AbstractThis study examines how variation in production is perceived and then (re)interpreted by listeners, thus providing the link between phonetic variation and sound change. We investigate whether listeners can detect the nasal leak that may accompany utterance-initial voiced stops in Spanish, and reinterpret it as a nasal segment. Such reinterpretation may account for a number of sound patterns involving emergent nasals adjacent to voiced stops in oral contexts. Oral pressure, nasal/oral airflow, and audio were recorded for utterance-initial /b d p t/ produced by 10 Spanish speakers. Tokens showing different degrees of nasal leak (nasal C, maximum, medium, and no nasal leak) were placed intervocalically, where both /C/ and /NC/ may occur. The stimuli were presented to Spanish listeners for identification as /VNCV/ or /V(C)CV/. Identification results indicate a higher number of VNCV responses with incremental changes in nasal leak in voiced but not voiceless stimuli. Reaction time analysis showed shorter latencies to nasal identification for larger velum leak stimuli. The results suggest that listeners can `hear' the nasal leak and fail to relate it to voicing initiation, interpreting a nasal segment. Thus a gesture aimed at facilitating voicing initiation may be interpreted as a new target goal.
The optimal time window of visual–auditory integration: a reaction time analysis
