Do Skewed Base-Rates Affect Advertising Effects on Brand Attitudes? On the Role of Pseudocontingencies in Evaluative Conditioning

• Main Authors: Hütter, Mandy, Kutzner, ian, Fiedler, Klaus
• Format: Conference Proceeding
• Language: English
• Subjects: Attitudes; Brands; Effects
• ISSN: 0098-9258

Evaluative conditioning (EC) serves as the prototypical paradigm to study advertising effects on product and brand attitudes (Sweldens, Van Osselaer, and Janiszewski 2010; Walther, Nagengast, and Trasselli 2005). In a typical EC procedure, an (often initially neutral) conditioned stimulus (CS; e.g., a product) repeatedly co-occurs in close spatio-temporal proximity with a valenced unconditioned stimulus (US; e.g., beautiful landscapes, smiling people, celebrity endorsers). As a result, evaluations of the CS change in the direction of the US-a phenomenon known as the EC effect. The investigation of EC has been mainly concerned with the effects on a specific CS (e.g., a product) in encapsulated pairing episodes. The conditioning of stimulus features across pairing episodes has received little attention. Recent findings demonstrate, however, that the pairing with valenced stimuli does not only have the potential to change attitudes towards individual CSs, but also towards cues that are shared by many CSs (Hütter, Kutzner, and Fiedler 2014). One shared cue in the consumer context is the brand by which an array of products is put on the market. In the paradigm adapted for the present research, participants make more positive experiences with products of one brand than with another brand. That is, there is a statistical contingency between brand and valence of experience. However, this contingency is not perfect. That is, consumers also make negative experiences with products of the positive brand and positive experiences with products of a negative brand. One can thus disentangle the evaluation of the brands from the evaluation of single products. That is, if many positive experiences with one brand increase the liking even in products that led to negative experiences (and vice versa), it can be assumed that the evaluation of the brand exerts a direct effect on the evaluation of the products that carry its brand. This is precisely what Hütter et al. (2014) have shown across several experiments. In the present research, we examine the impact of an intriguing contingency illusion that is caused by the joint skew of two variables irrespective of actual contingencies. This illusion is called a pseudocontingency. A pseudocontingeny constitutes a logically unwarranted inference that relates a rare (frequent) cue value to the rare (frequent) criterion value (i.e., valence; Fiedler, Kutzner, and Vogel 2013). For instance, if consumers sample many products of one brand