Neurophysiological and Behavioral Responses of Mandarin Lexical Tone Processing

Abstract

Language experience enhances discrimination of speech contrasts at a behavioral- perceptual level, as well as at a pre-attentive level, as indexed by event-related potential (ERP) mismatch negativity (MMN) responses. The enhanced sensitivity could be the result of changes in acoustic resolution and/or long-term memory representations of the relevant information in the auditory cortex. To examine these possibilities, we used a short (ca. 600 ms) vs. long (ca. 2,600 ms) interstimulus interval (ISI) in a passive, oddball discrimination task while obtaining ERPs. These ISI differences were used to test whether cross-linguistic differences in processing Mandarin lexical tone are a function of differences in acoustic resolution and/or differences in long-term memory representations. Bisyllabic nonword tokens that differed in lexical tone categories were presented using a passive listening multiple oddball paradigm. Behavioral discrimination and identification data were also collected. The ERP results revealed robust MMNs to both easy and difficult lexical tone differences for both groups at short ISIs. At long ISIs, there was either no change or an enhanced MMN amplitude for the Mandarin group, but reduced MMN amplitude for the English group. In addition, the Mandarin listeners showed a larger late negativity (LN) discriminative response than the English listeners for lexical tone contrasts in the long ISI condition. Mandarin speakers outperformed English speakers in the behavioral tasks, especially under the long ISI conditions with the more similar lexical tone pair. These results suggest that the acoustic correlates of lexical tone are fairly robust and easily discriminated at short ISIs, when the auditory sensory memory trace is strong. At longer ISIs beyond 2.5 s language-specific experience is necessary for robust discrimination.

The Duration of Auditory Sensory Memory for Vowel Processing: Neurophysiological and Behavioral Measures

“…As predicted, English listeners showed the poorest discrimination and identification for the vowel contrast /y/ vs. /u/, and poorer performance in the long ISI condition. In contrast to Yu et al (2017) , however, we found no effect of ISI reflected in the neural responses, specifically the mismatch negativity (MMN), P3a and late negativity ERP amplitudes. We did see a language group effect, with Mandarin listeners generally showing larger MMN and English listeners showing larger P3a.…”

“…The MMN amplitude was greatly reduced in high repeaters under the long ISI condition compared to that of the short ISI condition, and no MMN was observed for either the short or long ISI conditions in the low repeaters. In Yu et al (2017) , we found that native speakers of English failed to show an early negativity (i.e., MMN) to a Mandarin lexical tone contrast, which is phonemic in Mandarin but not English, when the ISI was greater than 2.5 s; however, English listeners did show MMN to the lexical tone contrast under a short ISI of approximately 500 ms. The Mandarin native listeners showed comparable MMNs for both the long and short ISI.…”

“…The interactions between sensory memory trace decay and auditory discrimination, as measured by the MMN responses, have been investigated using pure tones (e.g., Sams et al, 1993 ), consonant contrast ( Čeponienė et al, 1999 ) and lexical tones ( Yu et al, 2017 ). The accumulated evidence thus far suggests that the rate of sensory memory trace decay may differ depending on the nature of the stimulus.…”

“…Recent evidence suggested that consonants, vowels and lexical tones are weighted differently ( Wiener and Turnbull, 2016 ). In this paper, we examined whether the rate of sensory memory decay for vowels differs from the rate of decay for lexical tone reported in Yu et al (2017) .…”

Effects of native language experience on Mandarin lexical tone processing in proficient second language learners

“…In order to explore the processing of acoustic and phonological information in lexical tones, previous studies have adopted several methods to distinguish the acoustic information from the phonological information in lexical tones, such as the cross-language (native vs. nonnative) comparison (e.g., Chandrasekaran, Krishnan, & Gandour, 2007), speech and nonspeech (hum) comparison (e.g., Jia, Tsang, Huang, & Chen, 2015), interstimulus interval manipulation (Y. H. Yu, Shafer, & Sussman, 2017), and categorical perception paradigm (e.g., Xi et al, 2010).…”