Introduction

Customers who are searching for adequate products and services in bricks-and-mortar stores are supported by human sales experts throughout the entire process, from preference construction to product selection. In online sales scenarios, such an advisory support is given by different types of recommender systems (Häubl and Murray 2006, Xiao and Benbasat 2007). These systems increasingly take over the role of a profitable marketing instrument, which can help to increase a company's turnover because of intelligent product and service placements. Users of online sales environments have long been identified as a market segment, and the understanding of their purchasing behavior is of high importance for companies (Jarvenpaa and Todd 1996, Thompson and Yeong 2003, Torkzadeh and Dhillon 2002). This purchasing behavior can be explained by different models of human decision making (Gigerenzer 2007, Payne et al. 1993, Simon 1955); we discuss selected models in the following sections.

Traditional models of human decision making are based on the assumption that consumers are making optimal decisions on the basis of rational thinking (Grether and Plott 1979, McFadden 1999). In those models, consumers would make the optimal decision on the basis of a formal evaluation process. One major assumption is that preferences remain consistent and unchangeable. In contradiction to those economic models, research has clearly pointed out that preference stability in decision processes does not exist. For instance, a customer who purchases a digital camera could first define a strict upper limit for the price of the camera, but because of additional technical information about the camera, the customer could change his or her mind and significantly increase the upper limit of the price.