Perception of coda voicing: glottalisation, vowel duration, and silence

Recent research from Australian English has shown that glottalisation of vowels preceding coda stops results in increased perception of coda voicelessness. However, the addition of glottalisation results in a shorter portion of the vowel being modally voiced, raising the question of whether listeners may parse glottalisation as belonging to the coda rather than the preceding vowel. If so, listeners would perceive a shorter preceding vowel duration therefore increasing the perception of coda voicelessness.

This study thus compared listeners' coda voicing responses for words containing glottalised vowels with words containing vowels in which glottalisation was replaced with silence. The results suggest that both glottalisation and shorter vowel duration/longer coda closure duration result in increased voiceless percepts, but that listeners respond differently to these two conditions. The findings indicate that listeners are sensitive to glottalisation and utilise this as a cue to coda voicing rather than simply perceiving shorter modally voiced vowels.