutor helping a student-tutee to solve a problem, and then we showed the videos to pairs of student-observers. We found not only that the student-observers can learn as well as the average tutees, but that the observers learned even more when the tutee is a good student.”

There are two explanations for why this instructional method facilitates observers’ learning. One explanation is that the pairs of student-observers can learn as well as the student being tutored because the paired observers could interact constructively with their partners. “In prior work, I have shown that student-tutees benefited from tutoring not necessarily because the tutors excelled at tailoring their explanations to their tutees, but because the tutees had more opportunities to interact with the tutor in an active and constructive way,” she said.

She recently submitted a research proposal to the National Science Foundation to further understand how students can learn from observing and overhearing dialogue. The proposal suggests that poor students can learn from observing this tutor-student dialogue because they learn to model a good student’s interactive learning behavior. “It has important application because it implies that we don’t need expensive one-to-one tutoring. Not just one student learns, but many students can learn by watching this videotape.”

 “If Dr. Chi and her colleagues can develop a theoretical understanding of what aspects of the teacher/student interactions and dialog cause student learning than we can incorporate those features into instruction and pedagogy,” Nokes commented. “Learning by observing other students learn while engaged in active problem solving has broad implications for distance-learning as technological advances and societal changes make the internet and on-line courses a larger and larger part of student learning repertoire.”

Chi’s research has applications for distance and online learning. It suggests that a more effective online delivery of materials would be to present a teacher-student dialogue, rather than present a teacher’s lecture and explanations in a monologue. There is concern in online courses that students who lurk in the background rather than contribute to class discussions aren’t learning the content.

“Some students just don’t contribute, but there’s a hint from this work that they can still learn by watching the contribution of others, even if they don’t make contributions themselves,” Chi said.