The representation of segmental information: An fMRI investigation of the consonant–vowel distinction

Recent studies suggest that consonants and vowels are represented separately in cognitive/neural space. Much of the evidence comes from research on dysgraphia (for review, see Miceli & Capasso, submitted). In the first place, letter substitution errors preserve the consonant/vowel (CV) status of the target (e.g., cinema fi cirema or cinoma, but not cintma). Second, there are reports of selective impairment for consonants or vowels. Additional evidence comes from disorders of phonology, demonstrating the dissociability between consonants and vowels (Caramazza, Chialant, Capasso, & Miceli, 2000). However, while neuropsychological observations clearly support the C/V distinction, the anatomical correlates of such distinction are still unknown. We investigated this issue by means of an event-related fMRI experiment, with a grapheme verification task.