Research reportSyllabic encoding during overt speech production in Cantonese: Evidence from temporal brain responses

- ERPs reveal the temporal dynamics of overt Cantonese disyllabic word production.
- Participants in the study name aloud pictures and ignore phonologically related/unrelated word distracters.
- Phonological relatedness between a picture and its distractor affects ERP amplitudes 500-650 ms post target onset.
- Phonological encoding in Cantonese speech production is operated incrementally, at the speed of 40 ms per syllable.

The time course of phonological encoding in overt Cantonese disyllabic word production was investigated using a picture-word interference task with concurrent recording of the event-related brain potentials (ERPs). Participants were asked to name aloud individually presented pictures and ignore a distracting Chinese character. Participants' naming responses were faster, relative to an unrelated control, when the distractor overlapped with the target's word-initial or word-final syllables. Furthermore, ERP waves in the syllable-related conditions were more positive-going than those in the unrelated control conditions from 500 ms to 650 ms post target onset (i.e., a late positivity). The mean and peak amplitudes of this late positivity correlated with the size of phonological facilitation. More importantly, the onset of the late positivity associated with word-initial syllable priming was 44 ms earlier than that associated with word-final syllable priming, suggesting that phonological encoding in overt speech runs incrementally and the encoding duration for one syllable unit is approximately 44 ms. Although the size of effective phonological units might vary across languages, as suggested by previous speech production studies, the present data indicate that the incremental nature of phonological encoding is a universal mechanism.