Towards a taxonomy of abstract concepts

A large portion of human knowledge comprises abstract concepts (e.g., “love” and “justice”) that lack perceivable properties. Due to their lack of such properties, abstract concepts have historically been treated as an undifferentiated category of knowledge in the psychology and neuropsychology literatures. We collected a large set of implicit judgments about a set of fifty abstract nouns using an odd-one-out similarity task to determine the representational space of the concepts. We then identified category boundaries within the representational space using a clustering procedure that required categories to replicate across two independent data sets. In a separate experiment, we used automatic semantic priming to further validate the categories and to demonstrate that abstract category boundaries based on explicit judgments are inadequate for mapping their representational space. These results demonstrate that abstract concepts have an intrinsic category structure that can be characterized beyond their negative relation to concrete concepts.