The concept of neurodiversity was proposed in the late 1990s by Judy Singer. Since then the meaning of the term has been hotly contested, but it is often associated with the accommodation of ‘natural variation’ in the human genome. Here I explore the various meanings of the term, and its usefulness for research. I suggest that it may be helpful to move on from associations with ‘natural’ variations, and moreover that the concept may be best thought of as having an epistemic utility for research.