A well known virtue of similarity-scaling techniques such as multidimensional scaling, Thurstonian modeling, and clustering is that they reveal hidden structure underlying psychological data. By applying these techniques, complex matrices of similarity data such as similarity ratings, identification confusions, and same-different errors can be efficiently described, summarized, and displayed, and a deeper understanding into the underlying basis of the similarity data can be derived.
Beyond describing and summarizing data, however, scaling techniques can be viewed as psychological models for the mental representation of interobject similarity. The main theme of this chapter is to review the role of similarity-scaling techniques as components in formal psychological models of perceptual and coqnitive processes.