Debate Over How Children Learn is Reignited
An article in a well-known research journal has rekindled debate over how children learn and solve problems.
Writing in this month's issue of Educational Researcher, three leading cognitive-science researchers--John R. Anderson, Lynne M. Reder, and Herbert A. Simon--offer a pointed critique of a movement in their field known as situated learning.
Proponents of situated learning contend that much of what people learn is specific to the environment in which they learn it. For example, students often fail to apply what they learn in school in the workplace or in other situations outside the classroom.
Although situated learning is far from being a household term, the movement has been a driving force in school reforms--particularly in math education--that are aimed at grounding classroom learning in real-world and social contexts.
However, Mr. Anderson and his co-authors say many of the more sweeping claims made by situated-learning theorists are "misguided" and "inaccurate" when scrutinized against existing research.
"One of the reasons we wrote this article was to point out the fact that there are reasons to be more cautious and that we shouldn't take on these ideas whole cloth," Mr. Anderson, a professor of psychology and computer science, said last week in an interview. All three authors teach at Carnegie Mellon University in Pittsburgh.
The article is expected to generate more controversy in an already fractious field.
"It's a pretty significant paper given who's written it and given the kinds of issues that are being debated in the educational-research community right now," said John T. Bruer, the president of the James S. McDonnell Foundation, a St. Louis-based philanthropy that supports experiments in cognitive science.
Mr. Anderson, for example, is known for his theory of cognition and his work on "intelligent" computer tutors for students. Ms. Reder has done prominent research on memory and skills acquisition, and Mr. Simon, a recipient of the Nobel Memorial Prize in Economic Science, is a pioneer in the field of cognitive science.
Some observers said the views the three researchers expressed could have implications for many of the education reforms taking place in schools nationwide.
"If they are correct, educational reformers should no longer feel compelled to promote teaching strategies and approaches to learning in schools that, history tells us, are exceedingly difficult to implement and almost impossible to sustain," writes Robert Donmoyer, the journal's features editor. But the authors say their intent was not to call for an end to those reforms but only to caution against "overdoing" them.
The idea that learning and problem-solving cannot be separated from the cues, tools, and people in an individual's environment is the central claim of situated-learning proponents, but it is not the only one. The Carnegie Mellon researchers also attribute to this school of thought the idea that people do not readily carry over or "transfer" the knowledge and skills they gain from one context to another. In one classic study, for example, a group of homemakers who could easily calculate bargains in the supermarket fared much worse at doing the same kinds of calculations on paper-and-pencil tests.
Situated-learning theorists also contend that abstract instruction, such as the teaching of mathematical formulas, produces little payoff in real learning. Ideally, they say, learning should take place in complex, social environments, such as apprenticeships or "learning communities."
Not all supporters of situated-learning theory maintain that all learning must follow these principles, but some do.
In their paper, the Carnegie Mellon researchers cite studies that contradict such blanket statements, as well as those like the homemakers' study that support them. "In general, situated learning focuses on some well-documented phenomena in cognitive psychology and ignores many others," the authors conclude.
"While cognition is partly context-dependent, it is also partly context-independent," the researchers say. "While there are dramatic failures of transfer, there are also dramatic successes; while concrete instruction helps, abstract instruction also helps; while some performances benefit from training in a social context; others do not."
For example, the authors say, a student who wants to play a violin in an orchestra would find it difficult if all the practices took place with the rest of the orchestra. The opposite is also true: Practicing in isolation would not prepare a student for playing in an orchestra.
"We just want to defend a place for individual learning and abstract instruction and to say that, it, too, has a place in education," Ms. Reder said last week in an interview.
James Greeno, a Stanford University professor who has researched situated learning, said he does not entirely disagree with the Carnegie Mellon researchers' points.
"The big change with the situative view is trying to see how that private practice--going off in a music room--is connected to broader aspects of social practice," said Mr. Greeno, a principal scientist at the Institute for Research on Learning in Menlo Park, Calif. "Traditionally, in mathematics, it's been the whole practice."
But, Mr. Greeno and other proponents of situated learning say, the authors are using a different set of lenses to analyze the debate.
"The studies they cite are done in a cognitive paradigm," said Allan M. Collins, a professor of education and social policy at Northwestern University in Evanston, Ill. In such studies, researchers try to hold variables such as students' motivation constant, he said.
"In a real classroom, you want to see what you can do to make the kids interested enough so they can learn something," said Mr. Collins, who considers himself a moderate in this debate. Research in situated learning tries to account for such social factors, he said.