Set-Fit Effects in Choice

In five experiments, we show how the “fit” of an item with a set of similar items affects choice. We find that people have a notion of a set that fits together—one where all items are the same, or all items differ, on salient attributes. One consequence of this notion is that in addition to preferences over the set’s individual items, choices take set-fit into account. This leads to predictable shifts in preferences when choosing between different sets as opposed to individual items, sometimes even resulting in people choosing normatively inferior options over superior ones. Even though many products are manufactured and sold as grouped sets, virtually no research has investigated how the composition of such sets influences choice. Other literature, however, suggests that the composition of a set could have strong effects on preference and choice. Early research by the gestalt psychologists (Wertheimer, 1923) shows that people categorize and group stimuli rapidly and automatically, and that some sets of stimuli are grouped together more easily than others. Furthermore, we know from research on fluency, that things that are processed more easily are liked better (e.g., Winkielman & Cacioppo, 2001). Therefore, it could be expected that sets that are easier to group, will be liked better. Similar arguments can be found in work by philosophers on aesthetics. For example Klintsch (2012) argues that the lower the complexity of a stimulus, the more it will be liked. Even though these claims have not been tested empirically, cognitive psychologists (e.g., Garner, 1970) have investigated the liking of patterns and found a strong negative relation between the complexity and liking. Taken together, this suggests that sets of products that are low in complexity will be liked better than sets that are more complex. More specifically, we expected that sets in which all products are either all-similar or all-different on every salient attribute, would be liked better than sets which are not. A pre-test in which participants rated sets of geometric shapes confirmed our expectations. Participants consistently rated sets which were all-similar or all-different on shape and color as much more liked than sets which were not.