Similarity, interactive activation, and mapping.

Abstract

The question of What makes things seem similar? is important both because of similarity's pivotal role in theories of cognition and because of an intrinsic interest in how people make comparisons. This dissertation argues that psychological assessments of similarity involve more than listing the features of the things to be compared and comparing the lists for overlap. Instead, when comparing things that are hierarchically or propositionally structured, the parts of one thing must be aligned or placed in correspondence with the other thing's parts. In eleven experiments on human subjects this need for structural alignment is exhibited in similarity ratings, same/different responses, ease of comparison judgments, and feature match identifications. Three quantitative models of similarity are developed and applied to the empirically obtained data. The model with the best overall fit assumes an interactive activation process whereby correspondences between the parts of compared things mutually and concurrently influence each other. An essential aspect of this interactive activation model is that matching features increase similarity more if they belong to parts that are placed in correspondence. In turn, parts are placed in correspondence if they have many matching features in common, and if they are consistent with other developing correspondences. Predictions made by this model are empirically confirmed and the model is applied to results in problem solving, similarity comparisons, perception, and analogical reasoning.