Phonemic processing in compensatory responses of French and English speakers to formant shifted auditory feedback

Past studies have shown that speakers modify their vowel formant production when auditory feedback is altered in order to make the feedback more consistent with the intended sound. This behavior was thought to minimize acoustic error overall; however, Mitsuya et al. (2011) showed different magnitudes of compensation for altered F1 across two language groups depending on the direction of perturbation. Their results seem to reflect how the target vowel is represented in relation to other vowels around it. From this observation, they proposed that compensation is to maintain perceptual identity of the produced vowel, requiring some phonological processes for error reduction. Yet, the results might have been specific to the language groups examined, and/or unique to F1 production. To generalize Mitsuya et al.’s hypothesis, the current study examined 1) different language groups and 2) F2 production. We compared compensatory behavior of F2 for /ɛ/ among French speakers (FRN) and English speakers (ENG) with decreased F2 feedback. With this perturbation, the feedback sounded like /œ/, which is phonemic in French but not in English. The preliminary data suggest that FRN compensated in response to smaller perturbations and showed greater maximum compensations than ENG.