Multivariate analysis of sensory data: a comparison of methods

Publisher Website
Google Scholar
Abstract
Multivariate statistical analyses are designed to simplify the relationships that exist within a complex array of data. Within the chemical senses, multivariate methods have been used to address a number of problems, including the classification of neurons and the description of stimulus relationships. This paper presents a conceptual discussion of hierarchical cluster analysis, factor analysis, and non-metric multidimensional scaling, emphasizing how each of these procedures operates, how each is interpreted, and how they relate to one another. These techniques are applied to the stimulus relationships within a single set of chemosensory data. The results are used to provide a conceptual understanding of these multivariate procedures and to illustrate the similarities and differences among them.