What’s in a name? A large-scale computational study on how competition between names affects naming variation

Abstract

Different speakers often use different names to refer to the same entity (e.g., “woman” vs. “tennis player” for a given woman playing tennis). We study how visual typicality affects variation in naming behavior. We use a novel computational approach to estimate visual typicality from images, and analyze a large dataset containing naming data for realistic images. In contrast to previous work, we take into account the visual properties of both the object and the scene in which it appears; and factor in multiple candidate names. We show that visual typicality mediates competition between candidate names: high competition, induced by the relationship between the visual properties of the object and the visual representations associated to names, predicts higher naming variation. On a methodological level, we demonstrate the potential of using large-scale datasets with realistic images in conjunction with computational methods to shed light on how people name objects.