Cognitive Science Seminar Series

Linguistics graduate student Wil Gonzales will present proposed work on stop voicing in Hokkien. Title: Age and ethnicity in the perception of Hokkien stop voicing: A proposal.

Please visit the Seminar Series website for Zoom access information.

ABSTRACT

Can listeners’ knowledge of socially-condition linguistic variation influence how they perceive the speech of talkers from different social groups? Can the listeners’ perception be conditioned by their own social background? In this talk, I propose a study that hopes to answer these questions in the Sino-Philippine context. I focus on the potential effect of age (young vs. old) and ethnicity (Lannang vs. Mainlander) on the perception of Hokkien stop voicing in individuals with Southern Chinese heritage (SCH) - the Lannangs and the Mainlanders .
I plan to conduct a perception experiment to be administered over PsychoPy. I will subject my participants to a two-alternative forced choice task, where they will be first exposed to stimuli (created using a Hokkien voiced-voiceless stop continuum) and then decide which word they believe they heard from two alternatives on the computer screen. These are minimal pair Hokkien words that differ only by the voicing of the stop onset – /gau35/ ‘smart’ and /kau35/ ‘monkey’. In the experiment, participants will decide on the word based on audiovisual stimuli (social manipulation).
At the level of the talker, I hypothesize that individuals with SCH will differ in their perceptual judgments of stimuli varying along a voiced-to-voiceless stop continuum in ways that are sensitive to expected social characteristics of the talker. I expect there to be an asymmetrical effect: only talker ethnicity and not age will affect the judgments, as ethnicity is ‘socially marked’ in the Manila SCH community whereas age is not (McGowan and Babel 2019; Gonzales 2021). At the listener level, I hypothesize that the age and ethnicity of the listener can predict the perceptual judgments that they make. Young SCH listeners, having more exposure and experience with dialectal variation, may have significantly different /gau/-/kau/ perceptual crossover points in the Lannang and Mainlander talker conditions. Mainlander listeners, with little to no motivation to distinguish themselves from the Lannangs and no motivation to pay attention to the variation, may be inclined to have the similar crossover points for both Mainlander and Lannang talker conditions.