How People Compare an Item’s Placement in Two Alternative Categories

Abstract

This paper describes an experiment examining how people compare two alternative categories and decide whether a given item would be better placed in one category or the other. The experiment asks whether an item’s comparative classification in one of two alternative categories is related to the difference between the item’s typicality as a member of those two categories. The alternative categories compared in the experiment are single categories such as “bird” and combined categories such as “pet bird”. The experiment found that an item’s comparative classification in two such categories could not be predicted from the item’s typicality in those categories: some items had similar typicality in both single and combined categories but were judged to be much better placed in the combined rather than the single category. This suggests that some other factors such as category specificity or informativeness may be involved in comparative classification.