Although the ability to communicate through language about abstract concepts lies at the heart of what it means to be human, our knowledge of how abstract word meanings are represented and processed is extremely limited. In this paper we show that neither of the two dominant accounts (dual coding theory and the context availability model) put forward in order to explain differences between concrete and abstract words fully captures processing (and hence representational) differences between the two types of word meaning. Using lexical decision data, we show that this is, at least partly, because in both accounts abstract words are considered to be unrelated to experiential information. We show instead that there is one type of experiential information, namely affective information, which plays a crucial role in the processing and representation of abstract concepts: affect explains a residual advantage for abstract words, when variables such as imageability and rated context availability are held constant. We discuss our results with respect to embodied theories of cognition and language representation.