Indifference Curves that Travel with the Choice Set

Abstract

Normative and descriptive theories of choice assume that absolute attribute values are key determinants of preferences. Building on research on the compromise effect, we propose that absolute attribute values often play a relatively minor role in the construction of preferences and consumers rely primarily on the relative positions of product alternatives in particular choice sets. Consistent with this proposition, we present the results of three studies which show that consumers can predict with surprising accuracy the likelihood of selecting a compromise option based on the product category and relative positions of options on attributes, without knowing the absolute attribute values (i.e., the location of the set in the attribute space). We also show that providing information about the range of absolute values does not affect consumers' propensity to rely on the relative positions of options in choice sets. These results support the notion that, when constructing their preferences, consumers often retrieve global preference elements (e.g., the relative importance of attributes), and apply them locally with relatively little sensitivity to the location of the set - i.e. they behave as though their indifference curves "travel" with the choice set.