Throughout their career architects collect an extensive record of
architectural cases, which they use as a source of inspiration and
knowledge during design. Being novices, student architects do not yet
have such a record. In order to compensate for this lack of knowledge,
teachers in architecture engage their students into realistic yet simulated
projects and introduce them to relevant architectural precedents for
these projects. Within the realm of AI, case-based reasoning (CBR)
stresses the importance of cases too. So far, however, applications that
flow from CBR research have rarely found their way into architecture.
The experiment that is reported in this article examines the conditions
under which CBR technology can be useful in architectural education.
The results show that in order for students to benefit from this
technology, it should supply cases that are closely related to the project
at hand. These results are consistent with psychological theories of
knowledge representation in novices.