How do Different Types of Synesthesia Cluster Together? Implications for Causal Mechanisms.

It is unclear whether synesthesia is one condition or many, and this has implications for whether theories should postulate a single cause or multiple independent causes. Study 1 analyses data from a large sample of self-referred synesthetes (  =  2,925), who answered a questionnaire about   =  164 potential types of synesthesia. Clustering and factor analysis methods identified around seven coherent groupings of synesthesia, as well as showing that some common types of synesthesia do not fall into any grouping at all (mirror-touch, hearing-motion, tickertape). There was a residual positive correlation between clusters (they tend to associate rather than compete). Moreover, we observed a "snowball effect" whereby the chances of having a given cluster of synesthesia go up in proportion to the number of other clusters a person has (again suggesting non-independence). Clusters tended to be distinguished by shared concurrent experiences rather than shared triggering stimuli (inducers). We speculate that modulatory feedback pathways from the concurrent to inducers may play a key role in the emergence of synesthesia. Study 2 assessed the external validity of these clusters by showing that they predict performance on other measures known to be linked to synesthesia.