The usefulness of the averaging rule in consumer studies: A partial replication with an eye at neuroeconomics

Abstract

IntroductionInformation Integration Theory models how different pieces of information are combined by people into unified judgments. Its contribution to decision-making and to consumer studies distinctly relies on a set of empirically established algebraic models for reflecting the cognitive structure of judgments and decisions (cognitive algebra). Among these, the averaging model has particular properties, allowing for operationally distinguishing between the elusive notions of psychological value and importance. ObjectivesThe main goal of the paper is to support the applicability and robustness of the averaging model in consumer studies, by partially replicating a published study that documented averaging. Additionally, it tentatively explores potential contributions of the averaging model to neuroeconomics, by simultaneously recording ratings and electroencephalographic (EEG) responses. MethodsTwenty-five participants rated on a satisfaction scale (format 0-20) a set of experimental conditions arising from the full factorial combination of three levels of sneakers’ branded models and three levels of purchase prices, expanded with conditions in which brands were presented in isolation (sub-design). Simultaneous electrophysiological monitoring was performed with an EEG mask (international system 10-20, referenced to the mastoid apophyses). Segmentation and processing of evoked response potentials were done in Matlab. Both ratings and magnitudes of event-related potentials (ERP) components were analyzed through inspection of the factorial plots with the support of repeated-measures ANOVAs. ResultsThe rating responses revealed an averaging model, with equal weighting in a cluster of subjects and differential weighting in another, but more overall importance of the branded model variable in both cases. The only tentative suggestions of relevant neural correlates involved the LPP and P300 components of the ERP and a reduced number of EEG channels (P4, P3, Fz and T3). ConclusionsThe averaging model is robustly applicable and useful in consumer studies, both for measurement (value and importance) and consumer segmentation (cognitively based clusterization) purposes. While calling attention to the later components of the ERP at specific topographies, the paucity of EEG results suggests that ERP peak amplitude may not be a useful measure for the purpose. Apart from that, a functional measurement approach to neuroeconomics seems feasible in principle and an explorable path in the future.