MULTIDIMENSIONAL-SCALING OF HIERARCHICAL SORTING DATA APPLIED TO FACIAL EXPRESSIONS

A new algorithm for multidimensional scaling analysis of sorting data and hierarchical-sorting data is tested by applying it to facial expre ssions of emotion. We construct maps in ''facial expression space'' fo r two sets of still photographs: the I-FEEL series (expressions displa yed spontaneously by infants and young children), and a subset of the Lightfoot series (posed expressions, all from one actress). The analys is avoids potential artefacts by fitting a map directly to the subject 's judgments, rather than transforming the data into a matrix of estim ated dissimilarities as an intermediate step. The results for both sti mulus sets display an improvement in the extent to which they agree wi th existing maps. Some points emerge about the limitations of sorting data and the need for caution when interpreting MDS cofigurations deri ved from them.