A perceptual pathway for voicing-conditioned vowel duration

Abstract

When codas and vowels are cross-spliced, vowels originally produced with voiced codas are perceived as longer than vowels of the same duration produced with voiceless codas. The spliced coda has the opposite effect: Vowels presented with voiced codas are perceived as shorter. To explain what characteristics make vowels produced with voiced codas sound longer than vowels produced with voiceless codas, four experiments tested how acoustic correlates of voicing affect English speakers' perception of vowel duration. Vowels were manipulated in a ten-step duration continuum, and listeners categorized each vowel as 'long' or 'short.' Study 1 tested effects of vowel height categories (/ae, ε, ɪ/) and within-category F1. Study 2 tested effects of intensity contour. Study 3 tested effects of spectral tilt. Perceived vowel duration increased with vowel height and rising intensity. Perceived vowel duration decreased with falling intensity and with higher spectral tilt. There was no effect of within-category F1. Study 4 confirmed that the effect of the original coda environment is not specific to English stimuli. The effects of spectral tilt and intensity contour on perceived duration provide a possible perceptual pathway for the development of voicing-conditioned vowel duration; coda voicing influences these characteristics and they in turn influence perceived vowel duration.

How do headphone checks impact perception data?

“…Note that unlike the previous two studies, including these random slopes was possible, because the correlation between the intercept and the random slopes was low. Most of the same effects found by Sanker (2020) were observed regardless of participants' performance on the headphone checks. Listeners were more likely to identify vowels with longer duration as being long, less likely to identify vowels as long when they had originally been produced with a voiceless coda, and more likely to identify vowels as long when they were presented with a voiceless coda.…”

“…Participants who passed the Huggins check had a larger effect of spectral tilt on perceived vowel duration, replicating the previously observed effect; participants who failed the Huggins check did not exhibit a significant effect. This expected effect depends on the relationship between frequency and perceived loudness and between loudness and perceived duration; perceived intensity increases with frequency, so a sound with lower spectral tilt is likely to be perceived as louder and subsequently longer (Sanker, 2020).…”

“…Task 3 tests the effect of spectral tilt on perceived vowel duration. Based on previous work (Sanker, 2020), listeners should be more likely to identify vowels as long when they have a lower spectral tilt. Actual duration should also be a predictor of perceived duration, and listeners should compensate for the voicing of the coda, more often identifying vowels as long when they are presented with a voiceless coda.…”

A Reanalysis of the Voicing Effect in English: With Implications for Featural Specification

“…However, there seems to be little evidence to support the latter hypothesis, and enhancement explanations are unable to account for why it is preceding vowel length, and not obstruent length, degree of voicing, presence of audible release, or aspiration that are used to enhance the contrastiveness of the obstruents themselves. More recently, Sanker (2020) has proposed an explanation based on the interaction between acoustics and articulation: a subset of the features in the vowel that are affected by the voicing of the following obstruent (spectral tilt, and intensity contour) also affect perception of vowel duration, presumably for unrelated articulatory reasons. Thus, in the presence of those cues, independent of the presence of a following obstruent, listeners perceive the vowel as being longer/ shorter than some baseline duration.…”