Sign Theory: A Non-Extensional Theory of Preference

Abstract

A non-extensional theory of preference, called Sign Theory, is proposed and tested. Sign Theory considers preference construction as an evidence-building process, in which local valuation of evidence combines additively to form a global judgment. At the heart of this treatment is the notion that local evaluation generates positive or negative valuation outcomes (signs). Thus, signs are building blocks for preference construction. Study 1 confirmed the notion that valuation processes are based on signs derived from reference points. Studies 2 and 3 demonstrated the robustness and importance of the sign-frequency effect, which shows that the attractiveness of an option is positively related to the number of the positive signs, but negatively related to the number of negative signs. Studies 4 and 5 were designed to reduce the sign-frequency effect. Study 4 provided information about attribute weights to help participants appreciate the part-whole relationship between partitioned attributes and summarized attributes. In Study 5, decision makers were clearly informed about the two display formats of the same information. Nevertheless, in both studies, the presence of sign-frequency effects remained evident.

Keywords: Sign Theory, reference point effects, sign-frequency effects, preference construction