A. Didierjean et E. Cauzinillemarmeche, REASONING BY ANALOGY - IS IT SCHEMA-MEDIATED OR CASE-BASED, European journal of psychology of education, 13(3), 1998, pp. 385-398
One of the ''classical'' ways of learning consists of studying example
s of already solved problems. In two experiments, we analyzed the degr
ee of abstraction of the knowledge used by ninth grade students to sol
ve algebra problems after studying worked examples. The results showed
that there are two processes underlying reasoning by analogy, one tha
t uses abstract knowledge and another that involves case-based reasoni
ng. Both experiments pointed out interindividual differences in the po
pulation under study: when given examples, some subjects seem to extra
ct the structure of the solving process by comparing the worked exampl
es, while others focus more on the specifics of each example. To these
two processes correspond two levels of transfer: correctly solve prob
lems that have the same structure as the examples, regardless of how s
imilar they are, or be better at solving problems that resemble the ex
amples the most. Experiment 2 used a dual-task paradigm to show that s
ome subjects implement both processes, in which case th mental load is
greater. This experiment also showed that both processes can lead to
the long-term acquisition of the principles behind the examples.