Our aim is to create a neuroimaging repository consisting of 100+ synaesthetic brains that will be made freely available to researchers around the world. By using publicly available protocols, other groups are encouraged to add to this resource (as well as access it). This would be the first repository of its kind for synaesthesia. We will use state-of-the-art international imaging protocols (the Human Connectome Project, HCP). By using these protocols, we can ultimately make direct comparisons between synaesthetic brains and the brain images from a wide variety of clinical (e.g. in autism) and normative samples, via freely available datasets also acquired using the same HCP protocols. In these first stages of the project, we will create candidate brain-based biomarkers based on structural and functional (resting state) MRI data. These will be used to predict group membership (synaesthete v. non-synaesthete) using machine learning procedures with cross-validation (in the development phase) and direct replication of the most promising biomarkers (in the demonstration phase). A detailed behavioural assessment will be carried out including clinical profile, personality, and cognitive performance. The characteristics of the synaesthesia (e.g. types possessed) will be documented in detail. These differences will be linked to the neural biomarkers.