Visuo-spatial representations of the alphabet in synaesthetes and non-synaesthetes

Visuo-spatial representations of the alphabet (so-called ‘alphabet forms’) may be as common as other types of sequence–space synaesthesia, but little is known about them or the way they relate to implicit spatial associations in the general population. In the first study, we describe the characteristics of a large sample of alphabet forms visualized by synaesthetes. They most often run from left to right and have salient features (e.g., bends, breaks) at particular points in the sequence that correspond to chunks in the ‘Alphabet Song’ and at the alphabet mid-point. The Alphabet Song chunking suggests that the visuo-spatial characteristics are derived, at least in part, from those of the verbal sequence learned earlier in life. However, these synaesthetes are no faster at locating points in the sequence (e.g., what comes before/after letter X?) than controls. They tend to be more spatially consistent (measured by eye tracking) and letters can act as attentional cues to left/right space in synaesthetes with alphabet forms (measured by saccades), but not in non-synaesthetes. This attentional cueing suggests dissociation between numbers (which reliably act as attentional cues in synaesthetes and non-synaesthetes) and letters (which act as attentional cues in synaesthetes only)