KEER2010, PARIS | MARCH 2-4 2010 INTERNATIONAL CONFERENCE ON KANSEI ENGINEERING AND EMOTION RESEARCH 2010 MODELS FROM PSYCHOLOGY AND MARKETING APPLIED TO KANSEI ENGINEERING Richard Gonzalez *a , Sookyung Cho a , Tahira Reid a , and Panos Papalambros a a University of Michigan, USA ABSTRACT This paper considers new models from Psychology and Marketing that extend the general Kan- sei approach. We sketch a framework that views design as understanding (1) the consumer’s representation of the product space, (2) the consumer’s choice function, (3) the designer’s repre- sentation of the design space, (4) the designer’s choice function, and a connection between those four properties. This framework helps organize some existing methods and points to areas ripe for development of new models. Keywords: Psychology, Choice, Multivariate Analysis 1. INTRODUCTION Kansei methodology provides a scientific basis for relating user-defined characteristics to engi- neering variables. The ability to convert user-defined concepts to concrete engineering attributes is an important aspect of any analytic design process that uses inputs from the end-user in a quan- titative way. Kansei engineering uses several techniques from the social sciences and statistics, including multidimensional scaling (MDS), the semantic differential procedure and advanced re- gression techniques. In this paper, we bring modern methods from the social sciences, such as discrete choice models and mixture models, and methods from engineering, such as functional de- pendence table analysis, to relate user-defined characteristics to engineering attributes. We show how Kansei and related methods can model consumer choice and heterogeneity across consumers. Traditional Kansei methodology has focused on the semantic differential procedure, which is an important tool from psychology for understanding meaning and concepts as defined by users. The typical analysis of semantic differential data uses techniques such as linear regression and * Corresponding author: Department of Psychology, University of Michigan, Ann Arbor, MI 48108. gonzo@umich.edu.