Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Task demands influence phonological representation during syllable perception in tonal languages
Poster B37 in Poster Session B, Friday, October 25, 10:00 - 11:30 am, Great Hall 4
ling liu1, hua fan1, yuting meng1; 1Beijing Language and Culture University
The ability to perceive speech is essential for effective human communication as it allows individuals to differentiate meanings conveyed by different speech sounds. In tonal languages, semantic information can be communicated through three phonological features: consonants, vowels, and lexical tone. Despite numerous studies on the topic, the precise phonological representation during the perception of syllables remains a subject of ongoing investigation. In the present study, we used magnetoencephalography (MEG) to record the responses of participants engaged in a speech perception task in which participants were asked to judge whether the two preceding (audio1) and following (audio2) syllables were the same in terms of their overall (syllables task, subject n = 27), tone (tone task, subject n = 30), or consonants (consonant task, subject n = 19) (three parallel experiments). To explore how task modulates phonological representation, we focus our anlysis on brain responses when subject listened to audio2.Through multivariate pattern analysis (MVPA), we aim to decode phonemic features from MEG sensor signals for either audio1 or audio2. After the onset of audio2, we successfully decoded not only its phonological features but also certain phonological aspects of audio1, indicating that the perception of audio2 triggers the automatic reactivation of phonological information from audio1. Specifically, after the onset of audio2 for 200-400ms, the syllable task reactivated consonant, vowel, and tone. Conversely, the tone task only reactivated tone and vowel, while the consonant task reactivated only consonant and vowel. Collectively, these findings imply that phonological attributes of consonants, vowels, and lexical tone may remain active during the perception of syllables in tonal languages. The encoding of consonants and tone seems to be separate, while the representation of consonant and vowel, or vowel and tone, may be interconnected. We propose a novel hypothesis: the syllabic representation structure of consonant-vowel-tone may stored distinctly as consonant-vowel (C-V) and vowel-tone (V-T) forms.
Topic Areas: Speech Perception, Phonology