Abstract

Theories of analogical reasoning have viewed relational structure as the dominant determinant of analogical mapping and inference, while assigning lesser importance to similarity between individual objects. An experiment is reported in which these two sources of constraints on analogy are placed in competition under conditions of high relational complexity. Results demonstrate equal importance for relational structure and object similarity, both in analogical mapping and in inference generation. The human data were successfully simulated using a computational analogy model (LISA) that treats both relational correspondences and object similarity as soft constraints that operate within a limited-capacity working memory; but not with a model (SME) that treats relational structure as pre-eminent. Analogies provide a valuable source of new inferences and a means of expanding knowledge. Analogical reasoning is generally viewed as involving four major subprocesses: (1) retrieving an appropriate source analogue from long-term memory to compare with a novel target analogue, (2) mapping elements of the two analogues, (3) making inferences about the target as a function of its mapping to the source, and (4) using the source and target together to induce a more general schema or rule (e.g., Carbonell, 1983;

Citations