It's not what you hear but how often you hear it: on the neglected role of phonological variant frequency in auditory word recognition.

Recognition of a frequently heard spoken word variant in American English (flapping) was investigated in a phoneme identification experiment. Listeners identified the initial segment (b or p) of word-nonword continua (e.g., pretty-bretty) that was embedded in either a flap or a [t] variant carrier word (e.g., preDy-breDy or preTTy-breTTy). The results showed more identification responses forming a real word when the to-be-identified speech sound occurred in the more frequently experienced flap carrier. These results support the claim that lexical representation of spoken words includes the flap variant. Listeners do not recode the flap variant into an underlying /t/ version but recognize the flap, in its surface form, via a preexisting representation in lexical memory.