The mirror effect in online survey data: Evidence and implications for marketing theory and strategy

Abstract

This research reveals the presence, in online survey data, of a key pattern documented in psychology lab research: the Mirror Effect. The Mirror Effect occurs when unfamiliar stimuli are unexpectedly recognized as accurately as familiar stimuli, or more accurately. Using a set of familiar and unfamiliar words (as determined by lexical frequency), we first report that we can robustly replicate psychology lab research in an online survey, detecting the Mirror Effect. We then apply the same analytical approach to surveying consumer recognition of everyday brands (supermarkets, banks and car brands). We find that unfamiliar brands can be recognized with the same level of accuracy as familiar brands, and this effect is stronger than age and gender memory biases present in the data. However, we detect a boundary condition for branded apps, which are extremely unfamiliar brands competing in highly fragmented marketplaces, so very few get downloaded or used. For these digital brands, we find a Concordant Effect, as most respondents find it difficult to recognize highly unfamiliar branded apps (i.e., those with fewer than 5000 downloads). The Mirror Effect re-emerges for highly experienced app users. These results demonstrate the implications of a generalizable empirical pattern from cognitive psychology for branding and advertising theory. The outcomes of this research also translate into practical guidelines for brand performance measurement via online surveys, mitigating recognition memory bias for the development of marketing strategies based on more accurate interpretation of empirical evidence.