Multi-level processing of phonetic variants in speech production and visual word processing: evidence from Mandarin lexical tones

Abstract

AbstractTwo picture–word interference experiments provide new evidence on the nature of phonological processing in speech production and visual word processing. In both experiments, responses were significantly faster either when distractor and target matched in tone category, but had different overt realisations (toneme condition) or when target and distractor matched in overt realisation, but mismatched in tone category (contour condition). Tone 3 sandhi is an allophone of Beijing Mandarin Tone 3 (T3). Its contour is similar to another tone, Tone 2. In Experiment 1, sandhi picture naming was faster with contour (Tone 2) and toneme (low Tone 3) distractors, compared to control distractors. This indicates both category and context-specific representations are activated in sandhi word production. In Experiment 2, both contour (Tone 2) and toneme (low Tone 3) picture naming was facilitated by visually presented sandhi distractors, compared to controls, evidence that category and context-specific instantiated representations are automatically activated during processing of visually presented words. Combined, the results point to multi-level processing of phonology, whether words are overtly produced or processed visually. Interestingly, there were differences in the time course of effects.Keywords: speech productionMandarin Chineselexical tonephonetic variationsub-phonemic detailphonological processingpicture-word interference AcknowledgementsWe would like to express our gratitude to members of the Psychology Institute of the Chinese Academy of Science for their kind assistance while the first author was testing in Beijing. In particular, thanks to Yuan Yuan, Wang Cheng and Li Xiaoqing for help with participant recruitment and lab set-up and to Xia Jinyan for running additional participants. We also wish two thank two anonymous reviewers for their helpful and constructive suggestions on previous versions of this article. This research was conducted while the first author was at Leiden University. She is currently based at the Quantitative Linguistics Group, University of Tuebingen. This work was supported by the European Research Council under the ERC Starting [Grant 206198] to YC.Notes1. Mandarin tones are referred to using a number system (tones 1–4). Here, the numeral following the syllable represents the tone number, in this case, Tone 2.2. The third tone is sometimes described as 'falling-rising'. However, the rising part of the contour is optional and does not usually occur when there is a following syllable. In addition, the gradient of the fall is very shallow. For these reasons and for simplicity, we refer to the contour as 'low'.3. Significance is reported at the 95% confidence level, unless otherwise specified.4. In LME, it is unclear what the appropriate degrees of freedom should be. Therefore, in the lme4 package, p values are not provided in the output. However, the t value provides confidence intervals for sufficiently large datasets (1000 data points or more). T values below −2 or above 2 can be taken as significant at the 95% confidence level (see, e.g. Baayen, Citation2008; Baayen, Davidson, & Bates, Citation2008; Baayen & Milin, Citation2010 for full discussion).5. A Bayesian factor (BF) of 3 or more shows a substantial preference for the model over the alternative. A BF of less than 1 indicates lack of evidence or a preference for the alternative model.